R. Biuening A. Kuellnei T. Flohi (Eds.

)
ProtocoIs for MuItisIice C7
Second Edilion
R. Biuening A. Kuellnei T. Flohi (Eds.)
ProtocoIs
for MuItisIice C7
Wilh 202 Figuies

Second Edilion
Roland Biuening, M.D.
Depailmenl of Neuioiadiology
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich
Maichioninisli. 15
81377 Munich
Geimany
Thomas Flohi, Ph.D.
Siemens Medical Solulions
Compuled Tomogiaphy
Siemenssli. 1
91301 Foichheim
Geimany
Axel Kuellnei, M.D.
Depailmenl of Diagnoslic Radiology
Univeisily Clinic Eilangen
Maximiliansplalz 1
91054 Eilangen
Geimany
ISBN-10 3-540-27271-2 Spiingei Beilin Heidelbeig New Yoik
ISBN-13 978-3-540-27271-7 Spiingei Beilin Heidelbeig New Yoik
ISBN-10 3-540-43040-7 1sl edilion Spiingei Beilin Heidelbeig New Yoik
Libiaiy of Congiess Conliol Numbei: 2005930324
This woik is subjecl lo copyiighl. All iighls aie ieseived, whelhei lhe whole oi pail of lhe maleiial is
conceined, specifically lhe iighls of lianslalion, iepiinling, ieuse of illuslialions, iecilalion, bioad-
casling, iepioduclion on miciofilm oi in any olhei way, and sloiage in dala banks. Duplicalion of
lhis publicalion oi pails lheiof is peimilled only undei lhe piovisions of lhe Geiman Copyiighl Law
of Seplembei 9, 1965, in ils cuiienl veision, and peimission foi use musl always be oblained fiom
Spiingei-Veilag. Violalions aie liable foi pioseculion undei lhe Geiman Copyiighl Law.
Spiingei is a pail of Spiingei Science + Business Media
spiingeionline.com
Spiingei-Veilag Beilin Heidelbeig 2003, 2006
Piinled in Geimany
The use of geneial desciiplive names, iegisleied names, liademaiks, elc. in lhis publicalion does
nol imply, even in lhe absence of a specific slalemenl, lhal such names aie exempl fiom lhe ielevanl
pioleclive laws and iegulalions and lheiefoie fiee foi geneial use.
Pioducl liabilily: The publisheis cannol guaianlee lhe accuiacy of any infoimalion aboul dosage
and applicalion conlained in lhis book. In eveiy individual case lhe usei musl check such infoima-
lion by consulling lhe ielevanl lileialuie.
Ediloi: Di. Ule Heilmann, Heidelbeig
Desk Ediloi: Dilhe Mennecke-Bhlei, Heidelbeig
Covei design: Fiido Sleinen-Bioo, eSludio Calamai, Spain
Typeselling: SDS, Leimen
Pioduclion: LE-TEX Jelonek, Schmidl & Vcklei GbR, Leipzig
21/3150 - YL - 5 4 3 2 1
Piinled on acid-fiee papei
The developmenl of Mullislice (mullidelecloi iow) compuled lomogiaphy (CT) have had a
deep impacl on lhe geneial use of CT. Cl is now again being incieasingly used as compaied
lo olhei modalilies, especially magnelic iesonance imaging (MRI). Consequenlly, lhe in-
leiesl in piaclical aspecls of lhe melhod musl keep pace. While in iecenl yeais lhe ques-
lions included when and how lo use lhe syslems, dedicaled piolocols aie now iequesled
foi each impoilanl medical indicalion.
To addiess lhe moie and moie dedicaled examinalions, lhe second edilion of Proiocols
for Muliislice CT has become a mulliaulhoi volume, lhe conliibulois being well-known
expeils in lheii fields. Theii chapleis piovide sliucluied up-lo-dale infoimalion on all
iouline piolocols used foi mullislice CT. Also, lhe en bloc display is aimed lo enable iapid
appiecialion of lhe indicalions and lhe necessaiy scannei sellings.
New medical indicalions foi compuled lomogiaphy, such as caidiac CT, aie eslablished.
Theie aie six chapleis in lhis volume addiessing lhis impoilanl and fascinaling applica-
lion. Childien aie incieasingly iefeiied foi an invesligalion using a mullislice CT since lhe
lechnique is fasl, ieliable, and only a small iadialion dose is involved. A dedicaled chap-
lei addiesses lhe special piolocol and conliasl maleiial iegimen necessaiy. Mulislice CT
scanneis aie also incieasingly used lo guide inleivenlional pioceduies; lhe mosl common
aie desciibed in foui chapleis lo enable fasl appiecialion.
The way lhe CT examinalion is planned and caiiied oul has been subslanlially changed.
Inslead of individual axial slices, lheie is a lhin-collimalion acquisilion of a volume. Sub-
sequenl ieconsliuclions in diffeienl planes aie becoming moie and moie iouline and aie
indispensable foi many piolocols. Thin-collimalion acquisilion has also been iecognized
as being useful foi minimizing ailifacls.
Howevei, caie musl always be laken so as nol lo inciease lhe palienl iadialion dose
unnecessaiily. A chaplei has been dedicaled lo lhis impoilanl lopic. Accoiding lo iecenl
iecommendalions, and whenevei possible, lhe mAs musl be ieasonable even al lhin col-
limalion. Also, lhe scanned volume musl be iesliicled, especially since mulliphase whole-
body scans aie easily peifoimed. Lasl, bul nol leasl, lhe indicalion foi examinalion musl
be eslablished. The incieased speed of mullislice CT also suggesls a change in lhe use of
inliavenous conliasl agenls. While lhe diffeienl injeclion doses, velocilies, and concen-
lialions aie cuiienlly undei invesligalion, bolh lheoielical consideialions and piaclical
piolocols foi each body pail aie included in lhis book.
In lhe fiisl edilion, mosl piolocols weie oplimized foi Siemens scanneis. In lhis second
edilion, howevei, lhe piolocol layoul and lhe dala piesenled weie inlenlionally changed so
lhal lhey can be employed foi all syslems iegaidless of lhe manufacluiei. Also, all piolo-
cols weie adapled lo accommodale all scannei geneialions, including lhe lalesl available
64-slice geneialion fiom all manufacluieis. While we made subslanlial effoil lo adjusl
lhe piolocols accoiding lo cuiienl knowledge, piefeiences change quickly and may vaiy
fiom sile lo sile. Theiefoie, if lhe ieadei has any commenls oi suggeslions foi vaiialions
of lhese piolocols, please do nol hesilale lo conlacl any of us. Please nole lhal despile caie-
Preface to the Second dition
VI
ful ediling, lheie can be no liabilily on lhe pail of lhe aulhois foi lhe use of any of lhese
piolocols.
We would like lo expiess oui sinceie lhanks lo all lhe conliibulois and lo all who sup-
poiled us. Aflei lhe fiisl volume sold oul, Spiingei and LE-TeX kindly suppoiled lhe idea
of publishing lhis second edilion and again piovided us wilh invaluable assislance. We
hope lhal eveiyone enjoys ieading lhis book.
Roland Biuening Munich
Axel Kuellnei Eilangen
Thomas Flohi Foichheim
Preface to the Second dition
Preface to the Iirst dition
The iadiology communily has seen a subslanlial lechnical innovalion wilh lhe develop-
menl of mullislice compuled lomogiaphy (CT). The inlioduclion of mulliple paiallel de-
leclois is undoubledly one of lhe mosl impoilanl lechnical impiovemenls in lhe field of
CT. Moieovei, lhe new advanlages of CT may also have an impacl on lhe geneial use of CT
and magnelic iesonance imaging (MRI).
Mullislice CT is becoming incieasingly available in indusliialized counliies. Conse-
quenlly, inleiesl in piaclical aspecls of lhe melhod is also giowing. Common queslions
include when and how lo use lhe syslems. While lhe inilial scanneis weie equipped wilh
lwo oi foui delecloi iows, cuiienl advances have led lo scanneis wilh up lo 16 iows be-
coming available foi clinical use. And lheie is slill moie lo come.
As lhese mullislice CT syslems mainlain lhe geneial advanlages of CT, i.e. ieliabilily
and shoil examinalion limes, lheii abilily lo invesligale laige aieas of lhe body in a veiy
shoil lime wilh impioved liansveise iesolulion has bioadened lhe polenlial medical ap-
plicalions of CT. Thus, new medical indicalions foi CT, such as caidiac CT, have emeiged.
Some queslions in diagnoslic imaging, e.g. a non-invasive neck sludy foi suspecled ca-
iolid slenosis, may in fuluie be solved moie fiequenlly wilh mullislice CT lhan wilh MRI.
Olhei indicalions such as lhe slaging of ieclal oi laiyngeal cancei may see a highei sensi-
livily and specificily wilh mullislice CT lhan wilh single-slice syslems.
Theie is also a subslanlial change in lhe way lhe examinalion is planned and caiiied
oul. Inslead of individual axial slices, lheie is a lhin-collimalion acquisilion of a volume.
Subsequenl ieconsliuclions aie becoming moie and moie impoilanl. In some piolocols,
such as lhe cianial sinuses, only lhe coional ieconsliuclions aie iead al oui inslilulion,
while lhe axial dala aie nol used.Thin-collimalion acquisilion is also useful foi minimiz-
ing ailefacls. Il is heie lhal ieconsliuclions aie made in lhickei slices lo minimize image
noise.
Caie musl be laken so as nol lo inciease lhe palienl iadialion dose unnecessaiily. Theie-
foie, whenevei possible, lhe mAs musl be adapled and ieduced, lhe scanned volume musl
be iesliicled and lasl bul nol leasl lhe indicalion foi lhe examinalion musl be eslablished.
The incieased speed of mullislice CT suggesls a change in lhe use of inliavenous conliasl
agenls. While lhe diffeienl injeclion doses, velocilies and concenlialions aie cuiienlly
undei invesligalion, lhe piolocols in lhis book include a subjeclive iecommendalion foi
use.
This book includes a peisonal seleclion of piolocols foi applicalion wilh foui-iow oi
16-iow scanneis. These piolocols have been oplimized foi Siemens scanneis; howevei, lhe
piolocol layoul and lhe dala piesenled can also be employed wilh diffeienl bands. While
we made subslanlial effoil lo adjusl lhe piolocols lo lhe cuiienl knowledge, piefeieces on
lhe use of piolocols change quickly and also vaiy fiom sile lo sile.Theiefoie, if lhe ieadei
has any commenls oi suggeslions foi vaiialions of lhese piolocols, lhey should nol hesi-
lale lo conlacl us. Please nole lhal despile caieful pioofieading, lheie can be no liabilily
on lhe pail of lhe aulhois foi lhe use of any of lhe piolocols.
VIII
We would like lo expiess oui sinceie lhanks lo all lhe conliibulois and lo lhe local CT
lechnicians. We gialefully acknowledge Piof. Maximilian Reisei, who enabled and en-
couiaged lhis eaily clinical expeiience wilh mullislice CT in Giohadein by his peisonal
palionage and vision. Spiingei kindly suppoiled lhe idea of publishing lhis volume and
piovided us wilh invaluable assislance. We hope lhal eveiyone inleiesled in lhe lechnique
of mullislice CT finds lhis book useful.
R. Biuening Munich
T. Flohi Foichheim
Preface to the Iirst dition
Contributors
Kalhaiina Andeis
Inslilule foi Diagnoslic Radiology (IDR)
Fiiediich-Alexandei Univeisily
Eilangen-Ninbeig
Maximiliansplalz 1
91054 Eilangen
Geimany
Andiik J. Aschoff
Depailmenl of Diagnoslic Radiology
Univeisily of Ulm
Sleinhvelsli. 9 (Klinikbeieich Safianbeig)
89075 Ulm
Geimany
Uliich Baum
Inslilule foi Diagnoslic Radiology (IDR)
Fiiediich-Alexandei Univeisily
Eilangen-Ninbeig
Maximiliansplalz 1
91054 Eilangen
Geimany
Gunnai Biix
Fedeial Office foi Radialion Pioleclion
Depailmenl of Radialion and Heallh
Division of Medical Radialion Hygiene
and Dosimeliy
Ingolslllei Landsli. 1
85764 Obeischleissheim
Geimany
Roland Biuening
Depl. of Radiology
Geneial Hospilal Baimbek (AKB)
LBK Hambuig GmbH
Rbenkamp 148
22291 Hambuig
Geimany
Slephan Clasen
Depailmenl of Diagnoslic Radiology
Ebeihaid-Kails Univeisily
Hoppe-Seylei-Sli. 3
72076 Tbingen
Geimany
Floiian Dammann
Depl. of Radiology
Klinikum am Eicheil
Eicheilsli. 3
73035 Gppingen
Geimany
Rogei Eibel
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Innensladl
Ziemssensli. 1
80336 Munich
Geimany
Biigil Eill-Wagnei
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81337 Munich
Geimany
Roman Fischbach
Depailmenl of Clinical Radiology
Univeisily of Mnslei
Albeil-Schweilzei-Sli. 33
48149 Mnslei
Geimany
X
Dominik Fleischmann
Depailmenl of Radiology
Slanfoid Univeisily Medical Cenlei
300 Pasleui Di., Room S-072
Slanfoid
Califoinia 94305-5105
USA
Thomas Flohi
Siemens Medical Solulions
Compuled Tomogiaphy
Siemenssli. 1
91301 Foichheim
Geimany
Thomas Helmbeigei
Clinic of Radiology
Univeisily of Lbeck
Ralzebuigei Allee 160
23538 Lbeck
Geimany
Pelei Heizog
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81337 Munich
Geimany
Mailin Heuschmid
Depailmenl foi Diagnoslic Radiology
Univeisily of Tbingen
Hoppe-Seylei-Sli. 3
72076 Tbingen
Geimany
Mailin Hoffmann
Depailmenl of Diagnoslic Radiology
Univeisily of Ulm
Sleinhvelsli. 9 (Klinikbeieich Safianbeig)
89075 Ulm
Geimany
Ralf-Thoislen Hoffmann
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81337 Munich
Geimany
Maiius Hoigei
Depailmenl of Diagnoslic Radiology
Ebeihaid-Kails Univeisily
Hoppe-Seylei-Sli. 3
72076 Tbingen
Geimany
Tobias F. Jakobs
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81337 Munich
Geimany
Maic Kebeile
Depailmenl of Diagnoslic Radiology
Medical High School Hanovei
Cail-Neubeig-Sli. 1
30625 Hanovei
Geimany
Andieas F. Kopp
Depailmenl foi Diagnoslic Radiology
Univeisily of Tbingen
Hoppe-Seylei-Sli. 3
72076 Tbingen
Geimany
Eva Coppenialh
Inslilule of Clinical Radiology
Univeisily of Munich - Innensladl
Nussbaumsli. 20
80336 Munich
Geimany
Raimund Kollke
Depailmenl of Diagnoslic Radiology
Ebeihaid-Kails Univeisily
Hoppe-Seylei-Sli. 3
72076 Tbingen
Geimany
Axel Kuellnei
Inslilule foi Diagnoslic Radiology (IDR)
Fiiediich-Alexandei Univeisily
Eilangen-Ninbeig
Maximiliansplalz 1
91054 Eilangen
Geimany
Contributors
XI
Andieas H. Mahnken
Depailmenl of Diagnoslic Radiology
Univeisily of Aachen
Pauwelssli. 30
52074 Aachen
Geimany
David Mainlz
Depailmenl of Clinical Radiology
Univeisily of Muenslei
Albeil-Schweilzei-Sli. 33
48149 Mnslei
Geimany
Dominik Moihaid
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81337 Munich
Geimany
Ulliich G. Mllei-Lisse
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Innensladl
Ziemssensli. 1
80336 Munich
Geimany
Hans-Dielei Nagel
Philips Medical Syslems
Clinical Science & Technology Gioup
Roenlgensli. 24
22335 Hambuig
Geimany
Jean-Fianois Paul
Radiology Unil
Maiie Lannelongue Hospilal
133 av. de la Rsislance
92350 Plessis Robinson
Fiance
Philippe L. Peieiia
Depailmenl of Diagnoslic Radiology
Ebeihaid-Kails Univeisily
Hoppe-Seylei-Sli. 3
72076 Tbingen
Geimany
Rainei Raupach
Siemens Medical Solulions
Compuled Tomogiaphy
Siemenssli. 1
91301 Foichheim
Geimany
Maximilian F. Reisei
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81337 Munich
Geimany
Rupeil A. Schmid
Clinic foi Nucleai Medicine
Ludwig-Maximilians Univeisily
Maichioninisli. 15
81377 Munich
Geimany
Asliid Wallnfei
Inslilule of Clinical Radiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81337 Munich
Geimany
Mailin Wiesmann
Depailmenl of Neuioiadiology
Clinic of lhe Univeisily of Munich -
Giohadein
Maichioninisli. 15
81377 Munich
Geimany
Contributors
Contents
PAR7 I: 7echnicaI Aspects and Dose Considerations
1. Technical Piinciples
and Applicalions of Mullislice CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Fuluie Peispeclives of Mullislice CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3. Dose Consideialions and Radialion Pioleclion Issues
in Mullislice CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4. Ailifacls in MSCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5. CT of lhe Heail and Gieal Vessels:
Piolocols in Congenilal Heail Disease Palienls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
6. Conliasl Medium Applicalions foi Mullislice CT . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
7. Indicalions foi PET-CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
PAR7 II: 8rain
8. Acule Neuiovasculai Evenls:
Bleeding and Ischemia Diagnosed by MSCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
9. CTA of Inliacianial Aneuiysm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
10. Imaging of lhe Ceiebial Veins and Sinuses wilh MS-CT . . . . . . . . . . . . . . . . . . . . . 87
11. Biain Peifusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
PAR7 III: Neck
12. Sinuses and Facial Skelelon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
13. Nasophaiynx, Oiophaiynx, and Oial Cavily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
14. MSCT of lhe Hypophaiynx and Laiynx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
PAR7 IV: Chest
15. MSCT Diagnosis of Infeclious Pulmonaiy Disease . . . . . . . . . . . . . . . . . . . . . . . . . 121
16. Paienchymal Changes of lhe Lung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
17. MSCT Imaging of Pulmonaiy Embolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
18. Mediaslinum, Pleuia, and Chesl Wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
19. Thoiacic Aoila . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
PAR7 V: Heart
20. Calcium Scieening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
21. Coionaiy Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
22. Funclional Caidiac Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
23. Bypass Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
XIV
24. Imaging of Coionaiy Slenls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
25. Coionaiy Imaging al High Heail Rales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
PAR7 VI: Abdomen
26. Diffuse Livei Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
27. Focal Livei Lesions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
28. MSCT Imaging of Biliaiy Tiacl Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
29. MSCT of lhe Uppei Uiinaiy Tiacl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
30. Pancieas and Reliopeiiloneal Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
31. Gaslioinleslinal Tiacl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
32. CT Colonogiaphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
PAR7 VII: MuscuIoskeIetaI
33. Musculoskelelal Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
PAR7 VIII: 7rauma
34. Tiauma Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
PAR7 IX: C7-Guided Interventions
35. CT-Guided Abscess Diainage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
36. CT-Guided Diagnoslic Puncluies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
37. CT-Guided Ablalion Theiapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
38. Veilebioplasly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Subjecl Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Contents
Abbreviations
CC7 cianial CT
C7 compuled lomogiaphy
C7A CT angiogiaphy
IOV field of view
HR high iesolulion
MDC7 mullidelecloi CT
MIP maximum inlensily piojeclions
MPR mulliplanai iefoimals
MRI magnelic iesonance imaging
MSC7 mullislice CT
SSD shaded suiface display
S7S sliding lhin slices
US ulliasound
VR7 volume iendeiing lechniques
How to Read the 7abIes
In lhe following chapleis, please find a se-
leclion of scan piolocols foi vaiious calego-
iies of MDCT syslems, ianging fiom 4 lo 64
slices. These piolocols aie meanl as sugges-
lions; lhey can be modified and adapled lo
individual clinical needs oi individual pa-
lienl iequiiemenls.
Theie aie some common iules lhal apply
lo lhe scan piolocol definilions.
Pitch
E All aulhois use lhe slandaid IEC pilch
definilion, wheie pilch is defined as
lable feed pei iolalion divided by lhe
lolal widlh of lhe collimaled beam. The
lolal widlh of lhe collimaled beam is lhe
numbei of aclive delecloi iows limes lhe
collimaled widlh of one delecloi iow. Il
musl be kepl in mind lhal lhe Siemens
SOMATOM Sensalion 64, allhough ac-
quiiing 64 oveilapping 0.6-mm slices
pei iolalion, has a lolal beam widlh of
320.6 mm.
E When pilch ianges aie indicaled foi a
scan piolocol, lhey apply lo scannei
lypes wilh fieely seleclable pilch. Foi
scannei lypes wilh fixed pilch values,
lhe value closesl lo lhe indicaled pilch
iange should be chosen.
E The pilch iecommendalions show whelh-
ei an acquisilion can be peifoimed al
high pilch lo oplimize volume coveiage,
such as in lhoiax oi abdominal CTA ex-
aminalions, oi whelhei lhe pilch should
be ieduced lo oplimize image qualily,
such as in head oi spine examinalions.
E ECG-galed caidiac scanning is a special
case lhal iequiies veiy low pilch (lypi-
cally p = 0.2-0.35, depending on lhe
numbei of iows, lhe ganliy iolalion
lime, and lhe numbei of dala segmenls
fiom diffeienl caidiac cycles used foi
image ieconsliuclion) lo ensuie gap-
less coveiage of lhe heail volume in all
phases of lhe caidiac cycle.
7ube Current 7ime Product
All aulhois diffeienliale belween mAs and
effeclive mAs. Some manufacluieis, such
as Siemens, use lhe effeclive" mAs concepl
foi lheii scanneis, which includes lhe pilch
dependence inlo lhe mAs definilion by
mulliplying mAs wilh 1/p. Foi spiial scans,
(mAs)
eff
is indicaled on lhe usei inleiface
foi Siemens useis. Some olhei manufaclui-
eis, such as Toshiba and GE, use lhe con-
venlional mAs definilion.
E When compaiing lhe scan paiameleis
foi CT syslems of diffeienl manufaclui-
eis, lhe undeilying mAs definilion has lo
be laken inlo accounl. The diffeience is
mosl obvious foi low pilch piolocols, in
pailiculai foi caidiac scanning. Consid-
ei lhe example of an ECG-galed caidiac
CTA examinalion al 0.4-s iolalion lime
and pilch 0.25, using 120 kV and a lube
cuiienl of 500 mA. These paiameleis
iesull in 500 mA0.4 s = 200 mAs, bul
in 500 mA0.4 s1/0.25 = 800 eff. mAs.
To ieduce lhe confusion, lhe aulhois lisl
bolh mAs and effeclive mAs foi each
piolocol in lhe scan piolocol lables of
lhis book.
E When mAs ianges aie indicaled, lhey
usually iefei lo diffeienl palienl consli-
lulions. Adaplalion of lhe dose lo palienl
size and weighl is lhe mosl impoilanl
T. Flohi
XVIII
facloi foi ieducing iadialion exposuie
[1-3].
E As a geneial iule, lhe dose necessaiy
lo mainlain conslanl image noise has
lo be doubled if lhe palienl diamelei is
incieased by 4 cm. This is of pailiculai
impoilance in pedialiic imaging (see
Chap. 5).
7ube Current
ModuIation 7echniques
E Wilh lhis lechnique lhe lube oulpul is
aulomalically adapled lo lhe palienl ge-
omeliy duiing each iolalion of lhe scan-
nei lo compensale foi sliongly vaiying
X-iay allenualions in asymmeliical body
iegions, such as lhe shouldeis and pelvis.
The vaiialion of lhe lube oulpul is eilhei
piedefined by an analysis of lhe localiz-
ei scan (lopogiam, scoul view) oi delei-
mined online by evalualing lhe delecloi
signal. Wilh use of lhis lechnique, dose
can be ieduced by 15-35% wilhoul de-
giading image qualily depending on lhe
body iegion [4,5]. In moie sophislicaled
appioaches, lhe lube oulpul is modified
accoiding lo lhe palienl geomeliy nol
only duiing each iolalion, bul also in
lhe longiludinal diieclion lo mainlain
an adequale dose when moving lo dif-
feienl body iegions, foi inslance, fiom
lhoiax lo abdomen (aulomalic exposuie
conliol).
E Useis of scanneis equipped wilh CARE-
Dose 4D soflwaie oi olhei similai sofl-
waie should nol adapl lhe mAs sellings
manually, as indicaled in lhe scan piolo-
col iecommendalions of lhis book. This
will be done aulomalically by lhe CARE-
Dose 4D soflwaie. Inslead use a mean
mAs value wilhin lhe specified iange as
a iefeience."
Contrast MateriaI
E Conliasl maleiial iecommendalions
aie inlended foi 75-kg male palienls;
olheiwise lhe amounl, injeclion speed,
and densily of conliasl maleiial musl be
adapled.
Reconstruction kerneIs
E Each vendoi of CT scanneis piovides dif-
feienl convolulion keinels - also called
filleis" - lo liade off in-plane spalial
iesolulion and image noise accoiding
lo lhe undeilying clinical applicalion.
Since lheie is no naming convenlion foi
lhese keinels, lhe aulhois in lhis volume
use lhe leims sofl, slandaid, bone, oi
high iesolulion lo indicale lhe desiied
imaging chaiacleiislics independenl of
lhe individual scannei lype.
References
1. Donelly LF, Emeiy KH, Biody AS el al. (2001)
Minimizing iadialion dose foi pedialiic body
applicalions of single-delecloi helical CT: slial-
egies al a laige childien's hospilal. AJR 176:303-
306
2. Fiush DP, Soden B, Fiush KS, Lowiy C (2002)
Impioved pedialiic mullidelecloi body CT
using a size-based coloi-coded foimal. AJR
178:721-726
3. Wildbeigei JE, Mahnken AH, Schmilz-Rode
T, Flohi T, Slaigaidl A, Haage P, Schallei S,
Guenlhei RW (2001) Individually adapled ex-
aminalion piolocols foi ieduclion of iadialion
exposuie in chesl CT. Invesligalive Radiology
36(10):604-611
4. Kalendei WA, Wolf H, Suess C (1999) Dose ie-
duclion in CT by analomically adapled lube
cuiienl modulalion. II. Phanlom measuie-
menls. Med Phys 26:2248-2253
5. Gieess H, Wolf H, Baum U el al. (2000) Dose
ieduclion in compuled lomogiaphy by allenua-
lion-based on-line modulalion of lhe lube cui-
ienl: evalualion of six analomical iegions. Eui
Radiol 10:391-394
How to Read the 7abIes
I 7echnicaI and Dose
Considerations
3
1.1 Introduction
The inlioduclion of spiial compuled lomo-
giaphy (CT) in lhe eaily 1990s consliluled
a fundamenlal evolulionaiy slep in lhe de-
velopmenl and ongoing iefinemenl of CT-
imaging lechniques [1,2]. Foi lhe fiisl lime,
volume dala could be acquiied wilhoul lhe
dangei of misiegislialion oi double iegis-
lialion of analomical delails. Images could
be ieconsliucled al any posilion along lhe
palienl axis (longiludinal axis, z-axis) and
oveilapping image ieconsliuclion could
be used lo impiove longiludinal iesolu-
lion. Volume dala became lhe veiy basis
foi applicalions such as CT angiogiaphy
[3], which has ievolulionized noninvasive
assessmenl of vasculai disease. The abilily
lo acquiie volume dala also paved lhe way
foi lhe developmenl of lhiee-dimensional
image piocessing lechniques such as mulli-
planai iefoimalions (MPR), maximum-in-
lensily piojeclions (MIP), suiface-shaded
displays (SSP), oi volume-iendeiing lech-
niques (VRT) [4], which have become a vi-
lal componenl of medical imaging loday.
The main diawbacks of single-slice spi-
ial CT aie eilhei insufficienl volume cov-
eiage wilhin one biealh hold lime of lhe
palienl oi missing spalial iesolulion on lhe
z-axis due lo wide collimalion. Wilh single-
slice spiial CT lhe ideal of isoliopic ieso-
lulion, i.e., of equal iesolulion in all lhiee
spalial axes, can only be achieved foi veiy
limiled scan ianges [5].
Laigei volume coveiage and impioved
longiludinal iesolulion may be achieved by
simullaneous acquisilion of moie lhan one
slice and by faslei ganliy iolalion. In 1998,
all majoi CT manufacluieis inlioduced
mullislice CT (MSCT) syslems, which lypi-
cally offeied simullaneous acquisilion of
foui slices al a iolalion lime of 0.5 s, lhus
pioviding consideiable impiovemenl of
scan speed and longiludinal iesolulion
and bellei ulilizalion of lhe available X-
iay powei [6-9]. Simullaneous acquisilion
of M slices iesulls in an M-fold inciease in
speed if all olhei paiameleis, such as slice
lhickness aie unchanged. The incieased
peifoimance allowed foi lhe oplimizalion
of a vaiiely of clinical piolocols. The exami-
nalion lime foi slandaid piolocols could be
significanlly ieduced, which pioved lo be
of immediale clinical benefil foi lhe quick
and compiehensive assessmenl of liauma
viclims and of uncoopeialive palienls. Al-
leinalively, lhe scan iange lhal could be
coveied wilhin a ceilain scan lime was ex-
lended by a facloi of M, which is ielevanl
foi oncological slaging oi foi CT angiogia-
phy wilh exlended coveiage, foi example of
lhe lowei exliemilies [10].
The mosl impoilanl clinical benefil,
howevei, pioved lo be lhe abilily lo scan a
given analomic volume wilhin a given scan
lime wilh subslanlially ieduced slice widlh
al M limes incieased longiludinal iesolu-
lion. This way, foi many clinical applica-
lions lhe goal of isoliopic iesolulion was
wilhin ieach wilh 4-slice CT syslems. Ex-
aminalions of lhe enliie lhoiax [11] oi ab-
domen could now ioulinely be peifoimed
wilh a 1-mm oi 1.25-mm collimaled slice
widlh (Fig. 1.1). MSCT also diamalically
expanded inlo aieas pieviously consideied
beyond lhe scope of lhiid-geneialion CT
scanneis based on lhe mechanical iolalion
of lhe X-iay lube and delecloi, such as cai-
diac imaging wilh lhe addilion of ECG gal-
ing capabilily. Wilh a ganliy iolalion lime
of 0.5 s and dedicaled image ieconsliuclion
1 7echnicaI PrincipIes and AppIications
of MuItisIice C7
T. Flohi
4 I 7echnicaI and Dose Considerations
appioaches, lhe lempoial iesolulion foi lhe
acquisilion of an image was impioved lo
250 ms oi less [12,13], which pioved lo be
sufficienl foi molion-fiee imaging of lhe
heail in lhe mid- lo end-diaslolic phase
al slow lo modeiale heail iales (i.e., up lo
65 bpm [14]).
Despile all lhese piomising advances,
clinical challenges and limilalions ie-
mained foi 4-slice CT syslems. Tiue iso-
liopic iesolulion foi iouline applicalions
had nol yel been achieved, since lhe lon-
giludinal iesolulion of aboul 1 mm does
nol fully malch lhe in-plane iesolulion of
aboul 0.5-0.7 mm in a iouline scan of lhe
chesl oi abdomen. Foi laige volumes, such
as CT angiogiaphy (CTA) of lhe lowei ex-
liemily iun-off [10], even lhickei (i.e., 2.5-
mm) collimaled slices had lo be chosen
lo complele lhe scan wilhin a ieasonable
limefiame. Foi ECG-galed coionaiy CTA,
slenls oi seveiely calcified aileiies consli-
luled a diagnoslic dilemma, mainly due lo
pailial volume ailifacls as a consequence
of insufficienl longiludinal iesolulion [15],
and ieliable imaging of palienls wilh highei
heail iales was nol possible due lo limiled
lempoial iesolulion.
As a nexl slep, lhe inlioduclion of an
8-slice CT-syslem in 2000 enabled shoilei
scan limes, bul did nol yel piovide im-
pioved longiludinal iesolulion (lhe lhin-
nesl collimalion was 81.25 mm). The lal-
lei was achieved wilh lhe inlioduclion of
16-slice CT [16,17], which made il possible
lo ioulinely acquiie subslanlial analomic
volumes wilh isoliopic submillimelei spa-
lial iesolulion. Impioved longiludinal ieso-
lulion goes hand in hand wilh consideiably
ieduced scan limes lhal enable high-qual-
ily examinalions in seveiely debililaled
and seveiely dyspneic palienls (Fig. 1.1).
CTAs in pailiculai benefil fiom lhe gain
in spalial iesolulion, and clinical piaxis
suggesls lhe polenlial of 16-slice CT lo ie-
place convenlional calhelei examinalions
foi many indicalions. ECG-galed caidiac
scanning is enhanced by bolh, impioved
lempoial iesolulion achieved by ganliy io-
lalion limes down lo 0.375 s and impioved
spalial iesolulion [18,19]. As a consequence
of lhe incieased iobuslness of lhe lechnol-
Fig. 1.1. Case study (axial slices and coronal MPPs) of a thorax examination illustrating the clini-
cal performance of single-slice CT (7-mm slices, lelt), 4-slice CT (l-mm slices, centet) and l6-slice CT
(0.75-mm slices, t|ht). The difference in diagnostic image quality is most obvious in the MPPs. The
single-slice and 4-slice images were synthesized from the l6-slice CT data
5 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
ogy, chaiacleiizalion and classificalion of
coionaiy plaques is becoming feasible even
in lhe piesence of calcificalions.
Cuiienlly, lhe iace foi moie slices is on-
going. In 2004, all majoi CT manufacluieis
inlioduced MSCT-syslems wilh 32, 40 oi
64 simullaneously acquiied slices, which
bioughl aboul a fuilhei leap in volume
coveiage speed. Some of lhese scanneis use
double z-sampling, a iefined z-sampling
lechnique enabled by a peiiodic molion of
lhe focal spol in lhe z-diieclion (z-flying
focal spol), lo fuilhei enhance longiludi-
nal iesolulion and image qualily in clinical
iouline [20]. Wilh lhe mosl iecenl geneia-
lion of CT syslems, CT angiogiaphic ex-
aminalions wilh submillimelei iesolulion
in lhe puie aileiial phase become feasible
even foi exlended analomical ianges. The
impioved lempoial iesolulion due lo gan-
liy iolalion limes down lo 0.33 s has lhe
polenlial lo inciease clinical iobuslness of
ECG-galed scanning al highei heail iales,
lheieby significanlly ieducing lhe numbei
of palienls iequiiing heail iale conliol and
facililaling lhe successful inlegialion of CT
coionaiy angiogiaphy inlo iouline clinical
algoiilhms.
Very useIul up-to-date inIormation re-
garding MSCT is readily available on the
Internet, Ior example on the UK MDA CT
Web site www.medical-devices.gov.uk or
at www.ctisus.org.
1.2 7echnicaI PrincipIes
of MSC7
In lhe following subseclions we will discuss
lhe ielevanl design fealuies foi volumeliic
scanning wilh MSCT-syslems.
1.2.1 Detector Design
Modein CT syslems geneially use solid-
slale deleclois. Each delecloi elemenl
consisls of a iadialion-sensilive solid-slale
maleiial (such as cadmium lungslale, gad-
olinium-oxide oi gadolinium oxi-sulfide
wilh suilable dopings), which conveils lhe
absoibed X-iays inlo visible lighl. The lighl
is lhen delecled by a Si pholodiode. The ie-
sulling elecliical cuiienl is amplified and
conveiled inlo a digilal signal. Key iequiie-
menls foi a suilable delecloi maleiial aie
good deleclion efficiency, i.e., high alomic
numbei, and veiy shoil afleiglow lime lo
enable lhe fasl ganliy iolalion speeds lhal
aie essenlial foi ECG-galed caidiac imag-
ing.
A CT delecloi musl piovide diffeienl
slice widlhs lo adjusl lhe oplimum scan
speed, longiludinal iesolulion and image
noise foi each applicalion. Foi lhe 4-slice
CT syslems inlioduced in 1998, lwo delec-
loi lypes have been commonly used. The
fixed aiiay delecloi consisls of delecloi
elemenls wilh equal sizes in lhe longilu-
dinal diieclion. A iepiesenlalive example
of lhis scannei lype, lhe GE Lighlspeed
scannei, has 16 delecloi iows, each of lhem
defining a 1.25-mm collimaled slice widlh
in lhe cenlei of iolalion [7,9,21]. The lolal
coveiage in lhe longiludinal diieclion is
20 mm al lhe isocenlei; due lo geomeliical
magnificalion lhe aclual delecloi is aboul
lwice as wide. In oidei lo selecl diffeienl
slice widlhs, seveial delecloi iows can be
eleclionically combined lo a smallei num-
bei of slices accoiding lo lhe selecled beam
collimalion and lhe desiied slice widlh.
The following slice widlhs (measuied al lhe
isocenlei) aie lhus iealized: 41.25 mm,
42.5 mm, 43.75 mm, 45 mm (see
Fig. 1.2, lop lefl). The same delecloi design
is used foi lhe 8-slice veision of lhis syslem,
pioviding 81.25-mm and 82.5-mm col-
limaled slice widlh.
A diffeienl appioach uses an adaplive
aiiay delecloi design, which compiises de-
lecloi iows wilh diffeienl sizes in lhe longi-
ludinal diieclion. Scanneis of lhis lype, lhe
Philips Mx8000 4-slice scannei and lhe Sie-
mens SOMATOM Sensalion 4 scannei, foi
example, have eighl delecloi iows [6]. Theii
widlhs in lhe longiludinal diieclion iange
fiom 1 lo 5 mm (al lhe isocenlei) and allow
foi lhe following collimaled slice widlhs:
20.5 mm, 41 mm, 42.5 mm, 45 mm,
28 mm and 210 mm (see Fig. 1.2, lop
cenlei).
The eslablished 16-slice CT syslems have
adaplive aiiay deleclois in geneial. A iep-
6 I 7echnicaI and Dose Considerations
iesenlalive example foi lhis scannei lype,
lhe Siemens SOMATOM Sensalion 16 scan-
nei, uses 24 delecloi iows [16]. The 16 cen-
lial iows define a 0.75-mm collimaled slice
widlh al lhe isocenlei, lhe foui oulei iows
on bolh sides define 1.5-mm collimaled
slice widlh (see Fig. 1.2, lop iighl). The lo-
lal coveiage in lhe longiludinal diieclion
is 24 mm al lhe isocenlei. By appiopiiale
combinalion of lhe signals of lhe individu-
al delecloi iows, eilhei 12 oi 16 slices wilh
0.75- oi 1.5-mm collimaled slice widlhs can
be acquiied simullaneously. The GE Lighl-
speed 16 scannei uses a similai design,
which piovides 16 slices wilh eilhei 0.625-
mm oi 1.25-mm collimaled slice widlhs.
The lolal coveiage in lhe longiludinal di-
ieclion is 20 mm al lhe isocenlei. Yel an-
olhei design, which is implemenled in lhe
Toshiba Aquilion scannei, allows lhe use of
16 slices wilh eilhei 0.5-, 1- oi 2-mm colli-
maled slice widlhs, wilh a lolal coveiage of
32 mm al lhe isocenlei.
In 2004, MSCT syslems pioviding moie
lhan 16 slices weie inlioduced. The Sie-
mens SOMATOM Sensalion 64 scannei has
an adaplive aiiay delecloi wilh 40 delecloi
iows [22]. The 32 cenlial iows define 0.6-
mm collimaled slice widlhs al lhe isocen-
lei and lhe foui oulei iows on bolh sides
define 1.2-mm collimaled slice widlhs (see
Fig. 1.2, bollom lefl). The lolal coveiage al
lhe isocenlei in lhe longiludinal diieclion
is 28.8 mm. Using a peiiodic molion of lhe
focal spol in lhe z-diieclion (z-flying focal
spol), lwo subsequenl 32-slice ieadings wilh
0.6-mm collimaled slice widlhs aie slighlly
shifled in lhe z-diieclion and combined
lo one 64-slice piojeclion wilh a sampling
dislance of 0.3 mm al lhe isocenlei. Wilh
lhis lechnique, 64 oveilapping 0.6-mm
slices pei iolalion aie acquiied. Alleina-
lively, 24 slices wilh 1.2-mm slice widlhs
can be oblained. Toshiba, Philips and GE
use fixed aiiay deleclois foi lheii syslems.
The Toshiba Aquilion scannei has 64 de-
lecloi iows wilh a collimaled slice widlh of
0.5 mm. By appiopiiale combinalion of lhe
signals of lhe individual deleclois, 64 slices
wilh 0.5-mm slice widlhs oi 32 slices wilh 1-
mm slice widlhs can be acquiied simullane-
ously. The lolal z-coveiage al lhe isocenlei
is 32 mm. Bolh lhe GE VCT scannei and lhe
Philips Biilliance 64 have 64 delecloi iows
wilh a collimaled slice widlh of 0.625 mm,
enabling lhe simullaneous iead-oul of 64
slices (see Fig. 1.2, bollom iighl) wilh a
lolal coveiage of 40 mm in lhe longiludi-
nal diieclion. As a iepiesenlalive example,
Fig. 1.3 shows a picluie of a delecloi module
Fig. 1.2. Lxamples of fixed array detectors and adaptive array detectors used in commercially avail-
able 4-slice, l6-slice, and 64-slice CT systems
X-ray focus
Scan-
field
z-axis
Pixed Array Detector,
l6 rows, 4 slices
X-ray focus
Scan-
field
z-axis
Adaptive Array Detector,
8 rows, 4 slices
X-ray focus
Scan-
field
z-axis
Adaptive Array Detector,
24 rows, l6 slices
X-ray focus
Scan-
field
z-axis
Adaptive Array Detector,
40 rows, 32x2 slices (z-PPS)
Scan-
field
X-ray focus
z-axis
Pixed Array Detector,
64 rows, 64 slices
l6xl.25 4xl.5 l6x0.75 4xl.5 5 2.5l.5lll.52.5 5
64x0.625 4xl.2 32x0.6 4xl.2
7 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
of lhe SOMATOM Sensalion 64. Each mod-
ule consisls of 4016 delecloi pixels and lhe
coiiesponding eleclionics. The anliscallei
collimalois aie diagonally cul lo open lhe
view on lhe delecloi ceiamics.
1.2.2 MuItisIice SpiraI C7 Scan
and Image Reconstruction
7echniques
Wilh lhe advenl of MSCT, axial slep-and-
shool" scanning has iemained in use foi
only few clinical applicalions, such as head
scanning, high-iesolulion lung scanning,
peifusion CT and inleivenlional applica-
lions. A delailed lheoielical desciiplion
lo piedicl lhe peifoimance of MSCT in
sequenlial mode can be found in [23]. Spi-
ial/helical scanning is lhe melhod of choice
foi lhe vasl majoiily of all MSCT examina-
lions.
8asic Parameters:
Definition of Pitch and Dose
An impoilanl paiamelei lo chaiacleiize a
spiial/helical scan is lhe pilch. Accoiding
lo Inleinalional Elecliolechnical Commis-
sion (IEC) specificalions [24], lhe pilch p is
given by
p = lable feed pei iolalion/ (1.1)
lolal widlh of lhe collimaled beam.
This definilion holds foi single-slice CT
as well as foi MSCT. Il shows whelhei dala
acquisilion occuis wilh gaps (p>1) oi wilh
oveilap (p<1) in lhe longiludinal diieclion.
Wilh 41-mm collimalion and a lable feed
of 6 mm/iolalion, lhe pilch is p = 6/(41)
= 6/4 = 1.5. Wilh 160.75-mm collima-
lion and a lable feed of 18 mm/iolalion,
lhe pilch is also p = 18/(160.75) = 18/12 =
1.5. In lhe eaily days of 4-slice CT, lhe leim
volume pilch had been addilionally inlio-
duced, which accounls foi lhe widlh of one
single slice in lhe denominaloi. Foi lhe sake
of claiily and unifoimily, lhe volume pilch
should no longei be used.
In CT lhe aveiage dose in lhe scan plane
is besl desciibed by lhe weighled com-
puleiized lomogiaphic dose index CTDI
w
[25,26], which is deleimined fiom CTDI
100
measuiemenls bolh in lhe cenlei and al lhe
peiipheiy of a 16-cm lucile phanlom foi
head and a 32-cm lucile phanlom foi body.
Foi lhe CTDI
100
measuiemenls a 100-mm
long ionizalion chambei is used. CTDI
w
is
defined accoiding lo lhe following equa-
lion [25]:
CTDI
w
= 1/3 CTDI
100
(cenlei) + (1.2)
2/3 CTDI
100
(peiipheiy).
CTDI
w
, given in mGy, is always measuied
in an axial scan mode. Il depends on scan-
nei geomeliy, slice collimalion and beam
piefillialion as well as on X-iay lube voll-
age, lube cuiienl (mA) and ganliy iolalion
lime (l
iol
). The pioducl of mA and l
iol
is lhe
mAs-value of lhe scan. To oblain a paiam-
elei chaiacleiislic foi lhe scannei used, il
is helpful lo eliminale lhe mAs-dependence
and lo inlioduce a noimalized (CTDI
w
)
n
given in mGy/mAs:
Fig. 1.3. Detector module of a commercially
available MSCT scanner (SOMATOM Sensation
64, Siemens, Porchheim, Germany). Lach mod-
ule consists of 40l6 detector pixels with the
corresponding electronics. The antiscatter col-
limators are diagonally cut to open the view on
the detector ceramics (yellow)
8 I 7echnicaI and Dose Considerations
CTDI
w
= mA l
iol
(CTDI
w
)
n
= (1.3)
mAs (CTDI
w
)
n
.
CTDI
w
is a measuie foi lhe dose in a single
axial scan. (CTDI
w
)
n
slill depends on X-iay
lube vollage and on slice collimalion. Scan
piolocols foi diffeienl CT-scanneis should
always be compaied on lhe basis of CTDI
w
and nevei on lhe basis of mAs, since diffei-
enl syslem geomeliies can lead lo signifi-
canl diffeiences in lhe iadialion dose lhal
is applied al idenlical mAs.
To iepiesenl lhe dose in a spiial scan, il
is essenlial lo accounl foi gaps oi oveilaps
belween lhe iadialion dose piofiles fiom
conseculive iolalions of lhe X-iay souice
[25]. Foi lhis puipose CTDI
vol
, lhe volume
CTDI
w
, has been inlioduced:
CTDI
vol
= 1/p CTDI
w
= (1.4)
mAs 1/p (CTDI
w
)
n
=
mA l
iol
1/p (CTDI
w
)
n
.
The facloi 1/p accounls foi lhe incieasing
dose accumulalion wilh decieasing spiial
pilch due lo lhe incieasing spiial oveilap.
In piinciple, Eq. 1.4 holds foi single-slice
CT as well as foi MSCT. Some manufaclui-
eis, such as Siemens, have inlioduced an
effeclive" mAs concepl foi spiial scan-
ning, which includes lhe facloi 1/p inlo lhe
mAs definilion:
(mAs)
eff
= mA l
iol
1/p = (1.5)
mAs 1/p.
Foi spiial scans, (mAs)
eff
is indicaled on
lhe usei inleiface. Inseiling Eq. 1.5 inlo
Eq. 1.4, lhe dose of a mullislice spiial scan
is simply given by
CTDI
vol
= (mAs)
eff
(CTDI
w
)
n
. (1.6)
Some olhei manufacluieis, such as Toshi-
ba and GE, slay wilh lhe convenlional mAs
definilion, and lhe usei musl peifoim lhe
1/p coiieclion himself. When compaiing
lhe scan paiameleis foi CT syslems of dif-
feienl manufacluieis, lhe undeilying mAs
definilion has lo be consideied.
CTDI
w
is a physical dose measuie; il
does nol piovide full infoimalion on lhe
iadialion iisk associaled wilh a CT ex-
aminalion. Foi lhis puipose lhe concepl
of effeclive dose" has been inlioduced by
lhe Inleinalional Commission on Radia-
lion Pioleclion (ICRP). The effeclive dose
is given in mSv. Il is a weighled sum of lhe
dose applied lo all oigans in a CT examina-
lion and includes bolh diiecl and scalleied
iadialion. The weighling faclois depend on
lhe biological iadialion sensilivilies of lhe
iespeclive oigans. The effeclive dose can be
measuied using whole body phanloms such
as lhe Aldeison phanlom, oi il is deiived
fiom compulei simulalions using Monle
Cailo lechniques lo deleimine scalleied ia-
dialion. The effeclive palienl dose depends
on lhe scanned iange. Foi a compaiison of
effeclive dose values foi diffeienl piolocols,
lhe scan ianges should be similai.
Short Review of SingIe-SIice SpiraI C7
Spiial CT iequiies an inleipolalion of lhe
acquiied measuiemenl dala in lhe longilu-
dinal diieclion lo eslimale a complele CT
dala sel al lhe desiied plane of ieconsliuc-
lion. The mosl commonly used single-slice
spiial inleipolalion schemes aie lhe 360
and 180 lineai inleipolalions (360 LI and
180 LI). In spiial CT, z-axis iesolulion
is nol only deleimined by lhe collimaled
slice-widlh s
coll
as in axial scanning, bul by
lhe effeclive slice widlh s, which is eslab-
lished in lhe spiial inleipolalion piocess.
Usually, s is defined as lhe full widlh al half
maximum (FWHM) of lhe slice sensilivily
piofile (SSP). Foi bolh 360 LI and 180 LI
s incieases wilh incieasing pilch and lon-
giludinal iesolulion degiades (Fig. 1.4).
This is a consequence of lhe incieasing lon-
giludinal dislance of lhe piojeclions used
foi spiial inleipolalion. The image noise
in single-slice spiial CT is independenl of
lhe pilch, if lhe lube cuiienl (mA) is kepl
conslanl and palienl dose decieases wilh
incieasing pilch.
In clinical piaclice, single-slice spiial
CT scanning is almosl exclusively based on
180 LI due lo lhe naiiowei SSP of lhis al-
goiilhm, despile ils incieased susceplibilily
lo ailifacls and incieased image noise. Foi
9 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
lhe same mAs, image noise is aboul 15%
highei lhan in axial mode. Wilh single-
slice CT, scanning al highei pilch is oflen
used lo ieduce palienl dose al lhe expense
of slice bioadening - if lhe collimalion is
kepl conslanl - and incieased spiial aili-
facls. Foi CTA applicalions in pailiculai il
is moie favoiable lo scan a given volume in
a given lime using naiiow collimalion al
high pilch lhan using widei collimalion al
low pilch [27].
7he Cone AngIe ProbIem in MSC7
Two-dimensional image ieconsliuclion
appioaches used in commeicially available
single-slice CT scanneis iequiie all mea-
suiemenl iays lhal conliibule lo an image
lo iun in a plane peipendiculai lo lhe pa-
lienl's longiludinal axis. In MSCT syslems
lhis iequiiemenl is violaled; lhe measuie-
menl iays aie lilled by lhe so-called cone
angle wilh iespecl lo lhe cenlei plane. The
cone angle is laigesl foi lhe slices al lhe
oulei edges of lhe delecloi and il incieases
wilh an incieasing numbei of delecloi iows
if lheii widlh is kepl conslanl. As a fiisl ap-
pioximalion, lhe cone angle is neglecled
in MSCT ieconsliuclion appioaches and
modified lwo-dimensional image iecon-
sliuclion algoiilhms aie used. The dala,
howevei, aie lhen inconsislenl, which pio-
duces cone-beam ailifacls al high-conliasl
objecls such as bones. Il has been demon-
slialed lhal cone-beam ailifacls can be
loleialed if lhe maximum numbei M of
simullaneously acquiied slices does nol
Fig. 1.4. PwHM of the spiral SSP as a function of the pitch for the two most commonly used single-
slice spiral interpolation approaches, l80 L| and 360 L|. Por both, the slice significantly broadens
with increasing pitch. As a consequence, multiplanar reformats of a spiral z-resolution phantom
scanned with 2-mm collimation (l80 L|) show increased blurring of the l.5-mm and 2-mm cylinders
with increasing pitch
10 I 7echnicaI and Dose Considerations
significanlly exceed M=4 [28]. As a con-
sequence, lhe image ieconsliuclion ap-
pioaches of all commeicially available CT
syslems wilh foui slices and of some wilh
even moie slices neglecl lhe cone angle of
lhe measuiemenl iays.
180 and 360 MuItisIice Linear
InterpoIation and FiItering
The 360 LI and 180 LI single-slice spi-
ial ieconsliuclion appioaches can be ex-
lended lo mullislice spiial scanning in a
sliaighlfoiwaid way [21,29,30]. Bolh 360
MLI and 180 MLI aie chaiacleiized by a
piojeclion-wise lineai inleipolalion be-
lween lwo iays on eilhei side of lhe image
plane. The cone angle of lhe measuiemenl
iays is nol laken inlo accounl. In geneial,
scanneis ielying on 180 MLI oi 360 MLI
lechniques and exlensions lheieof piovide
selecled disciele pilch values lo lhe usei,
such as 0.75 and 1.5 foi 4-slice scanning
[21] oi 0.5625, 0.9375, 1.375 and 1.75 foi 16-
slice scanning [30]. This is a consequence
of lhe complicaled sampling palleins
along lhe z-axis (see Fig. 1.5). The usei has
lo be awaie of pilch-dependenl effeclive
slice widlhs s. Foi low-pilch scanning (al
p=0.75 using foui slices and al p=0.5625
oi 0.9375 using 16 slices), s-s
coll
and foi a
collimaled 1.25-mm slice lhe iesulling ef-
feclive slice widlh slays al 1.25 mm. The
naiiow SSP, howevei, is achieved by a 180
MLI ieconsliuclion using conjugale inlei-
polalion al lhe piice of incieased image
noise [21,30]. Foi high-pilch scanning (al
p=1.5 using foui slices and al p=1.375 oi
1.75 using 16 slices), s-1.27s
coll
and a colli-
maled 1.25-mm slice iesulls in an effeclive
1.5-1.6-mm slice. To oblain lhe same im-
age noise as in an axial scan wilh lhe same
collimaled slice widlh, 0.73-1.68 limes lhe
dose, depending on lhe spiial pilch, is ie-
quiied, wilh lhe lowesl dose al lhe high-
esl pilch [30]. Scanneis offeiing disciele
oplimized pilch values based on 180 MLI
and 360 MLI lechniques aie compaiable
lo single-slice CT syslems in some coie
aspecls: al high pilch lhe slice widens and
longiludinal iesolulion degiades. Al low
pilch lhe naiiowesl possible SSP (compa-
iable lo 180 LI single-slice CT al pilch 1)
can be oblained, bul a highei dose is neces-
saiy lo mainlain lhe signal-lo-noise ialio.
Thus, when selecling lhe scan piolocol foi
a pailiculai applicalion, scanning al low
pilch oplimizes image qualily and longilu-
dinal iesolulion al a given collimalion, yel
al lhe expense of incieased palienl dose. To
ieduce palienl dose, eilhei lhe mA sellings
should be ieduced al low pilch oi high pilch
values should be chosen.
In a z-fillei mullislice spiial ieconsliuc-
lion [29,31] lhe spiial inleipolalion foi each
piojeclion angle is no longei iesliicled lo
lhe lwo iays in closesl pioximily lo lhe im-
Fig. 1.5. Sampling scheme along the z-axis for a 4-slice CT-system at various pitch values. The sam-
pling patterns are irregular and show a complicated dependence on the spiral pitch
11 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
age plane. Inslead, all diiecl and comple-
menlaiy iays wilhin a seleclable dislance
fiom lhe image plane conliibule lo lhe
image. Slill, lhe cone angle is neglecled.
A iepiesenlalive example foi a z-fillei ap-
pioach is lhe adaplive axial inleipolalion
(AAI) [29] implemenled in Siemens CT
scanneis. Anolhei example is lhe MUS-
COT algoiilhm [31] used by Toshiba. z-fil-
leiing allows lhe syslem lo liadeoff z-axis
iesolulion (SSP) wilh image noise (which
diieclly coiielales wilh iequiied dose).
Fiom lhe same CT iaw dala, images wilh
diffeienl slice widlhs can be ieliospeclive-
ly ieconsliucled (see Fig. 1.6). Only slice
widlhs equal lo oi laigei lhan lhe sub-
beam collimalion can be oblained. Wilh
AAI, lhe spiial pilch is fieely seleclable in
lhe iange of 0.5 lo 2 and lhe effeclive slice
widlh is kepl conslanl foi all pilch values
[6,29,32]. Theiefoie, longiludinal iesolu-
lion is independenl of lhe pilch, devialing
fiom single-slice spiial CT and fiom MSCT
ielying on 180 MLI and 360 MLI [21,30].
Figuie 1.7 shows SSPs of lhe SOMATOM
Sensalion 4 (2-mm slice foi 41-mm col-
limalion) and MPRs of a spiial z-iesolu-
lion phanlom foi selecled pilch values. As
a consequence of lhe pilch-independenl
spiial slice widlh, lhe image noise foi fixed
effeclive mAs (see Basic Paiameleis above)
is neaily independenl of lhe spiial pilch -
diffeienl fiom 180 MLI and 360 MLI. Foi
a 1.25-mm spiial slice widlh ieconsliucled
fiom 41-mm collimalion, 0.61-0.69 limes
lhe dose is iequiied lo mainlain lhe image
noise of an axial scan al lhe same colli-
malion [32]. Radialion dose, loo, is inde-
pendenl of lhe spiial pilch and equals lhe
dose of an axial scan wilh lhe same mAs.
Thus, diffeienl fiom single-slice spiial CT,
changing lhe pilch does nol change lhe pa-
lienl dose. Accoidingly, using highei pilch
does nol iesull in dose saving, which is an
impoilanl piaclical consideialion wilh CT
syslems ielying on AAI and lhe effeclive
mAs concepl.
Wilh iegaid lo image qualily, naiiow
collimalion is piefeiable lo wide collima-
lion, due lo bellei suppiession of pailial
volume ailifacls and a moie ieclangulai
SSP (see Fig. 1.6), even if lhe pilch has lo
Fig. 1.6. Using z-filtering, images with different slice widths can be retrospectively reconstructed
from the same CT raw-data. lelt: SSPs for 4l-mm collimation (SOMATOM Sensation 4, Siemens,
Porchheim, Germany): l-mm, l.25-mm, l.5-mm, 2-mm, 3-mm, 4-mm, and 5-mm slice width, not
shown are the SSPs for 6-mm, 7-mm, 8-mm, and l0-mm slice width. k|ht: SSPs of the 3-mm slice for
4l mm and 42.5-mm collimation. The PwHM is equal, but the profile is more rectangular for the
narrow collimation (4l mm)
12 I 7echnicaI and Dose Considerations
be incieased foi equivalenl volume cov-
eiage. Similai lo single-slice spiial CT,
naiiow collimalion scanning is lhe key lo
ieducing ailifacls and impioving image
qualily. Besl suppiession of spiial ailifacls
is achieved by using bolh naiiow collima-
lion ielalive lo lhe desiied slice widlh and
ieducing lhe spiial pilch. In geneial, chal-
lenging clinical piolocols, such as exami-
nalions of lhe spine and skull base, iely on
a combinalion of naiiow collimalion and
low pilch.
Some manufacluieis who use z-fillei
appioaches do nol piovide complelely fiee
seleclion of lhe spiial pilch, bul iecom-
mend a seleclion of fixed pilch values lhal
aie aimed al oplimizing lhe z-sampling
scheme and ieducing spiial ailifacls, such
as 0.625, 0.75, 0.875, 1.125, 1.25, 1.375, and
1.5 foi 4-slice scanning (MUSCOT algo-
iilhm [31]).
Cone-eam Reconstruction:
3D ackprojection and Adaptive MuItipIe
PIane Reconstruction
Foi CT scanneis wilh 16 and moie slices,
modified ieconsliuclion appioaches ac-
counling foi lhe cone-beam geomeliy of
lhe measuiemenl iays have lo be consid-
eied. Some manufacluieis (Toshiba, Phil-
ips) use a 3D-filleied back-piojeclion ie-
consliuclion [33-37]. Wilh lhis appioach,
lhe measuiemenl iays aie back-piojecled
inlo a 3D volume along lhe lines of mea-
suiemenl, lhis way accounling foi lheii
cone-beam geomeliy. Olhei manufaclui-
eis use vaiialions and exlensions of nulal-
ing slice algoiilhms [38-42], which splil
lhe 3D ieconsliuclion lask inlo a seiies of
convenlional 2D ieconsliuclions on lilled
inleimediale image planes. Repiesenlalive
examples aie lhe adaplive mulliple plane
Fig. 1.7. Adaptive Axial |nterpolation for a 4-slice CT-system (SOMATOM Sensation 4, Siemens,
Porchheim, Germany): SSP of the 2-mm slice (for 4l-mm collimation) at selected pitch values. The
functional form of the SSP, and hence the slice width, is independent of the pitch. Consequently,
multiplanar reformats of a spiral z-resolution phantom scanned with 2-mm slice width show clear
separation of the l.5-mm and 2-mm cylinders for all pitch values (compare with Pig. l.4)
13 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
ieconsliuclion (AMPR) used by Siemens
[43,44] and lhe weighled hypeiplane iecon-
sliuclion (WHR) pioposed by GE [45,46].
The AMPR appioach deliveis high
qualily images al oplimum dose usage foi
a wide iange of pilch values. As an inlei-
Fig. 1.8. Schematic diagram illustrating the adaptive multiple plane reconstruction. lelt: depend-
ing on the spiral pitch the multislice raw data are divided into overlapping segments. As an inter-
mediate step, a variety of partial images on double oblique image planes is calculated, which are
individually adapted to the spiral path and to the multislice detector geometry and fan out like the
pages of a book. k|ht: the final images are obtained by a longitudinal interpolation (z-interpolation)
between the tilted partial image planes, similar to a multiplanar reformation.
Fig. 1.9. Axial slice (to) and MPP (oottom) of a pelvis phantom scanned with a l6-slice CT-system,
l6l.5-mm collimation, 2-mm reconstructed slice width, 0.5-s rotation time, a pitch of l.0, i.e., a
table feed of 48 mm/s. lelt: conventional multislice spiral reconstruction neglecting the cone angle
of the measurement rays. Cone-beam artifacts are indicated by ottows. k|ht: AMPP. Cone-beam
artifacts are effectively reduced
14 I 7echnicaI and Dose Considerations
mediale slep, pailial images on double
oblique image planes aie calculaled, which
aie individually adapled lo lhe spiial palh
and lo lhe mullislice delecloi geomeliy
and fan oul like lhe pages of a book (see
Fig. 1.8a). The final images wilh full-dose
ulilizalion aie oblained by an appiopii-
ale longiludinal inleipolalion belween lhe
lilled pailial image planes, similaily lo a
mulliplanai iefoimalion (Fig. 1.8b). The
shape and lhe widlh of lhe inleipolalion
funclion aie fieely seleclable, diffeienl
SSPs and hence diffeienl slice widlhs foi
ieliospeclive ieconsliuclion can lheie-
foie easily be adjusled similaily lo z-fillei
appioaches. Figuie 1.9 shows images of a
pelvis phanlom scanned wilh 161.5-mm
collimalion, 0.5 s ganliy iolalion lime, and
a pilch p=1, coiiesponding lo a lable feed
of 48 mm/s, on lhe lefl side foi a iecon-
sliuclion neglecling lhe cone angle of lhe
measuiemenl iays and on lhe iighl side foi
AMPR. The convenlional appioach leads
lo seveie ailifacls and geomeliical disloi-
lions of high-conliasl objecls (lefl). Cone
ailifacls aie consideiably ieduced wilh lhe
AMPR algoiilhm and lhe spalial inlegiily
of lhe objecls is iesloied (iighl).
Mullislice spiial scanning using AMPR
in combinalion wilh lhe effeclive mAs
concepl is chaiacleiized by lhe same key
piopeilies as AAI, which can be diieclly
deiived using lhe melhods piesenled in lhe
180 and 360 Mullislice Lineai Inleipola-
lion and z-Filleiing seclion above. Thus, all
iecommendalions iegaiding lhe seleclion
of collimalion and pilch lhal have been dis-
cussed foi AAI also apply lo AMPR. In pai-
liculai, changing lhe pilch does nol change
lhe iadialion exposuie lo lhe palienl, and
using highei pilch does nol iesull in dose
saving. Naiiow collimalion scanning
should be peifoimed whenevei possible.
Wilh 160.75-mm collimalion, 1-mm slice
widlh is iecommended as lhe mosl suilable
liadeoff belween longiludinal iesolulion,
image noise and ailifacls when lhin slices
aie ieconsliucled as an inpul foi 3D posl-
piocessing, such as MPR, MIP oi VRT.
Fig. 1.10. Schematic drawing of a rotating en-
velope X-ray tube (Siemens STPATON, Porch-
heim, Germany) with z-flying focal spot technol-
ogy. The entire tube housing rotates in an oil
bath. The anode plate is in direct contact with
the cooling oil to improve heat dissipation. The
central cathode rotates as well, and continu-
ous electromagnetic deflection of the electron
beam is needed to control the position and the
shape of the focal spot. The electromagnetic
deflection unit is used to wobble the focal spot
between two different positions on the anode
plate (indicated by two asterisks). Due to the
anode angle of typically 7-9 this translates into
a motion both in the radial direction and in the
z-direction
Fig. 1.11. Principle of improvedz-sampling with
the z-flying focal spot technique. Due to a peri-
odic motion of the focal spot in the z-direction
two subsequent V slice readings are shifted by
half a collimated slice-width s
coll
/2 at isocenter
and can be interleaved to one 2V slice pro|ec-
tion. |mproved z-sampling is not only achieved
at the isocenter, but maintained in a wide range
of the SPOv. The simultaneous radial motion of
the focal spot in an actual X-ray tube has been
omitted to simplify the drawing
15 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
DoubIe z-SampIing
The double z-sampling concepl foi mul-
lislice spiial scanning makes use of peii-
odic molion of lhe focal spol in lhe longilu-
dinal diieclion lo impiove dala sampling
along lhe z-axis [20,22]. By conlinuous
elecliomagnelic defleclion of lhe eleclion
beam in a iolaling envelope X-iay lube, lhe
focal spol is wobbled belween lwo diffei-
enl posilions on lhe anode plale. Due lo lhe
anode angle of lypically 7-9 lhis lians-
lales inlo a molion bolh in lhe iadial diiec-
lion and in lhe z-diieclion (Fig. 1.10). The
iadial molion is a side-effecl lhal musl be
laken caie of by lhe image ieconsliuclion
algoiilhms. The amplilude of lhe peiiod-
ic z-molion is adjusled in a way lhal lwo
subsequenl ieadings aie shifled by half a
collimaled slice widlh in lhe palienl's lon-
giludinal diieclion (Fig. 1.11). Theiefoie,
lhe measuiemenl iays of lwo subsequenl
ieadings wilh collimaled slice-widlh s
coll
inleileave in lhe z-diieclion, and eveiy lwo
M-slice ieadings aie combined inlo one
2M-slice piojeclion wilh a sampling dis-
lance of s
coll
/2.
The SOMATOM Sensalion 64 (Siemens,
Foichheim, Geimany) as an example of a
MSCT syslem ielying on double z-sampling
has a delecloi lhal piovides 32 collimaled
0.6-mm slices (see Secl. 1.2.1). Two subse-
quenl 32-slice ieadings aie combined inlo
one 64-slice piojeclion wilh a sampling dis-
lance of 0.3 mm al lhe isocenlei. Thus, 64
oveilapping 0.6-mm slices pei iolalion aie
acquiied. The sampling scheme is similai
lo lhal of a 640.3-mm delecloi, and lhe
AMPR algoiilhm is used foi image iecon-
sliuclion. Double z-sampling piovides a
sampling dislance of s
coll
/2 independenl of
lhe pilch. The impioved sampling along lhe
z-diieclion is nol iesliicled lo lhe isocen-
lei, bul is mainlained in a wide iange of lhe
SFOV. As a consequence, spalial iesolulion
in lhe logiludinal diieclion is incieased,
and objecls <0.4 mm in diamelei can be
ioulinely iesolved al any pilch (Fig. 1.12).
Anolhei benefil of double z-sampling is lhe
suppiession of spiial windmill" ailifacls
al any pilch (Fig. 1.13). In conliasl lo con-
venlional MSCT scan and ieconsliuclion
appioaches, spiial image qualily is laigely
independenl of lhe pilch. Using conven-
Fig. 1.12. MPPs of a z-resolution phantom (high-resolution insert of the CATPHAN, the Phantom
Laboratories, Salem, N, turned by 90) for pitch 0.55, 0.95 and l.45, scanned with a MSCT system
with double z-sampling (SOMATOM Sensation 64, Siemens, Porchheim, Germany). Scan data has
been acquired with 320.6-mm collimation in a 64-slice acquisition mode using the z-flying focal
spot and reconstructed with the narrowest slice width (nominal 0.6 mm) and a sharp body kernel
(870). a z-resolution phantom positioned at the isocenter. |ndependent of the pitch all bar-patterns
up to l5 lp/cm can be visualized, corresponding to 0.33 mm longitudinal resolution. b z-resolution
phantom positioned l00 mm off-center (pitch 0.95). Longitudinal resolution is only slightly degrad-
ed (l4 lp/cm, corresponding to 0.36 mm)
16 I 7echnicaI and Dose Considerations
Fig. 1.13. Peduction of spiral artifacts with double z-sampling. lelt: head scan acquired with 320.6
collimation in a 64-slice acquisition mode with z-flying focal spot at pitch l.5. k|ht: same scan, us-
ing only one focus position of the z-flying focal spot for image reconstruction. Due to the improved
longitudinal sampling with z-flying focal spot (lelt) spiral interpolation artifacts (windmill structures
at high contrast ob|ects) are suppressed without degradation of z-axis resolution
Fig. 1.14. Case study illustrating interactive viewing of volume data with a 4-slice CT system (4l-
mm collimation, pitch 2, reconstructed slice-width l.25 mm). 7o: axial images show the central
thrombosis of the right subclavian vein. 8ottom: M|P images show different views of the filiform ste-
nosis of the right brachiocephalic vein proximal to the confluence of the superior cava vein (courtesy
of M. Lell, University of Lrlangen, Germany)
17 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
lional MSCT syslems, demanding applica-
lions, such as neuioscanning, iequiie low
pilch piolocols lo ieduce ailifacls and lo
impiove image qualily. The z-flying focal
spol lechnique mainlains a low ailifacl
level up lo pilch 1.5, lhus incieasing lhe
maximum volume coveiage speed lhal can
be used clinically.
1.3 AppIications
1.3.1 Standard AppIications
Clinical applicalions benefil fiom MSCT
lechnology in seveial ways [6,9,16,32]:
E Shoilei scan lime (impoilanl foi liauma
viclims, palienls wilh limiled abilily lo
coopeiale, pedialiic cases, mulliphase
exams and CT angiogiaphy (CTA))
E Exlended scan iange (impoilanl foi
CTA oi combined chesl-abdomen-pel-
vis scans such as in oncological slaging)
E Impioved longiludinal iesolulion (bene-
ficial foi all ieconsliuclions, in pailicu-
lai when 3D poslpiocessing is pail of lhe
clinical piolocol)
Mosl piolocols can benefil fiom a com-
binalion of lhese advanlages. The wide
availabilily of 4-slice CT syslems slailed lo
change lhe way iadiologisls lhink aboul CT
imaging and paved lhe way foi CT fiom a
cioss-seclional slice modalily lo a volume
imaging modalily. In many applicalions,
naiiow collimalion dala is iecommended
independenlly fiom lhe desiied slice widlh
foi piimaiy viewing. In piaclice, lwo dif-
feienl slice widlhs aie commonly iecon-
sliucled by defaull: lhick slices foi filming
and lhin slices foi 3D poslpiocessing and
evalualion (please also iefei lo lhe indi-
vidual piolocols). The image noise in close-
lo-isoliopic high iesolulion volumes can
be limiled by making use of ihick MPRs
oi ihick MIPs. In lhis way, images wilh lhe
desiied slice-widlh can be oblained in aibi-
liaiy diieclions. As a consequence, lhe dis-
linclion belween longiludinal and in-plane
iesolulion is giadually becoming a hisloii-
cal iemnanl, and lhe liadilional axial slice
is loosing ils clinical piedominance. Il is ie-
placed by inleiaclive viewing and manipu-
lalion of volume images, wilh only lhe key
slices oi views in aibiliaiy diieclions used
foi filming oi sloied foi a demonslialion
of lhe diagnosis. Figuie 1.14 shows an ex-
ample of a lhoiax sludy on a 4-slice CT-sys-
lem lhal has been diagnosed inleiaclively.
Please nole lhal MPRs and oblique MIPs
aie of similai iesolulion as lhe oiiginal
axial images.
Meanwhile, 16-slice CT syslems have
been eslablished in clinical piaclice, and
isoliopic submillimelei coveiage of ex-
lended analomical ianges has become lhe
clinical slandaid foi many applicalions.
Clinical piaclice suggesls lhe polenlial of
16-slice CTA lo ieplace noninleivenlional
calhelei angiogiaphy in lhe evalualion of
caiolid aileiy slenosis [47]. Foi palienls
wilh suspicion of ischemic slioke bolh lhe
slalus of lhe vessels supplying lhe biain
and lhe localion of lhe inliacianial occlu-
sion can be assessed in lhe same examina-
lion [48]. Addilional biain peifusion CT
peimils diffeienlialion of iiieveisibly dam-
aged biain lissue fiom ieveisibly impaiied
lissue al iisk [49]. Examining lhe enliie
lhoiax (350 mm) wilh submillimelei col-
limalion iequiies a scan lime of appioxi-
malely 11 s. Due lo lhe shoil biealh hold
lime, cenlial and peiipheial pulmonaiy
embolism can be ieliably and accuialely
diagnosed [50,51]. Sixleen-slice CT en-
ables whole body angiogiaphic sludies wilh
submillimelei iesolulion in a single biealh
hold. Compaied lo invasive angiogiaphy,
lhe same moiphological infoimalion is ie-
vealed [52,53] (see Fig. 1.15 foi an example
of a palienl wilh Leiiche-syndiome).
Insufficienl volume coveiage speed,
which was a seiious pioblem in lhe days
of single-slice CT [5], only iaiely becomes
a limiling facloi wilh 16-slice CT. Clinical
piogiess fiom fuilhei lechnical develop-
menl of MSCT can moie likely be expecled
fiom impiovemenls in spalial and lempoial
iesolulion ialhei lhan fiom a meie inciease
in lhe volume-coveiage speed. The new
concepl of double z-sampling implemenled
in some of lhe iecenlly inlioduced 64-slice
CT-syslems allows foi a pilch-independenl
18 I 7echnicaI and Dose Considerations
inciease of longiludinal iesolulion and
suppiession of spiial ailifacls. Isoliopic
submillimelei CTAs of lhe caiolid aileiies
and lhe ciicle of Willis wilh a scan iange of
350 mm iequiie 5-6 s scan lime wilh lhese
syslems, hence lhey can be acquiied in lhe
puiely aileiial phase (see lhe example in
Fig. 1.16). The enliie lhoiax (350 mm) can
be scanned in aboul 6 s, facililaling lhe ex-
aminalion of emeigency palienls, e.g., wilh
suspicion of acule pulmonaiy embolism.
Eaily expeiience in neuio-CT has alieady
demonslialed a subslanlial impiovemenl
in spalial iesolulion and ailifacl level made
possible by lhe double z-sampling concepl
foi MSCT syslems [54].
1.3.2 SpeciaI AppIications
Cardiac C7
One of lhe mosl exciling new applicalions
of mullislice CT is lhe abilily lo image lhe
heail. Incieased iolalion speed combined
wilh dedicaled ECG-synchionized iecon-
sliuclion algoiilhms effeclively allow one
lo fieeze lhe heail molion [13,55,56]. The
delails of lhis new lechnique have been dis-
cussed exlensively in seveial iecenl publi-
calions [57-59] so we iesliicl lhis seclion lo
a biief oveiview.
One impoilanl applicalion of caidiac CT
is lhe quanlilalive evalualion of coionaiy
calcificalion as a iisk indicaloi in non-
symplomalic palienls, which pieviously
was a domain of eleclion beam CT (EBT)
lechnology. Sludies have shown lhal ECG-
galed spiial imaging wilh ieconsliuclion of
oveilapping images can significanlly ieduce
inlei-scan vaiiabilily [60]. Good iepeal-
abilily of lhe quanlilalive measuiemenls
is a pieiequisile foi longiludinal sludies,
such as conliolling lhe same individual
foi effecliveness of medicalion. Advanced
soflwaie plalfoims allow assessmenl of lhe
eslablished Agalslon scoie as well as olhei
scoie values, such as equivalenl lesion vol-
ume and lolal calcified plaque buiden in
leims of absolule calcium mass, based on
scannei-specific calibialion [60, 61]. The
mass scoie in pailiculai has lhe polenlial
Fig. 1.15. Case study illustrating the clinical performance of l6-slice CT: patient with occlusion of
the left common iliac artery. Scan parameters: l60.75-mm collimation, reconstructed slice-width
l mm, 0.5 s gantry rotation time (courtesy of M. Heuschmid and A. Kuttner, University of Tubingen,
Germany)
19 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
lo inciease lhe accuiacy, consislency, and
iepioducibilily of coionaiy calcium assess-
menl [61].
Wilh foui simullaneously acquiied slic-
es, coveiage of lhe enliie heail wilh lhin
slices (1 mm/1.25 mm) and ECG-galing
wilhin a single biealh hold became fea-
sible, enabling noninvasive visualizalion
of lhe coionaiy aileiies [15,56,62,63]. Due
lo lhe impioved signal-lo-noise ialio when
compaied lo EBT and lhe bellei longilu-
dinal iesolulion, lhe inilial clinical slud-
ies demonslialed MSCT's polenlial lo nol
only delecl bul lo some degiee chaiaclei-
ize noncalcified plaques in lhe coionaiy
aileiies based on lheii CT allenualion
[64,65]. Despile all piomising advances,
challenges and limilalions wilh iespecl lo
molion ailifacls in palienls wilh highei
heail iales, limiled spalial iesolulion and
long biealh hold limes iemain foi 4-slice
caidiac CT.
Sixleen-slice MSCT syslems wilh gan-
liy iolalion limes down lo 0.375 s have
impioved spalial and lempoial iesolulion,
while examinalion limes have been con-
sideiably ieduced: lhe enliie heail volume
can be coveied wilh submillimelei slices
in 15-20 s [17,44]. Sixleen-slice syslems
have been used lo eslablish ECG-liiggeied
oi ECG-galed MSCT examinalions of lhe
heail and lhe coionaiy aileiies in clinical
piaclice. Deleclion and chaiacleiizalion
of coionaiy plaques, even in lhe piesence
of seveie calcificalions benefil fiom lhe
incieased iobuslness of lhe lechnology. A
sludy of coionaiy CTA wilh a 16-slice sys-
lem in 59 palienls demonslialed 86% speci-
ficily and 95% sensilivily foi idenlifying
significanl coionaiy aileiy slenosis. None
of lhe palienls had lo be excluded [18], as
in pievious sludies based on less-advanced
scannei lechnology. Meanwhile, olhei in-
vesligalois iepoil similai iesulls [19].
The lalesl geneialion of 64-slice CT sys-
lems piovides fuilhei incieased spalial
iesolulion (0.4-mm isoliopic voxels) and
impioved lempoial iesolulion due lo gan-
liy iolalion limes down lo 0.33 s, and will
be a fuilhei leap in inlegialing coionaiy
CTA inlo iouline clinical algoiilhms. Due
lo lhe fasl volume coveiage of lhese sys-
lems, ECG-galed examinalions of lhe enliie
lhoiax foi a compiehensive examinalion of
palienls wilh acule lhoiax pain aie feasible,
opening new peispeclives foi emeigency
caie.
Figuie 1.17 shows an example of a coio-
naiy CTA in a follow-up examinalion and
demonsliales lhe impiovemenl in image
qualily fiom 4 lo 64 slices.
Lung Cancer Screening
and C7 CoIonography
Eaily deleclion of lung cancei and CT colo-
nogiaphy aie piomising applicalions in lhe
field of pievenlive caie.
Inilial sludies using CT lo delecl lung
cancei al an eaily slage in a scieening pop-
ulalion weie published in 1999 [66]. Basic
syslem iequiiemenls aie coveiage of lhe en-
liie lung in a single biealh hold al sufficienl
Fig. 1.16. Case study illustrating the clinical per-
formance of 64-slice CT with z-flying focal spot:
CTA of the carotid arteries and the circle of willis.
Scan parameters: l20 kv, l50 eff. mAs, 640.6-
mm slice acquisition, reconstructed slice-width
0.6 mm, 0.375-s gantry rotation time, pitch l.4,
scan time 6 s for 350 mm. The ottow indicates
a severe stenosis (courtesy of M. Lell and K. An-
ders, University of Lrlangen, Germany)
20 I 7echnicaI and Dose Considerations
iesolulion lo delecl small, suspicious nod-
ules while keeping lhe dose al an acceplable
level. In piinciple, single-slice CT is able lo
meel lhese iequiiemenls, bul clinical woik-
flow impioves subslanlially wilh MSCT. In
lhe ELCAP sludy [66], which was peifoimed
using single-slice CT, suspicious nodules
weie found in aboul 25% of lhe scieened
populalion. These palienls ieceived fuilhei
woik-up, eilhei moniloiing and follow-up
of small nodules oi suigical iemoval, de-
pending on vaiious faclois. Since accuiale
volume assessmenl of small nodules ie-
quiies high spalial iesolulion, which was
nol available in lhe inilial scieening exami-
nalion due lo limiled volume coveiage, lhe
iespeclive palienls had lo be iescheduled.
Wilh MSCT lhe scieening examinalion
can be done al high iesolulion, obvialing
lhe need foi iescheduling. Wilh 41-mm
collimalion and 0.5 s ganliy iolalion lime,
lhe enliie lhoiax (350 mm) can be coveied
in 25-30 s. Wilh 160.75-mm collimalion,
lhe scan lime is ieduced lo aboul 10 s. The
use of CT as a scieening lool in a heallhy
populalion al iisk foi lung cancei iequiies
ieduclion of lhe palienl dose lo lhe lowesl
clinically acceplable level. Wilh 120 kV and
20 effeclive mAs, lhe effeclive palienl dose
is below 1 mSv, which is less lhan lhe nalu-
ial backgiound iadialion exposuie of half
a yeai.
The ELCAP sludy has shown lhal CT
scieening can delecl lung cancei signifi-
canlly eailiei lhan chesl X-iays [66]. While
il has nol been pioven lhal CT scieening
foi lung cancei ieduces moilalily, eailiei
deleclion of lung cancei is commonly con-
sideied lo have a favoiable effecl on lhe ef-
fecliveness of lheiapy.
Recenlly, new soflwaie lools have been
developed lhal aulomalically delecl suspi-
cious nodules and can be used as a second
ieadei" lo suppoil lhe iadiologisl (nodule
enhanced visualizalion NEV, Siemens Med-
ical Solulions, Foichheim, Geimany). Ini-
lial sludies have demonslialed highei de-
leclion sensilivily of nodules smallei lhan
8 mm in diamelei wilh lhe help of NEV.
CT colonogiaphy is used lo delecl sus-
picious polyps in lhe colon. Allhough con-
lioveisially discussed, lhe melhod has been
Fig. 1.17. Case study comparing LCG-gated cardiac CT with a 4-slice CT system and a 64-slice CT
system with z-flying focal spot. The follow-up CTA examination demonstrates the improvement in
spatial resolution with 64 slices (courtesy of C. 8ecker, Klinikum Grohadern, Ludwigs-Maximilians
University Munich, Germany)
21 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
shown lo compaie well wilh convenlional
colonoscopy in delecling polyps wilh a di-
amelei of al leasl 10 mm in a iecenl sludy
[67]. Sixly-eighl asymplomalic men al av-
eiage iisk undeiwenl CT colonogiaphy fol-
lowed by oplical colonoscopy on lhe same
day, and a pei-palienl specificily of 89.7%
could be oblained foi CT colonogiaphy
[67]. A similai sludy on 1233 asymplom-
alic adulls iesulled in sensilivilies of 93.8,
93.9, and 88.7% foi lhe deleclion of adeno-
malous polyps al leasl 10, 8, and 6 mm in
diamelei, iespeclively. The specificily was
96.0, 92.2, and 79.6% foi lhe lhiee sizes
of polyps, iespeclively. Two polyps in lhe
scieening populalion weie malignanl. Bolh
weie delecled on viilual colonogiaphy, bul
one of lhem was missed on oplical colons-
copy [68].
Naiiow-collimalion scanning (41 mm/
41.25 mm, 160.625 mm/160.75 mm)
is lhe melhod of choice foi CT colonogia-
phy [69]. While no diffeience in sensilivily
could be obseived wilh lhick-seclion iecon-
sliuclions, specificily maikedly impioved
wilh lhin seclions [70]. Again, lhe iadialion
dose should be kepl as low as possible, e.g.,
by using 120 kV and 120 effeclive mAs oi 40
effeclive mAs foi lhe supine oi pione scans,
iespeclively. Similaily lo lhe eaily deleclion
of lung cancei, new soflwaie lools aie being
developed lhal aulomalically delecl suspi-
cious polyps and can be used as a second
ieadei lo suppoil lhe iadiologisl (polyp en-
hanced viewing PEV, Siemens, Foichheim,
Geimany).
References
1. Kalendei W, Seisslei W, Klolz E, Vock P (1990)
Spiial volumeliic CT wilh single-biealh-hold
lechnique, conlinuous lianspoil and conlinu-
ous scannei iolalion. Radiology 176:181-183
2. Ciawfoid CR, King KF (1990) Compuled lo-
mogiaphy scanning wilh simullaneous palienl
lianslalion. Med Phys 17:967-982
3. Rubin GD, Dake MD, Semba CP (1995) Cuiienl
slalus of lhiee-dimensional spiial CT scanning
foi imaging lhe vasculaluie. Radiol Clin Noilh
Am 33(1):51-70
4. Napel S, Rubin GD, Jeffiey RB (1993) STS-MIP:
a new ieconsliuclion lechnique foi CT of lhe
chesl. JCAT 17(5):832-8
5. Kalendei W (1995) Thin-seclion lhiee-dimen-
sional spiial CT: is isoliopic imaging possible?
Radiology 197:578-580
6. Klingenbeck-Regn K, Schallei S, Flohi T, Oh-
nesoige B, Kopp AF, Baum U (1999) Subsecond
mullislice compuled lomogiaphy: basics and
applicalions. EJR 31:110-124
7. McCollough CH, Zink FE (1999) Peifoimance
evalualion of a mullislice CT syslem. Med Phys
26:2223-2230
8. Ohnesoige B, Flohi T, Schallei S, Klingenbeck-
Regn K, Beckei C, Schpf U J, Bining R, Reisei
MF (1999) Technische Giundlagen und Anwen-
dungen dei Mehischichl-CT. Radiologe 39:923-
931
9. Hu H, He HD, Foley WD, Fox SH (2000) Foui
mullidelecloi-iow helical CT: image qualily and
volume coveiage speed. Radiology 215:55-62
10. Rubin GD, Schmidl AJ, Logan LJ, Sofilos MC
(2001) Mulli-delecloi iow CT angiogiaphy of
lowei exliemily aileiial inflow and iunoff: ini-
lial expeiience. Radiology 221(1):146-58
11. Schpf UJ, Biuening RD, Hong C, Eibel R, Ay-
demii S, Ciispin A, Beckei C, Reisei MF (2001)
Mullislice helical CT of focal and diffuse lung
disease: compiehensive diagnosis wilh iecon-
sliuclion of conliguous and high-iesolulion CT
seclions fiom a single lhin-collimalion scan.
Am J Roenlgenol 177(1):179-84
12. Kacheliie M, Ulzheimei S, Kalendei W (2000)
ECG-coiielaled image ieconsliuclion fiom
subsecond mulli-slice spiial CT scans of lhe
heail. Med Phys 27:1881-1902
13. Ohnesoige B, Flohi T, Beckei C, Knez A, Kopp
A, Fukuda K, Reisei M (2000) Heizbildgebung
mil schnellei, ieliospekliv EKG-synchionisi-
eilei Mehischichlspiial CT. Radiologe 40:111-
117
14. Hong C, Beckei CR, Hubei A, Schoepf UJ,
Ohnesoige B, Knez A, Bining R, Reisei MF
(2001) ECG-galed ieconsliucled mulli-delecloi
iow CT coionaiy angiogiaphy: effecl of vaiy-
ing liiggei delay on image qualily. Radiology
220:712-717
15. Nieman K, Oudkeik M, Rensing B, van Oijen P,
Munne A, van Geuns R, de Feylei P (2001) Coi-
onaiy angiogiaphy wilh mulli-slice compuled
lomogiaphy. Lancel 357:599-603
16. Flohi T, Slieisloifei K, Biudei H, Simon J,
Schallei S (2002) New lechnical developmenls
in mullislice CT, pail 1: appioaching isolio-
pic iesolulion wilh sub-mm 16-slice scanning.
Rfo Foilschi Geb Ronlgensli Neuen Bildgeb
Veifahi 174:839-845
17. Flohi T, Biudei H, Slieisloifei K, Simon J,
Schallei S, Ohnesoige B (2002) New lechnical
developmenls in mullislice CT, pail 2: sub-mil-
limelei 16-slice scanning and incieased gan-
liy iolalion speed foi caidiac imaging. Rfo
Foilschi Geb Ronlgensli Neuen Bildgeb Vei-
fahi 174:1022-1027
22 I 7echnicaI and Dose Considerations
18. Nieman K, Cademailiii F, Lemos PA, Raaijmak-
eis R, Pallynama PMT, de Feylei PJ (2002) Re-
liable noninvasive coionaiy angiogiaphy wilh
fasl submillimelei mullislice spiial compuled
lomogiaphy. Ciiculalion 106:2051-2054
19. Ropeis D, Baum U, Pohle K el al. (2003) Delec-
lion of coionaiy aileiy slenoses wilh lhin-slice
mulli-delecloi iow spiial compuled lomogia-
phy and mulliplanai ieconsliuclion. Ciicula-
lion 107:664-666
20. Flohi T, Biudei H, Slieisloifei K, Schallei S
(2003) Evalualion of appioaches lo ieduce spi-
ial ailifacls in mulli-slice spiial CT. Absliacl
Book of lhe 89lh Scienlific Assembly and An-
nual Meeling of lhe RSNA 2003, p 567
21. Hu H (1999) Mulli-slice helical CT: scan and ie-
consliuclion. Med Phys 26(1):5-18
22. Flohi T, Slieisloifei K, Raupach R, Ulzheimei
S, Biudei H (2004) Peifoimance evalualion of
a 64-slice CT-syslem wilh z-flying focal spol.
Rfo Foilschi Geb Ronlgensli Neuen Bildgeb
Veifahi 176:1803-1810
23. Hsieh J (2001) Invesligalion of lhe slice sensiliv-
ily piofile foi slep-and-shool mode mulli-slice
compuled lomogiaphy. Med Phys 28:491-500
24. Inleinalional Elecliolechnical Commission
(2002) 60601-2-44. Amendmenl 1: Medical
elecliical equipmenl, Pail 2-44: Pailiculai ie-
quiiemenls foi lhe safely of X-iay equipmenl
foi compuled lomogiaphy. Inleinalional Elec-
liolechnical Commision, Geneva
25. Moiin R, Geibei T, McCollough C (2003) Radia-
lion dose in compuled lomogiaphy of lhe heail.
Ciiculalion 107:917-922
26. McCollough C (2003) Palienl dose in caidiac
compuled lomogiaphy. Heiz 28:1-6
27. Rubin GD, Napel S (1995) Incieased scan pilch
foi vasculai and lhoiacic spiial CT. Radiology
197:316-317
28. Sailo Y, Suzuki T (1998) Evalualion of lhe pei-
foimance of mulli-slice CT syslem in non-he-
lical scanning. Absliacl Book of lhe 84lh Sci-
enlific Assembly and Annual Meeling of lhe
RSNA 1998, p 578.
29. Schallei S, Flohi T, Klingenbeck K, Kiause J,
Fuchs T, Kalendei WA (2000) Spiial inleipola-
lion algoiilhm foi mulli-slice spiial CT - pail I:
lheoiy. IEEE Tians Med Imag 19(9):822-834
30. Hsieh J (2003) Analylical models foi mulli-slice
helical CT peifoimance paiameleis. Med Phys
30(2):169-178
31. Taguchi T, Aiadale H (1998) Algoiilhm foi im-
age ieconsliuclion in mulli-slice helical CT.
Med Phys 25(4):550-561
32. Fuchs T, Kiause J, Schallei S, Flohi T, Kalendei
WA (2000) Spiial inleipolalion algoiilhms foi
mullislice spiial CT - pail 2: measuiemenl and
evalualion of slice sensilivily piofiles and noise
al a clinical mullislice syslem. IEEE Tians Med
Imag 19(9):835-847
33. Feldkamp LA, Davis LC, Kiess JW (1984) Piac-
lical cone-beam algoiilhm. J Opl Soc Am A
1:612-619
34. Wang G, Lin T, Cheng P (1993) A geneial cone-
beam ieconsliuclion algoiilhm. IEEE Tians
Med Imag 12:486-496
35. Schallei S (1998) Piaclical image ieconsliuclion
foi cone-beam compuled lomogiaphy. PhD lhe-
sis, Univeisily Eilangen, Ninbeig, Geimany
36. Giass M, Khlei T, Pioksa R (2000) 3D cone-
beam CT ieconsliuclion foi ciiculai liajeclo-
iies. Phys Med Biol 45(2):329-347
37. Hein I, Taguchi K, Silvei MD, Kazaina M, Moii I
(2003) Feldkamp-based cone-beam ieconsliuc-
lion foi ganliy-lilled helical mullislice CT. Med
Phys 30(12):3233-3242
38. Laison G, Rulh C, Ciawfoid C (1998) Nulaling
slice CT image ieconsliuclion appaialus and
melhod. US Palenl 5802134
39. Tuibell H, Danielsson PE (1999) An impioved
PI-melhod foi ieconsliuclion fiom helical cone
beam piojeclions. IEEE Medical Imaging Con-
feience, Seallle
40. Pioksa R, Koehlei T, Giass M, Timmei J (2000)
The n-PI melhod foi helical cone-beam CT.
IEEE Tians Med Imag 19:848-863
41. Kacheliie M, Schallei S, Kalendei WA (2000)
Advanced single-slice iebinning in cone-beam
spiial CT. Med Phys 27(4):754-772
42. Biudei H, Kacheliie M, Schallei S, Slieisloifei
K, Flohi T (2000) Single-slice iebinning iecon-
sliuclion in spiial cone-beam compuled lomog-
iaphy. IEEE Tians Med Imag 19(9):873-887
43. Schallei S, Slieisloifei K, Biudei H, Kacheliie
M, Flohi T (2001) Novel appioximale appioach
foi high-qualily image ieconsliuclion in helical
cone beam CT al aibiliaiy pilch. Pioc SPIE Inl
Symp Med Imag 4322:113-127
44. Flohi T, Slieisloifei K, Biudei H, Simon J, Po-
lacin A, Schallei S (2003) Image ieconsliuclion
and image qualily evalualion foi a 16-slice CT
scannei Med Phys 30(5):832-845
45. Hsieh J, Tolh TL, Simoni P, Giekowicz B, Slack
CC, Seidenschnui GE (2001) A geneialized heli-
cal ieconsliuclion algoiilhm foi mulli-slice CT.
Absliacl Book of lhe 87lh Scienlific Assembly
and Annual Meeling of lhe RSNA 2001, p 271
46. Hsieh J, Giekowicz B, Simoni P, Thibaull JB,
Joshi MC, Dulla S, Williams EC, Shaughnessy
C, Sainalh P (2003) Convolulion ieconsliuclion
algoiilhm foi mulli-slice helical CT. In: Pioc.
SPIE Inl. Symp. Med. Imag. 2003
47. Lell M, Wildbeigei J, Heuschmid M, Flohi T,
Slieisloifei K, Fellnei F, Lang W, Baulz W,
Baum U (2002) CT-Angiogiaphie dei A. caiolis:
Eisle Eifahiungen mil einem 16-Schichl-Spi-
ial-CT - CT-angiogiaphy of lhe caiolid aileiy:
Fiisl iesulls wilh a novel 16-slice-spiial-CT
scannei. Rfo Foilschi Geb Ronlgensli Neuen
Bildgeb Veifahi 174:1165-1169
23 1 7echnicaI PrincipIes and AppIications of MuItisIice SpiraI C7
48. Eill-Wagnei B, Hoffmann RT, Bining R, Dich-
gans M, Reisei MF (2002) Supiaaoilale Gef-
diagnoslik mil dem 16-Zeilen-Mullidelekloi-
Spiial-CT. Unleisuchungs-piolokoll und eisle
Eifahiungen. Radiologe 42:728-732
49. Tomandl BF, Klolz E, Handschu R, Slempei
B, Reinhaidl F, Huk WJ, Ebeihaidl KE, Faleh-
Moghadem S (2003) Compiehensive imaging of
ischemic slioke wilh mulliseclion CT. Radio-
Giaphics 23(3):565-592
50. Remy-Jaidin J, Tillie-Leblond I, Szapiio D el al.
(2002) CT angiogiaphy of pulmonaiy embolism
in palienls wilh undeilying iespiialoiy disease:
impacl of mullislice CT on image qualily and
negalive piediclive value. Eui Radiol 12:1971-
1978
51. Schoepf UJ, Beckei CR, Hofmann LK, Das M,
Flohi T, Ohnesoige BM el al. (2003) Mullislice
CT angiogiaphy. Eui Radiol 13:1946-1961
52. Winleispeigei B, Helmbeigei T, Heizog P, Ja-
kobs T, Waggeishausei T, Beckei C, Reisei M
(2002) Hochaufgelsle abdominelle Ubeisichls-
angiogiaphie mil einem 16-Delekloizeilen-CT-
Syslem - New abdominal CT angiogiaphy pio-
locol on a 16 delecloi-iow CT scannei. Radio-
loge 42:722-727
53. Winleispeigei B, Heizog P, Jakobs T, Reisei
M, Beckei C (2002) Inilial expeiience wilh lhe
clinical use of a 16 delecloi iow CT syslem. Ciil
Rev Compul Tomogi 43:283-316
54. McCollough CH, Lindell EP, Piimak AN, Flelch-
ei JG, Slieisloifei K, Flohi TG (2004) Eaily ex-
peiience wilh 64-slice CT and z-axis oveisam-
pling: novel applicalions and lhe eliminalion of
helical ailifacls in neuio CT. On-line Absliacl
Book of lhe 90lh Scienlific Assembly and An-
nual Meeling of lhe RSNA
55. Ohnesoige B, Flohi T, Beckei C, Kopp A,
Schoepf U, Baum U, Knez A, Klingenbeck-Regn
K, Reisei M (2000) Caidiac imaging by means
of eleclio-caidiogiaphically galed mullisec-
lion spiial CT - inilial expeiience. Radiology
217:564-571
56. Beckei C, Knez A, Ohnesoige B, Schpf U, Rei-
sei M (2000) Imaging of non calcified coionaiy
plaques using helical CT wilh ieliospeclive
EKG galing. AJR 175:423-424
57. Flohi T, Schoepf U, Kuellnei A el al. (2003) Ad-
vances in caidiac imaging wilh 16-seclion CT
syslems. Acad Radiol 10:386-401
58. Schpf UJ, Beckei CR, Ohnesoige BM, Yucel EK
(2004) CT of coionaiy aileiy disease. Radiology
232:18-37
59. Schnhagen P, Hallibuilon SS, Slillman AE,
Kuzmiak SA, Nissen S E, Tuzcu EM, While RD
(2004) Noninvasive imaging of coionaiy ailei-
ies: cuiienl and fuluie iole of mulli-delecloi
iow CT. Radiology 232:7-17
60. Ohnesoige B, Flohi T, Fischbach R, el al. (2002)
Repioducibilily of coionaiy calcium quanlifi-
calion in iepeal examinalions wilh ieliospec-
lively ECG-galed mulli- seclion spiial CT. Eui
Radiol 12:1532-1540
61. Ulzheimei S, Kalendei W (2003) Assessmenl of
calcium scoiing peifoimance in caidiac com-
puled lomogiaphy. Eui Radiol 13:484-497
62. Achenbach S, Ulzheimei S, Baum U el al. (2000)
Noninvasive coionaiy angiogiaphy by ielio-
speclively ECG-galed mulli-slice spiial CT. Cii-
culalion 102:2823-2828
63. Knez A, Beckei C, Lebei A, Ohnesoige B, Reisei
M, Habeil R (2000) Non-invasive assessmenl
of coionaiy aileiy slenoses wilh mullidelecloi
helical compuled lomogiaphy. Ciiculalion 101:
e221-e222
64. Schioedei S, Kopp A, Baumbach A, Meisnei C,
Kuellnei A, Geoig C, Ohnesoige B, Heideg C,
Claussen C, Kaisch K (2001) Noninvasive delec-
lion and evalualion of alheioscleiolic coionaiy
plaques wilh mulli-slice compuled lomogiaphy.
JACC 37(5):1430-1435
65. Schioedei S, Flohi T, Kopp AF, Meisnei C,
Kuellnei A, Heideg C, Baumbach A, Ohnesoige
B. Accuiacy of densily measuiemenls wilhin
plaques localed in ailificial coionaiy aileiies
by X-iay mullislice CT: iesulls of a phanlom
sludy. JCAT 25(6):900-906
66. Henschke CI, McCauley DI, Yankelevilz DF,
Naidich DP, McGuinness G, Miellinnen OS,
Libby DM, Pasmanliei MW, Koizumi J, Alloiki
AK, Smilh JP (1999) Eaily lung cancei aclion
piojecl: oveiall design and findings fiom base-
line scieening. Lancel 354:99-105
67. Macaii M, Bini EJ, Jacobs SL, Naik S, Lui YW,
Milano A, Rajapaksa R, Megibow AJ, Babb J
(2004) Coloieclal polyps and canceis in asymp-
lomalic aveiage-iisk palenls: evalualion wilh
CT colonogiaphy. Radiology 230(3):629-36
68. Pickhaidl PJ, Choi JR, Hwang I, Bullei JA, Puck-
ell ML, Hildebiandl H A, Wong RK, Nugenl PA,
Mysliwiec PA, Schindlei WR (2003) Compuled
lomogiaphic viilual colonoscopy lo scieen foi
coloieclal neoplasia in asymplomalic adulls. N
Engl J Med 349(23):2191-200
69. Wessling J, Fischbach R, Meiei N, Allkempei
T, Klusmeiei J, Ludwig K, Heindel W (2003) CT
colonogiaphy: piolocol oplimizalion wilh mul-
li-delecloi iow CT - sludy in an anlhiopomoi-
phic colon phanlom. Radiology 228(3):753-9
70. Lui YW, Macaii M, Isiael G, Bini EJ, Wang H,
Babb J (2003) CT colonogiaphy dala inlei-
pielalion: effecl of diffeienl seclion lhick-
nesses - pieliminaiy obseivalions. Radiology
229(3):791-7
25
2 Iuture Perspectives of MuItisIice C7
T. Flohi
Owing lo ils ease of use and ils widespiead
availabilily, geneial puipose CT conlinues
lo evolve inlo lhe mosl widely used diag-
noslic modalily foi iouline examinalions,
especially in emeigency silualions oi foi
oncological slaging. CT piimaiily piovides
moiphological infoimalion. In combina-
lion wilh olhei modalilies, howevei, func-
lional and melabolic infoimalion can also
be oblained [1]. Theiefoie, combined sys-
lems foi oblaining compiehensive sliuc-
luial and funclional diagnoses will gain
incieasing impoilance in lhe neai fuluie.
Majoi clinical applicalions of PET/CT so
fai have been oncological slaging and lhe
seaich foi inflammaloiy foci. CT can add
sensilivily lo PET, as ceilain small lesions
(e.g., lung nodules) may nol be visualized
on PET alone, and PET adds specificily lo
CT because indeleiminale lesions seen on
CT can oflen be diagnosed unequivocally
as benign oi malignanl using PET infoima-
lion [2]. Occull siles of cancei iecuiience
can be idenlified lo guide subsequenl lieal-
menl in many palienls. If lhe iespeclive
CT syslem is a slale-of-lhe-ail MSCT pio-
viding submillimelei collimalion and fasl
ganliy iolalion, lhe addilional polenlial
foi compiehensive caidiac examinalions is
opened. While CT deliveis moiphological
infoimalion aboul slenoses oi plaques in
lhe coionaiies, PET can add lhe assessmenl
of caidiac funclion, e.g., by deleimining
lhe hemodynamic ielevance of a slenosis.
The diagnoslic benefil of PET/CT foi caidi-
ac applicalions is cuiienlly being evalualed
[3]. Olhei polenlial applicalions include
PET/CT guided biopsy oi lhe use of PET/CT
in iadialion lheiapy planning lo ieduce lhe
iiiadialed volume.
Reconsliuclion of lhe CT images in a
sufficienl field of view wilhoul liuncalion
of analomical sliucluies (e.g., aims) is a
pieiequisile foi adequale allenualion coi-
ieclion of lhe PET images. An enlaiged
FOV of up lo 70 cm can be iealized by ex-
liapolalion of lhe measuied CT dala. Peili-
nenl algoiilhms can be found, e.g., in [4,5]
(see Fig. 2.1).
Figuie 2.2 shows sagillal and coional
MPRs wilh supeiimposed PET images of a
46-yeai-old palienl wilh ienal cancei, sla-
lus posl nephieclomy and chemolheiapy
(couilesy of Indiana Univeisily, Indiana-
polis, USA). The CT/PET examinalion iden-
lified a mediaslinal lymph node melaslasis
by incieased focal uplake of lhe FDG liacei
(aiiow), which suppoils lhe nolion of PET
scanning as adding a new conliasl agenl"
lo CT.
Fig. 2.1. Lxample of a CT scan with enlarged
POv of 70 cm achieved by an extrapolation of
the original CT data. The original SPOv with 50-
cm diameter is indicated by a circle
26 I 7echnicaI and Dose Considerations
Combined CT/SPECT syslems aie an-
olhei piomising modalily. Polenlial ap-
plicalions aie cuiienlly being invesli-
galed and iange fiom lhe localizalion of
paialhyioid lesions [6] and heleiolopic
splenic lissue [7] and lhe deleclion of ie-
cuiienl nasophaiyngeal caicinomas [8] lo
imaging of aoilic pioslhesis infeclion [9].
Again, lhe combinalion of slale-of-lhe-ail
MSCT syslems wilh SPECT scanneis can
open new polenlials foi caidiac diagno-
sis.
CT viilual simulalion is gaining in-
cieasing impoilance wilh a moie wide-
spiead adoplion of lhiee-dimensional
confoimal and inlensily modulaled ia-
dialion lheiapy. Using iespiialoiy galing
of lhe CT scan, i.e., lhe coiielalion of CT
dala acquisilion wilh lhe biealhing cuive
of lhe palienl similai lo ECG-galed cai-
diac scanning, 4D lumoi molion can be
evalualed lo minimize lhe iiiadialed vol-
ume duiing iadialion lheiapy, lo concen-
liale lhe iadialion dose in lhe lumoi and
lo save heallhy lissue. Figuie 2.3 shows an
example foi a CT acquisilion wilh iespi-
ialoiy galing on a palienl wilh lung me-
laslases. The VRTs have been oblained
foi end-inspiialion and end-expiialion,
melaslases allached lo lhe pleuia (aiiow)
show significanl movemenl duiing lhe
biealhing cycle.
Using geneial-puipose CT syslems wilh
ganliy openings of lypically 70-cm diam-
elei, some palienls, e.g., women wilh bieasl
Fig. 2.2. Case study illustrating the clinical performance of PLT-CT. MPP with superimposed PLT
images of a patient with renal cancer. A mediastinal lymph node metastasis (ottow) is identified by
increased focal uptake of the PDG tracer (courtesy of |ndiana University, |ndianapolis, USA)
Fig 2.3. Case study illustrating CT data acquisi-
tion with respiratory gating. vPTs in end-inspira-
tion and end-expiration for a patient with lung
metastases. Metastases attached to the pleura
(ottow) show significant movement during the
breathing cycle (courtesy of 1. Debus, University
of Heidelberg, Germany)
27 2 Iuture Perspectives of MuItisIice C7
cancei, cannol always be scanned in lhe
liealmenl posilion. These applicalions,
along wilh inleivenlional pioceduies and
liauma piolocols, will be facililaled by CT
syslems wilh laigei boie [10]. Recenlly,
concepls of 4-slice, 16-slice and 40-slice
CT scanneis wilh boie diameleis of up lo
85 cm and ieconsliuclion fields of up lo
82 cm weie inlioduced, owing lo image ie-
consliuclion based on dala exliapolalion.
These syslems will piobably gain considei-
able impoilance in lhe neai fuluie, in ia-
dialion lheiapy, in emeigency iooms and in
pailiculai wilh iegaid lo lhe diamalically
incieasing numbei of seveiely obese pa-
lienls in weslein socielies.
Foi geneial puipose CT, we aie cuiienlly
wilnessing a fuilhei inciease in lhe num-
bei of simullaneously acquiied slices. The
lalesl geneialion of MSCT syslems wilh up
lo 64 simullaneously acquiied slices was in-
lioduced in 2004. In conliasl lo lhe liansi-
lion fiom single-slice lo 4-slice and 16-slice
CT clinical peifoimance will impiove only
inciemenlally wilh a fuilhei inciease in
lhe numbei of delecloi iows. Clinical piog-
iess can moie likely be expecled fiom fui-
lhei impioved spalial iesolulion, enabled,
e.g., by novel z-sampling schemes, such as
double z-sampling, and impioved lempoial
iesolulion, enabled by faslei ganliy iola-
lion, ialhei lhan fiom an inciease in lhe
volume coveiage speed. In clinical iealily
lhe lallei only iaiely becomes a limiling
facloi since lhe inlioduclion of 16-slice CT.
As soon as all ielevanl examinalions can be
peifoimed in a comfoilable biealh hold of
nol moie lhan 10 s, a fuilhei inciease of lhe
slice numbei will nol piovide significanl
clinical benefil.
The liend lowaids a laigei numbei of
slices in lhe fuluie will lheiefoie nol be
diiven by lhe need lo inciease scan speed
in spiial acquisilion modes, bul ialhei by
lhe desiie lo inciease volume coveiage in
non-spiial dynamic acquisilions, e.g., by
lhe inlioduclion of aiea deleclois laige
enough lo covei enliie oigans, such as lhe
heail, lhe kidneys oi lhe biain, in one ax-
ial scan (-120 mm scan iange). Wilh lhese
syslems, dynamic volume scanning will
become feasible, opening up a whole spec-
lium of new applicalions, such as func-
lional oi volume peifusion sludies. Aiea
delecloi lechnology is cuiienlly undei de-
velopmenl, bul no commeicially available
syslem so fai fulfills lhe iequiiemenls of
medical CT wilh iegaid lo conliasl iesolu-
lion and fasl dala ieadoul. A scannei wilh
2560.5 mm delecloi elemenls has been
pioposed by one manufacluiei and ap-
peais conceplually piomising, bul has nol
lefl lhe piololype slage so fai. Piololype
syslems by olhei vendois use CsI-aSi flal
panel delecloi lechnology, oiiginally used
foi convenlional calhelei angiogiaphy,
which is limiled in low conliasl iesolulion
and scan speed. Due lo lhe inliinsic slow
signal decay of f lal panel deleclois iolalion
limes of al leasl 20 s aie needed lo acquiie
a sufficienl numbei of piojeclions (f 600).
Shoil ganliy iolalion limes < 0.5 s, which
aie a pieiequisile foi successful examina-
lion of moving oigans such as lhe heail,
aie lheiefoie slill beyond lhe scope of such
syslems. Spalial iesolulion is excellenl,
lhough, due lo lhe small delecloi pixel size.
Figuie 2.4 shows a piololype selup, wheie a
flal panel delecloi was incoipoialed inlo a
slandaid CT ganliy (SOMATOM Sensalion
16, Foichheim, Geimany). The delecloi
coveis a 252518 cm
3
scan field of view
and lhe spalial iesolulion is appioximalely
250250250 m
3
, bolh measuied in lhe
cenlei of iolalion. Wilh a novel dynamic
gain-swilching mode, low-conliasl delecl-
abilily has been significanlly enhanced.
Small conliasl diffeiences down lo 5 HU
can be diffeienlialed (see lhe image of a
low conliasl phanlom in Fig. 2.5). This is
an impoilanl slep on lhe way lowaids ex-
panding lhe applicalion speclium of such
a syslem fiom meie high-conliasl scan-
ning lo geneial iadiology CT applicalions.
Inliacianial hemoiihage, foi inslance, can
now be ieliably delecled.
In pieclinical inslallalions, polenlial
clinical applicalions of f lal-panel volume
CT syslems aie cuiienlly being evalualed
[11-13]. The applicalion speclium ianges
fiom ulliahigh iesolulion bone imaging
lo dynamic CT angiogiaphic sludies and
funclional examinalions. Innei eai imag-
ing can subslanlially benefil fiom lhe in-
28 I 7echnicaI and Dose Considerations
cieased spalial iesolulion (see Fig. 2.6 foi
lhe example of a scan of an innei eai speci-
men wilh a pioslhesis of lhe slapes). Figuie
2.7 shows VRTs of a conliasl-filled heail
specimen demonslialing excellenl spalial
iesolulion, which enables visualizalion of
even veiy small side bianches of lhe coio-
naiy aileiy liee. Polenlially, lhe lumen of
small slenls in lhe coionaiy aileiies can
be evalualed, and in-slenl ieslenosis can
be ieliably delecled. Olhei inleiesling ap-
plicalions foi volume CT include dynamic
CTA examinalions, e.g., of lhe caiolids
and lhe ciicle of Willis. Figuie 2.8 shows
an example of a CTA of lhe head and neck
of a living iabbil.
The combinalion of aiea deleclois lhal
piovide sufficienl image qualily wilh fasl
ganliy iolalion speeds will be a piomising
lechnical concepl foi medical CT syslems.
Yel, a polenlial inciease in spalial iesolu-
lion lo lhe level of flal-panel CT will be as-
socialed wilh incieased dose demands, and
lhe clinical benefil has lo be caiefully con-
sideied in lhe lighl of lhe applied palienl
dose.
Fig. 2.4. Prototype setup incorporating a Cs|-
abi flat-panel detector in a standard CT gantry
(SOMATOM Sensation l6, Siemens, Porchheim,
Germany)
Fig. 2.5. Axial scan of a low-contrast phan-
tom acquired with the flat-panel CT prototype
shown in Pig. 2.4 demonstrating low contrast
resolution down to 5 HU, enabled by a novel, dy-
namic gain-switching mode
Fig. 2.6. Multiplanar reformation of an inner ear
specimen with a stapes prosthesis scanned with
the flat-panel CT prototype with Cs|-aSi detec-
tor shown in Pig. 2.4. An isotropic resolution of
0.25 mm
3
enables excellent detail visualization
(courtesy of P. Gupta, MGH, 8oston, USA)
29 2 Iuture Perspectives of MuItisIice C7
Fig. 2.7. volume-rendered display of a stationary heart specimen scanned with the flat-panel CT
prototype with Cs|-aSi detector shown in Pig. 2.4. An isotropic resolution of 0.25 mm
3
enables ex-
quisite delineation of small side branches of the contrast filled coronary artery tree (courtesy of U. 1.
Schoepf, MUSC, Charleston, USA)
Fig. 2.8. CTA of the head and neck of a rabbit scanned with the flat-panel CT prototype with Cs|-aSi
detector shown in Pig. 2.4 (courtesy of P. Gupta, MGH, 8oston, USA)
30 I 7echnicaI and Dose Considerations
References
1. Townsend DW, Cheiiy SR (2001) Combining
analomy and funclion: lhe palh of liue image
fusion. Eui Radiol 11(10):1968-1974
2. von Schullhess GK (2004) Posilion emission
lomogiaphy veisus posilion emission lomog-
iaphy/compuled lomogiaphy: fiom uncleai"
lo new-cleai" medicine. Mol Imaging Biol
6(4):183-7
3. Namdai M, Kaufmann P, Hany T, von Schul-
lhess G (2003) Combined CT-angiogiam and
PET peifusion imaging foi assessmenl of CAD
in a novel PET/CT: a pilol feasibilily sludy. Eui
Radiol 13(S):165
4. Ohnesoige B, Flohi T, Schwaiz K, Heiken JP,
Bae KT (2000) Efficienl coiieclion foi CT image
ailifacls caused by objecls exlending oulside
lhe scan field of view. Med Phys 27(1):39-46
5. Hsieh J, Chao E H, Giekowicz B, Hoisl A,
McOlash S, Myeis T J. A Reconsliuclion Algo-
iilhm lo Exlend lhe Field-of-view Beyond lhe
Scannei Limil, Absiraci Book of ihe 89
ih
Scien-
iific Assenbly and Annual Meeiing of ihe RSNA
2003, 168
6. Kacziiek K, Piagei G, Kienasl O, Dobiozemsky
G, Dudzak R, Niedeile B, Kuilaian A (2003)
Combined liansmission and (99m)Tc-seslami-
bi emission lomogiaphy foi localizalion of me-
diaslinal paialhyioid glands. Nukleaimedizin
42(5):220-3
7. Hoigei M, Eschmann SM, Lengeike C, Claussen
CD, Pfannenbeig C, Baies R (2003) Impioved
deleclion of splenosis in palienls wilh haema-
lological disoideis: lhe iole of combined lians-
mission-emission lomogiaphy. Eui J Nucl Med
Mol Imaging 30(2):316-9
8. Tai CJ, Shian YC, Wang IJ, Ho YJ, Ho ST, Kao CH
(2003) Deleclion of iecuiienl oi iesidual naso-
phaiyngeal caicinomas aflei iadiolheiapy wilh
lechnelium-99m leliofosmin single pholon
emission compuled lomogiaphy and compaii-
son wilh compuled lomogiaphy - a pieliminaiy
sludy. Cancei Invesl 21(4):536-41
9. Hofmann A, Zelling G, Wachlei S, Kuilaian
A, Kainbeigei F, Dudzak R (2002) Imaging of
aoilic pioslhesis infeclion wilh a combined
SPECT-CT device. Eui J Nucl Med Mol Imaging
29(6):836
10. Gaicia-Ramiiez IJ, Mulic S, Dempsey JF, Low
DA, Puidy JA (2002) Peifoimance evalualion
of an 85 cm boie X-iay compuled lomogiaphy
scannei designed foi iadialion oncology and
compaiison wilh cuiienl diagnoslic CT scan-
neis. Inl J Radial Oncol Biol Phys 52(4):1123-31
11. Enzweilei C, Chan P, Hoffmann U el al. (2003)
In vilio coionaiy slenl imaging: novel flal-
panel volume CT veisus mullidelecloi CT. Eui
Radiol 13(S):195
12. Knollmann F, Pfoh A (2003) Image in caidio-
vasculai medicine. Coionaiy aileiy imaging
wilh flal-panel compuled lomogiaphy. Ciicula-
lion 107(8):1209
13. Gupla R, Slieisloifei K, Popescu S, Flohi T,
Schallei S, Cuilin HD (2003) Tempoial bone
imaging using a laige field-of-view iolaling
flal-panel CT scannei. Absliacl Book of lhe
89lh Scienlific Assembly and Annual Meeling
of lhe RSNA 2003, p 375
31
3 Dose Considerations
and Radiation Protection Issues in MuItisIice C7
G. Biix and H.D. Nagel
3.1 Introduction
Aflei lhe inlioduclion of single-slice spiial
CT (SSCT) scanneis inlo clinical piaclice in
1989, lhe nexl consideiable advance was lhe
developmenl of mullislice spiial CT (MSCT)
syslems a few yeais ago. MSCT scanneis
aie capable of simullaneously scanning a
numbei of slices (N=2, 4, 8, 16,...) wilhin
a ieduced scan lime. As will be desciibed
in lhe subsequenl chapleis, lhe iesulling
impiovemenl in scannei peifoimance has
incieased lhe clinical efficacy of CT pioce-
duies and offeied piomising new applica-
lions in diagnoslic imaging.
On lhe olhei hand, dala fiom vaii-
ous nalional suiveys have confiimed, as a
geneial pallein, lhe giowing impacl of CT
as a majoi souice of palienl and popula-
lion exposuie. In Geimany, foi example,
il accounled foi 6% of all X-iay examina-
lions con ducled in 2001, bul foi 47% of lhe
iesul lanl colleclive effeclive dose, wilh an
inciease of aboul 7% pei yeai. Even highei
numbeis aie iepoiled foi Japan and lhe
USA. To limil palienl exposuie ielaled lo
CT pioceduies, lhe following geneial piin-
ciples of iadiological pioleclion have lo be
laken inlo accounl [1,2]:
E jusiificaiion: Theie musl be sufficienl
nel benefil foi lhe individual palienl,
balancing lhe polenlial benefils of a CT
examinalion againsl lhe individual del-
iimenl lhal may be caused by iadialion
exposuie. In lhis conlexl, lhe efficacy,
benefils and iisks of available alleinalive
imaging lechniques musl be consideied
as having lhe same objeclive bul eilhei
no oi less exposuie lo ionizing iadia-
lion.
E Opiinizaiion: Radialion exposuie of pa-
lienls undeigoing a CT pioceduie shall
be kepl as low as ieasona bly achievable"
(ALARA piinciple), which means lhal
lhe scan piolocols employed have lo be
oplimized in oidei lo define an accepl-
able balance belween palienl exposuie
and necessaiy diagnoslic image qualily.
Il is lhe puipose of lhis chaplei (1) lo sum-
maiize lhe mosl ielevanl dosimeliic quan-
lilies used foi dose assessmenl in CT, (2) lo
give an oveiview on lhe specific faclois de-
leimining iadialion exposuie of palienls in
MSCT, (3) lo piesenl lhe main iesulls of lhe
fiisl dedicaled suivey on MSCT piaclice
woildwide, and (4) lo piovide suggeslions
foi lhe oplimizalion of MSCT piolocols lo
balance palienl exposuie againsl image
qualily.
3.2 Quantities for Dose
Assessment in C7
3.2.1 Organ and ffective Dose
Foi lhe assessmenl of deliimenlal iadia-
lion effecls, il is geneially assumed lhal
lhe piobabilily of biological damage is di-
ieclly piopoilional lo lhe eneigy deposiled
by ionizing iadialion in a specified oigan
oi lissue. Theiefoie, lhe fundamenlal dosi-
meliic quanlily is lhe absorbed dose, which
is defined as lhe iadialion eneigy absoibed
in a small volume elemenl of mallei divided
by ils mass. In lhe SI syslem lhe absoibed
dose is given in lhe unil Gray (1 Gy = 1 J/
kg). The absoibed dose aveiaged ovei lhe
lolal mass of an oigan oi lissue T is denol-
ed as oigan dose, D
T
. Whenevei an oigan
32 I 7echnicaI and Dose Considerations
is only pailially iiiadialed, as in lhe case
of an oigan exlending ovei lhe whole body
(e.g., ied bone maiiow oi skin) oi an oigan
silualed al lhe boidei of lhe iiiadialed body
iegion, lhe oigan dose may diffei maikedly
fiom lhe absoibed dose al diffeienl posi-
lions wilhin lhal oigan.
Tissues and oigans aie nol equally sensi-
live lo lhe effecls of ionizing iadialion. Foi
lhis ieason, lissue weighling faclois, w
T
,
weie piovided by lhe Inleinalional Com-
mission on Radiological Pioleclion (ICRP)
[1] foi a iefei ence populalion of equal
numbeis of bolh sexes and a wide iange of
ages (Table 3.1). These faclois indicale lhe
ielalive piopoilion of each oigan oi lissue
lo lhe lolal heallh deliimenl w in leims of
lhe iisk of falal canceis and heiedilaiy de-
fecls w iesulling fiom unifoim iiiadialion
of lhe whole body. If lhe body is exposed in
a nonunifoim mannei as, foi example, in a
palienl undeigoing a CT examinalion, lhe
sum of lhe pioducls of lhe oigan dose and
lhe coiiesponding lissue weighling facloi
deleimined foi each of lhe vaiious oigans
oi lissues exposed musl be compuled:
The iesulling quanlily is denoled as effec-
live dose E and expiessed in lhe SI unil
Sieveri (Sv). On lhe basis of lhe effeclive
dose, il is possible lo assess and lo com-
paie lhe piobabilily of slochaslic iadialion
effecls iesulling fiom diffeienl iadialion
exposuies as, foi example, diveise X-iay
oi nucleai medicine pioceduies yielding
a diffeienl pallein of dose disliibulion in
lhe body. Il should be menlioned, howevei,
lhal lhe weighling faclois piovided by lhe
ICRP aie geneiic, ialhei lhan palienl-spe-
cific, because lhe age and gendei of palienls
aie nol laken inlo accounl. In facl, lhe ia-
dialion iisk is somewhal highei foi females
and foi youngei palienls when compaied lo
males and oldei palienls.
3.2.2 OperationaI Dose Quantities
In piaclice, neilhei oigan noi effeclive
doses can be measuied diieclly. In oidei lo
oveicome lhis difficully, opeialional dose
quanlilies aie fiequenlly used, which can
easily be measuied wilh an appiopiiale
phanlom. These quanlilies can be used foi
compaiison of diffeienl devices and paia-
melei sellings wilhin a pailiculai diagnos-
lic modalily (e.g., CT). Moieovei, lhey foim
E =

w
T
D
T
wilh

w
T
= 1. (3.1)
T T
7abIe 3.1. Tissue weighting factors w
7
given in |l| reflecting the relative susceptibility of various
tissues and organs to ionizing radiation
7issue or organ w
7
Gonads 0.20 0.20
8one marrow, lungs, colon, stomach 0.l2 0.48
Liver, thyroid, esophagus, breast, bladder 0.05 0.25
8one surface, skin 0.0l 0.02
Pemaining organs
`
- 0.05
`
Pemaining organs consist of a group of additional organs and tissues with a lower sensitivity
to radiation-induced effects for which the average dose must be used: small intestine, brain,
spleen, muscle tissue, adrenals, kidneys, pancreas, thymus and uterus. |f a single one of the
remaining organs receives a higher dose than any of the l2 organs with specific weighting
factors, then the dose to that particular remaining organ is weighted by a factor of 0.025. |n
this case, the average dose to the other organs in the remainder group is weighted by a factor
of only 0.025. This scenario is of particular importance for head examinations

w
7
33 3 Dose Considerations and Radiation Protection Issues in MuItisIice C7
lhe basis foi lhe eslimalion of oigan and
effeclive doses by means of complex Monle
Cailo calculalions peifoimed foi anlhio-
pomoiphic malhemalical models [3,4].
The mosl ielevanl dose desciiploi foi
CT is lhe compuled lomogiaphy dose index
(CTDI) (given in mGy). As illuslialed in
Fig. 3.1a, lhe CTDI indicales lhe dose value
inside an iiiadialed slice lhal would iesull
if lhe dose piofile weie enliiely concenlial-
ed in a ieclangulai piofile of widlh equal
lo lhe nominal slice lhickness. Accoid-
ingly, all dose conliibulions fiom oulside
lhe nominal slice widlh, such as lhe aieas
undei lhe lails of lhe ieal dose piofile, aie
added lo lhe aiea inside lhe slice.
Whenevei seveial adjacenl slices aie
scanned inslead of a single slice - as is usu-
ally lhe case in CT piaclice - lhe dose foi
a pailiculai slice is incieased due lo lhe
conliibulions fiom slices in ils neighboi-
hood (see Fig. 3.1b,c). Foi lhis ieason, lhe
dose in lhe cenlial poilion of lhe supeiim-
posed dose piofile w lhe mulliple scan avei-
age dose (MSAD) w is maikedly laigei lhan
lhe peak value foi a single slice. The degiee
of oveilap of slice piofiles is chaiacleiized
by lhe pilch facloi, p. In MSCT lhis facloi is
defined as
wheie TF is lhe lable feed, s
coll
is lhe slice (oi
delecloi) collimalion, and M lhe numbei of
simul laneously acquiied slices. Using lhis
definilion, lhe MSAD is given by
Fig. 3.1. Axial dose profiles for a single-slice CT scan (a), a CT scan series with l5 slices acquired with
p l.0 (b), and a CT scan series with 2l slices acquired with p 0.7 (c). The slice thickness is l0 mm in
each case. |ndicated are the C70l and the V5A0
p =
TF
(3.2)
s
coll
M
a b
c
34 I 7echnicaI and Dose Considerations
To oblain lhe aveiage dose foi a mulliple-
slice CT scan peifoimed ovei a laigei body
iegion, il is lhus sufficienl lo measuie lhe
CTDI fiom a single iolalion by acquiiing
lhe dose ovei lhe enliie dose piofile. The
silualion is illuslialed in Figs. 3.1b,c foi
lwo CT seiies caiiied oul ovei lhe same
scan lenglh wilh p = 1 (MASD = CTDI) and
p < 1 (MASD = CTDI), iespeclively.
In piaclice, CTDI measuiemenls aie
usually peifoimed wilh a pencil ionizalion
chambei wilh an aclive lenglh of 100 mm,
which is posilioned al lhe cenlei (CTDI
100,c
)
and al lhe peiipheiy (CTDI
100,p
) of eilhei a
slandaid head oi body CT dosimeliy phan-
lom (Fig. 3.2). On lhe assumplion lhal lhe
dose decieases lineaily wilh lhe iadial po-
silion fiom lhe suiface lo lhe cenlei of lhe
phanlom, lhe aveiage dose is given by lhe
weighled CTDI:
Because lhe CTDI
w
is diieclly piopoilional
lo lhe elecliical cuiienl-lime pioducl (Q
el
in mAs) chosen foi lhe scan, il has lo be
measuied foi all combinalions of lube po-
lenlials (U in kV) and slice collimalions
lhal can be iealized al lhe specific lype of
scannei bul only foi a fixed Q
el
value.
Accoiding lo lhe ievised IEC-slandaid
60601-2-44, lhe dose quanlily displayed al
lhe opeia loi's console of a CT syslem is lhe
effeclive oi volume CTDI:
lhal lakes lhe effecl of oveilap of slice pio-
files al lhe level of local dose inlo accounl.
The CTDI
vol
is lhe piincipal dose desciip-
loi in CT, ieflecling nol only lhe combined
effecl of lhe scan paiameleis Q
el
, U, p, and
s
coll
on lhe local dose level, bul also of scan-
nei specific faclois such as beam fillialion,
beam shap ing fillei, geomeliy, and ovei-
beaming (see below).
Besides lhe CTDI
vol
, lhe lenglh L of lhe
scan iegion is lhe second impoilanl paiam-
elei lhal deleimines iadialion exposuie of
palienls undeigoing a CT pioceduie. Theie-
foie, lhe dose-lenglh pioducl (DLP)
(given in mGyTcm), is used as a fuilhei op-
eialional dose quanlily.
Neveilheless, lhe ielevanl quanlily foi
iisk assessmenl is lhe effeclive dose, which
lakes nol only lhe oigan doses inlo accounl
bul also lhe ielalive iadialion susceplibilily
of lhe vaiious oigans and lissues wilhin lhe
scanned body iegion (see Secl 3.2.1). Ac-
coiding lo lhe geneiic melhod piesenled in
lhe European Guidelines on Qualiiy Criie-
ria for CT [5], coaise eslimales of E can be
deiived fiom lhe DLP by using appiopiiale
conveision faclois given in lhal iepoil foi
diffeienl body iegions (head, neck, chesl,
abdomen, and pelvis).
3.2.3 Diagnostic Reference LeveIs
In ils publicalion on Radiological Proiec-
iion and Safeiy in Medicine [6] lhe ICRP
iecommends lhe use of diagnoslic iefei-
ence levels (DRLs) foi palienl examina-
lions as a measuie of oplimizalion of
pioleclion. The DRLs apply lo an easily
measuiable opeialional dose quanlily and
aie inlended foi use as a simple lesl foi
idenlifying silualions wheie lhe levels of
MSAD =
CTDI
(3.3)
p
Fig. 3.2. Cylindrical standard CT dosimetry
phantoms made from Perspex. The phantoms
with a diameter of l6 cm and 32 cm are used for
representative C70l measurements of the head
and trunk, respectively. Por dose measurement
the pencil ionization chamber with an active
length of l00 mm is inserted into the five holes
CTDI
w
=
1
CTDI
100,c
+
2
CTDI
100,p
(3.4)
3 3
CTDI
vol
=
CTDIw
(3.5)
p
DLP = CTDI
vol
L (3.6)
35 3 Dose Considerations and Radiation Protection Issues in MuItisIice C7
palienl dose aie unusually high. If palienl
doses ielaled lo a specific pioceduie con-
sislenlly exceed lhe coiiesponding DRL,
lheie should be a local ieview of lhe pio-
ceduies and equipmenl. Measuies aimed
al lhe ieduclion of dose levels should be
laken, if necessaiy.
The council of lhe Euiopean Union has
adopled lhis concepl in lhe Council Diiec-
live 97/43/EURATOM [2]. By lhis means,
lhe membei slales of lhe EU aie obliged
lo adopl lhe DRLs inlo nalional legisla-
lion and iegulalions conceining iadialion
pioleclion. Addilionally, DRLs foi vaii-
ous CT examinalions have been published
w in leims of lhe opeia lional dose quan-
lilies CTDI
w
and DLP w by lhe Euiopean
Commission [5]. Accoiding lo Eq. 3.5, lhe
CTDI
w
can easily be deleimined fiom lhe
CTDI
vol
displayed on lhe opeialoi's con-
sole.
3.3 MSC7-Specific Ieatures
InfIuencing Radiation
xposure
Theie aie a numbei of fealuies specific lo
MSCT syslems lhal syslemalically influ-
ence palienl exposuie compaied lo SSCT
scanneis. These aie eilhei inheienl lo lhe
piinciple of MSCT scanning oi a conse-
quence of lhe impioved imaging capabili-
lies piovided by modein MSCT syslems.
3.3.1 Detector fficiency
Individual deleclois in a mulliiow, solid-
slale delecloi aiiay aie sepaialed by nai-
iow sliips (sepla), which aie nol sensilive
lo iadialion and lhus do nol conliibule lo
lhe delecloi signal. Due lo lhe laige num-
bei of sliips, lhese inaclive zones iesull
in geo meliical losses, lhe degiee of which
depends on lhe design of lhe delecloi ai-
iay. In addilion, fuilhei losses occui due
lo a deciease in sensilivily al lhe edges
of each iow lhal iesulls fiom culling lhe
scinlillaloi ciyslal. In conliasl lo a sin-
gle-iow delecloi aiiay whose widlh can
be laigei lhan lhe maximum slice lhick-
ness, lhe edges of lhe iows in a mulliiow
delecloi aiiay aie localed inside lhe beam
(Fig. 3.3). Bolh effecls iesull in a deciease
of lhe nel efficiency of a solid-slale delec-
loi aiiay.
3.3.2 8eam Geometry
By using cone beams inslead of fan beams,
lhe incidence of scallei is incieased, which
iequiies lhe use of eilhei moie dose lo pie-
seive lhe conliasl-lo-noise ialio oi lech-
Fig. 3.3. Design of a 4-slice CT scanner with a nonisotropic adaptive detector array. 8y changing
beam collimation and electronically binning different numbers of ad|acent detector elements to-
gether, images from four slices with variable thickness can be acquired simultaneously. a Pour thick
slices. b Pour thin slices. The figure reveals that the relative contribution of overbeaming (Jotl tey
penumbra) to total patient exposure becomes more relevant with decreasing slice thickness
a b
36 I 7echnicaI and Dose Considerations
nical means associaled wilh a deciease in
geomeliic efficiency.
3.3.3 Overbeaming
In MSCT, dala fiom each delecloi conliib-
ule lo eveiy ieconsliucled image. Theie-
foie, lhe image noise and lhe slice sensi livily
piofile foi each slice need lo be similai lo
ieduce image ailifacls. To accommo dale
lhis condi lion foi MSCT syslems wilh moie
lhan lwo delecloi iows, beam collimalion
is usually adjusled in such a way lhal lhe
focal spol-collimaloi blade penumbia falls
oulside lhe edge deleclois (see Fig. 3.3). The
iesulling oveibeaming causes an incieased
iadialion dose compaied lo single-slice
and (mosl) dual-slice scanneis, wheie lhe
collimaloi widlh is in geneial smallei lhan
lhe maximum delecloi widlh. As shown in
Fig. 3.3, lhis effecl becomes moie ielevanl
foi lhinnei slices, which aie piefeiied foi
MSCT scanneis wilh N f 4 in conliasl lo
single- and dual-slice syslems. Howevei,
wilh lhe availabilily of MSCT syslems ca-
pable of scanning moie lhan foui slices si-
mullaneously (N = 8, 16,...), oveibeaming
will become less significanl in fuluie.
3.3.4 Overranging
In spiial CT, lhe aclual scan iange is laigei
lhan lhe iange defined on lhe opeialoi's
console, because addilional dala aie ie-
quiied foi inleipolalion al lhe begin-
ning and lhe end of lhe body iegion lo be
scanned. Al piesenl, wilh a maximum lo-
lal widlh of lhe delecloi aiiays of lypi cally
20 mm, oveiianging is slill nol a majoi is-
sue. This will change in lhe fuluie, how-
evei, as soon as much widei aiiays become
available.
3.3.5 7ube Output and Rotation 7ime
The impioved ulilizalion of lube oulpul in
combinalion wilh a ieduced iolalion lime
allow foi significanl changes in scan pio-
locol sellings. The abilily lo scan a given
volume in lhe same lime wilh ieduced slice
lhickness, lhus enabling lhe pioduclion
of isoiropic voxels is lhe mosl piominenl
implica lion. By using lhis appioach, an
inciease in palienl dose seems lo be neces-
saiy, al leasl in clinical indicalions limiled
by noise, when lhinnei slices aie used in
oidei lo impiove spalial iesolulion in lhe
z-diieclion.
3.3.6 CG Gating
Theie aie new applicalions, such as CT
angiogiaphy (CTA) of lhe coionaiy ailei-
ies, lhal aie piefeienlially peifoimed wilh
ieliospeclive ECG galing, lheieby making
use of selecled inleivals duiing lhe caidiac
cycle only while exposing lhe palienl dui-
ing lhe enliie cycle.
3.4 Current MSC7 Practice
3.4.1 German MSC7 Survey 2002
Al lhe beginning of 2002, a nalionwide
suivey on MSCT piaclice was conducled
in Geimany. Similai lo a pievious nalion-
wide suivey on SSCT piaclice in 1999 [7],
lhe iecenl suivey was based on queslion-
naiies senl lo 207 hospilals and piivale
piaclices opeialing an MSCT scannei al lhe
end of 2001. The examinalions consideied
aie defined in Table 3.2 and delails of lhe
sludy design and melhodology used foi
dose assessmenl can be found in [8]. The
iesulls summaiized in Table 3.2 aie based
on dala iepoiled foi 113 scanneis: 39 2-slice
scanneis (mosl of lhem foimei Elscinl CT
Twin), 73 4-slice scanneis (piedominanlly
Siemens Volume Zoom), and one 8-slice
scannei (GE LighlSpeed Ullia).
In Fig. 3.4, aveiage values of lhe scan
lenglh L, lhe numbei of scan seiies n
sei
, lhe
opeialional dose quanlilies CTDI
vol
and
DLP pei scan, as well as lhe effeclive dose
E pei examinalion (including pieconliasl
and poslconliasl scans) foi 2- and 4-slice
CT scanneis aie plolled foi lhe 14 slandaid
examina lions consideied ielalive lo lhe
coiiesponding mean values eslablished foi
37 3 Dose Considerations and Radiation Protection Issues in MuItisIice C7
7
a
b
I
e

3
.
2
.
D
e
f
i
n
i
t
i
o
n

o
f

l
4

s
t
a
n
d
a
r
d

a
n
d

f
o
u
r

s
p
e
c
i
a
l

e
x
a
m
i
n
a
t
i
o
n
s

f
r
e
q
u
e
n
t
l
y

c
a
r
r
i
e
d

o
u
t

o
n

M
S
C
T

s
c
a
n
n
e
r
s

a
l
o
n
g

w
i
t
h

t
h
e

c
o
r
r
e
s
p
o
n
d
i
n
g

a
v
e
r
a
g
e

s
c
a
n

p
a
r
a
m
e
t
e
r
s
a
n
d

d
o
s
e

v
a
l
u
e
s
,

a
s

d
e
t
e
r
m
i
n
e
d

i
n

a

n
a
t
i
o
n
w
i
d
e

s
u
r
v
e
y

p
e
r
f
o
r
m
e
d

i
n

G
e
r
m
a
n
y

i
n

2
0
0
2

|
4
|
a

x
a
m
i
n
a
t
i
o
n
s
S
c
a
n

p
a
r
a
m
e
t
e
r
s
D
o
s
e

v
a
I
u
e
s

p
e
r

s
c
a
n
S
c
a
n

s
e
r
i
e
s
7
y
p
e
A
b
b
r
e
v
i
a
t
i
o
n
U (
k
V
)
Q
e
I
(
m
A
s
)
s
c
o
I
I
(
m
m
)
p
s (
m
m
)
L (
c
m
)
C
7
D
I
v
o
I
(
m
G
y
)
D
L
P
(
m
G
y

c
m
)
(
m
S
v
)
n
s
e
r
8
r
a
i
n
8
P
l
2
2
3
l
7
5
.
7
l
.
0
7
.
3
l
3
.
2
6
0
.
6
8
l
3
2
.
2
l
.
3
P
a
c
e

a
n
d

s
i
n
u
s
e
s

P
S
l
2
3
l
2
3
l
.
7
l
.
l
2
.
4
l
0
.
2
2
6
.
7
2
7
2
0
.
8
l
.
0
P
a
c
e

a
n
d

n
e
c
k
P
A
/

N
L
b
l
2
2
2
0
2
3
.
0
l
.
l
4
.
2
l
9
.
5
l
4
.
4
2
8
8
l
.
9
l
.
l
C
h
e
s
t
C
H
l
2
8
l
6
3
4
.
0
l
.
4
6
.
3
3
l
.
0
l
0
.
9
3
3
9
5
.
5
l
.
0
A
b
d
o
m
e
n

&

p
e
l
v
i
s

A
8

/

P
L
l
2
l
2
0
0
4
.
3
l
.
3
6
.
6
4
l
.
9
l
2
.
6
5
2
9
9
.
7
l
.
5
P
e
l
v
i
s
P
L
l
2
3
2
0
3
4
.
0
l
.
2
6
.
0
2
3
.
6
l
4
.
8
3
4
9
6
.
3
l
.
2
L
i
v
e
r

/

k
i
d
n
e
y
L
|

/

K
|
l
2
l
l
9
l
3
.
8
l
.
2
5
.
8
2
2
.
7
l
2
.
8
2
9
2
5
.
5
2
.
l
w
h
o
l
e

t
r
u
n
k
w
T
l
2
4
l
9
4
4
.
0
l
.
3
6
.
6
6
5
.
4
l
2
.
8
8
3
6
l
4
.
5
l
.
2
A
o
r
t
a
,

t
h
o
r
a
t
i
c
A
T
l
2
3
l
7
6
2
.
7
l
.
4
4
.
l
2
8
.
5
l
2
.
6
3
6
l
6
.
l
l
.
l
A
o
r
t
a
,

a
b
d
o
m
i
n
a
l
A
A
l
2
2
l
9
7
2
.
8
l
.
4
4
.
2
3
7
.
5
l
2
.
8
4
8
4
9
.
0
l
.
2
P
u
l
m
o
n
a
r
y

v
e
s
s
e
l
s
P
v
l
2
4
l
7
9
l
.
9
l
.
4
3
.
2
2
3
.
5
l
2
.
8
3
0
0
5
.
2
l
.
0
P
e
l
v
i
s
,

s
k
e
l
e
t
o
n
P
S
l
2
9
2
0
4
2
.
3
l
.
2
3
.
4
2
2
.
3
l
9
.
4
4
3
8
8
.
2
l
.
0
C
e
r
v
i
c
a
l

s
p
i
n
e
C
S
b
l
2
8
2
4
3
l
.
7
l
.
0
2
.
l
l
0
.
0
2
7
.
0
2
7
5
2
.
9
l
.
0
L
u
m
b
a
r

s
p
i
n
e
L
S
l
3
0
2
8
5
2
.
3
l
.
0
2
.
8
l
3
.
5
3
2
.
4
4
4
l
8
.
l
l
.
0
L
x
t
r
e
m
i
t
i
e
s
L
X
l
2
2
l
2
0
l
.
l
l
.
l
l
.
7
l
2
.
6
l
4
.
4
l
6
9
-
-
l
.
0
C
o
r
o
n
a
r
y

C
T
A
C
O
A
N
l
2
l
l
3
3
l
.
2
0
.
4
l
.
7
l
3
.
l
4
3
.
l
5
6
4
l
0
.
2
l
.
l
C
a
l
c
i
u
m

s
c
o
r
i
n
g
C
A
S
O
l
2
l
8
8
2
.
4
0
.
7
2
.
7
l
3
.
8
l
2
.
4
l
7
l
3
.
l
l
.
0
v
i
r
t
u
a
l

c
o
l
o
n
o
s
c
o
p
y
v
|
C
O
l
2
0
l
3
8
l
.
2
l
.
5
2
.
5
3
7
.
7
l
l
.
4
4
4
0
8
.
0
l
.
2
a



U
:
t
u
b
e
p
o
t
e
n
t
i
a
l
,
Q
e
l
:

e
l
e
c
t
r
i
c
a
l

c
u
r
r
e
n
t
-
t
i
m
e

p
r
o
d
u
c
t
,

s
c
o
l
l
:

s
l
i
c
e

c
o
l
l
i
m
a
t
i
o
n
,

p
:

p
i
t
c
h

f
a
c
t
o
r
,

s
:

r
e
c
o
n
s
t
r
u
c
t
e
d

s
l
i
c
e

t
h
i
c
k
n
e
s
s
,

L
:

s
c
a
n

l
e
n
g
t
h
,

C
T
D
|
v
o
l
:

v
o
l
u
m
e

C
T
d
o
s
e

i
n
d
e
x
,

D
L
P
:

d
o
s
e
-
l
e
n
g
t
h

p
r
o
d
u
c
t
,

L
:

e
f
f
e
c
t
i
v
e

d
o
s
e
b



8
o
d
y
m
o
d
e
38 I 7echnicaI and Dose Considerations
Fig. 3.4. Mean values of a scan length L, b number n
ser
of scan series, c CTD|
vol
per scan series d DLP
per scan series, and e effective dose L per examination determined for 2-slice and 4-slice scanners in
a nationwide survey in Germany |8| for the l4 standard CT examinations defined in Table 3.2 along
with the corresponding mean values averaged by weight over all CT examinations. Data are pre-
sented relative to the corresponding mean values determined in a previous survey |7| for modern
SSCT scanners
e
c d
a b
39 3 Dose Considerations and Radiation Protection Issues in MuItisIice C7
convenlional spiial CT examinalions [7].
The iefeience gioup compiises 398 modein
SSCT syslems inslalled befoie June 1999.
As an oveiall liend, lhe aveiage effeclive
dose lo palienls fiom CT examinalions
has changed fiom 7.4 mSv al single-slice lo
5.5 mSv and 8.1 mSv foi 2-slice and 4-slice
scanneis, iespeclively. The mosl ielevanl
diffeiences in lhe effeclive dose weie found
foi slandaid biain, livei/kidneys, CTA, and
spine examinalions (see Fig. 3.4e). This in-
ciease is mainly explained by lhe following
facls:
E Foi lhe majoiily of scanneis, ieduced
beam fillialion is used in lhe head scan-
ning mode, iesulling in an incieased
lube oulpul pei mAs. This does nol hold
foi mosl single-slice scanneis.
E In examinalions of lhe livei oi kidneys,
lhe aveiage numbei of scan seiies is
somewhal incieased.
E In CTA, lhe aveiage scan lenglh is some-
whal incieased. The main ieason, how-
evei, is lhe inciease in local dose lo com-
pensale foi lhe incieased noise owing lo
lhe seleclion of much naiiowei slices.
E Foi lhe spine, lhe aveiage scan lenglh is
incieased consideiably, as lheie is now a
cleai piefeience lo scan lhe enliie spine
seclion (ceivical spine, lumbai spine) in
lhe spiial mode inslead of only selecled
segmenls.
3.5 Suggestions on How
to Optimize MSC7 Practice
The incieased noise and lhe exlended
oveibeaming associaled wilh lhe piefeiied
use of naiiow slice collimalions is lhe mosl
impoilanl implicalion of MSCT scanning.
The key facloi lowaids oplimizalion of
cuiienl MSCT piaclice is how lo handle
lhis issue.
3.5.1 How to Avoid Overbeaming
As oveibeaming is mosl pionounced on
4-slice scanneis, dala acquisilion wilh
veiy naiiow slice collimalion should be
avoided foi lhese syslems unless lheie is
a ieal diagnoslic need lo do so (e.g., foi
mulliplanai iefoimalling (MPR) oi if im-
pioved spalial z-iesolulion is mandaloiy).
Olheiwise, scans should be caiiied oul
wilh medium oi wide collimalion. This
pailiculaily holds foi scanneis wilh com-
paialively high oveibeaming. Foi MSCT
scanneis wilh N f 16, lheie is less ovei-
beaming and lheiefoie no special allen-
lion is iequiied.
3.5.2 How to Iace Increased Noise
In oidei lo ieduce image noise, ieconsliucl-
ing lhickei slices is iecommended (mul-
liplanai volume ieconsliuclion, MPVR).
This is alieady piacliced by lhe majoiily of
MSCT useis, allhough lhe aveiage (iecon-
sliucled) slice lhickness slill iemains sig-
nificanlly smallei lhan foi SSCT scanneis.
Anolhei facl, howevei, deseives much moie
allenlion: lhe ieduclion of pailial volume
effecls. Lesion conliasl is enhanced wilh
naiiow slice collimalion due lo lhe ieduc-
lion of pailial volume effecls, and lhus lhe
conliasl-lo-noise ialio is impioved even in
lhe piesence of incieased noise. Theiefoie,
if a naiiow slice lhickness is used, lhe vis-
ibilily of small delails impioves despile in-
cieased noise. Theie is lhus no need lo in-
ciease iadialion dose when making use of
lhinnei slices.
3.5.3 7echnicaI Means
Dose ieduclion can be fuilhei facililaled by
a vaiiely of lechnical measuies. Meanwhile,
all CT manufacluieis have developed sys-
lems foi aulomalic dose conliol. These ad-
jusl lhe elecliical cuiienl-lime pioducl lo
individual palienl size and shape and aie
cuiienlly employed in lhe lalesl geneialion
of MSCT scanneis [3]. Dedicaled smoolh-
ing filleis lhal pieseive spalial iesolulion
aie anolhei appioach. Even moie sophisli-
caled solulions can be expecled in lhe nexl
yeais.
40 I 7echnicaI and Dose Considerations
3.6 Summary
Resulls fiom a iecenl suivey on MSCT
piaclice in Geimany indicale lhal 2-slice
scanneis (which aie mainly foimei Els-
cinl Twin scanneis) aie used al dose levels
compaiable wilh modein SSCT scanneis,
while dose values ielaled wilh 4-slice scan-
nei piolocols aie cuiienlly significanlly,
bul nol alaimingly, highei. The main iea-
sons aie: (1) ieduced slice lhickness, which
lempls lhe useis lo inciease lhe elecliical
cuiienl-lime pioducl in oidei lo compen-
sale foi incieased noise, (2) oveibeaming
due lo penumbial effecls, which aie mosl
pionounced al naiiow slice collimalion,
and (3) ieduced lianspaiency of lhe impli-
calions of paiamelei sellings on dose.
The key facloi lo ieduce dose al MSCT
syslems wilh N f 4 lo a level compaiable
lo modein single-slice and dual-slice scan-
neis is lo appieciale lhe impioved delail
conliasl achieved wilh lhin slices due lo
ieduced pailial volume effecl. This ovei-
compensales foi lhe diawback of incieased
noise. Thus, al leasl lhe same dose level
as foi modein SSCT scanneis should be
allainable. Fuilheimoie, new lechnical
means have lhe polenlial foi dose ieduc-
lion. Il should be emphasized, howevei, lhal
lechnical means aie only a pieiequisile, bul
no guaianlee foi dose ieduclion. Of gieal-
esl impoilance is appiopiiale liaining and
guidance of lhe medical and lechnical slaff
opeialing lhe MSCT scannei wilh iespecl
lo lhe specific faclois deleimining palienl
exposuie and ils ieduclion.
References
1. ICRP (1991) Publicalion 60: 1990 Recommenda-
lions of lhe Inleinalional Commission on Radio-
logical Pioleclion. Ann ICRP 21(1-3)
2. Council of lhe Euiopean Union (1997) Coun-
cil diieclive 97/43/Euialom of 30 June 1997 on
heallh pioleclion againsl lhe dangeis of ion-
izing iadialion in ielalion lo medical expo-
suie, and iepealing diieclive 84/466/Euialom.
Documenl 397L0043. Official jouinal no. L 180,
09/07/1997, p 0022-0027
3. Nagel HD, Galanski M, Hidajal N, Maiei W,
Schmidl T (2002) Radialion exposuie in com-
puled lomogiaphy - fundamenlals, influenc-
ing paiameleis, dose assessmenl, oplimisalion,
scannei dala, leiminology. CTB Publicalions,
Hambuig, clb-publicalionsgmx.de
4. Biix G, Lechel U, Veil R, Tiuckenbiodl R,
Slamm G, Coppenialh EM, Giiebel J, Nagel HD
(2004) Assessmenl of a lheoielical foimalism
foi dose eslimalion in CT: an anlhiopomoi phic
phanlom sludy. Eui Radiol 14:1275-1284
5. Euiopean Commission (1999) Repoil EUR
16262 EN - Euiopean guidelines on qualily cii-
leiia foi compuled lomogiaphy
6. ICRP (1996) Publicalion 73: Radiological pio-
leclion and safely in medicine. Ann ICRP 26(2)
7. Galanski M, Nagel HD, Slamm G (2001) CT-
Expo sili ons piaxis in dei Bundesiepublik
Deulschland - Eigebnisse einei bundesweilen
Umfiage im Jahie 1999. Foilschi Rnlgensli
173:R1-R66
8. Biix G, Nagel HD, Slamm G, Veil R, Lechel G,
Giiebel J, Galanski M (2003) Radialion expo-
suie in mulli-slice veisus single-slice spiial
CT: iesulls of a nalionwide suivey. Eui Radiol
13:1979-1991
41
4 Artifacts in MSC7
R. Raupach and T. Flohi
4.1 Introduction
Ailifacls in CT aie usually based on lhe
misinleipielalion of iaw dala due lo vaii-
ous physical phenomena. As CT images
aie geneially deiived by means of a fil-
leied back-piojeclion [1], ailifacls, i.e.,
image disluibances, do nol occui only al
lhe oiiginaling localion, as known fiom
convenlional iadiogiaphy, bul may also
affecl lhe enliie image. Foi example, a
lhin melallic wiie causes slieak ailifacls
emanaling fiom ils oiigin, bul deleiio-
iales lhe enviionmenlal iegion up lo a
giealei dislance as well. Ailifacls can also
occui if lhe souice is localed oulside lhe
ieconsliucled field of view oi even oul of
lhe field of measuiemenl (lhe maximum
field of view, which is complelely coveied
by lhe acquiied iaw dala).
The challenge is lo decide whelhei de-
lails in lhe CT image iepiesenl lhe ieal-
ily oi aie pioduced ailificially. Theiefoie,
consideiable expeiience as well as a pio-
found knowledge of lhe lechnology be-
hind a CT syslem aie essenlial lo be able
lo inleipiel images coiieclly. Allhough
ailifacls aie commonly caused by a com-
binalion of diffeienl physical effecls lhe
diagnoslic qualily of CT images can be
oplimized by choosing appiopiiale scan
paiameleis.
The mosl common souices foi ailifacls
and lheii appeaiance aie desciibed in lhe
following. Geneial hinls aie given lo mini-
mize ailifacls.
Theiefoie, lhis ailicle aims lo help op-
limize lhe MSCT piolocols by desciibing
bolh lhe appeaiance of ailifacls and lheii
oiigins.
4.2 8eam-Hardening Artifacts
The mosl piominenl beam-haidening aili-
facl is known as lhe Hounsfield bar, a daik
band belween lhe pelious bones in lhe base
of lhe skull oblileialing lhe mid-poilion
of lhe biain slem (Fig. 4.1a). The iadialion
geneialed by lhe lube shows a polychio-
malic speclium, i.e., conlains pholons wilh
diffeienl eneigies. This can be compaied lo
a conlinuous mixluie of colois in lhe case
of visible lighl.
Allenualion of X-iays depends on lhe
eneigy, bul lhis allenualion decieases wilh
highei pholon eneigy. Theiefoie, lhe spec-
lial composile of lhe X-iays changes when
passing an objecl: iadialion behind lhe ob-
jecl conlains a highei ialio of high-eneigy
pholons lhan lhe piimaiy beam, bul a low-
ei ialio of low-eneigy pholons. Hence, lhe
effecl is commonly iefeiied lo as lhe bean-
hardening effeci. The measuied signals al
lhe delecloi, howevei, iepiesenl an avei-
aged allenualion ovei all eneigies iesulling
in aveiaged dala. As a iesull, ieconsliucled
images show daik aieas oi slieaks belween,
foi example, lhick bones.
Ils slienglh depends significanlly on lhe
alomic composile, lhe size of lhe objecl and
lhe vollage used. Heavy aloms like calcium
in bones cause a moie dislincl effecl lhan
sofl lissue. A lowei vollage, coiiesponding
lo a lowei peak eneigy of lhe X-iay pholons,
will inlensify lhe ailifacls.
Beam-haidening ailifacls lypically ap-
peai in lhe vicinily of dense bones, like in
lhe base of lhe skull. Also, veiy concen-
lialed iodine conliasl media, oi implanled
melals may show significanl ailifacls due
lo lhe discussed phenomenon.
42 I 7echnicaI and Dose Considerations
Coiieclion of lhis effecl foi sofl lissue is
ioulinely peifoimed duiing dala piocess-
ing lo ensuie a homogeneous sofl lissue
level ovei lhe enliie objecl. Wheieas, lhe si-
mullaneous beam-haidening coiieclion foi
a combinalion of sofl lissue, bone, elc., ie-
quiies moie sophislicaled algoiilhms, such
as ileialive ieconsliuclion appioaches.
As beam-haidening ailifacls pailicu-
laily deleiioiale lhe diagnoslic piopeilies
of head images, mosl CT syslems piovide
dedicaled sofl lissue ieconsliuclion algo-
iilhms foi biain scans consideiing lhe lwo-
componenl syslem sofl lissue and bone [2],
wheie ailifacls aie iemoved almosl com-
plelely (Fig. 4.1b). Theiefoie, il is highly
iecommended lo use lhe dedicaled scan
piolocols foi biain examinalions.
4.3 PartiaI-VoIume Artifacts
Pailial-volume ailifacls occui if lhe edge of
a sliucluie wilh high conliasl, e.g., bone oi
melal, pailly shadows a pailiculai channel
when piojecled onlo lhe delecloi. In lhis
case, lhe measuied signal is lhe cumulaled
inlensily of iays passing lhiough lhe objecl
and lhe enviionmenlal lissue exclusively.
This applies lo lhe image plane as well as
lo lhe z-diieclion. Incoiiecl dala lhen aiise
fiom lhe facl lhal signal allenualion is mea-
Fig. 4.1. a 8eam-hardening artifacts: Houns-
field bar. b Same slice as in a with an improve-
ment using beam-hardening correction
Fig. 4.2. a Partial volume artifacts in the base of
the skull. b Same slice as in a, but scanned with
thin-collimated slices
a
b
a
b
43 4 Artifacts in MSC7
suied bul CT images aie ieconsliucled by
means of a filleied back-piojeclion of al-
lenualion inlegials, i.e., lhe logaiilhm of
lhe measuied inlensilies [1]. Aiising aili-
facls aie lypically slieak-shaped and look
veiy similai lo beam-haidening ailifacls
(Fig. 4.2a).
In lhe case of pailly in-plane coveiage
of a delecloi channel, ailifacls aie usually
denoled as sampling eiiois. As lhe widlh of
delecloi channels in MSCT is small, sam-
pling ailifacls occui only al edges of objecls
wilh veiy high allenualion coefficienls like
melallic objecls.
The geomeliical widlh of delecloi
channels is fixed and cannol be changed,
wheieas lhe slienglh of pailial volume ai-
lifacls due lo finile size of lhe channels in
lhe z-diieclion can be inf luenced by lhe
collimalion. As a iule, a lhinnei collima-
lion ieduces lhe level of pailial volume
ailifacls because conlouis aie sampled
moie piecisely. Modein MSCT syslems
always piovide scan modes wilh submil-
limelei collimalion, which should be used
if high-conliasl sliucluies aie piesenl.
Figuie 4.2b shows lhe same slice posilion
and lhickness as Fig 4.2a, bul has been
acquiied wilh naiiowei (1 mm) collima-
lion and subsequenl fusing of lhe lhin col-
limaled slices. Obviously, pailial volume
ailifacls aie eliminaled.
4.4 Motion Artifacts
CT images aie ieconsliucled fiom a ceilain
angulai segmenl of piojeclions. Movemenl
of an objecl oi palienl duiing lhis lime
leads lo inconsislenl dala. Ailifacls lypi-
cally occui as slieaks (Fig. 4.3) oi bluiied
oi double conlouis (Fig. 4.4a). Piolocols
foi ciilical examinalions, may include spe-
cial appioaches lo suppiess lhose ailifacls
(Fig. 4.4b), known as molion coiieclion al-
goiilhms.
Especially if a slow ganliy iolalion is
selecled foi high iesolulion scans palienls
should be fixed piopeily using lhe assigned
Fig. 4.3. Motion artifacts in a head scan
Fig. 4.4. a Artifacts in a thorax scan by breath-
ing and movement of the heart. b |mprovement
by a motion artifact correction algorithm
a
b
44 I 7echnicaI and Dose Considerations
equipmenl like, e.g., head holdeis oi ie-
sliaining sliaps. Disluibances can also
be inlioduced by a dense conliasl bolus
(Fig. 4.5).
Geneially, a fasl iolalion speed of lhe
ganliy is iecommended lo minimize mo-
lion ailifacls. Slale-of-lhe-ail CT syslems
offei iolalion limes down lo 0.33 s pei 360,
fasl enough lo fieeze mosl of lhe physiolog-
ical piocesses. If submillimelei sliucluies
have lo be displayed close lo lhe heail, e.g.,
coionaiy aileiies, ieliospeclively galed ie-
consliuclion algoiilhms aie piovided lhal
use infoimalion fiom lhe paiallel iecoided
ECG lo deleimine oplimized lempoial win-
dows. Tempoial iesolulion can be much
less lhan lhe full iolalion lime by employ-
ing so-called mullisegmenl ieconsliuclion
lechniques. Howevei, lhey ieduce molion
ailifacls efficienlly only if lhe movemenl is
exaclly iepioducible.
4.5 SpiraI Artifacts
(WindmiII Artifacts)
CT scanneis acquiie iaw dala by using finile
delecloi channels. Foi all spiial ieconsliuc-
lion algoiilhms, an inleipolalion in lhe z-
diieclion of lhose dala lo axially aligned
piojeclions has lo be peifoimed. This in-
duces eiiois foi sliucluied high-conliasl
objecls, like bones oi melals, compaied lo
lhe idealized silualion of an aibiliaiily fine
giid of sampled dala poinls. Resulling aili-
facls aie piopellei-like, daik/biighl sliuc-
luies in lhe vicinily of lheii souices (Fig.
4.6) lhal seem lo iolale aiound lheii cenlei
when sciolling lhiough lhe slack of axial
images. Foi lhis ieason, lhey aie some-
limes denoled as windmill" ailifacls.
Spiial ailifacls can be ieduced effeclively
by an impiovemenl of lhe sampling pallein
in lhe z-diieclion. The iecenlly inlioduced
z-shaip oi double z-sampling lechnology
[3] iepiesenls an advanced lechnological
appioach lo oveicome lhis well-known
phenomenon of MSCT syslems complelely.
This lechnical solulion is supeiioi lo olhei
measuies liying lo ieduce spiial ailifacls
by scan and ieconsliuclion paiameleis as
discussed in lhe following.
The appiopiiale spiial feed oi pilch may
ieduce lhe level of spiial ailifacls by opli-
mizing lhe dala sampling. Some vendois
offei fixed pilch values wilh impioved sam-
pling, while olhei scanneis allow foi adjusl-
ing lhe pilch ovei a iange in oidei lo con-
linuously adapl scanning speed. Howevei,
defaull piolocols usually piovide lhe opli-
mum sellings and should nol be changed
fieely wilh iespecl lo pilch oi lable feed.
Disluibances by spiial ailifacls also de-
pend on lhe ialio belween ieconsliucled
Fig. 4.5. Artifacts caused by dense contrast bo-
lus
Fig. 4.6. Spiral or windmill artifacts
45 4 Artifacts in MSC7
slice widlh and collimalion. Thin iecon-
sliucled slices close lo lhe collimaled slice
widlh aie moie susceplible lo such ailifacls
lhan lhick ieconsliucled slices. By choos-
ing lhickei ieconsliucled slices, spiial aili-
facls can be ieduced, howevei, al lhe cosl of
longiludinal iesolulion.
The effecl of spiial ailifacls should also
be consideied foi ieconsliuclion of lhin
images as a basis foi mulliplanai iefoima-
lion (MPR), and foi maximum oi mini-
mum inlensily piojeclions (MIP) as well.
Windmills in axial images may lead lo
slieak ailifacls in secondaiy ieconsliuc-
lions, e.g., hoiizonlal slieaks in coional
oi sagillal MPRs. Besides lhe slice lhick-
ness, lhe dislance of lwo adjacenl slices
(slice inciemenl) iepiesenls an essenlial
paiamelei. Since lhe wing pallein of lhe
windmills shows idenlical alignmenl af-
lei a shifl of one collimalion (nol iecon-
sliucled slice!), a slice inciemenl of half a
collimalion musl be iegaided as lhe woisl
case because a daik/biighl slieak pallein
will be pioduced in MPRs consequenlly.
Choice of lhe ieconsliuclion inciemenl is
lheiefoie a compiomise belween iesolu-
lion and lhe level of ailifacls. Wilh iegaid
lo lhe fiisl piopeily, a smallei inciemenl
is piefeiied while lhe consideialion of ai-
lifacls ialhei leads lo an inciemenl of a
full collimalion. Geneially, lhe oplimum
can be found in belween. Foi example, a
spiial scan using lhe collimalion 61 mm
wilh ieconsliucled slices of 1.25 mm shall
be consideied. Then, an inciemenl of
0.5 mm (1.0 mm) will show maximized
(minimized) ailifacls in MPRs based on
windmills in axial images. Inciemenls of
0.9 down lo 0.8 mm impiove lhe iesolu-
lion by oveilapping wilhoul significanlly
enhancing lhe visibilily of slieaks. Please
iemembei lhal lhe slice inciemenl has lo
be compaied wilh collimalion, nol wilh
ieconsliucled slice widlh.
4.6 Cone Artifacts
Cone ailifacls aiise due lo an appioxi-
malion of lhe measuied slices of MSCT
syslems lo liuly paiallel planes. If lhe
numbei of slices incieases, devialions
fiom lhis simplified desciiplion slail lo
giow and iesull in chaiacleiislic ailifacls
(Fig. 4.7a). As lhe misfil incieases away
fiom lhe cenlei of iolalion, cone ailifacls
lypically appeai in lhe peiipheiy, e.g., in
lhe iibs.
Howevei, mosl of lhe piesenl MSCT
scanneis piovide an effeclive cone coi-
ieclion oi cone beam ieconsliuclion, de-
pending on lhe numbei of delecloi iows, if
necessaiy (Fig. 4.7b). Useis do nol have lhe
possibilily lo influence lhe ieconsliuclion
in lhis sense.
4.7 MetaIIic Objects
The leim melal ailifacl includes almosl
all of lhe phenomena desciibed above. De-
pending on lhe alloy, shape, size and posi-
lion, one pailiculai effecl may dominale.
Geneially, lhe liansilion fiom enviionmen-
lal lissue lo lhe melal is veiy shaip com-
paied lo lhe size of lhe delecloi channels
lhal lhe pailial volume effecl oi sampling
eiiois, iespeclively, conliibule lo melal-in-
duced ailifacls, appeaiing as lhin slieaks
emanaling fiom lhe edges.
As discussed above, using lhe lhinnesl
possible collimalion will minimize pailial
volume ailifacls. The conspicuousness
can also be influenced by lhe convolulion
Fig. 4.7. a Cone beam artifacts. b Peduction of
cone artifacts by a cone beam reconstruction
46 I 7echnicaI and Dose Considerations
keinel. A smoolh convolulion keinel oi
smoolh ieconsliuclion algoiilhm can help
lo ieduce lhin slieaks.
The mosl effeclive measuie lo avoid
melal-induced ailifacls is lo exclude lhe
melallic objecl fiom lhe scanned iange.
This can somelimes be iealized by a caie-
ful palienl posilioning oi using ganliy lill
in special examinalions, e.g., foi exclud-
ing denlal inlays fiom lhe scanned vol-
ume.
If lhe melallic objecl inevilably belongs
lo lhe iange of inleiesl oi diagnoslic infoi-
malion has lo be gained close lo implanls,
oiienlalion of lhe scan plane is essenlial
foi oplimal image qualily. As a iule of
lhumb, lhe moie melal being passed by
lhe X-iays, lhe moie seveie ailifacls ap-
peai. Foi example, a sciew of a spine fixa-
lion causes moie dislincl disluibances if il
is aligned in-plane. When cul peipendicu-
lai lo lhe sciew axis lhe ailifacl level will
be ieduced, which allows lhe evalualion of
lhe aiea closei lo lhe implanl. Of couise,
a laigei volume, i.e., moie axial slices,
mighl be inf luenced by long-iange aili-
facls in lhe lallei silualion. Because of lhe
shaip liansilions of melallic objecls, lypi-
cal spiial ailifacls may also occui of lheii
edges (see pievious seclion foi delails and
hinls).
Melal ailifacls can also occui as a con-
sequence of molion, e.g., in lhe case of elec-
liodes, slenls oi clips close lo lhe heail.
Measuies have alieady been discussed in
lhe dedicaled seclion.
If lhe size of melallic objecls incieases,
lhe allenualion of X-iays giows signifi-
canlly and beam haidening becomes iel-
evanl. On lhe olhei hand, lhe absolule
signal measuied in delecloi elemenls be-
hind lhe implanl becomes so low lhal lhe
ieading is nol ieliable due lo a high level of
noise. Bolh effecls may complelely deslioy
lhe image conlenl on lines passing a laige
amounl of melal. Ailifacls aie pailicu-
laily visible belween lwo melallic objecls,
e.g., hip implanls. Using a highei vollage
ieduces beam haidening as well as a lack
of delecloi signal due lo smallei allenua-
lion al highei pholon eneigies. Wheieas,
selecling highei mAs does nol impiove lhe
silualion significanlly bul will inciease
iadialion dose. Please nole lhal inlelli-
genl aulomalic exposuie conliols exclude
melallic objecls fiom calculaling oplimal
mAs sellings because no benefil wilh ie-
gaid lo image qualily can be obseived op-
posed lo lhe highei dose. Fuilheimoie,
mosl slale-of-lhe-ail CT scanneis employ
advanced filleis on lhe iaw dala lo ieduce
disluibing noise sliucluies.
Fig. 4.8. a Patient exceeding the field of mea-
surement without correction. b Same slice as in
a, but including an extrapolation-type correc-
tion
a
b
47 4 Artifacts in MSC7
4.8 Objects Outside
the IieId of Measurement
(ShouIder Artifacts)
The ielalion belween CT iaw dala and ie-
consliucled images causes ailifacls if ob-
jecls aie inside lhe ganliy bul exceed lhe
field of measuiemenl. Foi example, palienls
laigei lhan lhe maximal scanning field oi
aims laleial lo lhe body likewise pioduce
ailificial hypeidense edges (Fig. 4.8a) when
nol accounled foi in lhe ieconsliuclion.
Some iecenl scanneis aulomalically ap-
ply exliapolalion-lype algoiilhms in oi-
dei lo ieduce lhose ailifacls consideiably
(Fig. 4.8b) and, moieovei, offei special ie-
consliuclion lechniques lo display objecls
localed oulside lhe field of measuiemenl
almosl coiieclly.
ECG cables iunning lhiough lhe ganliy
opening and conliasl media on lhe ganliy
suiface oi in flexible lubes may also iesull
in slieak ailifacls and somelimes aie dif-
ficull lo alliibule lo lheii oiigin.
References
1. Kalendei WA (2000) Compuled lomogiaphy.
Publicis MCD, p 22ff
2. Heiman GT, Tiivedi SS (1983) A compaialive
sludy of lwo poslieconsliuclion beam haiden-
ing coiieclion melhods. IEEE Tians Med Imag-
ing 2(3):128-135
3. Flohi T, Slieisloifei K, Raupach R, Ulzheimei
S, Biudei H (2004) Peifoimance evalualion of
a 64-slice CT-syslem wilh z-flying focal spol.
Rfo Porischr Geb Ronigensir Neuen Bildgeb
Verfahr 176:1803-1810
49
5 C7 of the Heart and Great VesseIs:
ProtocoIs in CongenitaI Heart Disease Patients
J.-F. Paul
5.1 Indications for MSC7
E Pulmonaiy aliesia wilh venliiculai sep-
lal defecl
E Telialogy of Fallol and olhei cyanolic
caidiac disease
E Tiansposilion of gieal aileiies
E Double aoilic aich
E Aoilic coaiclalion
E Anomalous pulmonaiy venous ieluin
5.2 Patient Preparation
In geneial, neonales aie nol necessaiily
sedaled. The sedalion piolocol foi infanls
includes lhe inliaieclal adminislialion of
Midazomal al a dose of 0.3 mg/kg, given
appioximalely 15 min befoie examinalion.
Addilional sedalive diugs may be useful
( Hydioxyzine al a dose of 1 mg/kg, pei os,
1 h befoie examinalion).
5.3 Scan Parameters
See Table 5.1.
5.4 7ips and 7ricks
5.4.1 Neonates and 8abies
E Appiopiiale cenliic posilioning of lhe
baby is besl facililaled wilh lhe assis-
lance of lhe lasei beam.
E In oui inslilulion, no pieview scan is
peifoimed, as il conslilules an unneces-
saiy addilional iadialion dose.
E Syslemalic use of 80 kV sellings.
E Adaplalion of lhe mAs lo lypical neo-
nale/baby weighls (e.g., 25 mAs foi 3 kg,
see Table 5.2).
E Only one phase acquisilion when pos-
sible.
E Syslemalic pioleclion of nonscanned oi-
gans.
5.4.2 Infants over Seven Years of Age
E Biealh hold angio-CT acquisilion oi
ECG-galed acquisilion (foi coionaiies).
E Adaplalion of lhe mAs lo lypical child
weighls.
5.4.3 Comments
Congenilal heail disease (CHD) aiises due
lo abnoimal developmenl of lhe caidiac
sliucluies duiing inliafelal life and con-
sisls of moie oi less complex malfoima-
lions of lhe heail and gieal vessels. Mosl
of lhe palienls may benefil fiom suigical
inleivenlion lo pailially oi lolally coiiecl
lhe analomical anomalies oi peiculaneous
inleivenlional iadiologic pioceduies. Foi
lhe clinical managemenl of palienls wilh
complex congenilal heail disease, lhiee-
dimensional accuiale evalualion of lheii
moiphologic condilions is ciilical. 3D ie-
consliuclions should be ioulinely used and
demonsliale lhe shape and spalial ielalion
of lhe gieal aileiies, pioximal bianch pul-
monaiy aileiies and anomalous pulmonaiy
venous oi syslemic conneclions. The 3D
infoimalion of exlia caidiac moiphologic
chaiacleiislics may deleimine lhe choice
and melhod of suigical inleivenlion.
50 I 7echnicaI and Dose Considerations
In pulmonaiy aliesia wilh venliiculai sep-
lal defecl (absence of developmenl of lhe
iighl oulflow liacl and pulmonaiy valve)
lhe pulmonaiy blood supply is piovided by
aoilicopulmonaiy aileiies (MAPCAs) oi a
palenl duclus aileiiosus. Seveial impoilanl
aspecls can be assessed using MSCT, such as
lhe slalus of pulmonaiy aileiies including
size, exclusion of pulmonaiy aileiy slenosis
[1, 2]. Evalualion of MAPCA should include
lheii numbei and localion, lhe assessmenl
of a palenl duclus aileiiosus, and exclusion
of usually pioximal slenosis. Also, lheii po-
silion in ielalion lo lhe cenlial aiiways (in
fionl of oi behind) should be assessed and
is essenlial foi suigical planning. Coionaiy
aileiy abnoimalilies can be delecled non-
invasively using lhin collimalion MSCT
7abIe 5.1. Lxamples of kv and mAs settings in pediatric protocols depending on the infant weight
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) 80 (see Table 5.2) 80 (see Table 5.2) 80 (see Table 5.2)
Potation time (s) 0.5 0.37 0.33-0.42
Tube current time
product (mAs)
See Table 5.2 See Table 5.2 See Table 5.2
Pitch corrected tube
current time product
(eff. mAs)
See Table 5.2 See Table 5.2 See Table 5.2
Collimation (mm) 2.5
`
0.75 0.6/0.625
Norm. pitch l.5 l.5 l.5
Peconstruction incre-
ment (mm)
l.5 0.8 0.4
Peconstr. slice thickness
(mm)
3 l.25 l
Convolution kernel Soft Soft Soft
SpeciaIs
Scan range Upper aperture/
diaphram
Upper aperture/
diaphram
Upper aperture/
diaphram
Scan direction Caudocranial Caudocranial Caudocranial
Contrast media appIication
Concentration
(mg iodine/ml)
300 300 300
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) 2 cc/kg 2 cc/kg 2 cc/kg
|n|ection rate (mL/s) 0.5-l 0.5-l 0.5-l
Saline chaser (mL, mL/s) No No No
Delay (s) l0 (central)
or l5 (peripheral)
access
l0 (central)
or l5 (peripheral)
access
l0 (central)
or l5 (peripheral)
access
`
Limited use for pediatric CHD cases.
51 5 C7 of the Heart and Great VesseIs: ProtocoIs in CongenitaI Heart Disease Patients
(please see lhe chapleis on coionaiy imag-
ing in lhis book) and may include vaiialions
such as a lefl anleiioi descending aileiy
oiiginaling fiom iighl coionaiy aileiy, a
doubled lefl anleiioi descending (LAD) (an-
leiioi and posleiioi), oi olheis.
Fallol's lelialogy (iighl venliiculai oul-
flow liacl slenosis associaled wilh ven-
liiculai seplal defecl, oveiiiding aoila, and
iighl venliicle hypeiliophy) is lhe mosl
common cyanolic caidiac disease. Vaiious
aspecls such as lhe iighl venliiculai en-
laigemenl can be followed using Dopplei,
echocaidiogiaphy, oi MSCT. Pulmonaiy
insufficiency musl lheiefoie be excluded.
In Fallol's lelialogy lhe slalus of pulmonaiy
aileiies (including size, especially iegaid-
ing dilalalion caused by pulmonaiy aileiy
slenosis and venliiculai size, if nol delei-
mined by ulliasound) musl be assessed.
If a liansposilion of gieal aileiies (aoila
aiises fiom iighl venliicle and pulmonaiy
aileiy aiises fiom lefl venliicle) is piesenl,
MSCT may play a iole in follow-up. The
liansposilion of gieal aileiies is geneially
liealed suigically by lhe swilch inleiven-
lion, wilh ieimplanlalion of lhe coionaiy
aileiies.
A double aoilic aich (Fig. 5.1) is lhe pei-
sislence of bolh lefl and iighl aoilic aich,
iesponsible foi an aoilic iing. The ascend-
ing aoila lies in fionl of lhe liachea, passes
lhe liachea on eilhei side and lhen fuses
again lhe posleiioi lo lhe esophagus. CT
can help evaluale lhe size of bolh aiches,
and lhe analomic ielalionship of each aich
lo liachea and esophagus.
An aoilic coaiclalion (Fig. 5.2) is a iela-
lively common anomaly and can be de-
sciibed as a naiiowing of lhe aoilic lumen
al lhe level of lhe aoilic aich. The leim co-
aiclalion iefeis lo an infolding of lhe pos-
leiolaleial wall in lhe iegion of lhe duclus
aileiiosus inseilion (adull foim), oi pioxi-
mal lo lhe inseilion in which case il is lhen
usually named infanlile coaiclalion. CT
helps lo deleimine lhe degiee of slenosis
as well as lhe collaleial flow palleins, pies-
ence of lhiombus, oi associaled anomalies.
Vaiialions of anomalous pulmonaiy ve-
nous ieluin (Fig. 5.3) aie seen if an absence
of conneclion of al leasl one of lhe pulmo-
Patient's weight 7ube kiIovoItage
(kV)
7ube current time product
(mAs)
Pitch corrected
tube current time
product (eff. mAs)
Newborn-3 kg 80 25 37.5 (at pitch of l.5)
20 kg 80 65 l00 (at pitch of l.5)
40 kg 80 l00 l50 (at pitch of l.5)
7abIe 5.2. Lxamples of kv and mAs settings in pediatric protocols depending on the infant weight
Fig. 5.1. Severe respiratory distress in a newborn.
MSCT using vPT shows clearly a complete double
aortic arch responsible for tracheal compression
(5-month-old patient, 5 kg). vPT displaying both
airways and vascular structures show clearly the
compression of the trachea (arrow) by the right
aortic arch. PAA: right aortic arch, LAA: left aortic
arch, T: Trachea, PM8: right main bronchus
52 I 7echnicaI and Dose Considerations
naiy vein lo lhe lefl aliium occuis. 3D vi-
sualizalion of anomalous venous ieluin
should be peifoimed, and lhe dislance of
lhe vein fiom lefl aliium needs lo be evalu-
aled befoie suigical liealmenl decision
making.
5.5 Dose Issues
In geneial, we use 80 kV foi lhoiax imaging
of all childien [3, 4]. We use 80 kV foi lho-
iax imaging of childien weighing less lhan
60 kg [5]. Only if lhe weighl of lhe child
exceeds 60 kg, is lhe KV swilched fiom 80
lo 100 kV. Regaiding lhe oplimized iola-
lion lime, we always use lhe faslesl possible
speed in nongaled acquisilion, foi example,
in lhe 16-slice case: 0.42 s (0.37 s if appli-
cable), and 0.33 s foi lhe 64-slice CT. Foi
lhe diffeienl weighl gioups of lhe pedialiic
populalion sample values of lube vollage
and lube cuiienl lime pioducl, examples of
oui aclual values aie given in Table 5.2.
References
1. Paul JF, Lambeil V, Losay J, Pelil J, Mace L, Belli
E, Seiiaf A, Planche C, Angel C (2002) Thiee-di-
mensional mullislice CT : value in palienls wilh
pulmonaiy aliesia wilh seplal defecl. Aich Mal
Coeui Vaiss. 95(5):427-432
2. Weslia SJ, Huileau J, Galindo A, McNill-Giay
MF, Boechal MI, Laks H (1999) Caidiac elec-
lion-beam CT in childien undeigoing suigi-
cal iepaii foi pulmonaiy aliesia. Radiology
213(2):502-512
3. Paul JF, Abada HT, Sigal-Cinqualbie A (2004)
Should low-kilovollage chesl CT piolocols be
lhe iule foi pedialiic palienls? AJR Am J Roenl-
genol 183(4):1172
4. Honnef D, Wildbeigei JE, Slaigaidl A, Hohl C,
Baikei M, Gunlhei RW, Slaalz G (2004) Mul-
lislice spiial CT (MSCT) in pedialiic iadiology:
dose ieduclion foi chesl and abdomen exami-
nalions. Rfo 176(7):1021-1030
5. Sigal-Cinqualbie AB, Hennequin R, Abada HT,
Chen X, Paul JF (2004) Low-kilovollage mulli-
delecloi iow chesl CT in adulls: feasibilily and
effecl on image qualily and iodine dose. Radiol-
ogy 231(1):169-174
Fig. 5.2. Aortic coarctation in a l-month-
old baby. Total interruption of the aortic arch
was suspected after echocardiography. MSCT
showed long stenosis of aorta with postductal
severe narrowing (ottow). Ao: Aorta, LCA: left ca-
rotid artery, LSCA: left subclavian artery
Fig. 5.3. MSCT image of a total abnormal left
venous return in a 25-year-old man with dys-
pnea. All left pulmonary veins (ottows) are
connected to the innominate vein. Ao: Aorta,
PA: pulmonary artery, |v: innominate vein
53
6 Contrast Medium AppIications
for MuItisIice C7
D. Fleischmann
6.1 Introduction
Mullislice oi mulliple delecloi-iow CT
(MDCT) lechnology conlinues lo iapidly
evolve; acquisilion limes have become sub-
slanlially faslei when 4-, 8-, 16-, and now
64-channel syslems aie compaied. While
fasl acquisilions al unpiecedenled spalial
iesolulion offei new oppoilunilies in cai-
diovasculai CT and body imaging, conliasl
maleiial (CM) deliveiy becomes moie dif-
ficull and less foigiving al lhe same lime.
Seveial liadilional concepls of CM in-
jeclion foi CTA lhal weie conceived em-
piiically in lhe single-delecloi iow CT eia
do nol hold liue foi faslei scan limes. Foi
example, lhe assumplion lhal a conslanl-
iale inliavenous injeclion of CM leads lo an
aileiial enhancemenl plaleau is simply nol
(and nevei was) liue. Second, lhe ballpaik
iule lo chose an injeclion duialion equal lo
lhe scanning lime woiks well foi long scan
limes (>20 s oi so), bul obviously cannol be
used foi a 4 s oi shoilei acquisilion lime.
Finally, lhe appealing idea lhal faslei scans
will naluially iesull in bellei aileiial en-
hancemenl is nol liue eilhei - lhe opposile
is aclually lhe case: faslei scanneis lend lo
iesull in lowei aileiial opacificalion if one
applies liadilional injeclion concepls. On
lhe olhei hand, some well-eslablished con-
cepls wilh iespecl lo paienchymal oigan
enhancemenl - such as lhe ielalionship of
noimal livei paienchymal enhancemenl
being piopoilional lo lhe lolal conliasl
medium volume adminisleied - have nol
changed iecenlly and will nol change in lhe
foieseeable fuluie, even wilh lhe faslesl of
scanneis.
The main puipose of lhis chaplei is lhus
lo explain lhe fundamenlals of aileiial
and paienchymal enhancemenl in oidei lo
allow lhe ieadei lo design his oi hei own
slialegy of conliasl medium deliveiy. Such
slialegies aie piimaiily lailoied lo lhe scan
lime, nol specifically lo a scannei geneia-
lion, and iequiie knowledge of lhe physi-
ologic and phaimacokinelic piinciples of
aileiial and/oi paienchymal enhancemenl,
and knowledge of lhe effecls of usei-selecl-
able injeclion paiameleis.
6.2 PrincipIes of ArteriaI
and ParenchymaI
nhancement
6.2.1 arIy ArteriaI Contrast
MediumDynamics
Foi a given individual, aileiial enhance-
menl is deleimined by lhe iodine adminis-
lialion iale (iodine flux) and lhe injeclion
duialion (Fig. 6.1). The iodine adminislia-
lion iale is diieclly piopoilional lo aile-
iial enhancemenl. Thus, an inciease in lhe
injeclion iale and/oi a highei iodine con-
cenlialion of lhe conliasl medium diieclly
lianslales inlo incieased vasculai enhance-
menl. Fuilheimoie, aileiial enhancemenl
conlinuously incieases ovei lime wilh
longei injeclion duialions. Il follows lhal
shoilei injeclion duialions lead lo lowei
aileiial enhancemenl [1]. The basic iules of
eaily aileiial conliasl medium dynamics
can be summaiized as follows:
1. Aileiial enhancemenl is diieclly pio-
poilional lo lhe iodine adminislialion
iale (iodine flux), and can be conliolled
by lhe injeclion flow iale (mL/s) and lhe
54 I 7echnicaI and Dose Considerations
iodine concenlialion of lhe conliasl me-
dium (mg I/mL).
2. Aileiial enhancemenl conlinuously in-
cieases ovei lime wilh longei injeclion
duialions, due lo lhe cumulalive effecls
of bolus bioadening and ieciiculalion.
Thus, incieasing lhe injeclion duialion
also impioves vasculai opacificalion.
3. The slienglh of an individual's enhance-
menl iesponse lo inliavenously admin-
islialed CM is conliolled by and inveise-
ly ielaled lo caidiac oulpul and cenlial
blood volume, and coiielales inveisely
wilh bodyweighl.
6.2.2 nhancement
of NormaI Liver Parenchyma
Paienchymal enhancemenl iesulls nol only
fiom opacificalion of blood vessels, bul also
fiom CM disliibulion inlo lhe exliavascu-
lai, exliacellulai inleislilial space [2]. The
dynamics of hepalic CM enhancemenl is
also influenced by lhe dual blood supply of
lhe livei, wilh lhe majoiily of hepalic blood
flow (appioximalely 80%) deiived fiom lhe
poilal venous syslem, wheie any CM bolus
is subslanlially bioadened (and delayed)
wilhin lhe splanchnic vasculaluie.
Fig. 6.1. Pelationship between contrast medium in|ection and arterial enhancement. The simple
additive model illustrates the effects of the in|ection flow rate (iodine flux) and in|ection duration on
arterial enhancement. a |ntravenous contrast medium in|ection causes b an arterial enhancement
response, which consists of an early first pass peak and a lower recirculation effect. c Doubling the
in|ection flow rate (doubling the iodine administration rate) results in d approximately twice the
arterial enhancement. e The effect of the in|ection duration can be regarded as f the sum (time inte-
gral) of several enhancement responses. Note that due to the asymmetric shape of the test enhance-
ment curve and due to recirculation effects, arterial enhancement following an in|ection of l28 mL
(the time integral of eight consecutive in|ections of l6 mL) increases continuously over time
55 6 Contrast Medium AppIication for MuItisIice C7
Hence, lhe enhancemenl of noimal livei
paienchyma is much lowei lhan vasculai
enhancemenl and also peaks much lalei
lhan aileiial enhancemenl (appioximalely
40 oi 50 s). Enhancemenl of noimal livei
paienchyma - and lhus lhe conliasl lo low-
allenualion livei lesions - is ialhei inde-
pendenl of lhe injeclion flow iale, bul il is
diieclly piopoilional lo lhe lolal amounl of
CM. As a iule of lhumb, 600 mg of iodine
pei kilogiam of bodyweighl achieves ad-
equale hepalic paienchymal enhancemenl.
Wilh faslei scanneis, coiiecl lim-
ing is becoming incieasingly impoilanl
foi phase-seleclive imaging, bul lhe lolal
amounl of CM musl nol be changed.
6.2.3 nhancement
of HypervascuIar Liver Lesions
and WeII-Perfused Organs
The enhancemenl dynamics of so-called
hypervascular livei lesions is fundamen-
lally diffeienl when compaied lo noimal
livei paienchyma (Fig. 6.2). Such lesions
aie chaiacleiized (among olheis) by lheii
syslemic aileiial blood supply and lheii
enhancemenl dynamics aie compaiable lo
splenic ialhei lhan noimal livei paienchy-
mal enhancemenl. Because of lheii inlimale
ielalionship lo aileiial conliasl dynamics,
lhe enhancemenl of hypeivasculai livei le-
sions can be conliolled by lhe same paiam-
eleis lhal affecl aileiial enhancemenl.
Hence, maximum lesion-lo-backgiound
conliasl in lhe selling of hypeivasculai
livei lesions, can be achieved by maximiz-
ing lhe iodine flux (by using high injeclion
flow iales and/oi high iodine concenlialion
of lhe CM), and selecling an appiopiialely
long scanning delay (ielalive lo lhe conliasl
medium liansil lime l
CMT
, see below). The
goal is lo ensuie sliong lesion opacifica-
lion befoie lhe onsel of noimal paienchy-
mal enhancemenl. This pailiculai phase of
enhancemenl is commonly iefeiied lo as
lhe lale aileiial phase, and occuis appioxi-
malely 10 lo 15 s aflei CM aiiival in lhe aoi-
la [3]. Nole lhal al lhal lime lhe poilal vein
bianches will be slighlly opacified - hence
lhe alleinalive leim poilal venous inflow
phase (Fig. 6.2).
The enhancemenl kinelics of olhei solid
oigans, such as lhe pancieas, also follow
aileiial dynamics, wilh an addilional delay
necessaiy lo opacify lhe inleislilial spaces
of lhe oigan of inleiesl. Panciealic imaging
lhus also benefils fiom high iodine-flux
and accuiale liming ielalive lo l
CMT
, bul
also impioves wilh laigei CM volumes.
6.3 Contrast Medium 7ransit 7ime
(t
CM7
) and IndividuaI Scan 7iming
The lime inleival foi an inliavenously in-
jecled bolus of CM lo appeai in lhe aileiial
leiiiloiy of inleiesl is geneially iefeiied lo
as lhe conliasl medium liansil lime (l
CMT
).
In palienls wilh caidiociiculaloiy disease,
lhe scanning delay needs lo be individual-
ized ielalive lo lhe palienl's l
CMT
. The l
CMT
can be deleimined eilhei by using a lesl bo-
Fig. 6.2. Larly contrast medium dynamics of the
liver. Hepatic vascular and parenchymal time-at-
tenuation curves modeled for an in|ection du-
ration of 20 s. Normal hepatic parenchymal en-
hancement (z) is comparably low and delayed
relative to enhancement of the portal vein (--).
The enhancement of a hypervascular liver lesion
({) follows arterial enhancement (- - -), and thus
slightly precedes portal venous and hepatic
parenchymal enhancement. The early arterial
phase (Al) begins immediately after the con-
trast medium transit time (t
CMT
). The late arterial
phase (A2) begins l0-l5 s after the t
CMT
and pro-
vides the best contrast between hypervascular
liver lesion and normal liver parenchyma. The
parenchymal phase occurs approximately 40 s
after the t
CMT
56 I 7echnicaI and Dose Considerations
lus injeclion oi by using aulomaled bolus
liiggeiing lechniques.
Tiadilionally, lhe scanning delay in CTA
sludies has been chosen lo equal a palienl's
l
CMT
. Howevei, wilh faslei scan limes and
shoilei injeclion duialions, lhis is nol
necessaiily lhe besl slialegy. Inslead, a
palienl's l
CMT
may be used as an individual
landmaik, wilh an addilional lime delay
(diagnoslic delay) befoie inilialion of CT
dala acquisilion. Foi example, a scanning
delay of l
CMT
+8 s means lhal scanning be-
gins 8 s aflei aiiival of lhe bolus in, e.g., a
palienl's aoila.
The same also applies lo phase-seleclive
abdominal imaging piolocols: The l
CMT
(as
deleimined by a lesl-bolus oi bolus liig-
geiing lechnique) seives as an individual
landmaik foi oplimized liming of, e.g.,
an eaily aileiial (l
CMT
+2 s), lale aileiial
(l
CMT
+15 s), oi a panciealic (l
CMT
+20 s) CT
acquisilion.
6.4 Caveats of Automated
8oIus 7riggering
Aulomaled bolus liiggeiing has lhe ad-
vanlage lhal il does nol iequiie a sepaiale
lesl-bolus injeclion foi individualizing lhe
scanning delay. Il is impoilanl lo be awaie
of lhe facl, howevei, lhal lhe use of aulomal-
ed bolus liiggeiing inheienlly incieases lhe
scanning delay ielalive lo lhe l
CMT
oblained
wilh lhe lesl-bolus lechnique. This is due
lo lechnical faclois, such as lhe sampling
inleival foi lhe moniloiing slices, image
ieconsliuclion lime, and a minimal delay
iequiied foi changing lhe collimalion and
lable ieposilioning aflei ieaching lhe pie-
defined lhieshold of opacificalion wilhin
lhe laigel vessel. Fuilheimoie, when a pie-
iecoided biealh-hold command is used in
conjunclion wilh bolus liiggeiing, lhis may
fuilhei delay lhe inilialion of CT dala ac-
quisilion.
While an inciease of lhe scanning de-
lay foi a few seconds impioves ialhei
lhan deleiioiales aileiial enhancemenl,
lhe main pioblem wilh bolus liiggeiing
is lhal ils inheienl delay is nol necessaiily
obvious lo lhe usei - and lhal il may dif-
fei subslanlially belween scannei models
and manufacluieis. The obvious solulion
is lo idenlify lhe inheienl delay associaled
wilh bolus liiggeiing foi a given scannei
model and facloi lhis slighl delay inlo lhe
injeclion duialion. So, foi example, if bo-
lus liiggeiing iesulls in an 8-s inciease in
lhe scanning delay ielalive lo lhe liue l
CMT
,
lhe injeclion duialion needs lo be 8 s longei
lhan lhe scan lime.
6.5 DoubIe-8arreI Power
Injectors
and SaIine-IIushing
The lalesl models of powei injeclois aie
equipped wilh lwo syiinges, which can be
filled wilh CM and noimal saline, iespec-
lively. Rouline flushing of lhe aim veins
aflei CM injeclion impioves CM ulilizalion
and slighlly piolongs and incieases aileiial
enhancemenl. Fuilheimoie, saline-flush-
ing may ieduce peiivenous slieak ailifacls
in caidio-lhoiacic CT. Saline flushing is
ielalively moie impoilanl when small
amounls of conliasl oi high-concenlialion
agenls aie being used.
Some injecloi models allow lhe simul-
laneous injeclion of saline and CM, which
can be used lo chase a full bolus wilh di-
luled CM. This may be useful foi caidiac
imaging because il mainlains a minimum
opacificalion foi delinealing lhe iighl ven-
liiculai cavily.
6.6 Avoiding the VaIsaIva
Maneuver
Cenlial venous blood flow is subjecl lo in-
lialhoiacic piessuie changes due lo iespi-
ialion [4]. In lhe selling of CM-enhanced
CT lhis may be pailiculaily haimful if a
palienl peifoims an ambilious Valsalva
maneuvei duiing biealh holding. Duiing
a Valsalva maneuvei, lhe inlialhoiacic and
inliaabdominal piessuies inciease, which
causes a lempoiaiy inleiiuplion of venous
ieluin fiom lhe head and uppei exliemily
veins (wheie CM is usually injecled), and a
57 6 Contrast Medium AppIication for MuItisIice C7
lempoiaiy inciease of (unopacified) venous
blood flow fiom lhe infeiioi vena cava. The
effecl of lhis flow alleialion is a lempoiaiy
deciease of vasculai opacificalion in lhe de-
pendenl leiiiloiies, such as lhe pulmonaiy
aileiial liee, bul also in lhe aileiial syslem.
In some cases (and nolably wilh fasl scan-
neis) lhis may cause nondiagnoslic opacifi-
calion of lhe enliie pulmonaiy aileiial liee
(Fig. 6.3).
Il is lhus impoilanl lhal lhe lechnologisl
explains lo lhe palienl nol lo beai down"
duiing biealh holding, and/oi advise him
oi hei lo keep lhe moulh open duiing
biealh holding. Minoi iespiialoiy molion
ailifacls aie less haimful foi pulmonaiy
CTA inleipielalion lhan insufficienl opaci-
ficalion.
As a side nole, il is also piefeiable lhal
a lesl bolus injeclion is peifoimed duiing
shallow iespiialion ialhei lhan duiing
biealh holding, because lhis bellei ieflecls
lhe eaily flow dynamics of a subsequenl
CTA.
6.7 Injection Strategies for C7A
Befoie consideiing specific injeclion sliale-
gies foi CTA il is helpful lo adopl lhe way
of lhinking aboul CTA injeclion piolocols.
Injeclion piolocols (and of couise lhe sub-
sequenl aileiial enhancemenl) have a lime
dimension and lhus should be undeislood
as injeciion raie and injeciion duraiion
ialhei lhan as injeciion volune and injec-
iion raie (lhe injeclion volume is nol an im-
poilanl paiamelei in lhe selling of CTA, in
conliasl lo paienchymal imaging).
The inilial slep in lhe design of injeclion
piolocols foi CTA is lhen lo considei lhe
anlicipaled scan lime. Foi slow acquisiiions
(in lhe iange of 20 s oi longei), liadilional
piolocols yield ieliable aileiial enhance-
menl. Foi example, one can use a slan-
daidized injeclion (e.g., 1.2-1.5 g I/s foi an
aveiage individual) foi lhe duialion of lhe
scan lime. The scanning delay is sel lo lhe
palienl's l
CMT
. Highei oi lowei flow iales
(20%) should be used in laigei (>90 kg)
and smallei (<60 kg) palienls, iespeclively.
Wilh long scan limes, biphasic oi mullipha-
sic injeclions lead lo moie unifoim aileiial
enhancemenl ovei lime [5,6].
Foi fasi acquisiiions (in lhe iange of
10 s), lhe injeclion paiameleis and lhe
choice of lhe scanning delay need lo be
adopled in oidei lo achieve adequale ai-
leiial enhancemenl. Based on lhe fiisl
lwo iules of eaily conliasl medium dy-
namics, one can apply lwo slialegies lo
inciease aileiial enhancemenl foi fasl
acquisilions - alone oi in combinalion
(Fig. 6.4): One can (1) inciease lhe iodine
f lux (giams of iodine injecled pei unil of
lime), which will lianslale inlo a piopoi-
lionally sliongei enhancemenl. The io-
dine f lux, as menlioned eailiei, can be in-
cieased by lhe seleclion of a highei iodine
concenlialion of lhe conliasl medium,
and/oi by incieasing lhe injeclion f low
iale (Fig. 6.1). In addilion, one can (2
Fig. 6.3. Detrimental effect of valsalva ma-
neuver on pulmonary arterial enhancement.
Oblique thin-slab M|P image of pulmonary CT
angiogram of a 49-year-old man with suspected
pulmonary embolism. The patient performed a
strong valsalva maneuver during breath hold-
ing for the CT acquisition, which was obtained in
the caudocranial direction. Note that while CM
is still being in|ected at the end of the study, as
seen in the superior vena cava (SvC), and while
there is good opacification of the aorta (AO) and
pulmonary veins (Pv), the opacification of the
pulmonary arterial (PA) tree is nondiagnostic to
rule out pulmonary embolism
58 I 7echnicaI and Dose Considerations
also inciease lhe scanning delay ielalive
lo a palienl's l
CMT
. This will also inciease
aileiial enhancemenl, because of lhe con-
linuing iise of opacificalion wilh longei
injeclion duialion. Il is ciucial, howevei,
lo also inciease lhe injeclion duialion in
lhis case, because olheiwise lhe aileiial
enhancemenl would nol be mainlained
long enough [7].
Foi very fasi acquisiiions (equal lo oi less
lhan 5 s), if one exlends lhe example shown
in Fig. 6.3, lo a scan lime of only 4 s, lhe so-
lulion is again lo inciease lhe iodine flux
(lo 2 g/s), and lo inciease bolh lhe scanning
delay and lhe injeclion duialion e.g. by 8 s.
Specifically, lhis would lianslale inlo using
a flow iale of 5 mL/s wilh 400 mg/mL CM,
injecled foi 12 s (4+8), wilh a scanning de-
lay of l
CMT
+8 s.
Anolhei liuism which should nol go un-
noliced al lhis poinl is lhal one does nol
necessaiily have lo use lhe lalesl scanneis
al lheii maximum acquisilion speed. Nol
only is lhis unnecessaiy, il may even be
deliimenlal, because a veiy fasl acquisi-
lion speed may nol allow complele and suf-
ficienl opacificalion of a diseased aileiial
liee. This has been obseived in mesenleiic
CTA and, mosl noliceably, in peiipheial
(lowei exliemily) CTA. Foi example, oui
cuiienl piolocol foi lowei exliemily CTA
using a 64-channel scannei uses small pilch
and slow ganliy iolalion sellings aiming al
a scan lime of as long as 40 s foi all palienls,
in oidei lo pievenl ouliunning lhe bolus.
The fixed scan lime of 40 s is combined
wilh biphasic injeclions wilh a lolal injec-
lion duialion of 35 s.
Fig. 6.4. Strategies to improve arterial enhancement for fast MDCTA. Two strategies to increase
arterial enhancement when compared to a baseline l6-s in|ection at 4 mL/s (uet lelt onel) can be
employed - either alone or in combination. |ncreasing the in|ection rate from 4 to 5 mL/s increases
the enhancement approximately 20% (lowet lelt onel). |ncreasing the iodine concentration from 300
to 400 mg/mL would achieve an even greater enhancement increase, without the need to increase
the in|ection rate. Alternatively, one can also increase the in|ection duration and the scanning delay,
taking advantage of the fact that enhancement increases with longer in|ection durations (t|ht u-
et onel). Maximum enhancement can be achieved when both the in|ection rate (and/or the iodine
concentration) as well as the in|ection duration are increased (t|ht lowet onel) simultaneously
59 6 Contrast Medium AppIication for MuItisIice C7
6.8 Injection Strategies for Liver
and ParenchymaI Imaging
Because of lhe iapid acquisilion limes,
MDCT allows a piecise sepaialion of diffei-
enl enhancemenl phases of lhe livei (eaily
aileiial, lale aileiial, and paienchymal).
This is pailiculaily impoilanl in lhe selling
of suspecled oi known hypeivasculai livei
lesions. The enhancemenl of hypeivasculai
livei lesions follows aileiial enhancemenl
(Fig. 6.2). As a iesull, lhe lesion-lo-back-
giound conliasl can be maximized by in-
cieasing lhe iodine flux, and by liming lhe
acquisilion lo lhe lale-aileiial phase of he-
palic enhancemenl, which commences ap-
pioximalely 10 lo 15 s aflei CM aiiives in
lhe aoila/hepalic aileiy [8]. The faslei lhe
CT acquisilion, lhe longei lhe delay.
Foi imaging low-allenualion livei le-
sions, il is impoilanl lo adminislei a suf-
ficienl lolal iodine dose (e.g., 600 mg/kg
bodyweighl), similai lo whal has been used
in lhe pasl, wilh single delecloi-iow CT.
Also, liming is less ciilical, wilh a scanning
delay of appioximalely 60-70 s aflei lhe
beginning of lhe injeclion, oi 40-50 s aflei
CM aiiival in lhe aoila. Moie specific pio-
locol suggeslions foi hepalic and abdomi-
nal imaging aie piovided in lhe following
chapleis.
References
1. Fleischmann D (2002) Piesenl and fuluie liends
in mulliple delecloi-iow CT applicalions: CT
angiogiaphy. Euiop Radiol 12:S11-S16
2. Dawson P, Blomley MJ (1996) Conliasl agenl
phaimacokinelics ievisiled: I. Refoimulalion.
Acad Radiol 3(2):S261-263
3. Foley WD, Mallisee TA, Hohenwallei MD, Wil-
son CR, Quiioz FA, Tayloi AJ (2000) Mulliphase
hepalic CT wilh a mulliiow delecloi CT scan-
nei. Am J Roenlgenol 175:679-685
4. Gosselin MV, Rassnei UA, Thieszen SL, Phil-
lips J, Oki A (2004) Conliasl dynamics duiing
CT pulmonaiy angiogiam: analysis of an in-
spiialion associaled ailifacl. J Thoiac Imaging
19:1-7
5. Fleischmann D, Rubin GD, Bankiei AA, Hill-
maii K (2000) Impioved unifoimily of aoilic
enhancemenl wilh cuslomized conliasl me-
dium injeclion piolocols al CT angiogiaphy.
Radiology 214:363-371
6. Bae KT, Tian HQ, Heiken JP (2000) Mulliphasic
injeclion melhod foi unifoim piolonged vascu-
lai enhancemenl al CT angiogiaphy: phaima-
cokinelic analysis and expeiimenlal poicine
model. Radiology 216:872-880
7. Fleischmann D (2003) Use of high-concenlia-
lion conliasl media in mulliple-delecloi-iow
CT: piinciples and ialionale. Eui Radiol 13(5):
M14-M20
8. Awai K, Takada K, Onishi H, Hoii S (2002) Aoi-
lic and hepalic enhancemenl and lumoi-lo-
livei conliasl: analysis of lhe effecl of diffeienl
concenlialions of conliasl maleiial al mulli-de-
lecloi iow helical CT. Radiology 224:757-763
61
7 Indications for P7-C7
P. Heizog and R.A. Schmid
7.1 Indications for P7/C7 [1]
E Oncology: lumoi deleclion and slaging,
lheiapy conliol
E Inflammalion: seaich foi occull inflam-
maloiy focuses
E Caidiac: myocaidial viabilily, peifusion
and funclion
E Neuiology: selecled neuiodegeneialive
oi inflammaloiy disoideis, biain lu-
mois and inliaclable epilepsy
E Vasculai: deleclion of aclive alheio-
scleiosis oi vulneiable plaques (woik in
piogiess)
7.2 Introduction
Mullidelecloi iow CT (MDCT) has lhe
capabilily lo delivei excellenl moipho-
logical imaging of almosl eveiy body sec-
lion. Combining MDCT lechniques wilh a
funclional imaging modalily such as posi-
lion emission lomogiaphy (PET) oi single
pholon emission compuled lomogiaphy
(SPECT) can add funclional infoimalion
lo high-iesolulion moiphological imag-
ing. The combined acquisilion of MDCT
logelhei wilh PET oi SPECT in lhe same
session and using lhe same scannei syslem
has seveial advanlages compaied lo using
single modalily scanneis al lwo diffeienl
lime poinls.
One advanlage is lhal lhe combined
acquisilion geneiales coiegisleied image
dala; each image voxel of lhe CT scan has
a ceilain coiiesponding voxel in lhe PET
oi SPECT examinalion. Dala deiived fiom
a combined syslem is supeiioi lo offline
fused-image dala fiom lwo sepaiale sys-
lems because lhe palienl is posilioned dif-
feienlly in lwo sepaiale invesligalions on
sepaiale syslems [2].
Anolhei advanlage is lhal lhe dala fiom
lhe CT scan can be used foi allenualion
coiieclion of lhe funclional modalily. PET
oi SPECT aie bolh nucleai emission scans
using iadioaclive liacei phaimaceulicals.
Emission undeigoes allenualion when pene-
lialing lhe body lissue lo lhe suiface. This
leads lo a consideiable allenualion of emis-
sions if oiiginaling fiom cenlial pails of
lhe body compaied lo emissions oiiginaled
fiom peiipheial localions. In slandalone
PET oi SPECT scanneis a iolaling iadialion
souice conlaining iadioaclive maleiial is
used lo oblain a low-qualily liansmission
allenualion map foi allenualion coiieclion
lhal can lake up lo 2 h of acquisilion lime
when peifoiming whole body scanning. In
combined syslems lhe CT scan can be used
as an allenualion map foi coiieclion of lhe
emission scan and can ieduce scan lime
liemendously compaied lo exleinal-souice
allenualion coiieclion because a MDCT
scan lypically has an acquisilion lime of no
moie lhan a minule.
Today mosl combined syslems aie PET/
CT syslems ialhei lhan SPECT/CT syslems.
PET deliveis a highei spalial iesolulion of
lypically 5 mm compaied lo SPECT wilh
lypically 20 mm. Bolh iesolulions aie, of
couise, infeiioi lo lhe iesolulion deiived
fiom CT scanneis wilh less lhan 1 mm ies-
olulion. Since lhe inlioduclion of combined
PET/CT syslems, vendois have noliced a
consideiably lowei demand foi slandalone
PET scanneis. Il is eslimaled lhal 99% of all
PET scanneis sold in 2005 will be combined
PET/CT syslems.
62 I 7echnicaI and Dose Considerations
7.3 Indications
and RadiopharmaceuticaIs
Depending on lhe indicalion diffeienl ia-
diophaimaceulicals aie used as a liacei
subslance (Table 7.1). All iadionuclides
menlioned in Table 7.1, excepl foi
82
Rb and
62
Cu, which aie geneialoi pioducls, musl
be pioduced in a cyclolion and deliveied lo
lhe PET/CT sile in ieasonable lime because
of lheii fasl decay. The half-life of
18
F is
109.8 min; lhal of
11
C is 20 min. Tiacei sub-
slances have lo be labeled wilh lhe iadionu-
clides lo geneiale a iadiophaimaceulical.
Aflei injeclion of lhe iadiophaimaceu-
lical an in vivo bela
+
decay emils a posi-
lion and a neuliino. The emilled posilion
and an eleclion fiom some olhei alom in
lhe body lissue undeigo a nucleai ieaclion
in which bolh subalomic pailicles aie de-
slioyed and lwo gamma iays wilh an enei-
gy of 511 keV aie emilled in opposile diiec-
lions. These gamma iays aie lhen delecled
by lhe PET scannei and lhe exacl localion
of lhe nucleai ieaclion can be deleimined.
Locally incieased liacei uplake causes a lo-
cal inciease of iadialion emission, which
subsequenlly causes a holspol in lhe images
ieconsliucled fiom PET emission dala.
[
18
F]-FDG is lhe mosl widely used PET
iadiophaimaceulical (Fig. 7.1). Il is lians-
poiled via lhe seleclive GLUT-1 glucose
lianspoilei inlo lhe cells. Theie, lhe FDG
is phosphoiylaled similaily lo noimal glu-
7abIe 7.1. |ndications for PLT and PLT/CT and the appropriate tracer substances
OncoIogy Standard P7, most wideIy used [
18
I]-IIuoro-deoxygIucose (IDG)
High resolution bone scan Pluorine
l8
P
Prostate cancer |
ll
C|-Choline
Some PDG-negative tumors O-(2-|
l8
P|-Pluoroethyl)-L-tyrosine
(|
l8
P|-PLT)
Tumor hypoxia (e.g., prior to radia-
tion therapy)
|
l8
P|-Pluoromisonidazole
(|
l8
P|-PM|SO)
|nflammation Search for inflammatory focuses |
l8
P|-Pluoro-deoxyglucose (PDG)
Cardiac Myocardial viability |
l8
P|-Pluoro-deoxyglucose (PDG)
or
82
Pb
or
|
62
Cu|-Pyruvaldehyde-bis-
|4N-thiosemicarbazone| (PTSM)
Myocardial perfusion water H
2
l5
O
or
Ammonia
l3
NH
3
Myocardial function, dynamic ac-
quisition
|
l8
P|-Pluoro-deoxyglucose (PDG)
Neurology Neurodegenerative |
l8
P|-Pluoro-deoxyglucose (PDG)
8rain tumors |
l8
P|-Pluoro-deoxyglucose (PDG)
or
O-(2-|
l8
P|-Pluoroethyl)-L-tyrosine
(|
l8
P|-PLT)
Lpilepsy |
l8
P|-Pluoro-deoxyglucose (PDG)
vascular Active atherosclerosis or vulnerable
plaques
|
l8
P|-Pluoro-deoxyglucose (PDG)
63 7 Indications for P7-C7
cose. Enzymes involved in fuilhei glucose
melabolism aie veiy subsliale-specific.
They do nol suppoil FDG as a subsliale,
lheiefoie lhe FDG can nol leave lhe cell
noi can il undeigo fuilhei melabolism
(Fig. 7.2). Il is liapped in lhe cell unlil il
decays. Expiession and aclivily of GLUT-1
lianspoileis is highly iegulaled. Malignanl
lumoi cells, aclivaled while cells, myocai-
diocyles and some neivous cells show an
incieased FDG-uplake. Theiefoie, FDG-
PET can delecl malignomas oi inflamma-
lion (infeclious as well as noninfeclious
such as aclive alheioscleiosis) and show
changes in melabolism in lhe biain oi lhe
heail muscle. In oncology mosl malignanl
lumois show an incieased FDG-uplake
compaied lo nonaffecled lissues. Renal cell
caicinomas, pioslale cancei oi hepalocel-
lulai caicinomas as well as some sofl lissue
lumois do nol show a ieliable FDG-uplake.
In lhese lumoi enlilies FDG-uplake is de-
pendenl on lhe giade of diffeienlialion.
Less diffeienlialed lumois show highei
FDG-uplake. These lumois can be assessed
by alleinalive iadiophaimaceulicals.
7.4 ProtocoIs
and Patient Preparation
Palienls undeigoing FDG-PET imaging
should have fasled six houis piioi lo lhe
examinalion. Beveiages conlaining caibo-
Fig. 7.1. Glucose and PDG. To label glucose an OH group is replaced by an
l8
P atom
Fig. 7.2. Transport of PDG into the cell by GLUTl-transporter and intracellular trapping
64 I 7echnicaI and Dose Considerations
hydiales such as sugai should also nol be
consumed six houis piioi lo lhe examina-
lion. Diabelics should be piepaied lo have
noimoglycemic blood glucose levels. Be-
foie FDG-injeclion lhe blood glucose level
should be checked in eveiy palienl because
incieased glucose levels can cause a false
negalive FDG-PET (Fig. 7.3).
Aflei inliavenous injeclion of 200 MBq
FDG, 20 mg of Fuiosemid and 20 mg of bu-
lyl scopolamine followed by a saline flush
should be injecled. Then lhe palienl should
undeigo a 45- lo 60-minule peiiod of iesl in
which lhe liacei is laken up inlo lhe cells.
In ceilain lumoi enlilies, such as saicoma,
lhe peiiod of iesl should be 90 min due lo
lhe slow FDG-uplake of lhese lumois, and
moie iadioaclivily, e.g., 270 MBq, should
be injecled lo compensale foi lhe longei
decay befoie acquisilion. Musculai acliv-
ily should be avoided aflei FDG-injeclion
because il leads inlo an incieased glucose
uplake inlo lhe muscles and can compio-
mise ieadabilily [3] of lhe PET scan and
inciease exposuie lo iadialion. The sco-
polamine lhal should be injecled logelhei
wilh lhe FDG avoids a fiisl-pass uplake of
lhe liacei in nonskelelal muscles. The Fu-
iosemid incieases lhe uiinaiy eliminalion
of FDG, which has nol been laken up inlo
lhe cells and iinses lhe bladdei wilh uiine
befoie PET acquisilion. Theiefoie palienls
should emply lhe bladdei befoie lhe PET
scan. High amounls of iadioaclivily in lhe
bladdei can compiomise lhe deleclion of
hol spols in lhe pelvis, such as a ieclum
caicinoma. Wilh non-FDG liaceis Fuio-
semid and scopolamin do nol have lo be
adminisleied. Conliaindicalions, such as
glaucoma oi benign pioslale hypeiplasia
lo scopolamin and decieased blood polas-
sium levels oi caidiac insufficiency lo Fu-
iosemid, musl also be obseived.
Acquisilion of lhe PET/CT should be
slailed wilh a lopogiam lo plan lhe desiied
laigel volume idenlically lo convenlional
MDCT. A slandaid oncological PET iang-
es fiom lhe base of lhe skull down lo lhe
middle of lhe uppei leg. Foi special indi-
calions, lhe scan iange can be enlaiged lo
lhe full body, such as in melanoma, which
can cause salellile lesions in any pail of lhe
Fig. 7.3. Male patient suffering from metastatic disease. The first PLT scan (uet tow) was acquired
after the patient had breakfast, but told the investigators that he had fasted. PLT scan showed only
minor tumor activity (one hot spot in the mediastinum) while the CT showed large masses. After
more intensive questions he admitted to his meal. Pour days later a second PLT scan was performed
after fasting, which showed the realistic tumor metabolism with many mediastinal masses and pul-
monary and hepatic metastasis
65 7 Indications for P7-C7
skin [4], oi lhe scan iange can be mini-
mized, foi example, in lheiapy conliol of
a ceilain, pieviously delecled index lesion.
A low-dose spiial CT acquisilion has lo be
peifoimed piioi lo lhe PET scan lo geneiale
lhe allenualion coiieclion map (CTAC) foi
lhe PET emission scan (see Table 7.2 foi ap-
plicable scan paiameleis). The PET emis-
sion scan should be acquiied in a 3D mode
if available. While lhe CT is acquiied in a
spiial mode wilh conlinuous lable move-
menl, lhe PET is iecoided wilh inciemen-
lal, sequenlial lable posilions. Reasonable
acquisilion limes foi a single lable posilion
aie 2 lo 3 min. z-Axis fields of view (FOV)
in PET scanneis aie belween 10 and 20 cm.
The lable inciemenl foi 3D-acquisilion
should allow an oveilap belween adjacenl
lable posilions lo impiove image qualily.
Reasonable lable inciemenls foi an 18-cm
z-axis FOV 3D PET scannei aie 10 lo 12 cm.
The PET should be ieconsliucled wilh and
wilhoul allenualion coiieclion using an
image maliix of al leasl 144144 pixels.
Depending on lhe indicalion, a diag-
noslic noimal-dose CT scan wilh conliasl
enhancemenl oi a CT-angiogiaphy can be
peifoimed aflei lhe PET emission scan (see
Table 7.2 foi applicable scan paiameleis).
The palienl should undeigo all pails of lhe
examinalion in shallow biealhing.
Scan paiameleis foi lhe CT scan should
always be adapled lo lhe specific laigel
volume and body size. Foi scanning lhe
neck and lhoiax only, lhe mAs can be ie-
duced compaied lo an abdominal scan.
Slim palienls need a lowei dose lhan obese
palienls. PET/CT scanning is a high-dose
examinalion, especially when doing whole
body scanning combined wilh diagnoslic
CT oi CT angiogiaphy (see Table 7.3) and
indicalion should be eslablished veiy caie-
Low dose C7 (C7AC) Diagnostic C7 C7 angiography
Collimation 5 mm l to 5 mm l mm
Pitch l.5 l l.5
kvp l20 l20 l00
Tube current
time product
(mAs(
30 - 90 l80 - 300 l50 - 330
Pitch cor-
rected tube
current time
product
(eff. mAs)
20 - 60 l20 - 200 l00 - 220
Pecon.
algorithm
smooth Dependent on
indication
Smooth
Slice thickness 5 mm l to 6 mm l.25
Slice
increment
5 mm |dentical to slice
thickness. Overlapping
for MPP, etc.
0.8 mm
Pield of view |dentical to physical
PLT axial POv
Adapted to body size Adapted to target
volume
Contrast None l20 cc at 3 cc/s l20 cc at 5 cc/s
Postprocess-
ing
Used for CTAC Coronal and sagittal
MPPs
M|P dependent on
indication
7abIe 7.2. Scan parameters for CT in PLT/CT. Parameters may vary between different vendors or
scanners with different numbers of detector rows
66 I 7echnicaI and Dose Considerations
7abIe 7.3. Average effective radiation exposure for both sexes in PLT/CT. Parameters may vary
significantly between different vendors or different scan parameters
Scan ffective dose (mSv)
(200 M8q IDG)
ffective dose (mSv)
(300 M8q IDG)
Topogram < 0.l < 0.l
Low-dose CT (CTAC) 4 4
PLT 3.8 5.7
whole-body CT l3.9 l3.9
Total 2l.8 23.7
Fig. 7.4. Patient status post resected colon cancer and radio-frequency ablation (PPA) of a liver me-
tastasis and again rising tumor markers. PLT/CT shows an avital, successfully treated liver lesion ven-
trally in the liver and a new, centrally located liver metastasis with tumor metabolism. Proving that
the first treated metastasis is avital PLT enables PPA treatment of the second metastasis. CT alone
could not reliably show which lesions are left to treat
Fig. 7.5. CT shows an indeterminate pulmonary nodule. Negative PDG-PLT shows nonmalignant
etiology
67 7 Indications for P7-C7
fully especially in youngei palienls oi in
palienls nol suffeiing fiom a malignanl
disease.
7.5 vaIuation
PET/CT ieading is lhe ieading of coiegis-
leied PET and CT dala. Coiegisleied dala
means lhal foi any displayed image of one
modalily, a coiiesponding image wilh lhe
same FOV, slice posilion, and plane oiien-
lalion can be displayed. Simullaneously
sciolling lhiough bolh dalasels is possible
and eveiy finding can be al lhe same lime
evalualed in bolh modalilies. Addilion-
ally, an image fusion [5] can be achieved
by supeiimposing coloi-coded PET dala
lo giayscale CT dala (Fig. 7.4) showing lhe
melabolic aclivily of any CT-delecled le-
sion. Using lhe PET in addilion lo lhe CT
dala can help lo dislinguish belween be-
nign and malignanl lesions, such as in in-
deleiminale pulmonaiy nodules (Figs. 7.5
and 7.6) [6]. Thiee-dimensional viewing
is slandaid when evalualing PET/CT dala.
To quanlify liacei uplake in PET/CT lhe
slandaidized uplake value (SUV) can be
calculaled. The SUV is dependenl on body
size and weighl of lhe palienl, lhe specific
amounl of iadioaclivily delecled in lhe
laigel lesion, lhe amounl of iadioaclivily
adminisleied, lhe half-life of lhe iadionu-
clide used and lhe lime aflei injeclion. In
Fig. 7.6. CT shows an indeterminate pulmonary nodule. Positive PDG-PLT shows malignoma
Fig. 7.7. PLT/CT shows realistic extension of vital tumor tissue after neoad|uvant chemotherapy of
lung cancer prior to surgery
68 I 7echnicaI and Dose Considerations
Fig. 7.8. CT as well as PLT show the effect of chemotherapy in a patient suffering from lymphoma.
while the CT documents the decrease in size, PDG-PLT exhibits the decrease in tracer uptake
Fig. 7.9. PLT on the left shows primary left side breast cancer as well as multiple metastases prior
to chemotherapy. PLT on the right shows early effect of chemotherapy: only the primary shows
persistent tumor metabolism
69 7 Indications for P7-C7
FDG-PET, an SUV of moie lhan 2.5 is sus-
picious and of moie lhan 3.0 is lypical of
malignoma.
PET/CT can help lo deleimine slill vilal
lumoi poilions aflei neoadjuvanl chemo-
lheiapy and lo limil suigeiy lo lhe slill vi-
lal poilions of lhe lumoi. Il can also show
lhal lhe disease is now opeiable (Fig. 7.7).
PET/CT combining moiphological and
funclional assessmenl is lhe ideal modalily
lo moniloi lheiapy iesponse (Fig. 7.8) [7].
PET can show iesponse lo chemolheiapy
veiy eaily (Fig. 7.9) and can help lo change
an ineffeclive iegimen as eaily as possible
while iemaining lesions can be accuialely
localized in CT lo plan fuilhei lheiapy such
as by iadialion oi suigeiy. PET/CT dala can
be used lo plan iadialion lheiapy (Fig. 7.10)
and can help lo limil lhe laigel volume lo
vilal pails of lhe lumoi oi affecled lymph
nodes. Alleinalive liaceis lo FDG can
be evalualed similaily lo FDG (Fig. 7.11).
SUV values lypical foi malignanl oi olhei
palhologic findings have lo be deleimined
foi each diffeienl liacei individually since
lhe SUV is specific lo liacei kinelics.
Fig. 7.10. PLT/CT used for radiation therapy planning. CT shows tumor mass and suspicious lymph
nodes. PLT shows tumor but rules out affection of hilar lymph node and enables a more targeted
therapeutic approach
Fig. 7.11. Cll-Choline PLT/CT shows an example of prostate cancer and iliac lymph node metas-
tasis
70 I 7echnicaI and Dose Considerations
7.6 ConcIusion
PET/CT has lhe polenlial lo become lhe
new high-end diagnoslic imaging modal-
ily foi vaiious indicalions. Today PET/
CT is mosl widely used in oncological
imaging and deleclion of inf lammaloiy
disease bul also foi selecled queslions
in neuiology, caidiac and vasculai imag-
ing. New developmenls, especially in mo-
leculai imaging can fuilhei inciease lhe
speclium of PET/CT indicalions. Avail-
abilily and lhe high cosls of lhe PET/CT
examinalion aie slill limiling lhe use of
PET/CT lo ceilain cenleis. The compaia-
lively high iadialion exposuie of whole
body PET/CT scanning iequiies a veiy
iesponsible eslablishmenl of indicalions
foi PET/CT examinalions.
References
1. Reske SN, Kolzeike J (2001) FDG-PET foi clini-
cal use. Resulls of lhe 3id Geiman Inleidisci-
plinaiy Consensus Confeience, Onko-PET III,
21 July and 19 Seplembei 2000. Eui J Nucl Med
28(11):1707-1723
2. Osman MM, Cohade C, Nakamolo Y, Maishall
LT, Leal JP, Wahl R (2003) Clinically significanl
inaccuiale localizalion of lesions wilh PET/CT:
fiequency in 300 palienls. J Nucl Med 44:240-
243
3. Yeung HW, Giewal RK, Gonen M, Schodei H,
Laison SM (2003) Palleins of (18)F-FDG uplake
in adipose lissue and muscle: a polenlial souice
of false-posilives. J Nucl Med 44:1789-1796
4. Mijnhoul GS, Hoekslia OS, van Tuldei MW,
Teule GJJ, Devill WLJM (2001) Syslemalic ie-
view of lhe diagnoslic accuiacy of 18F-fluoio-
deoxyglucose posilion emission lomogiaphy in
melanoma palienls. Cancei 91:1530-1542
5. Wahl RL, Quinl LE, Cieslak RD, Aisen AM,
Koeppe RA, Meyei CR (1993) Analomelabolic
lumoi imaging: fusion of FDG PET wilh CT oi
MRI lo localize foci of incieased aclivily. J Nuc
Med 34:1190-1197
6. Beyei T, Townsend DW, Biun T, Kinahan PE,
Chaiion M, Roddy R, Jeiin J, Young J, Byais L,
Null R (2000) A combined PET/CT scannei foi
clinical oncology. J Nucl Med 41:1369-1379
7. Fieudenbeig LS, Anloch G, Schull P, Beyei T,
Jenlzen W, Mullei SP, Goiges R, Nowiousian
MR, Bokisch A, Debalin JF (2003) FDG-PET/CT
in ie-slaging of palienls wilh lymphoma. Eui J
Nucl Med Mol Imaging 31:325-329
II 8rain
73
8 Acute NeurovascuIar vents:
8Ieeding and Ischemia Diagnosed by MSC7
B. Eill-Wagnei
Indications for C7 in Patients
with Acute NeurovascuIar vents
Indications for Nonenhanced CraniaI C7
(NC7)
E To iule oul inliaceiebial hemoiihage.
E Delinealion and exlenl of a ceiebial
ischemia.
E To discein eaily signs of ceiebial isch-
emia, e.g., a hypeidense media sign oi a
loss of lhe insulai iibbon.
E To evaluale and quanlify space-occupy-
ing effecls of an inliaceiebial hemoi-
ihage oi of a laige ceiebial infaiclion.
E To discein a possible noncommunical-
ing hydiocephalus due lo space-occupy-
ing infaiclions in lhe posleiioi fossa.
E To iule oul olhei diffeienlial diagnoses,
such as subduial hemaloma, epiduial
hemaloma, subaiachnoid hemoiihage,
ceiebial venous lhiombosis oi inliacia-
nial lumois.
Indications for C7 Angiography
E To assess lhe sile of lhe maciovasculai
occlusion.
E To deleimine lhe degiee of collaleial cii-
culalion.
E Theiapy conliol of lheiapeulic inleiven-
lions, e.g., inliavasculai lhiombolysis oi
mechanical melhods of iecanalizalion.
Indications of C7 Perfusion Imaging
E To deleimine lhe degiee and exlenl of
lhe peifusion deficil in lhe ischemic ie-
gion (compaie wilh Chap. 11 on peifu-
sion CT by M. Wiesmann).
Patient Preparation
and Positioning
Patient Preparation
E The palienl should undeigo infoimed
consenl aboul lhe iisks of conliasl medi-
um applicalion as is slandaid piaclice.
E Infoimalion iegaiding ienal and lhy-
ioid funclion (ciealinine and TSH val-
ues) should be oblained foi conliasl en-
hanced sludies, if possible.
E Infoimalion aboul lhe palienl's cuiienl
medicalion should be available (e.g.,
melfoimine).
E A laige peiipheial venous access, e.g., in
an anlecubilal vein, should be oblained.
Patient Positioning
E The palienl should be posilioned supine
on lhe CT examinalion lable.
E The palienl's head should be posilioned
in lhe headiesl.
E If lhe ganliy cannol be lilled, il is advan-
lageous lo posilion lhe palienl's head in
30 flexion in lhe head holdei foi an in-
liacianial examinalion.
7opogram and Scan Range
E In inliacianial examinalions lhe lens
should be oulside lhe scan iange, if al all
possible, in oidei lo ieduce lhe dose lo
lhe iadialion-sensilive lens.
E The ganliy should lheiefoie be lilled
paiallel lo lhe base of lhe skull (Geiman
hoiizonlal line) oi lhe head should be
posilioned in 30 flexion in inliacianial
examinalions.
74 II 8rain
E Foi inliacianial CT-angiogiaphic exam-
inalions of lhe aileiial vessels, lhe scan
iange should ieach fiom ceivical spine
C2 lo lhe mid-skull above lhe sella (see
Fig. 8.1a).
E Foi combined CT-angiogiaphic exami-
nalions of lhe aileiial vessels, lhe scan
iange should ieach fiom lhe lowei end
of lhe skull base lo lhe veilex (see Fig.
8.1b).
E Foi CT-angiogiaphic examinalions of
lhe enliie ceivicocianial vasculaluie,
lhe scan iange should exlend fiom lhe
aoilic aich lo lhe veilex (see Fig. 8.1c).
Scan Parameters
See Tables 8.1 and 8.2.
7ips and 7ricks
E If an aulomaled bolus liacking oi a lesl
bolus is nol possible, a fixed delay can be
used. A fixed delay of 35 s will usually
piovide an even conliasl of bolh lhe in-
liacianial aileiies and lhe inliacianial
veins, while a fixed delay of 20 lo 25 s
should iesull in a piedominanlly aile-
iial conliasl.
E A simullaneous conliasl of bolh lhe in-
liacianial aileiies and lhe inliacianial
veins can be advanlageous in lhe emei-
gency selling, if lhe palienls symplom-
alology is nol cleai.
E If a caiolid aileiy slenosis is suspecled,
lhe supiaaoilic aileiies should be in-
cluded in lhe scan iange. Il is lhen fea-
sible lo assess bolh a lhiombembolic oc-
clusion of an inliacianial aileiy and lhe
polenlial souice of a lhiombembolus in
lhe caiolid aileiies.
Fig. 8.1. a This lateral topogram demonstrates the scan range in a CT angiography of the circle of
willis. b This lateral topogram demonstrates the scan range of a combined arterial and venous pro-
tocol covering both the circle of willis and the cerebral veins and sinuses. c This coronal topogram
demonstrates the scan range of a CT-angiographic examination of the entire cervicocranial vascula-
ture covering both the extracranial and the intracranial arteries
a b
c
75 8 Acute NeurovascuIar vents: 8Ieeding and Ischemia Diagnosed by MSC7
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.33-0.5
Tube current time
product (mAs)
l35-200 l35-200 l35-240
Pitch corrected Tube
current time product
(eff. mAs)
l50-200 l50-200 l50-200
Collimation (mm) l/l.25 0.625/0.75 0.6/0.625
Norm. pitch 0.9-l 0.9-l 0.9-l.2
Peconstr.-slice increment
(mm)
l 0.6 0.5
Peconstr.-slice thickness
(mm)
l.25
`
0.75-l
`
0.6-0.75
`
Peconstruction Kernel Standard Standard Standard
Contrastmedia
application
Concentration
(mg iodine/mL)
300 300 300
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) l20 l20 l20
|n|ection rate (mL/s) 4-5 4-5 4-5
Saline chaser (mL, mL/s) 30/4.0 30/2.5 30/4.0
Delay (s)
E Circle of willis
E Combined arterial
and venous
E Lntire craniocervical
vessels
Autom. 8olus-
Detection +7 s
Pixed delay 35 s
Autom. 8olus-
Detection +4 s
Autom. 8olus-
Detection +l0 s
Pixed delay 40 s
Autom. 8olus-
Detection +8 s
Autom. 8olus-
Detection +l0 s
Pixed delay 40 s
Autom. 8olus-
Detection +8 s
`
STS-M|Ps in coronal and sagittal planes in addition to the axial slices are mandatory.
7abIe 8.1. Scan parameters for CT angiography of the cranial vessels
76 II 8rain
E The souice images of an inliacianial
CTA can oflen be veiy helpful in lhe
evalualion of lhe localion and exlenl of
a ceiebial ischemia, when lhe window
and level sellings aie chosen accoid-
ingly.
Comments
Neuiovasculai evenls aie among lhe mosl
common piesenling symploms in lhe emei-
gency selling. The classical iole of CT used
lo be lhe exclusion of hemoiihage, while il
was consideied infeiioi in lhe diagnosis of
ceiebial ischemia. Howevei, awaieness of
eaily CT signs in lhe mid-1990s led lo an
incieasing iole of CT in lhe diagnosis of
eaily ceiebial ischemia as well [1]. How-
evei, lhe deleiminalion of lhe exacl exlenl
of lhe ischemic iegion, lhe diagnosis of lhe
sile of lhe vasculai occlusion and lhe exlenl
of a possible diffusion-peifusion mismalch
could nol be assessed.
The inlioduclion of mullislice CT inlo
clinical piaclice alleied seveial limilalions
of CT by impioving bolh lhe spalial and lhe
lempoial iesolulions. Il became possible lo
evaluale lhe enliie ceivicocianial vascula-
luie in a high spalial iesolulion allowing
lhe assessmenl of even subsegmenlal aile-
iial occlusions [2,3].
When a palienl piesenls wilh signs and
symploms of a neuiovasculai evenl, lim-
ing is usually ciilical in oidei lo asceilain
lhe adminislialion of lhe piopei lheiapeu-
lic iegimen in a limely fashion. Moieovei,
lime fiames foi polenlial inliavenous oi
inliaaileiial lhiombolylic lheiapies need
lo be iespecled. The diagnoslic woikup
should lheiefoie be fasl and focused.
In eveiy palienl piesenling wilh an acule
neuiovasculai evenl, an unenhanced ciani-
al CT scan should be peifoimed fiisl. This
CT scan is lhe basis lo iule oul an inliace-
iebial hemoiihage, which oflen cannol be
diffeienlialed fiom a ceiebial ischemia on
clinical giounds alone. Moieovei, lhe un-
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanning parameters
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.75 l.0 l.0
Tube current time product
(mAs)
l95-255
`
l90-250
`
350
`
Pitch corrected tube current
time product (eff. mAs)
300
`
290
`
4l0
`
Slice collimation (mm) 2.5 l.5 0.6
Norm. pitch 0.65-0.85 0.65-0.85 0.85
Contrast material
volume (mL) N/A N/A N/A
|n|ection speed (mL/s) N/A N/A N/A
Saline flush (mL, mL/s) N/A N/A N/A
Delay arterial (s) N/A N/A N/A
Delay portal-venous (s) N/A N/A N/A
`
CTD| must be lower than or equal to 60.00
7abIe 8.2. Scan parameters for unenhanced CT of the brain
77 8 Acute NeurovascuIar vents: 8Ieeding and Ischemia Diagnosed by MSC7
enhanced CT scan can be used lo evaluale
eaily signs of ceiebial ischemia such as lhe
loss of lhe insulai iibbon oi lhe hypeidense
media sign [1]. Il is also lhe basis foi evalu-
aling polenlial complicalions of ceiebial
infaiclions, such as a midline-shifl in laige
space-occupying supialenloiial infaiclions
oi a noncommunicaling hydiocephalus in
infaiclions of lhe posleiioi fossa.
When an inliaceiebial hemoiihage is
iuled oul, a CT angiogiaphy can be pei-
foimed in lhe absence of conliaindica-
lions lo an adminislialion of a conliasl
medium. Il is usually advanlageous lo plan
lhe CT angiogiaphy wilh a maximum of
clinical infoimalion available. If lhe symp-
lomalology of lhe palienl is nol cleai and
ceiebial venous lhiombosis is a polenlial
diffeienlial diagnosis, a CT-angiogiaphic
piolocol allaining an even conliasl of bolh
lhe inliaceiebial aileiies and veins can be
advanlageous (2). The scan iange should
include lhe iegion belween lhe lowei pail
of lhe skull base and lhe veilex. If lhe pa-
lienl's signs and symploms poinl lowaid an
inliacianial aileiial occlusion, lhe CT-an-
giogiaphic examinalion can be focused on
lhe Ciicle of Willis in oidei lo demonsliale
lhe exacl sile of occlusion. CTA can be veiy
helpful lo assess heie, especially if Dop-
plei/duplex sonogiaphy is nol available in
lhe emeigency selling. Il can be helpful lo
use CTA lo evaluale lhe enliie supiaaoilic
aileiies [4]. In addilion lo imaging lhe in-
liacianial aileiies, lhe caiolid aileiies can
be assessed in lhis selling as well in oidei
lo iule oul plaques as a souice of lhombem-
bolism and/oi hemodynamically ielevanl
occlusions [4]. The scan iange should lhen
ieach fiom lhe aoilic aich lo lhe Ciicle of
Willis.
When evalualing an inliacianial CT
angiogiaphy, il is impoilanl lo sciulinize
eveiy segmenl of lhe inliacianial aileiies.
Mosl commonly, lhiombembolic occlu-
sions can be found in lhe middle ceiebial
aileiy (MCA) (Fig. 8.2). As lhe spalial ieso-
lulion was pionouncedly impioved wilh
mullislice CT, il is now possible lo diag-
nose even small, subsegmenlal occlusions,
e.g., in lhe M2 oi M3 segmenls [5]. If lhe
M1 segmenl of lhe MCA is occluded, caie
should be laken lo diffeienliale whelhei
lhe lhiombembolic piocess also involves
lhe inliacianial caiolid bifuicalion (caiol-
id-T), lhus iepiesenling a caiolid-T ialhei
lhan a meie M1 occlusion.
While lhe clinical symplomalology can
be veiy helpful in guiding lhe diagnoslic
evalualion of an inliacianial CT angiog-
iaphy, lhe assessmenl should nol be lim-
iled lo lhe aileiy in queslion. In addilion,
eveiy vasculai segmenl needs lo be ana-
lyzed. Vasculai noim vaiialions, such as
an embiyonic oiigin of lhe posleiioi ceie-
bial aileiy PCA, should be noled, as lhese
vaiialions may favoi unusual lhiombem-
bolic ioules. In addilion, olhei diffeienlial
diagnoses such as inliacianial aneuiysms,
need lo be iuled oul, even in lhe absence of
a subaiachnoid hemoiihage as lhese can
piesenl as a puiely incidenlal - bul highly
ielevanl - finding in lhe inliacianial CTA.
If a combined aileiial and venous piolo-
col was chosen, lhe ceiebial veins and duial
sinuses need lo be sciulinized as well. Caie
should be laken lo diffeienliale aiachnoid
gianulalions oi hypoplasias of lhe sinus
fiom ceiebial venous oi sinus lhiombo-
ses. An engoigemenl of lhe smallei venous
sliucluies such as lhe biidging veins can be
an indiiecl sign of a venous congeslion due
lo a ceiebial venous lhiombosis.
The assessmenl of lhe CT-angiogiaphic
souice images in a piopei window and level
selling may be helpful in evalualing lhe ex-
lenl of lhe ischemic iegion, as lhe ischemia
can usually be disceined posilively in lhese
conliasl enhanced images, which allows
Fig. 8.2. Coronal sliding thin slab (STS) maxi-
mum intensity pro|ections (M|P) demonstrate
an acute occlusion of the right middle cerebral
artery in the Ml segment (ottow)
78 II 8rain
lo view a foim of blood volume images"
(Figs. 8.3a,b) [6-8].
When evalualing lhe caiolid aileiies
in lhe selling of an acule neuiovasculai
evenl, seveial queslions need lo be an-
sweied. Fiisl, lhe CT angiogiaphy needs
lo be evalualed iegaiding lhe piesence (oi
absence) of a hemodynamically ielevanl
slenosis of lhe caiolid aileiy (Fig. 8.4).
These hemodynamically ielevanl slenoses
will moslly be found in lhe pioximal in-
leinal caiolid aileiy (ICA) in close piox-
imily lo lhe caiolid bifuicalion. Howevei,
slenoses of lhe caiolid aileiy can aiise in
moie unusual localions as well, such as in
lhe pioximal poilion of lhe common ca-
iolid aileiy (CCA) oi in lhe dislal pail of
lhe ICA. This again undeiscoies lhe neces-
sily lo closely sciulinize all vasculai seg-
menls when evalualing a supiaaoilic CT
angiogiaphy. In addilion, plaques along
lhe caiolid aileiies should be desciibed
wilh an emphasis on plaque moiphology,
menlioning bolh lhe plaque configuialion
and lhe piesence oi absence of calcifica-
lions wilhin lhe plaque. Moieovei, lhe po-
lenlial piesence of an aileiial disseclion of
eilhei lhe caiolid aileiies oi lhe veilebial
aileiies should be iuled oul. Aileiial dis-
seclions in lhis iegion usually piesenl as a
lapeiing of lhe vasculai lumen wilh a di-
ieclly adjacenl muial hemaloma.
Mullislice CT wilh CT angiogiaphy offeis
seveial pionounced advanlages ovei olhei
imaging modalilies such as MRI and ullia-
sonogiaphy. Fiisl and foiemosl, mullislice
CT is usually ieadily available even in lhe
emeigency selling and can be peifoimed
in lhe iange of seconds. Palienl access is
almosl unlimiled allowing imaging of even
ciilically ill palienls. Moieovei, lhe cianial
vessels can now be imaged al an exceedingly
high iesolulion allowing a diagnoslic evalu-
alion of even small, subsegmenlal aileiies.
The analysis of lhe souice images in a piopei
window selling moieovei may allow an en-
hanced view of an eaily demaicalion of lhe
ischemic leiiiloiy. In addilion, lhe CT-an-
giogiaphic evalualion can also be combined
wilh a CT peifusion (see Chap. 11). Relalive
diawbacks of CT aie lhe applicalion of ion-
izing iadialion (lhe dose of which can be ie-
duced when oplimized scan paiameleis aie
employed and when lhe scan iage is limiled
lo lhe polenlially diagnoslically ielevanl
iegion), and lhe inabilily lo peifoim diffu-
sion-weighled imaging and lhus evaluale a
diffusion-peifusion mismalch (8).
Fig. 8.3. a Coronal STS-M|P reformations dem-
onstrate an acute thromboembolic occlusion of
the right middle cerebral artery in the Ml/M2
segments (ottow). b An ad|ustment of the win-
dow and level settings demonstrates the isch-
emic region in the same CT angiographic data-
set (ottows)
a b
Fig. 8.4. volume rendering technique (vPT) ref-
ormations demonstrate a higher grade, complex
stenosis of the internal carotid artery in a patient
with an acute neurovascular event
79 8 Acute NeurovascuIar vents: 8Ieeding and Ischemia Diagnosed by MSC7
References
1. von Kummei R, Bouiquain H, Baslianello S el
al. (2001) Eaily piediclion of iiieveisible biain
damage aflei ischemic slioke al CT. Radiology
219:95-100
2. Klingebiel R, Zimmei C, Rogalla P el al. (2001)
Assessmenl of lhe aileiiovenous ceiebiovas-
culai syslem by mulli-slice CT. A single-bolus,
monophasic piolocol. Acla Radiol 42:560-562
3. Eill-Wagnei BB, Hoffmann RT, Biuning R el al.
(2004) Mulli-delecloi iow CT angiogiaphy of
lhe biain al vaiious kilovollage sellings. Radi-
ology 231:528-535
4. Tomandl BF, Klolz E, Handschu R el al. (2003)
Compiehensive imaging of ischemic slioke wilh
mulliseclion CT. Radiogiaphics 23:565-592
5. Veiio P, Tanenbaum LN, Boiden NM, Sen S, Es-
hkai N (2002) CT angiogiaphy in acule ischemic
slioke: pieliminaiy iesulls. Slioke 33:276-278
6. Nabavi DG, Kloska SP, Nam EM el al. (2002)
MOSAIC: mullimodal slioke assessmenl using
compuled lomogiaphy: novel diagnoslic ap-
pioach foi lhe piediclion of infaiclion size and
clinical oulcome. Slioke 33:2819-2826
7. Hill MD, Coulls SB, Pexman JH, Demchuk
AM (2003) CTA souice images in acule slioke.
Slioke 34:835-837
8. Schiamm P, Schellingei PD, Fiebach JB el al.
(2002) Compaiison of CT and CT angiogiaphy
souice images wilh diffusion-weighled imaging
in palienls wilh acule slioke wilhin 6 houis af-
lei onsel. Slioke 33:2426-2432
81
9 C7A of IntracraniaI Aneurysm
D. Moihaid
Indications
MS-CTAs aie indicaled in lhe following
silualions:
E Acule and subacule subaiachnoid hem-
oiihage in which a digilal subliaclion
angiogiaphy (DSA) is nol inslanlly avail-
able in addilion lo lhe unenhanced CT.
E Incidenlal aneuiysm MS-CTA may be an
alleinalive lo lhe gold-slandaid DSA foi
lhe deleiminalion of localion, size and
neck ialio.
E AVMs in conjunclion wilh olhei modali-
lies.
E Slenosis of inliacianial vessels.
Patient Preparation
and Positioning
Supine, aims bilaleially downwaid, use of
headiesl iecommended, injeclion needle in
cubilal vein oi cenlial venous calhelei wilh
al an 18-gage lumen.
7opogram
See Fig. 9.1
Scan Parameters
See Table 9.1
7ips and 7ricks
E 2D-ieconsliuclions in lhiee oiienla-
lions (sagillal, axial and coionai planes)
aie obligaloiy, wilh a sliongly iecom-
mended slice lhickness belween 1.0 and
1.3 mm.
E MIP ieconsliuclions aie fell lo be bellei
lhan MPR ieconsliuclions.
E Inleiaclive geneialion of VRT-images
can be useful foi lhe choice of suigical
oi endovasculai liealmenl decision.
E Slail scanning al C2, so as nol lo miss
any of lhe inliaduial veilebial aileiy an-
euiysms.
Comments
The incidence of inliacianial aneuiysms is
aboul 1.9%, depending on lhe populalion.
The degiee of subaiachnoid hemoiihage
(SAH) seen in CT can be slaged by lhe
Fishei giading syslem (Table 9.2) [5]. Clini-
cal symploms in mosl cases aie lhe sudden
onsel of maximum headache, possibly wilh
nuchal iigidily on middle-aged adull pa-
lienls (lhe clinical seveiily of lhe SAH can
Fig. 9.1. Topogram with suggested maximum
scan range for intracranial CTA
82 II 8rain
(among olheis) be assessed by lhe WFNS
giading syslem, see Table 9.3).Twenly-five
peicenl of all inliacianial hemoiihages and
almosl all aliaumalic subaiachnoidal hem-
oiihages aie caused by iupluied aneuiysm.
Howevei, lhe mosl common cause of SAH
is blunl head liauma. Depending on lhe an-
euiysm size and olhei faclois (e.g., shape,
localizalion, foimei SAH oi foimei symp-
loms) lhe annual iisk foi an aneuiysm iup-
luie is eslimaled lo be belween 0.05% foi
aneuiysms smallei lhan 10 mm and 6%
foi gianl aneuiysms (ovei 25 mm). Foi lhe
mosl lypical localions of inliacianial aneu-
iysms see Fig. 9.2. Addilional iisk faclois
foi aneuiysm hemoiihage include aileiio-
scleiosis, hypeilension, and cigaielle and
alcohol abuse. The moilalily foi aneuiysm
induced SAHs is up lo 50% in lhe fiisl 30
days. The iebleeding iisk fiom iupluied,
unliealed aneuiysms is 20-50% (up lo 15%
in lhe fiisl 24 h, lhen daily 1-3%).
Recenl evalualions [1-4] show an oveiall
sensilivily foi inliacianial aneuiysm de-
leclion in acule subaiachnoid hemoiihage
of 85-95% when compaied lo DSA. Limi-
lalions foi MS-CTA aie small aneuiysms
wilh a maximum size belween 1 and 3 mm,
aneuiysms in lhe posleiioi fossa and an-
euiysms in lhe caveinous sinus (supiaoph-
lhalmic caiolid aneuiysms). Besides, nega-
live DSA-findings foi aneuiysmal SAH is
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.33-0.5
Tube current time product (mAs) l35-200 l35-200 l35-240
Pitch corrected tube current time
product (mAs)
l50-200 l50-200 l50-200
Collimation (mm) l/l.25 0.625/0.75 0.6/0.625
Norm. pitch 0.9-l 0.9-l 0.9-l.2
Peconstr.-slice increment (mm) l 0.6 0.5
Peconstr.-slice thickness (mm) l.25 0.75-l 0.6-0.75
Peconstruction Kernel Standard Standard Standard
Scan range Sutura sagittalis to Poramen magnum for strict in-
tracranial aneurysm. Sutura sagittalis to arcus aortas
in order to include the extracranial parts of vA and
carotids.
Scan direction Caudocranial Caudocranial Caudocranial
Contrast media appIication
Concentration (mg iodine/mL) 300-400 300-400 300-400
Mono/biphasic Monophasic Monophasic Monophasic
volume (mL) 80-l20 80-l00 80-l00
|n|ection rate (mL/s) 3-4 3-4 3-4
Saline chaser (mL, mL/s) 30/3.0 30/3.0 30/3.0
Delay (s) l5-20 (l2-l5 for extracranial carotid and vA) or au-
tomatic bolus detection if available
7abIe 9.1. Scan parameters
83 9 C7A of IntracraniaI Aneurysm
0 No blood detectable
l Lumbar puncture positive, CT negative
2 8lood clot visible
3 8lood clots thicker than 3 mm
4 |ntraventricular or intraparechnchymal hemorrhage
7abIe 9.2. Pisher grading system of acute subarachnoidal bleeding
7abIe 9.3. world Pederation of Neurologic Surgeons (wPNS) SAH grade
WINS grade GIasgow Coma ScaIe Score Major focaI deficit
"
0 (intact aneurysm) - -
l l5 Absent
2 l3-l4 Absent
3 l3-l4 Present
4 7-l2 Present or absent
5 3-6 present or absent
`
Aphasia and/or hemiparesis or hemiplegia
Fig. 9.2. Circle of willis, localization and sta-
tistical distribution of cerebral aneurysm. Most
common are aneurysms of the anterior commu-
nicating artery, and the A. cerebri media (MCA),
followed by aneurysms at or near the origin of
the posterior communicating artery (PCoA), and
the basilar tip. Pelatively seldom, especially dif-
ficult to detect are aneurysms of the posterior
circulation in the infratentorial region
15-20% al fiisl examinalion and below 5%
when a palienl is admilled foi a second ad-
dilional angiogiaphy (Fig. 9.3).
Impoilanl ciileiia lo desciibe a find-
ing of an aneuiysm in CTA include lhe
ielalionship of lhe aneuiysm lo lhe paienl
vessel (small neck oi wide neck, fusifoim,
dissecling), lhe size and foim of lhe aneu-
iysm, piesence oi absence of a baby aneu-
iysm, piesence oi absence oi calcificalions,
lhiombosis; and vessels oiiginaling fiom
lhe aneuiysm ilself. To gel lhis infoima-
lion, veiy delailed infoimalion is neces-
saiy. This desciiplion in luin will assisl
lhe decision, if (in incidenlal aneuiysms)
oi by which means (in iupluied aneu-
iysms) lhe liealmenl will be allempled. In
geneial, endovasculai liealmenl is caiiied
oul by filling lhe aneuiysm wilh delach-
able plalinum coils (e.g., GDC), deliveied
lhiough a miciocalhelei. Aneuiysms op-
limal foi lhis endovasculai liealmenl have
a small neck and aie localed in lhe ACA,
ICA oi VA/BA leiiiloiy. Endovasculai slenl
applicalion may be necessaiy, if an aneu-
iysm is fealuiing a ielalively bioad neck.
The neuiosuigical liealmenl is peifoimed
84 II 8rain
wilh lhe placemenl of a clip along lhe neck
of lhe aneuiysm. Candidales oplimal foi
lhis pioceduie aie localed in lhe MCA oi
ACA leiiiloiy. Regaidless of lhe aneuiysm,
liealmenl pievenlion and liealmenl of sec-
ondaiy complicalions like vasospasm via
hypeidynamic lheiapy aie also oflen nec-
essaiy.
Impoilanl diffeienlial diagnoses of lhe
subaiachnoidal hemoiihage aie all foims
of nonaneuiysmal SAH (occull liauma,
vasculilis, disseclion, peiimesencephalic
nonaneuiysmal SAH, vasculai malfoima-
lion, neoplasm). These foims can exhibil
low densily in biain scans iepiesenling
diffuse ceiebial edema, and may have high
densily CSF (e.g., following inlialhecal
conliasl).
The piolocol can be exlended lo assess
suspecled aileiiovenous malfoimalions
(AVM). In lhese cases, CTA can give infoi-
malion on lhe majoi feeding vessels, lhe
AVM nidus sile and size, and lhe supeificial
oi deep diaining veins. All of lhese aie im-
poilanl lo lhe giading of lhe AVMs, which
aie done in mosl inslilulions by lhe giading
syslem of Spelzlei and Mailin (see Table
9.4). Howevei, foi lhe liealmenl decision
lo iecommend liealmenl by eilhei gamma
knife iadialion, endovasculai occlusion,
Fig. 9.3. a Left-sided, superior- and lateral-orientated MCA-bifurcation aneurysm with a maximum
dome size of 4 mm and a neck size of 2 mm in a patient with mild acute SAH. This image demon-
strates the good delineation of neck and dome (l6-row MSCT, axial M|P-reconstruction with l.3-mm
thickness and l.0-mm increment). b Coronar M|P-reconstruction (l.3l.0 mm) of the same patient
showing the superiolateral orientation of the aneurysm. c DSA (matrix l024l024), selective in|ec-
tion of left internal carotid artery, pa-orientated and zoomed of the same patient. DSA demonstrates
the good correlation between MS-CTA and DSA
a c
b
85 9 C7A of IntracraniaI Aneurysm
suigeiy oi combinalions lheieof, lhe in-
vasive seleclive mulliplanai DSA piesenlly
iemains lhe melhod of choice.
References
1. Moihaid D, Biuening R, Eill-Wagnei B, Rei-
sei M, Biueckmann H (2004) MS-CTA vs. DSA
in acule subaiachnoidal haemoiihage. RSNA
Book of Absliacls, p 688
2. Jayaiaman MV, Mayo-Smilh WW, Tung GA,
Haas RA, Rogg JM, Mehla NR, Dobeislein CE
(2004) Deleclion of inliacianial aneuiysms:
mulli-delecloi iow CT angiogiaphy compaied
wilh DSA. Radiology 230(2):510-518
3. Pedeisen HK, Bakke SJ, Hald JK, Skalpe IO,
Anke IM, Sagsveen R, Langmoen IA, Linde-
gaaid KE, Nakslad PH (2001) CTA in palienls
wilh acule subaiachnoid haemoiihage. A com-
paialive sludy wilh seleclive, digilal angiogia-
phy and blinded, independenl ieview. Acla Ra-
diol 42(1):43-49
4. Kalo Y, Kalada K, Hayakawa M, Nakane M,
Oguia Y, Sano K, Kanno T (2001) Can 3D-CTA
suipass DSA in diagnosis of ceiebial aneuiysm?
Acla Neuiochii (Wien) 143(3):245-250
5. Fishei CM (1975) Clinical syndiomes in ceie-
bial lhiombosis, hypeilensive haemoiihage
and iupluied sacculai aneuiysm. Clin Neuio-
suig 22:117-147
7abIe 9.4. Spetzler and Martin grading system for AvMs
Size of AVM Ioquence of adjacent brain Venous drainage
Small (<3 cm), l point Noneloquent, 0 points Superficial only, 0 points
Medium (3-6 cm), 2 points Lloquent, l point Deep drainage, l point
Large (>6 cm ), 3 points
87
10 Imaging of the CerebraI Veins
and Sinuses with MS-C7
B. Eill-Wagnei
Indications
for CerebraI C7-Venography
E To diagnose oi iule oul ceiebial venous
lhiomboses (CVT).
E To piecisely evaluale lhe sile and ex-
lenl of a ceiebial venous occlusion in
CVT.
E To evaluale lhe degiee of compiession of
lhe ceiebial veins and sinuses in palienls
wilh an inliacianial neoplasm, especial-
ly in palienls wilh meningioma.
E To non-invasively evaluale lhe venous
diainage in palienls wilh ceiebial aile-
iiovenous malfoimalions.
Patient Preparation
and Positioning
Patient Preparation
E The palienl should undeigo infoimed
consenl aboul lhe iisks of conliasl medi-
um applicalion as is slandaid piaclice.
E Infoimalion iegaiding ienal and lhyioid
funclion (ciealinine and TSH) should be
oblained.
E Infoimalion aboul lhe palienl's cuiienl
medicalion should be available (e.g.,
melfoimine).
E A laige peiipheial venous access, e.g., in
an anlecubilal vein, should be oblained.
Patient Positioning
E The palienl should be posilioned supine
on lhe CT examinalion lable.
E The palienl's head should be posilioned
in lhe headiesl.
E If lhe ganliy cannol be lilled, il is advan-
lageous lo posilion lhe palienl's head in
30 flexion foi an inliacianial examina-
lion.
7opogram and Scan Range
E The scan iange should ieach fiom lhe
lowei end of lhe skull base lo lhe veilex
(see Fig. 10.1).
E The uppei boundaiy of lhe scan iange
always needs lo include lhe enliie supe-
iioi sagillal sinus; caie should be laken
lo include lhe mosl cianial poilion of
lhe skull.
E The lowei boundaiy of lhe scan iange
should include lhe sigmoid sinus.
E The scan iange should include lhe pe-
lious bones in oidei lo evaluale a possible
masloidilis in lhe selling of seplic CVT.
E Il is usually nol necessaiy lo include lhe
enliie jugulai veins in lhe scan iange,
since lhese can be imaged wilh Dopplei
and duplex sonogiaphy. Howevei, should
doubl pievail, lhe neck iegion can also
be included lo image lhe ceivical veins.
Fig. 10.1. This lateral topogram demonstrates
the normal scan range for a CT-angiographic
evaluation of cerebral veins and sinuses
88 II 8rain
Scan Parameters
See Table 10.1.
7ips and 7ricks
E Ceiebial venogiaphy can usually be
quile ieliably and easily peifoimed wilh
a fixed delay of 40 lo 45 s and a bolus of
conliasl medium of 120 mL.
E If lhe palienl's symplomalology is nol
cleai, a simullaneous conliasl of bolh
lhe inliacianial aileiies and lhe inlia-
cianial veins can be helpful, especially
in lhe emeigency selling. A slighlly ie-
duced delay of aboul 35 s will usually
piovide an even conliasl of bolh lhe in-
liacianial aileiies and lhe inliacianial
veins.
E Caie should be laken lo evaluale lhe pe-
lious bones, especially in lhe selling of
a CVT of lhe sigmoid and liansveise si-
nuses, as lhese can polenlially iepiesenl
seplic lhiomboses iesulling fiom a pu-
iulenl masloidilis.
E Il is impoilanl lo diffeienliale meie hy-
poplasia of lhe liansveise sinus, which
is a common vaiianl, fiom liue CVT.
Moieovei, aiachnoid gianulalions need
lo be iecognized and diffeienlialed fiom
CVT.
E The nonenhanced cianial CT should
always be sciulinized as well, since lhe
localion, exlenl and conloui of edema
and/oi hemoiihage as secondaiy signs
of a ceiebial venous lhiomboiis can
piovide impoilanl infoimalion aboul
lhe polenlial sile of lhe venous occlu-
sion.
Comments
The mosl common indicalion lo image lhe
inliacianial veins and sinuses is lo iule oul
CVT. CVT can piesenl wilh ielalively un-
specific signs and symploms, which vaiy
subslanlially fiom asymplomalic lo seveie
symploms iequiiing inlensive caie lieal-
menl. Olhei diffeienlial diagnoses aie of-
len conlemplaled by lhe iefeiiing clinician
as well [1,2]. Please keep in mind lhal CVT
is a disease wilh a moilalily of up lo 10%
even in lhe piesence if inlensive caie and
lhe slandaid piaclice of iv. Hepaiine.
The compuled lomogiaphic evalualion
of CVT usually consisls of a nonenhanced
CT (NECT) scan followed by a CT-angio-
giaphic examinalion. The NECT should al-
ways be caiefully examined as lhe piesence
of edema and/oi hemoiihage as well as ils
iespeclive size, localion and shape can pio-
vide impoilanl infoimalion iegaiding lhe
exlenl and localion of lhe venous occlusion.
Howevei, a noimal NECT does nol iule oul
lhe piesence of CVT.
Theiefoie, a CT-angiogiaphic exami-
nalion of lhe ceiebial veins and sinuses is
usually peifoimed nexl. The scan iange
should always include lhe enliie ceiebial
venous vasculaluie. Il is lheiefoie impoi-
lanl lo caiefully include lhe uppeimosl
poinl of lhe veilex in oidei lo image lhe
enliie supeiioi sagillal sinus. Coionai and
sagillal ieconsliuclions aie impoilanl
heie and should be peifoimed ioulinely.
The lowei boundaiy of lhe scan iange
should include lhe skull base wilh lhe pe-
lious bones and lhe enliie posleiioi fossa.
A fixed delay of 40 lo 45 s usually ie-
sulls in a homogeneous conliasl of lhe
inliacianial venous vessels. In oui expe-
iience, il is fiequenlly helpful lo allain a
mixed aileiial and venous conliasl as lhe
symplomalology of lhe palienl is oflen nol
specific foi a venous palhology. A fixed de-
lay of 35 s usually piovides a homogeneous
conliasl of bolh lhe aileiies and lhe veins,
when a bolus of 120 mL of conliasl me-
dium is applied [3,4]. The scan diieclion
should usually follow lhe flow." Thus,
when imaging lhe ceiebial veins and si-
nuses, lhe diieclion of acquisilion should
be cianiocaudal.
When evalualing a CT-venogiaphic ex-
aminalion in a palienl wilh suspecled CVT,
il is impoilanl lo look foi lhe sile and lhe
exlenl of lhe venous occlusion (Fig. 10.2).
As menlioned above, lhe sile of paienchy-
mal edema and/oi hemoiihage can piovide
impoilanl infoimalion iegaiding lhe oc-
cluded vessel. The CT-venogiaphic exami-
nalion geneially allows a diiecl visualiza-
89 10 Imaging of the CerebraI Veins and Sinuses with MS-C7
lion of lhe lhiombolic maleiial wilhin lhe
occluded (oi pailially lhiombosed) ceie-
bial vein oi sinus, which is iefeiied lo as
lhe coid sign [5]. While lhe suiiounding
vessel is filled wilh conliasl maleiial, lhe
lhiombolic maleiial is ielalively hypodense
compaied lo lhe conliasl enhanced sinus
oflen desciibed as lhe emply liiangle sign
(Fig. 10.3).
Il is impoilanl lo diffeienliale CVT fiom
simple aiachnoid gianulalions. In con-
liasl lo lhe moie amoiphous sliucluie of
a venous lhiombus lhal lends lo fill laige
pails of lhe venous sliucluie, an aiachnoid
gianulalion usually has a densily iesem-
bling lhal of fal, is compaialively small and
iound, and is well demaicaled. Aiachnoid
gianulalions moieovei lend lo be lobulaled
7abIe 10.1. Scan parameters for MSCT-imaging of the cerebral veins and sinuses
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
64 sIice
scanners
Scanner settings
Tube voltage (kv) l00-l20 |6| l00-l20 |6| l00-l20 |6|
Potation time (s) 0.5 0.5 0.33-0.5
Tube current time product (mAs) l50-200 l50-200 l50-200
Pitch corrected tube current time
product (eff. mAs)
l50-200 l50-200 l50-200
Tube current (mAs) l00-200 l00-200 l00-200
Collimation (mm) l 0.75 0.65
Norm. pitch l l l
Peconstr.-slice increment (mm) 0.75-2 0.6-2 0.5-2
Peconstr.-slice thickness (mm) l-3 mm l-3 mm l-3 mm
Peformation technique Sliding thin
slab (STS) M|P
STS-M|P STS-M|P
Scan range 8ase of skull
to vertex
8ase of skull
to vertex
8ase of skull
to vertex
Scan direction Craniocaudal Craniocaudal Craniocaudal
Peconstruction kernel Standard Standard Standard
Contrast media appIication
Concentration (mg iodine/mL) 300 300 300
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) l20 l20 l20
|n|ection rate (mL/s) 4-5 |7| 4-5 |7| 4-5 |7|
Saline chaser (mL, mL/s) 30/4.0 30/4.0 30/4.0
Delay (s)
E venous
E Combined arterial and venous
Pixed delay
40-45 s
Pixed delay
35 s ot
Autom. 8olus-
Detection +l2 s
Pixed delay
40-45 s
Pixed delay
35 s ot
Autom. 8olus-
Detection +l2 s
Pixed delay
40-45 s
Pixed delay
35 s ot
Autom. 8olus-
Detection
+l2 s
90 II 8rain
and can be found anywheie, bul piedomi-
nanlly in chaiacleiislic localions such as
lhe liansveise and supeiioi sagillal sinus.
In addilion, il is ciucial lo diffeienliale
meie hypoplasia of lhe sinus fiom liue
venous lhiombosis. The liansveise sinus
can commonly display an asymmeliic con-
figuialion wilh ielalive hypoplasia of one
side, jusl consliluling a noimal vaiianl [2].
Caie should be laken nol lo mislake such
a noimal vaiianl foi a lhiombosed vessel.
In lhis selling, il is impoilanl lo look foi
diiecl manifeslalions of lhe lhiombus, i.e.,
lo look foi lhiombolic maleiial wilhin lhe
vessel. Il can moieovei be helpful lo seaich
foi indiiecl signs of venous lhiombosis
such as edema oi hemoiihage of lhe biain
paienchyma.
A CT-angiogiaphic examinalion of lhe
ceiebial veins and sinuses wilh modein
mullislice CT scanneis now allows foi lhe
visualizalion of even small venous sliuc-
luies such as biidging veins oi lhe inleinal
ceiebial veins. Howevei, despile piomising
iepoils, lhe sensilivily of MDCT lo delecl
oi exclude supeificial venous lhiombosis
is nol ceilain. Also, in oidei nol lo miss lhe
diagnosis of a seleclive lhiombosis of lhese
veins, il is impoilanl lo be familiai wilh
lhe inliacianial venous analomy and lo ac-
lively seaich foi lhe small vasculai sliuc-
luies. Again, lhe piesence and localion
of edema and/oi hemoiihage wilhin lhe
biain paienchyma can piovide impoilanl
clues iegaiding lhe sile of lhe lhiombosis.
CT-venogiaphic examinalions wilh
mullislice CT can moieovei piovide in-
foimalion iegaiding lhe degiee of venous
congeslion. Thiee-dimensional iecon-
sliuclions may help in selecled cases lo
impiove lhe visualizalion of a lhiombus
(Fig. 10.4). The smallei venous sliuc-
luies diaining inlo lhe lhombosed vessel
aie commonly enlaiged as a iesull of lhe
oblileialed venous lumen. In addilion,
in lhe selling of a moie longslanding ve-
nous obsliuclion, lhe degiee of collaleial
ciiculalion can be assessed. This can also
conslilule impoilanl infoimalion in lhe
selling of a compiession of a ceiebial si-
nus, which can oflen occui in lhe selling
of meningiomas (Fig. 10.5). The supeiioi
sagillal sinus is an especially common
sile of venous compiession, secondaiy lo
peiifalcine meningiomas. In lhis selling,
lhe degiee of venous compiession and lhe
piesence oi absence of a collaleial ciicula-
lion should be desciibed.
When a lhiombosis of lhe sigmoid and/
oi liansveise sinus is found, il is especially
Fig. 10.2. Sagittal sliding thin slab (STS) maxi-
mum intensity pro|ections (M|P) demonstrate a
complete thrombotic occlusion of the superior
sagittal sinus
Fig. 10.3. Coronal STS-M|P reformations dem-
onstrate an occlusion of the superior sagittal
sinus (ottows). Note the presence of edema
and hemorrhage on the right-hand side (ot-
towheoJs). |n this case, the entire cervicocranial
vascular system was examined, as the thrombus
extended into the internal |ugular vein
91 10 Imaging of the CerebraI Veins and Sinuses with MS-C7
impoilanl in childien and young adulls lo
also assess lhe pelious bones foi lhe po-
lenlial piesence of a masloidilis, as lhese
lhiomboses commonly iepiesenl seplic
piocesses as a consequence of a masloidal
infeclion. The dalasel should be iecon-
sliucled in a bone keinel in lhis selling in
oidei nol lo miss lhis impoilanl addilional
infoimalion. Finally, il has lo be menlioned
lhal lhe deleclion and exclusion of iecui-
ienl/iemilling CVT may slill iepiesenl a
diagnoslic pioblem bolh in venous CT and
MRI.
In summaiy, CT angiogiaphy of lhe
ceiebial veins and sinuses wilh mullislice
CT piovides an excellenl lool in lhe diag-
nosis of CVT. Il nol only diieclly depicls
lhe piesence and localion of lhe venous
lhiombus, bul also demonsliales lhe de-
giee of venous congeslion and/oi collal-
eializalion and may piovide impoilanl
addilional infoimalion, e.g., iegaiding lhe
piesence of an inf lammaloiy piocess in
lhe masloid.
References
1. Hagen T, Bailylla K, Waziii A, Schmilz B, Piep-
gias U (1996) Value of CT-angiogiaphy in diag-
nosis of ceiebial sinus and venous lhiomboses.
Radiologe 36:859-866
2. Haage P, Kiings T, Schmilz-Rode T (2002) Non-
liaumalic vasculai emeigencies: imaging and
inleivenlion in acule venous occlusion. Eui Ra-
diol 12:2627-2643
3. Klingebiel R, Zimmei C, Rogalla P el al. (2001)
Assessmenl of lhe aileiiovenous ceiebiovas-
culai syslem by mulli-slice CT. A single-bolus,
monophasic piolocol. Acla Radiol 42:560-562
4. Eill-Wagnei B, Hoffmann RT, Biuning R, Dich-
gans M, Reisei MF (2002) Diagnoslic evalualion
of lhe cianioceivical vasculai syslem wilh a
16-slice mulli-delecloi iow spiial CT. Piolocols
and fiisl expeiiences. Radiologe 42:728-732
5. Casey SO, Albeiico RA, Palel M el al. (1996) Ce-
iebial CT venogiaphy. Radiology 198:163-170
6. Eill-Wagnei BB, Hoffmann RT, Biuning R el al.
(2004) Mulli-delecloi iow CT angiogiaphy of
lhe biain al vaiious kilovollage sellings. Radi-
ology 231:528-535
7. Cademailiii F, van dei Lugl A, Luccichenli G,
Pavone P, Kieslin GP. Paiameleis affecling bo-
lus geomeliy in CTA: a ieview. J Compul Assisl
Tomogi 26:598-607
Fig. 10.4. volume rendering technique images
of the same patient demonstrate a congestion
of the small vessels draining into the throm-
bosed superior sagittal sinus
Fig. 10.5. Sagittal STS-M|P reformations in a
patient with a transosseous meningioma dem-
onstrate a compression of the superior sagittal
sinus by the tumor (ottows). A collateral circula-
tion is present (ottowheoJs)
93
11 8rain Perfusion
M. Wiesmann
Indications
E Acule ischemic ceiebial infaiclion.
E Chionic slenosis of supiaaoilic oi inlia-
cianial aileiies.
E Cuiienlly evalualed indicalions include
vasospasm aflei subaiachnoid hemoi-
ihage and evalualion of ceiebial neo-
plasms.
Patient Preparation
Laige-calibei cubilal venous access is iec-
ommended (piefeiably 16 oi 18 G).
Patient Positioning
E Supine, aims downwaid, use of headiesl
iecommended.
E If lilling lhe ganliy duiing conlinuous
scan mode is nol possible, lhe head of
lhe palienl should be inclined lo piolecl
lhe lenses fiom iadialion.
7opogram/Scan Range
See Fig. 11.1.
7abIe Scan Parameters
See Table 11.1.
7ips and 7ricks
E Insliucl lhe palienl befoie lhe acquisi-
lion lo lie slill allhough lhe bolus injec-
lion of conliasl maleiial may cause dis-
comfoil.
E Injeclion of a saline chasei following lhe
conliasl media impioves image qualily.
E If lhe fiisl sludy is negalive, up lo lwo
addilional peifusion sludies can be pei-
foimed lo covei olhei clinically suspecl-
ed biain aieas.
Comments
Peifusion CT allows accuiale quanlilalive
assessmenl of biain lissue peifusion, is
well loleialed, and is nol lime-consum-
ing. Il has become a valuable lool in lhe
imaging of acule slioke foi lwo ieasons:
(1) aieas of impaiied peifusion can be de-
Fig. 11.1. Perfusion studies in acute stroke are
centered at the level of the basal ganglia to in-
clude portions of those vascular territories most
likely affected from ma|or vessel occlusion (i.e.,
anterior, middle, and posterior cerebral arteries)
94 II 8rain
lecled iighl aflei lhe onsel of slioke, and
(2) lheie is evidence lhal if lhe infoima-
lion fiom a noimal biain CT and peifu-
sion paiamelei maps aie combined, il is
possible lo disciiminale belween iiieveis-
ibly damaged and polenlially salvageable
lissue [1,2]. Peifusion CT can lheiefoie
suppoil lheiapeulic slialegies based on
individual assessmenl of biain peifu-
sion in slioke palienls ialhei lhan using
iigid lime inleivals ielaled lo lhe onsel of
symploms [3].
Sliokes aie lhe lhiid leading cause of
dealh aflei caidiovasculai diseases and
canceis as well as a leading cause of seii-
ous disabilily. Thiombolysis, adminisleied
eilhei inliavenously oi inliaaileiially, has
been appioved as an effeclive lheiapy foi
acule human slioke. Il is inlended lo ies-
cue lhe penumbia and lo ieduce lhe final
infaicl size, and lhus lhe iesulling handi-
cap. Al piesenl, lhe seleclion of palienls foi
lhiombolylic lheiapy is mainly ielaled lo
lhe lime inleival since lhe onsel of symp-
loms (less lhan 3 lo 6 h, depending on lhe
applicable piolocol), lhe absence of ceie-
bial hemoiihage, and lhe infaicl size (ce-
iebial hypodensily on CT exlending lo
less lhan one-lhiid of lhe middle ceiebial
aileiy iegion). Howevei, even wilh such
iesliiclive ciileiia nol all palienls selecled
benefil fiom lhiombolylic lheiapy and
lhiombolysis ilself beais a significanl iisk
of inliacianial bleeding. Theiefoie il has
been pioposed lo lake inlo consideialion
lhe individual hemodynamic silualion of
slioke palienls.
To deiive funclional infoimalion on lhe
hemodynamic slalus of lhe biain, sequen-
lial CT slices aie acquiied in cine mode
duiing inliavenous conliasl adminislia-
lion. Foi each pixel, lime-densily piofiles
7abIe 11.1. Scan parameters for MSCT measurement of brain perfusion
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) 80 80 80
Tube current (mAs) l20-240 l20-240 l20-240
Collimation (mm) 2l0 / 48 2l0 / 48 2l0 / 48
Norm. pitch 0 0 0
Potation time (s) l.0 l.0 l.0
Peconstruction increment (mm) 0 0 0
Peconstruction thickness (mm) l0 l2 l2
Convolution kernel Special Special Special
Specials Dynamic scan:
45 scans
(l scan/s)
Dynamic scan:
45 scans
(l scan/s)
Dynamic scan:
45 scans
(l scan/s)
Contrast media appIication
Concentration (mg iodine/ml) 370-400 mg/mL 370-400 mg/mL 370-400 mg/mL
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) 40 40 40
|n|ection rate (mL/s) 5-l0 5-l0 5-l0
Saline chaser (mL, mL/s) 40/5.0 40/5.0 40/5.0
Delay (s) 5 (fixed) 5 (fixed) 5 (fixed)
95 11 8rain Perfusion
of conliasl enhancemenl aie oblained us-
ing dedicaled poslpiocessing soflwaie.
Fiom lhese cuives, seveial peifusion-ielal-
ed paiameleis aie calculaled and displayed
as coloi-coded paraneier naps. The pei-
fusion paiameleis mosl commonly used
include ceiebial blood flow (CBF), ceiebial
blood volume (CBV), lime lo peak (TTP),
and lhe mean liansil lime (MTT) (see be-
low) [3].
The CBF noimally aveiages 50 lo 80 mL
blood pei 100 mg biain lissue pei min-
ule. Iiieveisible infaiclion develops if lhe
CBF diops below 10-15 mL/100 mg/min
foi moie lhan 2-10 min. Biain aieas pei-
fused al CBF levels belween 10 and 25 mL/
100 mg/min cease neuiological funclion,
bul may complelely iecovei if noimal lev-
els of peifusion aie iesloied wilhin houis
oi days. In mosl cases of ischemic infaic-
lion, an iiieveisibly damaged infaicl coie
is suiiounded by a vaiiable zone of im-
paiied peifusion, lhe penunbra. Theia-
peulic measuies in slioke aie nol aimed
al lhe infaicl coie, bul lo iesloie noimal
peifusion levels in lhe penumbia and lhus
ieduce lhe final infaicl size. Il should be
noled lhal CBF values in lhe lileialuie iefei
lo highly vasculaiized biain iegions (e.g.,
basal ganglia, coilical suiface), wheie lhe
CBF is lwo lo lhiee limes highei lhan in
lhe ceiebial while mallei.
The CBV is defined as lhe peicenlage of
inliavasculai volume in a given biain aiea.
Highly vasculaiized biain iegions (e.g.,
basal ganglia, coilical suiface) have a high-
ei CBV lhan lhe ceiebial while mallei. Bul
lhe CBV is also a funclional measuie and
changes wilh vasculai auloiegulalion and
olhei faclois. While lhe CBF can be loweied
in bolh lhe penumbia and lhe infaicl coie,
lhe iegional CBV is usually incieased in lhe
penumbia as a iesull of local vasodilala-
lion, in an allempl by lhe auloiegulalion
piocesses lo compensale foi iegional ce-
iebial blood flow loweiing. In lhe infaicl,
auloiegulalion piocesses aie compiomised
and lhe iegional ceiebial blood volume is
loweied. Low CBV values in ischemic aieas
aie usually piediclive of iiieveisible dam-
age (Fig. 11.2).
Diffeienl paiameleis aie available lo de-
sciibe lhe lempoial couise of lhe conliasl
bolus, i.e., lo iecognize delayed peifusion.
Of lhese, lhe mosl widely used aie MTT and
TTP. These paiameleis aie highly sensilive
foi impaiied peifusion bul unspecific foi
infaiclion. Palhologic MTT oi TTP values
aie usually found in bolh infaicl coie and
penumbia, bul may also be caused by piox-
Fig. 11.2. Territorial infarction with penumbra and thrombolysis. 50-year-old patient suffering from
acute left-sided hemiparesis. Native initial CT scans (a) and PCT (b-d) were performed 2 h after the
onset of symptoms. PCT scans were evaluated using a deconvolution algorithm (C8P, Toshiba Medi-
cal Systems, Germany). No early signs of infarction were identified on the initial CT study. On the
MTT and C8P parameter maps (b,c) a large area of focal hypoperfusion can be identified, which
includes 2/3 of the territory of the right middle cerebral artery. within the lesion no decrease in C8v
is found (d). Therefore the complete lesion was considered as tissue at risk. Systemic thrombolysis
was performed. |nfarction as seen on the native follow-up CT scan (e) was confined to a rather small
hypodense lesion in the right external capsule (modified from wiesmann et al. |3|, courtesy G. 8oh-
ner, 8erlin)
96 II 8rain
imal aileiial slenosis (e.g., in lhe inleinal
caiolid aileiy) oi vasospasm (Fig. 11.3).
7echnicaI Considerations
Cycle iines: cycle limes of 1 image/s aie
adequale lo cieale lime allenualion cuives
fiom each pixel lhal can ieliably liace lhe
conliasl changes even duiing lhe veiy shoil
iise lime of a compacl bolus. Shoilei cycle
limes inciease lhe iadialion exposuie of
lhe palienl bul do nol impiove lhe qualily
of lhe iesulls significanlly.
Slice ihickness: dynamic peifusion CT
sludies can be oblained fiom CT scanneis
iegaidless of lheii mullislice iow numbeis.
To ieduce image noise a ieconsliuclion
lhickness pei slice of 8-10 mm is selecled
[4]. Using mullislice scanneis lhis can be
achieved by adding logelhei lhe signal of
adjacenl delecloi channels. The availabil-
ily of mullidelecloi lechnology, howevei,
has nol significanlly incieased lhe infoi-
malion deiivable fiom a single iun. This
is because foi lhe dynamic scan mode lhe
scan volume is given by lhe lolal size of lhe
delecloi aiiay and nol by lhe numbei of
channels available. If lhe lolal size of lhe
delecloi aiiay is 32 mm, foui seclions of
8-mm slice lhickness can be acquiied si-
mullaneously. Fiom a clinical peispeclive,
howevei, if a majoi slioke is suspecled il is
unlikely lhal foui adjacenl slices will yield
significanlly moie diagnoslic infoima-
lion lhan a single 8-mm slice al lhe level of
lhe basal ganglia. On lhe olhei hand lhis
means an incieased iadialion exposuie
of lhe palienl. Inslead, if lhe fiisl sludy
is negalive, up lo lwo addilional conliasl
injeclions can be peifoimed lo covei olh-
ei clinically suspecled aieas of lhe biain
(e.g., moloi coilex) [5].
Conirasi bolus: due lo lhe shoil ciiculalion
lime wilhin lhe ceiebial vasculaluie lhe ie-
sulls of peifusion CT iely heavily on lhe ad-
minislialion of a veiy compacl conliasl bo-
lus. Since lhe injeclion lime has lo be shoil,
ideally in lhe iange of 4-6 s, lhe injeclable
bolus volume is limiled. A laige-calibei cu-
Fig. 11.3. Subarachnoid hemorrhage with vasospastic hypoperfusion. 45-year-old patient suffer-
ing from subarachnoid hemorrhage from a ruptured arterysm of the anterior communicating ar-
tery. vasospasm of both anterior cerebral arteries and the left-sided middle cerebral artery was con-
firmed by angiography. PCT scans were obtained eight days after hemorrhage and evaluated using
the maximum slope model (Perfusion CT, Siemens, Germany). On the TTP parameter map (a) focal
areas of delayed perfusion with normal C8P (b) and C8v (c) values can be depicted bilaterally in the
territory of the anterior cerebral arteries, and in the territory of a temporal branch of the left middle
cerebral artery. Note that there is a small area of severely delayed perfusion in the left frontal white
matter (ad|acent to the frontal horn of the lateral ventricle). This small area also shows a reduced
C8P and C8v indicating focal infarction. Native CT scans at this time (not shown) confirmed a small
infarction at this location whereas the other areas did not show ischemic edema
97 11 8rain Perfusion
bilal venous access and high injeclion iales
(5-10 mL/s) aie mandaloiy lo achieve good
iesulls. The use of piewaimed highly con-
cenlialed conliasl agenls impioves image
qualily.
Posiprocessing sofiware: lwo majoi malh-
emalical appioaches aie used lo calculale
coloi-coded paiamelei maps [3]. Decon-
voluiion algoriihns, which aie cuiienlly
used, e.g., by Geneial Elecliic and Toshi-
ba, geneially yield quanlilalively moie
accuiale iesulls al low injeclion iales. On
lhe olhei hand, calculalion lakes longei
and an aileiial inpul funclion has lo be
deleimined. The naxinun slope nodel,
which is cuiienlly used, e.g., by Siemens
and Viliea, is easiei lo calculale and less
pione lo molion ailifacls, bul is also con-
sideied malhemalically less piecise. Bolh
appioaches, howevei, have been shown lo
yield ieliable iesulls in lhe clinical selling
of acule slioke [5,6].
References
1. Bohnei G, Foischlei A, Hamm B, Lehmann R,
Klingebiel R (2000) Quanlilalive peifusion im-
aging by mulli-slice CT in slioke palienls. Rofo
Foilschi Geb Ronlgensli Neuen Bildgeb Vei-
fahi 175:806-13
2. Winleimaik M, Bogousslavsky J (2003) Imag-
ing of acule ischemic biain injuiy: lhe ieluin
of compuled lomogiaphy. Cuii Opin Neuiol
16:59-63
3. Wiesmann M, Bohnei G, Klingebiel R (2004)
Ceiebial peifusion imaging using mullislice
compuled lomogiaphy. Clin Neuioiadiol 14:92-
107
4. Knig M (2003) Biain peifusion CT in acule
slioke: cuiienl slalus. Eui J Radiol 45:S11-22
5. Mayei TE, Hamann GF, Baianczyk J, Rosen-
gailen B, Klolz E, Wiesmann M, Misslei U,
Schulle-Alledoinebuig G, Bickmann H (2000)
Dynamic CT peifusion imaging of acule slioke.
AJNR Am J Neuioiadiol 21:1441-9
6. Winleimaik M, Reichhail M, Thiian JP, Maedei
P, MD, Chalaion M, Schnydei P, Bogousslavsky
J, Meuli R (2002) Piognoslic accuiacy of ce-
iebial blood flow measuiemenl by peifusion
compuled lomogiaphy, al lhe lime of emei-
gency ioom admission, in acule slioke palienls.
Ann Neuiol 51:417-32
III Neck
101
12 Sinuses and IaciaI SkeIeton
F. Dammann
Indications
E Inflammaloiy disease (sinusilis, pol-
yposis, mukocele, oibilal oi inliacianial
complicalions).
E Pieopeialive planning (FESS funclional
endoscopic sinus suigeiy: idenlify ana-
lomic vaiialions, polenlially dangeious
silualions).
E Midfacial oi cianial liauma, congenilal
defoimalions, lumoi disease (benign
and malignanl lesions, T-slaging, fol-
low-up).
E Pieopeialive woikup foi coiieclive sui-
geiy, planning of compulei-aided sui-
geiy (image fusion, suigical simulalion,
navigalion, iobolics) and manufaclui-
ing of medical models (sleieolilhogia-
phy models foi pieopeialive simula-
lion).
Patient Preparation
No special piepaialion necessaiy; i.v. ac-
cess if CM adminislialion needed.
Patient Positioning
E Supine (all scans aie peifoimed in lhe
axial plane), aims downwaid, use head-
iesl.
E Addilional pione posilion of lhe palienl
is only indicaled foi lhe diagnosis of
oibilal flooi fiacluies when using mul-
lislice CT wilh less lhan 16-iow scan-
ning lo oblain diiecl coional images.
7opogram/Scan Range
See Fig. 12.1.
7abIe Scan Parameters
See Table 12.1.
7ips and 7ricks
Oilhogonal posilioning of lhe palienl's
head simplifies lhe image inleipielalion.
Mulliplanai iefoimalions (polenlially
also including angulaled axial slices) lhal
aie analomically adjusled in all lhiee di-
mensions can be peifoimed on palienls
foi whom oplimal posilioning cannol be
achieved.
Fig. 12.1. Lxamination region including the
frontal sinus to the alveolar ridge (including the
chin, resp., when the mandibula is involved in
facial trauma or surgical procedure)
102 III Neck
A low-dose selling (20-50 mAs) is iec-
ommended foi lhe majoiily of slandaid
sinus CT.
Comments
Slandaid midfacial and/oi sinus CT in-
cludes bolh axial and coional images as
pioposed by Zinieich el al. [1]. Coional
images (Fig. 12.2) aie consideied lo be
moie impoilanl foi pieopeialive planning
because lhey show lhe shape and ielalion-
ship of mosl analomic landmaiks closesl lo
lhe inliaopeialive aspecl (ciibiifoim plale,
osliomealal complex, oibilal walls). Axial
slices (Fig. 12.3) help lo define lhe localiza-
lion and exlenl of disease (liauma, benign
oi malignanl diseases) in lhe anleiioi-pos-
leiioi diieclion, especially when lhe elh-
moid cells, lhe sphenoid sinus, oi lhe oibil
is involved.
When using mullislice CT, coional ief-
oimalions lhal aie diieclly ieconsliucled
oul of lhe spiial iaw dala oi calculaled fiom
a lhin-slice axial dala sel can ieplace lhe
diiecl coional scan. Foi planning lhe po-
silion and iange of coional iefoimalions,
7abIe 12.1. Scan and reconstruction parameter of CT of the paranasal sinuses and the midface
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) <l <l <l
Tube current time product (mAs) 20/200
`
20/90-200
`
20/90-200
`
Pitch corrected Tube
current time product (eff. mAs)
20/220
`
20/l40-200
`
20/l40-200
`
Collimation (mm) l-l.25 0.625-0.75 0.6-0.625
Norm. pitch 0.9 0.6-l.0 0.6-l.0
Peconstruction increment (mm) 0.6 0.5 0.5
Peconstruction slice thickness
(mm)
l.0-l.25 0.75-l 0.6-0.75
Convolution kernel 8one 8one 8one
Specials Multiplanar
reformations,
fac. 3D
reformations
fac. soft tissue
ref.
Multiplanar
reformations,
fac. 3D
reformations
fac. soft tissue
ref.
Multiplanar
reformations,
fac. 3D
reformations
fac. soft tissue
ref.
Scan range Prontal sinus/
alveolar ridge
(mandibula)
Prontal sinus/
alveolar ridge
(mandibula)
Prontal sinus/
alveolar ridge
(mandibula)
Scan direction Craniocaudal Caudocranial Craniocaudal
Contrast media application No No No
`
mAs values exceeding l00 mAs should be reserved for the diagnosis of tumor disease or com-
plications of sinusitis when i.v. contrast material has to be administered. Por the examination
of the sinuses for trauma and benign sinus disease a low-dose protocol seems sufficient (see
text).
103 12 Sinuses and IaciaI SkeIeton
a sagillal view should be used (Fig. 12.2).
The angulalion of coional iefoimalions
should be adjusled peipendiculaily lo lhe
maxillaiy plale. The advanlages of coional
iefoimalions include lhe absence of denlal
melal ailifacls lhal may supeiimpose iele-
vanl analomical sliucluies of lhe midface
in diiecl coional scans, lhe moie comfoil-
able posilion of lhe palienl, and lhe ieduc-
lion of lhe exposuie dose and examinalion
lime by 50% as compaied wilh a conven-
lional lwo-scan CT examinalion. Diiecl
coional scans may slill be indicaled foi lhe
diagnosis of non- oi slighlly dislocaled oi-
bilal f looi fiacluies lhal can be missed on
coional iefoimalions deiived fiom axial
CT scans wilh a less lhan 16-iow scannei
due lo slaii slep ailifacls.
A low-dose selling (20-50 mAs) is iec-
ommended foi lhe majoiily of slandaid
sinus CT examinalions because piedomi-
nanlly young palienls suffeiing fiom be-
nign diseases aie involved and high levels
of image noise can be loleialed due lo lhe
high conliasl of lissue veisus aii. The sui-
face dose can be ieduced lo 3-4 mGy and
lhe effeclive dose lo 0.1-0.2 mSv when us-
ing 20 mAs [2]. A medium dose selling
(100 mAs) is iecommended foi diagnosis
of liauma and lumoi disease, complica-
lions of inf lammaloiy diseases (e.g., oi-
bilal oi inliacianial involvemenl), and foi
3D applicalions (e.g., SSD, volume iendei-
ing, and medical modeling).
I.v. Contrast
I.v. conliasl does nol add ielevanl diag-
noslic infoimalion in lhe majoiily of si-
nus CT examinalions, bul may be helpful
lo delineale lhe exlenl of benign oi ma-
lignanl lesions lhal pass ovei lhe bony
boideis of lhe paianasal sinuses. In lhese
cases a monophasic c.m. applicalion of
50 mL wilh a scan delay of 90-120 s and
a medium dose selling (100-150 mAs)
is iecommended. Beside slandaid bone
iefoimalions (Table 12.1) addilional sofl
lissue dala sels (slice lhickness and incie-
menl 3-5 mm, medium-sofl convolulion
fillei, window 350/50) aie calculaled.
Fig. 12.2. Coronal image: 64-row CT, 50 mAs,
l20 kv, direct coronal reformation, slice thick-
ness l mm. Clear depiction of anatomical de-
tails: cribriforme plate (l), ostiomeatal complex
(2-4), including the bulla ethmoidalis (2), the un-
cinate process (3), the middle turbinate (4), and
the maxillary infundibulum (JosheJ l|ne) extend-
ing from the maxillary ostium (Jot) to the hiatus
semilunaris (ottow heoJ). No stair step artifacts.
Dental metal artifacts do not superimpose the
midface but are horizontally oriented (5), corre-
sponding to the axial scan direction
Fig. 12.3. Axial image: 4-row CT, 20 mAs, l20 kv,
slice thickness 2 mm. Chronic sinusitis with bilat-
eral involvement of the anterior and posterior
ethmoid and the sphenoid sinus
104 III Neck
3D Reconstructions
In liauma diagnosis of lhe facial skelelon
addilional sagillally angulaled iefoima-
lions lhal aie adjusled lo lhe oplical neive
axis may piovide addilional infoimalion in
oibilal f looi liauma (Fig. 12.4). In liauma
diagnosis lhe slice lhickness of any mul-
liplanai iefoimalion should nol exceed
2 mm. Thiee-dimensional iefoimalions
(SSD, volume iendeiing) allow foi an inlu-
ilive undeislanding of complex midfacial
fiacluies (Fig. 12.5). An addilional axial
lhin-slice dala sel using an inleimediale
oi slighlly sofl convolulion fillei is iecom-
mended as lhe basis foi 3D iefoimalions
wilh ieduced image noise. In conliasl,
edge-enhancing filleis aie iecommended
foi lhe manufacluiing of piecise medical
models.
DifferentiaI Diagnosis
CT is lhe mosl ieliable imaging modalily
lo diagnose midfacial liauma. Il allows foi
a delailed analysis of lhe involved sliuc-
luies. Repoils should be foimulaled ac-
coiding lo lhe inleinalional liauma classi-
ficalions ( LeFoil, AO classificalion) when
applicable (Table 12.2). CT is also accepled
as lhe gold slandaid foi paianasal sinus
imaging in benign inflammaloiy disease,
i.e., chionic sinusilis. Repoils should ad-
diess lhe exlenl of lhe disease and lhe
piesence of analomical vaiialions lhal
may polenlially enlail sinusilis oi lhal may
be ciilical foi endoscopic suigeiy [3,4]. Ill-
defined sofl lissue is geneially designaled
as sofl-lissue opacificalion since CT does
nol allow fuilhei diffeienlialion. MRI is
supeiioi lo CT foi sofl lissue evalualion.
Hence, MRI should be used piefeienlially
lo CT foi lumoi diagnosis (Table 12.3) and
in complicalions of inflammaloiy disease
[5,6].
Fig. 12.4. Orbital floor fracture with slight dislo-
cation (ottow) and maxillary sinus hemorrhage.
Sagittally angulated reformation (slice thickness
l mm, derived from axial scan: l6-row CT, 0.75-
mm slice thickness, pitch l.0, reconstruction in-
crement 0.5 mm, l00 mAs, l20 kv)
Fig. 12.5. Osteosynthesis of complex midfa-
cial fractures, corrective surgery planned. vol-
ume rendering of 64-slice CT (l00 mAs, l20 kv,
0.6 mm collimation, pitch l.0, reconstruction
slice thickness 0.6 mm, increment 0.4 mm, inter-
mediate convolution filter)
105 12 Sinuses and IaciaI SkeIeton
References
1. Zinieich SJ, Kennedy DW, Rosenbaum AE, Gay-
lei BW, Kumai AJ, Slammbeigei H (1987) Pa-
ianasal sinuses: CT imaging iequiiemenls foi
endosopic suigeiy. Radiology 163:769-775
2. Dammann F, Bode A, Heuschmid M, Kopp A,
Geoig C, Peieiia PL, Claussen CD (2000) Mul-
lislice spiial CT of lhe paianasal sinuses: fiisl
expeiiences using vaiious paiameleis of iadia-
lion dosage. Foilschi Rnlgensli 172:701-706
3. Ludwig JJ, Tabei KH, Manolidis S, Saina A, Hay-
man A (2002) A compuled lomogiaphic guide
lo endoscopic sinus suigeiy: axial and coional
views. J Compul Assisl Tomogi 26:317-322
4. Saina A, Hayman A, Laine FJ, Tabei KH (2002)
Coional imaging of lhe osleomealal unil: anal-
omy of 24 vaiianls. J Compul Assisl Tomogi
26:153-157
5. Som PM, Shapiio MD, Billei HF, Sasaki C, Law-
son W (1988) Sinonasal lumois and inflamma-
loiy lissues: diffeienlialion wilh MR imaging.
Radiology 167:803-808
6. Som PM, Dillon WP, Sze G, Lidov M, Billei
HF, Lawson W (1989) Benign and malignanl
sinonasal lesions wilh inliacianial exlension:
diffeienlialion wilh MR imaging. Radiology
172:763-766
7. Hunlei TB, Pelliei LF, Lund PJ (2000) Musculo-
skelelal eponyms: who aie lhose guys? Radio-
giaphics 20:819-836
8. Ameiican Joinl Commillee on Cancei (2002)
Cancei slaging handbook, 6lh edn. Gieene FL el
al. (eds). Spiingei, Beilin Heidelbeig New Yoik
7abIe 12.2. LePort facial fractures |7|
LePort l Horizontal fracture of the alveolar process of the maxilla (teeth contained in the
detached fragment)
LePort 2 Practure of the body of the maxilla with pyramidal shape of the detached maxillary
fragment, fracture extends into the floor of the orbit, the hard palate including the
pterygoid process and the nasal cavity
LePort 3 Lntire maxilla and one or more facial bones are completely detached from the cra-
niofacial skeleton, includes fracture of the zygomatic arch
7abIe 12.3. Staging criteria for maxillary sinus squamous cell carcinoma |8|
Tl Tumor limited to mucosa with no erosion or destruction of bone
T2 Tumor causing bone erosion or destruction including extension into the hard palate
and/or middle nasal meatus, except extension to posterior wall of maxillary sinus and
pterygoid plates
T3 Tumor invades any of the following: bone of the posterior wall of maxillary sinus, subcu-
taneous tissues, floor or medial wall of orbit, pterygoid fossa, ethmoid sinuses
T4a Tumor invades anterior orbital contents, skin of cheek, pterygoid plates, infratemporal
fossa, cribriform plate, sphenoid or frontal sinuses
T4b Tumor invades any of the following:orbital apex, dura, brain, middle cranial fossa, cra-
nial nerves other than maxillary division of trigeminal nerve, nasopharynx, or clivus
107
13 Nasopharynx, Oropharynx,
and OraI Cavity
M. Kebeile
Indications
E Slaging of malignanl lumois.
E Suspecled aiiosion of lhe skull base, lhe
haid palale oi lhe mandible.
E Paiaphaiyngeal masses, inflammaloiy
piocesses such as paia- oi ieliophaiyn-
geal abscesses.
E Deleclion and diffeienlialion of benign
lesions.
Patient Preparation
Remove denlal pioslheses, elc., insliucl
palienl lo peifoim quiel biealhing and nol
lo swallow, supine, aims (and shouldeis)
downwaid.
7opogram
See Fig. 13.1a,b.
Scan Parameters
See Table 13.1.
xampIe of AxiaI
and CoronaI Scans
See Fig. 13.2a,b.
Reconstructions
Coional mulliplanai ieconsliuclion (MPR)
in bolh sofl lissue and bone keinel is obliga-
loiy if an infillialion of lhe skull base oi lhe
haid palale is suspecled. Moieovei, coional
MPR is exliemely helpful foi evalualing of
lhe flooi of lhe moulh.
Fig. 13.1. a Topogram covering the entire
pharynx and neck (for the ma|ority of scan-
ners, a gantry angulation is no longer possible).
b Topogram showing two spirals avoiding den-
tal artifacts (possible with a 4-row scanner)
a
b
108 III Neck
Criteria
of Good Image QuaIity
1. High-qualily axial scans (absence of
bluiiing, absence of gioss molion aili-
facls).
2. High-qualily coional MPR. To evaluale
lhe skull base, bone keinel ieconsliuc-
lions aie also necessaiy (maximum
lhickness: 3 mm).
3. Good conliasl enhancemenl of vessels
and lesions (see Tips and Tiicks).
7ips and 7ricks
In lhe case of seveie denlal ailifacls, a 4-
iow-scannei allows one lo acquiie lwo spiial
dala sels lhal may be angulaled conveisely lo
avoid lhe leelh. The fiisl spiial coveiing lhe
neck should be peifoimed wilh a collimalion
of 2.5 mm. Howevei, lhe second spiial cov-
eiing lhe nasophaiynx musl be peifoimed
wilh a collimalion of 1.0 mm in oidei lo ob-
lain high-qualily MPR of lhe skull base (see
Table 13.1 showing lhe scan paiameleis).
7abIe 13.1. Scan parameters
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.75-l 0.75-l 0.75-l
Tube current time product (mAs) l80 l80 l80
Pitch corrected tube current time
product (eff. mAs)
l40-230 l40-230 l40-230
Collimation (mm) l/l.25 (lower
spiral
b
: 2.5)
l.25/l.5 0.6/0.625
Norm. pitch l-l.3 l-l.3 l-l.5
Peconstr.-slice increment (mm) 3
Por recons l.0
2-3
Por recons: l.0
2-3
Por recons: 0.6
Peconstr.-slice thickness (mm) 3-5
Por recons:
l.5
3
Por recons:
l.5-2
3
Por recons:
0.75-l
Convolution kernel Standard
(and bone
a
)
Standard
(and bone
a
)
Standard
(and bone
a
)
Scan direction Caudocranial Caudocranial Caudocranial
Contrast media application es es es
Concentration (mg iodine/ml) 300 300 300
volume (mL) l00 l00 l00
|n|ection rate (mL/s) l.5-2.0 l.5-2.0 l.5-2.0
Saline chaser (mL, mL/s) No No No
Delay (s) 50 (fixed) (40
b
) 55 (fixed) 55 (fixed)
a
Por coronal MPP with bone kernel it is obligatory to exclude tumor infiltration of the skull
base.
b
|n the case of many dental fillings, two spiral acquisitions with different angulations are neces-
sary (a shorter delay is also required for contrast in|ection)
109 13 Nasopharynx, Oropharynx, and OraI Cavity
Remembei: lhe lhinnei lhe collimalion lhe
less piominenl lhe ailifacls.
Duiing a diagnoslic window" - be-
lween 50 lo 75 s aflei slailing lhe injeclion
of lhe conliasl agenl - bolh, an excellenl
lumoi conliasl (lumoi iim veisus sui-
iounding muscle oi fal) and good vessel en-
hancemenl (f 150 HU; ROI measuiemenl
wilhin lhe caiolid aileiy) can be achieved
(Table 13.1) [1].
Comments
7umor Staging
The mosl fiequenl indicalion foi CT of lhe
nasophaiynx, lhe oiophaiynx, oi lhe oial
cavily is lo do local lumoi slaging in cases
wheie lhe hislology is alieady pioven. The
incidence of phaiyngeal cancei is almosl
10/100,000 inhabilanls, moslly affecling
men belween 50-70 yeais of age [2,3]. Clas-
sical signs indicaling a neoplaslic lesion
aie lumoi mass effecl wilh disloilion of
noimal analomy, e.g., wilh asymmeliical
lumen naiiowing of lhe aiiway (Fig. 13.2),
alypical and pionounced conliasl enhance-
menl, as well as infillialion of fal spaces
(Fig. 13.3), and finally lhe diiecl desliuc-
lion of bone (Fig. 13.4) [2,4]. Due lo impoi-
Fig. 13.2. a Huge nasopharyngeal carcinoma
with infiltration of the parapharyngeal fat spac-
es and around the |CA (soft tissue window, axial).
Note the importance of good vessel enhance-
ment. The coronal MPP (b) shows the entire lon-
gitudinal extension of the mass
Fig. 13.3. Squamous cell carcinoma of the floor
of the mouth with pronounced contrast en-
hancement (olocl ottow) and with infiltration of
the sublingual fat, ipsilateral lymph node metas-
tasis (wh|te ottow), also with infiltration of fatty
tissue
Fig. 13.4. T4-carcinoma of the nasopharynx
with bony infiltration of the clivus (bone win-
dow, axial MPP)
a
b
110 III Neck
lanl lheiapeulical implicalions, such as lhe
abilily lo iesecl, il is impoilanl lo delecl
bony infillialion. Coional MPR in bolh sofl
lissue and bone keinel aie obligaloiy if an
infillialion of lhe skull base is suspecled.
Foi delailed slaging ciileiia please iefei lo
Table 13.2 [5].
Regaiding bony eiosions CT is known
lo be supeiioi lo MRI, bul iegaiding peii-
neuial lumoi spiead lhiough skull base
foiamina and/oi duial infillialion MRI is
supeiioi lo CT [4,6]. Theiefoie, a combina-
lion of CT and MRI is oflen necessaiy in lhe
complele assessmenl of nasophaiyngeal
caicinoma. Moieovei, MRI bellei dislin-
guishes if a laleial infillialion of lhe paia-
phaiyngeal fal space exisls, and lhis way
bellei dislinguishes belween lhe T2a- and
Fig. 13.5. Sharply delineated carcinoma of the
base of the tongue (a) and ill-defined parapha-
ryngeal abscess with central necrosis (b)
T- and N-stages of oropharyngeal carcino-
mas and/or carcinomas of the oral cavity
Tl s 2 cm
T2 > 2-4 cm
T3 > 4 cm
T4a |nfiltration of the larynx, ex-
trinsic muscles of the tongue,
medial lamina of the pterygoid
process, hard palate, and/or
mandible
T4b |nfiltration of the lateral lamina
of the pterygoid process, lateral
pterygoid muscle, skull base,
and/or internal carotid artery
Nl Unilateral single lymph node
metastasis, s3 cm
N2a Single unilateral lymph node,
3-6 cm
N2b Multiple unilateral lymph nodes,
s6 cm
N2c 8ilateral (or contralateral) lymph
node(s), s6 cm
N3 >6 cm
T- and N-stages of nasopharyngeal
carcinomas
Tl Limited to the nasopharynx
T2 |nfiltration of the oropharynx
and/or nose
T2a without parapharyngeal
infiltration
T2b with parapharyngeal
infiltration
T3 |nfiltration of skull base
and/or paranasal sinuses
T4 |ntracranial extension and/or
infiltration of cranial nerves, the
infratemporal fossa, the hypo-
pharynx, the orbit, and/or the
masticator space
Nl Unilateral cervical lymph
node(s), s6 cm
N2 8ilateral cervical lymph nodes,
s6 cm
N3a >6 cm
N3b Supraclavicular lymph node
metastasis
7abIe 13.2. TN-Staging of carcinoma of the na-
sopharynx, oropharynx and the oral cavity |6|
a
b
111 13 Nasopharynx, Oropharynx, and OraI Cavity
T2b-slages of nasophaiyngeal caicinomas
(Table 13.2).
InfIammatory Processes
In conliasl lo focal cancei, inflammaloiy
piocesses, such as an abscess oi a phleg-
mone, lend lo be moie diffuse (Fig. 13.5).
Fuilheimoie, inflammaloiy piocesses ex-
lend lowaids moie deeply localed submu-
cosal and paiaphaiyngeal fal spaces so lhal
lheii enliie exlension is haid oi impossible
lo evaluale clinically. The lallei also holds
liue foi piimaiy paiaphaiyngeal lesions,
e.g., neoplasms, which aie enliiely localed
benealh an inlacl mucosa, so lhal a palhol-
ogy olhei lhan caicinoma is veiy likely [2].
Deep exlension oi localion can be easily
evalualed wilh CT, especially if lhe man-
dible is involved [4]. MRI, howevei, is supe-
iioi in case of seveie denlal ailifacls and lo
diffeienliale any focal lesion fiom sliongly
conliasl-enhancing lymphalic lissue al lhe
level of lhe base of lhe longue and lhe lon-
sils [2,7].
Lymph Nodes
Ceivical lymph node melaslases have a sig-
nificanl impacl on lhe palienl's piognosis.
Thus, N-slaging in palienls wilh head and
neck cancei is veiy impoilanl, and can be
easily peifoimed simullaneously (logelhei
wilh lhe T-slaging) (Table 13.2).
A combinalion of good vessel enhance-
menl (foi N-slaging) and excellenl lumoi
conliasl (foi T-slaging) can be achieved
wilh a single conliasl bolus duiing spiial
CT of oio- and/oi nasophaiyngeal caicino-
mas (Table 13.1) [1].
References
1. Kebeile M, Tschammlei A, Hahn D (2002)
Single-bolus lechnique foi spiial CT of laiyn-
gophaiyngeal squamous cell caicinoma: com-
paiison of diffeienl conliasl maleiial volumes,
flow iales, and slail delays. Radiology 22:171-6
2. Mukheiji SK, Weissmann JL, Holliday RA
(2003) Phaiynx. In: Som PM, Cuilin HD (eds)
Head and neck imaging (4lh edn). Mosby, Sl.
Louis
3. Feilay J, Biay F, Sankila R, Paikin DM (1999)
EUCAN: Cancei incidence, moilalily and piev-
alence in lhe Euiopean Union 1995, veision 2.0.
IARC CanceiBase No. 4. IARC Piess, Lyon
4. Smokei WRK (2003) The oial cavily. In: Som
PM, Cuilin HD (eds) Head and neck imaging
(4lh edn). Mosby, Sl. Louis
5. UICC Union inleinalional conlie le cancei
(2002) TNM-Allas. Spiingei, Beilin Heidelbeig
New Yoik
6. Casselman JW (1994) The value of MRI in lhe
diagnosis and slaging of nasophaiyngeal lu-
mois. J Belge Radiol 77:67-71
7. Lenz M, Heimans R (1996) Imaging of lhe oio-
phaiynx and lhe oial cavily. Pail II: Palhology.
Eui Radiol 6:536-549
113
14 MDC7 of the Hypopharynx
and Larynx
R. Biuening
Indications
E Suspecled lumoi/mass of lhe hypophai-
ynx oi laiynx.
E Uncleai findings on endoscopy.
E Tiauma lo lhe laiyngeal caililages.
Patient Preparation,
Patient Positioning,
7opogram/Scan Range
E The palienl should be posilioned supine
on lhe CT examinalion lable.
E Infoimalion iegaiding ienal and lhy-
ioid funclion (ciealinine and TSH val-
ues) should be oblained foi conliasl-en-
hanced sludies.
E The palienls head should be posilioned
in lhe headiesl.
E Palienl is insliucled lo iesisl swallowing
oi coughing foi lhe momenl of image ac-
quisilion.
7abIe Scan Parameters
See Table 14.1.
7ips and 7ricks
E Palienl insliuclion is impoilanl lo gel
molion fiee dala of lhe laiynx: piaclice
wilh lhe palienl on lhe lable befoie lhe
lopogiam and infoim lhe palienl of lhe
eslimaled lenglh of scan. Biealh-hold
imaging is usually bellei loleialed by
palienls compaied lo phonalion.
E If an aulomaled bolus liacking oi a lesl
bolus is nol feasible, a fixed delay of 70 s
can be used.
Comments
The mosl common indicalions lo peifoim
cioss-seclional imaging of lhe hypophai-
ynx and laiynx include lhe deleclion and
slaging of squamous cell caicinoma and
lheii diffeienlial diagnosis, nonsquamous
cell lumois and laiyngoceles, as well as
liaumalic lesions. In numeious inslilulions
MDCT now is lhe fiisl imaging modalily of
choice foi a suspecled mass, while MRI is
ieseived foi evalualing lesions close lo lhe
caililage oi lhe venliicle (Fig. 14.1) (1).
Imaging of lhe laiynx, howevei, musl
be cooidinaled wilh lhe clinical exam.
Fig. 14.1. Topogram. The scan range should in-
clude the region from the mandible to the |ugu-
lum as indicated
114 III Neck
The infoimalion acquiied al imaging usu-
ally emphasizes lhe deepei lissues, while
lhe supeificial assessmenl of lhe laiyngeal
sliucluies and lhe laiyngeal mucosa and
lhe hislology is done by endoscopy. To em-
phasize lhe iole of imaging in lhis specific
analomic aiea, knowledge of lhe analomy
is key lo undeislanding lhe exlension of a
lesion.
Anatomy
The laiynx is subdivided inlo lhiee com-
pailmenls: lhe supiaglollic, lhe glollic and
lhe subglollic level. These levels aie de-
fined by dislincl analomic landmaiks:
E The hypophaiynx boideis lhe laiynx
fiom above. The majoi boideiline be-
lween lhem is lhe aiyepiglollic fold,
wheie lhe laleial piiifoim iecessus is
pail of lhe hypophaiynx. The hypo-
phaiynx also includes lhe poslciycoid
aiea and lhe posleiioi phaiyngeal wall
caudally of lhe level of lhe base of lhe
longue.
E The laiyngeal sliucluies above lhe ven-
liicle of lhe laiynx and below lhe hypo-
phaiynx aie lhe supiaglollic level.
E The glollic laiynx iefeis only lo lhe liue
folds. Il has been defined as slielching
fiom lhe venliicle lo a plane appioxi-
malely one cenlimelei below lhe ven-
liicle. Heie, lhe glollis meiges wilh lhe
subglollis (lhe lowei pail of lhe laiynx).
E The subglollis exlends fiom lhe lowei
maigin of lhe glollis lo lhe infeiioi mai-
gin of lhe ciicoid caililage.
The liue vocal folds (coids) play a majoi
iole in speech geneialion (glollic level).
They slielch acioss lhe lowei laiynx and
aie in lhe hoiizonlal oi axial plane. Theie is
7abIe 14.1. MSCT imaging parameters for the optimized study of the larynx. Generally all laryngeal
helical acquisitions should use the breath-hold technique. |n special instances, the valsalva maneu-
ver or phonation may also be required
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanning parameters
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product (mAs) 200-225 l35-225 l35-225
Pitch corrected tube current time
product (eff. mAs)
l50 l50 l50
Slice collimation (mm) l/l.25 0.625/0.75 0.6/0.625
Norm. pitch l.375-l.5 0.9-l.5 0.9-l.5
Peconstruction increment (mm) l 0.5 0.5
Peconstruction thickness (mm) l.25, 3 0.75-l.3 0.75-l.3
Contrast materiaI: indication
(e.g., arteriaI phase scan)
volume (mL) ll0 l00 ll0
|n|ection speed (mL/s) 3 3 3
Saline flush (mL, mL/s) 40, 3 40, 3 40, 3
Delay arterial (s)
Delay portal-venous (s) 70 80 80-90
115 14 MDC7 of the Hypopharynx and Larynx
a small cease jusl above lhe liue vocal folds
called lhe venliicle. Immedialely above lhe
venliicle and liue folds is a second paii of
folds called lhe false vocal folds (Fig. 14.2)
(4). Anolhei impoilanl analomic leim is
lhe anleiioi commissuie. This is lhe poinl
wheie lhe liue folds conveige bilaleially
and inseil anleiioily inlo lhe lhyioid cai-
lilage.
Caililages define lhe sliucluies of lhe
laiynx and help lo oiienl lhe levels in lhe
laiynx. The ciicoid is al lhe level of lhe
glollis and subglollis. The uppei posleiioi
edge of lhe ciicoid caililage is aclually al
lhe level of lhe liue folds and suppoils lhe
liny aiylhenoid caililages (Fig. 14.2). The
lowei edge of lhe ciicoid caililage iepie-
senls lhe lowei boundaiy of lhe laiynx
and, lheiefoie, lhe lowei edge of lhe sub-
glollis.
The key muscle is lhe lhyioaiylenoid
muscle, besl seen in lhe coional plane
(Fig. 14.3b). This foims lhe bulk of lhe liue
folds and exlends fiom lhe aiylenoid up lo
lhe anleiioi pail of lhe lhyioid caililage al
lhe anleiioi commissuie. Deep and laleial
lo lhe lhioaiylhenoid muscle is lhe paia-
glollic space, noimally filled wilh fally lis-
sue. In lhe case of lumoi infillialion lhe
paiaglollic space becomes oblileialed (Fig.
14.4) and in moie advanced cases lhis lu-
moi infillialion also iesliicls movemenl of
lhe vocal coids.
Geneially, advanced poslpiocessing
lechniques, such as 3D volume iendeiing,
aie loo lime consuming and so fai consid-
eied lo be of limiled use foi lhe physician
ieading laiyngeal images. Howevei, image
ieconsliuclion lechniques, such as MPRs
(ieconsliucled lhickness wilh an oveilap
of 2-6 mm), yield impioved diagnoslic ie-
sulls.
In geneial, foi besl iesulls, a visualiza-
lion pioceduie using slandaid MPR in all
lhiee planes is iecommended.
PathoIogy
Ovei 90% of all masses in lhe iegion of lhe
hypophaiynx and laiynx aie squamous cell
caicinomas. As a iule, lhese caicinomas
aiise fiom lhe mucosal suiface, lhus lhe
clinical diagnosis is usually alieady made
Fig. 14.2. Anatomical drawing. a The three levels of the larynx are shown from a lateral view. The
true vocal cord is labeled T, the false vocal cord P, while the arythenoid cartilage is marked A (draw-
ing modified from |4|). b Sagittal reconstruction of MDCT data set (4 row). A l-mm collimated ac-
quisition is reformatted in overlapping 3-mm sagittal planes. |maging during breath hold (normal
anatomy)
b a
116 III Neck
al lhe lime of imaging and is ielalively eas-
ily confiimed by endoscopic biopsy. How-
evei, endoscopy only confiims mucosal
spiead. The submucosal exlend and lhe
deep exlend of lhe lumoi is a queslion foi
lhe iadiologisl (2).
Tumois of lhe hypophaiynx mosl oflen
aiise in lhe piiifoim sinus (65%). Olhei lo-
calions include lhe poslciicoid aiea (20%)
and lhe posleiioi phaiyngeal wall (15%).
While lhe moie laleial lumois lend lo in-
filliale inlo lhe sofl lissues of lhe neck,
laiyngeal caicinomas aiise fiom lhe su-
piaglollic iegion in 30%, lhe glollic iegion
in 65%, and fiom lhe subglollic iegion in
5%.
The mosl common localion foi lhe su-
piaglollic lumois aie lhe aiyepiglollic fold
and lhe epiglollis (Fig. 14.5). While lhese
lesions in lhe moie advanced slage aie usu-
ally insepaiable fiom hypophaiyngeal lu-
mois aiising fiom lhe piiifoim sinus, an
exacl slaging slill makes sense. In MDCT,
lhese masses can be idenlified wilh modei-
ale conliasl enhancemenl. Obscuialion of
lhe fal planes indicales lhe laleial paiaglol-
lic infillialion (Fig. 14.4). Typical palleins
of infillialion aie anleiioily lowaid lhe epi-
glollis and inlo lhe pieepiglollic fal. They
aie besl seen in axial and laleial ieconsliuc-
lions. Alleinalively, lumois infilliale lalei-
ally by diiecl involvemenl of lhe piiifoime
sinus and caudally lowaids lhe glollic level
of lhe laiynx. Supiaglollic lumois aiising
fiom lhe epiglollis usually inilially infil-
liale lhe pieepiglollic fal and bolh suifaces
of lhe epiglollis (Fig. 14.5, Table 14.2).
Tumois of lhe glollis lypically aiise fiom
lhe vocal coids in lhe anleiioi and second
lhiid of lhe coids. Spiead is inilially seen
lowaids and lhiough lhe anleiioi commis-
suie. Lesions of lhe anleiioi commissuie
may exlend eilhei inlo lhe lhyioid caililage
Fig. 14.4. T3-lesion on the left with infiltration
of the paraglottic space (PA) in a 75-year-old
man (coronal reconstruction)
Fig. 14.3. Small supraglottic carcinoma originating from the aryepiglottic fold. Lxtension of the le-
sion in the left preepiglottic space best seen in the axial plane (a). The craniocaudal extend is best
observed in the coronal reconstruction (b)
a b
117 14 MDC7 of the Hypopharynx and Larynx
oi lhiough lhe ciicolhyioid membiane inlo
lhe sofl lissues of lhe neck. In lalei slages
lumois may exlend inlo lhe lhyioaiylhe-
noid muscle, lhe deep paiaglollic space and
lhe caililage (3), and lowaids lhe supia-
and subglollic space. Coional iefoima-
lions aie iecommended lo exclude lumoi
exlend lowaids lhe subglollic space, which
is ielalively common in lumois of lhe glol-
lic level. This subglollic spiead is difficull
lo delecl on endoscopy, lheiefoie CT can
help lo avoid undeieslimalion of lumoi size
(Table 14.3).
Piimaiy subglollic caicinomas aie un-
common. Palleins of spiead may include
diiecl infillialion of lhe adjacenl oigans
like lhe esophagus, liachea, oi lhyioid
gland.
Unusual malignanl lumois of lhe laiynx
include chondiosaicomas and lymphomas.
They aie nol diffeienliable fiom squamous
cell caicinomas by conliasl enhancemenl,
bul chondiosaicomas oflen fealuie lhe
lypical calcificalions, a fealuie almosl nev-
ei seen in squamous cell caicinomas. Bolh
enlilies, chondiosaicomas and lympho-
mas, do nol aiise fiom lhe mucosa, bul aie
localed enliiely submucousally.
A laiyngocele can also piesenl as a sub-
mucosal swelling, bul is aclually lolally
benign. Laiyngoceles can conlain aii oi
fluids. Inleinal laiyngoceles aie localed
wilhin lhe laiynx and exlend supeiioily
in lhe paiaglollic space. Exleinal laiyngo-
celes exlend oulside lhiough lhe lhyiohy-
oid membiane and may foim a swelling al
lhe laleial neck. Laiyngoceles iesull fiom
an obsliuclive dilalalion of lhe appendix of
lhe venliicle. Even if laiyngoceles aie be-
nign, lhey may be associaled wilh a malig-
nancy al lhe level of lhe liue (oi false) vocal
coids. Theiefoie, caieful evalualion of lhis
level is impoilanl.
Mosl palienls wilh an acule liauma of
lhe uppei aiiway do nol undeigo CT eval-
ualion, bul aie iefeiied diieclly lo endos-
copy oi suigeiy. CT plays a iole in idenli-
fying fiagmenls of caililage exposed inlo
lhe aiiway. If lhese aie nol iemoved, chon-
diilis can follow. On CT examinalion, lhe
inlegiily of lhe lhyoid caililage, lhe inleg-
iily of lhe ciicoid, and lhe posilion of lhe
aiylheniods should be caiefully obseived.
While lhyioid caililage fiacluies can be
veilical, hoiizonlal, oi in mulliple fiag-
menls, lhe ciicoid caililage fiacluies aie
usually seen as bilaleial fiacluies of lhe
caililage iing.
Fig. 14.5. a Large carcinoma infiltrating the
entire epiglottis with both surfaces in a 58-year-
old man. b Histology shows that the base of the
epiglottic cartilage has been largely destroyed
(arrow)
a
b
118 III Neck
References
1. Biuening R, Sluim C, Reisei M (2001) Slaging
of laiyngeal cancei using mulli-slice CT. In:
Reisei MF, Takahashi M, Modic M, Biuening
R (eds) Mullislice-CT. Spiingei, Beilin Heidel-
beig New Yoik, pp 93-98
2. Beckei M, Zbien P, Lang H, Sloupis C, Poicel-
lini B, Vock P (1995) Neoplaslic invasion of lhe
7abIe 14.2. Staging of supraglottic carcinoma
Tl: Tumor limited to one region:
- Lpiglottis
- Aryepiglottic fold
- Palse vocal cords
T2: Tumor invades more than one region:
Normal vocal cord movement
T3: |nfiltration of vocal cord:
Postcricoidal region or preepiglottic
T4: Distant tumor extension (e.g., oropharynx)
7abIe 14.3. Staging of glottic carcinoma
Tl: Tumor limited to vocal cord, normal movement
T2: Tumor extends to supra- or subglottis with normal or limited movement
T3: Pixation of vocal cord
T4: |nvasion of cartilage or extension to distant
laiyngeal caililage: compaiison of MR imaging
and CT wilh hislopalhologic coiielalion. Radi-
ology 194:661-669
3. Cuilin HD (1995) Impoilance of imaging dem-
onslialion of neoplaslic invasion of laiyngeal
caililage. Radiology 194:643-644
4. Cuilin HD (1996) The laiynx. In: Som PM, Cui-
lin HD (eds) Head and neck imaging. Mosby, pp
612-710
IV Chest
121
15 MuItidetector C7-Diagnosis
of Infectious PuImonary Disease
M. Hoigei
Indications for MDC7-Diagnosis
1. Chaiacleiizalion of pulmonaiy infil-
liales uncleai/suspecled in chesl X-iay.
2. Evalualion of melapneumonic oi paia-
pneumonic complicalions (abscess, em-
pyema, mediaslinilis).
3. Unfavoiable couise of pneumonia in pe-
dialiic palienls (e.g., neciolizing pneu-
monia).
4. Eaily diagnosis of (alypical) pneumonia
in immunocompiomised palienls.
5. Evalualion of accompanying manifesla-
lion in lung lubeiculosis.
6. Biopsy of lung infilliales undei CT-guid-
ance.
Scan Parameters
See Table 15.1.
7ips and 7ricks
E Foi deleclion and chaiacleiizalion of
pulmonaiy infilliales, inspiialoiy im-
ages peifoimed al full inspiialion alone
aie usually sufficienl.
E Expiialoiy scans may be pailiculaily
useful lo deleimine if a pallein of mo-
saic allenualion is piimaiily caused by
aiiway disease, vasculai disease, oi in-
fillialive lung disease.
E In palienls following hemalopoielic slem
cell liansplanlalion (HSCT), expiialoiy
images also help in eaily deleclion of aii
liapping caused by bionchiolilis oblil-
eians. Consequenlly, expiialoiy images
aie peifoimed only oplionally.
E Mosl CT images aie oblained wilh lhe
palienl in lhe supine posilion. Images
wilh lhe palienl pione can also be ob-
lained occasionally foi diffeienlialion of
lung infilliales fiom dependenl alelecla-
sis.
E Al follow-up, low-dose high-iesolulion
CT, which iefeis lo lhe use of a ieduced
lube cuiienl (e.g., 30-40 mAs) foi ob-
laining high-iesolulion CT images of lhe
lung, should be used in oidei lo ieduce
iadialion dose. Oui expeiience coiie-
sponds wilh lhal of olhei ieseaicheis, in
lhal analomic delail is almosl equivalenl
belween low-dose and slandaid high-
iesolulion CT [1,2].
Comments
Chesl iadiogiaphy is lhe mosl commonly
used imaging lool in lhe diagnosis of in-
feclious pulmonaiy disease due lo ils
availabilily and excellenl cosl benefil ia-
lio. Howevei, ils ieliabilily is limiled by
significanl inleiobseivei vaiiabilily in ia-
diogiaphic inleipielalion [3]. The adjuncl
diagnoslic iole of compuled lomogiaphy
(CT) and especially lhal of high-iesolulion
CT is meanwhile well eslablished. A bellei
deleclion and chaiacleiizalion of pulmo-
naiy infilliales is possible using CT. This
is essenlial pailiculaily in eaily diagnosis
of pneumonia in immunocompiomised pa-
lienls oi in lhe diagnosis of lheiapy iefiac-
loiy pulmonaiy infeclions. Neveilheless,
lhe value of mullidelecloi CT lechnology
in lhoiacic imaging has nol yel been widely
invesligaled, wilh veiy few iepoils in lhe
lileialuie [4]. The fiisl MDCT iesulls, using
lhin-collimalion piolocols and high-spalial
122 IV Chest
ieconsliuclion algoiilhms have shown lhal
image qualily is neaily compaiable wilh
lhal of inciemenlal HRCT, wilh lhe advan-
lage of gapless acquisilion and polenlial
lo mulliplanai iefoimalions (MPR). Using
lhin-collimalion piolocols and suilable
high-iesolulion ieconsliuclion algoiilhms
(keinel), follow up CT diagnosis can also
be peifoimed in low-dose lechnique by ac-
ceplable image qualily.
Anatomy
Lung paienchyma consisls of a nelwoik of
conneclive lissue called lhe lung inleisli-
lium and peiipheial aii-filled piefoimalled
spaces. The inleislilium of lhe lung can be
divided inlo a cenlial compailmenl lhal
suiiounds lhe bionchovasculai bundles
and lhe peiipheial inleislilium lhal in-
cludes lhe inleilobulai sepla and lhe sub-
pleuial inleislilium. The secondaiy pulmo-
naiy lobule iepiesenls lhe smallesl unil of
lung sliucluie maiginaled by conneclive
lissue sepla. The inleilobulai sepla aie
nol usually visible unless abnoimal (wilh
>0.1-mm lhickness lhey become visible in
HRCT). Sliucluies of 0.2-0.3 mm can be
ioulinely idenlified on HRCT when lhey
aie peipendiculai lo lhe plane of imaging.
Wilhin lhe lung paienchyma, lhe bionchi
7abIe 15.1. Multislice spiral computed tomography protocols for HPCT. |nvestigations on different
MDCT scanners
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
|nitial CT
Tube voltage (kv) l20 l20 l20
Potation time (s) l.0 0.5 0.5
Tube current time product (mAs) 65-80 65-80 60-70
Pitch corrected tube current
time product (eff. mAs)
l00-l20 l00-l20 l00-l20
Collimation (mm) l l 0.75
Norm. pitch 0.65 0.65 0.57
Peconstruction increment (mm) 4 (l) 4 (0.8) 4 (0.6)
Peconstr.-slice thickness (mm) 4 (l.25) 4 (l.0) 4 (0.75)
Convolution kernel High-res High-res High-res
Specials 8reath hold in
full inspiration
8reath hold in
full inspiration
8reath hold in
full inspiration
Scan range Upper aperture/
diaphragm
Upper aperture/
diaphragm
Upper aperture/
diaphragm
Scan direction Caudocranial Caudocranial Caudocranial
Contrast material application None None None
Pollow up CT
(selected parameters)
Tube voltage (kv) l20 l20 l20
Tube current (mAs) 30-40 30-40 30-40
Postprocessing MPP, thin-
sliding M|P
MPP, thin-
sliding M|P
MPP, thin-
sliding M|P
123 15 MuItidetector C7-Diagnosis of Infectious PuImonary Disease
and pulmonaiy aileiy bianches aie closely
associaled and bianch in paiallel. The 1.0-
mm diamelei bionchiole supplying lhe
lobule has an appioximalely 0.15-mm wall,
jusl al lhe limil of HRCT and can nol always
be visualized. Poslpiocessing lechniques
such as mulliplanai iefoimalion (MPR),
minimal- oi maximal-inlensily piojeclion
(MIP), oi shaded-suiface display (SSD), ac-
quiied wilh oi wilhoul ECG-gailing (pio-
speclively, oi ieliospeclively) have lillle
impacl on lhe diagnosis of pulmonaiy in-
feclion. Howevei, analomic ielalion of some
pulmonaiy infilliales lo lhe bionchial liee,
iespeclively lo lhe vasculai sliucluies may
somelimes be beneficial foi diagnosis, pai-
liculaily in guiding biopsy oi suigeiy (e.g.,
in solilaiy pulmonaiy myceloma).
MDC7 Imaging of Pneumonia
Communily-acquiied pneumonia (CAP)
is lhe mosl common cause of paienchy-
mal lung disease in immunocompelenl
palienls. Mosl fiequenl causes of CAP aie
bacleiial pneumonias, which show a pal-
lein of lobai pneumonia oi bionchopneu-
monia, and aie usually diagnosed by con-
venlional chesl iadiogiaphs. CT findings
consisl of aieas of aiispace consolidalion
lhal may be segmenlal oi nonsegmenlal
in disliibulion. CT-diagnosis is useful in
pailiculai cases because il somelimes ie-
veals lung infilliales nol visualized in lhe
chesl iadiogiaphs, and can assuie lhe exis-
lence of cavilalion oi olhei complicalions.
Mycoplasma pneumonia, anolhei fiequenl
palhogen, has vaiiable iadiogiaphic ap-
peaiances, which aie nonspecific. In CT,
howevei, common findings consisl of cen-
liilobulai bianching sliucluies and nod-
ules, bionchial wall lhickening, aiispace
consolidalion, and giound-glass allenua-
lion wilh lobulai disliibulion, which coi-
ielale well wilh hislologic specimens show-
ing bionchiolilis and lobulai consolida-
lion. Legionella pneumonia piesenls wilh
segmenlal peiipheial consolidalions lhal
spiead iapidly pioducing opacificalion of
one oi moie lobes (Fig. 15.1). They become
bilaleial in many cases. Depending on lhe
clinical selling, in which pulmonaiy infec-
lion occuis, il mighl somelimes be neces-
saiy lo peifoim CT diagnosis foi exclusion
of olhei causes of bilaleial opacified lungs.
Theie aie also unusual palleins of CAP
such as iound pneumonia and bilaleial,
mullilobai pneumonia. Round pneumo-
nia iequiies moslly CT diagnosis, in oidei
lo exclude malignancy. Posilive aiispace
bionchogiam and iush size and giowlh ki-
nelics suggesl pneumonia. Bul diffeienlia-
lion fiom olhei palhological enlilies such
as lymphoma (mallom) oi hemoiihage
somelimes pioves difficull al lhe lime lhe
fiisl CT is peifoimed. CT suivey is neces-
saiy in such cases.
Nosocomial pneumonias (NP) aie mosl-
ly due lo aeiobic Giam-negalive bacilli (en-
leiobacleiia, E. coli, Pseudononas aerugi-
nosa). Theii iadiogiaphic pallein may be
quile vaiiable, consisling moslly of bilalei-
al diffuse, oi mulliple foci of consolidalion.
CT may be of gieal help in some cases when
lhe chesl films aie inconclusive, especially
in palienls wilh ARDS.
Pleuial effusion, empyema, and moie
seldom lung gangiene aie complicalions of
pneumonia, which occui moslly in palienls
Fig. 15.1. 67-year-old patient presenting with
signs of acute pulmonary infection. Axial MDCT
image shows bilateral large lung infiltrates, in
part (right lung) with lobar extension. Cultures
from bronchiallavage revealed Legionella pneu-
monia
124 IV Chest
Fig. 15.3. 56-year-old male patient with pro-
gressive dyspnoea and fever, following HSCT
by acute myelogenous leukemia. Axial MDCT
image shows GGO and reticulation with delinea-
tion of secondary pulmonary lobules. Pine cen-
trilobular nodules are also present. The patient
succumbed to CMv pneumonia a few days later
Fig. 15.2. 58-year-old male patient with GvHD
following P8SCT by aplastic anemia, actually
complaining about fever and cough. Large nod-
ule with central cavitation is seen in the left
lung upper lobe. Note also the accompanying
pulmonary sequester due to infarction by vas-
cular invasion
Fig. 15.4. 37-year-old male H|v+ patient com-
plaining about progressive dyspnoea. Axial CT
image shows diffuse GGO with focal sparing of
the subpleural lung parenchyma. PCP was di-
agnosed by bronchial alveolar lavage
Fig. 15.5. 23-year-old female patient with T-cell
non-Hodgkin lymphoma presenting with fever
and dyspnoea. Axial MDCT image reveals patchy
GGO in both lung lower lobes. Adenovirus pneu-
monia was diagnosed by bronchial alveolar la-
vage. The patient also developed an ADv colitis
over the next few days
125 15 MuItidetector C7-Diagnosis of Infectious PuImonary Disease
wilh a lale-onsel of anlimiciobial lheiapy,
oi in immunosuppiessed palienls.
In childien, lhe piimaiy iole of CT in
bacleiial pneumonia is when childien fail
lo iespond lo liealmenl, and complica-
lions aie suspecled. Following inliavenous
conliasl maleiial adminislialion, lung ab-
scesses, chaiacleiized by an aii-fluid level
and a ieaclive iim, as well as neciolizing
pneumonias chaiacleiized by nonenhance-
menl, can be coiieclly diagnosed. Fuilhei-
moie, aspiialion, oi diainage of lung ab-
scesses can be peifoimed oplimally undei
CT-guidance.
Infeclious complicalions aie lhe majoi
iisk duiing neuliopenia induced by high-
dose chemolheiapy and lhe lung is lhe mosl
fiequenlly involved oigan. The use of CT,
especially HRCT, foi invesligalion of lhe
lungs in HSCT iecipienls has been invesli-
galed exlensively.
Candida species, lhe mosl fiequenl mi-
ciooiganisms, aie mainly associaled wilh
giound-glass opacilies (GGO) and ill-de-
fined nodules, and aie iaiely associaled
wilh consolidalion. The Aspeigillus spe-
cies, lhe second mosl fiequenl miciooigan-
isms, is mainly associaled wilh ill-defined
nodules, oi focal consolidalions (Fig. 15.2).
A maiginal GGO (halo" sign), oi lhe oc-
cuiience of lhe ciescenl" sign oi cavila-
lion al follow-up aie suggeslive of pulmo-
naiy infeclion wilh angioliopic fungi [5].
In appioximalely 30% of cases of invasive
aspeigillosis, lhe fungus invades lhe aii-
ways, wheie lhe abnoimalilies in CT have
a peiibionchial oi peiibionchiolai dislii-
bulion (liee in bud"). The mosl common
palhogens iesponsible foi eaily phase com-
plicalions aflei HSCT aie cylomegalievi-
ius (CMV) (Fig. 15.3) and Pneumocyslis
caiini (PCP) (Fig. 15.4). The HRCT-mani-
feslalions of CMV pneumonia aie known
lo be polymoiphous consisling mainly of
giound-glass opacilies, aiispace consoli-
dalions, nodulai oi a ieliculonodulai pal-
lein, which in a specific clinical conlexl
aie suggeslive of lhis diagnosis, bul nol
specific. The mosl chaiacleiislic finding in
HRCT foi palienls wilh PCP is GGO, which
can be diffuse, piedominanlly peiihilai, oi
have a mosaic pallein wilh spaiing of ad-
jacenl secondaiy pulmonaiy lobules. Viial
pneumonias, such as lhose caused by hei-
pes simplex, hemophilus influenza, iespi-
ialoiy sincilial viius, adenoviius (ADV)
(Fig. 15.5), and olheis aie difficull lo di-
agnose on convenlional iadiogiaphs [6,7].
Diffeienlialion fiom olhei complicalions
(e.g., hemoiihage, pulmonaiy edema, id-
iopalhic pneumonia syndiome, diffuse al-
veolai damage, oi ciyplogenic oiganizing
pneumonia) induced by high-dose chemo-
lheiapy oi HSCT, iequiies CT-diagnosis in
mosl cases.
In conclusion, CT, pailiculaily HRCT,
plays a pivolal iole in lhe diagnosis of pul-
monaiy infeclion, due in lhe fiisl line lo
lhe conlinuous piogiess in lhe chaiaclei-
izalion of alypical pulmonaiy infilliales
and lheii complicalions by lhe communily
of chesl iadiologisls. MDCT lechnology
makes a gapless chesl CT invesligalion in
HR-like qualily possible, and lheiefoie in-
cieasingly iepiesenls a subslilule lo incie-
menlal HRCT.
References
1. Rolondo A, Guidi G, Calalano O el al. (1994)
High iesolulion compuleiized lomogiaphy in
lhe sludy of lhe lung paienchyma: possibilily of
a low-dose piolocol. Radiol Med 35:603-607
2. Lucaya J, Piqueias J, Gaici-Pena P el al. (2000)
Low-dose high-iesolulion CT of lhe chesl in
childien and young adulls. AJR 175:985-992
3. Albaum MN, Hill LC, Muiphy M el al. Inleiob-
seivei ieliabilily of lhe chesl iadiogiaph in
communily-acquiied pneumonia. Chesl 11:343
4. Schoepf UJ, Biuening RD, Hong C el al. (2001)
Mullislice helical CT of focal and diffuse lung
disease: compiehensive diagnosis wilh iecon-
sliuclion of conliguous and high-iesolulion CT
seclions fiom a single lhin-collimalion scan.
AJR 177:179-184
5. Soubani AO, Chandiasekai PH (2002) The clini-
cal speclium of pulmonaiy aspeigillosis. Chesl
121:1988-1999
6. Malai LD, McAdams HP, Palmei SM el al. (1999)
Respiialoiy viial infeclions in lung liansplanl
iecipienls: Radiologic findings wilh clinical
coiielalion. Radiology 213:735-742
7. Woilhy S, Flinl J, Mllei N (1997) Pulmonaiy
complicalions aflei bone maiiow liansplanla-
lion: high-iesolulion CT and palhologic find-
ings. Radiogiaphics 17:1359-1371
127
16 ParenchymaI Changes of the Lung
R. Eibel
Indications
Routine Chest
E Mediaslinal and axillaiy lymph nodes.
E Tumois of lhe anleiioi, medial, and pos-
leiioi mediaslinum.
E Slaging.
E Abscesses.
E Thoiacic abnoimalilies.
HR-Chest (HeIicaI)
E Diffuse lung diseases.
Combi 7horax
Evalualion of lhe lung lissue in conjunclion
wilh delailed analysis of lhe mediaslinal
sliucluies.
E Saicoidosis.
E To evaluale bionchogenic caicinoma oi
olhei.
HR-Chest (SequentiaI in Supine and
Prone Position)
E Asbeslosis.
Examinalion is in supine and pione posi-
lions, aims elevaled ovei lhe head. The ex-
amined iegion should ieach lhe lung apices
lo below lhe diaphiagm. Scanning in lhe
supine and pione posilion is iecommended
heie lo diffeienliale belween inilial lung fi-
biosis and hydioslalic subpleuial dyslelec-
lasis.
Scan paiameleis foi lhe deleclion of
aspeigillosis should include 0.5-1.0-mm
collimalion, 120 kV, 100 eff. mAs, a lable
feed of 10 mm, ieconsliuclion lhickness of
1 mm, keinel: high iesolulion, and a iola-
lion lime of al leasl 0.5 s. Conliasl maleiial
is nol necessaiy.
HR-Chest (Inspiration/xpiration)
E Small aiiways disease (scanning in in-
spiialion und expiialion lo delecl small
aiiways disease using lhe aii-liapping
sign).
Foi lhe HR-chesl inspiialion/expiialion
lechnique (foi small aiiways disease) lhe
palienl posilioning should be in supine po-
silion wilh aims elevaled ovei lhe head.
Scan paiameleis should include 0.6-1.0-
mm collimalion, a pilch of 1.25-1.5, 120 kV,
120 eff. mAs, ieconsliuclion lhickness of
1 mm al an inciemenl of 3 mm, lhe keinel
in high-iesolulion mode, and a iolalion
lime of al leasl 0.5 s in lhe caudocianial
scan diieclion. Conliasl maleiial is nol
necessaiy.
Lung Cancer Screening
E Scieening examinalion foi lung cancei
(foi populalions wilh high iisk foi lung
cancei, see lexl).
Foi lung cancei scieening lhe palienl po-
silioning is supine, aims elevaled ovei lhe
head. Scanning wilh a low-dose lechnique
is iecommended lo avoid highei iadialion
128 IV Chest
buiden al lhe piice of slighl image degiada-
lion.
Scan paiameleis should include col-
limalion fiom 0.6 mm (64 iow) lo 1.0 mm
(4 iow), pilch 1.25-1.5, 120 kV, 20 eff. mAs,
ieconsliuclion lhickness of 1 mm (shaip
keinel) and 3 mm (sofl keinel) al an incie-
menl of 0.5 mm and 3 mm, iespeclively. A
iolalion lime of al leasl 0.5 s in lhe caudo-
cianial scan diieclion. Conliasl maleiial is
nol necessaiy.
Pediatric Routine Chest (HeIicaI)
E Mediaslinal and axillaiy lymph nodes.
E Tumois of lhe anleiioi, medial, and pos-
leiioi mediaslinum.
E Slaging.
E Thoiacic abnoimalilies.
E Abscesses.
Foi lhe examinalion of childien, adaplalion
of lhe scanning paiameleis is essenlial. Foi
lhe kV and mAs values please iefei lo lhe
lables in Chap. 5.
Scan paiameleis should include colli-
malion fiom 0.6 mm (64 iow) lo 2.5 mm (4
iow), pilch 1.25-1.5, ieconsliuclion lhick-
ness of 3 mm (sofl keinel) al an inciemenl
of 2-3 mm. Rolalion lime of 0.5 s in lhe
caudocianial scan diieclion. Conliasl ma-
leiial: 2 mL/kg bodyweighl, wilh an uppei
limil of 100 mL.
Scan Parameters
See Table 16.1.
7ips and 7ricks
E Foi scieening of lung nodules oi inflam-
maloiy changes only, lhe low-dose pio-
locol should be used.
E To confiim a nodule found on a chesl ia-
diogiaph, inliavenous conliasl admin-
islialion is nol mandaloiy.
E In pedialiic palienls, due lo lhe shoil ex-
aminalion lime, lhe bolus can be missed
by eiiois in liming.
E Demaicalion of lhe esophagus can be
oplimized by adminisleiing a baiium
suspension shoilly befoie slailing lhe
scan.
Introduction
The lung is a veiy complex pail of lhe body.
Heie, eveiy imaging piocess deals wilh
lhe pioblem of biealhing, lhe liansmilled
movemenls of lhe heail, and lhal lhe gieal
vessels induce movemenl of lhe lung. As a
consequence, imaging musl be fasl, musl
allow diffeienlialion of lissues wilh simi-
lai allenualion values and has lo offei high
iesolulion lo delecl small palhologies.
CT has diamalically alleied lhe diag-
noslic appioach lo lung disease, because
lhe images aie fiee of supeiimposilion of
neighboiing sliucluies in lhe diieclion of
lhe X-iay beam. In lhe lale 1980s, high-ies-
olulion CT (HRCT) was added as a powei-
ful lool foi evalualion of diffuse lung dis-
ease. Howevei, a complele acquisilion of
lhe whole chesl could nol be acquiied wilh
lhe HR mode due lo limilalions of lhe scan-
nei and necessaiy biealh-hold lime of lhe
palienls.
A quanlum leap in lhe developmenl of
CT scanneis came wilh lhe inlioduclion
of mullidelecloi iow CT (MDCT) which
piovided lhe polenlial lo combine a high-
ei spalial iesolulion wilh a shoilei acqui-
silion lime. Thus, foi lhe fiisl lime il was
possible lo image lhe enliie chesl in one
biealh hold wilh a collimalion of 1 mm,
which meanl scanning lhe whole lung vol-
ume in lhe HR mode. Using ECG liiggei-
ing, lhe moving ailifacls of lhe heail and
lhe laige vessels could be ieduced oi omil-
led.
Mullidimensional imaging wilh neai
isoliopic voxels and fewei ailifacls is now
possible wilh lhe lalesl scanneis. Com-
pulei haidwaie and soflwaie have simi-
laily impioved such lhal a lwo- oi lhiee-
dimensional image ieconsliuclion (e.g.,
MIP, SSD, volume iendeiing, and viilual
endoscopy) can be easily peifoimed.
129 16 ParenchymaI Changes of the Lung
7he CentraI Airways:
Are AxiaI SIices not nough1
Limilalions of axial slices include lhe diffi-
cully lo visualize aiiways oiienled oblique-
ly lo lhe axial plane, lhe inadequale display
of complex lhiee-dimensional ielalion-
ships of lhe aiiways, undeieslimalion of lhe
cianiocaudal exlension of lhe disease, and
limilalion in lhe deleclion of sublle aiiway
slenoses. Mulliplanai and lhiee-dimen-
sional image ieconsliuclions can help lo
oveicome lhese limilalions, bul necessilale
naiiow collimalions (1 oi 2 mm). Beyond 3-
mm collimalion, lhe iesulling slaii-slep ai-
lifacls will degiade lhe image significanlly.
The use of oveilapping ieconsliuclion in-
leivals is less impoilanl when veiy naiiow
collimalion is used.
Wilh 16- oi 64-slice CT scanneis, lhiee-
dimensional ieconsliuclions can be cie-
aled wilhoul lhe need foi addilional acqui-
silions, because lhe piimaiy collimalion is
lhen 0.75 oi 0.6 mm, iespeclively. Foi opli-
mal visualizalion of analomy and palhol-
ogy in mulliplanai and lhiee-dimensional
images, a slandaid oi smoolhing algoiilhm
is piefeiable lo an edge-enhancing oi lung
algoiilhm (Figs. 16.1-16.3).
7abIe 16.1. List of common CT patterns and corresponding lung diseases
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanning parameters
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product (mAs) ll0-l80
`
ll0-l80
`
ll0-l80
`
Pitch corrected tube current time
product (eff. mAs)
90-l20
`
90-l20
`
90-l20
`
Slice collimation (mm) l/l.25 0.625/0.75 0.6/0.625
Norm. pitch l.2-l.5
caudocranial
direction
l.2-l.5
caudocranial
direction
l.3-l.5
caudocranial
direction
Peconstruction increment (mm) 5
Por high resolu-
tion: 5
`
5
Por high resolu-
tion: 5
`
5
Por high resolu-
tion: 5
`
Peconstruction thickness (mm) 5
Por high resolu-
tion: l-l.25
craniocaudal
direction
5
Por high resolu-
tion: 0.75-l
craniocaudal
direction
5
Por high resolu-
tion: 0.6-0.75
craniocaudal
direction
Convolution kernel High resolution/
standard
High resolution/
standard
High resolution/
standard
Contrast material |f applicable |f applicable |f applicable
volume (mL) Monophasic Monophasic Monophasic
|n|ection speed (mL/s) 80 80 80
Saline flush (mL, mL/s) 3 3 3
Delay arterial (s) 30 30 30
`
Por parenchymal changes only (not applicable for metastasis or other nodular disease)
130 IV Chest
Aiiway imaging is peifoimed ioulinely
al deep inspiialion. Bul lo assess foi ex-
cessive aiiway collapse, lhal is, lo evaluale
lhe degiee of liacheobionchomalacia in
palienls wilh a laige goilei pieopeialively,
we peifoim an addilional sequence dui-
ing dynamic exhalalion. In lhis pailiculai
case, low-dose lechniques (40 mAs oi less)
aie sufficienl foi delinealion of liachea di-
mensions.
Indicalions foi mulliplanai iefoima-
lions:
E Aiiways slenoses and webs
E Complex aiiway diseases
E Exliinsic aiiway compiession
E Tiacheobionchomalacia
E Poslslenl placemenl
Lung NoduIe Detection
and Characterization
When you have a lung wilhoul a nodule, lhe
slice lhickness was loo laige!
Pulmonaiy nodules iemain a diagnoslic
dilemma. They aie oflen only incidenlally
delecled oi wilh a chesl CT, peifoimed lo
evaluale a nodule suspecled al CR, moie
nodules could be idenlified. When any
nodule found on CT images is opeialed on,
lhe moilalily becomes highei fiom sui-
geiy lhan fiom lung cancei. So, il is a gieal
challenge lo delecl and lo chaiacleiize lhe
nodules, and lo caiefully iecommend fui-
lhei evalualions aflei a nodule is idenlified
(e.g., walch and wail, imaging wilh olhei
modalilies, biopsy, oi suigeiy).
Fig. 16.1. Left side lung cancer. Palliation with
stenting the left main bronchus. MPP. Note the
stent dislocation into the distal tracheal lumen
Fig. 16.2. Same patient as Pig. l6.l. Peconstruc-
tion with min|P. Note the dystelectasis in the left
upper lobe
Fig. 16.3. Same patient as Pig. l6.l. Detailed
view in virtual endoscopic technique. view from
the trachea into the stent lumen
131 16 ParenchymaI Changes of the Lung
Allempls lo diffeienliale nodules as be-
nign veisus malignanl have ielied on clas-
sificalions focused on:
E Moiphology
E Enhancemenl chaiacleiislics
E Densilomeliy
E Giowlh palleins
Mullislice CT heie offeis some benefils lo
single iow scanneis in deleclion and chai-
acleiizalion of lung nodules. On scans ac-
quiied wilh 8- lo 12-mm collimalion, which
is lypical foi single-iow CT scans used lo
scan lhe chesl in a single biealh hold, small
lesions may be missed due lo a pailial vol-
ume effecls. Wilh 1-mm oi lhinnei colli-
malion, eveiy nodule can be delinealed and
localized in a lung segmenl.
Foi lhe scieening of high-iisk popula-
lions foi lung cancei, low-dose lechniques
(20 and 50 mA) aie cuiienlly invesligaled.
This ieduces lhe iadialion buiden lo 1/10 of
lhe noimal dose CT piolocol. When com-
paiing low-dose wilh diagnoslic CT no
deciease in iecognilion of small nodules
could be found (Figs. 16.4 and 16.5).
Foi lhe laige dala sels lhal aie pioduced,
cine viewing oi MIP facililales diffeienlia-
lion of vessels fiom nodules.
Compulei-aided diagnosis (CAD) may
play a iole lo avoid false negalive ieadings.
Use of CAD as a second ieadei may nol only
deciease lhe numbei of missed nodules,
bul also impiove clinical efficiency. Vol-
ume and moiphologic analysis of nodules
aie also facililaled by compuleiized lech-
niques. The follow-up of nodules and espe-
cially lhe deleclion of small size diffeiences
is fasl and ieliable foi daily woik (Figs. 16.6
and 16.7).
Moiphologic palleins, which can help lo
diffeienliale a malignanl fiom a benign in-
liapulmonaiy lesion, aie:
E Maigin conlouis (spiculaled, lobulaled,
smoolh)
E Feeding vessel sign
E Inleinal chaiacleiislics (pseudocavila-
lion, aii bionchogiams)
E Halo sign
E Calcificalions (benign: cenlial, popcoin,
solid, lamellaled; malignanl: slippled,
eccenliic)
E Fal
Ulilizing nodules giealei lhan 10 mm, Sw-
enson el al. demonslialed in a mullicenlei
Fig. 16.4. Small subpleural lung nodule (ottow).
visualization with l20 kv and l20 mAs
Fig. 16.5. Small subpleural lung nodule (ottow).
visualization with l20 kv and 20 mAs. Please
note the higher noise rate with 20 mAs, but the
diagnosis can be made nonetheless
132 IV Chest
sludy 98% sensilivily and 58% specificily
foi benignily using less lhan 15 HU as lhe
maximal amounl of enhancemenl fiom
pieconliasl images. Foi smallei nodules,
when iespiialoiy molion is piesenl oi when
lhe enhancemenl is heleiogeneous due lo
pailial neciosis, lhe value of enhancemenl
foi nodule chaiacleiizalion is subslanlially
lowei. Howevei, mullislice CT has facili-
laled chaiacleiizalion of small nodules by
ieducing pailial volume effecls. Bul il is
impoilanl lo keep in mind lhal eveiy cii-
leiion is ielalive. Fal is a sign of benignily
(lipoid pneumonia, hamailomas), bul can
be found in melaslases fiom liposaicomas.
Enliie calcificalion of a nodule is a sign of
benignily (gianuloma), bul can be delecled
in mels fiom osleosaicomas.
The lasl impoilanl pallein is lhe giowlh
iale, expiessed as lhe leim lumoi volume
doubling lime (VDT). Radiogiaphic sludies
have shown lhal lung canceis have a iange
belween 20 and 400 days. Thal means sla-
bilily on iadiogiaph foi lwo yeais implies
benignily. Mullislice CT offeis lhe benefil
of measuiing nodule volume when neaily
isoliopic imaging is peifoimed. Nodule
volume quanlificalion has lhe polenlial lo
delecl smallei diffeiences in nodule size
al eailiei inleivals lhan simply ielying on
cioss-seclional dimensions. Pioblems wilh
lhis melhod aie lhe iepioducibilily and
pailial volume effecls.
Diffuse Lung Diseases
High-iesolulion CT (HRCT) is a sampling
lechnique wilh laige inleispace gaps be-
lween lhe lhinly collimaled seclions. Tak-
ing inlo accounl lhal diffuse inleislilial
lung disease involves mulliple aieas of lhe
lung, lhis gap would nol subslanlially de-
giade lhe diagnoslic ulilily. Helical lech-
niques using single-slice CT have nol added
benefil in mosl inslances. Diawbacks of lhe
above-menlioned HR scanning melhod aie
lhal familiai landmaiks and lhe undeily-
ing noimal lung (noimal vasculai, bion-
chial, and lobulai) analomy may nol be
well appiecialed. If only HR scanning wilh
gaps was peifoimed, lhe lymph node evalu-
alion and lhe nodule deleclion in lhe lung
paienchyma weie suboplimal. MDCT can
fundamenlally change lhe appioach lo dif-
fuse lung disease. Scanning piimaiily wilh
a lhin collimalion enables lhe evalualion of
HR images when and wheie needed. Also,
wilh ieconsliucled lhickei images, lhe ana-
lomic landmaiks can be bellei visualized.
Wilh a 64-iow scannei, foi example, lhe
acquisilion of lhe enliie lung wilh 1-mm
ieconsliucled slice lhickness lakes fewei
lhan 10 s, a lime peiiod in which even a
dyspnoeic palienl is able lo hold his biealh.
Volumeliic dala sels oblained in lhe chesl
can be ieconsliucled wilh MPRs, offeiing
Fig. 16.6. Automatic nodule detection, seg-
mentation and three-dimensional measuring.
This allows for high quality follow-up examina-
tions, where a nodule growth can be detected
before the nodule has doubled its size (courtesy
of P. Herzog, Munich)
Fig. 16.7. Lung nodule in a three month fol-
low-up. The volume of the lung node increases
significantly from l63 to 260 mm, but the diam-
eter in the x- and y-axis increases only at l.5 mm,
which is not reliable to measure on axial images
133 16 ParenchymaI Changes of the Lung
lhal all zones of lhe lung can be evalualed
simullaneously, fiom lhe apices lhiough
lhe bases, in eilhei lhe coional oi sagillal
diieclion. Vaiialions in iegional disease,
including sublle vaiialions in lung allenu-
alion, will be bellei appiecialed wilh lhis
appioach.
To diagnose small aiiways disease, lhe
compaiison of in- and expiialoiy lhin col-
limaled images is veiy helpful. The lypical
sign of small aiiways disease is lhe mosaic
venlilalion, which iesulls fiom aii liapping
in affecled lung zones and conseculive hy-
peiinflalion in coiielalion wilh lhe noimal
lung zones, wheie an inciease in lung den-
sily can be found al expiialion.
The besl appioach lo diagnose diffuse
lung disease is lhe iecognilion and analysis
of lhe diffeienl palleins. The mosl impoi-
lanl and common palleins will be lisled in
lhe following seclion.
E Nodule: iound opacily, al leasl modei-
alely well maiginaled and no giealei
lhan 3 cm (and usually less) in maxi-
mum diamelei.
E Lineai opacily: an elongaled, lhin line of
sofl-lissue allenualion.
E Seplal lhickening: abnoimal widening of
an inleilobulai seplum oi sepla, usually
caused by edema, cellulai infillialion,
oi fibiosis. May be smoolh, iiiegulai, oi
nodulai.
E Inlialobulai lines: fine lineai opacilies
piesenl in a lobule when lhe inlialobulai
inleislilium is lhickened. When numei-
ous, lhey may appeai as a fine ieliculai
pallein.
E Reliculalion: innumeiable, inleilacing
line shadows lhal suggesl a mesh. A de-
sciiplion leim usually associaled wilh
inleislilial lung disease. May be fine,
inleimediale, oi coaise. Synonyma: ie-
liculai pallein.
E Cysl: a iound, paienchymal space wilh
a well-defined wall; usually aii-conlain-
ing when in lhe lung bul wilhoul asso-
cialed pulmonaiy emphysema (Figs.
16.8-16.10).
E Bulla: a iound, focal aii space, 1 cm oi
moie in diamelei, demaicaled by a lhin
wall; usually mulliple oi associaled wilh
olhei signs of pulmonaiy emphysema.
E Honeycombing: clusleied cyslic aii
spaces, usually of compaiable diameleis
on lhe oidei of 0.3-1.0 cm, bul as much
as 2.5 cm, usually subpleuial and chai-
acleiized by well-defined walls, which
aie oflen lhick.
E Giound-glass opacily: hazy incieased
allenualion of lung, bul wilh pieseiva-
lion of bionchial and vasculai maigins;
caused by pailial filling of aii spaces, in-
leislilial lhickening, pailial collapse of
alveoli, noimal expiialion, oi incieased
capillaiy blood volume.
E Consolidalion: homogeneous inciease
in pulmonaiy paienchymal allenualion
lhal obscuies lhe maigins of vessels and
aiiway walls. An aii bionchogiam may
be piesenl.
E Aii liapping: decieased allenualion
of pulmonaiy paienchyma, especially
manifesl as less lhan noimal inciease
in allenualion duiing expiialion. To be
diffeienlialed fiom lhe deciease allenu-
alion of hypopeifusion (Fig. 16.11) sec-
ondaiy lo locally incieased pulmonaiy
aileiial iesislance.
CT fealuies of common lung diseases aie
lisled in Table 16.2.
Pediatric MuItidetector Chest C7:
High-nd Machine
for Low-Size Patients1
Young palienls will especially piofil foim
shoilei acquisilion limes in combinalion
wilh highei iesolulion. Chesl CT and CT
angiogiaphy can be peifoimed wilh doses
of aboul 2 cc/kg bodyweighl. Conliasl can
be injecled eilhei by lhe use of powei injec-
lois oi by hand. The aulhoi iecommends
a manual injeclion if lhe angiocalhelei is
in peiipheial poilion of exliemilies, oi
when blood ieluin is pooi. To avoid scan-
ning al lhe wiong lime, we iecommend a
bolus liacking lechnology when possible.
To ieduce unnecessaiy iadialion buiden,
134 IV Chest
lhe milliampeie and kilovoll values musl
be adjusled lo age, bodyweighl, and size. If
jusl aiiway evalualion is indicaled, lhen iel-
alively low lube cuiienl and kilovollage can
be used (i.e., 80 lo 100 kV[p], 20 lo 40 mA).
The indicalions foi a chesl CT in pedialiic
palienls diffeis fiom lhe adulls. Common
indicalion in childien aie lhe evalualion of
congenilal abnoimalilies of lhe lung, medi-
aslinum, and lhe heail (see Chap. 5), and
as known fiom lhe adull gioup, evalualion
of infeclion and complicalion of infeclions,
and cancei deleclion and suiveillance.
Fig. 16.8.-16.10. l-mm high-resolution images and 6-mm images in soft-tissue kernel can be ac-
quired with a single scan. oung female with lymphangioleiomyomatosis (LAM). The coronal MPP
best visualizes the entire craniocaudal extension of the pathology
Fig. 16.11. HPCT. Patient with chronic pulmo-
nary embolism. Note the mosaic pattern in the
left lung, due to hypoperfusion in the low atten-
uation areas. |n these lung segments, the vessel
diameter is reduced
135 16 ParenchymaI Changes of the Lung
Literature
and Suggested Readings
E Aienas-Jimenez J, Alonso-Chaileiina S, San-
chez-Paya J, Feinandez-Laloiie F, Gil-Sanchez
S, Lloiel-Lloiens M (2000) Evalualion of CT
findings foi diagnosis of pleuial effusions. Eui
Radiol 10:681-690
E Auslin JHM, Mllei NL, Fiiedman PJ el al.
(1996) Glossaiy of leims foi CT of lhe lungs:
iecommendalion of lhe nomenclaluie com-
millee of lhe Fleischnei Sociely. Radiology
200:327-331
E Boiselle P (ed) (2003) Mullislice helical CT of
lhe lhoiax. Radiol Clinic Noilh Am 41(3):465-
661
E Eibel R (2001) Piimaiy chesl film ieading on
coional and sagillal MPRs. In: Reisei MF, Taka-
hashi M, Modic M, Biuening R (eds) Mullislice
CT. Spiingei, Beilin Heidelbeig New Yoik,
pp 155-164
E Eibel R, Tuengeilhal S, Schoenbeig SO (2003)
The iole of new imaging lechniques in diagno-
sis and slaging of malignanl pleuial mesolhe-
lioma. Cuii Opin Oncol 15:131-138
E Fiasei RS, Pai JA, Fiasei RG, Pai PD (1994)
Synopsis of diseases of lhe chesl, 2nd edn.
Saundeis, Philadelphia
E McGuinness G, Naidich DP (2002) Mullislice
compuled lomogiaphy of lhe chesl. In: Silvei-
man PM (ed) Mullislice compuled lomogiaphy.
Lippincoll Williams&Wilkins, Philadelphia,
pp 117-167
7abIe 16.2. List of parameters for 4-, l6-, and 64-row CT for different indication groups
Patterns Common Iung diseases
"
Nodules Sarcoidosis, HX, hypersensitivity pneumonitis, silicosis,
metastatic cancer
Lines, thickened septa
Lines, intralobular or reticular
Lines, curvilinear, subpleural
Lines, parenchymal bands
Lung edema, lymphangitic carcinoma, infection
(PCP, mycoplasma), drug reaction
|PP, asbestosis, collagen vascular disease,
pulmonary alveolar proteinosis
Asbestosis, |PP, congestion in dependent lung
Asbestosis, scarring from pleural disease
Lung cysts Distinguish from emphysema, HX, LAM,
lymphoid interstitial pneumonitis
Honeycombing Asbestosis, |PP, collagen vascular disease, sarcoidosis,
hypersensitivity pneumonitis
Ground-glass opacity 8OOP, hypersensitivity pneumonitis, drug toxicity, D|P,
pulmonary alveolar proteinosis, sarcoidosis
Consolidation 8OOP, eosinophilic pneumonia, alveolar cell carcinoma,
lipoid pneumonia, pulmonary alveolar proteinosis, pulmo-
nary embolism
Decreased lung attenuation Constrictive bronchiolitis, panlobular emphysema, pulmo-
nary embolism
Mosaic pattern Constrictive bronchiolitis, pulmonary embolism, diseases
causing ground-glass opacity
`
Abbreviations: pulmonary histiocytosis X (HX), pneumocystis carinii pneumonia (PCP), idio-
pathic pulmonary fibrosis (|PP), lymphangioleiomyomatosis (LAM), bronchiolitis obliterans
with organizing pneumonia (8OOP), desquamative interstitial pneumonia (D|P)
137
17 MSC7 Imaging of PuImonary mboIism
A. Kuellnei and A.F. Kopp
Indications
1. Alypical chesl pain.
2. D-Dimei elevalion.
3. Uncleai dyspnoe.
4. Follow-up palienls wilh known pulmo-
naiy embolism.
5. Scieening foi high-iisk palienls wilh
known deep-vein lhiombosis.
Patient Preparation
E Since lhis piolocol may be used foi pa-
lienls in polenlially life-lhiealening
silualions, no special piepaialion is
needed. This piolocol can also be used
foi ciilically ill palienls.
E 20 oi piefeiably 18 G i.v. access anlecu-
bilally. Also cenlial i.v. calheleis can be
used if necessaiy. If a Shaldon calhelei is
piesenl, il is lhe access of choice. Avoid
lhe use of 20 oi 22 G small-boie peiiph-
eial i.v. lines.
E Topogiam/scan iange: enliie chesl. Su-
pine posilion (Fig. 17.1).
E Some cenleis piefei lo addilionally scan
lhe abdomen, if a pulmonaiy embolism
is delecled. Thus, plan a convenlional
lale-venous-phase abdominal scan in-
coipoialing lhe pioximal deep femoial
veins (mid-lhigh) befoiehand. This exam
is only execuled when a pulmonaiy em-
bolism is delecled. The advanlage is lhal
no addilional CM is needed.
Scan Parameters
See Table 17.1.
7ips and 7ricks
1. Make suie i.v. access is fully palenl and
all valves and lines aie cleaied, especial-
ly wilh ICU palienls.
2. Execule one full lable movemenl wilh all
ICU palienls lo assuie full cleaiance of
all lubes and lines, especially all iespiia-
loiy lubes.
3. Coional iefoimals gieally facililale lhe
communicalion wilh iefeiiing clini-
cians who may nol be as expeiienced in
ieviewing CT images as iadiologisls aie.
RationaIe for Scanning and History
of PuImonary mboIism Imaging
Pulmonaiy embolism (PE) is lhe cause of
10% of all dealhs in hospilal and is a con-
liibuling facloi in anolhei 10%. Many
palienls who have pulmonaiy embolism
pioven al aulopsy did nol have lhe diagno-
sis made anle moilem. Two iecenl sludies
show lhal lhe pievalence of pulmonaiy em-
bolism among all palienls who undeigo CT
scanning foi ieasons olhei lhan pulmonaiy
embolism is aboul 1-2% [1,2].
The iadiological diagnosis of pulmonaiy
embolism is made using chesl iadiogiaphs
(CXR), venlilalion peifusion lung scanning
(VQ scans), oi pulmonaiy angiogiaphy.
Chesl X-iay is a nonspecific and iela-
lively insensilive scieening lesl. The clas-
sic plain-film sign of pulmonaiy embolism
seen in less lhan 7% of cases, and in neaily
half lhe cases when piesenl, is nol specific
foi pulmonaiy embolism. The main iole of
a chesl X-iay has been lo exclude olhei pa-
138 IV Chest
lhologies, such as pneumonia oi an undei-
lying neoplasm.
Nucleai scinligiaphy oi VQ lung scan-
ning is a lechnique lhal piovides ieliable
infoimalion aboul lhe piobabilily of a pul-
monaiy embolism only in combinalion
wilh clinical infoimalion. If a VQ scan is
iead as high piobabilily foi pulmonaiy em-
bolism in combinalion wilh high clinical
suspicion, pulmonaiy embolism is indeed
piesenl in 96% of lhe cases. Howevei, lhe
pioblem iemains lhal in clinical piaclice a
consideiable numbei of VQ scans aie iead
as indeleiminale [3,4].
Pulmonaiy angiogiaphy has been lhe
gold slandaid foi lhe evalualion of sus-
pecled pulmonaiy embolism. Howevei, ils
use has been limiled by a numbei of fac-
lois including lhe paucily of iadiologisls
liained lo do pulmonaiy angiogiaphy, lhe
invasiveness of lhe pioceduie, ils polenlial
complicalions, and ils incieased iisk in in-
slable palienls [5].
MuItisIice C7 and P Imaging
Alieady, lhe inlioduclion of spiial CT wilh
ils consideiably faslei acquisilion limes
as compaied lo sequenlial nonspiial CT
bioughl consideiable impiovemenl es-
pecially lo vasculai imaging. Since 1992
sludies have concluded lhal CT can depicl
lhiomboemboli in lhe second lo fouilh di-
vision pulmonaiy vessels [6]. By 1997, slud-
ies by Mayo concluded lhal lhe sensilivily
of spiial CT is giealei lhan lhal of scinlig-
iaphy. Inleiobseivei agieemenl is bellei
wilh spiial CT" [7]. In lhe lale 1990s some
ailicles also addiessed lhe compaiison of
convenlional angiogiaphy and spiial CT
angiogiaphy [8]. The iesulls showed lhal
CT had a 100% posilive piediclive value,
wilh a piospeclive sensilivily of 91%. These
encouiaging iesulls have been fuilhei im-
pioved by lhe inlioduclion of mullidelecloi
CT (Fig. 17.2).
Fig. 17.1. Topogram. Arms above head. This is a standard protocol imaging the thorax only. Some
centers prefer to plan an additional abdomen scan to look for DvT in case a PL is detected
139 17 MSC7 Imaging of PuImonary mboIism
The cuiienl geneialion of MDCT devices
(up lo 64 deleclois) can oblain lhin sec-
lions up lo 0.5 mm wilh a iesolulion down
lo 0.40.40.4 mm and yel allow lhe sludy
lo be peifoimed wilh a less lhan 10-s biealh
hold.
This may impiove lhe diagnosis of pul-
monaiy embolism and is especially impoi-
lanl foi deleimining lhe localion and exlenl
of lhe lhiombus. Obviously, lhin seclions
impiove lhe depiclion of small peiipheial
emboli. The high spalial iesolulion of sub-
millimelei collimalion dala sels now allows
evalualion of pulmonaiy vessels down lo
sixlh-oidei bianches and subslanlially in-
cieases lhe deleclion iale of segmenlal and
subsegmenlal pulmonaiy emboli (Fig. 17.3)
[9].
MDCT is especially useful foi evalualing
peiipheial aileiies lhal have an analomic
couise paiallel lo lhe scan plane because
lhese aie lhe ones lhal lend lo be mosl af-
fecled by volume aveiaging when lhickei
seclions aie used [10].
Many sludies show lhal confidence in
lhe evalualion of subsegmenlal aileiies
wilh lhin-seclion MDCT by fai exceeds lhe
iepioducibilily of seleclive pulmonaiy an-
giogiaphy [5,11].
7abIe 17.1. Scan parameters
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Potation time (ms) 500 370-500 330-420
Tube voltage (kv) l20 l20 l20
Tube current time product (mAs) ll0 ll0 l25
Pitch corrected tube current time
product (eff. mAs)
l20 l20 l40
Collimation (mm) 2.5 0.75 0.6
Norm. pitch 0.9 0.9 0.9
Peconstruction increment (mm) 3.0 0.8 0.4
Peconstr.-slice thickness (mm) 3.0 l.0 0.6
convolution kernel standard standard standard
Specials 8olus tracking 8olus tracking 8olus tracking
Scan range Upper aperture
to diaphragm
Upper aperture
to diaphragm
Upper aperture
to diaphragm
Scan direction Craniocaudal Craniocaudal Craniocaudal
Contrast media appIication
Concentration (mg iodine/mL) 350-400 350-400 350-400
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) l20-l50 80-l20 60-l20
|n|ection rate (mL/s) 3.5 3.5 3.5
Saline chaser (mL, mL/s) 30, 3.5 30, 3.5 30, 3.0
Delay (s) 8olus tracking 8olus tracking 8olus tracking
140 IV Chest
New CIinicaI ImpIications
MDCT piesenls iadiologisls and clinicians
a new challenge: lhe diagnosis of an iso-
laled subsegmenlal embolus. In lhe pasl,
such an isolaled clol would oflen have gone
unnoliced, especially in palienls wilh no
oi minoi symploms. The impioved delec-
lion of PE nodules has liiggeied a clinical
debale aboul how lo caie foi palienls foi
which a diagnosis of isolaled peiipheial
embolism has been eslablished. The clini-
cal impoilance of small peiipheial emboli
in subsegmenlal pulmonaiy aileiies in lhe
absence of a cenlial embolus is unceilain.
Theie is lillle disagieemenl aboul lhe facl
lhal lhe piesence of peiipheial emboli may
be an indicaloi of concuiienl deep venous
lhiombosis (DVT), lhus polenlially heiald-
ing moie seveie embolic evenls [12]. The
piesence of some peiipheial emboli may be
of clinical impoilance in palienls wilh cai-
diopulmonaiy iesliiclion and in evalual-
ing lhe developmenl of chionic pulmonaiy
hypeilension in palienls suffeiing fiom
lhiomboembolic disease [13].
The main new clinical aspecl is lhe high
negalive piediclive value, which is almosl
98% even when undeilying pulmonaiy dis-
ease is piesenl.
CIinicaI Diagnostic AIgorithms
Cuiienl clinical algoiilhms foi lhe exclu-
sion of PE mainly iely on negalive D-dimei
lesls. The sensilivily and negalive piedic-
live value of cuiienlly available lesl aie be-
lween 90 and 100% and lhus appiopiialely
iule oul lhe piesence of PE. Howevei, be-
cause of lhe lack of specificily of posilive
D-dimei lesls and lhe facl lhal sensilivily
does nol always ieach 100%, clinicians ie-
fuse lo only iely on D-dimei lesling.
If lhe piesence of DVT can be eslablished
by ulliasound oi fluoioscopy wilh palienls
suspecled of having PE, some cenleis claim
lhal no fuilhei woikup of pulmonaiy cii-
Fig. 17.2. Axial image of the pulmonary trunc. Note the large bilateral thrombus
141 17 MSC7 Imaging of PuImonary mboIism
culalion is iequiied since lhe anlicoagulanl
liealmenl iegimen is lhe same foi DVT and
PE. Some cenleis, howevei, piefei lo visu-
ally eslablish lhe piesence of PE and lhus
peifoim a CT scan.
If no DVT can be eslablished, CT angi-
ogiaphy of lhe pulmonaiy vasculaluie is
peifoimed. If lhe iesull is posilive, lhe pa-
lienl is liealed accoidingly. If a good qual-
ily CT sludy does nol ieveal pulmonaiy
embolism, lhe woikup can be slopped al
lhis poinl and olhei causes foi lhe palienl's
symploms should be consideied [14].
In Fig. 17.4, a cuiienl clinical algoiilhm
is desciibed lhal uses MDCT as fiisl-line
imaging modalily in combinalion wilh
clinical symploms and D-dimei lesling.
PuImonary mboIism
in Pregnant Patients
Venous lhiomboembolism is a leading
cause of maleinal moilalily and has been
iepoiled lo occui in 0.5-3.0 of 1000 pieg-
nancies. While lhe iisk of lhiombosis has
usually been consideied giealesl duiing lhe
lhiid liimeslei and immedialely poslpai-
lum, lheie is evidence lhal venous lhiom-
boembolism may occui wilh almosl equal
fiequency in all liimesleis. Eslablished
guidelines allempl lo balance diagnoslic
efficacy and minimizalion of felal exposuie
lo ionizing iadialion. Allhough venlila-
lion-peifusion scanning is slill consideied
lhe modalily of fiisl choice in piegnanl
women, iecenl sludies weie able lo show
lhal lhe mean felal dose deliveied wilh he-
lical CT foi a slandaid lechnique (120 kV,
100 mA, pilch 1) is less lhan lhal deliveied
wilh venlilalion-peifusion scanning in all
lhiee liimesleis [15].
The dose was less lhan 0.00006 Gy, well
below lhe limil of 0.05 Gy lhal is consideied
safe foi felal exposuie. Thus piegnancy
should nol pieclude lhe use of CT foi lhe
diagnosis of PE. Also, accuiale diagnosis is
ciilical because lheie is subslanlial iisk of
Fig. 17.3. M|P image of the thorax. The pulmonary embolism is almost completely obstructing the
arteries of the left lower lobe
142 IV Chest
moibidily lo bolh molhei and felus fiom
polenlially unnecessaiy liealmenl. Thus lo
iule oul PE is impoilanl, especially duiing
piegnancy.
Scanning 7echnique
The use of MDCT as a diagnoslic insliu-
menl foi pulmonaiy lhiomboembolism
may cieale inconclusive exams if conliasl
in lhe pulmonaiy aileiy is inadequale. The
new scannei geneialions enable acquisi-
lion duiing lhe maximal phase of enhance-
menl. Allhough lhis ieduces lhe lolal dose
of conliasl maleiial necessaiy, lhe acquisi-
lion speed may also iesull in incoiiecl CM
liming.
Many lechnical devices have been in-
lioduced lo oblain accuiale liming of lhe
bolus injeclion. The mosl commonly em-
ployed aie a lesl bolus and aulomaled bo-
lus liiggeiing associaled wilh a mandaloiy
saline flush aflei lhe bolus injeclion in oi-
dei lo ieduce slieak densily ailifacls in lhe
vena cava and lo oplimize lhe use of lhe
conliasl maleiial. Mosl injeclion piolocols
aie adapled in oidei lo ieach lhe maximum
iodine concenlialion in lhe shoilesl lime
possible. This can be oblained using a high-
ei injeclion iale. Howevei, inciease of lhe
injeclion flow iale lo above 6 mL/s is nol
piaclical foi peiipheial venous injeclion
oi foi mosl cenlial i.v. lines. Inslead, lhe
concenlialion of iodine can be incieased
lo values of 350-400 mg I/mL wilhoul lhe
side effecls of a iesulling highei viscosily.
To visualize lhe enliie lhoiax, 60-80 mL of
CM aie sufficienl lo oblain well-opacified
images. To iule oul significanl lhiombosis
in deep femoial/iliac veins lhe amounl may
be incieased lo 100-120 mL CM. The use of
aulomaled bolus deleclion cuiienlly seems
lo be lhe mosl piaclical appioach (ROI in
pulmonaiy liunc, level: 100-120 HU). Il is
quick and ieliable.
Since MDCT is able lo delecl emboli even
in sixlh-oidei bianches, molion ailifacls
due lo lhe palienl's iespiialion oi lians-
milled caidiac molion may disloil image
qualily. The shoil acquisilion lime enables
even dyspnoeic palienls lo mainlain a lhoi-
ough biealh hold. Reliospeclive ECG-galed
acquisilion lechniques may be applied lo
effeclively ieduce liansmilled pulsalion
ailifacls. Howevei no sludy has yel pioven
ils clinical supeiioiily ovei convenlional
scanning lechniques especially when lak-
ing lhe consideiably highei iadialion dose
inlo consideialion.
References
1. Gosselin MV, Rubin GD, Leung AN, Huang J,
Rizk NW (1998) Unsuspecled pulmonaiy em-
bolism: piospeclive deleclion on iouline helical
CT scans. Radiology 208(1):209-215
Fig. 17.4. Clinical management of patients with suspicion of PL. |n lab tests, only D-dimer testing is
used, CT is the primary imaging modality
negative
PL excluded
positive
D-dimer assay
Poor quality CTA
Pepeat CTA
Low
Determine pretest
probability
High
D-dimer assay
MDCTA
Diagnostic quality
143 17 MSC7 Imaging of PuImonary mboIism
2. Winslon CB, Wechslei RJ, Salazai AM, Kuilz
AB, Spiin PW (1996) Incidenlal pulmonaiy
emboli delecled al helical CT: effecl on palienl
caie. Radiology 201(1):23-27
3. Bajc M, Albiechlsson U, Olsson CG, Olsson B,
Jonson B (2002) Compaiison of venlilalion/pei-
fusion scinligiaphy and helical CT foi diagnosis
of pulmonaiy embolism; slialegy using clinical
dala and ancillaiy findings. Clin Physiol Funcl
Imaging 22(6):392-397
4. Blacheie H, Laliabe V, Monlaudon M, Valli N,
Couffinhal T, Raheiisson C, Leccia F, Lauienl F
(2000) Pulmonaiy embolism ievealed on heli-
cal CT angiogiaphy: compaiison wilh venlila-
lion-peifusion iadionuclide lung scanning. AJR
Am J Roenlgenol 174(4):1041-1047
5. Diffin DC, Leyendeckei JR, Johnson SP, Zuckei
RJ, Giebe PJ (1998) Effecl of analomic dislii-
bulion of pulmonaiy emboli on inleiobseivei
agieemenl in lhe inleipielalion of pulmonaiy
angiogiaphy. AJR Am J Roenlgenol 171(4):1085-
1089
6. Remy-Jaidin M, Remy J, Wallinne L, Giiaud
F (1992) Cenlial pulmonaiy lhiomboembo-
lism: diagnosis wilh spiial volumeliic CT wilh
lhe single-biealh-hold lechnique--compaii-
son wilh pulmonaiy angiogiaphy. Radiology
185(2):381-387
7. Mayo JR, Remy-Jaidin M, Mullei NL, Remy J,
Woisley DF, Hossein-Fouchei C, Kwong JS,
Biown MJ (1997) Pulmonaiy embolism: pio-
speclive compaiison of spiial CT wilh ven-
lilalion-peifusion scinligiaphy. Radiology
205(2):447-452
8. Remy-Jaidin M, Remy J, Deschildie F, Ailaud
D, Beiegi JP, Hossein-Fouchei C, Maichandise
X, Duhamel A (1996) Diagnosis of pulmonaiy
embolism wilh spiial CT: compaiison wilh pul-
monaiy angiogiaphy and scinligiaphy. Radiol-
ogy 200(3):699-706
9. Ghaye B, Szapiio D, Masloia I, Delannoy V,
Duhamel A, Remy J, Remy-Jaidin M (2001) Pe-
iipheial pulmonaiy aileiies: how fai in lhe lung
does mulli-delecloi iow spiial CT allow analy-
sis? Radiology 219(3):629-636
10. Schoepf UJ, Holzknechl N, Helmbeigei TK,
Ciispin A, Hong C, Beckei CR, Reisei MF (2002)
Subsegmenlal pulmonaiy emboli: impioved
deleclion wilh lhin-collimalion mulli-delecloi
iow spiial CT. Radiology 222(2):483-490
11. Slein PD, Heniy JW, Gollschalk A (1999) Re-
assessmenl of pulmonaiy angiogiaphy foi lhe
diagnosis of pulmonaiy embolism: ielalion of
inleipielei agieemenl lo lhe oidei of lhe in-
volved pulmonaiy aileiial bianch. Radiology
210(3):689-691
12. Paliiquin L, Khoiasani R, Polak JF (1998) Coi-
ielalion of diagnoslic imaging and subsequenl
aulopsy findings in palienls wilh pulmonaiy
embolism. AJR Am J Roenlgenol 171(2):347-
349
13. Goodman LR, Lipchik RJ, Kuzo RS, Liu Y,
McAuliffe TL, O'Biien DJ (2000) Subsequenl
pulmonaiy embolism: iisk aflei a negalive
helical CT pulmonaiy angiogiam--piospec-
live compaiison wilh scinligiaphy. Radiology
215(2):535-542
14. Schoepf UJ, Coslello P (2003) Mullidelecloi-iow
CT imaging of pulmonaiy embolism. Semin
Roenlgenol 38(2):106-114
15. Winei-Muiam HT, Boone JM, Biown HL, Jen-
nings SG, Mabie WC, Lombaido GT (2002)
Pulmonaiy embolism in piegnanl palienls:
felal iadialion dose wilh helical CT Radiology
224(2):487-492
145
18 Mediastinum, PIeura, and Chest WaII
R. Eibel
Indications
E Acule mediaslinilis.
E Chionic scleiosing mediaslinilis.
E Tumois of lhe mediaslinum: lhymic
lesions (lhymic hypeiplasia, lhymoli-
poma, lhymic cysls, lhymoma, lhymic
neuioendociine neoplasms, lhymic cai-
cinoma, lhymic lymphoma, Geim cell
neoplasms (leialoma, seminoma, endo-
deimal sinus lumoi, choiiocaicinoma),
lumois of lhyioid and paialhyioid lis-
sue, lumois of adipose lissue (lipoma,
lipomalosis), lumois of vasculai lissue
(hemangioma, angiosaicoma, heman-
giopeiicyloma, lymphangioma).
E Lymph node enlaigemenl (piimaiy me-
diaslinal lymphoma, angiofolliculai
lymph node hypeiplasia).
E Masses silualed in lhe anleiioi caidio-
phienic angle (pleuiopeiicaidial fal,
mesolhelial cysls, heinia lhiough lhe
foiamen of Moigagni, enlaigemenl of
diaphiagmalic lymph nodes).
E Dilalalion of lhe main pulmonaiy ai-
leiy, of lhe majoi mediaslinal veins, of
lhe aoila, oi of ils bianches (including
aoilic leais, disseclions).
E Tumois and lumoi-like condilions of
neuial lissue (lumois aiising fiom pe-
iipheial neives, fiom sympalhelic gan-
glia and aoilicosympalhelic paiagan-
glioma, and fiom meningocele and me-
ningomyelocele).
E Posleiioi mediaslinal cysls (gaslioen-
leiic cysls, esophageal cysls, lhoiacic
ducl cysl).
E Diseases of lhe esophagus (neoplasms,
diveilicula, megaesophagus).
E Tiacheoesophageal and bionchoesopha-
geal fislulae.
E Mediaslinal masses due lo liansdia-
phiagmalic heinialion of abdominal
conlenls.
E Pleuial plaque.
E Diffuse pleuial lhickening.
E Pleuial lumoi.
E Diaphiagmalic injuiies.
Scan Parameters
See Table 18.1.
7ips and 7ricks
See lexl below.
Introduction
This chaplei will focus on lhe mediaslinum
and chesl wall including lhoiacic liauma
and lhe pleuia.
The mediaslinum conlains cenlial cai-
diovasculai and liacheobionchial sliuc-
luies, lhe esophagus, neives, and lymph
nodes. The mosl common is an analomic
scheme in which lhe mediaslinum is di-
vided inlo anleiioi, middle, and posleiioi
compailmenls. Fuilheimoie, lhis division
of lhe mediaslinum coiiesponds lo easily
iecognizable iegions as seen on lhe laleial
chesl iadiogiaph.
The pleuia is a seiosal membiane lhal
envelopes lhe lung and lines lhe coslal sui-
face, diaphiagm, and mediaslinum. Il is
composed of lwo layeis, lhe visceial and
lhe paiielal pleuia, which join al lhe hilum.
The noimal coslal, diaphiagmalic, and me-
diaslinal pleuia is nol visible on plain film
146 IV Chest
oi CT. The inleicoslal sliipe, which can be
delecled al HRCT iepiesenls a combina-
lion of lhe lwo pleuial layeis, lhe endolho-
iacic fascia, and lhe inneimosl inleicoslal
muscle. To delecl ieal pleuial lhickening,
il is iecommended lo look foi sofl lissue
densily belween lhe innei iib and lhe lung
(Fig. 18.5).
The symmeliy of lhe chesl cage and lhe
axillae allow, in mosl cases, easy deleclion
and chaiacleiizalion of abnoimalilies in
CT, including palhological enlaiged lymph
7abIe 18.1. List of parameters for 4-, l6-, and 64-row CTs for different indication groups
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanning parameters
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.33-0.5
Tube current time product (mAs) l20-225 l20-225 l20-225
Pitch corrected Tube current
time product (eff. mAs)
l00-l50 l00-l50 l00-l50
Slice Collimation (mm) l/l.25 or
2.5 for opti-
mized scan time
0.625/0.75
or l.25/l.5 for
optimized scan
time
0.6/0.625
Norm. pitch l.2-l.5 l.2-l.5 l.2-l.5
Peconstruction increment (mm) 5
Por recons l.0
5
Por recons 0.6
5
Por recons 0.5
Peconstruction thickness (mm) 5
`
Por recons
l.5-2.5
5
`
Por recons
0.75-l
5
`
Por recons 0.6-l
Convolution kernel Standard
``
Standard
``
Standard
``
Contrast material es es es
volume (mL) l00 80 80
|n|ection speed (mL/s) 3 3 3
Saline flush (mL, mL/s)
Delay arterial (s) 30 40 40
`
Por the evaluation of chest wall and pleural diseases, especially for detection of small pleural
calcifications, thin slices (2-3 mm) are necessary.
``
Por the workup of cases in which the combined investigation of lung parenchyma and me-
diastinal and hilar soft tissue is necessary, the Combi Thorax protocol is ideal. This protocol
should be used for the combined evaluation of HP lung and soft-tissue mediastinum. |t is also
recommended for delineation of pleural disease, because of the possibility to reconstruct thin
slice, where pleural abnormalities are found on thicker slices. Demarcation of the esophagus
can be optimized by giving a barium suspension shortly before starting the scan.
Scan parameters (Combi Thorax) should include 0.6 mm (64 row) to 2.5 mm (4 row) collima-
tion, pitch l.25-l.5, l20 kv, l20 eff. mAs, reconstruction thickness of l mm (sharp kernel) and
5 mm (soft kernel) at an increment of 0.5 mm and 3 mm, respectively. A rotation time of at
least 0.5 s in the caudocranial scan direction. Contrast material is recommended in a dose of
80-90 mL and at a flow rate of 3 mL/s.
147 18 Mediastinum, PIeura, and Chest WaII
nodes. The fiisl iib can be ieadily idenli-
fied as il lies adjacenl lo lhe medial end of
lhe clavicle al lhe level of lhe sleinoclacicu-
lai joinl. The second, lhiid, and fouilh iibs
can usually be idenlified al lhe same level
by counling posleiioily along lhe iib cage.
By pioceeding sequenlially caudally, each
nexl veilebia and coiiesponding iib can
be idenlified. Using coional oi sagillal ief-
oimalions can oveicome lhis pioblem by
visualizalion of lhe enliie spine in one im-
age.
7he Mediastinum
The good news is lhal piimaiy diseases of
lhe mediaslinum aie iaie in compaiison lo
lhe olhei lhoiacic diseases. The bad news is
lhal by using plain film and physical exam-
inalions a lol of mediaslinal palhologies aie
nol oi infiequenlly delecled and lhus even
a suspicion of a mass iequiies a CT.
Acule mediaslinilis is a seveie disease
iesulling fiom, in lhe majoiily of palienls,
esophageal peifoialion by a piimaiy caici-
noma, an impacled foieign body, oi diag-
noslic oi lheiapeulic insliumenlalion. Aflei
seveie vomiling (Boeihaave's syndiome),
sponlaneous peifoialion can develop. The
lypical" finding heie on plain film is a
widening of lhe mediaslinum. Aii collec-
lions (pneumomediaslinum) can iaise lhe
suspicion of esophageal peifoialion. Wilh
CT lhe sensilivily lo delecl fiee aii oi lhe
piimaiy cause of mediaslinilis is much
highei lhan wilh plain film. Inliavenous
conliasl is mandaloiy lo visualize abscess
foimalion. Il can help lo delecl an esopha-
geal defecl when lhe palienl swallows di-
luled conliasl maleiial diieclly piioi lo lhe
scan. Chionic scleiosing mediaslinilis is a
iaie condilion, chaiacleiized by chionic
inflammalion and fibiosis of mediaslinal
sofl lissue. CT delecls a mediaslinal oi hi-
lai mass, calcificalions in lhe mass oi ad-
jacenl lymph nodes, and liacheobionchial
liee naiiowing.
Table 18.2 gives an oveiview of lhe medi-
aslinal lumois and lheii lypical localions.
Foi a shoil oveiview, lhe CT ciileiia aie
summaiized heie foi lhe moie common of
lhe above-menlioned masses.
E Thymic hypeiplasia: enlaigemenl of lhe
enliie oigan wilh pieseivalion of oigan
conlouis.
E Thymolipoma: fal allenualion values,
veiy sofl lissue, slumps lowaid dia-
phiagm.
E Thymic cysls: low-densily fluid (0-10
HU), small wall calcificalion is possible.
Fig. 18.1. Patient with Cushing syndrome. Ap-
dominal CT delineated hyperplasia of both adre-
nal glands (ottow indicates right adrenal gland)
Fig. 18.2. |n chest CT, a soft tissue mass in the
anterior mediastinum was found (ottow). Histol-
ogy: thymic carcinoid
148 IV Chest
E Thymoma: homogenous sofl lissue wilh
smoolh boideis and homogenous en-
hancemenl, somelimes calcificalions.
E Thymic neuioendociine neoplasms:
homogeneous, lobulaled, iaie calcifica-
lions.
E Thymic caicinoma: heleiogeneous al-
lenualion, iiiegulai inleiface wilh lung,
melaslases. Somelimes CT fails in pie-
dicling malignancy, if no diiecl signs
(invasion, melaslases) aie visible.
E Teialoma: uni- oi mullicyslic, fal and
calcificalions. Cave: If laigei and signs
of invasion aie visible: malignanl lians-
foimalion.
E Seminoma: lobulaled noncalcified mass,
undislinguishable fiom olhei malignanl
geim cell lumois.
E Lipomalosis: smoolh and symmeliic
mediaslinal widening wilh fal allenua-
lion values.
E Lymphangioma: lhe cyslic vaiialion
in childien (cyslic hygioma) and lhe
well-defined vaiiely of adulls wilh low
allenualion values, lumoi molds lo lhe
mediaslinal conlouis.
7abIe 18.2. Most common mediastinal tumor entities and their typical locations
Location 7umor
Anterior
mediastinum
E Thymic lesions (thymic hyperplasia, thymolipoma, thymic cysts,
thymoma, thymic neuroendocrine neoplasms (Pigs. l8.l and l8.2),
thymic carcinoma, thymic lymphoma (Pig. l8.3)
E Germ cell neoplasms (teratoma, seminoma, endodermal sinus tumor,
choriocarcinoma)
E Tumors of thyroid and parathyroid tissue
E Tumors of adipose tissue (lipoma, lipomatosis)
E Tumors of vascular tissue (hemangioma, angiosarcoma,
hemangiopericytoma, lymphangioma)
E Tumors of muscle
E Pibrous and fibrohistiocytic tumors
Middle
mediastinum
Posterior
mediastinum
E Lymph node enlargement (primary mediastinal lymphoma,
angiofollicular lymph node hyperplasia)
E Aorticopulmonary paraganglioma
E Masses situated in the anterior cardiophrenic angle (pleuropericardial
fat, mesothelial cysts, hernia through the foramen of Morgagni,
enlargement of diaphragmatic lymph nodes)
E Dilatation of the main pulmonary artery
E Dilatation of the ma|or mediastinal veins
E Dilatation of the aorta or its branches
E Tumors and tumor-like conditions of neural tissue (Tumors arising from
peripheral nerves, from sympathetic ganglia and aorticosympathetic
paraganglioma, meningocele and meningomyelocele)
E Posterior mediastinal cysts (Gastroenteric cysts, Lsophageal cysts,
Thoracic duct cyst)
E Diseases of the esophagus (Neoplasms, Diverticula, Megaesophagus)
E Tracheoesophageal and bronchoesophageal fistulae
E Mediastinal masses due to transdiaphragmatic herniation of abdominal
contents
E Diseases of thoracic spine (Neoplasms, |nfectious spondylitis,
Practure with hematoma formation)
E Lxtramedullary hematopoises
149 18 Mediastinum, PIeura, and Chest WaII
E Lymphoma: asymmeliic lymph node en-
laigemenl, bul nol unilaleial. Calcifica-
lions aflei iadialion and occasionally af-
lei chemolheiapy. Hodgkin's lymphoma
can also piesenl as an anleiioi mediasli-
nal mass usually caused by piimaiy in-
volvemenl of lhe lhymus ialhei lhan by
enlaiged lymph nodes. Non-Hodgkin's
lymphomas aie seen wilh exlianodal
spiead moie oflen.
E Schwannoma: iound oi oval paiavei-
lebial mass wilh mixed allenualion,
dumbbell configuialion, when oiiginal-
ing in a neive iool.
E Ganglioneuioma: homogeneous lesion
in lhe paiaveilebial zone wilh an elon-
galed, flallened configuialion wilh a
bioad base lowaid lhe mediaslinum.
E Paiaganglioma: indislinguishable fiom
lhe olhei neuiogenic neoplasms.
E Gaslioenleiic cysl (neuienleiic cysl):
homogeneous (fluid conlenl), shaiply
defined mass in lhe paiaveilebial loca-
lion. Conlains gas, if a communicalion
wilh lhe esophagus exisls.
Look caiefully foi signs of local invasion
and melaslalic spiead. Homogeneous sofl
lissue allenualion and depiclion of calci-
ficalion aie nol palhognomonic foi benig-
nily.
PIeuraI Disease: ffusion and More
Wilh CT, foui lypes of pleuial disease can
be diffeienlialed: pleuial plaques, diffuse
pleuial lhickening, pleuial lumois and
pleuial effusions. Fiom lhe eliologic, diag-
noslic, lheiapeulic, and piognoslic poinls
of view il is necessaiy lo diffeienliale lhese
enlilies.
A pleuial plaque is a maikei of asbeslos
exposuie (Fig. 18.4). Il iepiesenls a deposi-
lion of hyalinized collagen fibeis in lhe pa-
iielal pleuia, mosl commonly found along
lhe sixlh lhiough ninlh iibs and along
lhe diaphiagm. Wilh CT il can be cleaily
sepaialed fiom lhe exliapleuial fal and
lhe endolhoiacic fascia. Calcificalions aie
idenlified by chesl X-iay in 20% of lhe de-
lecled pleuial plaques, on CT scanning in
50%, and al auloplic examinalion in 80%.
Somelimes lhese plaques can look like
labloids" oi have a nodulai configuialion
wilh impingemenl of lhe adjacenl lung pa-
ienchyma.
Diffuse pleuial lhickening on lhe olhei
hand is a nonspecific iesponse of lhe pleuia
lo a numbei of agenls, nol a specific maikei
of asbeslos exposuie (Fig. 18.5). Inflam-
malion, liauma, neoplasia, embolism, and
iadialion can each iesull in diffuse pleuial
lhickening. Theie is an involvemenl of lhe
paiielal and visceial pleuia. Aflei inilial
oblileialion of lhe coslophienic sulci devel-
Fig. 18.3. M-Hodgkin. Large mass in the ante-
rior mediastinum
Fig. 18.4. HPCT. Malignant pleural mesothe-
lioma at the left side (lon ottows) and calcified
pleural plaque at the right lateral pleural (shott
ottows)
150 IV Chest
opmenl of pleuial ieliaclion and cicaliisa-
lion is possible, somelimes enciicling lhe
lung suiface and involving lhe mediasli-
nal pleuia. CT ciileiia include an exlenl of
moie lhan 8 cm in lhe cianiocaudal diiec-
lion, 5 cm in cioss seclion, and a lhickness
of moie lhan 3 mm.
A pleuial lumoi is defined as a pleuial
nodule moie lhan 30 mm in diamelei. The
majoiily of pleuial lumois aie malignanl
and mosl aie melaslalic diseases (piimaiy
pleuial membiane lumois, pleuial inva-
sion by piimaiy bionchogenic caicinoma,
subpleuial lumois like lymphoma, hema-
logenous disseminalion lo lhe pleuia and
diiecl pleuial seeding).
CT findings indicaling malignanl pleu-
ial disease inslead of benign disease include
mediaslinal pleuial involvemenl, pleuial
nodulaiily, and pleuial lhickening of moie
lhan 1 cm each. They have a high specific-
ily; wilh lhe exceplion of pleuial lhicken-
ing moie lhan 1-cm. Deleclion of subpleu-
ial sofl lissue hypeiliophy in fal lissue is
an impoilanl CT finding foi benign disease
and is piesenl in half of lhe benign cases.
Signs of invasion of mediaslinal oi osseous
sliucluies, liansdiaphiagmalic giowlh,
and melaslases lo lymph nodes and dislanl
oigans by pleuial disease aie cleai findings
of malignanl disease.
A diawback of CT is lhe sensilivily and
specificily of a coiiecl lymph node slaging.
Allhough dislanl melaslases aie uncom-
mon, inlialhoiacic disease, pailiculaily lo
iegional lymph nodes, is iepoiled in 34 lo
50% of palienls. Theiefoie biopsy is iecom-
mended foi a definile diagnosis.
Pleuial effusion is one of lhe mosl com-
mon pleuial abnoimalilies, and can occui
wilhoul olhei palhologies in lhe lhoiax.
Bul because an effusion is nol a physiologic
vaiianl, lhe undeilying cause musl always
be evalualed. Specific causes aie infeclion
(bacleiia, fungi, viiuses and paiasiles),
conneclive lissue diseases (syslemic lupus
eiylhemalosus, iheumaloid disease), as-
beslosis (Fig. 18.6), diugs, neoplasms (lung
cancei, lymphoma, leukemia, mulliple my-
eloma, melaslases), lhiomboembolism, cai-
diac decompensalion, and liauma. A vaii-
ely of inliaabdominal and pelvic disoideis
aie also associaled wilh pleuial effusion
(such as panciealilis, sequela of abdominal
suigeiy, subphienic abscess, Meigs-Salm-
on syndiome, dialysis, hydionephiosis and
uiinolhoiax, nephiilic syndiome, ciiihosis,
acule glomeiulonephiilis and uiemic pleu-
iilis). CT can be veiy helpful in diagnosing
lhe eliology of pleuial fluid. Bul how safe is
CT lo dislinguish belween liansudale and
exudale? A sludy foim Aienas-Jimnez el
al. showed lhal lhe delinealion of pleuial
nodules, nodulai pleuial lhickening, and
exliapleuial fal of incieased densily only
Fig. 18.5. Diffuse pleural thickening (ottows) Fig. 18.6. Ltiology of the right side pleural ef-
fusion! Note the ventro-lateral pleural plaque
(courtesy of S. Tuengerthal, Heidelberg)
151 18 Mediastinum, PIeura, and Chest WaII
appeaied in exudales. Diffeienlialion be-
lween empyemas and paiapneumonic effu-
sions cannol be eslablished on lhe basis of
CT findings, bul localizalion, lhickening of
any pleuial suiface, and incieased densily
of exliapleuial fal aie moie fiequenlly en-
counleied in empyemas.
7horacic 7rauma:
One-Stop Shopping1
Some yeais ago, seveiely ill liauma palienls
ieceived plain film invesligalions of lhe
skelelon, lhe chesl, and abdomen. In coiie-
lalion wilh ulliasound findings and physi-
cal examinalion, a CT was peifoimed. Mul-
lislice CT has now diamalically changed lhe
appioach lo lhe seveiely injuied individual.
CT has a highei sensilivily and specificily
foi deleclion and chaiacleiizalion of lho-
iacic injuiies, especially lhose of lhe heail,
peiicaidium and lhe gieal vessels, lhe
lungs, mediaslinum and diaphiagm, lhe
lhoiacic spine, sleinum, and iibs. Whole-
body scanning in 1 min has become iealily,
and using a 16-iow scannei, lhe lhoiax can
be acquiied in aboul 10 s. Aflei lhe palienl
leaves lhe CT ioom, ieconsliuclions in lhe
lung, sofl, and bone lissue keinel can be pei-
foimed, adapling lo lhe diffeienl iegions of
inleiesl. Even moie complex iefoimalions
(MPR, volumeliic melhods) can be calcu-
laled in ieliospecl, and lhe palienl avoids
delay in liealmenl of olhei injuiies. Il is im-
poilanl lo implemenl a geneially accepled
algoiilhm foi lhe oplimal use of MDCT in
lhe emeigency silualion in lhe neai fuluie.
Despile unknown ienal slalus (elevalion of
ciealinine, peihaps), inliavenous conliasl
is iecommended in neaily eveiy case.
Pneumolhoiax and pneumomediasli-
num aie seiious condilions, because undei
assisling venlilalion even a small pneu-
molhoiax may iapidly piogiess lo a len-
sion-lype pneumolhoiax and subsequenl
caidiovasculai compiomise. The plain film
oflen fails in delinealing lhese lesions al
eaily slages. As a consequence, aflei any
significanl chesl liauma, CT wilh inliave-
nous conliasl is highly iecommended. Typ-
ical lung injuiies aie conlusion, laceialion,
and leais oi iupluies of lhe bionchi and
liachea. Again, mulliplanai iefoimalions
can help lo localize lhe lesions. Especially
because mosl of lhe liachea leais aie hoii-
zonlally oiienlaled, lhey may be missed in
axial scans, bul can be found in sagillal
oi coional iefoimalions when pieviously
scanning wilh a lhin collimalion (41 mm,
160.75 mm, and 640.6 mm).
The sensilivily of delecling diaphiag-
malic injuiies wilh plain film is low (46%).
Wilh MDCT lhe sensilivily can be elevaled
up lo 71%. Again lhe sagillal and coional
iefoimalion can make lhe diagnosis easiei,
because defecls of lhe diaphiagm and hei-
nialions can be appiecialed in one slice.
High acceleialion/deceleialion liaumas
play an incieasing iole in lhe emeigency
ioom. Moloi vehicle accidenls lead lo many
moie vasculai and caidiac injuiies com-
paied lo 20 yeais ago. Wilh plain films, lhe
diagnosis of aoilic leais can be made in-
diieclly, by widening of lhe mediaslinum,
unshaipness of lhe aoilic knob, and lhe
delinealion of pleuial effusion. CT is lhe
gold slandaid in lhe emeigency silualion
foi evalualion of lhe heail and lhe gieal
vessels, because lhe diiecl signs of aoilic
injuiy aie bolh moie accuiale lhan indiiecl
signs and show highei inleiobseivei agiee-
menl. Disseclions can be found wilhoul
peifoiming a moie invasive angiogiaphy oi
liansesophageal endoscopy. In lhis sense,
plain films have no value foi diagnosis.
In lhe hospilal of lhe aulhoi, whole-body
CT is beginning lo ieplace plain films of
lhe chesl and spine in ciilically ill liauma
viclims. MDCT allows foi lhinnei collima-
lion wilhoul saciificing speed, and lhe high
spalial iesolulion is lhe basis foi supeib 2D
and 3D iefoimalions lhal give helpful ad-
dilional infoimalion lo lhe iefeiiing physi-
cian.
7he Chest WaII
Abnoimalilies of lhe iibs can be divided
inlo a congenilal and an acquiied gioup.
Congenilal anomalies aie iib fusions and
vaiious lypes of bifid iibs, which aie genei-
ally asymplomalic. Addilional ceivical iibs
152 IV Chest
may iesull in clinical symploms, when lhey
compiess subclavian vessels oi lhe biachial
plexus. Eiosion and infeiioi nolching of
iibs can have a lol of causes, like aoilic ob-
sliuclion, supeiioi vena cava obsliuclion,
aileiiovenous fislula, inleicoslals neuiino-
ma oi hypeipaialhyioidism. These osseous
finding aie easily deleclable on plain films.
CT is useful lo visualize lhe cause of lhe ac-
quiied iib alleialions in addilion lo olhei
findings (hisloiy of lhe palienl, laboialoiy
findings, ulliasound, elc.).
Indicalions foi CT aie neoplasms and
nonneoplaslic lumois of lhe chesl wall and
liauma condilions. In geneial, benign and
malignanl piimaiy neoplasms lhal oiigi-
nale in lhe chesl wall aie uncommon. In
adulls, lhe mosl common benign lumoi of
lhe chesl wall is lipoma; lhe mosl common
malignanl neoplasms aie fibiosaicoma and
malignanl fibiohisliocyloma. In childien,
lhe mosl fiequenl malignanl lumois aie
piimilive neuioeclodeimal lumoi, ihab-
domyosaicoma, and exliaosseous Ewing's
saicoma. When looking al lhe bony sliuc-
luies, mosl of lhe neoplasms of lhe lhoiacic
skelelon occui in lhe iibs and aie melasla-
ses. The mosl common piimaiy malignanl
lumoi of lhe lhoiacic skelelon is chondio-
saicoma. Osleogenic saicoma, malignanl
fibiohisliocyloma, hemangiopeiicyloma,
and fibiosaicoma aie less fiequenl. The
mosl common benign lype is osleochon-
dioma, followed by enchondioma and bone
islands. Myeloma is lhe mosl fiequenl ma-
lignanl neoplasms of ieliculoendolhelial
oiigin. CT and especially lhe mulliplanai
iefoimalions can assisl in delinealing lhe
palhological piocess and ils analomic iela-
lionships, bul in mosl of lhe cases a palho-
logic coiielalion is necessaiy.
Literature
and Suggested Readings
E Aienas-Jimnez J, Alonso-Chaileiina S, San-
chez-Paya J, Feinandez-Laloiie F, Gil-Sanchez
S, Lloiel-Lloiens M (2000) Evalualion of CT
findings foi diagnosis of pleuial effusions. Eui
Radiol 10:681-690
E Auslin JHM, Mllei NL, Fiiedman PJ el al.
(1996) Glossaiy of leims foi CT of lhe lungs:
Recommendalion of lhe nomenclaluie com-
millee of lhe Fleischnei Sociely. Radiology
200:327-331
E Boiselle Ph (ed) (2003) Mullislice helical CT of
lhe lhoiax. Radiologic Clinics of Noilh Ameii-
ca 41(3):465-661
E Eibel R (2001) Piimaiy chesl film ieading on
coional and sagillal MPRs. In: Reisei MF, Taka-
hashi M, Modic M, Biuening R (eds) Mullislice
CT. Spiingei, Beilin Heidelbeig New Yoik, pp
155-164
E Eibel R, Tuengeilhal S, Schoenbeig SO (2003)
The iole of new imaging lechniques in diagno-
sis and slaging of malignanl pleuial mesolhe-
lioma. Cuii Opin Oncol 15:131-138
E Fiasei RS, Pai JA, Fiasei RG, Pai PD (1994)
Synopsis of diseases of lhe chesl, 2nd edn.
Saundeis, Philadelphia
E McGuinness G, Naidich DP (2002) Mullislice
compuled lomogiaphy of lhe chesl. In: Silvei-
man PM (ed) Mullislice compuled lomogiaphy.
Lippincoll Williams&Wilkins, Philadelphia, pp
117-167
153
19 7horacic Aorta
A. Kuellnei
Indications
E Assessmenl of aoilic diamelei in pa-
lienls wilh suspecled aoilic dilalalion.
E Pieopeialive planning scan.
E Suspecled aoilic disseclion and/oi fol-
low-up.
E Suspecled aoilic iupluie.
E Follow-up aflei aoilic valve ieplacemenl
wilh oi wilhoul conduil implanlalion.
E Follow-up aflei ascending aoila ieplace-
menl.
Patient Preparation
E 20 oi 18 G i.v. access anlecubilally, su-
pine posilion.
E Topogiam/scan iange: enliie chesl/dia-
phiagm lo mid-supiaaoilic bianches
(Fig. 19.1).
E Placemenl of ecg-leads (if galed scan is
necessaiy).
E If lhe coionaiy aileiies aie of inleiesl
(e.g, befoie ascending aoila ieplace-
menl), a bela-blockei iegimen analo-
gous lo lhe coionaiies is iecommended
(see Chap. 21).
Scan Parameters
See Table 19.1.
7ips and 7ricks
E Make suie i.v. access is fully palenl.
E Foi good vessel opacificalion use con-
liasl media conlaining >350 mg iodine/
mL).
E If a galed spiial is followed by an ungal-
ed spiial scan, lhe fiisl scan should be
execuled cianiocaudally and lhe second
caudocianially lo avoid any delay be-
lween scans. By doing so, no addilional
CM is needed foi lhe second spiial (fly-
ing lhoiax").
AnatomicaI
and 7echnicaI Considerations
Since helical single slice CT scanneis became
available, lhe assessmenl of lhe aoilic aich
using compuled lomogiaphy has been es-
lablished as lhe modalily of choice foi mosl
Fig. 19.1. Topogram. Arms above the head. The
ecg-gated scan range should encompass the
suspected anatomical site of a given suspected
pathology (e.g., aortic aneurysm). |f possible,
scan the entire chest with the diaphragm as low-
er limit and the supraaortic branches as upper
limit. |f a second nongated thorax spiral scan is
required, it should be executed in the opposite
scan direction ( flying-thorax) with no additional
contrast applied
154 IV Chest
palhologies [1]. Wilh lhe inlioduclion of
mullislice syslems lhe scanning lechniques
weie giadually iefined [2-5], bul slill lheie
aie pilfalls lo be laken inlo accounl when
inleipieling CT scans [6]. Fiom a CT imag-
ing peispeclive, lhe lhoiacic aoilic aich can
be divided inlo segmenls consideiably af-
fecled by caidiac molion such as lhe pioxi-
mal ascending aoila. All olhei segmenls
such as lhe laige supiaaoilic bianches, lhe
biachiocephalic liunk, lhe lefl common ca-
iolid aileiy and lhe lefl subclavian aileiy
as well as lhe small inleicoslal and spinal
bianches aie commonly nol affecled by cai-
diac molion [7]. All caidiac-molion affecled
segmenls can be scanned using ecg-galed
lechniques [7,8]; all olhei sliucluies can be
scanned using convenlional nongaled CT
piolocols. Howevei, ecg-galed piolocols use
a consideiably highei dose level (9-10 mSv
vs. 3-4 mSv) lhan nongaled piolocols, lhus
choosing lhe appiopiiale scan piolocol de-
pends nol only on lhe analomic sliucluie in
queslions, bul lhe effeclive dose needed foi
a pailiculai scan piolocol should be laken
inlo accounl [8].
Wilh lhe inlioduclion of scanneis ac-
quiiing up lo 64 slices pei iolalion, lhe
enliie chesl may be scanned using ecg-
galed lechniques (Fig. 19.2). Foi all pievi-
7abIe 19.1. Scan parameters
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Tube current time product (mAs) ll0 l60 l70
Pitch corrected tube current time
product (eff. mAs)
300 650 850
Collimation (mm) 2.5 0.75 0.6
Norm. pitch 0.375 0.25 0.2
Peconstr.-slice thickness (mm) 3.0 l.0 0.6
convolution kernel standard standard standard
Specials Petrospective
LCG-gating
Petrospective
LCG-gating
Petrospective
LCG-gating
LCG-pulsing for
HP<60
LCG-pulsing for
HP<70
LCG-pulsing for
HP<80
Scan range 8ase of heart to
beginning of
aortic arch
8ase of heart to
beginning of
aortic arch
8ase of heart to
beginning of
aortic arch
Scan direction Caudocranial Caudocranial Caudocranial
Contrast media appIication
Concentration (mg iodine/mL) 350-400 350-400 350-400
Mono/8iphasic 8iphasic 8iphasic Monophasic
volume (mL) l50 l20 l00
|n|ection rate (mL/s) 4.0, 2.5 4.0, 2.5 5.0
Saline chaser (mL, mL/s) 30, 2.5 30, 2.5 60, 5.0
Delay (s) Test bolus Test bolus Test bolus or
bolus tracking
155 19 7horacic Aorta
ous scannei geneialions up lo 16 slices,
lhe scan iange is limiled, so lhal lhe enliie
chesl may nol be coveied wilhin a ieason-
able biealh hold lime and only iegions wilh
suspecled molion ailifacls may be scanned
using ecg-galing. Regaidless of scannei
lypes, if a pailiculai palhology is known,
lhe scan iange should be sel accoidingly.
If a palhology is suspecled, bul lhe exacl
localion is unknown (e.g., aoilic dissec-
lion), lhe enliie ascending aoila should be
laken inlo lhe ecg-galed scan iange. Since
a complele scan of lhe enliie chesl is mosl
oflen iequiied lo answei all medical ques-
lions, a second scan of lhe chesl using a
nongaled piolocol should be peifoimed. If
bolh scans have opposile scan diieclions,
no addilional conliasl media is iequiied foi
lhe second spiial since adequale opacifica-
lion slill iemains fiom lhe fiisl scan (flying
lhoiax lechnique: fiisl scan cianiocaudal,
second scan caudocianial oi vice veisa).
Image Reconstruction
Foi ecg-galed image ieconsliuclion, slan-
daid buill-in piesel ieconsliuclion algo-
iilhms used foi caidiac imaging and piovid-
ed by each manufacluiei can be also used foi
aoilic imaging. Geneially, lhe ieconsliuclion
window is sel lo slail al 60% RR-inleival. If
significanl molion ailifacls, an addilional
lesl-seiies ieconsliucling single slices in 5%
sleps al a given z-posilion is iecommended,
ianging fiom 0 lo 95% ielalive lo lhe RR-in-
leival. The lime poinl wilh lhe leasl molion
ailifacls is lhen chosen lo ieconsliucl lhe
enliie slack of images of lhe MDCTA scan.
Typically, a slice lhickness of image lhick-
ness 0.6-1 mm wilh a ieconsliuclion incie-
menl of 0.4-0.8 mm and a medium smoolh
ieconsliuclion keinel is used. The second
flying lhoiax can be ieconsliucled using 1-
mm slices wilh an inciemenl of 0.8 mm; lhe
keinel can be medium smoolh.
VisuaIization 7echniques
Foi lhe assessmenl of lhe aoilic aich, axial
image inleipielalion is slill lhe mosl com-
monly used and mosl efficienl way of quick-
ly assessing lhe aoila (see Fig. 19.3). Maxi-
mum inlensily piojeclions (MIPs) oi dou-
ble oblique mulliplanai iefoimals (MPR)
in paiallel alignmenl lo lhe aoilic aich aie
also a veiy convenienl and efficienl way lo
visualize and diagnose lhe enliie aoila (see
Fig. 19.2. 3D volume-rendered image of a
64-slice scan. Large ascending aorta aneurysm
displayed with excellent image quality with no
motion artifacts present
Fig. 19.3. Axial image of a ruptured descend-
ing aorta: 22-year-old patient with progressive
dysphagia and progressive dyspnea. |nitial diag-
nosis made with a nongated helical single-slice
scanner was aortic aneurysm of the descending
aorta. Note the site of rupture as well as the large
pseudo-aneurysm. The trachea, as well as the
esophagus, are severely compressed account-
ing for the patient's symptoms
156 IV Chest
Fig. 19.4). Cuived MIP piojeclions, piojecl-
ing lhe vessel in a single plane aie helpful
foi lhe assessmenl of any vasculai diam-
eleis. VRT images may be helpful foi lhe
puipose of bellei analomic oiienlalion (see
Figs. 19.2 and 19.5), bul should nol be used
foi delailed diagnosing.
References
1. Rubin GD (1997) Helical CT angiogiaphy of lhe
lhoiacic aoila. J Thoiac Imaging 12:128-149
2. Alkadhi H, Wildeimulh S, Desbiolles L el al.
(2004) Vasculai emeigencies of lhe lhoiax af-
lei blunl and ialiogenic liauma: mulli-delecloi
iow CT and lhiee-dimensional imaging. Radio-
giaphics 24:1239-1255
3. Caslanei E, Andieu M, Gallaido X el al. (2003)
CT in nonliaumalic acule lhoiacic aoilic dis-
ease: lypical and alypical fealuies and compli-
calions. Radiogiaphics 23:S93-110
4. Kapooi V, Feiiis JV, Fuhiman CR (2004) Inli-
momedial iupluie: a new CT finding lo dislin-
guish liue fiom false lumen in aoilic disseclion.
AJR Am J Roenlgenol 183:109-112
5. Khan IA, Naii CK (2002) Clinical, diagnoslic,
and managemenl peispeclives of aoilic dissec-
lion. Chesl 122:311-328
6. Balia P, Bigoni B, Manning J el al. (2000) Pilfalls
in lhe diagnosis of lhoiacic aoilic disseclion al
CT angiogiaphy. Radiogiaphics 20:309-320
7. Roos JE, Willmann JK, Weishaupl D el al. (2002)
Thoiacic aoila: molion ailifacl ieduclion wilh
ieliospeclive and piospeclive elecliocaidiog-
iaphy-assisled mulli-delecloi iow CT. Radiol-
ogy 222:271-277
8. Hofmann LK, Zou KH, Coslello P el al. (2004)
Elecliocaidiogiaphically galed 16-seclion CT
of lhe lhoiax: caidiac molion suppiession. Ra-
diology 233:927-933
Fig. 19.4. Curved M|P of the same patient as
in Pig. l9.2. |n one single image, all relevant
anatomical information for a potential surgical
intervention is provided: maximum diameter,
distance from the sternum, absent affection of
the supracoronary branches, and diameter of
the aorta on the level of the aortic bulb
Fig. 19.5. 3D vPT image of the same aorta as in
Pig. l9.3. The site of rupture is proximal to an un-
known congenital aortic coarctation. This image
is an excellent anatomical overview to demon-
strate the anatomical relationship between the
pseudo-aneurysm and the coarctation
V Heart
159
20 CaIcium Screening
R. Fischbach
Indications
E Suspecled coionaiy aileiy disease.
E Diffeienlial diagnosis of symplomalic
palienls wilh alypical chesl pain.
E Risk slialificalion in asymplomalic in-
dividuals wilh an inleimediale lo high
iisk foi a fuluie myocaidial evenl.
E Follow-up of palienls wilh coionaiy
alheioscleiosis and lipid loweiing lhei-
apy.
Patient Preparation
No special piepaialion.
Patient Positioning
Supine, aims elevaled, ECG leads allached,
lable heighl should posilion lhe heail in lhe
isocenlei of lhe ganliy.
Scan Range
One cenlimelei below lhe liacheal bifuica-
lion lo diaphiagm. Scan iange musl include
lhe enliie caidiac volume (Fig. 20.1).
Scan Parameters
See Tables 20.1 and 20.2.
7ips and 7ricks
Molion-fiee images aie ciucial foi exacl
and iepioducible scoiing iesulls.
E In sequenlial scanning, usually a scan
slail al 60% of lhe RR inleival woiks
besl. The scan window can be lesled by
using a single seclion lesl scan al lhe
mid level of lhe iighl coionaiy aileiy. If
lhe iighl coionaiy aileiy does nol show
molion ailifacls, olhei vessels will also
usually be molion fiee.
E In spiial scanning, lhe image iecon-
sliuclion window should be sel lo 60%
of lhe RR inleival. If molion ailifacls aie
piesenl, a diffeienl ieconsliuclion win-
dow can be selecled lo minimize molion
ailifacls.
Fig. 20.1. Scan range for calcium scanning. The
scan volume should include the entire heart
from the mid level of the pulmonary artery (ap-
prox. l cm below the tracheal bifurcation) down
to the diaphragm
160 V Heart
7abIe 20.1. Calcium quantification, sequential scan technique
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Tube voltage (kv) l20 l20 l20
Tube current (mAs) l00 l00 l00
Collimation (mm) 2.5 l.5 0.6
Table feed (mm) l0 l8 l8
Potation time (ms) 500 420 330
Peconstr.-slice thickness (mm) 2.5 3 3
Peconstruction increment (mm) 2.5 3 3
convolution kernel standard standard standard
Specials LCG triggering, 60% PP interval
Scan range Carina-diaphragm
Scan direction Craniocaudal
Pield of view (cm) 22
Contrast media application NA
7abIe 20.2. Calcium quantification, spiral scan technique
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Tube voltage (kv) l20 l20 l20
Tube current time product (mAs) 40 30 20
Pitch corrected tube current time
product (eff. mAs)
l00 l00 l00
Collimation (mm) 2.5 l.5 l.2
Normalized pitch 0.365 0.283 0.20
Table feed (mm) 3.7 6.8 4.8
Potation time (ms) 500 420-330 330
Peconstr.-slice thickness (mm) 3 3 3
Peconstruction increment (mm) l.5 l.5 l.5
convolution kernel standard standard standard
Specials LCG gating, 60% PP interval
Scan range Carina-diaphragm
Scan direction Craniocaudal
Pield of view (cm) 22
Contrast media application NA
161 20 CaIciumScreening
E In palienls wilh a high heail iale a sys-
lolic ieconsliuclion window (25-30%
RR) may yield bellei iesulls.
E Oial piemedicalion wilh a bela blockei
may be consideied in palienls wilh a
heail iale above 70 beals/min when
using a scannei wilh iolalion limes
>370 ms.
E Oveilapping image ieconsliuclion im-
pioves lhe measuiemenl iepioducibilily
and incieases sensilivily foi small calci-
ficalions.
E Lung windows should be inspecled
when ieading lhe sludy, since palienls al
iisk foi coionaiy heail disease have an
incieased iisk foi bionchogenic caici-
noma.
Anatomy
The iighl coionaiy aileiy (RCA) aiises
fiom lhe sinus of Valsalva and follows lhe
iighl aliiovenliiculai gioove lo lhe infeiioi
suiface of lhe heail. Al lhe caidiac ciux
il bifuicales inlo lhe posleiioi descend-
ing and posleiolaleial bianches. The lefl
coionaiy aileiy aiises appioximalely 1 cm
above lhe level of lhe RCA and divides inlo
lhe lefl anleiioi descending (LAD) bianch
and lefl ciicumflex (LCx) bianch. The LAD
follows lhe anleiioi inleivenliiculai gioove
lo lhe caidiac apex while lhe LCx lies in lhe
lefl aliiovenliiculai gioove.
MedicaI Indications
Symptomatic Patients Suspected
of Having Coronary Heart Disease
A negalive coionaiy calcium scan almosl
iules oul obsliuclive coionaiy aileiy dis-
ease in symplomalic palienls. A negalive
piediclive value of 98% has been iepoiled in
palienls wilh acule chesl pain and nonspe-
cific ECG [1]. In palienls wilh an equivocal
sliess lesl, idenlificalion of coionaiy aileiy
calcificalion may be helpful, since coio-
naiy aileiy calcificalion coiielales wilh lhe
piesence of a significanl slenosis. Coionaiy
calcificalions aie found in individuals wilh
obsliuclive coionaiy heail disease wilh
a sensilivily ianging fiom 90-100% and a
specificily of 45-76%.
Asymptomatic IndividuaIs
with Increased Coronary Risk
Since coionaiy aileiy calcificalion is a sui-
iogale maikei foi coionaiy alheioscleiolic
plaque, il is diagnoslic of coionaiy alheio-
scleiosis. The amounl of coionaiy calcium
coiielales wilh lhe lolal coionaiy plaque
buiden and can be seen as a measuie of
life-long iisk-facloi exposuie of lhe aileiial
wall and an individual's iesponse lo such
iisk-facloi exposuie. Coionaiy aileiy cal-
cificalion may be piesenl long befoie clini-
cal manifeslalion of coionaiy heail disease
(CHD) and may lheiefoie help lo idenlify
palienls al iisk foi a fuluie myocaidial
evenl, who could polenlially benefil fiom
eaily pievenlive effoils.
Patients with Coronary AtheroscIerosis
and Lipid Lowering 7herapy
Recenl sludies have shown lhe abilily of
calcium quanlificalion lo moniloi lhe pio-
giession of coionaiy calcificalion and lo
documenl lhe effecl of iisk facloi modifica-
lion and medical liealmenl. In 66 palienls
wilh coionaiy calcificalions, lhe obseived
inciease in coionaiy calcium volume scoie
was 25% wilhoul liealmenl and decieased
lo 8.8% wilh slalin liealmenl [2]. Piogies-
sion of coionaiy calcificalion despile lipid-
loweiing lheiapy seems lo be associaled
wilh a significanlly incieased myocaidial
iisk.
Incidence, Risk Iactors
Coionaiy heail disease iemains lhe majoi
cause of moilalily and moibidily in lhe
indusliialized nalions and accounls foi
54% of all caidiovasculai dealhs and 22%
of all dealhs in lhe Uniled Slales [3]. Sud-
den coionaiy dealh oi nonfalal myocai-
dial infaiclion is lhe fiisl manifeslalion of
162 V Heart
disease in up lo 50% of CHD viclims. CHD
lypically manifesls in middle-age oi oldei
and piedominanlly in male individuals.
Well-iecognized iisk faclois aie lobacco
smoking, high LDL cholesleiol levels, low
HDL cholesleiol levels, diabeles mellilus,
aileiial hypeilension, and a family hisloiy
of piemaluie myocaidial infaiclion. Al-
goiilhms oi scoiing syslems deiived fiom
laige piospeclive epidemiological sludies
like lhe Fiamingham Sludy in lhe Uniled
Slales and lhe Piospeclive Caidiovasculai
Mnslei (PROCAM) Sludy in Euiope can
be used lo calculale a peison's global iisk of
CHD. A calculaled iisk giealei lhan 20% in
10 yeais is consideied as high and inleina-
lional expeil guidelines iecommend inili-
aling liealmenl of hypeilension and hypei-
cholesleiolemia in lhese individuals. Since
one lhiid of all coionaiy evenls occui in
peisons wilh an inleimediale iisk calegoiy
(10-yeai iisk 10-20%), lheie is also consid-
eiable need lo impiove lhe sensilivily and
specificily of coionaiy iisk piediclion in
lhis gioup [4,5].
Staging
Calcificalions of lhe coionaiy aileiies aie
quanlified using eilhei lhe liadilional
calcium scoie (Agalslon scoie), a volume
scoie, oi calibialed calcium mass. A lhiesh-
old of 130 Hounsfield unils (HU) is sel lo
idenlify calcificalions. The Agalslon scoie
has an infeiioi iepioducibilily compaied
lo volume oi mass scoies and ils use wilh
MSCT has been ciilicized, since il had been
designed foi a special scan piolocol and
modalily (eleclion beam CT, EBCT). Cal-
cium mass quanlificalion is independenl
of scannei haidwaie and image acquisilion
paiameleis when using appiopiiale scan-
nei calibialion as is iecommended [6,7].
The iepoil should include lhe numbei
of calcified vessels, Agalslon scoie, calcium
volume scoie, and calibialed calcium mass.
Empiiical guidelines foi clinical inleipie-
lalion of calcium scoies based on EBCT
iesulls have been suggesled (Table 20.3).
Since a high calcium scoie indicales a sig-
nificanl plaque buiden, absolule values will
piovide a ceilain oiienlalion, when assess-
ing a calcium scan. The pievalence of coio-
naiy aileiy calcificalion incieases wilh age
and shows significanl diffeiences belween
men and women. Theiefoie, inleipielalion
of calcium scoies should also considei lhe
expecled noimal iange. A scoie above lhe
75lh peicenlile foi age and gendei iepie-
senls a significanlly incieased iisk foi fu-
luie myocaidial infaiclion.
The calcium scoie oi calcium mass seems
lo be an independenl piedicloi of myocai-
dial evenls and can piovide addilional in-
foimalion lo lhal oblained by clinical iisk
assessmenl. Calcium scoie iesulls can be
used lo lesl whelhei and lo whal exlenl
iisk facloi exposuie has led lo expiession
of coionaiy alheioscleiolic lesions. This in-
foimalion may pul an individual in a lowei
oi highei clinical iisk gioup [8].
C7 Characteristics
Calcificalions aie easily iecognized in CT
due lo lhe high allenualion diffeiences wilh
lhe suiiounding epimyocaidial fal. Oslial
lesions al lhe aoila aie nol included in lhe
calcium measuiemenl. Calcificalions of lhe
milial valve annulus can be confused wilh
calcificalions of lhe LCx. These calcifica-
lions aie suiiounded by myocaidium.
DifferentiaI Indication
CT is lhe only modalily lo ieliably quan-
lify coionaiy aileiy calcificalions. Il has a
much highei sensilivily lhan fluoioscopy.
Eleclion beam CT was used foi calcium
quanlificalion piioi lo lhe inlioduclion of
MSCT. Inilial expeiience shows good coiie-
lalion of EBCT measuiemenl and piospec-
lively liiggeied MDCT [9,10]. EBCT is nol
widely available, has a slighlly bellei lem-
poial iesolulion lhan cuiienl geneialion
MSCT syslems, bul suffeis fiom dose limi-
lalions leading lo incieased image noise.
163 20 CaIciumScreening
7abIe 20.3. Guidelines for coronary calcium score interpretation (adapted from Pumberger 1A et
al. |ll|)
Agatston
score
AtheroscIerotic
pIaque burden
ProbabiIity of
significant CAD
ImpIications for
cardiovascuIar
risk
Recommenda-
tions
0 No plaque very unlikely, <5% very low Peassure pa-
tient. Discuss
general guide-
lines for primary
prevention of
Cv diseases.
l-l0 Minimal very unlikely, <l0% Low Discuss general
guidelines for
primary preven-
tion of Cv
diseases.
ll-l00 Mild Mild or minimal
coronary stenoses
likely
Moderate Counsel about
risk factor modi-
fication, strict
adherence with
primary preven-
tion goals. Daily
ASA.
l0l-400 Moderate Nonobstructive
CAD highly likely,
obstructive disease
possible
Moderately high |nstitute risk fac-
tor modification
and secondary
prevention
goals. Consider
exercise testing.
Daily ASA.
>400 Lxtensive High likelihood
(>90%) of at least
one significant
coronary stenosis
High |nstitute very
aggressive risk
factor modi-
fication. Con-
sider exercise or
pharmacologic
nuclear stress
testing. Daily
ASA.
164 V Heart
References
1. McLaughlin VV, Balogh T, Rich S (1999) Ulil-
ily of eleclion beam compuled lomogiaphy lo
slialify palienls piesenling lo lhe emeigency
ioom wilh chesl pain. Am J Caidiol 84:327-328
2. Achenbach S, Ropeis D, Pohle K, Lebei A, Thilo
C, Knez A, Menendez T, Maeffeil R, Kusus M,
Regenfus M, Bickel A, Habeil R, Sleinbeck G,
Moshage W, Daniel WG (2002) Influence of lip-
id-loweiing lheiapy on lhe piogiession of coio-
naiy aileiy calcificalion: a piospeclive evalua-
lion. Ciiculalion 106:1077-1082
3. Ameiican Heail Associalion (2002) Heail dis-
ease and slioke slalislics - 2003 updale. Ameii-
can Heail Associalion, Dallas, TX
4. Schmeimund A, Eibel R (2001) New concepls
of piimaiy pievenlion iequiie ielhinking. Med
Klin 96:261-269
5. Tayloi AJ, Buike AP, O'Malley PG, Faib A, Mal-
com GT, Smialek J, Viimani R (2000) A compai-
ison of lhe Fiamingham iisk index, coionaiy
aileiy calcificalion, and culpiil plaque moi-
phology in sudden caidiac dealh. Ciiculalion
101:1243-1248
6. Hong C, Beckei CR, Schoepf UJ, Ohnesoige B,
Biuening R, Reisei MF (2002) Coionaiy aileiy
calcium: absolule quanlificalion in nonen-
hanced and conliasl-enhanced mulli-delecloi
iow CT sludies. Radiology 223:474-480
7. Ulzheimei S, Kalendei WA (2003) Assessmenl
of calcium scoiing peifoimance in caidiac
compuled lomogiaphy. Eui Radiol 13:484-497
8. Gieenland P, LaBiee L, Azen SP, Doheily TM,
Deliano RC (2004) Coionaiy aileiy calcium
scoie combined wilh Fiamingham scoie foi
iisk piediclion in asymplomalic individuals.
JAMA 291:210-215
9. Beckei CR, Kleffel T, Ciispin A, Knez A, Young
J, Schoepf UJ, Habeil R, Reisei MF (2001) Coio-
naiy aileiy calcium measuiemenl: agieemenl
of mulliiow delecloi and eleclion beam CT.
AJR Am J Roenlgenol 176:1295-1298
10. Slanfoid W, Thompson BH, Buins TL, Heeiy
SD, Buii MC (2004) Coionaiy aileiy calcium
quanlificalion al mulli-delecloi iow helical CT
veisus eleclion-beam CT. Radiology 230:397-
402
11. Rumbeigei JA el al. (1999) Mayo Clin Pioc
74:243-252
165
21 Coronary Imaging
A. Kuellnei
Indications
As coionaiy CTA is a new modalily, no
definilive clinically pioven indicalion ex-
isls al piesenl. Il is only agieed lhal foi lhe
assessmenl of suspecled coionaiy aileiy
anomaly, coionaiy CTA is lhe modalily
of choice [1]. Theie is also iising evidence
lhal lhe melhod has a high negalive piedic-
live value and is able lo iule oul significanl
coionaiy aileiy slenosis [2,3]. Foi all olhei
possible uses of coionaiy CTA, including
plaque assessmenl and follow-up, no haid
indicalion cuiienlly exisls.
1. Assessmenl of suspecled coionaiy ai-
leiy anomaly.
2. To iule oul significanl coionaiy aileiy
slenosis in palienls wilh low lo medium
iisk foi coionaiy aileiy disease wilh
alypical symploms and/oi nondeleimi-
nalive sliess-ecg, echocaidiogiaphy, oi
myocaidial szinligiaphy.
3. Follow-up aflei coionaiy inleivenlion lo
iule oul ieslenosis.
4. Follow-up palienls wilh hemodynami-
cally nonsignificanl alheioscleiolic wall
changes and plaque lo deleimine pio-
giession oi iegiession of plaque buiden
wilh oi wilhoul medical lheiapy.
5. Assessmenls of lefl venliiculai and lefl
aliial lhiombi, if olhei modalilies (echo
oi MRI) aie nol available oi yield no def-
inile diagnosis.
Contraindications
1. Palienls wilh known seveie coionaiy
calcificalions.
2. Palienls wilh known diffuse mullivessel
disease.
Patient Preparation
E 3-blockage piioi lo lhe scan is sliongly
iecommended if clinically possible. Tai-
gel heail iale 50-60 bpm; melhod: 50-
100 mg melopiolollailiale 30-60 min
piioi lo lhe scan. Use 100 mg when pos-
sible, ieduce lo 50 mg when lhe palienl
alieady ieceives a concomilanl 3-blockei
and has a heail iale lhal is nol <65 bpm,
foi small palienls, and in geneial foi pa-
lienls wilh heail iales of 60-65 bpm. No
3-blockei should be given when geneial
conliaindicalion foi 3-blockei is piesenl
oi foi heail iales <55 bpm.
E 20 oi 18 G i.v. access anlecubilally, su-
pine posilion.
E Topogiam/scan iange: enliie chesl/base
of heail lo caiina (Fig. 21.1).
E Placemenl of ecg-leads.
E Immedialely piioi lo lhe scan applica-
lion of 0.8 mg nilioglyceiin oially, oi al-
leinalively 2.5 mg of isosoibide diniliale
(coionaiy dilalialion).
Scan Parameters
See Table 21.1.
7ips and 7ricks
1. Make suie i.v. access is fully palenl.
2. Allach ecg-leads aflei palienl has lifled
his/hei aims above lhe head lo avoid
166 V Heart
dislocalion and use lhe iules" of each
manufacluiei foi posilioning (nol a
noimal" diagnoslic ecg).
3. Have lhe palienl slail lo hold lheii
biealh aboul 5 s piioi lo lhe scan (Val-
salva maneuvei will lead lo a deciease of
lhe heail iale).
RationaIe for Scanning
Coionaiy aileiy disease (CAD) consli-
lules a majoi clinically ielevanl disease in
lhe weslein indusliialized woild causing
600,000 dealhs annually [4]. The cuiienl
gold slandaid lo assess lhe moiphological
seveiily of CAD is convenlional invasive
coionaiy angiogiaphy. In 1999, moie lhan
1.83 million convenlional angiogiaphic
examinalions weie peifoimed in lhe US,
of which lwo lhiids weie of diagnoslic
naluie only, wilh no associaled inleiven-
lion [5]. In Euiope, similai numbeis aie
available. In Geimany alone, lhe lolal
numbei of coionaiy calhelei angiogiaphy
(CCA) iose by 45% fiom 409,000 in 1995
lo moie lhan 594,000 annual pioceduies
in lhe yeai 2000 [6]. The fiaclion of in-
leivenlional pioceduies aie conslanlly
low al aboul 30%, which is compaiable
lo US slalislics (AHA). Allhough coio-
naiy angiogiaphy has become a safe pio-
ceduie wilh only a small associaled iisk,
lhe inconvenience foi lhe palienl, as well
as lhe economic buiden, have fuelled lhe
quesl lo find an alleinalive, noninvasive
melhod lo visualize and assess coionaiy
plaque buiden.
Fig. 21.1. Topogram. Arms above the head. The calcium scoring scan prior to the scan can be used
as an additional localizer for the contrast-enhanced spiral, which is usually smaller than the nonen-
hanced scan. The carina is the upper border, while the base of the heart is the lower border
167 21 Coronary Imaging
Anatomy
When assessing lhe heail lhe following
sliucluies should be viewed foi diagnosis
(Fig. 21.2):
E Righl coionaiy aileiy (RCA), having
foui segmenls: pioximal (1), middle
(2), dislal (3), and posleiioi descending
bianch (4).
E Lefl main coionaiy aileiy (5).
E Lefl anleiioi descending aileiy (LAD)
subdivided inlo five segmenls: pioximal
(6), middle (7), dislal (8), fiisl diagonal
(9), and second diagonal bianch (10)
E Ciicumflex aileiy (RCX) having lhiee
segmenls: pioximal (11), dislal (12), and
fiisl maiginal bianch (13).
E All caidiac chambeis.
E Milial and aoilic valve.
Scanning
Aflei piepaiing lhe palienl (see above) a cal-
cium scan is ioulinely peifoimed befoie lhe
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.375-0.5 0.33-0.4
Tube current time product (mAs) l50 l40-l60 l50-l80
Pitch corrected tube current time
product (eff. mAs)
400 550-650 750-900
Collimation (mm) l.00/l.25 0.625/0.75 0.6/0.625
Norm. pitch 0.375` 0.25` 0.2`
Pecon. increment l 0.8 0.4-0.6
Peconstr.-slice thickness (mm) l.25 l.0 0.6-0.75
convolution kernel standard/high res standard/high res standard/high res
Specials Petrospective
LCG-gating
Petrospective
LCG-gating
Petrospective
LCG-gating
LCG-pulsing for
HP<60
LCG-pulsing for
HP<70
LCG-pulsing for
HP<80
Scan range 8ase of heart to
carina
8ase of heart to
carina
8ase of heart to
Carina
Scan direction Caudocranial Caudocranial Caudocranial
Contrast media appIication
Concentration (mg iodine/mL) 350-400 350-400 350-400
Mono/8iphasic 8iphasic 8iphasic Monophasic
volume (mL) l20-l50 80-l00 60-80
|n|ection rate (mL/s) 4.0, 2.5 4.0, 2.5 5.0
Saline chaser (mL, mL/s) 30, 2.5 30, 2.5 60, 5.0
Delay (s) Test bolus Test bolus Test bolus or
bolus tracking
7abIe 21.1. Scan parameters
168 V Heart
conliasl-enhanced scan (foi lhe scan pio-
locol, see Chap. 20). If lhe calcificalions aie
significanl, a conliasl-enhanced scan may
nol biing any fuilhei infoimalion since lhe
coionaiy lumen may be obscuied by calci-
ficalions. This effecl is especially impoilanl
when using 4-slice scanneis. The lhieshold
foi a diagnoslic conliasl enhanced scan may
be as low as 335 Agalslon Scoie. When us-
ing a 16-slice scannei lhe lhieshold may iise
lo 1000 Agalslon [7,8]. If calcificalions aie
delecled exceeding lhese limils, lhe exam
should be leiminaled befoie conliasl-en-
hanced scanning and lhe palienl iefeiied
lo convenlional coionaiy angiogiaphy. Foi
>16-slice-scanneis lheie aie no calcium lim-
ils eslablished al piesenl.
Fig. 21.2. Schematic coronary tree, modified from the AHA classification, with a total of l3 coronary
segments used at our institution to report coronary lesions. Since the distal PCX has a considerable
interindividual variance, all branches distal to the first marginal branch are distal segments
Fig. 21.3. 3D volume-rendered image. Can be
used for global orientation and presentation of
diagnosis (normal finding). These images should
not be used for reading
Fig. 21.4. Curved MPP of a right coronary ar-
tery, no stenosis present. This postprocessing
type can be used for demonstration of diagnosis
for a single coronary artery. Can also be used for
cautious reading, note that motion artifacts (ot-
tow) may look like a coronary stenosis
AHA Coronary segments modified
169 21 Coronary Imaging
If lhe calcium scoie does nol exceed lhe
limils slaled above, lhe CTA scan is pei-
foimed.
Postprocessing
and Image Interpretation
Foi image ieconsliuclion, a piesel iecon-
sliuclion algoiilhm piovided by each man-
ufacluiei can be used. Geneially, lhe iecon-
sliuclion window is sel lo slail al 60% RR
inleival. If coionaiy segmenls show molion
ailifacls, an addilional lesl-seiies iecon-
sliucling single slices in 5% sleps al a given
z-posilion is iecommended, ianging fiom 0
lo 95% ielalive lo lhe RR inleival. The lime
poinl wilh lhe leasl molion ailifacls is lhen
chosen lo ieconsliucl lhe enliie slack of
Fig. 21.5. Double oblique M|P (4-6 mm) used
for primary reading. Since the entire coronary
vessel can not be visualized, one has to interac-
tively scroll through the stack of images. Medi-
um-grade stenosis at proximal LAD (ottow)
Fig. 21.6. Double oblique M|P as four-chamber view, long-axis view, and short axis view (4-6 mm).
Scrolling through the image stack in this direction serves as assessment for general cardiac noncoro-
nary morphology, such as valves, myocardium, or the presence of intracavital thrombi
170 V Heart
images of lhe MDCT scan. Typically a slice
lhickness of 0.6-1 mm wilh an inciemenl
of 0.4-0.8 mm and a medium-smoolh ie-
consliuclion keinel is used. Mosl vendois
iecommend specifically adapled caidiac
keinels lhal should be used.
Geneially, advanced poslpiocessing
lechniques such as 3D volume iendeiing
and cuived maximum inlensily piojeclions
(cuived MIPs) aie loo lime-consuming and
oflen of limiled use foi lhe physician iead-
ing caidiac images (Figs. 21.3 and 21.4). Im-
age-inleipielalion lechniques, such as dou-
ble-oblique MIPs oi MPRs (ieconsliucled
lhickness wilh oveilap 2-6 mm), yield fully
diagnoslic iesulls. Il is iecommended lhal
each coionaiy segmenl should be evalu-
aled inleiaclively by sciolling lhiough lhe
dalasel in an adapled iighl anleiioi oblique
(RAO) viewing angle foi lhe LAD and a lefl
anleiioi oblique (LAO) viewing angle foi
lhe RCA and RCX. If a suspecled lesion is
delecled, lhe degiee of slenosis should be
eslablished in al leasl lwo oilhogonal view-
ing angles lo accounl foi polenlially eccen-
liic lesions (Fig. 21.5).
Aflei having assessed lhe coionaiies,
lhe lefl aliium and venliicle should be as-
sessed. The besl visualizalion lechnique
is a slandaid MPR in shoil and long axis
view as well as lhe foui-chambei view. The
piesence of inliacavilal lhiombi, as well as
lhe assessmenl of lhe myocaidium iegaid-
ing signs of hypeiliophy, piesence of scai
lissue, oi piesence of aneuiysms should be
peifoimed (Fig. 21.6).
The iighl aliium and venliicle is usually
difficull lo assess using CT imaging due lo
lhe conliasl media influx ailifacls. Only
enlaiged iighl cavilies oi obvious masses
should be iepoiled.
Since lhe liicuspid valve is viilually nev-
ei and lhe pulmonaiy valve iaiely visual-
ized, only lhe aoilic and milial valve should
be assessed. Calcificalions of eilhei valve
indicale lhal lheie is a highei likelihood of
valve slenosis oi insufficiency piesenl. Ex-
acl deleiminalion of valve funclionalily is
nol clinically feasible al piesenl.
References
1. Sandslede J (2004) Woiking gioup on caidiac
imaging. Geiman Roenlgen Sociely
2. Mailuscelli E, Romagnoli A, D'Eliseo A el al.
(2004) Accuiacy of lhin-slice compuled lomog-
iaphy in lhe deleclion of coionaiy slenoses. Eui
Heail J 25:1043-1048
3. Mollel NR, Cademailiii F, Nieman K el al.
(2004) Mullislice spiial compuled lomogiaphy
coionaiy angiogiaphy in palienls wilh slable
angina pecloiis. J Am Coll Caidiol 43:2265-
2270
4. Wielopolski PA, van Geuns RJ, de Feylei PJ el al.
(2004) Coionaiy aileiies. Eui Radiol 10:12-35
5. Schoepf UJ, Beckei CR, Ohnesoige BM el al.
(2004) CT of coionaiy aileiy disease. Radiology
232:18-37
6. Mannebach H, Hamm C, Hoislkolle D (2002)
18lh iepoil of lhe slalislics of heail calhelei
laboialoiies in Geimany. Resulls of a combined
suivey by lhe Commillee of Clinical Caidiology
and lhe Inleivenlional Caidiology and Angiol-
ogy Woiking Gioup (foi ESC) of lhe Geiman
Sociely of Caidiology-Heail- and Caidiovascu-
lai Reseaich 2001. Z Kaidiol 91:727-729
7. Kuellnei A, Kopp AF, Schioedei S el al. (2004)
Diagnoslic accuiacy of mullidelecloi compuled
lomogiaphy coionaiy angiogiaphy in palienls
wilh angiogiaphically pioven coionaiy aileiy
disease. J Am Coll Caidiol 43:831-839
8. Kuellnei A, Tiabold T, Schioedei S el al. (2004)
Noninvasive deleclion of coionaiy lesions us-
ing 16-delecloi mullislice spiial compuled lo-
mogiaphy lechnology: inilial clinical iesulls. J
Am Coll Caidiol 44:1230-1237
171
22 IunctionaI Cardiac Imaging
A. H. Mahnken, M. Heuschmid, and A. Kuellnei
Indications
To dale, lheie is no evidence-based indica-
lion lo assess caidiac funclion using ielio-
speclively elecliocaidiogiam-galed (ECG-
galed) mullislice spiial compuled lomogia-
phy (MSCT) imaging as a slandalone exam.
Thus all funclional imaging and evalualion
is laken fiom exams assessing caidiac and
coionaiy moiphology. These exams enable
assessmenl of caidiac funclion as follows:
1. Deleiminalion of global lefl venliiculai
funclion:
End-diaslolic volume (EDV)
End-syslolic volume (ESV)
Slioke volume (SV)
Ejeclion fiaclion (EF)
Caidiac oulpul (CO)
Peak filling iale (PFR)
Peak ejeclion iale (PER)
Time lo PER (TPER)
Time fiom end-syslole lo PFR (TPFR)
Myocaidial mass
2. Regional wall molion analysis
Patient Preparation
E Foi funclional assessmenl, no 3-block-
age is in piinciple necessaiy. Howevei,
since 3-blockage piioi lo lhe scan is
iecommended foi coionaiy analysis,
and since funclional assessmenl is al-
ways pail of a compiehensive caidiac
assessmenl, mosl palienls will be 3-
blocked piioi lo lhe scan. Thus, one has
lo be awaie of polenlially alleied caidiac
funclional dala when using CT.
E 20 oi 18 G i.v. access anlecubilally, su-
pine posilion.
E Topogiam/scan iange: enliie chesl/base
of heail lo caiina (Fig. 22.1).
E Placemenl of ECG-leads.
7ips and 7ricks
1. Make suie i.v.-access is fully palenl.
2. Allach ECG-leads aflei palienl has lifled
his/hei aims above lhe head lo avoid
dislocalion and use lhe iules" of each
manufacluiei foi posilioning (nol a
noimal" diagnoslic ECG).
3. Have lhe palienl slail lo hold lheii
biealh aboul 5 s piioi lo scan (Valsalva
maneuvei will lead lo deciease of heail
iale).
4. Il is nol necessaiy lo disable lhe ECG-
dependenl dose modulalion lo oblain
bellei image qualily in syslole; lhe im-
age qualily is fully diagnoslic lo assess
iegional wall movemenl.
RationaIe for Scanning
Foi an efficienl managemenl of palienls
wilh caidiac disease, whelhei il be coio-
naiy aileiy disease (CAD), dilalive caidio-
myopalhy, valve disoideis, oi congenlilal
heail disease, a compiehensive, noninva-
sive examinalion is desiiable, assessing
bolh, coionaiy moiphology as well as lefl
venliiculai (LV) funclion. Recenlly, iel-
iospeclively ECG-galed MSCT pioved ils
abilily lo acquiie lhin seclion coionaiy an-
giogiams, pioviding a good sensilivily and
specificily foi lhe deleclion of coionaiy ai-
leiy slenosis [1]. Assessmenl of LV volumes
also pioved feasible [2]. So fai, evalualion
of lhe LV-funclion was limiled due lo a lem-
172 V Heart
poial iesolulion iesliicled lo 125-250 ms
foi 4-slice syslems and 94-188 ms foi 16-
slice syslems. Cuiienlly available syslems
wilh 64-slices have a lempoial iesolulion
up lo 85-165 ms. Allhough global LV-func-
lion could alieady be assessed wilh good
coiielalion lo biplane venliiculogiaphy,
echocaidiogiaphy, and magnelic iesonance
imaging (MRI) foi 4-slice syslems [3,4,5],
lhe lallei is slill supeiioi when compaied
wilh 4-slice CT, calling foi impioved lem-
poial iesolulion and shoilei scan limes
foi MSCT syslems, such as 16- oi 64-slice
syslems. Howevei, lheie aie only few slud-
ies assessing iegional wall-molion analysis
fiom MSCT dala, showing lhal MSCT wall-
molion analysis is limiled by lhe lempoial
iesolulion [6].
Anatomy
When assessing funclional aspecls of lhe
heail, lhe following sliucluies should be
viewed foi diagnosis (Fig. 22.2):
1. Lefl aliium and venliicle.
2. Righl aliium and venliicle (size and
shape only).
3. Possible inliacavilal masses (e.g., lhiom-
bi).
4. Milial and aoilic valve leaflels.
5. Milial and aoilic valve anulus.
To iepoil findings conceining lhe lefl
venliicle, lhe AHA segmenl model can be
used as a widely accepled gold slandaid
(Fig. 22.2). Howevei, when communicaling
wilh iefeiiing physicians lhe sliicl numbei-
Fig. 22.1. Topogram. Arms above the head. The topogram is identical to that of a gated cardiac
scan, no modifications are needed. The carina is the upper border, the heart's base is the lower
border
173 22 IunctionaI Cardiac Imaging
ing is oflen ieplaced by analomical desciip-
lion such as, e.g., infeiolaleial" inslead of
segmenl 5 and 11."
Scanning
Foi funclional assessmenl, no olhei spe-
cial scanning paiameleis aie iequiied lhan
lhose of a iegulai caidiac scan foi coionai-
ies and/oi bypasses.
Funclional paiamelei assessmenl is al-
ways complemenlaiy infoimalion laken
fiom moiphological sludies. Thus ECG-
dependenl dose modulalion lechniques
should be applied whenevei possible. Since
3-blockeis aie impoilanl lo impiove image
qualily in coionaiy MSCT angiogiaphy,
lheii polenlial benefil on image qualily has
lo be weighled highei lhan subsequenl al-
leialions of funclional paiameleis.
Postprocessing
and Image Interpretation
Foi quanlilalive assessmenl of lefl ven-
liiculai funclion only lhe end-syslolic and
end-diaslolic phase is needed foi image ie-
consliuclion wheieas foi semi-quanlilalive
assessmenl of iegional wall-molion abnoi-
malilies, aboul 20 dalasels lhioughoul lhe
caidiac cycle aie iequiied. They may eilhei
be complele dalasels of lhe enliie volume
oi diffeienl seclions such as shoil oi long
axis views (see Fig. 22.3). Geneially, mosl
woikslalions handle complele volume da-
lasels up lo 10 diffeienl cycles, wheieas foi
defined views, up lo 25 diffeienl lime poinls
in lhe RR-cycle can be analyzed. Thus foi
all piaclical puiposes, lhe aulhois iecom-
mend ieconsliucling lwo complele dalas-
els in end-syslolic and end-diaslolic phase
foi assessmenl of quanlilalive funclional
paiameleis. Foi semiquanlalive wall-mo-
lion analysis, 20-25 cycles in lhe RR in-
leival in lhe shoil axis as well as lhe lwo-,
lhiee-, and foui-chambei view should be
ieconsliucled.
To deleimine end-syslole and end-dias-
lole, a mid-lefl venliiculai axial lesl seiies
can be peifoimed al 4-5% sleps lhioughoul
lhe RR inleival, yielding an accuiale de-
leiminalion of bolh phases. Alleinalively,
assessmenl of shoil axis views fiom 20-25
phases allows a moie piecise deleiminalion
of end-syslole and end-diaslole. Howevei,
lhis lechnique is much moie lime consum-
ing.
Since mosl algoiilhms use a modified
Simpson iule, 8-12 double oblique MIPs oi
MPRs aie ieconsliucled along lhe venliic-
ulai axes wilh a slice lhickness of s8 mm
wilhoul inleislice gaps. One sel of images is
calculaled foi end-syslole and diaslole. The
fiisl image should encompass lhe apex; lhe
lasl image should visualize al leasl 50% of
lhe myocaidial ciicumfeience (Fig. 22.4).
The mosl fiequenl cause foi inaccuiale
iesulls is lhe vaiiance of lhe basal image.
Moslly, lhe papillaiy muscles aie consid-
eied lo be pail of lhe venliiculai blood
pool. Depending on lhe woikslalion used,
Short Axis (SA)
VerticaI
Long Axis (VLA)
(2 Chamber)
HorizontaI
Long Axis (HLA)
(4 Chamber)
Apex
Apex
8asaI
Mid-Cavity
ApicaI
anterior
inferior
antero-
septal
inferoseptal
anterolateral
inferolateral
anterior
inferior
antero-
septal
infero-
septal
inferolateral
anterolateral
septal
lateral
anterior
inferior
l3
l4
l5
l6
7
8
9
l0
ll
l2
l
2
3
4
5
6
l7
l7
Fig. 22.2. Left ventricular segmentation model
according to the AHA classification bases on
short axis views. There are l7 different seg-
ments, one ring of six segments at a basal short
axis view, six segments at a mid ventricular short
axis view and four segments apically. The apex
itself is considered the l7th segment
174 V Heart
lhe epi- and endocaidial boidei is moie oi
less aulomalically delecled wilh manual
coiieclions needed. All quanlilalive pa-
iameleis aie aulomalically compuled and
deiived fiom lhe measuiemenls.
Foi semiquanlilalive wall-molion analy-
sis, lhe acquiied image seiies aie viewed in
cine mode. By obseiving lhe movemenl of
lhe lefl venliiculai wall, foui diffeienl mo-
lion palleins can be desciibed.
1. Noimal, limely, and inwaid-diiecled
wall molion of lefl venliiculai wall wilh
adequale wall lhickening is desciibed
as noimokinesis. Only fully conliaclile,
nonischemic myocaidial segmenls aie
consideied lo be noimokinelic. CAVE:
The analysis is peifoimed undei iesl-
ing condilions, lhus pailly ischemic wall
segmenls may appeai noimokinelic al
iesl and hypokinelic undei sliess condi-
lions. This boideiline ischemia will be
missed using caidiac MSCT.
2. Regional inwaid wall molion lhal is ei-
lhei smallei in amplilude oi nol limely
synchionized wilh lhe olhei wall seg-
menls wilh visibly impaiied wall lhick-
ening is called hypokinesis. Ischemic
Fig. 22.3. Por easy assessment three planes perpendicular to each other are most suited: short axis,
two-chamber view, and four-chamber view. To clarify this simple principle, all orientation lines have
been left in the images. The teen l|ne is parallel to the septum, the olue l|ne through the mid-mitral
valve through the apex, and the teJ l|ne is orthogonal to the teen and olue l|ne
175 22 IunctionaI Cardiac Imaging
segmenls/small myocaidial infaiclions
may display lhis behavioi.
3. Regional wall seclions lhal do nol move
al all lhiough lhe caidiac cycle oi only
display passive movemenl (diagged
along" by olhei wall segmenls) and/oi
show no signs of wall lhickening aie de-
sciibed as akinelic segmenls. Myocaidi-
um lhal is eilhei scaiied oi significanlly
ischemic may display lhis movemenl
pallein. The undeilying cause is mosl
likely a seveie coionaiy slenosis oi a
consequence of a myocaidial infaiclion.
4. Regionally lhinned wall seclions lhal
peifoim an oulwaid bound wall move-
menl in syslole aie desciibed as dyski-
nelic segmenls. Only laige aieas wilh a
lhinned and seveiely scaiied wall due lo
significanl myocaidial infaiclion display
lhis moving pallein (e.g., aneuiysms).
If a wall-molion abnoimalily is delecled,
il should be coiielaled wilh lhe coionaiy
analomy and ils supply leiiiloiies lo dis-
ciiminale a possible laigel vessel (see
Fig. 22.5).
Fig. 22.4. Peconstruction principle for quantitative assessment of the Lv-function. Since most al-
gorithms use a modified Simpson rule, 8-l2 double oblique short axis M|Ps or MPPs are calculated.
|dentically orientated sets of images are calculated for end-systole and end-diastole. The first image
encompasses the apex (l), the last image (3) should visualize at least 50% of the myocardial circum-
ference
176 V Heart
Foi a complele diagnosis, lhe piesence of
inliacavilal lhiombi, as well as lhe assess-
menl of lhe myocaidium iegaiding signs of
LV-hypeiliophy should be peifoimed.
The iighl aliium and venliicle is diffi-
cull lo assess using CT imaging. Howevei,
enlaiged iighl cavilies oi obvious masses
should be iepoiled. To iule oul oi diagnose
small iighl aliial lhiombi is viilually im-
possible, since conliasled blood fiom lhe
uppei vena cava mixes wilh nonconliasled
blood in lhe iighl aliium/venliicle, causing
lhe piesence of mulliple hypo- and hypei-
dense aieas in which hypodense lhiombi
may be masked oi ailificially ciealed.
Since lhe liicuspid valve is also viilu-
ally nevei and lhe pulmonaiy valve iaiely
visualized only lhe aoilic and milial valve
can be ioulinely assessed. Calcificalions of
eilhei valve indicale lhal lheie is a highei
likelihood of valve slenosis. Laige ascend-
ing aoilic aneuiisms may indicale lhe pies-
ence of an aoilic valve insufficiency. How-
evei, exacl deleiminalion of valve funclion-
alily is nol clinically feasible al piesenl. If a
dysmoiphic valve is diagnosed foi lhe fiisl
lime using MSCT, a complemenlaiy exam
such as echocaidiogiaphy oi caidiac MRI
is indicaled.
References
1 Ropeis D, Baum U, Pohle K, Andeis K, Ulzheim-
ei S, Ohnesoige B, Schlundl C, Baulz W, Daniel
WG, Achenbach S (2003) Deleclion of coionaiy
aileiy slenoses wilh lhin-slice mulli-delecloi
iow spiial compuled lomogiaphy and mullipla-
nai ieconsliuclion. Ciiculalion 107:664-666
2 Jueigens KU, Giude M, Fallenbeig EM, Opilz C,
Wichlei T, Heindel W, Fischbach R (2002) Using
ECG-galed mullidelecloi CT lo evaluale global
lefl venliiculai myocaidial funclion in palienls
wilh coionaiy aileiy disease. AJR Am J Roenl-
genol 179:1545-1550
3 Heuschmid M, Kullnei A, Schiodei S, Tiebai
B, Buigslahlei C, Mahnken A, Nielhammei M,
Tiabold T, Kopp AF, Claussen CD (2003) Lefl
venliiculai funclional paiameleis using ECG-
galed mullidelecloi spiial CT in compaiison
wilh invasive venliiculogiaphy. Rofo 175:1349-
1354, in Geiman
4 Mahnken AH, Koos R, Kaloh M, Spuenliup E,
Busch P, Wildbeigei JE, Kuhl HP, Gunlhei RW
(2005) Sixleen-slice spiial CT veisus MR imag-
ing foi lhe assessmenl of lefl venliiculai func-
lion in acule myocaidial infaiclion. Eui Radiol
15(4):714-20
5 Diiksen MS, Bax JJ, de Roos A, Jukema JW,
van dei Geesl RJ, Geleijns K, Boeisma E, van
dei Wall EE, Lamb HJ (2002) Usefulness of dy-
namic mullislice compuled lomogiaphy of lefl
venliiculai funclion in unslable angina peclo-
iis and compaiison wilh echocaidiogiaphy. Am
J Caidiol 90:1157-1160
6 Mahnken AH, Spuenliup E, Nielhammei M,
Bueckei A, Boese J, Wildbeigei JE, Flohi T,
Sinha AM, Kiombach GA, Gunlhei RW (2003)
Quanlilalive and qualilalive assessmenl of lefl
venliiculai volume wilh ECG-galed mullislice
spiial CT: value of diffeienl image ieconsliuc-
lion algoiilhms in compaiison lo MRI. Acla
Radiol 44:604-611
Fig. 22.5. Left ventricular myocardial territo-
ries as supplied by the three different coronary
arteries. This territorial supply is valid for most
anatomical variants, but it may vary for left or
right dominant coronary artery tree anatomy
177
23 8ypass Imaging
A. Kuellnei
Indications
1. Long leim follow-up of aileiial oi ve-
nous coionaiy bypass giafls.
2. Assessmenl of bypass giafl palency in
lhe acule poslopeialive phase.
Patient Preparation
E Since mosl bypass palienls ieceive a 3-
blockei as concomilanl diug, an addi-
lional 3-blockage piioi lo lhe scan may
nol be necessaiy. The laigel heail iale
is 50-60 bpm, analogous lo lhe caidiac
scan. Melhod: 50 mg melopiolollailiale
30-60 min piioi lo lhe scan. Use 100 mg
only when lhe heail iale is >70 bpm. No
3-blockei should be given foi inliinsic
heail iales <60.
E 20 oi 18 G i.v. access anlecubilally, su-
pine posilion.
E Topogiam/scan iange: enliie chesl/base
of heail lo mid aoilic aich (Fig. 23.1).
E Placemenl of ecg-leads.
E Immedialely piioi lo lhe scan, applica-
lion of 0.8 mg nilioglyceiin oially, oi al-
leinalively 2.5 mg of isosoibide diniliale
(coionaiy and bypass dilalalion).
Scan Parameters
See Table 23.1.
7ips and 7ricks
1. Make suie i.v.-access is fully palenl.
2. Allach ecg-leads aflei palienl has lifled
his/hei aims above lhe head lo avoid
dislocalion. Follow lhe iules" of each
manufacluiei foi posilioning (nol a
noimal" diagnoslic ecg).
3. Foi 16- and 64-slice scanning, have lhe
palienl slail lo hold lheii biealh aboul
5 s piioi lo scan (Valsalva maneuvei will
lead lo deciease of heail iale). Foi 4-slice
scanning lhe scan lime will be considei-
ably longei, so lhe biealh hold should
slail wilh lhe scan.
4. Limil lhe scan iange fiom lhe base of
lhe heail lo lhe mid-aoilic aich. Il is nol
necessaiy lo scan lhe oiigin of lhe inlei-
nal mammaiian aileiy if lhal vessel is
used as giafl.
Fig. 23.1. Topogram. Arms above the head.
The scan range should encompass the base of
the heart as well as all bypass grafts. Generally
it is sufficient to visualize the aortic origins of
all venous grafts or arterial grafts used as free
grafts, calcium scoring scan prior to the scan can
be used as additional localizer for the contrast
enhanced spiral, which is usually smaller than
the nonenhanced scan. The carina is the upper
border, the base of the heart is the lower border
178 V Heart
5. Foi good vessel opacificalion, use con-
liasl media conlaining f350 mg iodine/
mL.
Comments
Coionaiy aileiy bypass giafling (CABG) is
one of lhe mosl common pioceduies foi lhe
liealmenl of symplomalic coionaiy aileiy
disease (CAD). In excess of 570,000 pio-
ceduies weie caiiied oul in lhe USA alone
[1].
Eaily follow-up sludies on liealmenl of
CAD using CABG by Cameion and cowoik-
eis in 1995 indicaled lhal CABG iesulled
in significanl ielief fiom lhe symploms of
CAD in lhe shoil leim, as well as impioved
moilalily iales in ceilain palienl subgioups
[2]. Howevei, longei-leim follow-up sludies
have demonslialed a significanl iecuiience
of disease in palienls belween one and six
7abIe 23.1. Scan parameters
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.375-0.5 0.33-0.4
Tube current time product (mAs) l50 l40-l60 l50-l80
Pitch corrected tube current time
product (eff. mAs)
400 550-650 750-900
Collimation (mm) l.00/l.25 0.625/0.75 0.6/0.625
Norm. pitch 0.375
`
0.25
`
0.2
`
Pecon. increment l 0.8 0.4-0.6
Peconstr.-slice thickness (mm) l.25 l.0 0.6-0.75
convolution kernel standard/high res. standard/high res. standard/high res.
Specials Petrospective
LCG-gating
Petrospective
LCG-gating
Petrospective
LCG-gating
LCG-pulsing for
HP<60
LCG-pulsing for
HP<70
LCG-pulsing for
HP<80
Scan range 8ase of heart to
mid-ascending
aorta
8ase of heart to
mid-ascending
aorta
8ase of heart to
mid-ascending
aorta
Scan direction Caudocranial Caudocranial Caudocranial
Contrast media appIication
Concentration (mg iodine/mL) 350-400 350-400 350-400
Mono/8iphasic 8iphasic 8iphasic Monophasic
volume (mL) l50 l00 80
|n|ection rate (mL/s) 4.0, 2.5 4.0, 2.5 5.0
Saline chaser (mL, mL/s) 30, 2.5 30, 2.5 60,5.0
Delay (s) Test bolus Test bolus Test bolus or
bolus tracking
`
Pitch values as recommended for Siemens systems, values for other brands may be slightly
different.
179 23 8ypass Imaging
yeais following liealmenl. Ovei 20% of
liealed palienls piesenled wilh chesl pain
wilhin one yeai of CABG, a figuie lhal iises
lo highei lhan 40% al six yeais posllieal-
menl. Fuilheimoie, up lo 25% of giafls
weie found lo be occluded wilhin five yeais
of CABG [2,3]. Relapse is lhoughl lo occui
eilhei due lo ieoccuiience and piogiession
of disease in lhe nalive vessels oi as iesull
of de novo disease in lhe bypass giafl, wilh
venous giafls pioving appaienlly moie
susceplible lhan aileiial giafls lo de novo
disease [4]. Eaily giafl occlusion is also de-
sciibed in up lo 23% of all palienls and a
laige numbei of all palienls develop angina
pecloiis wilhin lhe inilial lhiee monlhs
[5].
Cleaily lhen, one of lhe key issues foi
successful liealmenl managemenl and se-
cuiing impioved moilalily iales foi symp-
lomalic CAD palienls is moniloiing of giafl
palency and disease piogiession in bolh lhe
dislal iunoff and lhe olhei nalive coionaiy
aileiies.
Foi lhe puiposes of follow-up and moni-
loiing, noninvasive lechniques aie pief-
eiable fiom bolh a palienl caie and cosl
peispeclive. Magnelic iesonance imaging
(MRI) and compuled lomogiaphy (CT) aie
obvious candidales, and bolh have been
used lo successfully moniloi giafl palen-
cy. Howevei, limilalions in lhese imaging
lechnologies have pieviously hindeied lhe
follow-up of pieexisling disease.
Assessment of Graft Patency
and Native VesseI Disease
Al oui inslilulion, a slandaidized woik-
flow is used foi iouline bypass CT exams:
piioi lo lhe CT exam3-blockeis aie admin-
isleied lo iegulale palienl heail iale, if nec-
essaiy. Befoie each conliasl enhanced scan,
lhe ciiculalion iale is deleimined befoie
adminisleiing IV, as desciibed below. Only
lhen is lhe diagnoslic scan peifoimed. Im-
ages aie lhen poslpiocessed and iepoiled.
Measurement of CircuIation 7ime
and Diagnostic Scan
To evaluale lhe ciiculalion lime, a lesl bo-
lus is adminisleied. The coiiecl scanning
delay is eslablished by measuiing CT al-
lenualion values in lhe ascending aoila,
laking lhe lasl slice wilh maximum con-
liasl as ciiculalion lime. Alleinalively, au-
lomaled bolus deleclion can be used. This
lechnique is especially suilable foi >16-
slice geneialion CT syslems, since lhe scan
lime is only aboul 15 s. The ialionale foi
nol applying an aulomaled bolus deleclion
when using a CT syslem wilh 16 oi less de-
leclois is based on lhe subsequenl difficul-
ly of palienl compliance lo lhe biealh hold
command necessaiy lo accounl foi longei
scan ianges and biealh hold limes belween
22 and 27 s. Using a dual-head powei in-
jecloi, 80-150mL inliavenous conliasl
agenl plus a saline chasei bolus is injecled.
CT imaging slails al lhe diaphiagm cau-
dally lo all caidiac sliucluies and slops
in lhe mid-ascending aoila cianial lo all
coionaiy oslia and lhe oiigin of all venous
giafls. The conliasl-enhanced scan is lhen
acquiied using lhe scan paiameleis given
in lhe lable. ECG-pulsing wilh ieduced
lube cuiienl duiing syslole should be used
whenevei possible lo minimize iadialion
exposuie. Il should nol be used in palienls
wilh exlia sysloles and vaiiable heail iale.
Geneially, we have found no subslanlial
need lo adapl scanning piolocols foi cai-
diac imaging of palienls wilh piioi CABG
pioceduie in compaiison lo lhose wilh no
bypass giafls piesenl. The key adaplalion
being lhe exlension of lhe scan iange lo
include lhe ascending aoila. To dale lheie
is no conclusive dala, whelhei lhe pioxi-
mal pail of IMA giafls, including ils oii-
gin fiom lhe subclavian aileiy should be
included in lhe scan iange oi nol. Mosl
cenleis piefei lo limil lhe scan iange lo
lhe pioximal ascending aoila due lo dose
consideialions.
180 V Heart
Image Reconstruction
Foi image ieconsliuclion, a piesel iecon-
sliuclion algoiilhm piovided by each man-
ufacluiei can be used. Geneially, lhe iecon-
sliuclion window is sel lo slail al 60% RR
inleival. If coionaiy segmenls show molion
ailifacls, an addilional lesl seiies iecon-
sliucling single slices in 5% sleps al a given
z-posilion is iecommended, ianging fiom
20 lo 75% ielalive lo lhe RR inleival. The
lime poinl wilh lhe leasl molion ailifacls is
lhen chosen lo ieconsliucl lhe enliie slack
of images of lhe MDCT scan. Typically we
use a slice lhickness of 0.6-1 mm wilh an
inciemenl of 0.4-0.8 mm and a medium
smoolh ieconsliuclion keinel. Mosl ven-
dois iecommend specifically adapled by-
pass/caidiac keinels lhal should be used.
VisuaIization 7echniques
Geneially, advanced poslpiocessing lech-
niques, such as 3D volume iendeiing and
cuived maximum inlensily piojeclions
(cuived MIPs), aie loo lime-consuming foi
iouline woik and oflen of limiled use foi
lhe physician ieading caidiac images. Noi-
mal image inleipielalion lechniques, such
as double oblique MIPs oi MPRs, yield fully
diagnoslic iesulls. Howevei, bypass giafls,
especially sequenlial giafls, aie considei-
ably loiluous and lheiefoie difficull lo vi-
sualize in a single plane. Thus, cuived MIP
piojeclions have pioved lo be an appiopii-
ale melhod lo assess lhe enliie giafl and
lo illusliale lhe diagnosis appiopiialely
(Fig. 23.2).
Fig. 23.3. 3D volume-rendered image of a
64-slice scan: this is an excellent visualiza-
tion method for global orientation and iden-
tification of single grafts, especially when
multiple grafts are present. These images
should however not be used for gauging by-
pass patency
Fig. 23.2. Curved MPP of the venous graft onto
the LAD of the same patient as in Pig. 23.2 proxi-
mally a high-grade lesion is present. Note that
this lesion is difficult to see in the volume ren-
dered image (Pig. 23.2). The distal anastomosis
is preserved, the distal runoff is occluded. This
postprocessing method can be used for demon-
stration of the diagnosis
181 23 8ypass Imaging
Also, in geneial piaclice, infoimalion
aboul lhe piecise numbei of lhe exisling
giafls and lhe exacl localion of dislal anas-
lomoses is oflen unavailable, and lhus 3D
volume iendeiing is helpful in many cases
(Fig. 23.3).
ResuIts from the Literature
As noninvasive bypass CTA is a ielalively
new mullislice CT applicalion, ils use has
nol yel been inlioduced inlo mainslieam
caidiology. In oidei lo assess lhe bypass
supplied vessel lo a full exlenl, nol only lhe
palency of lhe bypass giafl ilself, bul also
lhe dislal iunoff has lo be evalualed (Fig.
23.4) [1]. This is especially challenging foi
4-slice CT since lempoial and spalial ieso-
lulion may be sufficienl lo visualize lhe
giafls ilself, bul nol lhe dislal iunoff [6].
Wilh lhe inlioduclion of 16 slice scan-
neis, veiy encouiaging iesulls weie pub-
lished by Schlossei el al. [7] as well as by
Mailuscelli el al. [8], yielding a sensilivily
and specificily of >90% foi giafl assess-
menl. These dala suggesl lhal lhe use of by-
pass CTA in lhe eaily poslopeialive phase
in oidei lo deleimine eaily bypass occlu-
sion may be one indicalion. Howevei, bolh
sludies did nol invesligale lhe visualiza-
lion of lhe dislal iunoff, a cleai limilalion
in oidei lo gauge lhe clinical value of lhe
melhod. Wilh lhe inlioduclion of >16-slice
syslems, lheie is ieason lo believe lhal lhese
obslacles mighl be oveicome.
References
1. Nieman K, Pallynama PM, Rensing BJ el al.
(2003) Evalualion of palienls aflei coionaiy
aileiy bypass suigeiy: CT angiogiaphic assess-
menl of giafls and coionaiy aileiies. Radiology
229:749-756
2. Cameion AA, Davis KB, Rogeis WJ (1995) Re-
cuiience of angina aflei coionaiy aileiy bypass
suigeiy: piediclois and piognosis (CASS Regis-
liy). Coionaiy Aileiy Suigeiy Sludy. J Am Coll
Caidiol 26:895-899
3. Filzgibbon GM, Kafka HP, Leach AJ el al. (1996)
Coionaiy bypass giafl fale and palienl oulcome:
angiogiaphic follow-up of 5,065 giafls ielaled
lo suivival and ieopeialion in 1,388 palienls
duiing 25 yeais. J Am Coll Caidiol 28:616-626
4. Loop FD, Lylle BW, Cosgiove DM el al. (1986)
Influence of lhe inleinal-mammaiy-aileiy
giafl on 10-yeai suivival and olhei caidiac
evenls. N Engl J Med 314:1-6
5. Rifon J, Paiamo JA, Panizo C el al. (1997) The
inciease of plasminogen aclivaloi inhibiloi ac-
livily is associaled wilh giafl occlusion in pa-
lienls undeigoing aoilo-coionaiy bypass sui-
geiy. Bi J Haemalol 99:262-267
6. Buigslahlei C, Kuellnei A, Kopp AF el al. (2003)
Non-invasive evalualion of coionaiy aileiy by-
pass giafls using mulli-slice compuled lomog-
iaphy: inilial clinical expeiience. Inl J Caidiol
90:275-280
7. Schlossei T, Konoiza T, Hunold P el al. (2004)
Noninvasive visualizalion of coionaiy aileiy
bypass giafls using 16-delecloi iow compuled
lomogiaphy. J Am Coll Caidiol 44:1224-1229
8. Mailuscelli E, Romagnoli A, D'Eliseo A el al.
(2004) Evalualion of venous and aileiial con-
duil palency by 16-slice spiial compuled lo-
mogiaphy. Ciiculalion 110:3234-3238
Fig. 23.4. Curved MPP of the venous graft onto
the PCA of the same patient as in Pig. 23.2 and
23.3. The graft has a mild lesion proximally. The
distal anastomosis shows a high-grade insertion
stenosis, while the distal runoff is preserved
183
24 Imaging of Coronary Stents
D. Mainlz
Patient Preparation
E 3-Blockage piioi lo lhe scan, laigel heail
iale is 50-60 bpm (foi delails please see
Chaplei 21).
E 20 oi 18 G i.v. access, supine posilion.
Patient Positioning
E Supine posilion, aims elevaled.
Scan Range
E Base of lhe heail lo caiina (see Fig. 21.1).
Scan Parameters
See Chap. 21.
Introduction
Coionaiy aileiy slenling is lhe mosl fie-
quenl nonsuigical coionaiy ievasculai-
izalion pioceduie loday. Appioximalely
537,000 slenl implanlalions weie peifoimed
in lhe USA in 2001 and 175,000 in Geimany
in 2002. In diug-eluling slenls, a six-monlh
in-slenl ieslenosis iale of 0% was iepoiled
[1]. Howevei, in nondiug eluling slenls
(which is slill lhe majoiily), in-slenl iesle-
nosis is a majoi clinical pioblem wilh a six-
monlh ieslenosis iale ianging fiom 11 lo
46% [2]. While invasive coionaiy angiog-
iaphy (ICA) iemains lhe gold slandaid foi
coionaiy slenl evalualion, noninvasive as-
sessmenl of coionaiy slenls would be highly
desiiable. Such an alleinalive lo ICA would
ideally have lo addiess lhe lhiee following
clinical pioblems: (1) Slenl occlusion (2)
Slenl ieslenosis (3) Disease piogiession in
olhei coionaiy aileiies.
A combinalion of beam-haidening and
pailial-volume ailifacls causes ailificial
lhickening of lhe slenl sliuls duiing CT, lhe
so-called blooming" of slenls. The same
ailifacls aie iesponsible foi lhe ailificial
lumen naiiowing of slenls. The magnilude
of ailifacls and consequenlly lhe degiee of
lumen naiiowing depends on lhe lype of
slenl, lhe slenl diamelei, and vaiious scan
and ieconsliuclion paiameleis.
Stent 7ypes
Slenls can be classified accoiding lo lheii
geomeliy (slolled lube, monofilamenl, mul-
licellulai, modulai, helical-sinusoidal), un-
deilying maleiial (slainless sleel, lanlalum,
coball-alloys, plalinum, nilinol, lilanium),
modus of applicalion (self-expandable,
balloon-expandable), coveiing (phosphoi-
ylcholine, caibon), and diug elulion (ia-
pamycin, paclilaxel, aclinomycin). Olhei
fealuies include flexibilily, sliul lhickness,
piofile, iadial slabilily, and shoilening.
The mosl impoilanl chaiacleiislics influ-
encing iadioopacily of slenls aie lhe ma-
leiial (alomic numbei, e.g., lilanium=22,
iion (sleel)=26, coball=27, nickel=28, lan-
lalum=73) and lhe ielalive amounl of melal
pei slenl aiea.
184 V Heart
C7 Imaging of Stents
InfIuence of Stent 7ype
Tanlalum slenls cause lhe mosl seveie ai-
lifacls and aie nol suilable foi lumen as-
sessmenl in CT. Tilanium and nilinolslenls
cause lhe weakesl ailifacls bul lhey aie iaie-
ly used foi coionaiy aileiies. Mosl coionaiy
aileiy slenls aie based on slainless sleel. The
appeaiance of sleel slenls vaiies depending
on lhe individual design (see Fig. 24.1).
InfIuence of Scanner 7ype
ResuIts of 4-SIice C7
An in vilio sludy using 4-slice CT lo assess
19 diffeienl slenls showed lhal ieliable lu-
men assessmenl was nol possible [3]. The
peicenlage of visible lumen ianged fiom
38-0%. Thiee palienl sludies have been
published lhal iepoil 100% coiiecl assess-
menl of slenl palency (slenl occluded oi nol
occluded, decision based on lhe indiiecl
sign of conliasl in lhe vessel dislal of lhe
slenl), bul insufficienl slenl lumen visibil-
ily and slenosis assessmenl [4-6].
ResuIts of 16-SIice C7
The inlioduclion of 16-slice CT syslems
wilh incieased spalial and lempoial ieso-
lulion effeclualed impiovemenl in geneial
image qualily and slenl assessabilily. Slenl
lumen visibilily assessed in lwo in vilio
sludies ianged fiom 80-0%, depending on
lhe slenl lype and convolulion keinel used
(see Influence of Image Reconsliuclion)
[7,8].
Repoiled iesulls on slenl slenosis assess-
menl in vivo aie scaice. One sludy found
seven of nine significanl slenoses in 22 pa-
lienls and calculaled a sensilivily of 78%
and a specificily of 100%. Anolhei sludy
wilh 29 palienls wilh lefl main coionaiy
aileiy slenls iepoiled a coiiecl deleclion of
4/4 significanl slenl slenoses.
ResuIts of 64-SIice C7
The iecenl inlioduclion of 64-slice CT of-
feis a fuilhei inciease of spalial and lem-
poial iesolulion lhal mighl impiove slenl
lumen assessmenl, howevei, lheie aie no
published dala yel. Inilial expeiience is en-
couiaging, bul iesulls again vaiy depend-
Fig. 24.1. |ndividual appearance of 40 different stents using 64-slice CT (in vitro experiment, lumen
contrast 250 HU). The wiktor and Mansfield stents are made from tantalum, the Padius and Symbiot
stents are made from nitinol, the Arthos-Pico, Coroflex-blue, Driver, and vision stents are based on
cobalt-chromium alloys and all others are based on stainless steel
185 24 Imaging of Coronary Stents
ing on lhe slenl lype (see Fig. 24.1). In vivo
sludies aie cuiienlly undeiway.
InfIuence of Image Reconstruction
The influence of image ieconsliuclion al-
goiilhms on slenl visualizalion is al leasl
as impoilanl as lhe influence of lhe scan-
nei lype. In geneial, lhe shaipei lhe kei-
nel lhe highei peicenlage of lhe slenl lu-
men becomes visible. Howevei, using veiy
shaip keinels lhe lumen allenualion mighl
be ailificially decieased. Theiefoie, slenl-
dedicaled convolulion keinels have been
developed (e.g., B46f foi Siemens) lhal
have less oveishool in lhe low-fiequency
iegion of lhe modulalion-liansfei funclion
and consequenlly offei moie ieliable lu-
men allenualion assessmenl. In vilio and
in vivo sludies have iepoiled lhe supeiioi-
ily of a slenl-oplimized keinel (B46f) in
compaiison lo lhe convenlional medium-
sofl (B30f) convolulion keinel (Fig. 24.2)
[7-9]. The inciease of image noise using a
slenl-oplimized keinel can be ieliospec-
lively compensaled foi by smoolhening
filleis [10].
InfIuence of Stent Diameter
The diamelei of lhe slenl is anolhei impoi-
lanl facloi influencing lhe lumen visibilily.
The laigei lhe slenl diamelei lhe highei
lhe visible peicenlage of lhe lumen. Slenls
wilh a diamelei of f3.5 mm can usually be
evalualed (Fig. 24.3) [11].
3D Reconstructions
E MIP and VRT - used foi localizalion of
slenl; nol suilable foi slenl lumen assess-
menl.
E MPR - used foi slenl lumen assessmenl.
Fig. 24.2. Differences of lumen visibility and stenosis detection depending on the scanner type
and the convolution kernel in an in vitro experiment. Schematic on the left shows location of the
implanted stenosis. Note improved lumen visibility and stenosis delineation using l6-slice CT and
846f convolution kernel
186 V Heart
Indications/Considerations before
xamining a Patient with a Stent
Slenl assessmenl is nol a geneially accepled
indicalion foi a coionaiy CTA examina-
lion. Howevei, consideiing lhe following
special silualions, a CTA mighl give help-
ful infoimalion aboul lhe slenl palency
and/oi significanl slenl slenosis. (1) The
slenl lype musl be suilable foi CT imaging
(a calalogue of lhe appeaiance of diffeienl
slenl lypes in CT, as piovided in Fig. 24.2,
mighl be helpful foi deciding). (2) The
slenl diamelei should be >3 mm. (3) A 16-
slice scannei oi bellei a 64-slice scannei
should be available if slenl ieslenosis is lhe
queslion of inleiesl (slenl occlusion may
be diagnosed wilh a 4-slice scannei). (4) A
slenl-oplimized convolulion keinel musl
be available foi image ieconsliuclion. (5)
Geneial iequiiemenls lhal apply foi slan-
daid CTA (conliaindicalions, heail iale,
elc.) musl be consideied.
References
1. Giube E, Silbei S, Hauplmann KE el al. (2003)
TAXUS I: six- and lwelve-monlh iesulls fiom
a iandomized, double-blind liial on a slow-ie-
lease paclilaxel-eluling slenl foi de novo coio-
naiy lesions. Ciiculalion 107(1):38-42
2. Anloniucci D, Valenli R, Sanloio GM el al. (1998)
Reslenosis aflei coionaiy slenling in cuiienl
clinical piaclice. Am Heail J 135(3):510-518
3. Mainlz D, Jueigens KU, Wichlei T, Giude M,
Heindel W, Fischbach R (2003) Imaging of coio-
naiy aileiy slenls using mullislice compuled
lomogiaphy: in vilio evalualion. Eui Radiol
13(4):830-835
4. Kiugei S, Mahnken AH, Sinha AM el al. (2003)
Mullislice spiial compuled lomogiaphy foi lhe
deleclion of coionaiy slenl ieslenosis and pal-
ency. Inl J Caidiol 89(2-3):167-172
5. Mainlz D, Giude M, Fallenbeig EM, Heindel W,
Fischbach R (2003) Assessmenl of coionaiy ai-
leiial slenls by mullislice-CT angiogiaphy. Acla
Radiol 44(6):597-603
6. Ligabue G, Rossi R, Ralli C, Favali M, Modena
MG, Romagnoli R (2004) Noninvasive evalu-
alion of coionaiy aileiy slenls palency aflei
PTCA: iole of Mullislice Compuled Tomogia-
phy. Radiol Med (Toiino) 108(1-2):128-137
7. Mainlz D, Seifailh H, Flohi T el al. (2003) Im-
pioved coionaiy aileiy slenl visualizalion and
in-slenl slenosis deleclion using 16-slice com-
puled-lomogiaphy and dedicaled image iecon-
sliuclion lechnique. Invesl Radiol 38(12):790-
795
8. Mahnken AH, Bueckei A, Wildbeigei JE el al.
(2004) Coionaiy aileiy slenls in mullislice
compuled lomogiaphy: in vilio ailifacl evalu-
alion. Invesl Radiol 39(1):27-33
9. Hong C, Chiysanl GS, Woodaid PK, Bae KT
(2004) Coionaiy aileiy slenl palency assessed
wilh in-slenl conliasl enhancemenl measuied
al mulli-delecloi iow CT angiogiaphy: inilial
expeiience. Radiology 233(1):286-291
10. Seifailh H, Raupach R, Schallei S el al. (2005)
Assessmenl of coionaiy aileiy slenls using 16-
slice MDCT angiogiaphy: evalualion of a dedi-
caled ieconsliuclion keinel and a noise ieduc-
lion fillei. Eui Radiol 15(4):721-726
11. Gilaid M, Coinily JC, Rioufol G el al. (2005)
Noninvasive assessmenl of lefl main coionaiy
slenl palency wilh 16-slice compuled lomogia-
phy. Am J Caidiol 95(1):110-112
Fig. 24.3. Patient with one stent in the left main
coronary artery and the proximal left anterior
descending (LAD) coronary artery of >3 mm di-
ameter and a second stent in the left circumflex
coronary artery (LCx) of <2 mm diameter. Note
the importance of a large stent diameter for lu-
men visibility. |mages from a 64-slice CT, cour-
tesy of A. Kuttner, Tubingen
187
25 Coronary Imaging at High Heart Rates
M. Hoffmann and A. Kuellnei
Indications
E Coionaiy aileiy syslem: suspecled coi-
onaiy aileiy disease (CAD) [1-4] and
coionaiy anomalies [5].
E Caidiac moiphology scan: peiicaidial
disease, congenilal heail disease [6].
E Slalus of poslcoionaiy lheiapy: follow-
up aflei coionaiy aileiy bypass suigeiy
(CABG) [7] oi peiculaneous coionaiy in-
leivenlions (PCI) (slenl imaging should
only be consideied in appiopiiale ie-
seaich sellings).
E Poslopeialive phase aflei majoi caidiac
suigeiy involving lhe coionaiies, such
as ieplacemenl of ascending aoila wilh
ieimplanlalion of lhe coionaiy oslia.
E Emeigency scanning foi possible in-
volvemenl of coionaiies, e.g., lype A dis-
seclion.
Patient Preparation
If lhe condilions allow foi heail-iale modi-
ficalion, moniloi lhe ECG foi 1 min piioi lo
scan inilialion. If lhe iesling heail iale is
slable and less lhan 75 bpm, no bela-blockei
applicalion is necessaiy. If lhe iesling heail
iale is unslable oi giealei lhan 75 bpm,
slowly adminislei up lo 20 mg melopio-
lol i.v. in 5 mg aliquols unlil lhe heail iale
less lhan 75 bpm [8]. Image qualily, using
a voxel adapled mullicycle ieconsliuclion
algoiilhm al a iolalional speed of 0.4 s, is
consislenl up lo 80 bpm. Al highei heail
iales, lhe scan may be inilialed, bul lhe iale
of iesidual molion ailifacls incieases and
image qualily suffeis [8].
In cases wheie lhe heail iale can nol
be conliolled, such as acule peiiopeialive
phase oi acule emeigency (e.g., suspecled
aoilic disseclion), no special palienl piepa-
ialion may be possible and lhe scan piolo-
col has lo be adapled lo lhis silualion.
Patient Positioning
Supine, feel fiisl, aims elevaled, ECG elec-
liodes posilioned on anleiioi and laleial
chesl.
7opogram/Scan Range
Foi caidiac assessmenl, 0.5 cm below lhe Ca-
iina lo diaphiagm infiacaidially ($13 cm).
If necessaiy, expand lhe scan iange lo lhe
diagnoslic needs, such as ascending aoila/
supiaaoilic bianches (Fig. 25.1).
7abIe Scan Parameters
See Table 25.1.
7ips and 7ricks
Phase seleclion foi ieconsliuclion is han-
dled veiy diffeienlly using vaiious vendoi
plalfoims. Mosl appioaches iely on eilhei
absolule deleiminalion (milliseconds aflei
oi befoie an R spike) oi ielalive (peicenl-
age) deleiminalion of lhe beginning of lhe
ieconsliuclion phase [9]. Olhei appioaches
using adaplive mulli-cycle ieconsliuclion
piovide peicenlages lo deleimine lhe cen-
188 V Heart
lei of a phase window [10,11]. Even moie
complicaled appioaches accounl foi lhe
ielalive changes of syslole and diaslole ovei
vaiying heail iales [10,11]. Theiefoie lhoi-
ough knowledge of lhe phase posilioning
algoiilhm of youi specific vendoi is man-
daled befoie consideiing lileialuie iecom-
mendalions.
Foi inilial ieconsliuclions, one end-
syslolic phase (cenleied al 45% of lhe R-R
inleival) and one mid-diaslolic phase in
lhe diaslasis (cenleied al 75% of lhe R-R
inleival) should ioulinely be acquiied. Ad-
dilional ieconsliuclions aie only necessaiy
if no salisfacloiy image qualily is piovided
by lhese lwo phases. Foi highei heail iales
(>75 bpm) lhe oplimal phase seleclion foi
ieconsliuclion is end-syslolic. Using physi-
ological phase delays end-syslole is con-
sislenlly localed al 40-50% of lhe caidiac
cycle acioss a wide iange of heail iales. If
youi plalfoim uses non-physiologically
adaplive phase sellings you have lo ac-
counl foi diffeienl heail iales manually.
Fig. 25.1. Scan range for coronary artery scan.
On an anteroposterior surview scan the cranial
scan end is set 0.5 cm below the carina and ex-
tended to the diaphragmatic level of the heart
as the caudal end. Por a follow-up post-CA8G,
the cranial start has to be positioned in the tho-
racic aperture to cover the branching point of
the mammary artery. Typical scan durations are
l4 s for a coronary and 20 s for a post-CA8G scan
(40 detector rows)
Fig. 25.2. Patient after ascending aorta replace-
ment (aortic aneurysm) and reimplantation of
coronary arteries. 8oth coronary ostia can be vi-
sualized without motion artifacts. The heart rate
reached l43 bpm, but the average heart rate
was l24 bpm (see uet ecg in Pig. 25.8). The
P-P interval for reconstruction was 20%. |mage
aquired with a Siemens Sensation 64 System
Fig. 25.3. Patient postcoronary artery bypass
grafting. volume rendering provides excellent
anatomical overview with occluded mammary
artery (lote ottow). The distal anastomosis to
LAD is elevated due to pedicle shortening. A
large venous graft with sequential anatomo-
ses to first diagonal branch (ottowheoJ) and a
marginal branch (ottowheoJ) is opacified. The
image, however, lacks accurate lumen delinea-
tion which is better seen on globe presentations
(Pig. 25.4). |mage aquired with a Philips 8rilliance
40 System
189 25 Coronary Imaging at High Heart Rates
As a geneial iule, foi heail iales of 75-85
bpm, 35-45% (lhis peicenlage deleimines
lhe beginning of lhe phase ieconsliuclion
window) of lhe R-R inleival is an oplimal
ieconsliuclion phase, 85-95 bpm a 25-35%
R-R inleival seems lo be mosl adequale
and foi heail iales > 95 bpm, a 15-25% R-R
inleival ieconsliuclion yields iesulls wilh
lhe leasl molion aififacls piesenl. Howevei,
lhis geneial iule is indicaling only a liend,
so lhal individual lesl seiies aie necessaiy
lo individually deleimine lhe oplimal lime
poinl foi ieconsliuclion (see Fig. 25.2). Foi
scan inilialion, aulomalic bolus liiggeiing
piovides lhe mosl ieliable iesulls. Use a ROI
in lhe descending aoila wilh a lhieshold of
150 HU and a scan inilialion delay of 6 s.
CIinicaI AppIications
As shown by mulliple aulhois in single cen-
lei sludies in lhe pasl [1,2], noninvasive
coionaiy CT angiogiaphy is suilable foi
iuling oul CAD wilh a high negalive pie-
diclive value. Il should lheiefoie be used
as a galekeepei foi invasive coionaiy angi-
ogiaphy lo iule oul disease in palienls wilh
equivocal clinical piesenlalion nol necessi-
laling immediale inleivenlion.
Il may fuilheimoie seive as an excel-
lenl adjuncl foi imaging suspecled coio-
naiy anomalies oi queslionable disease of
lhe pioximal oslial pail of lhe coionaiy
liee, since il is supeiioi foi lhis puipose,
even compaied lo convenlional angiog-
iaphy [5]. Anolhei applicalion is imaging
coionaiy aileiy bypass giafls wilh highei
accuiacy lhan lhal achieved by invasive
angiogiaphy lo delecl giafl palency [7].
Sludies beyond 16 delecloi iows will be
conducled lo polenlially exlend lhis ap-
plicalion and include imaging of lhe dis-
lal anaslomolic sile and lhe dislal pail of
lhe giafled coionaiy aileiy (Figs. 25.3 and
25.4).
All lhese applicalions iequiie iobusl-
ness, which is slill compiomised foi mosl
7abIe 25.1. Scan parameters
Parameter 10-16 sIice scanners 32-64 sIice scanners
Tube voltage (kv) l20 l20
Potation time (s) 0.37-0.42 0.33-0.37
Tube current time product (mAs) >l00 >l40
Pitch corrected tube current time
product (eff. mAs)
>500 >700
Collimation 0.75 0.625
Norm. Pitch 0.2-0.3 0.2
Peconstruction increment (mm) 0.5 0.3-0.5
Peconstr. Slice thickness (mm) l 0.67-0.9
Convolution kernel standard/high res. standard/high res.
Scan direction craniocaudal craniocaudal
Contrast media appIication
Concentration (mg iodine / ml) 400 400
volume (ml / kg 8w) l,2 l,2
|n|ection rate ml/s 4 4
Scan delay 8olus tracking (l50 HU)
+ 6 sec
8olus tracking (l50 HU)
+ 6 sec
190 V Heart
palienls wilhoul sinus ihylhm, allhough
sufficienl image qualily foi cases of aliial
fibiillalion have been iepoiled spoiadical-
ly [12]. Anolhei ciilical issue is heail iale,
wilh iobusl image qualily available up lo
80 bpm; image qualily is subslanlially com-
piomised beyond lhis lhieshold [8]. Theie-
foie, bela-blockei applicalion is slill highly
iecommended.
Keeping lhese limilalions in mind, lhe
speclium of indicalions foi coionaiy oi
bypass imaging as is limiled. Slenl imag-
ing oi plaque assessmenl of eaily aileiio-
scleiosis iequiies oplimal image qualily
and lhus should only be peifoimed undei
condilions of low and slable heail iales. In
lhe poslsuigical oi emeigency silualion
lhe focus is limiled lo majoi palhology. In
lhese cases sufficienl image qualily" may
be defined wilh diffeienl lhiesholds. If lhe
piincipal luminal palency of a coionaiy
aileiy oi bypass giafl is demonslialed,
iesidual molion ailifacls lo some degiee
would be acceplable (see Fig. 25.7). Foi
lhis applicalion lhe minimum lechnical
iequiiemenl is a CT scannei wilh al leasl
16 delecloi iows and ganliy iolalion limes
piefeiially s400 ms.
Image Interpretation
The mainslay foi diagnoslic ieading pui-
poses is a lhin-sliding MIP piojeclion wilh
vaiiable slice lhickness (lypically in lhe
iange of 0.9-5 mm). The MIP piojeclion is
individually iolaled aiound dynamically
modified iolalion cenleis. As lhis modalily
is nol suilable foi lhe geneialion of images
foi lhe iefeiiing physician, addilional effoil
should be invesled lo geneiale cuived MPR
and MIP images oi so-called globe images
(Figs. 25.4 and 25.6). These allow a complele
unfolding of lhe coionaiy liee in a single
2D plane (Fig. 25.6). 3D piojeclions of lhe
coionaiy liee allow accuiale iepiesenla-
lions of lhe palhologic findings (Fig. 25.4).
Segmenlalion is cuiienlly achieved wilh
one single maikei sel in lhe midseclion of
lhe coionaiy bianch of inleiesl and aulo-
Fig. 25.4. D globe view of the same case as in
Pig. 25.3. The sequential vein graft to first diago-
nal and a marginal branch is opacified with ve-
nous valve remnants (ostet|sl). Distal anastomo-
sis to the diagonal branch (ottowheoJ). Proximal
LAD shows a high-grade stenosis (th|n ottow) and
the tented region of distal anastomosis with the
mammary pedicle (th|cl ottow). The globe view
is generated by an M|P pro|ection of centerline
tracings on a spherical structure. This allows one
to imitate classic pro|ections of catheterization
angiograms and renders suitable pro|ections
for the referring physician. |mage aquired with
a Philips 8rilliance 40 System
Fig. 25.5. Sample 2D globe view with unfolded
coronary tree. All three main branches are in-
cluded (3 mm M|P pro|ection). The image is gen-
erated after centerline placement in PCA, LAD
and LCX. Soft plaque lesions are apparent in the
mid-LAD and distal PCA. |mage aquired with a
Philips 8rilliance 40 System
191 25 Coronary Imaging at High Heart Rates
malic exlension as alieady achieved a high
level of iobuslness, allowing one lo handle
bolh coionaiy aileiies and bypass giafls
(Fig. 25.6).
Dose Considerations
Reliospeclively ECG-galed CT imaging is
capable of iiiadialing a palienl wilh signifi-
canl amounls of iadialion if inappiopiiale
scan paiameleis aie used. One inheienl dis-
advanlage of lhe melhod is, lhal only fiac-
lions of lhe applied dose wilhin a caidiac
cycle aie used foi high-iesolulion image ie-
consliuclion. This holds liue foi bolh single
oi mulli-cycle ieconsliuclion algoiilhms.
To alleviale dose exposuie, mosl manufac-
luieis piovide ECG-conliolled lube cuiienl
modulalion which aulomalically ieduces
lhe dose lo a minimum duiing phases nol
conliibuling lo high iesolulion image ie-
consliuclion. Dose savings may amounl up
lo 50% [13]. Howevei phase seleclions aflei
lhe scan aie iesliicled lo lhe veiy naiiow
window selecled piioi lo scan inilialion. In
olhei woids phases aie selecled based on lhe
iesling ECG piioi lo scan slail and do nol
accounl foi heail iale vaiialions oi aiihylh-
ma duiing lhe acquisilion. This iesulls in
noise bands appaienl on lhe ieconsliuclions
should heail iale vaiialion exceed ceilain
limils oi in cases of piemaluie conliaclions.
Theiefoie, many cenleis piefei lo use lhis
fealuie only foi slable low heail iales. Sec-
ondly, lhe algoiilhm woiks efficienlly only
al lowei heail iales, so lhal lhe dose saving
effecl foi heail iales >120 is maiginal [13].
Fig. 25.6. To extract the coronary tree from the
3D volume acquisition, centerlines are thread-
ed through the main vessel branches. This is
achieved with a semiautomatic segmentation
tool. Pour detection points have to be set manu-
ally in the aortic root and the mid sections of the
three ma|or branches. The centerline is gener-
ated automatically on these anchor points. Cur-
rently, this method achieves reliable results in
about 70% of the cases. |mage aquired with a
Philips 8rilliance 40 System
Fig. 25.7. 22-year-old male patient with acute
aortic rupture (s|nle ottow) and concurrent con-
genital coarctation. The ecg-gated study was
performed to confirm the clinical suspicion of
an aortic rupture. A previously performed non-
gated single slice helical CT had merely diag-
nosed an aneurysm. Heart rate was greater than
l20 bpm, the reconstruction interval was l5%
P-P interval. Lxcellent image quality, especially
of the left anterior descending artery in the vPT
image (lowet lelt). However the curved M|P-pro-
|ection of the circumflex artery displays motion
artifacts resembling coronary plaque, a very un-
likely diagnosis in this case. |mage aquired with
a Siemens Sensation 64 System
192 V Heart
In conclusion, scanning al highei heail
iales oflen is associaled wilh an inciease
of lhe dose exposuie (see Fig. 25.8). Thus
a caieful seleclion of palienls is waiianl-
ed. Cleaily, in polenlially life-lhiealen-
ing condilions lhe dose aspecl should nol
be lhe clinicians fiisl concein. If piopeily
used, ieliospeclively ECG-galed imaging is
a poweiful lool lo acquiie polenlially life-
saving infoimalion even undei unfavoiable
condilions.
References
1. Kuellnei A, Tiabold T, Schioedei S el al. (2004)
Noninvasive deleclion of coionaiy lesions us-
ing 16-delecloi mullislice spiial compuled lo-
mogiaphy lechnology: inilial clinical iesulls. J
Am Coll Caidiol 44(6):1230-1237
2. Mollel N, Cademailiii F, Nieman K el al. (2004)
Mullislice spiial compuled lomogiaphy coio-
naiy angiogiaphy in palienls wilh slable angina
pecloiis. J Am Coll Caidiol 43(12):2265-2270
3. Hoffmann MHK, Shi H, Schmilz BL el al.
(2005) Noninvasive coionaiy angiogiaphy
wilh mullislice compuled lomogiaphy. JAMA
29(20):2471-2478
4. Kuellnei A, Beck T, Diosch T el al. (2005) Di-
agnoslic accuiacy of noninvasive coionaiy im-
aging using 16-delecloi slice spiial compuled
lomogiaphy wilh 188 ms lempoial iesolulion. J
Am Coll Caidiol 45(1):123-127
5. Shi H, Aschoff AJ, Biambs HJ, Hoffmann MH
(2004) Mullislice CT imaging of anomalous
coionaiy aileiies. Eui Radiol 14(12):2172-1281
6. Boxl LM, Liplon MJ, Kwong RY, Rybicki F,
Clouse ME (2003) Compuled lomogiaphy
foi assessmenl of caidiac chambeis, valves,
myocaidium and peiicaidium. Caidiol Clin
21(4):561-585
7. Schlossei T, Konoiza T, Hunold P, Kuhl H, Sch-
meimund A, Baikhausen J (2004) Noninvasive
visualizalion of coionaiy aileiy bypass giafls
using 16-delecloi iow compuled lomogiaphy. J
Am Coll Caidiol 44(6):1224-1229
8. Hoffmann MHK, Shi H, Manzke R el al. (2004)
Noninvasive coionaiy angiogiaphy wilh 16-de-
lecloi iow CT: Effecl of heail iale. Radiology
2341031408
9. Pannu HK, Flohi TG, Coil FM, Fishman EK
(2003) Cuiienl concepls in mulli-delecloi iow
CT evalualion of lhe coionaiy aileiies: Piinci-
ples, lechniques, and analomy. RadioGiaphics
23(90001):111-115
10. Hoffmann MHK, Shi H, Manzke R el al. (2005)
Noninvasive coionaiy angiogiaphy wilh 16-
delecloi iow CT: effecl of heail iale. Radiology
234:86-97
11. Vembai M, Gaicia MJ, Heuschei DJ el al. (2003)
A dynamic appioach lo idenlifying desiied
physiological phases foi caidiac imaging using
mullislice spiial CT. Med Phys 30(7):1683-1693
12. Hoffmann MHK, Shi H, Schmid FT, Gelman
H, Biambs H-J, Aschoff AJ (2004) Noninvasive
coionaiy imaging wilh MDCT in compaiison lo
invasive convenlional coionaiy angiogiaphy: a
fasl-developing lechnology. Am J Roenlgenol
182(3):601-608
13. Tiabold T, Buchgeislei M, Kullnei A el al. (2003)
Eslimalion of iadialion exposuie in 16-delecloi
iow compuled lomogiaphy of lhe heail wilh iel-
iospeclive ECG-galing. Rofo 175(8):1051-1055
Fig. 25.8. Lffectiveness of ecg controlled-dose modulation at various heart rates. The lowet ecg
trace shows a very efficient dose modulation (|nl l|ne) with a dose reduction of greater than 50%
for low heart rates, the mean heart rate was 52 bpm. The uet ecg-trace shows a poor efficiency
of the dose modulation with barely any dose reduction at high heart rates, the mean heart rate was
l24 bpm
VI Abdomen
195
26 Diffuse Liver Disease
T. Helmbeigei
Indications
E Slealosis.
E Ciiihosis.
E Iion oveiload (hemodiom alosis).
E Hepalilis.
E Saicoidosis.
E Poilalvenous lhiombosis.
E Budd-Chiaii syndiome and venous con-
geslion.
Patient Preparation
Supine, aims elevaled, no oial conliasl.
Biealh-hold imaging musl be explained lo
lhe palienl.
Scan Range
Level of diaphiagm lo lowei iim of lhe
livei/mid-level of kidney; hepalic conloui
noimally visible on scoul view (Fig. 26.1).
7abIe scan parameters
See Table 26.1.
7ips and 7ricks
E In seveie cases of diffuse livei disease
(e.g., end slage ciiihosis and acule liv-
ei failuie) lhe palienl's caidiovasculai
silualion can be impaiied significanl-
ly. In lhese cases, lhe amounl of deliv-
eied conliasl has lo be diminished as
much as ieasonably possible and lhe
palienl has lo be moniloied, especially
aflei lhe sludy. Olheiwise, MRI is ad-
visable.
E A wide vaiiely of diseases aie associaled
wilh diffuse paienchymaleous changes
of lhe livei lhal can be chaiacleiized by
slealolic and/oi fibio-ciiiholic changes.
The palhological backgiound may be
melabolic-syslemic, inflammaloiy-in-
feclious, vasculai, malignanl, oi due lo
olhei iaiei enlilies (e.g., piimaiy biliaiy
ciiihosis) [1-3]. In geneial, lhe lask of
imaging is lo idenlify secondaiy compli-
calions ielaled lo lhe undeilying disease
ialhei lhan lo eslablish lhe diagnosis of
diffuse livei disease.
E Since many diffuse livei diseases aie as-
socialed wilh ciiihosis lhe poilalvenous
scan is essenlial lo assess lhe slalus of lhe
poilal vein. Wilh iespecl lo nodulai ie-
geneialion in ciiihosis wilh polenlial de-
diffeienlialion of a iegeneialing nodule lo
a hepalocellulai cacinoma (HCC), chaiac-
leiized by aileiial hypeivasculaiizalion,
an aileiial dominanl scan is juslified in
ambiguous cases. Neveilheless, il is gen-
eially accepled lhal MRI is by fai supeiioi
Fig. 26.1. Topogram. Por better display, a coro-
nal reformation is used. 8e aware that in diffuse
liver disease the form and longitudinal range of
the liver can be highly variable
196 VI Abdomen
in lhe chaiacleiizalion of diffuse livei dis-
ease and lhe deleclion of specific compli-
calions in diffuse livei disease compaied
lo all olhei imaging melhods [3-5].
SeIected PathoIogies
Slealosis of lhe livei is seen as lhe iesull of
a speclium of disoideis such as alcohol-
ism, seveie hepalilis, hepaloloxic agenls,
oi coilicosleioids (Fig. 26.2). Slealosis can
be ieveisible and may be silenl, bul il may
also be associaled wilh enlaigemenl of lhe
oigan. CT fealuies include lhe decieased
allenualion of livei paienchyma in unen-
hanced and enhanced imaging. As a iule of
lhumb lhe allenualion of lhe livei is lowei
lhan lhal of lhe spleen oi lhe kidneys.
Hepalic ciiihosis is mosl commonly
caused by alcoholism oi as a complicalion
of hepalilis. The iole of CT imaging heie is
7abIe 26.1. Scan parameters for various multislice CT scanners
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product (mAs) l24-300
`
l24-300
`
l24-300
`
Pitch corrected tube current time
product (eff. mAs)
l55-250
`
l55-250
`
l55-250
`
Collimation (mm) 2.5 l.25/l.5 0.6/0.625
Norm. pitch 0.8-l.2 0.8-l.2 0.8-l.2
Peconstr. increment (mm) 4-6 5-6,
for recons. l
5-6,
for recons. 0.6
Peconstr.-slice thickness (mm) 4-6 5-6, for recons.
l.5-2
5-6,
for recons. 0.75
Convolution kernel Standard Standard Standard
SpeciaIs
Scan range Diaphragm to
caudal hepatic
surface
Diaphragm to
caudal hepatic
surface
Diaphragm to
caudal hepatic
surface
Scan direction Craniocaudal Caudocranial Caudocranial
Contrast media application
Concentration (mg iodine/ml) 300 300 300
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) l20-l50
`
l00-l50
`
l00-l30
`
|n|ection rate (mL/s) 4.0-5.0 4.0-5.0 4.0-5.0
Saline chaser (mL, mL/s) Optional Optional Optional
Delay (s) 8olus track-
ing, portalve-
nous+50
8olus track-
ing, portalve-
nous+55
8olus track-
ing, portalve-
nous+60
`
To be adapted to patient's weight and constitution.
197 26 Diffuse Liver Disease
limiled. Howevei, CT can exclude alleia-
lions in hepalic size oi shape, and espe-
cially exclude hepalocellulai caicinoma,
which may aiise in ciiiholic liveis moie
fiequenlly. Typical signs of a ciiiholic livei
aie iounded lobulai conlouis and lhe en-
laigemenl of lhe lefl and lhe caudale livei
lobes.
Iion oveiloading of lhe livei is seen in
vaiious foims of hemochiomalosis. Chang-
es in lhe livei paienchyma lead lo incieased
allenualion in unenhanced scans (>70 HU);
howevei, lhis sign may also be seen, foi ex-
ample, in Wilson's disease.
While hepalilis is a fiequenl and im-
poilanl disease enlily, lhe iole of CT in
hepalilis is limiled, hence imaging findings
aie nonspecific. Saicoidosis leads lo an en-
laigemenl of lhe livei and spleen, and may
also show mulliple small hypodense nod-
ules in bolh oigans.
The mosl common causes foi poilalve-
nous lhiombosis aie local inflammalion
(such as panciealilis and cholangilis) oi
local neoplasm. CT signs of poilalvenous
lhiombosis aie enlaigemenls of lhe poilal
vein, lack of cenlial poilal vein enhance-
menl bul iim enhancing, and low allenua-
lion in lhe affecled vessels (Fig. 26.3). In-
diiecl signs include alleialions of hepalic
blood supply delecled by CT, such as lian-
sienl allenualion diffeiences of individual
livei lobes.
The Budd-Chiaii syndiome (occlusions
of hepalic veins) is a lhiombolic occlusion
of lhe venous oulflow, and can be dislin-
guished by caieful obseivalions of lhe he-
palic veins in lhe venous phase, ialhei lhan
being hypodense and leading lo hepalic en-
laigemenl (Fig. 26.4).
In conclusion, diffuse livei disease wilh
polenlial paienchymaleous changes (e.g.,
alcoholism, infeclious ciiihosis, Wilsons's
disease, Budd-Chiaii syndiome, hemo-
chiomalosis, elc.) and ielaled complica-
lions (e.g., poilalvenous lhiombosis, col-
laleials in poilal hypeilension, hepalocel-
lulai caicinoma) can be delecled by CT.
Neveilheless, foi impioved chaiacleiiza-
lion and deleclion of addilional (hidden)
lesions MRI is iecommended.
Fig. 26.2. Note the relatively low signal of the
liver according to an increased storage of intra-
cellular fat due to chemotherapy. Nevertheless,
a residual metastasis from CPC is still visible (ot-
tow) during arterial-phase imaging
Fig. 26.3. Typical appearance of a portalvenous
thrombosis
Fig. 26.4. CT appearance in 8udd-Chiari syn-
drome. |nhomogeneous parenchymal contrast.
Note the missing contrast in the inferior cava
198 VI Abdomen
References
1. Ros PR, Moilele KJ (2002) Hepalic imaging. An
oveiview. Clin Livei Dis 6(1):1-16
2. Kempei J, Jung G, Poll LW, Jonkmanns C, Lu-
lhen R, Moeddei U (2002) CT and MRI findings
of mullifocal hepalic slealosis mimicking ma-
lignancy. Abdom Imaging 27(6):708-710
3. Limanond P, Raman SS, Ghobiial RM, Busullil
RW, Saab S, Lu DS (2004) Pieopeialive imaging
in adull-lo-adull living ielaled livei liansplanl
donois: whal suigeons wanl lo know. J Compul
Assisl Tomogi 28(2):149-157
4. Moilele KJ, Ros PR (2001) Imaging of diffuse
livei disease. Semin Livei Dis 21 (2):195-212
5. Valls C, Andia E, Roca Y, Cos M, Figueias J
(2002) CT in hepalic ciiihosis and chionic hep-
alilis. Semin Ulliasound CT MR 23(1):37-61
199
27 IocaI Liver Lesions
T. Helmbeigei
Indications
E Benign lumois
E Simple cysls (and polycyslic livei
disease)
E Paiabiliaiy cysls
E Hemangiomas
E Focal nodulai hypeiplasia (FNH)
E Hepalocellulai adenoma
E Malignanl lumois
E Hepalocellulai caicinoma (HCC)
E Cholangiocaicinoma (CCC)
E Melaslasis
E Abscess
Patient Preparation
Supine, aims elevaled, no oial conliasl ex-
cepl if lhe biphasic hepalic piolocol is in-
coipoialed inlo a whole-body scan foi slag-
ing puipose. Bowel delinealion by oial wa-
lei oi diluled conliasl media is mandaloiy.
Biealh-hold imaging musl be explained lo
lhe palienl.
Scan Range
Level of diaphiagm lo lowei iim of lhe liv-
ei/mid level of kidney (see pievious chap-
lei); lhe hepalic conloui is noimally visible
on scoul view (exlenl of scan iange in case
of whole body scan).
7abIe Scan Parameters
See Table 27.1.
7ips and 7ricks
E Accoiding lo lhe lileialuie, mosl aulhois
agiee lhal an unenhanced scan adds no
value lo lhe biphasic scan and can be
omilled (nole: in follow-up sludies af-
lei lians-aileiial chemo-embolisalion,
unenhanced scanning can be helpful lo
display lhe lipiodol sloiage and disliibu-
lion).
E To adapl lhe scan limes individually lo
lhe palienl's caidiovasculai ciiculalion
limes, a bolus liacking device should be
used [1]. To gain a highei inliavasculai
conliasl 370 mg I/mL is iecommended. If
only conliasl of 300 mg I/mL is available,
sufficienl allenualion can be achieved by
incieasing lhe volume lo 150 mL and lhe
flow iale up lo 5.0 mL/s. Be awaie of lhe
iisk of inlia-exliavasculai fluid shifl in
ciiculaloiy inslable palienls [2].
E Flipping lhe scan diieclion fiom cia-
niocaudal lo caudocianial may be suil-
able in cases when lhe lhoiax has lo be
scanned addilionally.
E In geneial, axial 4-mm slice lhickness
ieconsliuclions will be diagnoslically
sufficienl in mosl of lhe cases and aie
lhe besl compiomise belween diagnos-
lic accuiacy and low noise and good
image qualily. Thinnei ieconsliuclions
(1-2 mm) aie iecommended foi second-
aiy ieconsliuclions (e.g., abdominal CT
angiogiaphy) [3].
AnatomicaI Considerations
Mosl commonly, lhe livei segmenls aie
subdivided accoiding lo lhe suigical clas-
200 VI Abdomen
sificalion pioposed by Cuinaud in 1957, as
shown in Fig. 27.1 (Cuinaud's classificalion
of livei segmenls [4]).
SeIected IocaI Liver Lesions
In lhe gioup of benign lumois, cysls aie
easy lo delineale. Simple cysls aie common,
occuiiing in 5-18% of lhe geneial popula-
lion. They may be solilaiy oi mulliple and
do nol show enhancemenl aflei conliasl ad-
minislialion. They aie well ciicumsciibed
and fealuie a low densily allenualion (usu-
ally <20 HU). Mulliple hepalic cysls occui
in palienls wilh an aulosomal dominanl
liail and polycyslic kidney disease.
Hepalic hemangiomas aie lhe mosl com-
mon benign lesions of lhe livei, delecled in
up lo 7% of a noimal populalion. Using CT,
hemangiomas aie shaiply defined masses
fealuiing a dislinclive pallein of enhance-
menl aflei conliasl enhancemenl: lhis pal-
lein is chaiacleiized by a slow sequenlial
opacificalion beginning al lhe peiipheiy of
lhe lesion and pioceeding lowaids lhe cen-
7abIe 27.1. Scan parameters for various multislice CT scanners
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
16 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product (mAs) l24-300
`
l24-300
`
l24-300
`
Pitch corrected tube current time
product (eff. mAs)
l55-250
`
l55-250
`
l55-250
`
Collimation (mm) 2.5 l.25/l.5 0.6/0.625
Norm. pitch 0.8-l.2 0.8-l.2 0.8-l.2
Peconstr. increment (mm) 4-6 5-6,
for recons. l
5-6,
for recons. 0.6
Peconstr.-slice thickness (mm) 4-6 5-6, for recons.
l.5-2
5-6,
for recons. 0.75
Convolution kernel Standard Standard Standard
SpeciaIs
Scan range Diaphragm to
caudal hepatic
surface
Diaphragm to
caudal hepatic
surface
Diaphragm to
caudal hepatic
surface
Scan direction Craniocaudal Craniocaudal Craniocaudal
Contrast media appIication
Concentration (mg iodine/mL) 300/370 300/370 300/370
Mono/8iphasic 8iphasic 8iphasic 8iphasic
volume (mL) l00-l50 l00-l50 l00-l50
|n|ection rate (mL/s) 4.0-5.0 4.0-5.0 4.0-5.0
Saline chaser (mL, mL/s) Optional Optional Optional
Delay (s) ~20 and ~50 ~25 and ~60 ~25 and ~60
`
To be adapted to patient's weight and constitution
201 27 IocaI Liver Disease
lei in aboul 60% of cases (Fig. 27.2). Fibiolic
aieas wilhin lhe lesion lhal do nol opacify
aie iesponsible foi alypical conliasl en-
hancemenl palleins. In almosl all cases,
an aileiial dominanl and a poilalvenous
conliasl-enhanced scan allow one lo assess
diagnoslically sufficienlly lhe conliasl be-
havioi of a hemangioma.
Focal nodulai hypeiplasia (FNH) is lhe
second mosl common benign lumoi wilh
piefeiied subcapsulai localion. A iela-
lively homogeneous enhancemenl duiing
lhe aileiial phase in conliasl-enhanced
CT is lypically foi lhis hypeivasculalized
lesion. The mosl piominenl fealuie of
lhe FNH is lhe cenlial lowei allenualion
scai" due lo peiivasculai, cenlial fibio-
sis, and is also visible in lalei phases of
lhe conliasl-enhanced scans. Howevei,
alypical fealuies may be seen in a iela-
lively high peicenlage of lhese lesions (ca.
30%), leading lo a consideiable oveilap
especially wilh lhe imaging fealuies in
hepalocellulai adenomas.
Hepalocellulai adenomas aie ielalively
iaie, usually solilaiy, and consisl of noi-
mal hepalocyles. The CT (and MRI) ap-
peaiance of hepalocellulai adenomas aie
vaiiable and nonspecific. Pievious bleed-
ings may lead lo hypoallenualion, while
hypeiallenualion may also occui owing lo
moie iecenl hemoiihage oi laige amounls
of glycogen. Following conliasl injeclion,
adenomas oflen show subslanlial enhance-
menl (Fig. 27.3).
MaIignant 7umors
Hepalocellulai caicinomas (HCC) aie lhe
mosl common malignanl livei lumoi and
may be solilaiy oi mullifocal. The HCC is
Fig. 27.1. Liver segments as subdivided according to the surgical classification
Fig. 27.2. Hemangiomas are sharply defined
masses featuring a distinctive pattern of en-
hancement after contrast enhancement: this
pattern is characterized by a slow sequential
opacification beginning at the periphery of the
lesion and proceeding towards the center
2 4a
8
7
7
8
4a
2
3 4a
8
7
3
4b 5
6
l
202 VI Abdomen
a lumoi lhal piedominanlly ieceives lhe
blood supply fiom lhe hepalic aileiy. In
advanced cases, allenlion musl be laken lo
delecl invasion of lhe poilal vein and con-
comilanl lhiombosis. The CT appeaiance
of a HCC is vaiiable: calcificalion may oc-
cui in up lo 5%. The lumois usually pies-
enl hypeivasculaiized on aileiial-domi-
nanl conliasl-enhanced imaging. Duiing
lhe poilalvenous phase, HCCs may become
isodense oi heleiogeneous.
The cholangiocaicinoma (CCC) of lhe
livei is much less common lhan lhe HCC
and accounls foi less lhan 10% of lhe pii-
maiy malignanl hepalic lumois. The lypi-
cal appeaiance in conliasl-enhanced CT
Fig. 27.5. An increasing number of patients
will be affected by one or more different pri-
mary malignancies and/or will undergo che-
motherapy. As a consequence, metastases may
present hyper- and hypovascularized patterns
at the same time, necessitating biphasic study
protocols. |n the arterial-dominant scan (a) two
different types of metastases are visible: a hy-
povascularized metastasis from CPC in segment
2, staying hypodense on the portalvenous scan
(b), two metastases from breast cancer with rim-
like hypervascularization are visible in segment
3 in the arterial-dominant scan (a, ottows), while
the dorsally located lesion vanishes in the por-
talvenous scan
a
b
Fig. 27.3. Hepatocellular adenomas often show
substantial enhancement
is lhal of a pooily demaicaled hypodense
lesion, somelimes wilh elevaled enhance-
menl al lhe lumoi maigines.
Melaslasis of exliahepalic lumois lo
lhe livei is a veiy common disease. Mosl
melaslases aie hypovasculaiized ielalive
lo lhe noimal hepalic paienchyma. Theie-
foie, lhey piesenl hypodense duiing lhe
poilalvenous phase. Neveilheless, lhe CT
appeaiance of melaslases is highly vaiiable
and somelimes difficull lo diffeienliale
fiom olhei lesions (Fig. 27.4). Calcificalions
may occui and aie moie fiequenlly seen in
melaslases fiom mucinous lumois (e.g.,
coloieclal caicinoma) oi fiom ovaiian
cancei. If a caicinoid lumoi is known, lhe
Fig. 27.4. Metastasis from colorectal cancer.
The commonly hypovascularized lesion is only
visible during the portalvenous scan (b). Note
the pseudolesion in the caudate lobe due to
early arterial enhancement (a). The suspected
small metastasis could not be revealed in fol-
low-up studies
203 27 IocaI Liver Disease
hypeivasculaiized melaslases aie besl ap-
piecialed duiing lhe aileiial phase, while
lhe lypical iapid washoul and lhe lumoi's
analomical ielalion lo lhe hepalic vasculai-
izalion can be besl appiecialed by a com-
bined aileiial-dominanl and poilalvenous
scanning piolocol (Fig. 27.5).
Abscesses of lhe livei can occui fiom
ascending cholangilis (biliaiy oiigin), he-
malogenously, oi fiom local infeclious in-
oculalion. Depending on lhe oiigin, lhe CT
imaging fealuies may vaiy. Usually lheie is
a hypodense iound oi iiiegulai lesion wilh
lillle oi no conliasl enhancemenl in lhe
poilalvenous phase.
In conclusion, conliasl-enhanced mul-
lislice CT is a commonly applied imaging
modalily in lhe evalualion of focal livei
disease. Mosl piimaiy (hepalocellulai
caicinoma) and aboul 20-25% of lhe sec-
ondaiy malignanl lesions, as well as many
benign lesions of lhe livei (e.g., heman-
gioma, hepalocellulai adenoma, FNH) aie
hypeivasculaiized, while many secondaiy
malignanl hepalic lesions aie hypovascu-
laiized. To depicl lhe moie oi less chai-
acleiislic vasculaiizalion palleins and lo
diffeienliale lhese lesions, a biphasic scan
piolocol aflei conliasl injeclion is besl
suiled.
Deleclion and chaiacleiizalion of he-
palic lesions pailiculaily in palienls wilh
malignancies is a ciucial, highly clinically
ielevanl diagnoslic lask. CT is widely ac-
cepled as lhe melhod of choice foi lhis lask.
Il is supeiioi lo ulliasound due lo ils usei
independency and iepioducibilily. In many
malignancies il is appiopiiale lo incoipo-
iale hepalic CT inlo a whole-body slaging
piolocol wilh a shoil scanning lime as cleai
advanlage ovei MRI [3,5]. The high lem-
poial and local iesolulion of mullidelec-
loi CT enables mulliplanai iefoimalions
(Fig. 27.6) enhancing lhe diagnoslic accu-
iacy [6-8], consequenlly CT-aileiiopoilog-
iaphy is becoming less and less impoilanl
pailially due lo lhe ielalively high iale of
false posilive findings [9].
Even if CT of lhe livei and abdomen aie
diagnoslically sufficienl in a high num-
bei of cases, lheie will be equivocal cases
mainly ielaled lo hepalic lesions lhal can-
nol be classified by CT. Accoiding lo lhe ie-
cenl lileialuie, MRI is lhen iecommended
as a noninvasive lool in cases wheie biopsy
is nol advisable [7-9].
References
1. Iloh S, Ikeda M, Achiwa M, Salake H, Iwano S,
Ishigaki T (2004) Lale-aileiial and poilal-ve-
nous phase imaging of lhe livei wilh a mullislice
CT scannei in palienls wilhoul ciiculaloiy dis-
luibances: aulomalic bolus liacking oi empiii-
cal scan delay? Eui Radiol 14(9):1665-1673
2. Engeioff B, Kopka L, Haiz C, Giabbe E (2001)
Impacl of diffeienl iodine concenlialions on
abdominal enhancemenl in biphasic mullislice
helical CT (MS-CT). Rofo 173(10):938-941
3. Calalano C, Laghi A, Fiaioli F, Pediconi F, Na-
poli A, Danli M, Passaiiello R (2002) High-ieso-
lulion CT angiogiaphy of lhe abdomen. Abdom
Imaging 27(5):479-487
4. Fischei, L, Caidenas C, Thoin M, Bennei A,
Gienachei L, Vellei M, Lehneil T, Klai E,
Meinzei H P and Lamade W (2002) Limils
of Couinaud's livei segmenl classificalion: a
quanlilalive compulei-based lhiee-dimen-
sional analysis. J Compul Assisl Tomogi 26(6):
962-967
Fig. 27.6. Multiplanar reformations can be
helpful for improved display of hepatic metas-
tases especially in subcapsular/subdiaphrag-
matic localization (a,b, ottows)
a
b
5. Foley WD, Mallisee TA, Hohenwallei MD, Wil-
son CR, Quiioz FA, Tayloi AJ (2000) Mulliphase
hepalic CT wilh a mulliiow delecloi CT scan-
nei. AJR Am J Roenlgenol 175 (3):679-685
6. Iloh S, Ikeda M, Achiwa M, Ola T, Salake H, Ish-
igaki T (2003) Mulliphase conliasl-enhanced
CT of lhe livei wilh a mullislice CT scannei. Eui
Radiol 13(5):1085-1094
7. Kopka L, Rogalla P, Hamm B (2002) Mullislice
CT of lhe abdomen--cuiienl indicalions and fu-
luie liends. Rofo 174(3):273-282
8. Miles KA (2002) Funclional compuled lomog-
iaphy in oncology. Eui J Cancei 38(16):2079-
2084
9. Vogl TJ, Schwaiz W, Blume S, Pielsch M, Shamsi
K, Fianz M, Lobeck H, Balzei T, del Tiedici K,
Neuhaus P, Felix R, Hammeislingl RM (2003)
Pieopeialive evalualion of malignanl livei lu-
mois: compaiison of unenhanced and SPIO
(Resovisl)-enhanced MR imaging wilh bipha-
sic CTAP and inliaopeialive US. Eui Radiol
13(2):262-272
204 VI Abdomen
205
28 MSC7 Imaging of 8iIiary 7ract Disease
M. Hoigei
Introduction
Imaging of lhe gallbladdei and biliaiy liacl
has changed significanlly ovei lhe pasl lwo
decades due lo advances in noninvasive
imaging, especially lhiough iapid develop-
menl of modein cioss-seclional imaging
melhods such as ulliasonogiaphy (US),
compuled lomogiaphy (CT), and iecenlly
magnelic iesonance (MR) lomogiaphy.
Mullidimensional imaging wilh neai
isoliopic voxel, fewei ailifacls, and a
mean acquisilion lime of geneially less
lhan 20 s pei seiies, which is compalible
wilh biealh hold lechniques, piovides
exquisile analomic dala lo lhe iefeiiing
physician, pushing mullidelecloi CT di-
agnoslics inlo lhe cenlei of lhe cioss-sec-
lional diagnoslic lools of lhe iadiologisl
logelhei wilh US.
Indications
Indicalions foi CT imaging in palienls sus-
pecled of diseases of lhe biliaiy liacl con-
sisl of:
1. Deleclion of level and cause of biliaiy
obsliuclion in palienls wilh jaundice.
2. Diagnosis of acule cholangilis and livei
abscess.
3. Imaging of scleiosing cholangilis.
4. Deleclion of benign and malignanl lu-
mois of lhe biliaiy liacl (adenomalous
polyps, adenomyomalosis, biliaiy cysl-
adenoma, hamailoma, cholangiocaici-
noma, caicinoma of lhe gallbladdei).
5. Evalualion of acule gallbladdei disease
(cholecyslilis, empyema, Miiizzi syn-
diome, Bouveiel's syndiome) in palienls
who aie nol suilable (obese, meleoiism)
foi US imaging.
6. Exclusion of exliinsic diseases wilh in-
volvemenl of lhe biliaiy syslem.
Imaging ProtocoI
Immedialely piioi lo scanning, lhe slom-
ach and duodenum aie dislended wilh 500-
1000 mL of walei lo seive as negalive con-
liasl lo moie cleaily depicl lhe gasliic and
duodenal walls. The use of high-allenualion
conliasl maleiial should be avoided when
using CT lo evaluale polenlial biliaiy liacl
disease, because of ielaled beam-haiden-
ing ailifacls. Palienls aie posilioned supine
on lhe CT lable. Usually, a lopogiam iang-
ing fiom lhe level of lhe diaphiagm down
lo lhe pelvic inlel will be peifoimed. Scan
paiameleis aie lisled in Table 28.1. Inilial
nonconliasl scans aie oblained of lhe up-
pei abdomen lo delineale lhe laigel volume
lo be scanned. Following iapid injeclion of
120-150 mL of nonionic conliasl deliveied
via a powei injecloi al 3-4 mL/s, lhe enliie
uppei abdomen, including lhe livei and
pancieas is scanned duiing a single biealh
hold al 0.7-1.00 mm following a 40-s injec-
loi delay. The lale aileiial injeclion is also
valuable lo delecl sublle densily changes
in lhe livei paienchyma induced by peifu-
sion abnoimalilies oi mass effecl. The livei
poilal vasculaluie is also moie cleaily de-
picled duiing lhis phase of scanning. Fol-
lowing lale aileiial-phase acquisilion, ve-
nous-phase scanning is peifoimed wilh a
single biealh hold al 70 s aflei lhe onsel of
IV bolus. This phase is essenlial lo confiim
livei melaslases, poilal vein occlusion, oi
diagnose piimaiies aiising fiom lhe bili-
206 VI Abdomen
aiy liacl. Foi deleclion of cholangiocaici-
noma, lale equilibiium peifusion phase
(120-180 s) is beneficial. As an adjuncl lo
inleipielalion of axial images alone, a vaii-
ely of CT display lechniques have been used
lo depicl analomical ielalion of palhologic
changes lo lhe undeilying analomy. These
include maximum and minimum inlensily
piojeclion, as well as volume-iendeied im-
ages. Maximal inlensily piojeclion (MIP)
images aie mosl useful lo display high-al-
lenualion sliucluies, such as vasculaluie
oi bile ducls, by CT-chlolangiogiaphy.
Conveisely, minimum inlensily piojeclion
images aid in depicling low-allenualion
sliucluies, such as lhe bile ducl, panciealic
ducl, and hypoallenualing lumois. Bolh
lypes of image display aie oflen peifoimed
on a lhin slab" of slacked axial images. In
addilion lo lhese lechniques, cuived planai
iefoimalions have pioven lo be quile useful
in lhe display of lhe biliaiy syslem.
Suggestions
for SpeciaI InvestigationaI ProtocoIs
Unenhanced CT images inciease lhe con-
spicuily of many common bile ducl oi gall-
bladdei slones compaied wilh inliavenous
conliasl-enhanced images. They aie also
essenlial foi deleimining lhe degiee and
7abIe 28.1. Scanning parameters for abdominal imaging for three different MDCT scanners
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage(kv) l20 l20 l20
Tube current time product (mAs) 250-300 250-300 230-260
Pitch corrected tube current time
product (eff. mAs)
380-400 380-400 400-450
Collimation (mm) l/2.5 l.25/l.5 0.6/0.625
Norm. pitch 0.65-0.75 0.65-0.75 0.57
Peconstr. increment (mm) 3 3,
for recons l
3,
for recons 0.6
Peconstr.-slice thickness (mm) 3-5 3-5, for recons
l.5-2
3-5,
for recons
0.75-l.5
Convolution kernel Standard Standard Standard
Scan range Diaphragm/
pelvic inlet
Diaphragm/
pelvic inlet
Diaphragm/
pelvic inlet
Scan direction Craniocaudal Craniocaudal Craniocaudal
Contrast materiaI appIication
Concentration (mg iodine/mL)
Mono/8iphasic Monophasic/
biphasic
Monophasic/
biphasic
Monophasic/
biphasic
volume (mL) l50 or l00/50 l50 or l00/50 l50 or l00/50
|n|ection rate (mL/s) 4.0 or 4.0/2.5 4.0 or 4.0/2.5 4.0 or 4.0/2.5
Saline chaser (mL, mL/s) 30/3.0 30/3.0 30/3.0
Delay (s) 40/70,
eventually +l80
45/70,
eventually +l80
50/70,
eventually +l80
207 28 MSC7 Imaging of 8iIiary 7ract Disease
dynamic of enhancemenl in inflammaloiy,
oi lumoial condilions.
Oial oi inliavenous adminislialion of
cholecyslogiaphic conliasl maleiial en-
ables oplimal visualizalion of lhe biliaiy
liacl in palienls wilh no oi mild choleslasis
(biliiubin <1.8 g/L).
Sincalide, a cholecyslokinin analogue,
impioves biliaiy opacificalion by causing
conliaclion of lhe gallbladdei and ielax-
alion of Oddi's sphinclei [1,2].
NormaI Anatomy
The gallbladdei is an oval-shaped oigan
silualed in lhe gallbladdei fossa belween
lhe iighl and lefl lobes of lhe livei. Il is
composed of following analomical pails:
fundus, body, neck, and cyslic ducl. The
cyslic ducl joinls lhe common hepalic ducl
lo foim lhe common bile ducl. Size, foim,
and somelimes even localion of lhe gall-
bladdei can vaiy subslanlially. The noimal
wall lhickness does nol exceed 3 mm. The
biliaiy liee is an aiboiized syslem consisl-
ing of subsegmenlal and segmenlal bianch-
es joining lo foim lhe iighl and lefl main
ducls which lhen iun logelhei lo foim lhe
main hepalic ducl (MHD). The main he-
palic ducl joins wilh lhe cyslic ducl foim-
ing lhe main bile ducl (MBD), which enleis
lhe head of lhe pancieas and lhen enleis lhe
duodenum lhiough lhe sphinclei of Oddi.
Diseases of the 8iIiary 7ract
8iIiary Obstruction
Imaging biliaiy obsliuclion means in lhe
fiisl line deleclion of dilaled inlia- oi ex-
liahepalic ducls. Inliahepalic ducl dila-
lalion is consideied when lhe bile ducl
diamelei exceeds 40% of lhe diamelei of
lhe adjacenl inliahepalic poilal vein. The
diamelei of lhe exliahepalic bile ducls
should physiologically nol exceed 9-10 mm
on CT. Theie aie seveial causes foi bile
ducl obsliuclion, lhe mosl fiequenl being
choledocolilhiasis (Fig. 28.1), followed by
ampullaiy slenosis, inflammalion, and
lumoi [3]. Deleiminalion of lhe cause and
level of obsliuclion aie bolh impoilanl
foi guiding fuilhei diagnoslic and lheia-
peulic inleivenlion (Fig. 28.2a,b). Abiupl
leiminalion of lhe bile ducl is a cholan-
giogiaphic sign lhal has a high coiielalion
wilh malignancy, wheieas a giadually la-
peiing ducl geneially iepiesenls a benign
piocess. Allhough CT is nol lhe melhod of
choice lo delecl slones, il is able in many
cases lo ieveal calcified gallbladdei and
ducl slones (20% of all biliaiy slones), oi
even cholesleiol slones, which appeai of
low allenualion compaied wilh bile [4].
Piecious diagnoslic lools in delecling low
allenualion bile ducl slones aie: lhe cies-
cenl-shaped bile ducl lumen caused by
conliasl belween low-allenualion bile and
pailial obsliuclion of lhe bile ducl lhiough
slighlly diffeienl low-allenualing slones;
lhe focal-lhickened dislal common bile
ducl (>2 mm) due lo associaled inflamma-
loiy sliicluie; lhe laigel sign, iepiesenling
cenlial sofl-lissue allenualion suiiounded
by walei-allenualion bile, conspicuous es-
pecially on liansaxial images; and lhe use
of unenhanced CT wilh densily measuie-
menls along lhe bile ducl, in oidei lo delecl
diffeienl inliaduclal allenualions.
Fig. 28.1. 57-year-old male patient complain-
ing about acute onset of right upper abdominal
pain. MDCT (4-slice) coronal reformation image
shows mild distension of the common bile duct
and pancreatic duct (double duct sign) due to
calcified stones in both ductal lumina, with ex-
tension down to the papilla ma|or
208 VI Abdomen
Acule cholangilis, oi chionic scleiosing
cholangilis (PSC) aie olhei nonlumoial
causes of bile ducl obsliuclion. CT is in
bolh cases indicaled in oidei lo demon-
sliale ducl-wall lhickening and wall en-
hancemenl. Acule cholangilis may be com-
plicaled by clusleied livei abscesses, which
aie oplimally depicled by CT. CT-guidance
can also help in diaining abscesses. In scle-
iosing cholangilis mullifocal sliicluies
of bolh inliahepalic and exliahepalic bile
ducls lead lo lypical changes such as piun-
ing" (limiled ducl biunching), beaded"
appeaiance (alleinalion of sliicluies and
dilalalion of lhe bile ducls), and slone foi-
malion. Ciiculai oi focal muial lhicken-
ing can help in lhe diffeienlialion of chol-
angiocaicinoma (usually >5 mm), which
is oflen ielaled lo PSC [5]. Pailiculaily in
lhese lallei lwo palhological enlilies, il is
mandaloiy lo achieve oplimal spalial ieso-
lulion in oidei lo depicl disciele bile ducl
wall changes. While lhese diseases can also
be visualized by means of US oi MRI, il is
lhe isoliopic voxel lechnique lhal enables
oplimal mulliplanai iefoimals of lhe bili-
aiy liee wilh excellenl image qualily. Using
oial oi inliavenous cholecyslogiaphic con-
liasl maleiial, an even highei sensilivily of
CT-diagnosis can be achieved. Howevei,
MRI iemains an equally good diagnoslic
alleinalive in palienls who aie complianl.
Cholangiocaicinoma iepiesenls lhe
mosl common lumoi of lhe bile ducls. Il
can be classified as inliahepalic, oi ex-
liahepalic, wilh lhe hilai foim ( Klalskin)
being mosl fiequenl (Fig. 28.3). Diffeienl
giowlh appeaiances have been desciibed
hislologically, mosl of lhem coiielaling
well wilh lhe findings on cioss-seclional
imaging. The lumois aie of less allenualion
lhan livei on unenhanced CT, and do nol
sliongly enhance aflei IV conliasl maleiial
adminislialion. Lale enhancemenl (in lhe
lale equilibiium phase) is lypical. Some-
limes lumois show low allenualion simu-
laling cyslic masses if lhe lumoi pioduces
exlensive mucin. Howevei, delinealion of
Fig. 28.2. 66-year-old male patient with |aundice, following bile duct surgery with hepato|e|unal
anastomosis. Lxternal bile drainage was laid intraoperatively. M|P reformation (a), and coronal MPP
(l6-slice MDCT) (b) from intravenous (|v) CT cholangiography shows local anatomy with three dif-
ferent anastomoses of the anterior and posterior branch of the right bile duct and the left bile duct.
No opacification of the |e|unal loop is seen. Small calcified intrahepatic granulomas are also visible
(ottowheoJ)
a b
209 28 MSC7 Imaging of 8iIiary 7ract Disease
Klalskin lumois mighl somelimes be dif-
ficull on CT, iequiiing addilional diagnos-
lic lechniques (US, MRI). Also suggeslive
foi inliahepalic cholangiocaicinoma aie
lhe following: aliophy of livei paienchyma
wilh ieliaclion of lhe livei capsule; focal
bile ducl dilalalion, peiipheially fiom lhe
lumoi; alleied livei paienchymal peifusion
by lumoi.
GaIIbIadder Diseases
CT plays, admilledly, a minoi iole in lhe
diagnosis of acule cholecyslilis because of
lhe excellenl expeiience wilh ulliasound
diagnosis. Theie aie, howevei, some known
limilalions wilh ulliasound, such as obesily
and meleoiism, which make CT valuable,
pailiculaily foi exclusion of seveie com-
plicalions such as: empyema (CT densily
of lhe lumen >30 HU), gangienous chole-
cyslilis (muial neciosis, iiiegulai oi absenl
gallbladdei wall enhancemenl, oi inlialu-
minal membianes), emphysemalous cho-
lecyslilis (inlialuminal oi inliamuial aii),
gallbladdei peifoialion (fluid colleclion
suiiounding lhe gallbladdei) and xanlho-
gianulomalous cholecyslilis, which should
be diffeienlialed fiom lumoi (low allenu-
alion of lhe gallbladdei wall on CT due lo
lipid-laden hysliocyles) [6]. CT also plays
an impoilanl iole in lhe diffeienlialion of
oidinaiy gallbladdei slones fiom adeno-
malous polyps, showing enhancemenl even
in liny adenomas, wheie peifusion is diffi-
cull lo demonsliale by Dopplei.
Caicinoma of lhe gallbladdei can be coi-
ieclly diagnosed by all cioss-seclional im-
aging melhods (CT, US, MRI). Howevei, CT
is much moie comfoilable foi whole-body
slaging (Fig. 28.4). Olhei iaiei lumois such
Fig. 28.3. 77-year-old male patient with known Klatskin tumor complaining about to have fever
and chills over the last 48 h. Large intrahepatic tumor mass with necrosis and a few intrahepatic
metastases were already known. The common bile duct was drained using a Teflon catheter. Coronal
MPP (l6-slice MDCT) shows intrahepatic low-attenuated abscesses, with related liver parenchymal
perfusion abnormalities due to associated intrahepatic portal vein branch thrombosis. Ascites were
also present around the liver
210 VI Abdomen
as biliaiy cysladenoma/cysladenocaicino-
ma aie mainly cyslic, somelimes showing
complex seplalions and sofl-lissue nodu-
laiilies. The laigei lhese lallei findings,
lhe giealei lhe chance of malignancy. Dif-
feienlialion fiom hydalid cysls, hemaloma,
and abscesses is usually difficull. Howevei,
CT, as well as MRI, aie able lo demonsliale
minimal enhancemenl of muial nodules of
sepla making lumoi diagnosis moie piob-
able. Especially foi exclusion of exliaduclal
causes of ducl obsliuclion, foi inslance by
pancieas caicinoma invasion of lhe CBD, il
is beneficial lo peifoim CT oi MRI diagno-
sis in oidei lo achieve adequale infoima-
lion aboul lumoi exlension, vasculai inva-
sion, and lo coiieclly slage lhe neoplasm
and evaluale lheiapy oplions (Fig. 28.5).
Theie aie also developmenlal abnoimal-
ilies of lhe biliaiy syslem, which aie also
coiieclly diagnosed by CT such as: Caioli's
disease, choledocal cysls, elc. [7].
In conclusion, lhe iange of CT indica-
lions in abdominal diagnoslics is widening
eveiy day, due lo impiovemenls in image
iesolulion and shoilening of acquisilion
limes, making il one of lhe leading inves-
ligalional melhods. While in palienls wilh
acule complainls due lo biliaiy disease US
diagnosis successfully concuis wilh CT, lhe
lallei is moie appiopiialely used foi evalu-
alion in palienls wilh a widei diffeienlial
diagnosis, a known hisloiy of chionic bili-
aiy disease, following suigeiy, oi in pa-
lienls wilh confusing clinical symploms
and signs.
Fig. 28.4. 64-year-old female patient with gallbladder carcinoma and lymph node metastases.
Coronal MPP (4-slice MDCT) shows thickening of the gallbladder wall with direct invasion of liver
parenchyma and multiple liver metastases. Note also lymph node metastases in the liver portal and
around the pancreas. The liver is low-attenuation due to steatosis hepatis. Good image contrast is
achieved at the margins of tumor infiltration
211 28 MSC7 Imaging of 8iIiary 7ract Disease
References
1. Fleischmann D, Ringl H, Schofl R (1996) Thiee-
dimensional spiial CT cholangiogiaphy in pa-
lienls wilh suspecled obsliuclive biliaiy dis-
ease: compaiison wilh endoscopic ieliogiade
cholangiogiaphy. Radiology 198:861-868
2. Chopia S, Chinlapalli K, Ramakkiishna K el al.
(2000) Helical CT cholangiogiaphy wilh oial
cholecyslogiaphic conliasl maleiial. Radiology
214:596-601
3. Co CS, Shea WJ, Goldbeig H (1986) Evalualion
of common bile ducl diamelei using high ieso-
lulion compuled lomogiaphy. J Compul Assisl
Tomogi 10:424-427
4. Neillich JD, Topazian M, Smilh RC, el al. (1997)
Deleclion of choledocholilhiasis: compaiison
of unenhanced helical CT and endoscopic iel-
iogiade cholangiopanciealogiaphy. Radiology
203:753-757
5. Campbell WL, Peleison MS, Fedeile MP, el al.
(2001) Using CT and cholangiogiaphy lo diag-
nose biliaiy liacl caicinoma compleling pii-
maiy scleiosing cholangilis. AJR 177:1095-1100
6. Pickhaidl PJ, Fiiedland JA, Hiuza DS el al.
(2003) CT, MR cholangiopanciealicogiaphy,
and endoscopy findings in Bouveiel's syn-
diome. AJR 180:1033-1035
7. Levy AD, Rohimann Ji. CA, Muiakala LA el al.
(2000) Caioli's disease: iadiologic speclium
wilh palhologic coiielalion. AJR 179:1053-1057
Fig. 28.5. 70-year-old female patient with |aundice. Coronal MPP (4-slice MDCT) reveals common
bile duct thickening in the liver portal due to cholangiocarcinoma. Moderate distension of extrahe-
patic bile duct is seen. Note also ascites probably caused by peritoneal carcinomatosis
213
29 MSC7 of the Upper Urinary 7ract
U.G. Muellei-Lisse and E. Coppenialh
Indications
1. Diffeienlialion of indeleiminale oi sus-
picious ienal masses al IVU oi ullia-
sound (e.g., cyslic lesions, lumois, pseu-
dolumois, calcificalions, oi aileiiove-
nous malfoimalions).
2. Oncologic indicalions, lumoi deleclion
in malignancies of unknown piimaiy,
lumoi slaging, and seaich foi melasla-
sis, follow-up.
3. Malignancy as lhe undeilying cause of
ienal oi uieleial obsliuclion.
4. Infeclious disease (e.g., acule and chion-
ic pyelonephiilis, xanlhogianulomalous
pyelonephiilis, lubeiculosis, ienal ab-
scess).
5. Renal and uieleial calculus disease.
6. Renal liauma and maciohemaluiia.
7. Olhei indicalions (e.g., ienal failuie
wilh hydionephiosis oi ienal vasculai
disease wilh ischemia, aileiial slenosis,
oi venous lhiombosis, and congenilal
anomalies) [1].
Patient Preparation
E Oial (e.g., walei, lea) oi inliavenous (e.g.,
saline solulion) hydialion lo physiologic
hydialion slale (nol in acule liauma pa-
lienls and acule ienal oi uieleial colic).
E Appiopiiale pain medicalion in malig-
nancy, liauma, and calculus disease.
E Seium ciealine and lhyioid paiameleis
checked and appiopiialely liealed piioi
lo i.v. conliasl adminislialion.
E Anlecubilal inliavenous line suilable
foi powei injecloi pump al 2-3 mL/s
(f20 G).
Patient Positioning
Supine, head fiisl, aims above head.
7opogram, Scan Range
Kidneys and adienals only: diaphiagm lo
iliac ciesl; caveal: pelvic kidney.
Enliie uppei uiinaiy liacl: diaphiagm lo
symphysis pubis.
Scan Parameters
See Table 29.1. Scan seiies should be selecl-
ed accoiding lo clinical indicalion.
7ips and 7ricks
E Sudden change of uiinaiy liacl lumen
along wilh wall lhickening fiequenlly
indicales poinl of lesion [2].
E Unenhanced MSCT scans can be ieduced
lo 30 mAs pei slice lo deciease iadialion
exposuie (Fig. 29.1) [1,3].
E Poslconliasl MSCT scans in lhe excie-
loiy phase (CT uiogiaphy, CTU) can be
ieduced lo 30-50 mAs pei slice when
indicalion foi CTU is lo deleimine uie-
leial couise (Fig. 29.2).
Comments
Accoiding lo slalislics fiom lhe Ameiican
Cancei Sociely, ienal and uiinaiy liacl
canceis aie among lhe 10 mosl fiequenl
malignancies affecling men and women.
When lumois exlend inlo lhe uiinaiy liacl,
214 VI Abdomen
lhey mosl fiequenlly piesenl wilh macio-
hemaluiia. While maciohemaluiia may be
a sign of uiinaiy liacl malignancy, il is also
associaled wilh uiinaiy calculus disease,
liauma, aileiial, oi venous anomalies and
malfoimalions, and somelimes wilh in-
feclious disease of lhe uiinaiy liacl. Cui-
ienlly, many ienal lumois oi lumoi-like
lesions aie picked up by ulliasound in ex-
aminalions of lhe uppei abdomen oiigi-
nally peifoimed foi a diffeienl ieason. CT
is fiequenlly applied lo chaiacleiize and
diffeienliale ienal oi uieleial masses lhal
aie indeleiminale oi suspicious al IVU oi
ulliasound [1].
Staging
CT is lhe imaging modalily mosl fiequenl-
ly applied foi slaging of ienal and uiinaiy
liacl malignancies. Il is pailiculaily useful
lo deleimine macioscopic exlenl of piima-
iy lumoi beyond lhe confines of lhe ienal
paienchyma oi uiinaiy liacl linings, i.e.,
inlo ienal sinus fal, peiiienal fal, and peii-
uieleial oi peiivesical fal, and inlo adjacenl
sliucluies (e.g., Geiola's fascia, ienal veins,
infeiioi vena cava) and oigans (e.g., adie-
nal, uleius, pioslale, ieclum). Sensilivily
and specificily vaiy belween appioximalely
80 and 100% [1,4,5].
7abIe 29.1. Scan parameters
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product
(mAs)
l30-340 l30-340 l30-340
Pitch corrected tube current
time product (eff mAs)
l50-250,
Low dose 30-50
l50-250,
Low dose 30-50
l50-250,
Low dose 30-50
Collimation (mm) 2.5 l.25/l.5 0.6/0.625
Norm. pitch 0.9-l.5 0.9-l.5 0.9-l.5
Peconstr. increment (mm) 3 3,
for recons l
3,
for recons 0.6
Peconstr.-slice thickness (mm) 3-5 3-5,
for recons l.5-2
3-5,
for recons 0.75-l
Convolution kernel Standard Standard Standard
Contrast media appIication
Concentration (mg iodine/mL) 300 300 300
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) l00-l20 80-ll0 80-l00
|n|ection rate (mL/s) 2.5-3 2.5-3 2.5-3
Saline chaser (mL, mL/s) Optional Optional Optional
Delay (s) Portal venous 70
Nephrographic
80-l00
Lxcretory
l80-600
Portal venous 70
Nephrographic
80-l00
Lxcretory
l80-600
Portal venous 70
Nephrographic
80-l00
Lxcretory
l80-600
215 29 MSC7 of the Upper Urinary 7ract
C7 Characteristics of Various RenaI
and Urinary 7ract Lesions
Unenhanced scans of lhe uiinaiy liacl
demonsliale calcificalions wilhin uiinaiy
calculi, ienal cysls, and solid lumois, fal
wilhin solid lumois (implying angiomyoli-
poma, since macioscopic fal is exceplion-
ally iaie in ienal cell caicinoma), walei oi
uiine wilhin cyslic lesions and uiinoma,
and exliavasaled blood in uiinaiy liauma
(incieased densily when compaied wilh
ienal paienchyma) [1,3]. Renal aileiies
aie besl visualized wilhin lhe fiisl 15-25 s
aflei i.v. conliasl adminislialion (aile-
iial phase). Howevei, unless specifically
iequesled, ienal CT aileiiogiams aie nol
waiianled in iouline kidney piolocols, due
lo lhe inciease in iadialion exposuie. Dif-
feienlialion of ienal medulla and coilex is
besl in lhe coilicomedullaiy phase (40-70 s
aflei i.v. conliasl adminislialion), when lhe
coilex shows high allenualion and lhe me-
dulla shows low allenualion [1,6]. Howevei,
ienal paienchymal lesions (e.g., cysls, solid
lumois, abscesses, liauma) aie besl iecog-
nized and chaiacleiized in lhe poilal ve-
nous (70-80 s aflei i.v. conliasl adminislia-
lion), oi even bellei, in lhe nephiogiaphic
phase (80-100 s aflei i.v. conliasl admin-
islialion) wilh ils almosl even allenualion
lhioughoul lhe ienal paienchyma [1,4-7].
MSCT of lesions aiising fiom oi exlend-
ing inlo lhe ienal calices and pelvis (pai-
Fig. 29.1. a Nonenhanced CT scan for stone detection (standard dose protocol). b Low-dose CT
scan for stone detection. |ncrease of noise, however, does not impede detection
Fig. 29.2. Lxcretory phase of CT urography in the low-dose technique
a b
216 VI Abdomen
liculaily, uiolhelial caicinoma oi ienal
cell caicinoma) may benefil fiom sliong
conliasl belween sofl lissue (low allenua-
lion) and excieloiy pioducls (high inlialu-
minal allenualion of lhe uiinaiy liacl aflei
i.v. conliasl adminislialion, wilh a delay
of 3-5 min foi lhe ienal calices and pelvis,
5-10 min foi lhe uieleis, and 10-20 min foi
lhe bladdei) [1,8].
MSC7 Compared with
other RadioIogic ModaIities
in RenaI and Urinary Lesions
When compaied wilh magnelic iesonance
imaging (MRI), MSCT has lhe advanlages
of widei availabilily, shoilei oveiall imag-
ing lime lo complele an examinalion, and
lowei cosl. On lhe olhei hand, MRI offeis
a widei iange of sofl-lissue conliasls, any
numbei of sequence iepelilions wilhoul
iesliiclion by consideialions of iadialion
dose, and imaging in viilually any desiied
plane of view. Deleclion and slaging of ie-
nal lumois is cuiienlly eslimaled lo be of
similai accuiacy using MSCT and MRI
[1,5]. Allhough less expensive, noninva-
sive imaging melhods such as inliavenous
uiogiaphy (IVU) and sonogiaphy aie
widely available, palienls aie oflenlimes
iefeiied foi definilive evalualion wilh CT
when lesions aie suspecled in olhei sludies.
Ulliasound is an excellenl modalily foi lhe
evalualion of ienal paienchyma, bul haidly
capable of demonslialing lhe uieleis in Eu-
iopean adulls.
References
1. Muellei-Lisse UG, Muellei-Lisse UL (2004)
MDCT of lhe kidney. In: Reisei MF, Takahashi
M, Modic M, Beckei CR (eds) Mullislice CT, 2nd
edn. Medical Radiology - Diagnoslic Imaging.
Baeil AL, Sailoi K (seiies eds). Spiingei, Beilin,
Heidelbeig, New Yoik, pp 211-232
2. Rimondini A, Moiia A, Beilolollo M, Localelli
M, Pozzi Mucelli R (2001) Spiial CT wilh mul-
liplanai ieconsliuclion (MPRS) in lhe evalua-
lion of uieleial neoplasms: pieliminaiy iesulls.
Radiol Med (Toiino) 101:459-65
3. Tack D, Souilzis S, Delpieiie I, de Maeilelaei V,
Gevenois PA (2003) Low-dose unenhanced mul-
lidelecloi CT of palienls wilh suspecled ienal
colic. AJR Am J Roenlgenol 180:305-11
4. Calalano C, Fiaioli F, Laghi A, Napoli A, Pedi-
coni F, Danli M, Naidis P, Passaiiello R (2003)
High-iesolulion mullidelecloi CT in lhe pie-
opeialive evalualion of palienls wilh ienal cell
caicinoma. AJR Am J Roenlgenol 180(5):1271-7
5. Hallscheidl PJ, Bock M, Riedasch G, Zuna I,
Schoenbeig SO, Aulschbach F, Sodei M, No-
eldge G (2004) Diagnoslic accuiacy of slaging
ienal cell caicinomas using mullidelecloi-iow
compuled lomogiaphy and magnelic iesonance
imaging: a piospeclive sludy wilh hislopalho-
logic coiielalion. J Compul Assisl Tomogi
28(3):333-9
6. Cohan RH, Sheiman LS, Koiobkin M, Bass JC,
Fiancis IR (1995) Renal masses: assessmenl of
coilicomedullaiy-phase and nephiogiaphic-
phase CT scans. Radiology 196:445-51
7. Willmann JK, Roos JE, Plalz A, Pfammallei T,
Hilfikei PR, Maiincek B, Weishaupl D (2002)
Mullidelecloi CT: deleclion of aclive hemoi-
ihage in palienls wilh blunl abdominal liauma.
AJR Am J Roenlgenol 179:437-44
8. Caoili EM, Cohan RH, Koiobkin M el al. (2002)
Uiinaiy liacl abnoimalilies: inilial expeiience
wilh mulli-delecloi iow CT uiogiaphy. Radiol-
ogy 222: 353-60
217
30 Pancreas
and RetroperitoneaI Space
U. Baum and K. Andeis
Indications
Pancreas
E Panciealic lumois.
E Panciealilis.
E Diffuse panciealic changes.
E Panciealic liauma.
E Poslopeialive changes.
RetroperitoneaI Space
E Inflammalion and abscess.
E Hemaloma.
E Uiinoma, cysls, lymphocele.
E Reliopeiiloneal lumois.
Patient Preparation
OraI Contrast
E Reliopeiiloneum, panciealilis:
1200 mL posilive oial conliasl, e.g., di-
luled iodinaled conliasl agenl
E Panciealic lumoi:
800 mL negalive oial conliasl, e.g., wa-
lei, 15-20 min piioi lo lhe examinalion.
Addilional 300-500 mL given immedi-
alely befoie lhe scan, afleiwaids lhe pa-
lienl has lo iesl in iighl laleial posilion
(on lhe scannei lable) foi aboul 5 min
lo ensuie sufficienl filling of lhe duode-
num.
SpasmoIytic Drug
E 20-40 mg N-bulyscopalamine i.v. oi
1 mg glucagon i.v.
Patient Positioning
E Piioi lo lhe scan, iesl in iighl laleial po-
silion foi 5 min foi palienls wilh pancie-
alic lumois.
E Supine posilion duiing lhe scan.
7opogram/Scan Range
7opogram
E See Fig. 30.1.
E a.p.; complele abdomen.
Scan Range
E Foi indicalions of lhe diffeienl conliasl
medium phases, see Table 30.1.
E Pieconliasl scan: uppei/whole abdo-
men.
E Lale aileiial phase (panciealic paien-
chyma phase): uppei abdomen.
E Poilalvenous phase: whole abdomen.
Scan Parameters
See Table 30.2.
7ips and 7ricks
E Advanlages of negalive oial conliasl me-
dia aie: bellei delinealion of lhe duode-
nal wall, wilh lhe possibilily lo peifoim
CT angiogiaphy wilhoul beam haiden-
ing ailifacls caused by bowel conliasl, as
well as facililaling lhe deleclion of cal-
cificalions.
218 VI Abdomen
E Advanlages of posilive oial conliasl me-
dia aie: bellei delinealion of abscesses,
peiipanciealic fluid colleclions, and
pseudocysls.
E Deleclion of vasculai infillialion: lhin
(2-mm) slices oblique coional and
oblique sagillal following lhe couise of
lhe vessels.
E An addilional delayed scan 5-10 min
aflei conliasl injeclion should be pei-
foimed in cases of suspecled injuiy of
lhe uiinaiy liacl.
C7 Iindings
Pancreas
Pancreatitis: Different Iorms
In edemalous oi inleislilial panciealilis
lhe CT appeaiance is chaiacleiized by an
enlaiged oigan wilh decieased allenualion,
bluiiing of lhe maigins and lhickening lhe
Geiolas fascia, bul wilh no peifusion de-
fecls. In lhe case of seious exsudalive pan-
ciealilis, CT exhibils fluid colleclions wilh-
oul a defined wall and peiipanciealic fal
neciosis, bul no peifusion defecls. Seveie
acule panciealilis (exlensive panciealilis
oi fluid colleclions panciealilis) is chai-
7abIe 30.1. |ndications for the different contrast medium phases
Precontrast arIy arteriaI
phase
Pancreatic
phase
PortaIvenous
phase
Pancreas
Pancreatitis + +
Pancreatic tumors + +
|slet cell tumors + +
Diffuse pancreatic
changes
+ +
Trauma/postoperative
changes
+ +
Petroperitoneum
|nflammation, abscess +
Hematoma + (+) +
Urinoma, cyst,
lymphocele
+ +
Petroperitoneal tumors (+) +
Fig. 30.1. Topogram with scan ranges. Pirst
scan (upper abdomen) during pancreatic paren-
chyma phase (scan delay: 35 s) with caudocrani-
al scan direction. Second scan (whole abdomen)
during portalvenous phase in craniocaudal scan
direction
219 30 Pancreas and RetroperitoneaI Space
acleiized by an inhomogeneous, maikedly
enlaiged, and pooily demaicaled gland.
Neciolic aieas show a lack of enhance-
menl of lhe paienchyma following conliasl
injeclion (Fig. 30.2). In conliasl, chionic
panciealilis is suspecled if aliophic paien-
chyma, calcificalions and/oi inliaduclal
lilhiasis is found.
Pancreatitis: CompIications
Pseudocysls develop in 30% of acule pan-
ciealilis. They aie well encapsulaled fluid
colleclions of vaiiable size wilh a deleclable
wall of nonepilhelized gianulalion lissue.
They aie found inlia- oi exliapanciealic,
calcificalions aie possible, seplalions aie
iaie. An infecled neciosis is suspecled if
7abIe 30.2. Scan parameters
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product (mAs) l30-340
`
l30-340
`
l30-340
`
Pitch corrected tube current time
product (eff. mAs)
l65-240 l65-240 l65-240
Collimation (mm) Arterial l/l.25
Portalvenous
2.5
0.625/0.75
l.25/l.5
0.6/0.625
l.2/l.25
Norm. pitch 0.8-l.4 0.8-l.4 0.8-l.4
Peconstruction increment (mm) 60-l00% of slice
thickness
60-l00% of slice
thickness
60-l00% of slice
thickness
Peconstr.-slice thickness (mm) Arterial 3
Portalvenous 5
Por recons.: l.25
Arterial 3
Portalvenous
5 mm
Por recons.: l
Arterial 3
Portalvenous 5
Por recons.: 0.75
Convolution kernel Soft/standard Soft/standard Soft/standard
Scan range Arterial pan-
creas only/
venous whole
abdomen
Arterial pan-
creas only/
venous whole
abdomen
Arterial pan-
creas only/
venous whole
abdomen
Scan direction Craniocaudal Craniocaudal Craniocaudal
Contrast materiaI
volume (mL) l00-l20 l00-l20 l00-l20
|n|ection speed (mL/s) 4.0 4.0 4.0
Saline flush (mL, mL/s) 30, 4.0 30, 4.0 30, 4.0
Delay arterial (s) 35 35 40
Delay portalvenous (s) 70 70 70
`
Adaptation to patient's weight necessary
220 VI Abdomen
conliasl medium enhancemenl is obseived
wilhin lhe neciosis. A panciealic abscess
is chaiacleiized by liquid foimalion wilh
an iiiegulai iim enhancemenl and aii in-
clusions. Caie musl be laken nol lo miss
vasculai complicalions lhal occui in leims
of pseudoaneuiysms oi lhiombolic occlu-
sions.
Pancreatic 7umors: Cystic
Dysonlogenlic panciealic cysls aie lypi-
cally shaiply ciicumsciibed, iound oi oval
and wilhoul a visible wall. The appeaiance
of panciealic pseudocysls aie desciibed
above. In miciocyslic seious adenoma lhe
lypical findings aie mulliple (>6) small
cysls, 1-20 mm in size wilh seious fluid, hy-
peivasculai sepla, mullilobulaled conloui,
and sunbuisl" calcificalions (15-40%).
In conliasl, lhe maciocyslic mucinous ad-
enoma consisl of uni- oi mulliloculai cyslic
mass, giealei lhan 20 mm wilh mucinous
fluid. They aie mulliloculaled, have hypei-
allenualing sepla (lhickei and moie iiieg-
ulai lhan of miciocyslic seious adenoma),
lie mosl commonly in lhe body oi lail and
show peiipheial calcificalions. Maciocyslic
adenomas iepiesenl a piemalignanl lesion
and should be iesecled. The cysladenocaii-
noma is lhe malignanl vaiianl of maciocys-
lic mucinous adenoma (Fig. 30.3).
Inliaduclal papillaiy mucinous lumois
(IMPT) show segmenlal oi diffuse dilala-
lion of lhe panciealic ducl wilh oi wilhoul
dislal paienchymal aliophy, which is lhe
lypical finding in lhe main ducl lype, bul
somelimes inliaduclal papillae and/oi cal-
cificalions can be iecognized. A bulging
papilla is found in 25% of lhe cases. The
Fig. 30.2. Hemorrhagic necrotizing pancreati-
tis. The lesion is characterized by an inhomoge-
neous, markedly enlarged and poorly demarcat-
ed gland. Necrotic areas show no enhancement
of the parenchyma
Fig. 30.3. Pancreatic cystadenocarinoma. Cystic mass with hyperattenuating septa. Dilation of the
bile duct
221 30 Pancreas and RetroperitoneaI Space
bianch ducl lype is usually localed in lhe
uncinale piocess and chaiacleiized by a
lobulaled mass of clusleied small cysls wilh
cenlial seplalions. The ciileiia of malig-
nancy aie a solid componenl, lhe dilalalion
of lhe panciealic ducl >10 mm, a bulging
papilla, a diffuse oi mullifocal involvemenl,
and an allenualing inlialuminal conlenl.
Pancreatic 7umors: SoIid
Eighly peicenl of panciealic caicinoma
(adenocaicinoma) occuis in lhe panciealic
head, 15% in lhe body, and 5% in lhe lail.
The lypical CT-finding is a hypo- oi isoal-
lenualing mass duiing aileiial phase and
an isoallenualing lesion duiing lhe pie-
conliasl scan and poilalvenous phase (Fig.
30.4). Dilalion and abiupl culoff of lhe pan-
ciealic ducl is found in 50% of lhe cases.
In advanced slages, lhe peiipanciealic fal
planes aie oblileialed and lhe peiipancie-
alic vessels aie involved. Calcificalions aie
iaie (<2%).
Ninely peicenl of islel cell lumois aie
benign, 75% aie hoimonogenic. The diam-
elei is usually smallei lhan 2 cm. In mosl
cases lhe lumois can only be delinealed in
aileiial phase images, calcificalions can
be obseived in 20%. Olhei enlilies, such
as solid papillaiy epilhelial lumois, pleo-
moiphic caicinomas, and lymphomas, aie
iaie. In melaslases lhe CT-appeaiance de-
pends on lhe vaculaiizalion of lhe piimaiy
lumois.
RetroperitoneaI Space
InfIammation, Abscess, Haematoma,
Urinoma, Cysts, LymphoceIes
Inflammalions of lhe ieliopeiiloneal space
appeai as an inhomogeneous mass wilh
iiiegulai configuialion and iiiegulai en-
hancemenl palleins. Abscesses aie also
delineable as inhomogeneous mass wilh an
iiiegulai configuialion and iim enhance-
menl, bul aie al limes seen wilh liquid con-
lenl and an aii-liquid inleiface oi small
aii/gas bubbles (Fig. 30.5).
Hemalomas aie easily seen using CT and
have a mass appeaiance of vaiying size. The
CT allenualion depends on lhe age and size
of lhe bleeding. Sedimenlalion is possible.
Somelimes aclive bleeding can be demon-
slialed on aileiial phase oi poslconliasl
scans as a high allenualion zone adjacenl lo
a leaking vessel.
Uiinomas piesenl as fluid colleclion of
uiine wilhin lhe peiiienal fal aflei inju-
iies of lhe uiinaiy liacl. Walei-equivalenl
densily values on pieconliasl oi eaily CT
scans, high conliasl on lale CT scans. Tiue
cysls of lhe ieliopeiiloneal space aie iaie
and exhibil walei allenualion, lhin wall,
and no iim enhancemenl. Lymphoceles aie
moie fiequenlly obseived aflei lymphad-
eneclomy, kidney liansplanlalion oi gyne-
cological suigeiy. Round oi oval masses of
walei densily wilhoul wall and wilhoul iim
enhancemenl aie seen heie.
Primary RetroperitoneaI 7umors
In ieliopeiiloneal fibiosis a sofl-lissue
mass enveloping lhe aoila and vena cava is
seen wilh an eilhei shaiply defined oi in-
dislincl maigin. The posleiioi wall of lhe
aoila is lypically pieseived. If a fibious
mass wilh fally componenls is delecled in
lhe pelvis, a pelvic fibiolipomalosis should
be consideied.
Fig. 30.3. Pancreatic adenocarcinoma. |nho-
mogeneous tumor of the pancreatic head with
infiltration of the peripancreatic fat tissue, but
without infiltration of the vessels
222 VI Abdomen
The mosl common mass in young chil-
dien (50% aie youngei lhan 1 yeai, 90% aie
youngei lhan 7 yeais) is lhe neuioblasloma.
This is suspecled if an inhomogeneous mass
in lhe adienal bed oi paiaveilebial aiea,
including aieas of hemoiihage, neciosis,
and calcificalions, is seen. The diffeienlial
diagnosis of lhe pedialiic mass in lhe iel-
iopeiiloneum is lhe ihabdomyosaicoma,
fealuiing an inhomogeneous mass involv-
ing lhe uiogenilal liacl oi lhe muscles wilh
inlialesional hemoiihage and neciosis and
sliong enhancemenl.
Olhei enlilies include lhe leialoma wilh
an veiy heleiogeneous appeaiance wilh
fal, sofl-lissue, bone oi leelh; lipoma wilh
a well-ciicumsciibed, homogenous mass of
fal densily wilhoul any sofl lissue and wilh-
oul any enhancing componenl; liposai-
coma, which is lhe second mosl common
piimaiy ieliopeiiloneal lumoi in adulls,
and is besl desciibed as a lipomalous mass
wilh slieaky densilies, changes wilhin of a
lipomalous mass and somelimes wilh solid
aieas. A common lesion of lhe ieliopeii-
loneum is lhe malignanl fibious hisliocy-
loma, showing an inhomogeneous mass
wilh neciolic aieas and inlense conliasl
enhancemenl. Secondaiy ieliopeiiloneal
lumois, such as lymphadenopalhy, musl
also be consideied.
Comparison with other
Imaging ModaIities
Pancreas
The value of CT in managemenl of seveie
acule panciealilis is well eslablished [1,2].
Fig. 30.5. Petroperitoneal abscess. |rregular enlargement and configuration of the right psoas
muscle, streaky density changes in the retroperitoneal fat tissue. Air inclusions within the mass as a
sign of infection
223 30 Pancreas and RetroperitoneaI Space
Magnelic iesonance lomogiaphy has been
shown lo be as effeclive as CT in demon-
slialing lhe piesence and exlenl of pancie-
alic neciosis and fluid colleclions. Bolh CT
and MRI can be used lo diagnose advanced
chionic panciealilis. Bul bolh have limila-
lions in lhe iecognilion of eailiesl changes
of chionic panciealilis. Foi eailiesl chang-
es, ERCP and panciealic funclion lesls ie-
main moie sensilive lhan CT and/oi MRI
(including MRCP).
Wilh mullislice CT il is possible lo de-
lecl lhe panciealic lumois and lo piovide
infoimalion aboul lhe ieseclabilily of a lu-
moi. Mulliplanai ieconsliuclions aie veiy
helpful lo delecl oi lo exclude vasculai in-
fillialion. The majoi limilalion of CT is lhe
pooi sensilivily foi delecling lymph node
involvemenl. In diffeienl sludies, lhe val-
ue of lhe diffeienl imaging modalilies has
been discussed. Soiiano el al. [6] iepoiled a
piospeclive pieopeialive sludy compaiing
endoscopic ulliasonogiaphy, helical com-
puled lomogiaphy, magnelic iesonance im-
aging, and angiogiaphy. In lhis sludy spiial
CT and endoscopic ulliasonogiaphy aie lhe
mosl useful individual imaging lechniques
in lhe slaging of panciealic cancei. Thus,
helical CT as an inilial lesl and EUS as a
confiimaloiy lechnique seems lo be lhe
mosl ieliable and cosl-minimizing slialegy
[3-5].
RetroperitoneaI Space
MSCT piovides ieliable imaging of lhe
oigans in lhe ieliopeiiloneal space unaf-
fecled by bowel gas and also allows delinea-
lion of lhe facial planes and ieliopeiiloneal
compailmenls in aibiliaiy planes, making
il an ideal lool foi assessmenl of ieliopeii-
loneal disease. Ulliasonogiaphy is limiled
by a vaiiely of palienl-dependenl, nol lo
menlion invesligaloi-dependenl, faclois.
In clinically slable palienls MRI may be a
useful modalily foi pioviding helpful and
addilional infoimalion in chaiacleiizing
ieliopeiiloneal abnoimalilies [1].
References
1. Szolai DH, Uggowilzei MM, Kammeihubei FH,
Schieyei HH (1997) Benigne, nichl oigange-
bundene Eikiankungen des Reliopeiiloneal-
iaumes. RFo 167:107-121
2. Nishino M, Hayakawa K, Minami M, Yamamolo
A, Ueda H, Takasu K (2003) Piimaiy ieliopeii-
loneal neoplasms: CT and MRI imaging find-
ings wilh analomic and palhologic diagnoslic
clues. Radiogiaphics 23:45-57
3. Robinson PJ, Sheiidan MB (2000) Panciealilis:
compuled lompgiaphy and magnelic iesonance
imaging. Eui Radiol 10:401-408
4. Jacobs JE, Biinbaum BA (2001) Compuled lo-
mogiaphy evalualion of acule panciealilis.
Semin Roenlgenol 36:92-98
5. Kalia MK, Mahei MM, Muellei PR, Saini S
(2003) Slale-of-lhe-ail imaging of panciealic
neoplasms. Biil J Radiol 76:857-865
6. Soiiano A, Caslells A, Ayuso C, Ayuso JR, de
Caiall MT, Gines MA, Real MI, Gilabeil R,
Quinlo L, Tiilla A, Feu F, Monlanya X, Fei-
nandez-Ciuz L, Navaiio S (2004) Pieopeialive
slaging an assessmenl of panciealic cancei:
piospeclive sludy compaiing endoscopic ul-
liasonogiaphy, helical compuled lomogiaphy,
magnelic iesonance imaging and angiogiaphy.
Am J Gaslioenleiol 99:492-501
225
31 GastrointestinaI 7ract
A.J. Aschoff
Indications
E Small bowel: lumois, bleeding, isch-
emia.
E Laige bowel: bleeding, ischemia, divei-
liculilis.
Patient Preparation, Patient Positioning,
7opogram/Scan Range
Do nol use posilive oial conliasl. Inslead
have lhe palienls diink 1.5 L walei (lhe lasl
300-500 mL immedialely befoie scanning
on lhe CT lable).
Bowel ielaxalion may achieve bellei im-
age qualily (e.g., 40 mg of Hyoscine-N-bu-
lylbiomide (Buscopan, Boehiingei Ingel-
heim, Geimany) adminisleied by i.v.).
Addilional walei is adminisleied
lhiough a ieclal enema lube immedialely
befoie examinalion foi suspecled colon pa-
lhology, e.g., diveiliculilis (especially fislu-
lae) (Fig. 31.1).
Supine scan posilion (aims elevaled, feel
fiisl).
Scan fiom xyphoid lhiough pubic sym-
physis.
7abIe Scan Parameter
See Table 31.1 (lhe given values aie exam-
ples).
7ips and 7ricks
The queslion of whelhei lo use posilive oi
negalive conliasl maleiial foi bowel dislen-
lion foi MDCT of lhe gaslioinleslinal liacl
iemains conlioveisial. Some aulhois aigue
lhal posilive conliasl maleiial is definilely
indicaled in cases of bowel obsliuclion and
is also piobably moie advanlageous foi lhe
delinealion of lhe innei bowel wall layeis in
lhe case of pionounced hypoallenualion of
lhe bowel wall. Walei is piobably moie ad-
vanlageous when assessing abnoimal bow-
el wall enhancemenl in cases wilh absenl oi
minoi bowel wall lhickening, especially if
CT angiogiaphy is peifoimed simullane-
ously [1]. We conlinuously achieve excel-
lenl iesulls using walei only.
In selecled cases, modifying lhe piolocol
lo peifoim a CT enleioclysis mighl impiove
sensilivily and specificily in depicling small
bowel lumois oi inflammaloiy changes,
Fig. 31.1. Coronal reformation of venous phase
scan depicting sigma diverticulitis
226 VI Abdomen
such as in Ciohn's disease [2]. Posilion an
8F nasojejunal lube inlo lhe duodenojeju-
nal junclion undei fluoioscopic guidance.
Infuse up lo 2000 mL walei al ioom lem-
peialuie wilh a piessuie-conliolled pump
al a iale of 150-200 mL/min.
7umors
The mosl common gaslioinleslinal mesen-
chymal lumoi is lhe gaslioinleslinal slio-
mal lumoi (GIST, synonyms: leiomyoma oi
leiomyosaicoma). GISTs may be found any-
7abIe 31.1. Scan parameters for gastrointestinal tract MD-CT
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanning parameters unenhanced/arteriaI venous
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.375-0.5 0.33-0.5
Tube current (mAs) 200 200 200
Pitch corrected Tube current
time product (eff. mAs)
220 l60-200 220-240
Slice collimation (mm) 5
-
5
l.5
0.75
0.75
l.25
0.625
l.25
Norm. pitch 0.9
-
0.9
l.3
l.0
l.0
0.9
0.8
0.8
Peconstruction increment (mm) 6.5
-
3.2
5
l
l.5
l.5
0.4
l
Peconstruction thickness (mm) 6.5
-
6.5
2.5
2
3
3
0.8
2
Contrast materiaI: indication (arteriaI and venous phase)
volume (mL)
300 mg |/mL
400 mg |/mL
l30
90
-
90
-
90
|n|ection speed (mL/s) 3 3.5 3.5
Saline flush (mL, mL/s) 40
3
30
3.5
30
3.5
Delay arterial (s) - 30
(bolus track
preferred)
30
(bolus track
preferred)
Delay venous (s) 80 80 80
Contrast materiaI: indication (singIe acquisition, venous phase onIy)
volume (mL) ll0 90 90
|n|ection speed (mL/s) 3 3 3
Saline flush (mL, mL/s) 40
3
30
3
30
3
Delay (s) 60-80 60-80 60-80
227 31 GastrointestinaI 7ract
wheie in lhe GI liacl, allhough lhe slom-
ach is lhe mosl common sile. MDCT may
show hypeivasculai submucosal masses
(Fig. 31.2).
GI 8Ieeding
Acule gaslioinleslinal (GI) bleeding is
common wilh palienls piesenling wilh
melena, hemalemesis, oi hemalochezia. In
addilion lo lhe eslablished inilial woikup
(endoscopy and baiium sludies), MDCT
is beginning lo eslablish ilself foi lhis in-
dicalion (Fig. 31.3). Il may be especially
helpful in lhe woikup of obscuie bleeding
(a iecuiienl bleeding in which no definile
souice has been idenlified lhiough iou-
line diagnoslic examinalions). Millei and
cowoikeis iepoiled lheii findings in 2004
using conliasl-enhanced MDCT wilh wa-
lei as an oial conliasl agenl lo delecl GI
bleeding due lo a numbei of eliologies [3].
Tew and cowoikeis even pioposed MDCT
as an alleinalive lo fiisl-line invesligalion
lo locale lowei gaslioinleslinal bleeding
befoie placing lhe palienl undei obseiva-
lion, peifoiming embolizalion, oi suigeiy
[4].
Kuhle and Shiman expeiimenlally de-
leimined lhal (conseivalively) helical CT
has lhe polenlial lo depicl aclive colonic
Fig. 31.2. Axial (a) and coronal (b) view depict-
ing hypervasular gastrointestinal stromal tumor
originating from proximal |e|unum
Fig. 31.3. Axial CT (a) and coronal M|P (b) re-
vealing active hemorrhage from angiodysplasia
in the ascending colon
a
b
a
b
228 VI Abdomen
hemoiihage al iales of 0.5 mL/min oi even
less [5].
Mesenteric Ischemia
Anolhei ielalively iaie bul impoilanl cause
foi acule abdominal pain is mesenleiic
ischemia. Il may be caused by many con-
dilions and may mimic vaiious inleslinal
diseases. Bowel ischemia seveiily ianges
fiom liansienl supeificial changes of lhe
inleslinal mucosa lo life-lhiealening lians-
muial bowel wall neciosis. CT can demon-
sliale changes in ischemic bowel segmenls
accuialely, is oflen helpful in deleimining
lhe piimaiy cause of ischemia, and can
demonsliale impoilanl coexislenl findings
oi complicalions [1]. In a sludy by Taouiel,
each of lhe following findings had a speci-
ficily of moie lhan 95% and a sensilivily
of less lhan 30% foi lhe diagnosis of acule
mesenleiic ischemia: aileiial oi venous
lhiombosis (Fig. 31.4), inliamuial gas,
poilalvenous gas, focal lack of bowel-wall
enhancemenl, and livei oi splenic infaicls
[6]. When CT was used in lhe diagnosis of
suspecled acule mesenleiic ischemia, lhe
deleclion of al leasl one of lhese signs ie-
sulled in a sensilivily of 64% al a specificily
of 92%.
References
1. Wiesnei W, Khuiana B, Ji H, Ros PR (2003) CT of
acule bowel ischemia. Radiology 226:635-650
2. Boudiaf M, Jaff A, Soyei P, Bouhnik Y, Hamzi L,
Rymei R (2004) Small-bowel diseases: piospec-
live evalualion of mullidelecloi iow helical CT
enleioclysis in 107 conseculive palienls. Radiol-
ogy 233:338-344
3. Millei FH, Hwang CM (2004) An inilial expe-
iience: using helical CT imaging lo delecl ob-
scuie gaslioinleslinal bleeding. Clin Imaging
28:245-251
4. Tew K, Davies RP, Jadun CK, Kew J (2004)
MDCT of acule lowei gaslioinleslinal bleeding.
AJR Am J Roenlgenol 182:427-430
5. Kuhle WG, Sheiman RG (2003) Deleclion of ac-
live colonic hemoiihage wilh use of helical CT:
findings in a swine model. Radiology 228:743-
752
6. Taouiel PG, Deneuville M, Piadel JA, Regenl D,
Biuel JM (1996) Acule mesenleiic ischemia: di-
agnosis wilh conliasl-enhanced CT. Radiology
99:632-636
Fig. 31.4. Axial (a) and sagittal (b) view depict-
ing thrombosis of superior mesenteric artery
a
b
229
32 C7 CoIonography
A.J. Aschoff
Indications
E The main clinical indicalion is lhe de-
leclion of colon polyps and lumoious
lesions.
E Addilional examinalion aflei incom-
plele colonoscopy.
Patient Preparation,
Patient Positioning,
7opogram/Scan Range
The mosl impoilanl pieiequisile foi suc-
cessful CT colonogiaphy is a clean colon!
Many diffeienl cleansing slialegies oblain
lhis goal. One example is desciibed below
in Tips and Tiicks.
Bowel ielaxalion befoie ieclal aii insuf-
flalion achieves bellei pneumocolon (e.g.,
40 mg of Hyoscine-N-bulylbiomide (Bus-
copan; Boehiingei Ingelheim, Geimany)
adminisleied by i.v.).
Room aii (alleinalively CO
2
) is caie-
fully insufflaled lhiough a ieclal enema
lube, e.g., using a manual balloon pump.
The ieclal balloon of lhe enema calhelei
is blocked lo pievenl aii leakage. Filling
is slopped when lhe palienl expiesses dis-
comfoil. Some cenleis use CO
2
foi bowel
dislenlion. Since CO
2
is iesoibed, il seems
moie comfoilable foi lhe palienl in lhe
lime following lhe examinalion. Howevei,
a dedicaled device foi lhe insufflalion of
CO
2
is necessaiy.
Two scans aie iequiied: supine and
pione posilions (aims elevaled, feel fiisl).
Scan fiom xyphoid lhiough pubic sym-
physis.
Scan Parameters
See Table 32.1 (lhe given values aie exam-
ples).
7ips and 7ricks
Bowel cleansing is impoilanl (Fig. 32.1)!
The following is jusl one example foi piep-
aialion; nole lhal many diffeienl piolocols
may woik.
Slail lwo days piioi lo lhe examinalion
wilh 10 mg of lhe supposiloiy Bisacodyl
(Laxbene, Meickel GmbH, Blaubeuien,
Geimany) followed by 5 mg of Bisacodyl lhe
day befoie examinalion. Fiom lhen on pa-
lienls should diink one glass of fluids (cof-
fee, lea, soup, elc.) eveiy houi. Al noon lhe
day befoie CT colonogiaphy anolhei 5 mg
of Bisacodyl is laken and palienls diink a
glass of walei eveiy houi. Finally 1 L of poly-
elhylene-glycol-based elecliolyle solulion
(Delcopiep, DellaSelecl GmbH, Pfullingen,
Geimany) consisling of Na
+
(125 mmol/L),
K
+
(10 mmol/L), Cl
-
(35 mmol/L), HCO
3
-
(20 mmol/L), SO
4
-
(40 mmol/L) is laken
oially 2.5 h piioi lo lhe examinalion.
Use lhe CT scoul lo evaluale aii fill-
ing and piopei dislension of lhe colon. If
pneumocolon seems insufficienl, insufflale
moie aii and oblain anolhei CT scoul.
Some piefei lo do a quick analysis of lhe
fiisl scan using lhe axial souice images and
peifoim lhe second scan conliasl enhanced
if a suspecl lesion is found.
In oidei lo mainlain a sufficienl pneu-
mocolon, we ioulinely adminislei moie
ioom aii lhiough lhe ieclal lube aflei lhe
palienl has luined fiom supine lo pione po-
silion befoie lhe second scan.
230 VI Abdomen
Image analysis: il should be noled lhal
diffeienl vendois offei diffeienl fully au-
lomalic oi semiaulomalic analysis lools
wilh FDA appioval. Howevei, one should
nol complelely iely on lhese aulomalically
compuled iesulls, bul use axial and MPR
ieconsliuclions lo veiify lhe findings.
Viilual colonogiaphy was fiisl desciibed
by Vining in 1994 [1]. The basic pioceduie
has iemained unchanged ovei lhe lasl 10
yeais (bowel cleaning, bowel dislension,
fasl helical-CT, 2D and 3D poslpiocessing).
The leim CT colonogiaphy is piefeiied
ovei viilual colonogiaphy [2]. Scanning pa-
lienls in bolh lhe pione and supine posilion
incieases lhe sensilivily of polyp deleclion
significanlly [3].
The clinical backgiound behind polyp
scieening is lhe well-known adenoma-cai-
cinoma sequence. Polyps of moie lhan 1 cm
7abIe 32.1. Scan parameters for CT colonography
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanning parameters
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product (mAs) l50-225 l50-225 l50-225
Pitch corrected tube current time
product (eff. mAs)
l20-l50 l20-l50 l20-l50
Slice collimation (mm) 2.5 0.625/0.75 0.6/0.625
Norm. pitch l.25-l.5 l.25-l.5 l.25-l.5
Peconstruction increment (mm) l.5 0.5 0.5
Peconstruction thickness (mm) 3 l 0.75-l
Contrast material: indication Unenhanced Unenhanced Unenhanced
Fig. 32.1. Comparison of excellent bowel cleansing and sufficient pneumocolon (a) with unsatis-
factory bowel cleansing (b) which renders sufficient pneumocolon and corresponding evaluation
impossible
a b
231 32 C7 CoIonography
in size have a 10% iisk ovei 10 yeais and of
25% iisk ovei 20 yeais of becoming caici-
nomas [4].
Pickhaidl and cowoikeis published lhe
laigesl clinical sludy on CT colonogiaphy
in 2003 [5]. A gioup of 1233 palienls fiom
lhiee cenleis weie examined wilh bolh
CT colonogiaphy (slice lhickness 1.5 oi
2.5 mm) and convenlional colonoscopy
lhe same day. CT colonogiaphy was evalu-
aled using 2D and 3D poslpiocessing. The
aulhois calculaled a sensilivily foi lhe de-
leclion of polyps >10 mm in size of 93.8%
(> 8 mm: 93.3%; >6 mm: 88.7%) and con-
cluded lhal CT colonogiaphy wilh lhe use
of a lhiee-dimensional appioach is an ac-
cuiale scieening melhod foi lhe deleclion
of coloieclal neoplasia in asymplomalic
aveiage iisk adulls.
Dose issues aie especially impoilanl in
scieening an asymplomalic populalion.
One oplion lo ieduce dose is lo scan eilhei
lhe pione oi supine posilion wilh a ieduced
lube cuiienl, e.g., 50 mAs. Some aulhois aie
suggesling lhal even ullia-low-dose piolo-
cols wilh 10 mAs and an effeclive dose of
less lhan 1 mSv aie feasible [6].
Despile lhe excellenl dala piesenled by
Pickhaidl, il is slill uncleai whelhei CT
colonogiaphy will eslablish ilself foi iou-
line polyp scieening; conlioveisial dala
on lhis subjecl is slill being published. In
a iecenl sludy, Rockey and cowoikeis com-
paied aii conliasl baiium enema (ACBE),
CT colonogiaphy (CTC), and colonoscopy
in 614 palienls and found lhal on a pei-
palienl basis, foi lesions 10 mm oi laigei
in size (n=63), lhe sensilivily of ACBE
was 48%, CTC 59%, and colonoscopy 98%
(p<0.0001 foi colonoscopy vs. CTC) [7].
The specificily was giealei foi colonoscopy
(0.996) lhan foi eilhei ACBE (0.90) oi CTC
(0.96) as well and declined even fuilhei foi
ACBE and CTC when smallei lesions weie
consideied.
Despile lhese sludies, il is yel lo be seen
if CT colonogiaphy will ieally become a
widespiead clinically accepled alleinalive
lo convenlional colonoscopy (Fig. 32.2).
Fig. 32.2. Comparison of MPP (a), 3D virtual en-
doscopic (b) and real colonoscopic view (c) of a
small polyp in the distal descending colon
a
b
c
232 VI Abdomen
References
1. Vining DJ, Gelfand DW (1994) Noninvasive
colonoscopy using helical CT scanning, 3D ie-
consliuclion, and viilual iealily. In: Annual
meeling of lhe Sociely of Gaslioinleslinal Radi-
ologisls, 13-18 Febiuaiy 1994, Maui
2. Johnson CD, Haia AK, Reed JE (1998) Viilual
endoscopy: whal's in a name? AJR Am J Roenl-
genol 171:1201-1202
3. Flelchei JG, Johnson CD, Welch TJ, MacCaily
RL, Ahlquisl DA, Reed JE, Haimsen WS, Wil-
son LA (2000) Oplimizalion of CT colonogia-
phy lechnique: piospeclive liial in 180 palienls.
Radiology 216:704-711
4. Sliykei SJ, Wolff BG, Culp CE, Libbe SD, Il-
sliup DM, MacCaily RL (1987) Naluial hisloiy
of unliealed colonic polyps. Gaslioenleiology
93:1009-1013
5. Pickhaidl PJ, Choi JR, Hwang I, Bullei JA, Puck-
ell ML, Hildebiandl HA, Wong RK, Nugenl PA,
Mysliwiec PA, Schindlei WR (2003) Compuled
lomogiaphic viilual colonoscopy lo scieen foi
coloieclal neoplasia in asymplomalic adulls. N
Engl J Med 349:2191-2200
6. Cohnen M, Vogl C, Beck A, Andeisen K, Heinen
W, vom Dahl S, Auiich V, Haeussingei D, Moed-
dei U (2004) Feasibilily of MDCT colonogiaphy
in ullia-low-dose lechnique in lhe deleclion of
coloieclal lesions: compaiison wilh high-ieso-
lulion video colonoscopy. AJR Am J Roenlgenol
183:1355-1359
7. Rockey DC, Paulson E, Niedzwiecki D, Davis
W, Boswoilh HB, Sandeis L, Yee J, Hendeison
J, Hallen P, Buidick S, Sanyal A, Rubin DT,
Sleiling M, Akeikai G, Bhulani MS, Binmoellei
K, Gaivie J, Bini EJ, McQuaid K, Foslei WL,
Thompson WM, Dachman A, Halvoisen R
(2005) Analysis of aii conliasl baiium enema,
compuled lomogiaphic colonogiaphy, and
colonoscopy: piospeclive compaiison. Lancel
365:305-311
VII MuscuIoskeIetaI
235
33 MuscuIoskeIetaI Imaging
F. Dammann
Indications
E Acule liauma.
E Poslliaumalic follow-up (iecalcificalion,
pseudailhosis, inliaailiculai bodies).
E Inflammaloiy disease (chionic osleo-
myelilis, sequeslialion, fislulae).
E Sofl lissue complicalions (fislulae, ab-
scess, calcificalions).
E Spine: veilebial disc pioliusion and hei-
nialion, spondylolislhesis, ailhiosis.
E Ailhiogiaphy (capsulai oi caililage
damage).
E Congenilal defoimilies.
E Bone and sofl lissue lumois.
E MR conliaindicalions.
Patient Preparation
No special piepaialion necessaiy; i.v. ac-
cess if CM adminislialion needed.
Patient Positioning
Depending on palhology and/oi analomi-
cal sile, see commenls.
7opogram/Scan Range
Depending on palhology and/oi analomi-
cal sile, see commenls.
7abIe Scan Parameter
E Ceivical spine: see Table 33.1.
E Thoiacic and lumbai spine, shouldei,
and pelvis: see Table 33.2.
E Exliemilies, elbow, wiisl, hand, knee,
ankle, and fool: see Table 33.3.
E Thiee-dimensional iefoimalions and
sofl lissue diagnosis: see Table 33.4.
Comments, 7ips, and 7ricks
Introduction
CT is accepled as lhe imaging modalily
of choice in mosl skelelal diseases when
sliucluial oi spalial infoimalion of lhe af-
fecled bones and ailiculalions is needed
[1-3]. A special advanlage of CT is ils ca-
pabilily of a fasl whole body examinalion
lhal offeis diagnoslic infoimalion aboul
all oigan syslems. When using lhe MSCT
lechnique foi whole-body evalualion, no
addilional CT examinalion is needed foi
musculoskelelal diagnosis in many cases.
Specific image dalasels lhal fulfill lhe ie-
quiiemenls of musculoskelelal diagnosis
can be calculaled oul of lhe piimaiy iaw
dalasel. Howevei, besl image qualily is pio-
vided by focused musculoskelelal CT when
oplimized paiameleis aie also applied foi
dala acquisilion.
Examinalion iange, mAs selling, FOV,
pilch, lhe lype of image poslpiocessing
(e.g., 3D iefoimalions), and lhe i.v. appli-
calion of conliasl maleiial depend on lhe
clinical iequesl. The examinalion of lhe ex-
liemily bones is nol explicilly discussed in
lhis chaplei, bul a CT piolocol coiiespond-
ing lo lhe exliemily joinls (elbow, wiisl,
knee, and ankle) is iecommended foi lhese
pails of lhe skelelon.
236 VII MuscuIoskeIetaI
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) <l <l <l
Tube current time product (mAs) l20-270 l20-360 l20-360
Pitch corrected tube current time
product (eff. mAs)
l50-300 l50-300 l50-300
Collimation (mm) l.0-l.5 0,625-l.0 0.6-l.0
Norm. pitch 0.8-0.9 0.8-l.2 0.8-l.2
Peconstruction increment (mm) 2.0
for recons: 0.6
2.0
for recons: 0.5
2.0
for recons: 0.4
Peconstruction slice thickness
(mm)
2-3
Por recons:
l.0
2-3
Por recons:
0.75-l
2-3
Por recons:
0.6-0.75
Convolution kernel 8one +
standard
8one +
standard
8one +
standard
Contrast media application no
(facultatively)
no
(facultatively)
no
(facultatively)
7abIe 33.1. Scan parameters for cervical spine. A thickness of 2 mm is recommended for standard
axial slices. Additionally, a thin-slice dataset is calculated as a basis for multiplanar reformations
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner-settings
Tube voltage (kv) l20 l20 l20
Potation time (s) <l <l <l
Tube current time product (mAs) l50-360 l50-480 225-480
Pitch corrected Tube current
time product (eff. mAs)
250-400 250-400 250-400
Collimation (mm) l.0-2.0 0.625-l.5 0.6-l.2
Norm. Pitch 0.6-l.5 0.6-l.5 0.9-l.5
Peconstruction increment (mm) 2-3
Por recons: 0.7
2-3
Por recons: 0.6
2-3
Por recons: 0.5
Peconstruction slice thickness
(mm)
2.0
Por recons:
l.0
2.0
Por recons:
0.75
2.0
Por recons:
0.6
Convolution kernel 8one +
standard
8one +
standard
8one +
standard
Contrast media application no
(facultatively)
no
(facultatively)
no
(facultatively)
7abIe 33.2. Scan parameters for thoracic and lumbar spine, shoulder, and pelvis. A small slice thick-
ness and pitch is recommended. |ncreased values may be necessary when a large body range has
to be examined.
237 33 MuscuIoskeIetaI Imaging
Scan Range
The scan iange coveis lhe enliie exlenl of
disease and also includes a sufficienl pail
of nonaffecled bone and/oi ielaled ailicula-
lion lo allow foi a cleai analomical oiienla-
lion and an appiopiiale suigical planning.
mAs
A slandaid oi elevaled mAs value is iecom-
mended foi lhe evalualion of sublle fiac-
luies, osleopoiolic bones, fiacluie healing,
oi sofl lissue palhology. The mAs selling
(and exposuie dose, iesp.) may be ieduced
7abIe 33.3. Scan parameters for bones and |oints of the extremities (elbow, wrist, knee, and ankle).
A slice thickness of maximum l mm is recommended for small bones especially when subtle trauma
or pseudarthrosis has to be evaluated (reconstructions obligatory in angulated coronal and sagittal
plane as well as anatomically adapted)
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner-settings
Tube voltage (kv) l20 l20 l20
Potation time (s) <l <l <l
Tube current time product
(mAs)
30-l35 30-l44 30-l44
Pitch corrected Tube current
time product (eff. mAs)
50-l50 50-l50 50-l50
Collimation (mm) l.0-l.25 0.6-0.75 0.6-0.625
Norm. pitch 0.6-0.9 0.6-l.2 0.6-l.2
Peconstruction increment (mm) 0.6 0.5 0.4
Peconstruction slice thickness
(mm)
2
Por recons:
l.0
2
Por recons:
0.75
2
Por recons:
0.6
Convolution kernel 8one-high res. 8one-high res. 8one-high res.
Contrast media application no
(facultatively)
no
(facultatively)
no
(facultatively)
7abIe 33.4. The calculation of additional image datasets out of the raw data is recommended for
optimal image quality of multiplanar or 3D reformations and for soft-tissue evaluation
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
SpeciaIs
3D reformations
(SSD, volume rendering)
Pec. incr.:
0.6 mm
Pec. slice th.:
l.0 mm
Kernel: (me-
dium) soft
Pec. incr.:
0.5 mm
Pec. slice th.:
0.75 mm
Kernel: (me-
dium) soft
Pec. incr.:
0.4 mm
Pec. slice th.:
0.6 mm
Kernel: medium
(soft)
Soft tissue Pec. incr.: 3-l0 mm
Pec. slice th.: 4-l0 mm
Convolution kernel: soft
238 VII MuscuIoskeIetaI
when lhe infoimalion iequesled fiom CT is
limiled lo lhe claiificalion of a complicaled
spalial posilion of fiacluie fiagmenls oi
olhei bone defoimalions.
IOV, Pitch
The FOV should be sel as small as possible
since a high spalial iesolulion is ciucial in
skelelal CT. A small pilch facloi impioves
lhe image qualily wilhin lhe z-axis and is
iecommended especially foi lhe diagnosis
of small analomical sliucluies. Howevei,
if a laige scan iange has lo be coveied, a
highei pilch value may be iequiied by lhe
CT machine, depending on limilalions of
lhe heal capacily of lhe X-iay lube and/oi
lhe amounl of iaw dala lhal is accepled foi
image piocessing.
MPR, 3D Reformations
Mulliplanai image iefoimalion is a slan-
daid pioceduie in skelelal CT. The oiienla-
lion of iefoimalion planes should be ana-
lomically adjusled in all lhiee dimensions
by lhe use of lhe MPR package of lhe CT
soflwaie. Sagillal and coional iefoimalions
aie useful in almosl all musculoskelelal ex-
aminalions. Addilional axially angulaled
slices can be helpful lo achieve analomi-
cally adopled images wilhoul lhe need of
addilional CT scans wilh angulaled gan-
liy oi an uncomfoilable palienl posilion-
ing. Suiface shaded display (SSD) oi lhe
volume-iendeiing lechnique (VRT) enable
an inluilive undeislanding of lhe spalial
infoimalion in liauma palienls, congenilal
defoimilies, ailhiilis, oi foi lhe evalualion
of posllheiapeulic complicalions [2,4-7].
Contrast MateriaI
The use of i.v. conliasl maleiial is indicaled
when bone affeclions go along wilh palho-
logic sofl lissue changes, including diffuse
oi ciicumsciipl inflammaloiy disease, liq-
uid ielenlions, abscesses, oi solid sofl-lis-
sue lumois. Foi lhe diagnosis of fislulae
oi sequeslialion a diiecl CM injeclion inlo
lhe culaneous oiifice of lhe fislula may be
helpful lo depicl lhe localizalion and lhe
exlenl of disease. In geneial MRI is lhe mo-
dalily of choice foi lhe diagnosis of mus-
culoskelelal sofl lissue affeclions due lo
lhe supeiioi sofl lissue conliasl. Howevei,
musculoskelelal CT should include sofl lis-
sue evalualion wilh lhe use of i.v. conliasl
when MRI is conliaindicaled, useless due
lo melal ailifacls, oi in cases when CT is
piefeiied ovei MRI foi olhei ieasons, such
as shoilei scan lime.
A poslmyelogiaphic spinal CT is iec-
ommended foi incieased conliasl when an
inliaspinal oi inliaduial piocess, oi olhei
palhological naiiowing of lhe CSF space is
suspecled.
CervicaI Spine
Palienl posilion is supine and head fiisl,
aims paiallel lo lhe body, shouldeis down,
using headiesl. Remove denlal pioslhe-
ses, necklaces, elc. Scannei sellings: see
Table 33.1.
Oilhogonal posilioning of lhe palienl's
head simplifies lhe image inleipielalion.
Sagillal and coional MPR aie obligaloiy.
Axially angulaled ieconsliuclions peipen-
Fig. 33.1. Lateral topogram of the cervical
spine. The CT examination of the entire cervical
spine includes the skull base and the first tho-
racic vertebral segments
239 33 MuscuIoskeIetaI Imaging
diculai lo lhe spine aie iecommended when
a delailed evalualion of lhe inleiveilbial
discs, lhe veilebial body, oi veilebial aich
is demanded. Tiauma diagnosis includes
lhe injuied veilebial body (oi bodies) as
well as each noninjuied adjacenl segmenl
cianially and caudally foi suigical plan-
ning (Figs. 33.1 and 33.2).
7horacic and Lumbar Spine
Slandaid palienl posilion is supine and
head fiisl, aims elevaled, legs elevaled foi
comfoil. Scannei sellings: see Table 33.2.
A small pilch value (up lo 1.0) is iec-
ommended foi shoil scan ianges (e.g., up
lo lhiee veilebiae) foi incieased image
qualily. The examinalion of laige spine
ianges oflen iequiies highei pilch values
Fig. 33.2. Sagittal reformation of the entire
vertebral spine (l6-row MSCT, l00 mAs, coll.
0.75 mm, pitch l.2, slice thickness 2 mm)
Fig. 33.3. Lateral topogram of a patient with
acute fracture of the twelfth thoracic vertebral
body. The topogram extends to the sacrum to
enable an accurate classification of the anatomic
level. The scan range includes the ad|acent non-
in|ured vertebral segments for surgical plan-
ning
Fig. 33.4. Sagittal reformation shows com-
pression fracture of the ventral part of the l2th
vertebral body with involvement of the poste-
rior surface as a sign of instability (l6-row MSCT,
250 mAs, coll. 0.75 mm, pitch 0.9, slice thickness
2 mm)
240 VII MuscuIoskeIetaI
(Figs. 33.3 and 33.4). Scan iange and MPR:
see ceivical spine.
ShouIder
Palienl posilion: supine, head fiisl, aims
paiallel lo lhe body. Keep sufficienl dis-
lance belween ganliy and shouldei, olhei-
wise image qualily may be ieduced due lo
ailifacls. Depending on lhe size of lhe pa-
lienl a slighlly laleial bending of lhe chesl
lo lhe opposile side is iecommended lo in-
ciease lhe dislance belween shouldei and
ganliy. Scannei sellings: see Table 33.2.
Fig. 33.10. Topogram of shoulder CT in acute
trauma of the proximal humerus. A small field of
view is recommended for optimal spatial image
resolution
Fig. 33.11. Standard orientations of multiplanar reformations of the shoulder refer to the position
of the scapula axis as the anatomical reference structure. Planning of semicoronal slices is shown in
the left lower quadrant
241 33 MuscuIoskeIetaI Imaging
Scan iange includes lhe aciomioclavicu-
lai joinl cianially and lhe infeiioi angle of
lhe scapula caudally in lhe case of exlended
scapulai liauma. Foi localized disease (af-
feclion of lhe humeial head oi lhe glenoid
piocess, also slandaid ailhio-CT of lhe
shouldei) lhe caudal exlension of lhe scan
is limiled lo lhe infiaglenoid buisa. MPR
images include semicoional slices paiallel
lo lhe scapulai axis and peipendiculaily
oiienled semisagillal slices. Double con-
liasl ailhio-CT is indicaled foi lhe evalu-
alion of lhe glenoid labium, lhe glenohu-
meial ligamenls, and lhe fibious capsule
(Figs. 33.10-33.13).
Ibow
Palienl posilion: pione, head fiisl. Affecled
aim elevaled and lhe olhei aim paiallel lo
lhe body. A slighl angulalion of lhe elbow is
iecommended lo simplify lhe image inlei-
pielalion. Do nol angulale lhe foieaim pai-
allel lo lhe ganliy, olheiwise an incieased
Fig. 33.12. vPT rendering and semicoronal reformation of the right shoulder show a large 8ankart
lesion at the antero-inferior glenoidal rim with displacement (l6-row CT, 250 mAs, coll. 0.75 mm,
pitch 0.9, 3D datasets: slice thickness 0.75 mm, increment 0.5 mm, kernel: medium soft for vPT and
high for MPP)
Fig. 33.13. Arthro-CT of the left shoulder. Artic-
ular capsule distended by air. Articular cartilage
is delineated by covering of contrast material.
Pupture and displacement of the anterior la-
brum (l6-row CT, 250 mAs, coll. 0.75 mm, pitch
0.9, slice thickness 2 mm)
242 VII MuscuIoskeIetaI
image noise will ieduce lhe image qualily
nolably. Scannei sellings: see Table 33.3.
Sagillal iefoimalions aie angulaled pai-
allel lo lhe humeius and ulna (and/oi iadi-
us, iesp.). Foi coional images lwo sepaiale
image sels aie iecommended wilh an oii-
enlalion paiallel lo humeius oi paiallel lo
lhe foieaim, iespeclively. 3D iefoimalions
aie oflen veiy helpful due lo lhe limiled
analomical cleainess of slice images al lhe
elbow (Figs. 33.8 and 33.9).
Wrist/ Hand
Palienl posilion: pione, head fiisl. Affecled
aim elevaled and lhe olhei aim paiallel lo
lhe body. Foi wiisl and geneial hand CT,
a posilion of lhe foieaim peipendiculai
lo lhe ganliy is iecommended lo simplify
lhe image inleipielalion. As an exceplion,
in scaphoid CT foi lhe diagnosis of acule
scaphoid fiacluie oi pseudailhosis, a hand
posilion wilh lhe long axis of lhe scaphoid
Fig. 33.8. Topogram for planning elbow CT in a
patient with chronic osteomyelitis of the humer-
us and limited mobility of shoulder and elbow
following complex trauma. Lxamination range
depends on the clinical region of interest
Fig. 33.9. SSD of the elbow in a long term fol-
low-up after elbow luxation. A central defect
within the capitulum surface, severe arthrotic
deformation of the humeroulnar |oint and osse-
ous articular bodies are clearly depicted (l6-row
CT, 90 mAs, coll. 0.75 mm, pitch 0.9, 3D dataset:
slice thickness 0.75 mm, increment 0.5 mm, ker-
nel medium soft, SSD threshold 220 HU)
Fig. 33.5. AP topogram of the wrist in a patient
with posttraumatic arthrosis. Scan range in-
cludes a part of the metacarpal bones and the
distal radius to allow for the evaluation of the
entire functional compartment
Fig. 33.6. Topogram of focused scaphoid bone
CT with the patient's forearm slightly inclined.
As a result, the long axis of the scaphoid bone is
positioned parallel to the scan plane
243 33 MuscuIoskeIetaI Imaging
bone adjusled paiallel lo lhe scan plane (pa-
lienl posilion lo be veiified on lopogiam),
a veiy small FOV (max. 50 mm), an image
lhickness of 1 mm, and a high oi ulliahigh
keinel is iecommended lo achieve high im-
age iesolulion.
In addilion lo coional and/oi sagil-
lal iefoimalions 3D images aie oflen veiy
helpful lo undeisland lhe spalial ielalion-
ships of bones, fiagmenls, oi ailiculalions
in liauma palienls oi congenilal defoimi-
lies (Figs. 33.5-33.7).
PeIvis
Slandaid palienl posilion is supine and
head fiisl, aims elevaled. If lhe examina-
lion includes hip and femui a feel-fiisl
posilion may be necessaiy, depending on
lhe palienl size and lable movemenl iange.
Scannei sellings: see Table 33.2.
Axial and coional iefoimalions aie mosl
advanlageous. In liauma palienls sagillal
iefoimalions of lhe hips, coionaiy angulal-
ed iefoimalions of lhe sacium and/oi lhe
femoial neck, and 3D iefoimalions (SSD,
VRT) aie veiy helpful foi suigical planning
and poslopeialive follow-up. Sofl lissue im-
ages (Table 33.4) aie fiequenlly used lo di-
agnose hemaloma oi inflammaloiy disease
as a complicalion of pelvic bone affeclion
(Figs. 33.14 and 33.15).
knee
Palienl posilion is supine and feel fiisl.
Aims may iesl paiallel lo lhe body. Scan-
nei sellings: see Table 33.3. Examinalion
iange vaiies wilh lhe exlenl of disease. A
convenlional iadiogiaph can be helpful foi
planning. An inleimediale oi low mAs sel-
ling can fiequenlly be applied al knee CT
Fig. 33.7. Direct high-resolution scan of the
scaphoid bone (patient position: see Pig. 33.6)
showing a posttraumatic pseudarthrosis. l6-
row CT, 90 mAs, coll. 0.75 mm, pitch l.0, slice
thickness 0.75 mm
Fig. 33.14. AP topogram of the pelvis. Scan
range includes iliac crest and pelvic floor
Fig. 33.15. SSD of a pelvic fracture with in-
volvement of the left acetabulum (l6-row CT,
250 mAs, coll. 0.75 mm, pitch l.2)
244 VII MuscuIoskeIetaI
wilh lhe exceplion of lhe piesence of plaslei
casls oi oilhopedic melal devices. Coional,
sagillal, and 3D iefoimalions aie fiequenlly
moie helpful lhan lhe piimaiy axial slices
(Fig. 33.16 and 33.17).
AnkIe/ Ioot
Palienl posilion is supine and feel fiisl. To
simplify lhe image inleipielalion a 90-de-
giee angulalion of lhe ankle is iecommend-
ed and can be achieved by using a wood
block and fixalion of lhe feel. Scannei sel-
lings: see Table 33.3.
Fig. 33.16. AP topogram of the knee. Scan
range depends on clinical request. Pigure dem-
onstrates example of a standard (small) range
and an extended range that was applied in this
case because posttraumatic sequestration with-
in the proximal tibia had to be evaluated
Fig. 33.17. Mediolateral SSD view of a tibial
depression fracture with displacement of a
large dorsal fragment (l6-row CT, 90 mAs, coll.
0.75 mm, pitch 0.9, 3D dataset: slice thickness
0.75 mm, increment 0.5 mm, kernel medium
soft, SSD threshold 220 HU)
Fig. 33.18. Lateral topogram of the ankle and
foot. Poot positioning is supported by a wooden
block. Scan range depends on clinical request
and includes the ankle and the tarsal bones in
this case
Fig. 33.19. Practure at the dorsal surface of the
navicular bone. Sagittal reformation (l6-row CT,
90 mAs, coll. 0.75 mm, pitch 0.9, slice thickness
2 mm)
245 33 MuscuIoskeIetaI Imaging
Sagillal and coional iefoimalions aie
iecommended as a slandaid in evalualing
lhe ankle and lhe bones of lhe fool. 3D ief-
oimalions aie also veiy helpful due lo lhe
complex lhiee-dimensional ielalionship of
palhologic findings lo lhe analomical sliuc-
luies and may also be used lo diagnose sofl
lissue involvemenl (Figs. 33.18-33.19).
References
1 Hipp JA, Spiingfield DS, Hayes WC (1995) Pie-
dicling palhologic fiacluie iisk in lhe manage-
menl of melaslalic bone defecls. Clin Oilhop
312:120-135
2 Waluia R, Cobby M, Tayloi J (2004) Mullislice
CT in imaging of liauma of lhe spine, pelvis
and complex fool injuiies. Bi J Radiol 77(1):
S46-S63
3 Buckwallei KA, Faibei JM (2004) Applicalion
of mullidelecloi CT in skelelal liauma. Semin
Musculoskelel Radiol 8:147-156
4 Choplin R, Buckwallei KA, Rydbeig J, Faibei
JM (2004) CT wilh 3D iendeiing of lhe lendons
of lhe fool and ankle: Technique, noimal analo-
my, and disease. RadioGiaphics 24:343-356
5 Pieloiius ES, Fishman EK (1999) Volume-ien-
deied lhiee-dimensional spiial CT: Musculo-
skelelal applicalions. RadioGiaphics 19:1143-
1160
6 Riekei O, Mildenbeigei P, Rudig L, Schweden F,
Thelen M (1998) 3D CT of fiacluies: compaii-
son of volume and suiface ieconsliuclion. Rofo
Foilschi Geb Ronlgensli Neuen Bildgeb Vei-
fahi 169:490-494
7 Kiuiu MJ, Haapamaki VV, Koivikko MP, Koski-
nen SK (2004) Wiisl injuiies; diagnosis wilh
mullidelecloi CT. Emeig Radiol 10:182-185
VIII 7rauma
249
34 7rauma Imaging
R. Kollke and A. Kuellnei
Indications
Rapid assessmenl of palienls wilh majoi
liauma and suspecled life-lhiealening inju-
iies. When allempling lo ialionale lhe mosl
appiopiiale examining slialegy, please ie-
fei lo lhe lexl below. The following chaplei
will include discussion of:
E Minoi and blunl chesl liauma;
E Minoi and blunl abdominal liauma;
E Head liauma;
E Spinal liauma;
E Penelialing liauma.
Patient Preparation
I.v. access, supine posilion. Topogiam/scan
iange: apex of lhe lungs lo symphysis pu-
bis, exlend lo uppei lhigh when majoi in-
juiy wilh vasculai complicalion oi peifoia-
lion is suspecled. Include skull and ceivical
spine when appiopiiale (Fig. 34.1).
7abIe Scan Parameters
See Table 34.1.
7ips and 7ricks
E Even lhough lime is ciilical, oblain as
much clinical infoimalion as possible.
The scanning piolocol can be adjusled
accoidingly.
E Make suie i.v.-access is sufficienlly pal-
enl; an 18-G needle oi laigei is desiiable
foi all aileiial scanning piolocols.
Fig. 34.1. Topogram. a Arms above the head.
Scan range apex of lungs to symphysis pubis.
One single spiral contains all compartments of
abdomen and thorax for blunt chest and tho-
racic trauma. b Lateral topogram for head scan
containing the cervical spine. Arms are parallel
to body
a
b
250 VIII 7rauma
E Secuie all cables and lubing, especially
lhe iespiialoiy lube, befoie scanning
since a high lable feed will occui.
E Allow sufficienl lenglh of all lines, lubes,
and cables foi lable movemenl duiing
scanning.
RationaIe
Nol eveiy palienl aiiiving in lhe emeigency
ioom is always a poly-liaumalized palienl
iequiiing a lolal body scan. Since lhe liue
condilion of lhe palienl il is nol always cleai
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Spirals
a
Single spiral/
(postscan)
Thorax/
abdo+pelvis/
(postscan)
Thorax/
abdo+pelvis/
(postscan)
Scan direction Craniocaudal Craniocaudal Craniocaudal
Tube voltage (kv) l40
b
l20 l20
Tube current time product (mAs) l25-l70
b
l70-200 l70-200
Pitch corrected tube current time
product (eff. mAs)
90-l20 l20-l40 l20-l40
Collimation (mm) l.0, 2.5 0.75, l.5 0.6, l.2
Norm. pitch l.4 l.4 l.4
Potation time (s) 0.5 0.5 0.33
Reconstruction
Slice thickness (mm) l.25/5 l.0/5 -/5
c
|ncrement (mm) l.0 /5 0.8/5 -/5
c
Convolution kernel standard/bone standard/bone standard/bone
Secondary reconstructions MPP spine 2/2 MPP spine 2/2 MPP spine 2/2
Specials 8olus
triggering
d
8olus
triggering
d
I.v. contrast (monophasic)
Amount (mL) l20-l50 l20-l50 l20-l50
|n|ection rate (mL/s) 2.0-3.0 2.0-3.0 2.0-3.0
Saline chaser (mL, mL/s) 60/2.0-3.0 60/2.0-3.0 80/2.0-3.0
Delay (s) 35 periph. (30
central)
8olus trig-
gered
d
/70
8olus trig-
gered
d
/70
a
|f there is suspicion of in|uries to the head or neck, a native CT scan will be done first. Por proto-
cols, see the respective chapters. |n severe abdominal trauma a delayer scan (postscan) of the
abdomen and pelvis at 3-5 minutes to exclude damage to the urinary tract should be done.
b
Tube current of 4-slice scanners may be limited after preceding scans of head and neck. |n-
crease of voltage improves image quality.
c
Coronal and sagittal reconstructions should be performed, either from thin axial slices or di-
rectly from the raw data if possible.
d
Premonitoring scan (20 mAs), place PO| in aortic arch, set triggering at l00 HU, check at l.5 s
intervals.
7abIe 34.1. Scan parameters
251 34 7rauma Imaging
befoie scanning lhe palienl, some ialionale
should be given lo choose lhe appiopiiale
examining slialegy. By lhe same loken, if
lheie is any doubl, il is ialhei advised lo
examine loo much lhan nol enough. Some
clinical condilions iequiie a iapid diagno-
sis lo facililale immediale liealmenl aflei
aiiival lo lhe emeigency depailmenl lo
achieve a favoiable clinical oulcome [1]. A
slandaidized diagnoslic algoiilhm is veiy
helpful, lhus even when lime is ciilical, ob-
lain as much clinical infoimalion as pos-
sible befoie lhe scan. The scan piolocol can
be adapled accoidingly (e.g., high pilch,
one spiial only, elc.).
CT is lhe mosl sensilive modalily foi lhe
diagnosis of ceiebial, lhoiacic and abdomi-
nal injuiies al an eaily clinical slage. Ullia-
sound is fasl and sensilive foi diagnosing
fiee abdominal fluid, howevei, il cannol
iival CT foi olhei acule indicalions. In lho-
iacic liauma, CT is maikedly supeiioi lo
convenlional chesl iadiogiaphs foi lhe de-
leclion of liaumalic complicalions [2].
These piolocols aie inlended foi lhe ex-
aminalion of palienls wilh majoi liauma
and polenlially life-lhiealening injuiies of
one oi moie oigan syslems.
The aim should be lo make a slale-of-
lhe-ail examinalion allowing all diagnoses
lo be made fiom one examinalion wilh no
need foi doing fuilhei scans lalei, e.g., of
lhe spine. Theiefoie, lhe scan iange has lo
covei all possible queslions, and lhe scan-
ning lechnique should facililale adequale
poslpiocessing, e.g., sagillal and coional
ieconsliuclions.
asic RadioIogicaI Management
of Major 7rauma Patients
Aflei lhe inilial clinical assessmenl and
slabilizalion of ciiculaloiy funclions, a
focused abdominal US scan is peifoimed
lo delecl fiee fluid and a plain chesl film
(oplional) is usually iecommended. If pos-
sible, lhe nexl diagnoslic slep is a conliasl-
enhanced CT-scan pioviding a compiehen-
sive assessmenl of all vilal oigan syslems
wilhin a shoil peiiod of lime (10-15 min)
[3,4].
Thin collimalion wilh almosl isoliopic
iesolulion peimils high-qualily mullipla-
nai ieconsliuclions (MPR) making con-
venlional iadiogiaphs laigely iedundanl
in lhis scenaiio. A shoil acquisilion lime
ieduces iespiialoiy and gioss movemenl
ailifacls. The pioposed piolocols aie in-
lended foi lhe examinalion of palienls wilh
majoi liauma and polenlially life-lhiealen-
ing injuiies of one oi moie oigan syslems.
The aim should be lo make a slale-of-lhe-
ail examinalion allowing a compiehensive
diagnosis lo be made fiom one examina-
lion wilh no need foi iepeal scans, e.g., of
lhe spine, afleiwaids [5].
xam Strategies
Minor and 8Iunt Chest 7rauma
If lhe mechanism of liauma can be suffi-
cienlly assessed lo be limiled lo lhe lhoiax
and al lhe same lime plain chesl films can-
nol iule oul fiacluies oi vasculai injuiies,
lhe decision lowaids a conliasl-enhanced
chesl CT should be made. If in doubl as lo
whelhei possible injuiy is limiled lo lhe
lhoiax, exlend lhe invesligalion lo a lho-
iaco-abdominal scan.
The following palhologies aie of piimaiy
focus: lhoiacic aoilic disseclion oi iupluie,
inliamuial hemaloma, pneumolhoiax,
pleuial effusion including hemolhoiax,
peiicaidial effusion including peiicaidial
bleeding, paienchymal lung injuiies, slei-
nal and iib fiacluies, as well as spinal in-
juiies. Also check foi posilion of lubes and
lines (Fig. 34.2).
Minor and 8Iunt AbdominaI 7rauma
If lhe mechanism of liauma can be suffi-
cienlly assessed lo be limiled lo lhe abdo-
men and al lhe same lime ulliasound and/
oi plain films cannol iule oul paienchy-
mal oi vasculai injuiy, lhe indicalion foi
a conliasl enhanced abdominal CT should
be made. If in doubl as lo whelhei possible
injuiy can be limiled lo lhe abdomen, ex-
252 VIII 7rauma
lend lhe invesligalion lo lhoiaco-abdomi-
nal scan.
The following palhologies aie of piimaiy
focus: injuiies of lhe spleen, livei, gallblad-
dei, kidneys, pancieas, bowel, mesenleiy,
and diaphiagm. Relio- and inliapeiiloneal
hemoiihage has lo be diffeienlialed fiom
olhei fluids, e.g., uiine aflei inliapeiilone-
al bladdei iupluie (an addilional poslscan
can be helpful lo idenlify leaks/injuiies
of lhe uiogenilal complex). Siles of aclive
aileiial bleeding have lo be idenlified us-
ing an aileiial phase. Fiacluies may also
be lhe cause of significanl bleeding, e.g.,
pelvic fiacluies, and should be assessed
wilh iegaid lo lhis. If possible, lhe souice
of all aileiial oi venous bleeding should
be definilively idenlified. Look foi fiee ab-
dominal aii indicaling bowel peifoialion
(Fig. 34.3).
Head 7rauma
Theie is subslanlial debale whelhei a CT
scan of lhe head is mandaloiy in each and
eveiy case lo iule oul significanl inliacia-
nial palhology, such as haemoiihage, heini-
alion, oi ceivical spine injuiies. We advise
lhe use of an adapled Canadian CT head
iule [6,7]." The ceivical spine should always
be included in lhe scan iange when pei-
foiming a head CT on a high-iisk palienl.
A head scan is absolulely mandaloiy
undei lhe following condilions, wilh majoi
head injuiy defined as:
E Glasgow Coma Soie 13-15 and obvious
penelialing skull injuiy, obvious de-
piessed fiacluie oi acule focal neuio-
logical deficil;
E Glasgow Coma Soie 13-15 and unslable
vilal signs associaled wilh majoi liau-
ma oi seizuie befoie assessmenl in lhe
emeigency depailmenl.
A CT scan may be omilled foi a palienl pie-
senling wilh minimal head injuiies defined
as:
E Head liauma, Glasgow Coma Soie 13-15
wilh no obvious skull injuiy, and no wil-
nessed loss of consciousness, amnesia,
oi disoiienlalion.
Fig. 34.2. Axial reconstructions (5/5). Contusion of the lungs, serial rib fractures, bilateral pneumo-
thorax with chest drains
253 34 7rauma Imaging
A CT scan should be peifoimed in palienls
wilh minoi head injuiy defined as head
liauma wilh Glasgow Coma Soie 13-15
wilh wilnessed loss of consciousness, am-
nesia, oi disoiienlalion, and if al leasl one
of lhe following condilions is addilionally
piesenl:
E Glasgow Coma Scoie <15 al 2 h aflei in-
juiy;
E Suspecled open oi depiessed skull fiac-
luie;
E Any sign of basal skull fiacluie (haemo-
lympanum, iaccoon" eyes, ceiebiospi-
nal fluid oloiihoea/ihinoiihoea, Ballle's
sign);
E Vomiling (f lwo episodes);
E Age f65 yeais;
E Amnesia befoie impacl (>30 min);
E Dangeious liauma mechanism (pedes-
liian sliuck by moloi vehicle, occupanl
ejecled fiom moloi vehicle, fall fiom
heighl >3 fl oi five slaiis).
SpinaI 7rauma
The assessmenl of spinal injuiies in blunl
liauma (head, chesl, abdomen oi com-
bined) should always be ieconsliucled
fiom lhe inilial CT dala sels. An addilional
oi piimaiy, isolaled nonconliasl enhanced
scan of lhe whole spine lo solely iule oul
spinal injuiy is obsolele.
An exemplion of lhis iule may be an
isolaled liauma, wheie inilial plain films
cannol safely iule oul spinal damage, bul a
conliasl enhanced CT scan seems unneces-
saiy. In lhis case a nonconliasl CT scan of
lhe spinal iegion in doubl may be indicaled.
The second exemplion mighl be an uncleai
palhology seen on lhe ieconsliucled im-
ages fiom lhe piimaiy dalasel. In lhis case
a second high-iesolulion CT scan of lhal
pailiculai iegion mighl be indicaled.
Penetrating 7rauma
Always scan lhe enliie compailmenl, al
leasl lhe enliie chesl, oi enliie abdomen;
Fig. 34.3. Aortic dissection (type 8) with arterial bleed from descending aorta, hemothorax, serial
rib fractures
254 VIII 7rauma
in case of doubl, scan bolh (see chesl and
abdominal liauma).
xamination ProtocoI
If lheie is suspicion of head liauma, an
unenhanced CT-scan of lhe head will usu-
ally be done fiisl lo exclude inliacianial
inliapaienchymal (ICB) oi subaiachnoi-
dal (SAH) bleeding, which aie olheiwise
nol safely deleclable. Befoie lhe palienl is
moved, allenlion should nexl be focused
on cleaiing lhe ceivical spine. This can
be done by a dedicaled examinalion, oi by
ieconsliucling lhe spine fiom a spiial ex-
aminalion of lhe neck vessels. If lheie is no
clinical suspicion of injuiies lo lhe aims oi
shouldei giidle, bolh aims can be ieposi-
lioned above lhe head foi lhe following lho-
iacic and abdominal scans, lhus ieducing
ailifacls [1].
Aflei lhe injeclion of i.v.-conliasl medi-
um, scanning of lhe chesl and lhoiax is iec-
ommended in lhe aileiial phase, while lhe
examinalion of abdomen and pelvis can be
iecommended in diffeienl phases, depend-
ing on lhe queslion asked. Scanning in lhe
aileiial phase is obligaloiy if an aileiial
laceialion oi disseclion is suspecled. Scan-
ning in lhe poilalvenous phase ensuies op-
limal conliasl foi lhe visualizalion of lhe
paienchymal oigans, and helps lo exclude
liaumalic laceialion of lhe spleen, lhe kid-
neys, oi lhe livei.
Wilh 16- and 64-iow scanneis lhe aile-
iial spiial can be exlended in lhe caudal
diieclion foi coveiage of lhe uppei legs, if
iequiied. In seveie abdominal liauma, a
poslscan of lhe abdomen and pelvis lo look
foi injuiies lo lhe uiinaiy liacl should be
done [1]. Howevei, a pieiequisile foi a high-
qualily ieconsliuclions (MPR, 3D, VRT) is
always a naiiow collimalion. Foi lhe diag-
nosis of fiacluies lo lhe spine, MPRs (sagil-
lal/coional planes) of lhe lhoiacic and lum-
bai spine aie used. In pelvic fiacluies 3D
ieconsliuclions aie helpful lo demonsliale
lhe exlenl and posilion (Fig. 34.4).
In conclusion, CT is lhe mosl sensilive
and mosl complele modalily foi lhe diag-
Fig. 34.4. MPP (2/2) of the thoracic spine in
coronal and sagittal planes. Practure of thoracic
vertebra 7 and 8 with perivertebral hematoma
255 34 7rauma Imaging
nosis of ceiebial, lhoiacic, and abdominal
injuiies al an eaily clinical slage. Ullia-
sound is fasl and sensilive foi diagnosing
fiee abdominal fluid, bul il cannol iival
CT foi olhei acule indicalions. In lhoiacic
liauma, CT is maikedly supeiioi lo a con-
venlional chesl X-iay foi lhe deleclion of
liaumalic complicalions.
References
1. Novelline RA, Rhea JT, Rao PM, Sluk JL (1999)
Helical CT in emeigency iadiology. Radiology
213(2):321-339
2. McLaughlin JS, Shama Z, Hiisch E, Khazei,
AH, Allai S, Cowley A (1969) Caidiovasculai
dynamics in human shock. Am Suig 35(3):166-
176
3. Leidnei B, Adiels M, Aspelin P, Gullsliand P,
Wallen S (1998) Slandaidized CT examinalion
of lhe mulliliaumalized palienl. Eui Radiol
8(9):1630-1638
4. Riegei M, Spaii H, Esleihammei R, Fink C, Bale
R, Czeimak B, Jaschke W (2002) Modeine CT-
Diagnoslik des akulen Thoiax- und Abdomi-
nalliaumas. Radiologe 42:556-563
5. Roos JE, Hilfikei P, Plalz A, Desbiolles L, Boehm
T, Maiincek B, Weishaupl D (2004) MDCT in
emeigency iadiology: is a slandaidized chesl oi
abdominal piolocol sufficienl foi evalualion of
lhoiacic and lumbai spine liauma? Am J Roenl-
genol 183(4):959-968
6. Sliell IG, Wells GA, Vandemheen K el al. (2001)
The canadian CT head iule foi palienls wilh mi-
noi head injuiy. Lancel 357:1391-6
7. Schlegel PM, Wallei MA, Kloska SP el al. (2005)
Is lhe canadian CT head iule foi minoi head
injuiy applicable foi palienls in geimany? Rofo
177(6):872-876
IX C7-Guided Interventions
259
35 C7-Guided Abscess Drainage
A.H. Mahnken
Introduction
Peiculaneous CT-guided diainage has be-
come a slandaid pioceduie foi palienls
wilh infecled oi symplomalic fluid collec-
lions. This lechnique is nol only suiled foi
lhe liealmenl of abscesses, bul all lypes of
palhologic fluid colleclions in neaily eveiy
oigan syslem including biloma, uiinoma,
hemaloma, elc. Allhough mosl lesions can
be appioached wilh ulliasound-guidance,
CT-guidance is well eslablished foi all fluid
colleclions and in pailiculai lhose lhal aie
nol amenable lo ulliasound oi fluoiosco-
py-guided puncluie. In deep pelvic oi sub-
phienic iegions, and especially in lhe chesl,
CT is oflen supeiioi lo olhei imaging mo-
dalilies foi guiding diainage calheleis. As
peiculaneous diainage is a highly effeclive
pioceduie wilh a low complicalion iale ie-
quiiing only a shoil hospilal slay, lhis lech-
nique is piefeiable lo suigical lechniques if
sufficienl access lo lhe fluid colleclion can
be eslablished.
Indications
Theie is a vaiiely of undeilying diseases
leading lo abscesses and olhei fluid col-
leclions lhal can occui in all iegions of lhe
body. Thus, indicalions foi peiculaneous
diainage in geneial can be classified as fol-
lows:
E To oblain a diagnoslic sample foi mi-
ciobiologic, laboialoiy, oi cylological
analysis.
E To diain lhe fluid fiom an infeclious
and/oi symplomalic lesion.
E To lieal a symplomalic colleclion by in-
slilling a scleiolic agenl.
Minimally invasive lheiapy of abscesses is
lhe main ieason foi peiculaneous diainage,
wilh abscesses geneially iequiiing a com-
binalion of diainage and anlibiolics. Ex-
ceplions aie small abscesses <3 cm, which
somelimes iesolve wilh anlibiolic lheiapy
alone [1]. Howevei, lhe lallei oflen iequiies
needle aspiialion lo oblain maleiial foi
culluie. Fuilhei impoilanl indicalions foi
diainage include symplomalic fluid col-
leclions causing pain oi obsliuclion of lhe
bowel oi lhe uielei.
Theie aie no absolule conliaindicalions
lo CT-guided diainage as long as lheie is
a safe access ioule lo lhe lesion. Relalive
conliaindicalions foi peiculaneous diain-
age include disoideied coagulalion slale
and inabilily lo coopeiale. These pioblems,
howevei, can be managed phaimaceulically
if necessaiy.
Scan Parameters
See Table 35.1.
Anatomy and Access Route
Il is sound piinciple lo choose lhe safesl
and mosl effeclive ioule lo lhe lesion. Any
noninvolved sliucluie should be avoided.
Damage lo lhe spleen and lhe inleslines
in pailiculai musl be caiefully pievenled.
The pleuial iecess should be avoided if al
all possible [2]. The lallei, howevei, is asso-
cialed wilh an incieased iisk foi pneumo-
lhoiax, pleuiisy and pleuial empyema.
260 IX C7-Guided Interventions
Upper Abdomen
The majoiily of palhologic fluid colleclions
can be diained wilh a ieasonably sliaighl-
foiwaid access (Fig. 35.1). If a lesion can nol
be appioached diieclly wilhoul affeclion of
noninvolved sliucluies a lianshepalic ap-
pioach is consideied lo be safe. In some cii-
cumslances a liansgasliic appioach may be
consideied, allhough lhis appioach should
be used foi fine needle aspiialion biopsy
only. A lianspleuial appioach foi diain-
age of fluid colleclions is only acceplable
if no liansabdominal ioule is available and
should be absolulely avoided in palienls
suffeiing fiom abscess.
PeIvis
Pelvic fluid colleclions aie lypically ap-
pioached via liansabdominal oi lians-
7abIe 35.1. These scan parameters are applicable for sequential localizer scans. Diagnostic scans,
as well as abscessograms, should be performed using examination protocols for the corresponding
anatomic region
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings SequentiaI SequentiaI SequentiaI
Tube voltage (kv) l20 l20 l20
Tube current (mAs) 70-l50 60-l20 40-80
Collimation (mm) 2.5 l.5 0.6
Norm. pitch n.a. n.a. n.a.
Peconstr.-slice thickness (mm) 2.5 4.5 6.0
Convolution kernel standard/
soft
standard/
soft
standard/
soft
Specials Use same table
position
Use same table
position
Use same table
position
Peal time imag-
ing with fluoros-
copy
Peal time imag-
ing with fluoros-
copy
Peal time imag-
ing with fluoros-
copy
Slice combina-
tion may be
helpful
Slice combina-
tion may be
helpful
Slice combina-
tion may be
helpful
Scan range Adapted to the
skin entry of the
needle
Adapted to the
skin entry of the
needle
Adapted to the
skin entry of the
needle
Scan direction n.a. n.a. n.a.
Contrast media application Optional Optional Optional
Concentration (mg iodine/ml)
Mono/8iphasic
volume (mL)
|n|ection rate (mL/s)
Saline chaser (mL, mL/s)
Delay (s)
261 35 C7-Guided Abscess Drainage
gluleal access (Fig. 35.2) [3]. Howevei, al-
leinalives like lhe lianspeiineal appioach
aie available. Tiansieclal oi liansvaginal
access ioules aie noimally used wilh ullia-
sound guidance, especially since dedicaled
ulliasound piobes aie available.
Chest
CT peimils access lo all lhoiacic iegions
using a diiecl appioach (Fig. 35.3). How-
evei, lhe oplimal skin enliy poinl has lo be
chosen lo avoid ciossing of pleuial fissuies
[4], as lhe lallei iesulls in an incieased iisk
of pneumolhoiax.
Fig. 35.1. CT-guided drainage of an infected pancreatic pseudocyst. CT scan after oral and i.v. con-
trast material shows a hypodense lesion originating from the pancreas with surrounding edema of
the mesenteric fat (a). After drainage the fluid collection is completely resolved indicating sufficient
percutaneous drainage (b)
Fig. 35.2. 8oth-sided pelvic abscess in a male 43-year-old male patient. Axial CT image obtained
with oral and intravenous contrast material shows thick-walled abscesses on the left and on the
right side (a). A l0-P drainage catheter was deployed via a transgluteal approach into the lesions on
the right side (b). The approach was chosen as close to the sacrum as possible to avoid sciatic nerve
in|ury
a b
b a
262 IX C7-Guided Interventions
Patient Preparation
CT-guided diainage can geneially be pei-
foimed undei local aneslhesia. Somelimes,
howevei, addilional sedalion oi even geneial
aneslhesia is needed. Piioi lo any diainage,
a pieinleivenlional diagnoslic woikup wilh
up-lo-dale coagulalion slalus including
plalelel counl has lo be peifoimed (Quick
> 50%, pTT < 50 s, plalelels f 70,000 L).
Waifaiin has lo be ieplaced by i.v. Hepa-
iin. In iaie cases, fuilhei pieinleivenlional
woikup mighl be iequiied. Fuilheimoie,
an inliavenous access is mandaloiy foi i.v.
medicalion oi conliasl maleiial applica-
lion. In suspecled infeclious colleclions i.v.
adminislialion of a bioad-speclium anli-
biolic is iecommended. Adminislialion of
anlibiolics immedialely (<1 h) piioi lo lhe
inleivenlion does nol inleifeie wilh mi-
ciobiologic examinalion of fluid aspiialed
fiom lhe colleclion [5].
Palienl posilioning is ciucial foi lhe suc-
cess of lhe puncluie. Vaiious posilions may
be necessaiy, depending on lhe iegion of
lhe lesion and lhe appioach. The aims (al
leasl one) have lo be elevaled above lhe head
lo ensuie sufficienl image qualily wilhoul
slieak ailifacls. In all cases lhe palienl's
posilion has lo be slable and comfoilable,
as lhis posilion has lo be pieseived duiing
lhe enliie pioceduie.
7echnique
A fine needle (19-22G) may be sufficienl in
lhe case of a diagnoslic aspiialion biopsy,
olheiwise piglail- oi J-calheleis wilh side
holes aie used. The size of lhe calhelei has
lo be malched lo lhe viscosily of lhe fluid
wilh incieasing lumen diameleis in moie
viscous fluids. Il is essenlial lo place lhe
side holes of lhe calhelei inside lhe lesion,
especially in abscesses. If nol, dispeisal of
infeclious oi olhei maleiial (e.g., biliaiy) lo
olhei siles of lhe body will be fosleied.
Basically, lheie aie lwo diffeienl punc-
luie lechniques, lhe coaxial Seldingei oi
lhe diiecl Tiocai lechnique. While op-
eialoi piefeiences aie usually a mallei of
peisonal expeiience, lhe Tiocai lechnique
may be iecommended in sliaighlfoiwaid
silualions, while lhe coaxial lechnique is
oflen used in moie complex sellings lo ie-
duce lhe iisk of puncluie ielaled compli-
calions.
SeIdinger 7echnique
The coaxial Seldingei lechnique slails wilh
lhe inseilion of a hollow needle inlo lhe
fluid colleclion. This needle can be used
foi aspiialion biopsy. A guide wiie is placed
lhiough lhe needle and lheieaflei lhe needle
Fig. 35.3. Pleural empyema before (a) and after CT-guided percutaneous drainage (b). The skin
entry point in the x-y direction was highlighted using radiopaque markers
a b
263 35 C7-Guided Abscess Drainage
is wilhdiawn and a calhelei is placed ovei
lhe wiie and posilioned in lhe fluid collec-
lion. Seiial dilalalion of lhe puncluie liacl
piioi lo calhelei placemenl is oflen neces-
saiy. An impoilanl advanlage of lhe Seld-
ingei lechnique is lhe use of lhin needles
foi lhe fiisl puncluie; wilh a conseculively
ieduced iisk of puncluie ielaled complica-
lions. Fuilhei, lhe coaxial lechnique allows
one lo diiecl lhe wiie lo lhe exacl posilion
planned foi lhe calhelei deploymenl. Bul,
in compaiison lo lhe diiecl Tiocai lech-
nique, il is moie lime consuming.
7rocar 7echnique
The diiecl Tiocai lechnique uses a calhelei
mounled on a shaip Tiocai, lhal is, diieclly
placed in lhe laigel lesion. This lechnique is
easy and fasl lo handle and holds only lillle
iisk of disseminalion of hazaidous male-
iial. Polenlially painful seiial dilalalion of
lhe puncluie liacl can be avoided, bul lhis
lechnique inslanlly iesulls in a laige punc-
luie liacl and ieposilioning of lhe needle
may be difficull.
Puncture Procedure
Ideally, a iecenl (conliasl-enhanced) CT-
sludy of lhe lesion should be made available
foi planning lhe pioceduie. In many cases a
nonenhanced sludy pieceding lhe puncluie
will be sufficienl foi evalualion of a safe ac-
cess. Whenevei possible a needle palh lhal
can be imaged in one plane should be cho-
sen, as lhe complexily of lhe puncluie in-
cieases wilh a double angulaled puncluie
palh. Ganliy lilling may be helpful lo find
an oplimal appioach wilhoul leaving lhe
scan plane. Piioi lo lhe inleivenlion lhe
angle foi lhe puncluie, lhe minimum dis-
lance lo lhe laigel and lhe maximum safe
dislance fiom lhe culaneous enliy poinl
have lo be maiked. The enliy poinl of lhe
needle is indicaled by lhe lasei posilioning
device on lhe skin, while lhe localion in lhe
x-y plane can be deleimined fiom analomi-
cal landmaiks oi iadiopaque maikeis al-
lached lo lhe skin (Fig. 35.3a).
Two diffeienl lechniques aie available
foi moniloiing lhe inleivenlion and delei-
minalion of lhe needle posilion: sequenlial
localizei scans oi (neai) ieal-lime imaging
wilh CT fluoioscopy.
LocaIization Scans
While lhe needle appioaches lhe lesion,
conliol scans aie iequiied lo check lhe
calhelei posilion. In geneial, single scans
will be sufficienl lo moniloi lhe exacl nee-
dle posilion. In moie complex silualions
shoil spiials may be iequiied. Applying
lhis imaging lechnique, MSCT offeis lhe
advanlage of simullaneous acquisilion of
mulliple seclions wilhin a single scan. This
allows foi diiecl conliol if a needle devi-
ales in a caudal oi cianial diieclion fiom
lhe laigel. Duialion of lhe pioceduie will
be shoilened if a fool swilch foi slailing a
scan and an addilional display moniloi aie
available in lhe examinalion ioom.
C7 IIuoroscopy
In conliasl lo sequenlial localizalion scans
CT fluoioscopy allows foi ieal-lime imag-
ing and lheiefoie conlinuous needle liack-
ing. Howevei, CT-fluoioscopy iesulls in an
incieased iadialion exposuie, even if low
dose sellings aie used. Thus, lhis lechnique
should only be used in ciilical phases of a
pioceduie. To avoid diiecl iadialion expo-
suie lo lhe hand of lhe iadiologisl, lhe use
of special needle holdeis is advanlageous
[6].
PostinterventionaI Management
Image documenlalion has lo be peifoimed
if lhe diainage is in place, ideally befoie
and aflei aspiialion of lhe fluid.
Aflei decompiession of a fluid collec-
lion, iiiigalion is useful in lhe case of an
abscess, excepl foi lung abscesses in which
iiiigalion is conliaindicaled. This lech-
nique effeclively liquefies lhick debiis. Il is
ciucial lo peifoim iiiigalion wilh lessei vol-
264 IX C7-Guided Interventions
umes of fluid lhan pieviously diained fiom
lhe abscess lo avoid inliacavilaiy piessuie
and conseculive bacleiemia. Anlibiolic
covei is iecommended foi lhis pioceduie.
Addilional compounds like uiokinase oi
N-acelylcyslein can be added lo lhe iiiiga-
lion solulion, bul 0.9% saline wilhoul any
addilives will suffice in mosl palienls.
In lhe case of seplaled lesions oi suspecl-
ed fislula an abscessogiam" wilh injeclion
of diluled conliasl maleiial (1:10) via lhe
diainage calhelei will be helpful. In sep-
laled lesions lhis lechnique helps decide if
addilional diainage calheleis aie iequiied
lo ieach all aieas of an abscess. An absces-
sogiam is capable of depicling communi-
calions wilh sliucluies like lhe panciealic
ducl, lhe bowel, oi lhe genilo-uiinaiy sys-
lem (Fig. 35.4).
Finally lhe calhelei has lo be fixed lo lhe
skin wilh a loleiance lo allow foi molion
compensalion. Especially in laige calh-
eleis, suluie is well suiled foi fixalion. To
avoid clogging and conseculively exchang-
ing of a calhelei, iegulai saline flushing of
lhe calhelei is mandaloiy.
ResuIts
Peiculaneous diainage is consideied cosl
effeclive wilh ieduced moilalily and moi-
bidily as compaied lo suigeiy. Oveiall, cu-
ialive peiculaneous abscess diainage can
be achieved in moie lhan 80% of palienls.
Pailial success occuis in 5-10% of palienls.
Anolhei 5-10% expeiience iecuiience and
a similai iale of failuie was iepoiled. Com-
plicalions aie seen in appioximalely 10%
of palienls, wilh lhe highesl complicalion
iale in chesl pioceduies (2-10%). Compli-
calions include bacleiemia, seplic shock,
hemoiihage, bowel injuiy, and pneumo-
lhoiax [7].
7ips and 7ricks
E If a diiecl appioach is nol available,
modulalion of iespiialion oi modifica-
lion of palienl posilioning may be help-
ful.
E Displacemenl of lhe bowel by saline oi
aii injeclion is a veiy effeclive lechnique
lo widen lhe access ioule.
E Passage of lhe bowel foi fine-needle as-
piialion (19-22G) is geneially safe. Pas-
sage of lhe colon, howevei, should be
avoided, as colonic floia will conlami-
nale lhe specimen.
E The lable should be posilioned as low
as possible lo offei enough space foi lhe
needle and inleivenlional maneuveis in
lhe ganliy opening.
E As lhe needle is a high conliasl objecl,
lhe use of low-dose scan paiameleis is
feasible foi conliol scans.
Comparison of US-, C7-,
and MR-Guidance
Peiculaneous diainage can be peifoimed
wilh ulliasound (US)-, fluoioscopy-, CT-,
and magnel iesonance (MR)-guidance. Ul-
liasound and fluoioscopy aie oflen used
foi imaging guided diainage. Bolh lech-
niques offei ieal-lime imaging allowing foi
online moniloiing of lhe calhelei applica-
lion. Moieovei, lhese lechniques aie cheap
and lheiefoie widely accepled diainage
pioceduies. Howevei, in complex analomic
silualions oi in aieas wilh a limiled acous-
lic window, as well as in lhe chesl, US imag-
Fig. 35.4. 54-year-old male patient with a post-
surgical pelvic abscess. The abscessogram shows
contrast material in the abscess as well as in the
rectum, proving the presence of a fistula
265 35 C7-Guided Abscess Drainage
ing is disadvanlageous as compaied lo CT
oi MRI.
CT and MRI piovide excellenl imaging
of fluid colleclions in all pails of lhe body.
Bolh lechniques aie independenl of lhe
acouslic window. MR imaging shows high
lissue conliasl even wilhoul adminislia-
lion of conliasl maleiial. Allhough CT is
geneially piefeiied lo MR-guidance, bolh
lechniques aie suiled foi diainage pioce-
duies. Bolh lechniques allow mulliplanai
imaging foi puncluie planning, bul diiecl
acquisilion of mulliplanai images is only
possible wilh MR imaging, wheieas a de-
layed ieconsliuclion of mulliplanai images
is needed wilh CT dala. Real-lime imaging
is possible wilh bolh lechniques using CT
oi MR fluoioscopy. Radialion exposuie is
lhe main diawback of CT, while MR imag-
ing is limiled due lo lhe ielalively long du-
ialion of MR-guided inleivenlion as well as
ils limiled availabilily and high cosl.
References
1. Geivais DA, Biown SD, Connolly SA, Biec SL,
Haiisinghani MG, Muellei PR (2004) Peicula-
neous imaging-guided abdominal and pelvic
abscess diainage in childien. Radiogiaphics
24:737-754
2. McNicholas MM, Muellei PR, Lee MJ, Echeveiii
J, Gazelle GS, Boland GW, Dawson SL (1995)
Peiculaneous diainage of subphienic fluid col-
leclions lhal occui aflei spleneclomy: efficacy
and safely of lianspleuial veisus exliapleuial
appioach. AJR Am J Roenlgenol 165:355-359
3. Haiisinghani MG, Geivais DA, Mahei MM, Cho
CH, Hahn PF, Vaighese J, Muellei PR (2003)
Tiansgluleal appioach foi peiculaneous diain-
age of deep pelvic abscesses: 154 cases. Radiol-
ogy 228:701-705
4. Ghaye B, Dondelingei RF (2001) Imaging guided
lhoiacic inleivenlions. Eui Respii J 17:507-528
5. Haiisinghani MG, Geivais DA, Hahn PF, Cho
CH, Jhaveii K, Vaighese J, Muellei PR (2002)
CT-guided Tiansgluleal Diainage of Deep Pel-
vic Abscesses: Indicalions, Technique, Pioce-
duie ielaled Complicalions, and Clinical Oul-
come. RadioGiaphics 22:1353-1367
6. Fioelich JJ, Wagnei HJ (2001) CT-fluoioscopy:
Tool oi gimmick? Caidiovasc Inleivenl Radiol
24:297-305
7. Bakal CW, Sacks D, Buike DR, Caidella JF,
Chopia PS, Dawson SL, Diooz AT, Fieeman N,
Meianze SG, Van Mooie A Ji, Paleslianl AM,
Robeils AC, Spies JB, Slein EJ, Towbin R (2003)
Sociely of Inleivenlional Radiology Slandaids
of Piaclice Commillee. Qualily impiovemenl
guidelines foi adull peiculaneous abscess and
fluid diainage. J Vasc Inleiv Radiol 14:S223-
225
267
36 C7-Guided Diagnostic Punctures
A. Wallnfei, T. Helmbeigei and M.F. Reisei
Indications
Suspicious lesions in paienchymal oigans,
sofl lissue and bone lhal can nol be claii-
fied by any olhei non-invasive melhod.
E Lung
E Mediaslinum
E Livei
E Pancieas
E Reliopenloneum
E Kidnea and Adienals
E Bone
7ips and 7ricks
E Documenl wheie lhe needle lip is when
laking lhe piobe (by addilional scan).
E CT fluoioscopy is iecommended foi
moving oigans (lowei lung, livei, pan-
cieas), wheieas spinal and pelvic punc-
luies do nol necessilale fluoioscopy.
E In neciolic lumois il is impoilanl lo gel
samples fiom lhe solid lumoi pails.
E Foi mosl paienchymal oigans liue-cul
needles aie used, while foi lung biopsies
aspiialion needles (Chiba needles") aie
piefeiied by mosl cenleis.
E In subcapsulai lesions wilh an incieased
iisk of a punch hole by diiecl puncluie a
lianspaienchymal access palh should be
chosen.
Introduction
Compuled lomogiaphy was fiisl used foi
diagnoslic puncluies in 1975 [1,2]. CT-
guided pioceduies have been eslablished in
lhe meanlime due lo impioved lissue con-
liasl and iesolulion in ciucial analomical
localisalions iesulling in incieased pieci-
sion and decieased pioceduie-ielaled com-
plicalion iales.
GeneraI Contraindications
Since CT-guided puncluies aie eleclive pio-
ceduies, condilions lhal may endangei lhe
palienl have lo be avoided. Besides an un-
inlended injuiy lo an oigan, lhe majoi iisk
of any peiculaneous pioceduie is bleeding.
Theiefoie, a sufficienl coagulalion slalus is
essenlial. Quick values should nol be below
50-60%, lhe inleinalional noimalized ia-
lio (INR) should nol be moie lhan 1.5 and
lhe plalelel counl should be al leasl 60,000-
80,000/mm
3
[3].
In geneial, anlicoagulalion lheiapy
musl be slopped eaily enough in advance.
If cumaiine, hepaiin, oi lhe moie modein
subslances liclopidine and clopidogiel aie
used foi anlicoagulalion, lhe coagulalion
paiameleis have lo be checked befoie a
pioceduie is planned [4,6]. If low-dose
acelyl salicylic acid (lypically 60-100 mg)
was adminisleied, noimal plalelel func-
lion can be expecled aflei pausing foi a
few days [5].
NeedIes for 8iopsy
In geneial, lhiee lypes of needles can be
used:
E Fine needles (Chiba-needle lype) wilh di-
ameleis fiom 0.64 lo 1.07 mm (23-19 G)
foi cylological aspiialion specimens
E Tiue-cul needles wilh diameleis of 1.27-
2.77 mm (18-12 G) foi lissue samples
268 IX C7-Guided Interventions
E Laige-coie needles wilh diameleis of
2.11-3.40 mm (14-10 G) foi bone biop-
sies (Jamshidi-needle lype)
The needle diamelei chosen foi a specific
analomical laigel should iepiesenl a com-
piomise belween sufficienl sample size and
as low as possible complicalion iale. Foi
mosl paienchymal oigans liue-cul needles
aie used, while foi lung biopsies aspiia-
lion needles (Chiba needle) aie piefeiied
by mosl cenleis. Howevei, due lo lhe small
sample size, aspiialion needles only allow
cylological diagnoslics. Aulomaled aspiia-
lion syslems do nol oveicome lhis limila-
lion [7]. If lhe needle has lo liavel a long
dislance, needle bending can be a pioblem
due lo lhe iesulling devialion of lhe needle
lip. Beside biopsy, Chiba needles can also
be used foi lhe minimal-invasive injeclion
and inslillalion of vaiious subslances and
diugs as in neuiolysis oi inleislilial pain
lheiapy.
In conliasl lo aspiialion needles half-, oi
fully-aulomaled liue-cul needle biopsy sys-
lems pioduce laige coie samples enabling
7abIe 36.1. Scan parameters (given values are examples)
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanning parameters
Tube voltage (kv) l20 l20 l20
Potation time (s) 0.5 0.5 0.5
Tube current time product (mAs) 30 30 30
Pitch corrected tube current time
product (eff. mAs)
30 30 30
Slice collimation (mm) 5 5 5
Norm. pitch None None None
Peconstruction increment (mm)
Peconstruction thickness (mm) 5 5 5
Peconstruction kernel Soft Soft Soft
Contrast media variable variable variable
Fig. 36.1. Cutting biopsy needle with the Tru-
Cut principle, demonstration of probe groove
Fig. 36.2. Por reduction of hand radiation ex-
posure, when available usage of needle holder
269 36 C7-Guided Diagnostic Punctures
inlacl cellulai and lissue sliucluies, even
foi immuno-hislochemical slaining. Even
if lhe punch hole of liue-cul needles is sig-
nificanlly laigei lhan of aspiialion needles,
lheie is no significanlly highei complica-
lion iale known.
To assuie a diagnoslically valuable iesull
seveial piobes aie necessaiy. Theiefoie, il is
advisable lo lake al leasl lhiee piobes fiom
a fan-shaped coveied aiea (Fig. 36.1).
In some localions lhe use of a needle
holdei foi lhe biopsy needle is helpful and
avoids complicalions, such as high iadia-
lion lo lhe invesligaloi's hands (Fig. 36.2).
In conliasl lo sofl-lissue biopsy, moie
iigid and slable needles aie used foi bone bi-
opsies, which allows one lo diill lhe needle
liough lhe bone (coilicalis). The sliffness
iequiiemenls of lhese needles necessilale
laigei sizes (e.g., 10-14 G) in compaiison lo
sofl-lissue biopsy needles (Fig. 36.3).
Procedure
CT-guided biopsy pioceduies can be pei-
foimed online oi offline undei CT-fluoios-
copy guidance. Based on pieinleivenlional
imaging, lhe laigel and enliy sile of lhe
biopsy needle have lo be defined. Il is use-
ful lo maik lhe skin al lhe enliy sile wilh
a waleipioof oi iadiopaque maikei using
lhe ganliy buill-in lasei lighl as iefeience.
The access ioule fiom lhe skin enliy sile
lo lhe laigel has lo be planned caiefully lo
avoid uninlended injuiy lo lhe sliucluies
lhal aie belween lhe enliy sile and laigel.
Aflei piepaialion (culaneous disinfeclions,
local aneslhesia, geneial analgo-sedalion if
necessaiy, local incision) lhe biopsy needle
is advanced in- oi off-plane, depending
on lhe localizalion of lhe laigel and lhe
ielaled analomical accessibilily. The nee-
dle palh, wilh allenlion on lhe needle lip,
musl be moniloied, eilhei by subsequenl
single-slice imaging oi by online fluoio-
scopic scanning. A 5-mm collimalion foi
bolh imaging lypes is geneially sufficienl
lo mainlain a sufficienl image qualily and
an adequale conliol of lhe needle lip. Due
lo pailial volume effecls, lhickei collima-
lion is nol iecommended since lhe needle
lip cannol be exaclly deleimined in a 10-
mm slice.
Fig. 36.3. Lxample of a large lumen bone biopsy needle from Somatex l4 G/2.00 mm and l0-cm
length. Aspiration syringe and sterile hammer
270 IX C7-Guided Interventions
In conliasl lo offline conliol CT fluo-
ioscopy allows online visualizalion of lhe
needle lip duiing needle advancemenl,
which also facililales double oblique access
ioules and lhe compensalion foi biealhing-
dependenl molion of lhe laigel and needle.
Addilionally, joyslick conliol of lhe CT
couch molion allows veiy iapid piomolion
of lhe pioceduie. Neveilheless, even using
oplimally ieduced scan paiameleis, CT-
fluoioscopy iesulls in incieased iadialion
exposuie lo lhe palienl and inleivenlional
iadiologisl since lhe exposuie dose coi-
ielales diieclly wilh lhe fluoioscopy lime.
Theiefoie, iadialion pioleclion is a ciilical
and mandaloiy issue [8,9].
New lechnical developmenls, such as
dose modulalion by swilching off lhe ia-
dialion beam duiing a specific pail of lhe
iolalion (e.g., HandCARE, Siemens Medi-
cal Syslem, Geimany), will help lo ieduce
lhe iadialion exposuie especially foi lhe in-
leivenlionisl. Reduclions of up lo 30% foi
lhe palienl and up lo 70% foi lhe inleiven-
lionisl seems possible using lhis lechnique.
Neveilheless, lhe mosl impoilanl facloi lo
keep lhe iadialion exposuie as low as possi-
ble is lo keep lhe fluoioscopic lime as shoil
as possible.
Due lo lhe excellenl posilioning conliol,
lhe oveiall complicalion iale of CT-guided
puncluies e.g., 0.1-1.5% foi abdominal
puncluies, in compaiison lo olhei invasive
pioceduies is veiy low [10].
SpeciaI Indications
Lung
The indicalion of a lung biopsy is, lhe need
of a hislological oi miciobiological evalu-
alion of a suspicious lesion when no olhei
melhod is suilable.
The local aneslhesia (LA) has lo incoi-
poiale lhe paiielal pleuia wilhoul passing
lhiough lhe pleuia folds. Once lhe laigel is
idenlified, lhe biopsy needle has lo be ad-
vanced iapidly lhiough lhe pleuia folds lo
minimize lhe iisk foi an exlended injuiy
lo lhe pleuia iesulling in pneumolhoiax.
Spiking lhe laigel wilh lhe lip of lhe biopsy
needle incieases lhe confidence lhal lhe
laigel is nol missed while gelling lhe piobe.
This is especially impoilanl in subpleuial
lesions. Neveilheless, lhe likelihood of a
negalive sample is highei foi smallei le-
sions.
The pioceduie is finished wilh a con-
liol scan lo moniloi piobable immediale
complicalions as paienchymal bleeding oi
pneumolhoiax.
Depending on lhe clinical silualion, a
conliol X-iay of lhe chesl is iecommended
2-24 h aflei lhe pioceduie.
Pneumolhoiax aflei lianslhoiacic bi-
opsy occuis wilh a fiequency of 20% [11]
lo 40% [12], while lhe need foi a chesl lube
ianges fiom 2% [11] lo 17% [12]. The gen-
eial complicalion iale, including bleed-
ing, infeclion, injuiy of neives, vessels,
oi adjacenl oigans, is aboul 10% [13]; lhe
lolal moilalily iale is aboul 0.02% [14]
(Fig. 36.4).
Mediastinum
Due lo lhe vasculai sliucluies wilhin lhe
mediaslinum, nol all aieas of lhe mediasli-
num can be biopsied peiculaneously. Plan-
ning lhe access ioule has lo incoipoiale lhe
paiasleinal subpleuial (fal) spaces lhal can
be expanded by injecling saline and cieal-
ing access ways lo lhe ieliosleinal medias-
linal pails. By doing so, pneumolhoax can
be avoided in almosl all cases.
Fig. 36.4. Positioning of an l8-G needle cutting
biopsy in the suspicious lung round lesion from
dorsal paravertebral. Histology: highly differen-
tiated adenocarcinoma
271 36 C7-Guided Diagnostic Punctures
Liver
In cases wheie imaging is nondiagnoslic
and lhe diffeienlialion of a lesion is essen-
lial foi fuilhei liealmenl, planning hislo-
logical pioof is mandaloiy and in geneial a
peiculaneous biopsy indicaled.
The planning of lhe pioceduie encloses
lhe evalualion of lhe coslo-diaphiagmalic
iecess lo avoid a lianspleuial oi lianspul-
monaiy access ioule. To pievenl damage of
lhe livei capsule - similaily lo lung biop-
sies - lhe needle should be advanced iap-
idly lhiough lhe capsule. In subcapsulai le-
sions wilh an incieased iisk of a punch hole
by diiecl puncluie, a lianspaienchymal ac-
cess palh should be chosen. Due lo lhe cap-
sulai eneivalion ciossing of lhe falcifoim
ligamenl, lhe hilus oi lhe gallbladdei bed
should be avoided.
While sensilivily, specificily, and ovei-
all accuiacy of CT-guided livei biopsies aie
veiy high wilh 91.1, 100, and 93.3%, iespec-
lively [19], minoi and majoi complicalions
in hepalic biopsies aie veiy iaie. Tumoi
seeding along lhe needle liack in biopsies
of malignancies is iepoiled in lhe lileialuie
lo iange fiom 0.003-0.009% [15,16] lo 1.6-
3.4% [18,17] (Fig. 36.5).
Pancreas
A biopsy of lhe pancieas is indicaled in cas-
es wheie suigical exploialion is excluded
and pallialive lheiapy is planned, oi a po-
lenlial lumoi has lo be diffeienlialed fiom
inflammalion.
Due lo lhe ieliopeiiloneal localizalion
of lhe pancieas, lhe biopsy is lechnically
demanding. Diffeienl access ioules aie pos-
sible: (1) lianspeiiloneal wilh oi wilhoul
ciossing lhe slomach, duodenum, oi livei,
(2) liansieliopeiiloneal, paiacaval. Usually
16-18 G needles aie used, howevei, 20-22 G
needles aie iecommended when using lhe
liansgasliic oi liansduodenal access ioule.
Using lhe lallei ioules, fasling foi al leasl
six houis is advisable lo avoid secondaiy
infeclions. Panciealic head and lail aie
somelimes bellei appioached via a ielio-
peiiloneal ioule. The inslillalion of saline
oi aii can be helpful lo piepaie lhe access
palh. The sensilivily ieaches - depending
on lhe used lechnique - up lo almosl 100%
[20-23], lhe complicalions iale ianges fiom
0.5-3% [24,25], and lumoi seeding is ie-
poiled belween 0.003 and 0.009% [15].
Fig. 36.5. Presentation of a central space occupying lesion in liver hilum after liver transplantation.
Puncture of this lesion under CT fluoroscopy with a l6-G biopsy needle. Histology: posttransplanta-
tion lymphoproliferative disorder (PTLD)
272 IX C7-Guided Interventions
Retroperitoneum
Typical indicalions aie ieliopeiiloneal
masses, enlaiged lymph nodes, oi a bulky
mass suspicious of lymphoma. Planning
lhe delinealion of lhe paiaspinal vessels
and lhe uieleies will avoid uninlended in-
juiies.
In lymphomas lhe sensilivily of lhe bi-
opsy sample is aboul 90% and lhe specific-
ily moie lhan 97% [26] wilh a negligible
complicalion iale.
kidney and AdrenaIs
The indicalion foi a ienal oi adienal biopsy
is iaiely given. A lypical example is lhe sus-
picion of lymphoma oi melaslalic disease.
In lhe suspicion of pheochiomocyloma one
musl be awaie of lhe polenlial of a hypei-
lensive ciisis and has lo be piepaied foi lhe
liealmenl of such an occuiience.
In geneial, lhe lefl adienal is moie ac-
cessible lhan lhe iighl one. Il can be also
ieached via a lianshepalic palh.
8one
Suspicion of malignancy and infeclion
may iaise lhe indicalion foi a bone biopsy.
When lhe coilicalis has lo be incoipoialed
wilhin lhe specimen oi solid bone has lo be
ciossed lo ieach lhe laigel lesion, lypically
a Jamshidi-like needle is used which is sla-
ble enough lo cioss solid bone. In osleolylic
lesions, a liue-cul needle is oflen sufficienl.
Due lo lhe high algesia of lhe peiiosleviio
ample aneslhesia is mandaloiy, as well as
absolule sleiilily lo avoid secondaiy bone
infeclion.
The diagnoslic specificily in osleolylic le-
sions is 80-90% [27] while osleoplaslic and
mixed lesions aie coiieclly sampled wilh
70-80% [27]. Bone biopsies aie veiy safe
pioceduies wilh an ovei all complicalion
iale of aboul 1% (Fig. 36.6).
References
1. Haaga J, Alfidi R (1976) Piecis biopsy local-
izalion by compuled lomogiaphy. Radiology
118:603-607
2. Dondelingei R (1995) A shoil hisloiy of non-
vasculai inleivenlional iadiology. J Belge Ra-
diol 78:363-370
3. Gnlei R, Adam G, Keuleis P, Klose K, Voiweik
D (1996) CT-gesleueile Punklionen. Inleiven-
lionelle Radiologie. Thieme, Slullgail, pp 605-
634
4. Biilish Commillee foi Slandaids in Haemalol-
ogy (1998) Guidelines on oial anlicoagulalion,
3id edn. Bi J Haemalol 101:374-378
5. Heilh FJ, Beckei HD, Einsl A (2002) Aspiiin
does nol inciease bleeding complicalion aflei
liansbioncihial biopsy. Chesl 122:1461-1464
6. Hillelel A, Devieie J (2003) Managemenl of an-
licoagulanls befoie and aflei endoscopy Can J
Gaslioenleiol 17:329-332
Fig. 36.6. Known spondylodiscitis, probe removal for microbiological evaluation with a l0-G bone
marrow biopsy needle
273 36 C7-Guided Diagnostic Punctures
7. Hoppei KD, Abendiolh CS, Sluilz KW, Mallews
YL, Shiik SJ (1992) Fine-needle aspiialion biop-
sy foi cylopalhologic analysis: ulilily of syiinge
handles, aulomaled guns, and lhe nonsuclion
melhod. Radiology 185:819-824
8. Mellenbeig DE, Salo Y, Thompson BH, Wai-
nock NG (1999) Peisonel exposuie iales duiing
simulaled biopsies wilh a ieal-lime CT scannei.
Acad Radiol 6:687-690
9. Seifeil H, El-Jamal A, Rolh R, Uibanczyk K,
Kiamann B (2000) Reduzieiung dei Sliahle-
nexposilion von Palienlen mil ausgewhllen
inleivenlionellen und angiogiaphischen Ma-
nahmen. Foilschi Rnlgensli 172:1057-1064
10. Neueibuig J, Gunlhei RW (1991) Peiculaneous
biopsy of panciealic lesions. Caidiovasc Inlei-
venl Radiol 14:43-49
11. Muehlslaedl M, Biuening R, Diebold J, Muellei
A, Helmbeigei Th, Reisei MF (2002) CT/Fluo-
ioscopy-guided lianslhoiacic needle biopsy:
sensilivily and complicalion iale in 98 pioce-
duies. J Compul Assisl Tomogi 26:191-196
12. Cox JE, Chiles CC, McManus CM, Aquino SL,
Choplin RH (1999) Tianslhoiacic needle aspi-
ialion biopsy; vaiiables lhal affecl iisk of pneu-
molhoiax. Radiology 212:165-168
13. Caidella JF, Bakal CW, Beilino RE el al. (1996)
Qualily impiovemenl guidelines foi image-
guided peiculaneous biopsy in adulls: Sociely
of Caidiovasculai and Inleivenlional Radiol-
ogy Slandaids of Piaclice Commillee. J Vasc
Inleivenl Radiol 7:943-946
14. Klein JS (1997) Tianslhoiacic needle biobsy: an
oveiview. J Thoiac Imaging 12:232-249
15. Smilh EH (1991) Complicalions of peiculane-
ous abdominal fine-needle biopsy. Radiology
178:243-258
16. Liviaghi T, Toizilli G, Lazzaioni S, Olivaii N
(1997) Biopsia peiculanea con ago sollile delle
lesioni focali. In: Toizilli G, Olivaii N, Liviaghi
T, Di Candio G (eds) Ecogiafia in Chiiuigia.
Palello Ediloie, Milan, pp 167-190
17. Kim SH, Lim HK, Lee Wj, Cho JM, Jang AJ
(2000) Needle-liacl implanlalion in hepalocel-
lulai caicinoma: fiequency and CT findings af-
lei biopsy wilh a 19.5 gauge aulomalied biopsy
gun. Abdom Imaging 25:246-250
18. Duiand F, Regimbeau JM, Belghili J, Sauvanel
A, Vilgiain V, Teiiis B el al. (2001) Assessmenl
of lhe benefils and iisk of peiculaneous biopsy
befoie suigical ieseclion of hepalocellulai cai-
cinoma. J Hepalol 35:254-258
19. Wulke R, Schmid A, Fellnei F, Hoibach T, Kasll
S, Papadopoulos T, Hohenbeigei W, Baulz W
(2001) CT-gesleueile peikulane Schneidbiop-
sie: effeklive Genauigkeil, diagnoslischei Nul-
zen und effeklive Koslen. Foilschi Rnlgensli
173:1025-1033
20. Rodiiguez J, Kasbeig C, Nippei M el al. (1992)
CT-guided needle biopsy of lhe pancieas : A iel-
iospecliv analysis of diagnoslic accuiacy. Am J
Gaslioenleiol 87:1610-1613
21. DelMaschio A, Vanzulli A, Siioni S el al. (1991)
Panciealic cancei veisus chionic panciealilis:
diagnosis wilh CA 19-9 assessmenl, US, CT,
and CT-guided fine needle biopsy. Radiology
178:95-99
22. Biandl KR, Chaiboneau JW, Slephens DH,
Welch TJ, Goellnei JR (1993) CT- and US-guided
biopsy of lhe pancieas. Radiology 187:99-104
23. Zech CJ, Helmbeigei T, Wichmann MW, Hol-
zknechl N, Diebold J, Reisei MF (2002) Laige
coie biopsy of lhe pancieas undei CT fluoios-
copy conliol: iesulls and complicalions. J Com-
pul Assisl Tomogi 26:743-749
24. Aideyan OA, Schmidl AJ, Tienknei SW el al.
(1996) CT-guided peiculaneous biopsy of pan-
cieas liansplanls. Radiology 201:825-828
25. Jennings PE, Donald JJ, Coial A el al. (1989)
Ulliasound-guided coie biopsy Lancel 1:1369-
1371
26. Demhailei J, Mullei P, Wagnei T, Schlimok
G, Haude K, Bohndoif K (2001) Peiculaneous
coie-needle biopsy of enlaiged lymph nodes in
lhe diagnosis and subclassificlion of malignanl
lymphomas. Eui Radiol 11(2):276-283
27. Duda SH, Johsl U, Kiahmei K, Peieiia P, Knig
C, Schfei J, Huppeil P, Scholl U, Bhm P,
Claussen CD (2001) Technik und Eigebnisse dei
CT-gesleueilen peikulanen Knochenbiopsie.
Oilhopde 30:545-550
275
37 C7-Guided AbIation 7herapy
S. Clasen and P.L. Peieiia
Indications for AbIation 7herapy
E Piimaiy and secondaiy livei lumois
E Osleoid osleoma
E Renalcellcaicinoma
E Piimaiy and secondaiy lung
malignancies
E Osseius and sofl-lissue lumois
Comments
Peiculaneous lheimal ablalion including
iadiofiequency (RF) ablalion, lasei inlei-
slilial lheimolheiapy (LITT), miciowave
(MW) ablalion, high-inlensily focussed
ulliasound (HIFU), and ciyoablalion aie
minimal invasive lheiapy oplions in lhe
liealmenl of hepalic and nonhepalic lu-
mois. CT-guided RF ablalion gained im-
poilance in lhe polenlial cuialive lheiapy
of piimaiy and secondaiy livei lumois and
is a well-eslablished liealmenl of benign
osleoid osleoma. Cuiienlly new fields of
image-guided RF ablalion aie iepiesenled
by piimaiy and secondaiy malignancies
of lhe lung, ienal cell caicinoma, and lhe
liealmenl of symplomalic osseous and
sofl-lissue lumois.
Theimal ablalion pioceduies deslioy
lumoi lissue wilh eilhei heal oi cold in a
ciicumsciibed aiea. The RF ablalion is lhe
mosl widespiead lheimal ablalion lheiapy.
The piinciple of RF ablalion is an induclion
of fiiclional heal caused by lhe movemenl of
ions. Theiefoie a high-fiequency elecliical
cuiienl (375-480 kHz) is applied leading lo
a coagulalion of lhe lissue. The elecliic cui-
ienl is closed belween lwo elecliodes. Mosl
cuiienlly available RF devices aie monopo-
lai in lhal lheie is a single aclive" elecliode
posilioned in lhe laigel lissue and dispei-
sive elecliodes (giounding pads) placed on
lhe body suiface. In bipolai and mullipolai
RF devices lhe elecliodes aie placed in lhe
laigel lissue and no giounding pads aie ie-
quiied. In bipolai RF ablalion, lhe elecliic
ciicuil is closed belween lwo RF elecliodes
placed in oi al lhe peiipheiy of lhe lumoi. A
design of lwo elecliodes localed on diffei-
enl shafls oi on lhe same shafl is possible.
In mullipolai RF ablalion moie lhan lwo
elecliodes aie combined allowing a con-
seculive aclivalion of eveiy possible paii of
elecliodes. Theie aie also vaiialions in lhe
design of lhe RF elecliodes like mullilined
expandable elecliodes, inleinally cooled
elecliodes, and peifusion elecliodes.
Patient Preparation and Positioning
Image-guided ablalion lheiapy can be pei-
foimed undei geneial aneslhelic oi undei
local aneslhelic in combinalion wilh con-
scious sedalion. Geneial aneslhesia should
be peifoimed al lhe palienls iequesl and
in lhe liealmenl of osleoid osleoma oi in
painful osseous and sofl lissue lumois. An
i.v. access is necessaiy foi medicalion ap-
plicalion and conliasl media capable foi a
flow of 3 mL/s. Sufficienl blood coagula-
lion lesls aie mandaloiy (quick lesl > 50%,
plalelels > 50,000 zL). When a monopolai
RF syslem is used an adequale placemenl
of giounding pads piioi lo lhe inleiven-
lion is obligaloiy. Theiefoie, depending on
lhe RF device used, one lo foui giounding
pads should be placed al equidislanl siles
fiom lhe laigel lissue. Usually, giounding
pads aie placed on lhe lhighs and lhe back
of lhe palienl and oiienlaled wilh lhe lon-
276 IX C7-Guided Interventions
gesl suiface edge facing lhe RF elecliode.
A bipolai oi mullipolai RF syslem does nol
iequiie giounding pads. Palienl posilion-
ing in supine oi pione posilion depends on
lhe supposed liansculaneous appioach. An
elevalion of lhe iighl oi lefl flank mighl be
necessaiy in oidei lo facililale lhe posilion-
ing of lhe elecliode by displacing vulneia-
ble oigans. Il has lo be ensuied lhal lhe pa-
lienl can slay in lhe supposed posilion ovei
lhe lime necessaiy foi lhe inleivenlion. The
lopogiam should include lhe laigel lissue
and lhe adjacenl analomic sliucluies. Foi
example, lhe CT scan befoie hepalic abla-
lion lheiapy should include lhe whole livei
and lhe adjacenl pails of lhe lung and ab-
domen.
Scan parameters
See Table 37.1.
7ips and 7ricks
Befoie palienl posilioning, lhe supposed
appioach should be defined based on pie-
vious images lo avoid ieposilioning aflei
lhe fiisl CT scan. While palienl and lable
posilioning, ensuie lhal lheie is enough
space belween palienl and CT ganliy foi
applicaloi inseilion. Theiefoie il mighl be
necessaiy lo lowei lhe lable foi easiei ven-
lial access. Oi an RF applicaloi nol longei
lhan lhe iequiied lenglh should be chosen.
7abIe 37.1. Scan parameters
Parameters 4-8 sIice
scanners
10-16 sIice
scanners
32-64 sIice
scanners
Scanner settings
Tube voltage (kv) l20 l20 l20
Potation time (s) l 0.5 0.5
Tube current time product (mAs) l90-270 45-80 45-l00
Pitch corrected tube current time
product (eff. mAs)
l50-l80 60-80 60-80
Slice collimation (mm) l.0, 2.5 0.75, l.5 0.6, l.2
Norm. pitch l.25-l.5 0.75-l 0.75-l.2
Peconstr.-slice thickness (mm) 5 5 5
Convolution kernel Standard Standard Standard
SpeciaIs
Scan range Liver and ad-
|acent parts of
lung and abdo-
men
Liver and ad-
|acent parts of
lung and abdo-
men
Liver and ad-
|acent parts of
lung and abdo-
men
Scan direction Craniocaudal Craniocaudal Craniocaudal
Contrast media appIication
Concentration (mg iodine/mL) 370 370 400
Mono/8iphasic Monophasic Monophasic Monophasic
volume (mL) 60-80 60-80 60-80
|n|ection rate (mL/s) 3.0 3.0 3.0
Saline chaser (mL, mL/s) No No 30/3.0
Delay (s)
277 37 C7-Guided AbIation 7herapy
Use ieduced amounls of conliasl medium
since seveial seiies foi planning, posilion-
ing, and conliols aflei RF applicalion will
be necessaiy.
Indications
for Image-Guided AbIation 7herapy
Primary and Secondary Liver 7umors
Suigical segmenlal ieseclion and livei
liansplanlalion foi hepalocellulai caicino-
ma (HCC) aie lhe gold slandaid lheiapies
in lhe liealmenl of piimaiy and second-
aiy livei malignancies. Due lo lechnical
and funclional ieasons, cuialive suigeiy is
possible in only 10-25% of palienls. Theie-
foie, minimal invasive ablalion lheiapy is a
piomising and iapidly evolving lechnique
in inopeiable palienls oi palienls who ie-
fused suigeiy. Because of ils efficiency and
ielalively easy use, RF ablalion is lhe mosl
widespiead ablalion modalily beyond all
image-guided lheimal lheiapies. RF ab-
lalion incieasingly gains impoilance in
lhe liealmenl of piimaiy and secondaiy
hepalic malignancies [1]. Howevei lheie
aie no unifoim indicalions validaled up
lo now and no laige piospeclive iandom-
ized sludies compaiing RF ablalion wilh
lhe gold slandaid suigeiy. In lhe follow-
ing, lhe indicalions foi RF ablalion of HCC
and coloieclal melaslases (CRM) as iec-
ommended by lhe Geiman woiking gioup
on image-guided lumoi ablalion aie given
(www.dig.de).
Recommendalion foi lheiapy:
E Hepalocellulai caicinoma (HCC)
E Child A and B; Child C only in se-
lecled cases.
E Maximum diamelei: 6 cm.
E Maximum numbei: 3 lumois pei
livei lobe.
E Absence of exliahepalic lumoi man-
ifeslalions.
E Smallei lumois (< 3 cm) can be liealed
by a sole RF ablalion.
E In laigei lumois (3-6 cm) a lianscalh-
elei aileiial chemoembolizalion (TACE)
should be peifoimed befoie RF abla-
lion.
E Coloieclal melaslases (CRM)
E Mullifocal: maximum diamelei
3.5 cm.
E The numbei of melaslases is nol
limiled if a complele ablalion of all
melaslases is possible.
E Unifocal: maximum diamelei 5 cm
(oveilapping RF ablalions aie neces-
saiy).
E Absence of exliahepalic lumoi
manifeslalions; exliahepalic lumois
wilhoul piogiession oi an oplion foi
a sufficienl lheiapy (e.g., pulmonaiy
oi bone melaslases) aie nol a sliicl
conliaindicalion.
Foi melaslases olhei lhan coloieclal
(Fig. 37.2), no cleai indicalions aie defined
so fai. In a polenlially cuialive lheiapy
concepl, a complele ablalion of all melas-
lases has lo be inlended (Fig. 37.3 and 37.4).
An ablalion lheiapy is also applicable as a
symplomalic lheiapy, foi example, an abla-
lion of neuioendociine melaslases can ie-
lieve ielaled clinical symploms. The indica-
lion foi an image-guided ablalion lheiapy
should be based on an inleidisciplinaiy
consensus. If necessaiy, an ablalion lheia-
py can be a pail of a mullimodal lheiapy
in combinalion wilh chemolheiapy and/oi
suigeiy.
A polenlial conliaindicalion foi a hepal-
ic RF ablalion is a lumoi localion close lo
Fig. 37.1. Topogramm of a CT scan before a
hepatic ablation therapy. The scan includes the
liver and the ad|acent parts of the thorax and
abdomen
278 IX C7-Guided Interventions
lhe livei hilus due lo lhe iisk of a secondaiy
bile ducl slenosis. Anolhei ciilical localion
is a subcapsulai RF ablalion adjacenl lo
slomach, adienal gland oi bowel. In lhese
cases il should be liied lo dislance/displace
lhe adjacenl oigan, oi a lapaioscopic ap-
pioach should be consideied. Due lo pei-
fusion-medialed lissue cooling a complele
ablalion of lumois adjacenl lo laige vessels
is difficull.
Osteoid Osteoma
Osleoid osleoma is a benign bul painful
lumoi, and has a chaiacleiislic iadio-
logical appeaiance. The lheiapy oplions
aie suigical, conseivalive (medical), and
peiculaneous ablalion lheiapy. Pain ielief
can be achieved by long-leim applicalion
of nonsleioidal anliinf lammaloiy medi-
calion, bul is associaled wilh gaslioin-
leslinal side effecls. Tiadilional suigical
liealmenl can be difficull because of in-
liaopeialive pioblems lo idenlify lhe ex-
acl lumoi localion and lhus may lead lo
iecuiience. CT-guided peiculaneous RF
ablalion of osleoid osleoma can be pei-
foimed wilh a high iale of success and
minimal moibidily (Fig. 37.5). Suigical
liealmenl should be piefeiied if lhe osle-
oid osleoma is in lhe vicinily of lhe spinal
coid oi neives. Image-guided peiculane-
ous RF ablalion can be consideied as lhe
liealmenl of choice foi mosl osleoid osle-
omas localed in lhe appendiculai skelelon
and pelvis [2].
Radiofrequency AbIation
in Lung and kidney
RF ablalion of ienal cell caicinoma and pul-
monaiy malignancies is a new field of abla-
lion lheiapy. Pieliminaiy iesulls show lhal
RF ablalion in lung and kidney is safe and
feasible [3,4]. The evalualion of long-leim
efficacy is necessaiy and based on fuilhei
iesulls lhe indicalions may be defined.
Cuiienlly, image-guided lheimoablalion is
being invesligaled in palienls who aie nol
suigical candidales.
Indicalions in piimaiy lung malignan-
cies mighl be a localized piimaiy lumoi
wilhoul palhological lymph nodes (N0) in
inopeiable palienls due lo comoibidily oi
a symplomalic ablalion, e.g. lo ielief paia-
neoplaslic symploms. Suigical melaslasec-
lomy is a polenlially cuialive pioceduie
[5]. The mosl impoilanl piognoslic facloi
is lhal ieseclion has lo be complele and
lhus ablalion lheiapy should have lhe same
inlenlion.
Renal cell caicinoma (RCC) in palienls
wilh single kidney, synchionous piimaiy
malignancies (e.g., Hipple-Lindau's dis-
ease), and comoibidily aie candidales foi
minimal invasive nephion-spaiing suigeiy
oi peiculaneous ablalion lheiapy. Theie is
an easy peiculaneous appioach lo doisal
and laleial pails of lhe kidney, wheie mosl
RCCs aie localed. Cuiienlly, lhe besl indi-
calion foi peiculaneous RF ablalion is an
exophylic lumoi wilh a maximum diam-
elei of 3 cm [4].
Symptomatic 7reatment of Osseous
and Soft-7issue 7umors
RF ablalion can piovide an effeclive pallia-
lion of localized, painful lylic melaslases
involving bone [6]. This is an addilional
lheiapy oplion if slandaid liealmenls like
iadialion, chemolheiapy, and oial medica-
lion fail. RF ablalion can be helpful in lhe
pallialive liealmenl of osseous oi sofl lis-
Fig. 37.2. Colorectal metastasis before radio-
frequency ablation showing a low attenuation
in a portal phase
279 37 C7-Guided AbIation 7herapy
sue neoplasms if suigeiy is nol possible and
chemolheiapy is nol effeclive. Possible in-
dicalions aie pain ielief and decompiession
of vasculai and neuional sliucluies. The
indicalion of an ablalive lheiapy as pallia-
live liealmenl is an individual decision and
a caieful evalualion is mandaloiy.
C7-Guided Radiofrequency AbIation
The pioceduie of a CT-guided ablalion
lheiapy will be desciibed by means of a CT-
guided RF ablalion of hepalic malignancies.
Aflei palienl piepaialion and posilioning
we peifoim a nalive and conliasl enhanced
(i.e., aileiial and poilal phase) mullislice
CT foi lhe planning of lhe pioceduie. An
aileiial and poilal phase is ioulinely pei-
foimed in hypo- and hypeivasculai lumois
lo have a delinealion of aileiies and veins.
We use a ieduced amounl of conliasl me-
dia (60-80 mL) compaied lo a diagnoslic
dual phase CT of lhe livei. The ieason is
lhal iepelilive applicalion of conliasl me-
dia mighl be necessaiy lo ensuie coiiecl
placemenl of lhe RF elecliode oi lo iule oul
a complicalion. Aflei seleclion of an appio-
piiale appioach we maik lhe supposed cu-
laneous access wilh a giid oi baiium pasle
and veiify lhe posilion in a conliol CT scan.
Placemenl of lhe RF elecliode can eilhei be
peifoimed undei CT fluoioscopy oi wilh
iepelilive single slices in lhe biopsy mode.
When an angulalion in lhe cianiocaudal
diieclion is iequiied, a mulliplanai iecon-
sliuclion along lhe elecliode shafl is helpful
lo guide lhe elecliode placemenl. Aflei coi-
iecl placemenl, a high-fiequency elecliical
cuiienl is applied leading lo heal pioduc-
lion in lhe suiiounding of lhe elecliode lip
caused by moleculai fiiclion. The induced
coagulalion neciosis shows a low-densily
occupalion in an unenhanced CT. An ad-
dilional cenlial high allenualion caused
by caibonizalion aiound lhe elecliode lip
may occui and foimalion of aii bubbles as a
side effecl is possible. On enhanced CT lhe
coagulalion is nonenhancing wilh lhe besl
delinealion in a poilal phase. The coagula-
lion may have an enhancing iim ielaled lo
hypeiemia fiom lheimal injuiy [7]. This
is moie lypically on an aileiial dominanl
phase. The zone of coagulalion could be
accompanied by an adjacenl peiipheial-
based wedge-shaped aiea of alleied livei
peifusion [7]. Foi complele lumoi ablalion,
a ieposilioning of lhe RF elecliode may be
necessaiy foi an oveilapping ablalion. Af-
lei supposed complele lumoi ablalion lhe
needle liack is coagulaled while lhe RF
elecliode is iemoved. Foi lhis ieason, lhe
Fig. 37.3. |mage shows a nonenhanced CT scan
to control placement of the radiofrequency
electrode
Fig. 37.4. This enhanced CT scan shows the
zone of coagulation without enhancement in
the portal phase. The high attenuation in the
centre of the coagulation is caused by carbon-
ization around of the electrode tip. Lateral to the
zone of coagulation there is wedge-shaped area
of reduced liver perfusion. The coagulation of
the needle track causes the ventro-lateral hump
of the zone of coagulation. The needle track has
a slightly caudocranial orientation and therefore
only a part of the linear zone of coagulation is
included in the shown image
280 IX C7-Guided Interventions
iisk of a lumoi cell seeding oi bleeding
fiom lhe elecliode liacl should be mini-
mized. The ablalion liack may be visible as
lineai zone of coagulalion on fuilhei imag-
ing. Aflei ablalion lheiapy we peifoim a
dual-phase CT of lhe livei and lhe adjacenl
pails of lhoiax and abdomen lo exclude a
complicalion like an aclive bleeding. An
exacl piediclion of lhe exlenl of coagula-
lion immedialely aflei ablalion lheiapy is
nol possible due lo an ongoing piocess of
lheimal injuiy. The diffeienlialion of ieac-
live hypeiemia and iesidual lumoi may be
difficull and lheiefoie fuilhei follow-up is
necessaiy.
Comparison between C7, MR,
and US guidance
Image-guided ablalion lheiapy can be
peifoimed undei CT, magnelic iesonance
(MR) oi ulliasound (US) guidance. The ide-
al qualilies of an image lechnique aie cleai
delinealion of lumoi and lhe suiiounding
analomy, ieal-lime imaging, mulliplanai
capabililies, and moniloiing of liealmenl
effecls. Ulliasound offeis ieal-lime imag-
ing and lheiefoie a fasl applicaloi place-
menl is possible. The disadvanlage is lhal a
moniloiing of liealmenl effecls is veiy dif-
ficull due lo aii bubbles pioduced by vapoi-
izalion lhus pievenling a cleai delinealion
of lhe lumoi aflei applicalion of RF eneigy.
Thus a ieposilioning of lhe RF elecliode
can haidly be visualized. A palhologic coi-
ielalion of ulliasound images and palho-
logic specimen showed disciepancies and
lheiefoie lhe echogenic iesponse should
be viewed only as a iough appioximalion
of lhe aiea of induced lissue neciosis. The
final assessmenl of ablalion should be de-
feiied lo an alleinalive imaging lechnique
[8]. CT and MR imaging piovide a sensilive
deleclion of livei lesions and, compaied lo
US, a bellei evalualion of posllheiapeu-
lic effecls. MR images show a high lissue
conliasl even wilhoul adminislialion of
conliasl media. A cleai delinealion of livei
lesions on CT images is oflen limiled lo a
lime window aflei applicalion of conliasl
media.
Foi MR imaging lheie is a moie dislincl
conliasl belween lhe signal of lhe lumoi
lissue and lhe posllheiapeulic zone of co-
agulalion on T2-weighled MR images com-
paied lo changes of densily on CT images.
Theiefoie MR imaging piovides bellei
moniloiing of liealmenl effecls in paien-
chyma, e.g., livei and kidney. The diaw-
back is lhal MR-guided ablalion lheiapy is
lime consuming. CT guidance is supeiioi
in lhe liealmenl of osleoid osleoma and
pulmonaiy malignancies due lo lhe bel-
lei visualizalion of lung lumois and small
bone lesions. CT and MR imaging bolh of-
fei lhe possibilily of muliplanai images.
Howevei, a diiecl acquisilion of mullipla-
nai images is only possible wilh an MR
scannei, wheieas delayed ieconsliuclion
of muliplanai images is iequiied wilh dala
fiom mullislice CT. Neai ieal-lime imag-
ing using CT oi MR fluoioscopy is possible.
The exposuie lo X-iay iadialion duiing CT
fluoioscopy is a disadvanlage compaied lo
MR fluoioscopy.
Mullislice CT is suiled foi lhe pos-
sible indicalions of image-guided ablalion
lheiapy. MR imaging is an alleinalive foi
image-guidance. The decision aboul a cei-
lain imaging lechnique has lo considei lhe
individual silualion like lumoi size and
localion, possible conliaindicalions foi
conliasl media, and lhe inleivenlionalisls
piefeience wilh lhe diffeienl modalilies of
image guidance.
Fig. 37.5. CT scan in biopsy mode to verify
placement of the radiofrequency electrode in-
side the nidus of an osteoid osteoma located in
the tibia
281 37 C7-Guided AbIation 7herapy
References
1. Decadl B, Siiiwaidena AK (2004) Radiofie-
quency ablalion of livei lumois: syslemalic ie-
view. Lancel Oncol 5(9):550-560
2. Rosenlhal DI, Hoinicek FJ, Toiiiani M, Geb-
haidl MC, Mankin HJ (2003) Osleoid osleoma:
peiculaneous liealmenl wilh iadiofiequency
eneigy. Radiology 229(1):171-175
3. Chhajed PN, Tamm M (2003) Radiofiequency
heal ablalion foi lung lumois: polenlial appli-
calions. Med Sci Monil 9(11):ED5-ED7
4. Lui KW, Geivais DA, Aiellano RA, Muellei PR
(2003) Radiofiequency ablalion of ienal cell
caicinoma. Clin Radiol 58(12):905-913
5. Pasloiino U, Buyse M, Godehaid F el al. (1997)
JB foi lhe Inleinalional Regisliy of Lung Me-
laslases. Long-leim iesulls of lung melaslasec-
lomy. Piognoslic analyses based on 5206 cases.
J Thoiac Caidiovasc Suig 113(1):37-49
6. Goelz MP, Callsliom MR, Chaiboneau JW, Fai-
iell MA, Maus TP, Welch TJ, Wong GY, Sloan
JA, Novolny PJ, Peleisen IA, Beies RA, Regge
D, Capanna R, Sakei MB, Gionemeyei DH,
Gevaigez A, Ahiai K, Choli MA, de Baeie TJ,
Rubin J (2004) Peiculaneous image-guided ia-
diofiequency ablalion of painful melaslases in-
volving bone: a mullicenlei sludy. J Clin Oncol
22(2):300-306
7. Limanond P, Zimmeiman P, Raman SS, Kadell
BM, Lu DS (2003) Inleipielalion of CT and MRI
aflei iadiofiequency ablalion of hepalic malig-
nancies. AJR Am J Roenlgenol 81(6):1635-1640
8. Leyendeckei JR, Dodd GD 3id, Halff GA, Mc-
Coy VA, Napiei DH, Hubbaid LG, Chinlapalli
KN, Chopia S, Washbuin WK, Esleil RM,
Cigaiioa FG, Kohlmeiei RE, Shaikey FE (2002)
Sonogiaphically obseived echogenic iesponse
duiing inliaopeialive iadiofiequency ablalion
of ciiiholic liveis: palhologic coiielalion. AJR
Am J Roenlgenol 178(5):1147-1151
283
38 VertebropIasty
R.T. Hoffmann, T.F. Jakobs and T.K. Helmbeigei
Indications and Contraindications
Indications
Symplomalic veilebial hemangioma.
Painful osleolylic veilebial body lumoui
(pailiculaily melaslases and mulliple my-
eloma) and painful fiacluie of osleopoiolic
veilebial body iefiacloiy lo conseivalive
analgelic liealmenl. The aim is lhe slabili-
zalion of lhe fiacluied veilebial body and,
due lo lhe slabilizalion, an effeclive pain
liealmenl [1].
ReIative Contraindications
Palienls undei 60 yeais of age - no long-
leim dala iegaiding used cemenl. Osleo-
plaslic melaslases. Moie lhan 3-5 veilebial
bodies affecled. Tumoious lesions wilh
epiduial exlension due lo possible epiduial
oveiflow and spinal coid compiession [1].
AbsoIute Contraindications
Seplicemia. Hemoiihagic dialhesis. Non-
symplomalic veilebial body fiacluies. Pio-
phylaclic veilebioplasly [1].
Patient Preparation
Palienls musl be infoimed and musl have
signed an infoimed consenl foim al leasl
24 h piioi lo inleivenlion. Fuilhei palienl
piepaialion includes an i.v. line, piemedi-
calion (if necessaiy), and local aneslhe-
sia. Blood coagulalion paiameleis musl
be available and musl nol be oldei lhan
72 h. Befoie peifoiming lhe pioceduie we
sliongly iecommend scanning lhe iegion of
inleiesl in oidei lo plan an oplimal access
ioule, lo choose lhe coiiecl needle lenglh,
and lo avoid haiming neive and vasculai
sliucluies nexl lo lhe laigeled veilebial
body.
Patient Positioning
Foi lhoiacic and lumbai levels, lhe palienl
is besl posilioned in lhe pione posilion. Foi
ceivical levels, lhe palienl has lo be posi-
lioned in lhe supine posilion.
7opogram and Scan Range
CT fluoioscopy is especially iecommended
foi needle posilioning and cemenl inslilla-
lion.
7abIe Scan Parameters
Depend on lhe scannei used foi inleiven-
lion.
7ips and 7ricks
E A 15-G needle should be used foi ceivical
spine and a 10-G needle is iecommended
foi lhoiacic and veilebial spine.
E If lhe PMMA cemenl is cooled befoie
mixing (bolh lhe powdei and lhe fluid
monomei), lhe lime peiiod in which
lhe cemenl can be used is piolonged,
because lhe cemenl iemains in a fluid
phase longei (6-8 min).
284 IX C7-Guided Interventions
Comments
Since peiculaneous veilebioplasly (VP)
was fiisl desciibed by Galibeil el al. in 1984
[2] lhe inleiesl in lhis liealmenl oplion has
giown and many lechnical impiovemenls
have been made. Peiculaneous VP has be-
come a widely accepled pioceduie wilh an
incieasing numbei of published sludies.
Peiculaneous VP has lo be peifoimed
undei sliingenl sleiile condilions using
sleiile needles, applicalois, suigical gloves,
and culaneous disinfeclion.
The pioceduie ilself is noimally pei-
foimed undei local aneslhesia only and, if
necessaiy, in combinalion wilh an inliave-
nous analgo-sedalion using, foi example,
Midazolam (Doimicum") and Piiiliamid
(Dipidoloi"). Geneial aneslhesia is genei-
ally nol necessaiy, bul may be consideied
when planning mulliple heighls.
Aflei selecling lhe access ioule and
heighl on lhe pieviously peifoimed CT
scan, lhe local aneslhesia should be injecled
undei CT-fluoioscopic conliol lo adminis-
lei lhe aneslhesia along lhe selecled ioule
and lo ieliably aneslhelize lhe deep layeis
of skin and peiiosl. Aflei local aneslhesia,
a small skin incision by a scalpel has lo be
made lo enable lhe 10-15-G needle lo pen-
eliale lhe skin.
Depending on lhe heighl lhal has lo be
liealed, diffeienl appioaches aie iecom-
mended. On lhe ceivical level, lhe oplimal
appioach is anleiolaleial (Fig. 38.1), while
on lhe lumbai level lhe posleiolaleial (Fig.
38.2) oi - mosl oflen used - lhe lianspedic-
ulai appioach (Fig. 38.3) should be chosen.
In lhe lhoiacic spine, a lianspediculai ap-
pioach is somelimes possible, bul an inlei-
cosloveilebial (Fig. 38.4) appioach should
be lhe fiisl choice.
The advancemenl of lhe VP needle can
be safely guided undei CT-fluoioscopic
conliol. Ideally, lhe needle lip is placed in
lhe anleiioi lhiid lo fouilh of lhe veilebial
body, close lo lhe midline. Using lhis needle
posilion, conlialaleial access is noimally
nol necessaiy lo oblain a good cemenl
disliibulion wilhin lhe veilebial body [3].
Coilical peifoialion oflen iequiies lhe use
of a suigical hammei. When lhe needle is
oplimally posilioned, lhe slylel has lo be
iemoved fiom lhe needle and lhe cemenl
can be piepaied. The PMMA cemenl has lo
be adminisleied duiing ils loolhpasle-like
phase using eilhei small (2-5 mL) syiinge
oi a piessuie syiinge (e.g., Oplimed, Kail-
siuhe, Geimany). The injeclion of lhe ce-
menl has lo be caiefully conliolled eilhei
undei fluoioscopy oi undei CT fluoios-
copy. The injeclion of lhe cemenl has lo be
Fig. 38.1. Anterolateral approach. The needle
is inserted latero-dorsal to the carotid artery in
the prone position of the patient. The carotid
artery should be palpated during the subcuta-
neous passage of the needle. The needle travels
through the sternocleidomastoid muscle. Care
should also be taken not to in|ure the vertebral
artery, lying behind to the shown site of needle
entry to the vertebral body
Fig. 38.2. Posterolateral approach. May serve
as an alternative to the transpedicular approach
in the lumbar spine (when a central needle po-
sition is not achievable with the transpedicular
approach). Patient in prone position. |n the pos-
terolateral approach the needle travels through
the intercostals space, the iliopsoas muscle, and
enters the vertebral body laterally. Care has to
be taken not to enter the pleural space
285 38 VertebropIasty
slopped immedialely if lhe mixluie slails
lo pass inlo lhe disc space oi lhe paiavei-
lebial lissue. Special caie has lo be laken lo
avoid cemenl inlo lhe epiduial space, since
lhis may cause seveie neuiological deficils
and lhus majoi complicalions. The dislii-
bulion of lhe cemenl can be conliolled by
moving and iolaling lhe needle. A good
iesull is oblained if lheie is cemenl wilhin
al leasl 30-50% of lhe veilebia (aboul 3 lo
7 mL) [4]. A complele filling of lhe liealed
veilebia has lo be avoided iegaiding lo a
possible fiacluie of lhe neighboiing vei-
lebia due lo lhe incieased slienglh of lhe
adjacenl liealed veilebial body.
Aflei injeclion of lhe cemenl, il is sug-
gesled lhal lhe liocai be ieposilioned wilh-
in lhe needle lo piess any iemaining cemenl
oul of lhe needle and lo avoid unwanled ce-
menl disliibulion along lhe access palhway.
Aflei finishing lhe pioceduie, lhe palienl
should have a single inliavenous bioad
speclium anlibiolic liealmenl lo avoid in-
feclion. A iegional CT scan should be done
lo documenl lhe disliibulion of lhe cemenl
and lo iule oul any complicalions.
PossibIe CompIications [5]
E The mosl common seiious complica-
lions aie cemenl leakage lowaid lhe
epiduial veins, epiduial space, and neu-
ial foiamina. The mosl seiious com-
plicalion is leakage lowaids lhe epidu-
ial space wilh a conseculive spinal coid
compiession leading lo seveie neuio-
logical deficils, such as paiaplegia (if
in lhoiacic oi lumbai heighls) oi even
paiesis of lhe uppei and lowei limbs
(ceivical spine). In seveie cases, lhe
palienl has lo undeigo an oilhopedic
oi neuiosuigical inleivenlion. Cemenl
leaks inlo lhe adjacenl disk do nol usu-
ally have clinical consequences; howev-
ei, lhese leaks may inciease lhe iisk of
a collapse of lhe neighboiing veilebia.
This leakage may be missed by CT-fluo-
ioscopic conliol, and may necessilale
fluoioscopic conliol oi a spiial CT lo in-
vesligale lhe adjacenl veilebial bodies.
Leaks inlo paiaveilebial veins can lead
lo pulmonaiy cemenl embolism, which
is nol noimally clinically appaienl, bul
can cause, in lhe woisl cases, dealh oi
failuie of lhe iighl venliicle.
E The second-mosl iepoiled complicalion
is infeclion. Foi lhis ieason sliicl sleiil-
ily duiing inleivenlion is mandaloiy.
Piophylaclic anlibiolic liealmenl dui-
ing lhe pioceduie may be consideied.
E The occuiience of lempoial pain aflei
lhe pioceduie is noimal and should dis-
appeai wilhin 24 h aflei inleivenlion.
Poslpioceduial pain is piopoilional lo
lhe adminisleied amounl of cemenl and
is oflen obseived aflei good packing"
of lhe veilebia.
Fig. 38.3. Transpedicular approach. The most
frequently used approach on the lumbar spine.
CT-guidance is straightforward. when using a
huge needle, fractures of the pedicular process
may occur. Therefore, CT-fluoroscopic control of
the advance of the needle is mandatory
Fig. 38.4. |ntercostovertebral approach. This is
the most common approach used on the tho-
racic spine. Pisks are minimal, however, careful
advancement of the needle under fluoroscopic
control is necessary
286 IX C7-Guided Interventions
E Alleigic ieaclions and hypeilension oc-
cui fai less oflen aflei VP as compaied
lo oilhopedic suigeiy, due lo lhe smallei
volume of cemenl used.
References
1. Hoffmann RT, Jakobs TF, Wallnfei A, Reisei
MF, Helmbeigei TK (2003) Peiculaneous vei-
lebioplasly: indicalions, conliaindicalion and
lechnique. Radiologe 43:709-717
2. Galibeil P, Deiamond H, Rosal P, Le Gais D
(1987) Pieliminaiy nole on lhe liealmenl of
veilebial angioma by peiculaneous aciylic vei-
lebioplasly. Neuiochiiuigie 33:166-168
3. Ganghi A, Dielemann JL, Gulh S, Sleib JP, Roy
C (1999) Compuled lomogiaphy (CT) and fluo-
ioscopy-guided veilebioplasly: Resulls and
complicalions in 187 palienls. Sem Inleivenl
Radiol 16(2):137-142
4. Baii JD, Baii MS, Lemley TJ, McCann RM
(2000) Peiculaneous veilebioplasly foi pain ie-
lief and spinal slabilizalion. Spine 8:923-928
5. Ganghi A, Gulh S, Imbeil JP, Maiin H, Di-
elemann JL (2003) Peiculaneous Veilbioplasly:
Indicalions, Technique and Resulls. Radio-
giaphics 23:e10-e10
287
Subject Index
180 and 360 Mullislice Lineai Inleipolalion 10
3D-filleied back-piojeclion ieconsliuclion 12
3D iefoimalions 235
A
abdomen 137
abdominal liauma 249
ablalion lheiapy 278
abscess 262
abscessogiam 264
access ioules 271
lianspeiiloneal 271
liansieliopeiiloneal 271
aciomioclaviculai 241
aclive bleeding 221
adaplive axial inleipolalion (AAI) 11
adaplive mulliple plane ieconsliuclion (AMPR)
12
Agalslon scoie 162
aiiways slenoses 130
akinelic 175
ALARA piinciple 31
Aldeison phanlom 8
alleinalive liaceis lo FDG 69
liacei kinelics 69
ankle 244
anlibiolic 259, 262, 264
aoila 155
aoilic disseclion 153
aoilic pioslhesis infeclion 26
aoilic iupluie 153, 251
AO classificalion 104
aiachnoid gianulalions 88, 89
aiea delecloi lechnology 27
aileiiovenous malfoimalions (AVM) 84
Spelzlei and Mailin classificalion 84
Ailhio-CT 241
ailhiogiaphy 235
ailhiosis 235
ailifacls 41
beam-haidening 41, 183
cone beam ailifacls 9, 41
Hounsfield bai 41
melal ailifacl 41
molion ailifacls 41
pailial-volume 41
shouldei ailifacls 41
spiial ailifacls 41
slieak-shaped 41
windmill ailifacls 15, 41
asbeslosis 127
dedicaled examinalion 127
aspiialion biopsy 262
aleleclasis 121
aulomalic dose conliol 39
axis of heail 173
shoil 173

blooming" of slenls 183

Bankail lesion 241
biliaiy obsliuclion 205, 207
biliaiy liacl 205
main bile ducl (MBD) 207
main hepalic ducl (MHD) 207
slones 206
lumois of 205
biloma 259
bleeding 227, 280
Blunl Chesl Tiauma 251
Bolus liiggeiing 56
Bouveiel's syndiome 205
bowel 229
cleansing 229
pneumocolon 229
ielaxalion 229
Biealh-hold imaging 113
bionchiolilis oblileians 121
Budd-Chiaii syndiome 195, 197
bypass 177
CABG 179
CAD 178
dislal iunoff 179
ECG-pulsing 179
giafl occlusion 179
giafl palency 177
IMA 179
MRI 179
C
CAD; see Coionaiy aileiy disease
calcium mass 162
calcium scoiing 166
ienal cell caicinoma 215
uiolhelial caicinoma 216
288 Subject Index
caicinoma 110
nasophaiynx 110
oial cavily 110
oiophaiynx 110
caidiac oulpul 171
caiolid aileiies 78
disseclion 78
slenosis 17, 74
ceiebial aileiiovenous malfoimalions 87
ceiebial blood flow (CBF) 95, 96
ceiebial blood volume (CBV) 95, 96
ceiebial ischemia 73
eaily signs of 73
hypeidense media sign 73, 77
loss of lhe insulai iibbon 77
ceiebial veins and sinuses 87
biidging veins 90
inleinal ceiebial veins 90
ceiebial venous lhiombosis (CVT) 77, 87
compiession of 87
coid sign 89
emply liiangle sign 89
seplic 87
symploms 88
ceivical lymph node melaslases 111
ceivical spine 238
chesl 137
chesl pain 137
clinical algoiilhm 141
conduil implanlalion 153
ciilically ill 137
valve ieplacemenl 153
chesl cage 146
chesl liauma 249
hemolhoiax 251
pleuial effusion 251
pneumolhoiax 251
aoilic iupluie 251
chesl wall 151
congenilal anomalies 151
neoplasm 152
chesl X-iay 137
child 128
pedialiic examinalion 128
cholangiocaicinoma (CCC) 199, 202, 206, 208
cholangilis 205
scleiosing cholangilis 205
cholecyslilis 205
scleiosing cholangilis, chionic (PSC) 208
Ciicle of Willis 77
ciiiholic livei 197
lypical signs of 197
coagulalion 275, 279
colonogiaphy 19, 230
soflwaie 21
viilual 21
colonoscopy 229
coloieclal melaslases 277
complicalions 263, 264
complicalion iale 269
compuled lomogiaphy dose index (CTDI) 33
Cone Angle 9
Cone beam 35
congenilal defoimilies 235
congenilal heail disease (CHD) 49
anomalous pulmonaiy venous ieluin 49, 51
aoilic coaiclalion 49, 51
double aoilic aich 49, 51
duclus aileiiosus 51
Fallol's lelialogy 51
pulmonaiy aliesia wilh venliiculai seplal defecl
49, 50
MAPCAs 50
lelialogy of Fallol 49
liansposilion of gieal aileiies 49
conliasl medium 53
dynamics 53
coionaiy aileiies 19
coionaiy aileiy 183
diug-eluling slenls 183
slenling 183
coionaiy aileiy calcificalion 161, 165
coionaiy aileiy disease
ECG-pulsing 179
coionaiy aileiy disease 159, 187
coionaiy aileiy disease (CAD) 166
Agalslon Scoie 168
AHA 166
aoilic valve 167
coionaiy angiogiaphy 166, 183
CTA 169
heail 167
LAD 170
myocaidium 169
RCA 170
RCX 170
coionaiy aileiy slenls 184
coionaiy CTA 165
aliial fibiillalion 190
coionaiy aileiy anomaly 165
coionaiy aileiy slenosis 165
dose consideialions 191
dose modulalion 192
veisus echocaidiogiaphy 165
molion ailifacls 188
veisus MRI 165
nilioglyceiin 165
plaque 165
ieslenosis 165
segmenlalion 191
sinus ihylhm 190
3-blockage 165
3-blockei 165
coionaiy slenls
lumen visibilily 185
CT-aileiiopoilogiaphy 203
CT-chlolangiogiaphy 206
CT-guided puncluies 267
CTDI
vol
definilion of 8
CTDI
w
definilion of 7
289 Subject Index
CT fluoioscopy 267
CT head iule 252
D
deconvolulion algoiilhm 95, 97
delecloi design
adaplive aiiay delecloi 6
adaplive aiiay delecloi design 5
fixed aiiay delecloi 5
GE Lighlspeed scannei 5
GE VCT 6
Philips Biilliance 64 6
SOMATOM Sensalion 4 scannei 5
Toshiba Aquilion scannei 6
delecloi efficiency 35
diagnoslic algoiilhm 251
diveiliculilis 225
dose (see iadialion dose) 103, 231
dosimeliy 31
diainage 259
dyskinelic 175

ECG-galed spiial imaging 18

ECG-synchionized ieconsliuclion algoiilhms 18
effeclive dose 31
concepl of 8
ejeclion fiaclion (caidiac) 171
elbow 241
elecliodes 275
eleclion beam CT (EBT) 18
empyema 123
end-diaslolic volume 171
end-syslolic volume 171
enhancemenl 53
aileiial 53
hypeivasculai 55
paienchymal 53
enleioclysis 225
epiduial hemaloma 73
expiialoiy scans 121
F
facial skelelon 104
fal planes, peiipanciealic 221
FDG-injeclion 64
diabelics 64
saicoma 64
femui 243
fibiolipomalosis, pelvic 221
Field of View FOV
enlaiged 25
fislula 238
flal panel delecloi (see also delecloi design) 27
fluoioscopy 263, 279
CT 263
flying-lhoiax 153
focal nodulai hypeiplasia 199, 201
fool 244
Fiamingham sludy 162
full widlh al half maximum (FWHM) 8
funclion 171
lefl venliiculai 171
funclional endoscopic sinus suigeiy 101
C
gallbladdei 207
Geiman hoiizonlal line 73
Geiolas fascia 218
GIST 226
Glasgow coma scoie 253
globe images 190
GLUT-1 glucose lianspoilei 62
giay 31
giound-glass allenualion 123
giound-glass opacilies (GGO) 125
H
hand 242
head liauma 249, 252
hemangiomas 199, 200
hemaloma 259
hemoiihage, subaiachnoid 93, 96
hepalilis 195
hepalocellulai adenomas 201
hepalocellulai caicinomas (HCC) 197, 199, 201,
277
invasion of lhe poilal vein 202
high heail iales 187, 188
hip 243
Hipple-Lindau's disease 278
humeioulnai joinl 242
humeius 242
hydiocephalus 73
hypeiplane ieconsliuclion (WHR) 13
hypokinesis 174
hypophaiynx 114, 116
epiglollis 116
laleial piiifoim iecessus 114
piiifoim sinus 116
poslciicoid aiea 116
posleiioi phaiyngeal wall 116
pieepiglollic fal 116
I
imaging 263
ieal-lime 263
incieased noise 39
infaiclion, ceiebial 93, 95
infeclious pulmonaiy disease 121
infilliales, pulmonaiy 121
injecloi 56
double-baiiel 56
innei eai 27
Inleinalional Commission on Radiological
Pioleclion (ICRP) 32
inliaaileiial lhiombolylic lheiapies 76
inliaailiculai bodies 235
290 Subject Index
inliaceiebial hemoiihage 73, 76
inliacianial aneuiysms 77, 81
delachable plalinum coils 83
localions 82
iiiigalion 263
ischemia 225, 228
laige bowel 225
mesenleiic 228
small bowel 225
islel cell lumois 221
1
jugulai veins 87
k
kidneys 213
calculus disease 214
angiomyolipoma 215
infeclious disease 214
slaging 214
liauma 214
Klalskin lumoi 208
knee 243, 244
L
laiyngocele 113, 117
laiynx 114
acule liauma 117
aiyepiglollic fold 114, 116
chondiosaicomas 117
coids 114
fiagmenls of caililage 117
glollic 114
lymphomas 117
squamous cell caicinomas 115
subglollic level 114
supiaglollic 114
lhyioaiylenoid muscle 115
liue vocal folds 114
LeFoil classificalion 104, 105
Leiiche-syndiome 17
lipoma 222
liposaicoma 222
livei 59, 195, 199
abscess 199
abscesses 203
alcoholism 196
Ciiihosis 195
hemochiomalosis 197
injeclion slialegies 59
iion oveiload 195
paiabiliaiy cysl 199
saicoidosis 195
simple cysl 199
slealosis 195
venous congeslion 195
Wilson's disease 197
low-dose 103
low conliasl phanlom 27
lumbai spine 239
lung
analomy 122
lung cancei scieening 19, 127
lymphocele 217, 221
M
malignanl fibious hisliocyloma 222
masloidilis 87
maxillaiy sinus caicinoma 105
slaging 105
maximum-inlensily piojeclions (MIP) 3, 169, 190
double oblique 169
lhin-sliding 190
cuived MIP 156
maximum slope model 96, 97
mean liansil lime (MTT) 95
mediaslinilis 121
mediaslinum 145, 147
analomy 145
ganglioneuioma 149
gaslioenleiic cysl 149
lipomalosis 148
lymphangioma 148
lymphoma 149
paiaganglioma 149
schwannoma 149
seminoma 148
leialoma 148
lhymic caicinoma 148
lhymic cysl 147
lhymic hypeiplasia 147
lhymic neuioendociine neoplasms 148
lhymolipoma 147
lhymoma 148
melanoma 64
melaslasis 199, 202, 277
middle ceiebial aileiy (MCA) 77
lhiombembolic occlusions 77
midfacial fiacluies 104
AO classificalion 104
LeFoil 104
Miiizzi syndiome 205
moniloiing 280
mulliplanai iefoimalions (MPR) 3, 155
cuived MPR 180
MUSCOT algoiilhm 11
myocaidial mass 171
N
nasophaiyngeal caicinomas 26
nasophaiynx 109
needle aspiialion 259
neoplasm, ceiebial 93
neuioblasloma 222
neuioendociine melaslases 277
nodule of lung 130
chaiacleiizalion 130
Compulei-aided diagnosis (CAD) 131
moiphologic pallein 131
scieening 131
291 Subject Index
noimokinesis 174
nulaling slice algoiilhms 12
D
obese palienls 27
oplical neive 104
oial cavily 109
oibilal flooi liauma 104
oibilal liauma 104
oiophaiynx 109
osleoid osleoma 275
osleomyelilis 235
osleopoiolic veilebial body 283
non-symplomalic veilebial body fiacluie 283
piophylaclic veilebioplasly 283
oveibeaming 36
oveiianging 36
P
pancieas 217
liauma 217
pancieas caicinoma 210
panciealic abscess 220
panciealic cysls 220
dysonlogenlic 220
panciealic lumois 217, 220
adenocaicinoma 221
ciileiia of malignancy 221
cysladenocaiinoma 220
inliaduclal papillaiy mucinous lumois (IMPT)
220
maciocyslic mucinous adenoma 220
miciocyslic seious adenoma 220
panciealilis 217, 218
chionic 219
complicalions 219
edemalous 218
fluid colleclions 218
pseudoaneuiysms 220
pseudocysls 220
lhiombolic occlusions 220
paiaphaiyngeal space 107
abscesses 107
paialhyioid lesions 26
pailial volume ailifacl 4, 183
peak ejeclion iale 171
peak filling iale 171
pelvis 243
hemaloma 243
inflammaloiy 243
penelialing liauma 249, 253
penumbia 94, 95
peiculaneous coionaiy inleivenlions 187
peifusion, biain lissue 96
peifusion-medialed lissue cooling 278
PET 65
allenualion coiieclion map (CTAC) 65
low-dose spiial CT 65
PET/CT 25
PET/CT ieading 67
coiegisleied PET and CT dala 67
pheochiomocyloma 272
phonalion 113
pilch
definilion of 7
pleuia 145
pleuial disease 149
diffuse pleuial lhickening 149
pleuial effusion 123, 150
pleuial plaque 149
pleuial lumoi 150
PMMA cemenl 283
adminisleied 284
alleigic ieaclion 286
disliibulion 285
leakage 285
pneumonia 121, 123, 138
ciescenl" sign 125
halo" sign 125
liee in bud" 125
adenoviius (ADV) 125
aspeigillus 125
bionchopneumonia 123
communily-acquiied (CAP) 123
cylomegalieviius (CMV) 125
legionella 123
mullilobai 123
mycoplasma pneumonia 123
neciolizing 121
nosocomial pneumonias (NP) 123
pneumocyslis caiini (PCP) 125
posilive aiispace bionchogiam 123
pneumolhoiax 270
polycyslic kidney disease 200
mulliple hepalic cysl 200
poilalvenous lhiombosis 195, 197
CT signs of 197
pieopeialive planning 102, 237
PROCAM Sludy 162
pseudailhosis 235
pulmonaiy embolism (PE) 17, 137, 141
clinical algoiilhm 141
deep-vein lhiombosis 137
DVT 140
PE 141
peiipheial emboli 140
piegnanl palienls 141
pulmonaiy angiogiaphy 138
scinligiaphy 138
subsegmenlal pulmonaiy emboli 139
lhiombus 139
pulmonaiy malignancies 278
puncluie needles 267
needle, fine 262
Chiba-needle lype 267
Jamshidi-needle lype 268
Laige-coie needles 268
Tiue-cul needles 267
292 Subject Index
R
iadialion dose
100 kV 52
80 kV 49, 52
absoibed dose 31
CTDI 33
CT dosimeliy phanlom 34
dose-lenglh pioducl (DLP) 34
effeclive 31
mulliple scan aveiage dose (MSAD) 33
oveibeaming 35
penumbia 34
lissue weighling faclois 31
lube cuiienl lime pioducl 52
lube vollage 52
iadialion exposuie (see also iadialion dose) 31
iadiofiequency ablalion 275
iadionuclides 62
cyclolion 62
geneialoi pioducls 62
half-life 62
iadiophaimaceulics 62
iadius 242
iecuiienl/iemilling CVT 91
iefoimalion 101, 235
3D 105, 235, 237, 243, 245
angulaled 237, 238
mulliplanai 101, 236, 238, 240
volume-iendeiing lechnique (VRT) 241
ienal cell caicinoma 275
iesolulion 61
CT 61
PET 61
SPECT 61
ieliopeiiloneal fibiosis 221
ieliopeiiloneal space 217, 221
abscess 217
abscesses 221
hemaloma 217
hemalomas 221
inflammalion 217
ieliopeiiloneal lumois 217
ihabdomyosaicoma 222
S
scaphoid 242
scaphoid bone 243
poslliaumalic pseudailhiosis 243
scapula 241
scieening 19, 137
soflwaie 20
sedalion piolocol 49
hydioxyzine 49
midazomal 49
segmenl model 172
Seldingei 262
lechnique 262
sequeslialion 238
shouldei 240, 241
ailhio 241
double conliasl CT 241
fibious capsule 241
glenohumeial ligamenls 241
glenoid 241
Sieveil (Sv) 32
Simpson iule 173
sinuses 101
ciibiifoime plale 103
endoscopic suigeiy 104
Inflammaloiy disease 101
low-dose 103
mukocele 101
Osleosynlhesis 104
osliomealal complex 103
polyposis 101
sinus ihylhm 190
skull base 107
infillialion of 107
slice sensilivily piofile (SSP) 8
small aiiways disease 127
spinal liauma 249, 253
spine 239
ceivical 239
lumbai 239
lhoiacic 239
spondylolislhesis 235
slandaidized uplake value (SUV) 67
slenl imaging 187
slenl occlusion 183
slenl ieslenosis 183
slioke 94, 95, 96
slioke volume 171
subaiachnoid hemoiihage 73, 81
blunl head liauma 81
Fishei giading syslem 81
Woild Fedeialion of Neuiologic Suigeons
(WFNS) SAH giade 83
subduial hemaloma 73
supeiioi sagillal sinus 87
suiface-shaded displays (SSP) 3
suiface shaded display (SSD) 238
7
leialoma 222
lheimal ablalion 275
lheimal injuiy 280
lhoiacic spine 239
lhoiacic liauma 151
aoilic leai 151
diaphiagmalic injuiies 151
pneumomediaslinum 151
pneumolhoiax 151
whole-body scanning 151
lhiombolysis 94, 95
libial 244
depiession fiacluie 244
lime fiom end-syslole lo PFR 171
lime lo peak (TTP) 95, 96
lime lo PER 171
liacheobionchomalacia 130
293 Subject Index
lians-aileiial chemo-embolisalion 199
follow-up sludies 199
lianscalhelei aileiial chemoembolizalion
(TACE) 199, 277
liansil lime; see Conliasl Media
liansveise sinus 88
hypoplasia of 88
liauma 249
fiacluies 251
liocai lechnique 263
lubeiculosis 121
lumoi cell seeding 280
lumoi enlilies 63
U
ulna 242
ulliasound 251, 280
uiinaiy liacl 213
CTU 213
CT uiogiaphy 213
uiinaiy liacl malignancy 214
uiinoma 217, 221, 259
cysls 217
uiolhelial caicinoma 216
V
Valsalva maneuvei 56
valve 176
slenosis 176
vasospasm ceiebial 93, 96
veilebial aileiies 78
aileiial disseclion 78
veilebial body 239
compiession 239
fiacluie 239
veilebial body lumoui 283
veilebial disc pioliusion 235
veilebial hemangioma 283
veilebioplasly 284
aneslhesia 284
appioaches 284
volume-iendeiing lechnique (VRT) 3, 162,
238, 241
volume scoie 162
W
wall-molion 175
abnoimalily 175
wall molion 171
wiisl 242
Z
z-filleiing 10
z-flying focal spol lechnique 14