ORIGINAL ARTICLE

Temporomandibular joint growth changes in

hyperdivergent and hypodivergent Herbst
subjects. A long-term roentgenographic
cephalometric study
Hans Pancherz, DDS, Odont Dr, FCDSHK (Hon),a and Chrysoula Michailidou, DDS, Dr med dentb
Giessen, Germany

The aim of this long-term study was to assess the amount and direction of glenoid fossa displacement,
condylar growth, and “effective” temporomandibular joint (TMJ) changes (⫽ the sum of glenoid fossa
displacement, condylar growth, and condylar position changes in the fossa) in 3 vertical facial-type groups
of Class II Division 1 malocclusions treated with the Herbst appliance. A comparison was made between 38
normodivergent (ML/NSL⫽ 26.5°-36.5°), 17 hypodivergent (ML/NSL ⱕ 26°), and 13 hyperdivergent (ML/NSL
ⱖ 37°) subjects. Lateral headfilms from before, after, and 5 years after treatment were scrutinized. Glenoid
fossa displacement, condylar growth, and “effective” TMJ changes were analyzed. Treatment changes: in all
facial-type groups, the glenoid fossa was displaced anteriorly and inferiorly. No differences existed between
the 3 groups. Condylar growth and “effective” TMJ changes were directed posteriorly and superiorly. The
changes in posterior direction were more apparent in the hyperdivergent group than in the normodivergent
and hypodivergent groups. Posttreatment changes: in all facial-type groups, the fossa was displaced
posteriorly. No differences existed between the 3 groups. Condylar growth and “effective” TMJ changes
were directed more vertically compared with the treatment changes. The changes in posterior direction were
more pronounced in the hyperdivergent group than in the other 2 groups. It was found that the amount and
direction of TMJ growth changes (fossa displacement, condylar growth, and “effective” TMJ changes) were
only temporarily affected favorably in the sagittal direction by Herbst treatment. Condylar growth and
“effective” TMJ changes were directed more posteriorly in hyperdivergent than in hypodivergent Herbst
subjects. This was true for both treatment and posttreatment period changes. (Am J Orthod Dentofacial
Orthop 2004;126:153-61)

T
he effects of the Herbst appliance on vertical increase in mandibular prognathism during Herbst
facial dimensions have been the focus of atten- treatment: (1) increased condylar growth due to condy-
tion in several articles.1-5 Windmiller 6 and Ruf lar remodeling, (2) anterior glenoid fossa displacement
and Pancherz7 showed that Herbst treatment is success- due to fossa remodeling, and (3) anterior positioning of
ful in both hypodivergent (small mandibular plane the condyle within the fossa.
angle) and hyperdivergent (large mandibular plane In previous clinical Herbst studies, the 3 TMJ adaptive
angle) Class II subjects. Furthermore, on a long-term mechanisms were analyzed as single factors by using
basis, Ruf and Pancherz7 demonstrated that the Herbst profile cephalometric roentgenograms,8-12 orthopantomo-
appliance in hypodivergent and hyperdivergent sub- grams,13 computerized tomography scans,14 bone scinto-
jects had no influence on mandibular rotation. graphs,15 and magnetic resonance imaging (MRI).3,16 In
Three adaptive processes in the temporomandibular 3 Herbst studies,3,16,17 the summation effect of the 3 TMJ
joint (TMJ) are thought to be responsible for the adaptive mechanisms (condylar growth, glenoid fossa
displacement, and condyle-fossa relationship changes)
From the Department of Orthodontics, University of Giessen, Giessen, Ger-
many.
was assessed by using lateral headfilms.
a
Professor and chairman. However, for the TMJ growth changes during
b
Assistant professor. Herbst therapy, no study takes into account patients’
Reprint requests to: Prof Dr Hans Pancherz, Department of Orthodontics,
School of Dentistry, University of Giessen, Schlangenzahl 14, D-35392
vertical facial types. The aim of this long-term cepha-
Giesssen, Germany; e-mail, [email protected]. lometric roentgenographic study was to compare hypo-
Submitted, January 2003; revised and accepted, July 2003. divergent and hyperdivergent Herbst subjects by ana-
0889-5406/$30.00
Copyright © 2004 by the American Association of Orthodontists. lyzing the amount and direction of condylar growth and
doi:10.1016/j.ajodo.2003.07.015 fossa displacement changes as single factors as well as
153
154 Pancherz and Michailidou American Journal of Orthodontics and Dentofacial Orthopedics
August 2004

the 3 TMJ adaptive processes (condylar growth, gle- Skieller.21 The condylar head was identified on the
noid fossa displacement, and condylar-fossa relation- mouth-open headfilm and transferred to the mouth-
ship changes) in combination. closed film after mandibular superimpositions of the
radiographs.
MATERIAL AND METHODS A reference line/reference-line perpendicular grid
The original patient sample of consecutive Class II was defined on the T1 headfilm and then transferred to
Division 1 malocclusions treated with the banded type the other headfilms in a series (T2 and T3) after
of Herbst appliance18 comprised 118 subjects. Herbst superimposing the films on the stable bone structures of
therapy resulted in Class I or overcorrected Class I the anterior cranial base.21 The treatment and posttreat-
dental-arch relationships in all 118 patients. From this ment changes of the measuring points were related to
sample, 68 subjects (42 boys and 26 girls) were the reference line/reference-line perpendicular grid.
selected for whom no other active treatment was The sagittal (x) and vertical (y) coordinates of the
performed after Herbst treatment and complete records measuring points were assessed on all films (T1 to T3).
existed at all examination times. The pretreatment age The lines of the grid were defined as follows:
of the subjects ranged from 10 to 16 years (mean age,
12.4 years). The average treatment time was 7 months Reference line (RL): a line that connects the incisal
(5 to 12 months), and no teeth were extracted. edge of the most prominent mandibular central
Lateral cephalometric roentgenograms in habitual incisor and the distobuccal cusp tip of the first
occlusion and with the mouth wide open were available permanent mandibular molar; it corresponds to the
for all subjects. The headfilms were analyzed at 3 x-axis of the grid.
times: before treatment (T1), after treatment (T2), and Reference line perpendicular (RLp): a line perpendic-
5 years (4 to 6 years) after treatment (T3). All roent- ular to RL through the midpoint of sella turcica; it
genograms were taken with the same apparatus. By corresponds to the y-axis of the grid.
using the mandibular plane angle (ML/NSL) measure- Measuring points were defined as follows:
ment at T1, the subjects were divided into 3 facial-type
groups: Anatomic condylar point (Co): the most superior-
posterior point of the condylar head in relation to the
● Normodivergent: ML/NSL ⫽ 26.5°-36.5° (38 sub-
RL/RLp grid. This point was defined on each head-
jects, 24 boys and 14 girls)
film in a series (T1 to T3).
● Hypodivergent: ML/NSL ⱕ 26° (17 subjects, 10
Arbitrary condylar point (Co-A): an arbitrary point in
boys and 7 girls)
the area of the condylar head.20 The point was
● Hyperdivergent: ML/NSL ⱖ 37° (13 subjects, 8 boys
defined on the T1 headfilm and transferred to the
and 5 girls)
other headfilms in a series (T2 and T3) after super-
The age distribution of the subjects was comparable imposing the headfilms on the stable bone structures
in the 3 groups. of the cranial base.21
To reduce the method error in defining the different
measuring points and reference structures, all headfilms Measuring variables
were analyzed twice by the same person (C.M.) with a To assess glenoid fossa displacement, the radio-
2-week interval between the recordings. The mean graphs (T1, T2, T3) were superimposed on the stable
value of the 2 recordings was used as final measuring bone structures of the cranial base.21 The changes of Co
value. Furthermore, at each recording session, all roent- in relation to its T1 position represent the fossa dis-
genograms in a series from each subject were traced placement (Fig 1). The validity of the method depends
and evaluated. Linear measurements were made with a on an unchanged condyle/fossa relationship at all times
ruler to the nearest 0.5 mm. The linear roentgeno- of registration (T1, T2, T3).
graphic enlargement of 7% was not corrected for. To assess condylar growth, the radiographs (T1, T2,
In evaluating the headfilms, the methods (in modi- T3) were superimposed on the stable bone structures of
fied form) of Buschang and Santos-Pinto19 and Creek- the mandible.21 The changes of Co in relation to its T1
more20 were used. position represent condylar growth (Fig 2).
The habitual occlusion headfilms in a series for To assess “effective” TMJ changes, Co-A marked on
each subject (T2 and T3) were superimposed on the T1 the T1 radiograph was transferred to the other radiographs
film. By using stable bone structures for orientation, (T2 and T3) after superimposing the headfilms on the
anterior cranial base and mandibular superimpositions stable bone structures of the cranial base.21 Thereafter, the
were performed according to the method of Björk and radiographs (T1, T2, T3) were superimposed by using the
American Journal of Orthodontics and Dentofacial Orthopedics Pancherz and Michailidou 155
Volume 126, Number 2

Fig 1. Glenoid fossa displacement. Changes of Co Fig 2. Condylar growth. Changes of Co point after
point after cranial base superimposition of radiographs. mandibular superimposition of radiographs.

stable bone structures of the mandible.21 The changes of were comparable in the 3 groups, no sex differentiation
Co-A in relation to its T1 position represent the “effec- was done.
tive” TMJ changes (Fig 3) that are the sum of glenoid Table I gives the position of the measuring points in
fossa displacement, condylar growth, and displacement of the 3 facial-type groups at the different examination
the condyle in the fossa.20 times. Table II shows the corresponding treatment and
posttreatment changes of the measuring points. The
Statistical methods comparisons of the 3 groups are given in Table III.
As an expression of glenoid fossa displacement
The mean value (mean) and the standard deviation
(Tables II and III; Fig 4), the position change of Co was
(SD) were calculated for each variable. Student t tests
assessed by using a cranial base superimposition tech-
for paired samples were used to test the significance of
nique of the headfilms (Fig 1).
changes during different examination periods. Student t
In all 3 facial-type groups, the fossa was, on
tests for unpaired samples were used to test the signif-
average, displaced in an anterior (P ⬍ .05) and inferior
icance of differences between the examination groups
(P ⬍ .05 to P ⬍ .001) direction during treatment
and between sexes. The significance levels were P ⬍
(T2-T1). For the amount of both sagittal and vertical
.001, P ⬍ .01, and P ⬍ .05; P ⬎ .05 was considered not
fossa displacements, no statistically significant (ns)
significant. All calculations were performed with the
group differences existed.
software SPSS for Windows (SPSS, Chicago, Ill).
At posttreatment (T3-T2) in all 3 groups, the fossa
was, on average, displaced in a posterior direction (P ⬍
RESULTS .001). No vertical fossa displacement could be verified
Most of the TMJ growth changes (glenoid fossa, statistically (ns). For the amount of sagittal fossa
condyle, “effective” TMJ changes) during the treatment displacement, no statistically significant group differ-
(T2-T1) and posttreatment (T3-T2) periods were insig- ences existed.
nificantly larger in the boys than in the girls. Statisti- As an expression of condylar growth (Tables II and
cally significant differences were found for sagittal III; Fig 5), the position change of Co was assessed by
fossa changes (T2-T1; P ⬍ .001) and vertical condylar using a mandibular base superimposition technique of
changes (T3-T2; P ⬍ .05) in the hypodivergent group. the headfilms (Fig 2).
However, because the proportions of boys and girls In all 3 facial-type groups, condylar growth was, on
156 Pancherz and Michailidou American Journal of Orthodontics and Dentofacial Orthopedics
August 2004

Fig 3. “Effective” TMJ changes. A, Transference of Co-A point from first headfilm to second
headfilm after cranial base superimposition of radiographs. B, Changes of Co-A point after
mandibular superimposition of radiographs.

average, directed posteriorly (P ⬍ .01 to P ⬍ .001) and As an expression of “effective” TMJ changes (the
superiorly (P ⬍ .01 to P ⬍ .001) during treatment sum of fossa displacement, condylar growth, and con-
(T2-T1). However, growth was directed more posteri- dylar displacement in the fossa) (Tables II and III; Fig
orly in the hyperdivergent group than in the normodi- 6), the position change of Co-A was assessed by using
vergent and hypodivergent groups (Fig 5). For the a mandibular superimposition technique of the head-
amount of sagittal condylar growth, no statistically films (Fig 3).
significant group differences existed. For vertical con- The direction of the “effective” TMJ changes in the 3
dylar growth, the amount was greater in the normodi- facial-type groups was, on average, similar to that of
vergent than in the hyperdivergent group (P ⬍ .05). condylar growth during reatment (T2-T1): the changes
At posttreatment (T3-T2) in all 3 groups, condylar were directed posteriorly (P ⬍ .001) and superiorly (P ⬍
growth was, on average, directed posteriorly (P ⬍ .05 to .001). Posteriorly directed changes were more apparant in
P ⬍ .001) and superiorly (P ⬍ .001). Condylar growth in the hyperdivergent group than in the normodivergent and
the posttreatment period was directed more posteriorly in hypodivergent groups. However, compared with the con-
the hyperdivergent group than in the normodivergent and dylar growth changes, the amount of sagittal and vertical
hypodivergent groups. However, compared with the treat- “effective” TMJ changes in all groups was greater. For the
ment changes, the posttreatment changes in all 3 facial- amount of sagittal “effective” TMJ changes, no statisti-
type groups were more vertically oriented. cally significant group differences existed. The amount of
The amount of sagittal condylar growth was greater vertical “effective” TMJ changes, on the other hand, was
in the hyperdivergent group than in the normodivergent greater in the hypodivergent than in the hyperdivergent
(P ⬍ .01) and hypodivergent (P ⬍ .01) groups. For the group (P ⬍ .01).
amount of vertical condylar growth, no statistically At posttreatment (T3-T2) in all 3 facial-type
significant group differences existed. groups, the “effective” TMJ changes were, on average,
American Journal of Orthodontics and Dentofacial Orthopedics Pancherz and Michailidou 157
Volume 126, Number 2

Table I. Position (mean, SD) of measuring points in 38 normodivergent (Normo), 17 hypodivergent (Hypo), and
13 hyperdivergent (Hyper) Herbst patients at T1 (before treatment), T2 (after treatment), T3 (5 years after
treatment)
T1 T2 T3

Variable Group Mean SD Mean SD Mean SD

Fossa position - sagittal (CBS) mm Normo 14.5 2.63 13.9 3.24 16.1 3.40
Co/RLp Hypo 16.7 2.53 16.2 2.58 18.3 2.61
Hyper 14.2 2.89 13.5 2.55 16.0 2.62
Fossa position - vertical (CBS) mm Normo 35.4 5.02 34.6 5.12 34.3 5.47
Co/RL Hypo 40.8 5.61 39.4 5.73 39.6 5.51
Hyper 33.3 3.22 32.5 3.91 32.6 3.49
Condylar position - sagittal (MBS) mm Normo 14.5 2.62 16.3 3.42 19.7 5.12
Co/RLp - anatomical point Hypo 16.6 2.52 18.0 2.57 19.2 3.18
Hyper 14.2 1.89 16.5 3.77 20.8 4.72
Condylar position - vertical (MBS) mm Normo 35.4 5.01 38.3 6.04 45.1 5.72
Co/RL - anatomical point Hypo 40.8 5.61 43.4 6.26 51.1 6.72
Hyper 33.3 3.22 35.1 4.26 41.8 5.07
Condylar position - sagittal (MBS) mm Normo 10.8 2.62 13.2 3.03 14.4 4.38
Co-A/RLp - arbitrary point Hypo 12.3 2.75 14.0 3.07 13.2 3.51
Hyper 10.1 3.03 13.0 4.43 14.9 5.43
Condylar position - vertical (MBS) mm Normo 31.8 4.81 35.4 5.85 42.6 5.75
Co-A/RL - arbitrary point Hypo 36.1 5.73 40.1 6.22 47.6 6.67
Hyper 29.6 3.36 32.2 3.76 38.9 4.77

CBS, Cranial base superimposition.

MBS, Mandibular base superimposition.

directed posteriorly (ns to P ⬍ .05) and superiorly (P ⬍ growth and “effective” TMJ changes are favorable in
.001). The “effective” TMJ changes in the posttreat- skeletal Class II treatment. All changes contribute to
ment period were directed more posteriorly in the increase mandibular prognathism. On the other hand,
hyperdivergent than in the normodivergent and hypo- posterior-directed fossa displacement and anterior-
divergent groups. Compared with the posttreatment directed condylar growth and “effective” TMJ changes
condylar growth changes, the postreatment “effective” are unfavorable for Class II correction. These changes
TMJ changes in all 3 groups were more vertically contribute to decrease mandibular prognathism. When
oriented. The amount of sagittal “effective” TMJ considering vertical TMJ changes, they will only indi-
changes was greater in the hyperdivergent group than in rectly contribute to the outcome of Class II treatment by
the normodivergent (P ⬍ .05) and hypodivergent (P ⬍ affecting mandibular growth rotation.17
.05) groups. For the amount of vertical “effective” TMJ For the analysis of glenoid fossa displacement
changes, no statistically significant group differences changes, the method of Buschang and Santos-Pinto19
existed. was used. The prerequisite for this method to be valid
is an unchanged condyle-fossa relationship at all exam-
DISCUSSION ination times (T1, T2, and T3). In case of a changed
This is the first study that examines the growth position of the condyle in the fossa (not possible to
effects of the Herbst appliance, or of any orthodontic or ascertain in the analysis of lateral headfilms), this will
orthopedic appliance, on the TMJ and compares Class misleadingly be appraised as a fossa displacement. Ruf
II subjects with different vertical facial morphologies. and Pancherz,3 using MRI of the TMJ, showed an
For this kind of research, the Herbst appliance has unaffected condyle-fossa relationship by Herbst ther-
many advantages when compared with removable func- apy; this means that the condyle had the same position
tional appliances such as the activator, the bionator, and before and after treatment. Furthermore, on a long-term
the Fränkel. The Herbst appliance is fixed to the teeth, basis, Hansen et al,22 using the method of lateral TMJ
works continuously 24 hours a day, and does not tomography, demonstrated a normal centered position
depend on compliance for its correct function. of the condyle in the fossa in Herbst patients 7.5 years
In interpreting the present findings, it must be posttreatment.
remembered that anterior-directed glenoid fossa dis- For the analysis of condylar growth changes, an
placement changes and posterior-directed condylar anatomic condylar point was defined. Because the
158 Pancherz and Michailidou American Journal of Orthodontics and Dentofacial Orthopedics
August 2004

Table II. Changes (mean) of measuring points in 38 normodivergent (Normo), 17 hypodivergent (Hypo), and 13
hyperdivergent (Hyper) Herbst patients; T1 (before treatment), T2 (after treatment), T3 (5 years after treatment)
T2-T1 T3-T2 T3-T1

Variable Group Mean t Sig Mean t Sig Mean t Sig

Fossa displacement - sagittal (CBS) mm Normo ⫺0.6 ⫺2.58 * ⫹2.2 ⫹7.99 *** ⫹1.6 ⫹6.72 ***
Co/RLp Hypo ⫺0.5 ⫺2.16 * ⫹2.1 ⫹7.40 *** ⫹1.6 ⫹4.26 **
Hyper ⫺0.7 ⫺2.41 * ⫹2.5 ⫹6.59 *** ⫹1.8 ⫹5.39 ***
Fossa displacement - vertical (CBS) mm Normo ⫺0.8 ⫺3.16 ** ⫺0.3 ⫺1.53 ns ⫺1.1 ⫺4.11 ***
Co/RL Hypo ⫺1.4 ⫺4.38 *** ⫹0.2 ⫹0.62 ns ⫺1.2 ⫺2.84 *
Hyper ⫺0.8 ⫺2.44 * ⫹0.1 ⫹0.11 ns ⫺0.7 ⫺1.40 ns
Condylar growth - sagittal (MBS) mm Normo ⫹1.8 ⫹7.16 *** ⫹3.4 ⫹5.15 *** ⫹5.2 ⫹7.77 ***
Co/RLp Hypo ⫹1.4 ⫹4.39 ** ⫹1.2 ⫹2.11 * ⫹2.6 ⫹4.47 ***
Hyper ⫹2.3 ⫹5.58 *** ⫹4.3 ⫹4.39 *** ⫹6.6 ⫹5.89 ***
Condylar growth - vertical (MBS) mm Normo ⫹2.9 ⫹9.06 *** ⫹6.8 ⫹9.71 *** ⫹9.7 ⫹15.95 ***
Co/RL Hypo ⫹2.6 ⫹5.50 *** ⫹7.7 ⫹4.81 *** ⫹10.3 ⫹6.40 ***
Hyper ⫹1.8 ⫹3.59 ** ⫹6.7 ⫹5.52 *** ⫹8.5 ⫹7.30 ***
“Effective” TMJ changes - sagittal (MBS) mm Normo ⫹2.4 ⫹9.41 *** ⫹1.2 ⫹2.12 * ⫹3.6 ⫹6.21 ***
Co-A/RLp Hypo ⫹1.7 ⫹4.88 *** ⫺0.8 ⫺1.35 ns ⫹0.9 ⫹1.87 ns
Hyper ⫹2.9 ⫹5.44 *** ⫹1.9 ⫹1.76 ns ⫹4.8 ⫹3.87 **
“Effective”TMJ changes - vertical (MBS) mm Normo ⫹3.6 ⫹11.08 *** ⫹7.2 ⫹9.4 *** ⫹10.7 ⫹15.39 ***
Co-A/RL Hypo ⫹4.0 ⫹12.78 *** ⫹7.5 ⫹4.72 *** ⫹11.5 ⫹7.71 ***
Hyper ⫹2.6 ⫹7.49 *** ⫹6.7 ⫹7.00 *** ⫹9.3 ⫹9.61 ***

CBS, Cranial base superimposition.

MBS, Mandibular base superimposititon.

Levels of significance Plus (⫹) means: Minus (⫺) means:

***(P ⬍ .001) Posterior condylar growth Anterior condylar growth
**(P ⬍ .01) Superior condylar growth Inferior condylar growth
*(P ⬍ .05) Posterior fossa displacement Anterior fossa displacement
ns (P ⱖ .05) Superior fossa displacement Inferior fossa displacement
Posterior “effective” TMJ change Anterior “effective” TMJ changes
Superior “effective” TMJ change Inferior “effective” TMJ changes

condylar head is difficult or impossible to identify on ically in mandibular-protrusion experiments in ani-

habitual occlusion headfilms, mouth-open headfilms mals24-30 and in earlier clinical Herbst studies with
were also used to visualize the condyle and then to MRI of the TMJ region.3,16
transfer its countour to the mouth-closed films after During the posttreatment period (T3-T2), fossa
mandibular superimposition of the radiographs. displacement in all facial-type groups was in a posterior
To circumvent the problem of defining reliable ana- direction. Therefore, in the total observation period
tomic measuring points for assessing fossa displacement (T3-T1), the direction of fossa displacement posteriorly
and condylar growth changes, the “effective” TMJ chang- and inferiorly corresponded to that of normal growth
es20 were calculated by using an arbitrary condylar point changes.19,25
(Co-A). Thus, it was possible to assess quantitatively the When comparing the 3 facial-type groups for the
sum of fossa displacement, condylar growth, and possible amount and direction of fossa dispalcement changes
condylar position changes in the fossa. during the treatment and posttreatment periods, no
differences were found.
Glenoid fossa displacement It seems that the Herbst appliance has only a
During normal growth, the glenoid fossa is dis- temporary impact on glenoid fossa displacement in an
placed in a posterior-inferior direction.19,23 During anterior direction during the active phase of treatment,
Herbst treatment (T2-T1), the fossa in all groups was thus contributing to mandibular advancement and fa-
displaced in an anterior and inferior direction. This was cilitating Class II correction. The amount and direction
most likely the result of remodeling processes at the of the fossa changes seem to be independent of the
posterior fossa wall as has been demonstrated histolog- patients’ vertical facial types.
American Journal of Orthodontics and Dentofacial Orthopedics Pancherz and Michailidou 159
Volume 126, Number 2

Table III. Changes (mean) of measuring points in 38 normodivergent (Normo), 17 hypodivergent (Hypo), and 13
hyperdivergent (Hyper) Herbst patients; T1 (before treatment), T2 (after treatment), T3 (5 years after treatment)
T2-T1 T3-T2 T3-T1

Variable Group Mean t Sig Mean t Sig Mean t Sig

Fossa displacement - sagittal (CBS) mm Normo-Hypo ⫺0.1 ⫺0.51 ns ⫹0.1 ⫹0.33 ns ⫹0.0 ⫹0.48 ns
Co/RLp Normo-Hyper ⫹0.1 ⫹0.36 ns ⫺0.3 ⫺0.74 ns ⫹0.2 ⫹0.53 *
Hypo-Hyper ⫹0.2 ⫹0.80 ns ⫺0.4 ⫺0.49 ns ⫺0.2 ⫺0.35 ns
Fossa displacement - vertical (CBS) mm Normo-Hypo ⫺0.6 ⫺1.52 ns ⫺0.5 ⫺1.48 ns ⫹0.1 ⫹0.61 ns
Co/RL Normo-Hyper ⫹0.0 ⫹0.19 ns ⫺0.4 ⫺0.73 ns ⫺0.4 ⫺0.59 ns
Hypo-Hyper ⫺0.6 ⫺1.12 ns ⫹0.1 ⫹0.19 ns ⫺0.5 ⫺0.57 ns
Condylar growth - sagittal (MBS) mm Normo-Hypo ⫹0.4 ⫹0.86 ns ⫹2.2 ⫹2.69 ** ⫹2.6 ⫹2.98 **
Co/RLp Normo-Hyper ⫺0.5 ⫺1.06 ns ⫺0.9 ⫺0.79 ns ⫺1.4 ⫺1.11 ns
Hypo-Hyper ⫺0.9 ⫺1.64 ns ⫺3.1 ⫺2.87 ** ⫺4.0 ⫺3.22 **
Condylar growth - vertical (MBS) mm Normo-Hypo ⫹0.3 ⫹0.42 ns ⫺1.1 ⫺0.56 ns ⫺0.6 ⫺0.42 ns
Co/RL Normo-Hyper ⫹1.1 ⫹1.89 * ⫹0.1 ⫹0.00 ns ⫹1.2 ⫹1.00 ns
Hypo-Hyper ⫹0.8 ⫹1.35 ns ⫹1.0 ⫹0.49 ns ⫹1.8 ⫹0.96 ns
“Effective” TMJ changes - sagittal (MBS) mm Normo-Hypo ⫹0.7 ⫹1.45 ns ⫹2.0 ⫹2.43 * ⫹2.7 ⫹3.55 ***
Co-A/RLp Normo-Hyper ⫺0.5 ⫺0.97 ns ⫺0.7 ⫺0.55 ns ⫺1.2 ⫺0.91 ns
Hypo-Hyper ⫺1.2 ⫺1.87 ns ⫺2.7 ⫺2.19 * ⫺3.9 ⫺2.93 **
“Effective” TMJ changes - vertical (MBS) mm Normo-Hypo ⫺0.4 ⫺1.14 ns ⫺0.3 ⫺0.19 ns ⫺0.8 ⫺0.51 ns
Co-A/RL Normo-Hyper ⫹1.0 ⫹1.92 ns ⫹0.5 ⫹0.39 ns ⫹1.4 ⫹1.15 ns
Hypo-Hyper ⫹1.4 ⫹3.00 ** ⫹0.8 ⫹0.43 ns ⫹2.2 ⫹1.24 ns

CBS, Cranial base superimposition.

MBS, Mandibular base superimposition.
Levels of significance
***(P ⬍ .001)
**(P ⬍ .01)
*(P ⬍ .05)
ns (P ⱖ .05)

Fig 4. Glenoid fossa displacement. Average changes of Co point in 38 normodivergent (Normo), 17

hypodivergent (Hypo), and 13 hyperdivergent (Hyper) Herbst subjects. T1, before treatment; T2,
after treatment; T3, 5 years after treatment.

Condylar growth jects and a sagittal condylar growth pattern (predomi-

In untreated subjects having different vertical facial nantly posterior-directed condylar growth) in
morphologies, Björk and Skieller21 demonstrated a hyperdivergent subjects. These growth patterns corre-
vertical condylar growth pattern (predominantly supe- spond to those of our hypodivergent and hyperdiver-
rior-directed condylar growth) in hypodivergent sub- gent Herbst subjects. However, Herbst treatment stim-
160 Pancherz and Michailidou American Journal of Orthodontics and Dentofacial Orthopedics
August 2004

The results indicate that Herbst treatment tempo-

rarily stimulates condylar growth, especially in the
posterior direction, and that hyperdivergent subjects
react more readily than hypodivergent subjects. Clini-
cally, this implies that, for the mandibular growth
contribution to Class II correction, the Herbst appliance
is more efficient in hyperdivergent than in hypodiver-
gent subjects.34
“Effective” TMJ changes
The “effective” TMJ changes represent the sum of
glenoid fossa displacement, condylar growth, and con-
dylar position changes in the fossa. The pattern of
“effective” TMJ changes in the 3 facial-type groups
Fig 5. Condylar growth. Average changes of Co point in was basically the same as that of condylar growth.
38 normodivergent (Normo), 17 hypodivergent (Hypo), However, during treatment (T2-T1) in all groups,
and 13 hyperdivergent (Hyper) Herbst subjects. T1, the amount of vertical “effective” TMJ changes was
before treatment; T2, after treatmentl; T3, 5 years after greater than the corresponding amount of vertical
treatment. condylar growth changes. Furthermore, during the
posttreatment period (T3-T2), the growth direction
changes to a more vertical pattern were more pro-
nounced for “effective” TMJ changes (Fig 6) than for
condylar growth (Fig 5). Additionally, in the normodi-
vergent and hypodivergent groups, the amount of
vertical “effective” TMJ change was greater than the
corresponding condylar growth changes.
When comparing these patterns of “effective” TMJ
changes and condylar growth, the differences could be
explained by the opposing glenoid fossa displacement
changes during and after Herbst thearpy (Fig 4). In
“effective” TMJ changes, anterior and inferior fossa
displacement changes during treatment (T2-T1) are
added to posterior and superior condylar growth,
whereas posterior fossa displacement changes post-
Fig 6. “Effective” TMJ changes. Average changes of
treatment (T3-T2) are subtracted from posterior condy-
Co-A point in 38 normodivergent (Normo), 17 hypodi-
vergent (Hypo), and 13 hyperdivergent (Hyper) Herbst
lar growth.
subjects. T1, before treatment; T2, after treatment; T3, 5 The results indicate that, in the 3 facial-type groups,
years after treatment. the “effective” TMJ changes mainly reflect the corre-
sponding condylar growth changes. Glenoid fossa dis-
placement changes during and after Herbst treatment
ulates condylar growth in a posterior direction,8,12,30 affect, however, the pattern of the “effective” TMJ
and this seems particularly to be the case in hyperdi- changes.
vergent subjects. A stimulation of condylar growth in
especially the posterior direction as a response to CONCLUSIONS
mandibular advancement appliances has also been ver- The amount and direction of TMJ growth changes
fied histologically in animals25,29-33 and with radiogra- (fossa displacement, condylar growth, and “effective”
phy13,14 and MRI3,16 in humans. TMJ changes) were only temporarily affected favorably
When comparing the treatment (T2-T1) and post- in the sagittal direction by Herbst treatment. For gle-
treatment (T3-T2) period changes, condylar growth noid fossa displacement changes, no differences existed
direction in all facial-type groups became more verti- between hypodivergent and hyperdivergent subjects at
cally oriented posttreatment. This could be due to any examination period. But condylar growth and
recovery after Herbst therapy. The amount and direc- “effective” TMJ changes, on the other hand, were
tion of growth return to their original patterns.9,12 directed more posteriorly in hyperdivergent than in
American Journal of Orthodontics and Dentofacial Orthopedics Pancherz and Michailidou 161
Volume 126, Number 2

hypodivergent Herbst subjects during treatment and radiographic investigation. Am J Orthod Dentofacial Orthop
posttreatment. 1999;115:607-18.
17. Pancherz H, Ruf S, Kohlhas P. “Effective condylar growth” and
chin position changes in Herbst treatment: a cephalometric
REFERENCES roentgenographic long-term study. Am J Orthod Dentofacial
1. Pancherz H. Vertical dentofacial changes during Herbst appli- Orthop 1998;114:437-46.
ance treatment. A cephalometric investigation. Swed Dent J 18. Pancherz H. The Herbst appliance—its biologic effects and
1982;15(Suppl 15):189-96. clinical use. Am J Orthod 1985;87:1-20.
2. Pancherz H, Anehus-Pancherz M. The head-gear effect of the 19. Buschang P, Santos-Pinto A. Condylar growth and glenoid fossa
Herbst appliance. A cephalometric long-term study. Am J Orthod displacement during childhood and adolescence. Am J Orthod
Dentofacial Orthop 1993;103:510-20. Dentofacial Orthop 1998;113:437-42.
3. Ruf S, Pancherz H. Temporomandibular joint growth adaptation 20. Creekmore T. Inhibition or stimulation of the vertical growth of
in Herbst treatment: a prospective magnetic resonance imaging the facial complex, its significance to treatment. Angle Orthod
and cephalometric roentgenographic study. Eur J Orthod 1998; 1967;37:285-97.
20:375-88. 21. Björk A, Skieller V. Normal and abnormal growth of the
4. Kruse J. Rotationsformen des Unterkiefers bei der Herbst- mandible. A synthesis of longitudinal cephalometric implant
Behandlung. Eine röntgenkephalometrische Untersuchung der studies over a period of 25 years. Eur J Orthod 1983;5:1-46.
Langzeitveränderungen der Total-, Matrix- und Intramatrix- 22. Hansen K, Pancherz H, Petersson A. Long-term effects of the
Rotation [dissertation]. Giessen, Germany: University of Gies- Herbst appliance on the craniomandibular system with special
sen; 1999. reference to the TMJ. Eur J Orthod 1990;12:244-53.
5. Croft RS, Buschang PH, English JD, Meyer R. A cephalometric 23. Björk A, Skieller V. Facial development and tooth eruption. An
and tomographic evaluation of Herbst treatment in the mixed implant study at the age of puberty. Am J Orthod 1972;62:339-
dentition. Am J Orthod Dentofacial Orthop 1999;116:435-43. 83.
6. Windmiller EC. The acrylic-splint Herbst appliance: a cephalo- 24. Breitner C. Experimentelle Veränderungen der mesiodistalen
metric evaluation. Am J Orthod Dentofacial Orthop 1993;104: Beziehungen der oberen und unteren Zahnreihen. Zschr Stomatol
73-84. 1930;28:620-35.
7. Ruf S, Pancherz H. The effect of Herbst appliance treatment on 25. Stöckli PW, Willert HG. Tissue reactions in the temporoman-
the mandibular plane angle: a cephalometric roentgenographic dibular joint resulting from anterior displacement of the mandi-
study. Am J Orthod Dentofacial Orthop 1996;110:225-9. ble in the monkey. Am J Orthod 1971;60:142-55.
8. Pancherz H. Treatment of Class II malocclusion by jumping the 26. Woodside DG, Metaxas A, Altuna G. The influence of functional
bite with the Herbst appliance. A cephalometric investigation. appliance therapy on glenoid fossa remodeling. Am J Orthod
Am J Orthod 1979;76:423-42. Dentofacial Orthop 1987;92:181-98.
9. Pancherz H. The effect of continuous bite jumping on the 27. Rabie ABM, Zhao Z, Shen G, Hägg U, Robinson W. Osteogen-
dentofacial complex: a follow-up study after Herbst appliance esis in the glenoid fossa in response to mandibular advancement.
treatment of Class II malocclusion. Eur J Orthod 1981;3:49-60. Am J Orthod Dentofacial Orthop 2001;119:390-400.
10. Wieslander L. Intensive treatment of severe Class II malocclu- 28. Rabie ABM, Chayanupatkul A, Hägg U. Stepwise advancement
sions with a headgear-Herbst appliance in the early mixed using functional appliances. Experimental perspective. Semin
dentition. Am J Orthod 1984;86:1-13. Orthod 2003;9:41-6.
11. Pancherz H, Hägg U. Dentofacial orthopedics in relation to 29. Peterson JE, McNamara JA. Temporomandbular joint adaptation
somatic maturation. An analysis of 70 consecutive cases treated associated with Herbst appliance treatment in juvenile rhesus
with the Herbst appliance. Am J Orthod 1985;88:273-87. monkeys (Macaca mulatta). Semin Orthod 2003;9:12-25.
12. Pancherz H, Littmann C. Morphologie und Lage des Unterkief- 30. McNamara JA, Peterson JE, Pancherz H. Histologic changes
ers bei der Herbst-Behandlung: eine kephalometrische Analyse associated with the Herbst appliance in adult rhesus monkeys
der Veränderungen bis zum Wachstumsabschluss. Inf Kiefer- (Macaca mulatta). Semin Orthod 2003;9:26-40.
orthop 1989;8:215-28. 31. McNamara JA, Carlson DS. Quantitative analysis of temporo-
13. Paulsen HU. Changes of the condyles of 100 patients treated with mandibular joint adaptation to protrusive function. Am J Orthod
the Herbst appliance from puberty to adulthood: a long-term 1979;76:593-611.
radiographic study. Eur J Orthod 1997;19:657-8. 32. McNamara JA, Hinton RJ, Hoffman DL. Histologic analysis of
14. Paulsen HU, Karle A, Bakke M, Herskind A. A CT-scanning and temporomandibular joint adaptation to protrusive function in
radiographic analysis of temporomandibular joints and cephalo- young adult rhesus monkeys (Macaca mulatta). Am J Orthod
metric analysis in a case of Herbst treatment in late puberty. Eur 1982;82:288-98.
J Orthod 1995;17:165-75. 33. Woodside DG, Altuna G, Harvold E, Herbert M, Metaxas A.
15. Paulsen HU, Rabol A, Sorensen S. Bone scintigraphy of human Primate experiments in malocclusion and bone induction. Am J
temporomandibular joints during Herbst treatment: a case report. Orthod 1983;83:460-8.
Eur J Orthod 1988;20:369-74. 34. Ruf S, Pancherz H. The mechanism of Class II correction during
16. Ruf S, Pancherz H. Temporomandibular joint remodeling in Herbst therapy in relation to the vertical jaw base relationship: a
adolescents and young adults during Herbst treatment: a prospec- cephalometric roentgenographic study. Angle Orthod 1997;67:
tive longitudinal magnetic resonance imaging and cephalometric 271-6.