Clin Auton Res (2008) 18:96–104

DOI 10.1007/s10286-008-0462-3 RESEARCH ARTICLE

Jean-Luc Elghozi Tuba players reproduce a Valsalva

Arlette Girard
Philippe Fritsch maneuver while playing high notes
Dominique Laude
Jean-Luc Petitprez

j Abstract Playing wind instru- high notes (24 beats/minute) and

Received: 30 November 2007
Accepted: 13 March 2008 ments requires expiratory efforts. Valsalva 60 (33 beats/minute).
Published online: 14 April 2008 Blowing low notes on a tuba High baroreflex sensitivity was
means a low resistance to expira- associated with a high HR
tion while playing high notes response. Total peripheral resis-
requires a strenuous expiratory tance was reflexly elevated to
strain. The resulting high intra- counteract cardiac output reduc-
thoracic pressure may reproduce a tions. During rebreathing, BP
Valsalva maneuver. Ten tuba overshoots reached 30 mmHg for
players were asked to blow med- the high note and 53 mmHg for
ium loud long (15 seconds) notes the Valsalva 60. Altogether, these
at three different pitches (low, findings indicate that blowing
middle, and high) and to perform notes on a tuba reproduces the
Valsalva maneuvers at 10, 40, and cardiovascular changes seen with a
60 mmHg. Blood pressure (BP) Valsalva maneuver with the effects
was measured continuously with a of blowing high notes being close
Finapres
monitor. The four to a classic (40 mmHg) Valsalva
classic phases of the Valsalva maneuver. In addition, the baro-
maneuver were reproduced with reflex sensitivity might be a pre-
J.-L. Elghozi (&) Æ A. Girard Æ D. Laude
INSERM U872 the notes. The expiratory effort dictor of the ability to surmount
Clinical Pharmacology Unit produced systolic BP and pulse the BP intolerance that could
Assistance Publique – Hôpitaux de Paris pressure falls reaching about occur during wind instrument
Hôpital Necker 24 mmHg with the high note or playing.
149 rue de Sèvres
75015 Paris, France
the Valsalva 60. BP and pulse
E-Mail: [email protected] pressure falls were linearly related
to stroke volume reductions. j Key words autonomic nervous
P. Fritsch Æ J.-L. Petitprez
Conservatoire National Supérieur Reflex heart rate (HR) changes system Æ baroreceptors Æ
de Musique were inversely related to BP falls, hypotension Æ music Æ
Paris, France with maximal increases for the venous pressure

function of wind-instrument players has been studied

Introduction for more than 40 years [3, 14, 29]. The general prin-
ciple for emitting a note is based on a sound source
Listening to music [1], singing [34], or playing music (vibrating lips or vibrating reed) and on a tube of
CAR 462

[3] may induce cardiovascular changes, which are variable geometry acting as a resonator [15]. The
largely dependent upon respiratory pattern. Pulmonary musician drives a regulated flow of air, which requires
 97

a high degree of voluntary breath control. Breath volunteers. Five subjects were light smokers (4.4 ± 0.6 cigarettes/
performances differ between the instruments. The day for a duration of 4 ± 2 years corresponding to 0.9 ± 0.3 pack-
years of smoking). None was on medication. All subjects gave in-
tenor tuba (small tuba) is of particular interest since formed consent to participate in the study which was performed
different breathing patterns are required to blow low with the approval of the Necker ethics committee (17/10/2006).
and high notes. Low notes require a high flow rate but
relatively low mouth pressure. After a deep inspira-
tion, the passive expiratory forces at high lung vol- j Procedures
umes are sufficient to generate a small pressure. The
embouchure is adjusted to loosely maintain the lips, Recordings were obtained with subjects in a seated posture in a
and a high flow of air passes through the bore of the quiet room dedicated to the cardiovascular recordings, at ambient
temperature (21C). Each participant brought his own instrument
mouthpiece. The low frequency of vibrating lips gives and mouthpiece. A practice mute for euphonium model DW5512
rise to a low note. For a high note, an active expira- (Denis Wick, Coventry, UK) was placed in the bell to limit noise in
tory effort is needed whenever the required mouth nearby patients’ rooms.
pressure is high. A high intrathoracic pressure results Blood pressure (BP) was measured continuously with a Fina-
pres
monitor (model 2300, Ohmeda, Limonest, France) from the
from the contraction of the diaphragm and abdominal third finger of the left hand, which held the instrument as usual.
muscles, and the subsequent compression of the lungs This position maintained the hand at the heart level. The right hand
produces a high expiratory pressure. The embouchure was positioned on the pistons. The RR interval was obtained from a
is tightened and lips vibrate at a high frequency to three-lead ECG montage. Respiratory movements were obtained by
produce the high note. The expiratory pressure is inductance plethysmography using a respiratory belt.
Notes were blown as follows. After an initial 5 minutes
high but the airflow remains small, since the tightened recording at rest, three natural notes (without the use of the pis-
lips hold the air into the lungs and pharyngeal cavity. tons) were blown for 15 seconds each, with a constant mf
Finally, the airflow may be modulated to adjust dynamics: a pedal note (B flat tone, 117 Hz, the lowest natural note
the sound level of a given note, and increasing of the instrument or fundamental of the harmonic series), a middle
note (B flat tone, 468 Hz, two octaves above the pedal note), and a
the airflow increases loudness. Therefore, it is not high note (B flat tone, 937 Hz, three octaves above the pedal tone).
easy to sustain loud low notes since airflow require- Each note was played three times, with 1–2 minutes between
ments in this mode may easily reach the limits of the each blow. Notes were blown in an ascending rank: three low, three
individual. middle, and three high notes. A recovery interval (10 minutes)
preceded each new set of three notes. After ending the notes, the
The aim of this study was to test whether the high instrument was put aside.
intrathoracic pressure achieved while playing high Valsalva maneuvers with an open glottis were performed as a
notes with a tenor tuba may induce cardiovascular second part of the study, in the same sitting position with the left
changes comparable to a Valsalva maneuver [16]. hand maintained at the heart level and the right hand holding a
This maneuver has been widely used when evaluat- polycarbonate mouthpiece 6½ AL from Kelly mouthpieces
(Fond du Lac, WI), close to the instrument mouthpiece, which
ing the function of the autonomic nervous system [6, was sealed to the rubber tubing of the manometer. This adapta-
25]. Medium loud (mezzoforte, mf) notes were pre- tion made the maneuver easy for the musicians. The subject
ferred to allow the production of long (15 seconds) voluntarily maintained the meniscus of a mercury U-manometer
notes within the breathing capacities of the musi- at three constant levels during 15 seconds: 10, 40 (classic Valsalva
maneuver), and 60 mmHg (named Valsalva 10, 40, and 60). In
cians. Low, middle, and high notes were compared contrast to the notes emission, the lips did not vibrate during the
to Valsalva maneuvers performed at 10, 40, and Valsalva maneuver and the expiratory effort was not giving rise to
60 mmHg. These pressures were selected to include a flow of air. The levels of the meniscus were very stable for each
the common range of pressure measured in the requested pitch for the whole 15 seconds period, which was not
surprising because these subjects were expert in controlling
mouthpiece of the instrument [3]. Estimates of intra- expiratory efforts.
arterial pressures were obtained using a non-invasive
finger plethysmographic device, validated under
resting conditions and during Valsalva maneuvers j Measurements
[4, 18].
The three signals (BP, electrocardiogram and respiration) were
digitized at 500 Hz and recorded using a MP 35
data acquisition
system from Biopac Systems (Goleta, CA). Systolic, mean, diastolic
Methods BP, pulse pressure, RR interval, or instant heart rate (HR) were
extracted using 3.7 Biopac Student Lab PRO
software from Biopac
j Subjects Systems. The Beatscope
software (Finapres Medical Systems,
Amsterdam, The Netherlands) was used for deriving the beat-to-
Ten (nine males, one female) healthy tuba players (saxhorn or beat stroke volume from the BP waveform [19, 24]. Mean BP,
euphonium) participated in the study. Their age was instantaneous HR, and stroke volume were used to calculate car-
29.7 ± 4.1 years (m ± SEM, age range 18–55 years), bodyweight diac output and total peripheral resistance.
was 74.8 ± 4.9 kg, height 173 ± 3 cm, body mass index was Resting values were obtained over the 30 seconds preceding
24.9 ± 1.0 kg/m2. Blood tests (creatinin, ionogram, blood count, each series of notes or Valsalva maneuvers. Measures obtained for
glucose level, cholesterol) were within normal limits for all the each set of three notes or three Valsalva maneuvers were averaged.
98

Table 1 Average control cardiovascular values (mean ± SEM) obtained in the ten tuba players calculated over the 30 seconds preceding the notes or the Valsalva
maneuvers

Low note Middle note High note Valsalva 10 Valsalva 40 Valsalva 60

Systolic BP (mmHg) 121.5 ± 3.8 118.7 ± 3.4 126.5 ± 4.9 117.5 ± 4.8 120.0 ± 4.9 122.1 ± 4.2
Diastolic BP (mmHg) 74.7 ± 1.8 73.5 ± 1.4 77.3 ± 2.4 70.5 ± 3.3 71.5 ± 3.5 73.7 ± 3.0
Pulse pressure (mmHg) 46.8 ± 2.6 45.3 ± 2.5 49.2 ± 3.3 47.0 ± 3.0 48.5 ± 2.7 48.3 ± 3.2
HR (beats/min) 79.9 ± 3.0a 79.3 ± 3.2 79.5 ± 3.5 77.6 ± 3.4 75.7 ± 3.7 75.8 ± 3.6
RR interval (ms) 763.9 ± 29.1b 770.9 ± 30.3 770.5 ± 31.9 785.9 ± 32.4 811.5 ± 35.8 813.3 ± 37.0
Mean BP (mmHg) 90.4 ± 2.2 89.0 ± 1.8 93.9 ± 2.9 85.7 ± 3.6 87.1 ± 3.7 88.5 ± 3.2
Stroke volume (ml) 71.0 ± 2.1 71.1 ± 2.8 72.8 ± 3.8 70.4 ± 3.4 73.4 ± 2.9 71.2 ± 3.5
Cardiac output (l/min) 5.6 ± 0.2 5.6 ± 0.2 5.7 ± 0.2 5.5 ± 0.3 5.6 ± 0.3 5.4 ± 0.3
Total peripheral resistance (mmHg Æ s/ml) 0.97 ± 0.04 0.96 ± 0.03 1.01 ± 0.05 0.97 ± 0.07 0.97 ± 0.06 1.03 ± 0.07

BP blood pressure, HR heart rate

A one-way ANOVA for repeated measures, followed by a post-hoc Tukey’s test for multiple comparisons was used to estimate the influence of the period on the
cardiovascular parameters
a
HR of the low note control period was significantly higher than the HR of the Valsalva 40 and 60 control periods
b
RR interval of the low note control period was significantly smaller than the RR interval of the Valsalva 60 control period

Spontaneous baroreflex sensitivity (BRS) was estimated as the resting value. Table 2 summarizes the BP and HR
gain of the transfer function between systolic BP and RR intervals data. The general picture was a gradation of responses
in the low frequency (LF) range [23]. The gain had units of ms/
mmHg. Gain was calculated on 205 s stationary segments of the from low to middle and high notes and from Valsalva
resting periods preceding the ventilatory strains. 10 to Valsalva 40 and Valsalva 60. Changes in systolic
BP, pulse pressure, HR, and RR intervals reached
significance for almost all phases versus the control
j Statistics
period and are not included in Table 2 for reasons of
Results are expressed as means ± SEM. A oneway ANOVA for re- clarity. Changes in diastolic BP, of lesser amplitude,
peated measures, followed by a post-hoc Tukey’s test for multiple usually reached significance for the high note and the
comparisons was used to estimate the influence of notes and Val- higher Valsalva maneuvers. When the changes in BP
salva maneuvers on the cardiovascular parameters. Relationships or heart rhythm resulting from a note or from
between stroke volume changes and HR or systolic BP changes
were evaluated by linear regression.
a Valsalva maneuver were compared, the highest
Valsalva (40 and 60) did not usually differ from the
high note. However, the maneuvers with a high
respiratory strain (Valsalva 40 and 60 and high note)
Results differed from the low note or from the Valsalva 10.
The greatest effects were seen firstly during the
Resting levels are summarized in Table 1. BP levels early expiratory effort (Phase IIa) when the systolic
were stable over the recording session and were found BP and pulse pressure falls reached about 24 mmHg
in the normal range (120/75 mmHg). A slight graded with the high note or the Valsalva 60, and secondly
HR decrease from 80 to 76 bpm occurred during the during rebreathing (Phase IV) with BP overshoots of
session, corresponding to a 68 ms increase in RR 30 mmHg for the high note and 53 mmHg for the
interval. The hemodynamic estimates were also stable. Valsalva 60. HR increased during BP falls, with
This indicated that the recovery periods were of long maximal increases observed during phase IIb, of
enough duration. 24 beats/minutes for the high notes and 33 beats/
Examples of the effects of the three different notes minutes for Valsalva 60. Conversely, the BP overshoot
and the three different Valsalva maneuvers obtained (Phase IV) produced a reflex HR decrease averaging
in one subject are shown in Figs. 1 and 2. 20 beats/minutes for the high note and 22 beats/
The analysis of each maneuver (note or Valsalva) minutes for the Valsalva 60.
followed the same pattern. The recording was divided The hemodynamic changes derived from phase IIa
into five periods corresponding to: (1) the initial BP were compared to the resting levels. The effects of the
rise following the deep inspiration (Phase I), (2) the three notes were combined as were those of the three
early BP fall until a minimal level (Phase IIa), (3) the Valsalva maneuvers in Fig. 3. Stroke volume
late phase II (IIb) associated with a graded BP res- decreases were associated with reductions in BP. The
toration, (4) the short phase III during which the note subsequent and reflexly mediated HR increases (RR
or Valsalva strain stopped and (5) the overshoot in BP interval decreases) as well as the peripheral resistance
or phase IV at rebreathing. Individual variations were increases were also related to the decrease in stroke
expressed as absolute changes from the corresponding volume. The relation between HR changes and stroke
 99

Fig. 1 Time course of the changes in low note middle note high note
respiration (arbitrary units) and
cardiovascular variables following a low, Respiration
a middle and a high note blown by one au
subject for 15 seconds from 0 second.
The roman numbers correspond to the 220
phases described in the text Blood I IIb IV
Pressure 140
mmHg
60
IIa III
80
Stroke Volume
40
ml
0
140
Heart Rate
100
beats/min
60
8
Cardiac
Output 4
l/min
0
Total 2
Peripheral
Resistance
mmHg.s/ml
0
−30 −15 0 +15 +30 −30 −15 0 +15 +30 −30 −15 0 +15 +30
Time s

Fig. 2 Time course of the changes in Valsalva 10 Valsalva 40 Valsalva 60

respiration and cardiovascular variables
following a 15 seconds Valsalva Respiration
maneuver performed at 10, 40 and au
60 mmHg by the subject selected for
160 I IV
Fig. 1. The roman numbers correspond IIb
to the phases described in the text Blood
Pressure 90
mmHg
40 IIa III
140
Stroke Volume
70
ml
0
120
Heart Rate
80
beats/min
40
8
Cardiac
Output 6
l/min
4
Total 2.5
Peripheral
Resistance
mmHg.s/ml
0
−30 −15 0 +15 +30 −30 −15 0 +15 +30 −30 −15 0 +15 +30
Time s

volume changes was linear while peripheral resistance Finally, the spontaneous BRS values (LF gain in ms/
changes (nearly exponential rises) occurred with mmHg) obtained during a resting period were com-
marked stroke volume decreases (below )40%). pared to the maximal HR increase (RR interval de-
100

Table 2 Average phase by phase changes in cardiovascular values (mean ± SEM) resulting from the notes or the Valsalva maneuvers obtained in the ten tuba
players

Low note Middle note High note Valsalva 10 Valsalva 40 Valsalva 60

Phase I
Systolic BP (mmHg) 15.7 ± 1.5 15.0 ± 2.2 18.6 ± 2.0 17.8 ± 2.7 36.8 ± 3.1* 44.7 ± 4.6*
Diastolic BP (mmHg) 9.9 ± 1.1 10.4 ± 1.6 22.7 ± 3.4* 10.6 ± 1.2 25.0 ± 1.8* 36.1 ± 2.7*
Pulse pressure (mmHg) 5.8 ± 1.3 3.3 ± 1.6 )3.8 ± 4.1* 7.2 ± 2.2 11.8 ± 2.8 8.6 ± 3.2
HR (beats/min) )9.3 ± 1.9 )4.1 ± 1.6 )0.7 ± 2.0* )9 ± 1.6 )6.3 ± 2.7 )7.4 ± 2.9
RR interval (ms) 88.9 ± 18.3 44.8 ± 17.5 7.2 ± 19.8* 106.0 ± 17.5 80.7 ± 34.8 89.6 ± 37.5
Phase IIa
Systolic BP (mmHg) )5.2 ± 2.7 )11.5 ± 3.6 )24.1 ± 3.8* )9.9 ± 3.4 )25.4 ± 5.3* )23.9 ± 4.7*
Diastolic BP (mmHg) )6.4 ± 1.4 )7.0 ± 1.9 )2.9 ± 2.9 )6.5 ± 1.0 )4.7 ± 3 2.0 ± 3.5*
Pulse pressure (mmHg) 1.2 ± 1.9 )5.6 ± 2.9 )21.3 ± 3.8* )3.4 ± 2.5 )20.8 ± 3.2* )25.9 ± 2.8*
HR (beats/min) )0.5 ± 2.4 4.8 ± 2.6 13.7 ± 3.1* )1.5 ± 3.0 13.4 ± 3.2* 18.08 ± 3.3*
RR interval (ms) 2.4 ± 21.2 )41.5 ± 25.0 )107.4 ± 24.9* 19.5 ± 29.0 )127.1 ± 31.0* )163.5 ± 33.2*
Phase IIb
Systolic BP (mmHg) 14.6 ± 3.4 16.1 ± 3.6 29.9 ± 8.8 5.6 ± 2.7 9.4 ± 7.5 10.2 ± 5.7
Diastolic BP (mmHg) 11.5 ± 3.3 12.3 ± 2.9 30.6 ± 5.4* 3.1 ± 1.1 16.4 ± 3.7 23.4 ± 3.8*
Pulse pressure (mmHg) 3.1 ± 1.1 3.1 ± 1.6 )0.2 ± 5.4 2.5 ± 2.3 )7.0 ± 4.4 )13.2 ± 4.0*
HR (beats/min) 4.8 ± 2.9 10.6 ± 2.6 24 ± 3.0* )0.7 ± 4.5 22.8 ± 3.7* 32.9 ± 4.5*
RR interval (ms) )44.9 ± 26.7 )81.2 ± 24.0 )177.6 ± 21.5* 14.1 ± 40.6 )193.1 ± 32.1* )252.5 ± 37.5*
Phase III
Systolic BP (mmHg) 1.7 ± 2.6 0.1 ± 2.3 )3.8 ± 7.3 )5.0 ± 2.8 )21.6 ± 7.3* )34.0 ± 7.2*
Diastolic BP (mmHg) )0.0 ± 1.9 0.3 ± 1.6 2.5 ± 2.7 )3.5 ± 1.2 )6.3 ± 2.8 )11.1 ± 2.6*
Pulse pressure (mmHg) 1.8 ± 1.6 )0.6 ± 1.8 )6.2 ± 5.1 )1.5 ± 2.2 )15.2 ± 5.2* )22.8 ± 5.3*
HR (beats/min) 5.3 ± 3.5 8.3 ± 2.6 17.1 ± 3.3 0.5 ± 3.7 25.5 ± 4.3* 35.3 ± 4.8*
RR interval (ms) )43.4 ± 30.6 )84.2 ± 18.9 )129.0 ± 19.4 5.6 ± 33.6 )207.3 ± 34.2* )263.1 ± 38.6*
Phase IV
Systolic BP (mmHg) 14.9 ± 3.3 18.4 ± 2.6 30.4 ± 5.2* 16.4 ± 2.5 45.3 ± 4.9* 53.0 ± 7.7*
Diastolic BP (mmHg) 7.7 ± 1.5 8.1 ± 1.6 13.0 ± 2.1 6.0 ± 0.9 14.8 ± 2.2* 15.7 ± 3.2*
Pulse pressure (mmHg) 7.2 ± 2.1 10.0 ± 1.5 17.5 ± 3.5* 10.4 ± 1.8 30.5 ± 3.4* 37.2 ± 5.3*
HR (beats/min) )8.8 ± 1.8 )13.2 ± 2.3 )20.1 ± 2.7* )10.3 ± 1.7 )20.3 ± 2.8* )21.9 ± 3.6*
RR interval (ms) 92.9 ± 17.5 161.4 ± 35.8 246.9 ± 33.3* 119.9 ± 18.3 303.3 ± 50.0* 341.4 ± 63.3*

A one-way ANOVA for repeated measures, followed by a post hoc Tukey’s test for multiple comparisons was used to estimate the influence of notes and Valsalva
maneuvers on the cardiovascular parameters. For clarity, the significance of the changes is only indicated (*P < 0.05 or less) for the comparison of medium and high
notes with low notes and for Valsalva 40 and 60 when compared to Valsalva 10

crease in ms) observed during the expiratory strain. pressures or at least additional skills such as better
Figure 4 illustrates the value of the BRS gain as an respiratory perception and ventilatory neuromuscular
indicator of the reflex HR response. The greater the gain control as compared with lay people [3, 14]. Differ-
was, the greater the HR increase (RR interval decrease). ences exist between the wind instrument families. The
airflow rate is generally larger with brass instruments
(like the tenor and bass tubas) than with the wood-
Discussion winds (like oboe). Nevertheless, the intrathoracic
pressure of the oboe player may reach 50 mmHg with
The main result of the study is that the cardiovascular high notes [3]. The range of mouth pressure of brass
effects of playing long mf high notes with a tuba instrument is much larger than that of woodwinds,
resemble a Valsalva maneuver. It is known among starting from low levels of 10 mmHg and reaching
wind players that playing high loud notes for more high levels, especially with trombones and trumpets.
than a few seconds may cause dizziness or occasion- The highest mouth pressure associated with a loud
ally black out. high note is around 60–70 mmHg for the tuba. In the
There are reports in trumpet and French horn present study, the Valsalva maneuver was applied at
players documenting the cardiovascular changes 10, 40, and 60 mmHg to reproduce this usual range,
resulting from blowing long notes or phrases using a and the four classic phases of the Valsalva maneuver
single breath [2, 9, 12, 17, 22]. [16] were indeed present in the recordings obtained
The analogy between the respiratory strain of a while the musicians were playing.
Valsalva maneuver and wind instrument practice The BP elevation at the onset of straining (phase I)
highlights the importance of breathing in wind- reflects the mechanical effect of the increased intra-
instrument players. Wind instrumentalists may thoracic pressure. A quantity of blood is expelled
exhibit high vital capacities and maximum respiratory from the thoracic vessels by the increase in intra-
 101

Fig. 3 Up Linear relationship between changes in Notes Valsalva

pulse pressure and stroke volume resulting from the 40
three notes or from the three Valsalva maneuvers
obtained in the ten tuba players with r = 0.95, 0
P < 0.001 for the notes and r = 0.96, P < 0.001 for Pulse
the Valsalva maneuvers. Middle Linear relationship Pressure −40
between changes in heart rate (HR) and stroke % changes
volume changes resulting from the three notes or −80
from the three Valsalva maneuvers obtained in the
ten tuba players with r = 0.74, P < 0.001 for the −120
notes and r = 0.65, P < 0.001 for the Valsalva 120
maneuvers. Low Polynomial relationship between
changes in peripheral resistance and stroke volume 80
changes resulting from the three notes or from the Heart Rate
40
three Valsalva maneuvers obtained in the ten tuba % changes
players with r = 0.98, P < 0.001 for the notes and 0
r = 0.92, P < 0.001 for the Valsalva maneuvers
−40
500
400
Total 300
Peripheral
200
Resistance
% changes 100
0
−100
−100 −80 −60 −40 −20 0 20 −100 −80 −60 −40 −20 0 20
Stroke Volume % changes Stroke Volume % changes

Fig. 4 Linear relationship between changes in heart Notes Valsalva

rate (HR) and spontaneous baroreflex sensitivity in 100
the ten tuba players. Changes in HR observed 80
following each of the three notes and each of the
three Valsalva maneuvers were linearly related to 60 low note Valsalva 10
spontaneous baroreflex sensitivity with r = 0.64, 0.85 40
and 0.69 for the low, middle and high notes and 0.66, 20
0.70 and 0.69 for the 10, 40 and 60 Valsalva
maneuvers, P < 0.05 or less 0
Maximum Heart Rate changes in %

–20
100
80
middle note Valsalva 40
60
40
20
0
–20
100
80 high note
60
40
20
Valsalva 60
0
–20
0 5 10 15 0 5 10 15
Baroreflex sensitivity ms/mmHg Baroreflex sensitivity ms/mmHg

thoracic pressure and the aorta is compressed. This consideration the differences in ventilatory strain
rise was greater with the Valsalva 40 and 60, although between the note emission and the Valsalva maneu-
the reflex HR decrease was similar with the two ver, since blowing a note means an airflow, while the
maneuvers (notes and Valsalva). One may take into Valsalva maneuver means no airflow.
102

The expiratory strain corresponds to phase II. relaxation, at the release of the strain would be
During the early phase II (IIa), the impeded venous expected.
return [21, 26], by lowering the preload, causes a BP During Phase IV, deep breathing and continuation
decrease, with a predominant fall in systolic BP of the venous return of blood which has been dam-
resulting in a reduced pulse pressure. After a few med back markedly increase the stroke volume. The
seconds, arterial baroreflexes compensate BP, in the increased cardiac output results from the increased
late phase IIb. The BP decreases associated with the stroke volume and from the increased cardiac sym-
two maneuvers (notes, Valsalva) reached 24 mmHg pathetic tone initiated during the strain. The
for the highest note and the Valsalva 40 and 60. The remaining constriction of the arteriolar bed, i.e., the
reduced pulse pressure (predominant fall in systolic rise in total peripheral resistance (increased vascular
BP) observed with the high note and the Valsalva 40 sympathetic tone which initiated during the strain)
and 60 were causally related to the reduction in stroke together with the rise in cardiac output result in a BP
volume [24]. This supports the view that mechanical overshoot and, subsequently, in a reflex vagally
circulatory effects are responsible for the hypotension mediated bradycardia. Large rises in BP occurred in
observed on Valsalva-type straining [26]. the present study, the highest being observed with the
During phase II, reflex HR increases and rises in Valsalva 40 and 60, together with marked reflex HR
peripheral resistance were observed. BP was restored decreases.
during late phase II (IIb), although stroke volume Altogether, blowing long mf high notes on a tuba
remained low. The cardiac and the vascular responses produced cardiovascular changes close to a classic
did not follow the same pattern: the HR increase (40 mmHg) Valsalva maneuver. Long mf low notes
(cardiac sympathetic activation and possibly vagal evoked changes closer to a Valsalva 10. In this study,
withdrawal) was linearly related to the stroke volume the note pitch and the dynamics were adjusted to fit
diminution. By contrast, the peripheral resistance the usual practice of the instrument. The posture for
exhibited steep increases when the stroke volume playing a wind instrument is the seated or standing
decreased more than 40%, indicating that for marked position. The effect of the posture on the cardiovas-
reductions in stroke volume, the vasoconstrictor cular response to a 40 mmHg Valsalva maneuver has
response was predominant and explained almost all been specifically addressed [26, 33]. In the upright
the BP restoration. This is another demonstration that position, straining may reduce cerebral blood velocity
the intensity of straining, translated into stroke vol- further to ~50% of supine rest, which is close to the
ume reductions, conditions the vascular sympathetic critical level of cerebral perfusion where healthy
activation [32]. A pharmacological approach also subjects loose consciousness [35]. Another cause of
demonstrated the crucial role of the reflex vascular dizziness exists specifically in (bass) tuba players,
sympathetic activation during phase II, showing that a who play long phrases with low notes, which require
Valsalva maneuver determined profound hypoten- little mouth pressure but very high flow rates, and
sion, reaching 40 mmHg, in control subjects after who hyperventilate during such periods.
alpha-adrenoceptor blockade [20, 27]. Besides dizziness, the practice of wind instruments
Interestingly, the maximal HR changes observed may also induce episodic headache, blurred vision, or
during phase II were linearly related to the sponta- visual defects [28, 30]. Headache may be a conse-
neous BRS (LF gain). Previous studies used the HR quence of increased intracranial pressure caused by
decrease resulting from the BP overshoot occurring in decreased cerebral venous drainage [11]. The com-
phase IV to estimate the BRS but this index was bined effects of an increased intraocular pressure (rise
shown to be poorly indicative of the spontaneous BRS in venous pressure) and a reduced ocular perfusion
[36]. This relationship between spontaneous BRS pressure (BP fall) may explain the commonly reported
derived from a resting period preceding the maneuver transient blurred vision, during a long musical phrase
and the HR rises during phase II substantiates the (phase II). The magnitude of intraocular pressure
reflex nature of the HR response. increase is greater in high resistance (oboe, bassoon,
Phase III was brief and corresponded to the French horn, trumpet, tenor trombone) wind players
immediate release of the strain. As intrathoracic versus low resistance (flute, clarinet, saxophone, bass
pressure suddenly drops, there is a pooling of blood trombone, and tuba) wind players [30].
in the pulmonary vessels, causing a further fall in BP. Another source of problem for the wind instru-
This fall was marked with Valsalva 40 and 60 and this mentalist may potentially occur with the BP over-
may well mirror the phase I changes. If markedly shoot, which may reach BP levels high enough to
contracted lungs are propelling large amounts of cause headache. This BP level is considered to be
blood into the arterial circulation, at the initiation of lower than the BP reached while coughing, but some
the expiratory strain of the Valsalva maneuver, a rare incidents (such as cervical artery dissection and
larger passive abrupt filling during lung parenchymal stroke) have been related to marked BP rises, usually
 103

related to the practice of high resistance wind treatment [27]. This BP overshoot is due to a rapid
instruments [11]. Valsalva retinopathy (retinal hem- increase in stroke volume, as described above, and a
orrhage) has not been reported, to the best of our lingering cardiac beta-adrenoceptor activation. Also,
knowledge. beta-blockers do not accentuate the BP decreases
Many factors such as preexisting hypertension or during the early phase of the Valsalva maneuver [27].
age may affect the vascular and HR responses to the This strengthens the view that the most efficient
Valsalva maneuver [8, 20, 31]. It is evident that an compensatory adjustment to the stroke volume
autonomic neuropathy or drugs affecting the auto- reduction is the vascular sympathetic activation,
nomic nervous system function may alter the re- confirmed by microneurography [7] and catechol-
sponses to a Valsalva maneuver and supposedly to amine elevations [13].
blowing notes on a wind instrument. Alpha-adreno- More interesting for the musician is the search for
ceptors should be logically avoided in wind instru- a better note quality, based on optimizing the expi-
ment players since these drugs are known to ratory strain. Repeated respiratory exercises and
accentuate the BP decreases during the Valsalva instrument playing contribute to improving lung
maneuver [21, 27]. These data support the concept functions [3, 14]. Subjects who exhibit attenuated BRS
that the reflex rises in total peripheral resistance are have more frequent episodes of intolerance when
counteracting the fall in cardiac output. The effects of exposed to orthostatic challenges [5]. Convertino
alpha-adrenoceptor blockade on the BP overshoot are et al. [5] demonstrated that periodic elevations of BP
more complex and differ according to the position of resulting from repeated respiratory straining maneu-
the subject (supine or seated) [10, 27]. In contrast, vers, similar to those used by high-performance air-
beta-adrenoceptor blockers may not represent a car- craft pilots, resulted in baroreflex sensitization. By
diovascular contra-indication to wind instrument extrapolating to wind instrument practice, one could
practice. Beta-adrenoceptor blockade may improve recommend to subjects with a low BRS to repeatedly
the symptoms resulting from the Valsalva maneuver, blow loud high notes for short periods of time, in
especially the BP overshoot which is reduced after order to improve their BP tolerance to long notes.

References
1. Bernardi L, Porta C, Sleight P (2006) 6. de Jong-de Vos van Steenwijk CC, 13. Fernandez PG, Snedden W (1991)
Cardiovascular, cerebrovascular, and Imholz BP, Wesseling KH, Wieling W Converting enzyme inhibition attenu-
respiratory changes induced by differ- (1997) The Valsalva manoeuvre as a ates the noradrenaline response to the
ent types of music in musicians and cardiovascular reflex test in healthy Valsalva maneuvre. Can J Cardiol
non-musicians: the importance of si- children and teenagers. Clin Auton Res 7:214–222
lence. Heart 92:445–452 7:167–171 14. Fiz JA, Aguilar J, Carreras A, Teixido A,
2. Borgia JF, Horvath SM, Dunn FR, von 7. Delius W, Hagbarth KE, Hongell A, Haro M, Rodenstein DO, Morera J
Phul PV, Nizet PM (1975) Some phys- Wallin BG (1972) Manoeuvres affecting (1993) Maximum respiratory pressures
iological observations on French horn sympathetic outflow in human muscle in trumpet players. Chest 104:1203–
musicians. J Occup Med 17:696–701 nerves. Acta Physiol Scand 84:82–94 1204
3. Bouhuys A (1964) Lung volumes and 8. Denq JC, O’Brien PC, Low PA (1998) 15. Gilbert J, Ponthus S, Petiot JF (1998)
breathing patterns in wind-instrument Normative data on phases of the Val- Artificial buzzing lips and brass
players. J Appl Physiol 19:967–975 salva maneuver. J Clin Neurophysiol instruments: experimental results. J
4. Constant I, Laude D, Elghozi JL, Murat 15:535–540 Acoust Soc Am 104:1627–1632
I (1999) Assessment of short-term 9. Dimsdale JE, Nelesen RA (1995) 16. Hamilton WF, Woodbury RA, Harper
blood pressure variability in anesthe- French-horn hypertension. N Engl J HT (1936) Physiologic relationships
tized children: a comparative study Med 333:326–327 between intrathoracic, intraspinal and
between intraarterial and finger blood 10. Elisberg EI, Miller G, Weinberg SL, arterial pressures. JAMA 107:853–856
pressure. J Clin Monit Comput 15:205– Katz LN (1953) The effect of the Val- 17. Harris LR (1996) Horn playing and
214 salva maneuver on the circulation. II. blood pressure. Lancet 348:1042
5. Convertino VA, Ratliff DA, Doerr DF, The role of the autonomic nervous 18. Imholz BP, van Montfrans GA, Settels
Ludwig DA, Muniz GW, Benedetti E, system in the production of the over- JJ, van der Hoeven GM, Karemaker JM,
Chavarria J, Koreen S, Nguyen C, Wang shoot. Am Heart J 45:227–236 Wieling W (1988) Continuous non-
J (2003) Effects of repeated Valsalva 11. Evers S, Altenmuller E, Ringelstein EB invasive blood pressure monitoring:
maneuver straining on cardiac and (2000) Cerebrovascular ischemic events reliability of Finapres device during the
vasoconstrictive baroreflex responses. in wind instrument players. Neurology Valsalva manoeuvre. Cardiovasc Res
Aviat Space Environ Med 74:212–219 55:865–867 22:390–397
12. Faulkner M, Sharpey-Schafer EP (1959)
Circulatory effects of trumpet playing.
Br Med J 1:685–686
104

19. Jellema WT, Imholz BP, Oosting H, 24. Leonetti P, Audat F, Girard A, Laude D, 31. Shimada K, Kitazumi T, Ogura H,
Wesseling KH, van Lieshout JJ (1999) Lefrere F, Elghozi JL (2004) Stroke Sadakane N, Ozawa T (1986) Effects of
Estimation of beat-to-beat changes in volume monitored by modeling flow age and blood pressure on the cardio-
stroke volume from arterial pressure: a from finger arterial pressure waves vascular responses to the Valsalva
comparison of two pressure wave mirrors blood volume withdrawn by maneuver. J Am Geriatr Soc 34:431–434
analysis techniques during head-up tilt phlebotomy. Clin Auton Res 14:176– 32. Smith ML, Beightol LA, Fritsch-Yelle
testing in young, healthy men. Clin 181 JM, Ellenbogen KA, Porter TR, Eckberg
Auton Res 9:185–192 25. Looga R (2005) The Valsalva manoeuvre– DL (1996) Valsalva’s maneuver revis-
20. Korner PI, Blombery PA, Bobik A, cardiovascular effects and performance ited: a quantitative method yielding
Tonkin AM, Uther JB (1976) Valsalva technique: a critical review. Respir Physiol insights into human autonomic con-
vasoconstrictor reflex in human Neurobiol 147:39–49 trol. Am J Physiol 271:H1240–1249
hypertension and after beta-adrenore- 26. Pott F, van Lieshout JJ, Ide K, Madsen 33. ten Harkel AD, van Lieshout JJ, van
ceptor blockade in conscious rabbits. P, Secher NH (2000) Middle cerebral Lieshout EJ, Wieling W (1990) Assess-
Clin Sci Mol Med 3(suppl):365s–368s artery blood velocity during a Valsalva ment of cardiovascular reflexes: influ-
21. Korner PI, Tonkin AM, Uther JB (1979) maneuver in the standing position. J ence of posture and period of
Valsalva constrictor and heart rate re- Appl Physiol 88:1545–1550 preceding rest. J Appl Physiol 68:147–
flexes in subjects with essential hyper- 27. Sandroni P, Novak V, Opfer-Gehrking 153
tension and with normal blood TL, Huck CA, Low PA (2000) Mecha- 34. van Lieshout JJ, Imholz BP, Wesseling
pressure. Clin Exp Pharmacol Physiol nisms of blood pressure alterations in KH, Speelman JD, Wieling W (1991)
6:97–110 response to the Valsalva maneuver in Singing-induced hypotension: a com-
22. Larger E, Ledoux S (1996) Cardiovas- postural tachycardia syndrome. Clin plication of a high spinal cord lesion.
cular effects of French horn playing. Auton Res 10:1–5 Neth J Med 38:75–79
Lancet 348:1528 28. Sbeity ZH, Mansour AM (2004) 35. van Lieshout JJ, Wieling W, Karemaker
23. Laude D, Elghozi JL, Girard A, Bellard E, Recurrent retinal vein occlusion after JM, Secher NH (2003) Syncope, cere-
Bouhaddi M, Castiglioni P, Cerutti C, playing a wind instrument. Graefes bral perfusion, and oxygenation. J Appl
Cividjian A, Di Rienzo M, Fortrat JO, Arch Clin Exp Ophthalmol 242:428–431 Physiol 94:833–848
Janssen B, Karemaker JM, Leftheriotis 29. Schorr-Lesnick B, Teirstein AS, Brown 36. Zollei E, Paprika D, Rudas L (2003)
G, Parati G, Persson PB, Porta A, Quintin LK, Miller A (1985) Pulmonary func- Measures of cardiovascular autonomic
L, Regnard J, Rudiger H, Stauss HM tion in singers and wind-instrument regulation derived from spontaneous
(2004) Comparison of various tech- players. Chest 88:201–205 methods and the Valsalva maneuver.
niques used to estimate spontaneous 30. Schuman JS, Massicotte EC, Connolly Auton Neurosci Basic Clin 103:100–105
baroreflex sensitivity (the EuroBaVar S, Hertzmark E, Mukherji B, Kunen MZ
study). Am J Physiol 286:R226–231 (2000) Increased intraocular pressure
and visual field defects in high resis-
tance wind instrument players. Oph-
thalmol 107:127–133