J Appl Physiol 107: 1840–1846, 2009;
doi:10.1152/japplphysiol.00334.2009.
Involuntary breathing movements improve cerebral oxygenation during apnea
struggle phase in elite divers
Zeljko Dujic,1 Lovro Uglesic,1 Toni Breskovic,1 Zoran Valic,1 Karsten Heusser,2 Jasna Marinovic,1
Marko Ljubkovic,1 and Ivan Palada1
1
Department of Physiology, University of Split School of Medicine, Split, Croatia; and 2Institute for Clinical Pharmacology,
Hannover Medical School, Hannover, Germany
Submitted 30 March 2009; accepted in final form 15 October 2009
breath hold; cerebral autoregulation; near-infrared spectroscopy; diaphragmatic movements
ELITE BREATH-HOLD DIVERS ARE extreme athletes trained to maintain the prolonged periods of hypoxia/hypercapnia, lasting up
to 10 min (current world record is 10 min 12 s). Breath-hold
diving is associated with a diving response comprised of
bradycardia, decreased cardiac output (CO), peripheral vasoconstriction, and increased arterial blood pressure (AP) (15,
16). These cardiovascular adaptations are caused by the simultaneous activation of sympathetic and parasympathetic nervous
systems (8, 10), with consequent reduction of oxygen in
peripheral tissues and sufficient oxygen supply of the vital
organs (brain and heart). Although the arterial oxygen saturation (SaO2) is reduced with the long breath holds, the compensatory increase in cerebral blood flow (CBF) could potentially
Address for reprint requests and other correspondence: Z. Dujic, Dept. of
Physiology, Univ. of Split School of Medicine, Soltanska 2, 21000 Split,
Croatia (e-mail: [email protected]).
1840
offset the reduced SaO2 and maintain the cerebral tissue
oxygenation. Near-infrared spectroscopy (NIRS) is the method
that continuously and noninvasively monitors the cerebral
oxygenation and cerebral blood volume (CBV). Regional oxygen delivery and consumption in the brain can be assessed by
measuring the changes in oxyhemoglobin (bHbO2) and deoxyhemoglobin (bdHb), whereas the relative changes in CBV can
be assessed by measuring the total hemoglobin (bTHb) in the
absence of major changes in hematocrit (2, 3, 7, 23). Our
laboratory has shown previously, using the transcranial Doppler (TCD) and NIRS techniques, that, despite the large increases in cerebral perfusion in the later phase of the breath
hold, the regional cerebral desaturation may become a factor
limiting the maximal breath-hold duration (18). Therefore, the
cerebral oxygenation is reduced only slightly until the end of
the breath hold at the expense of reduced peripheral blood flow
and oxygenation.
Breath-hold diving research can be performed in the field
(wet) head-out immersion and dry laboratory conditions, with
and without face immersion in cold water. Under all of these
conditions, diving response can be elicited, although the magnitude of observed cardiovascular changes may differ. Since
the methodological approach used in this study was not compatible with the field conditions, we investigated the physiological mechanisms of interest in the model of dry apnea in the
laboratory.
Maximal voluntary apnea is divided into two phases: the
initial or easy-going phase that lasts until the physiological
breaking point when the accumulated carbon dioxide (CO2)
stimulates the respiratory drive (15), and the struggle phase,
during which the subject feels a growing urge to breathe and
shows progressive involuntary breathing movements (IBM)
(5). Recently, our laboratory reported that the IBM are involved in restoration of the venous return by improving the
inferior vena cava blood flow, leading to augmentation of
the stroke volume (SV) and normalization of the CO (17). The
influence of the IBM on cerebral oxygenation and CBV is
presently unknown.
Therefore, in the present study, we tested the hypothesis
that IBM, together with peripheral vasoconstriction and
progressive hypercapnic cerebral vasodilatation, contribute
to centralization of the blood volume, thus maintaining the
cerebral oxygenation during the extended breath holds. To
investigate this hypothesis, the CO, cerebral, muscle, and
arterial oxygenation, as well as blood pressure and IBM,
were measured during the maximal apnea in trained apnea
divers.
8750-7587/09 $8.00 Copyright © 2009 the American Physiological Society
http://www. jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on April 27, 2017
Dujic Z, Uglesic L, Breskovic T, Valic Z, Heusser K, Marinovic J,
Ljubkovic M, Palada I. Involuntary breathing movements improve cerebral
oxygenation during apnea struggle phase in elite divers. J Appl Physiol 107:
1840–1846, 2009; doi:10.1152/japplphysiol.00334.2009.—We investigated whether the involuntary breathing movements (IBM) during the
struggle phase of breath holding, together with peripheral vasoconstriction and progressive hypercapnia, have a positive effect in maintaining cerebral blood volume. The central hemodynamics, arterial
oxygen saturation, brain regional oxyhemoglobin (bHbO2), deoxyhemoglobin, and total hemoglobin changes and IBM were monitored
during maximal dry breath holds in eight elite divers. The frequency
of IBM increased (by ⬃100%), and their duration decreased (⬃30%),
toward the end of the struggle phase, whereas the amplitude was
unchanged (compared with the beginning of the struggle phase). In all
subjects, a consistent increase in brain regional deoxyhemoglobin and
total hemoglobin was also found during struggle phase, whereas
bHbO2 changed biphasically: it initially increased until the middle of
the struggle phase, with the subsequent relative decline at the end of
the breath hold. Mean arterial pressure was elevated during the
struggle phase, although there was no further rise in the peripheral
resistance, suggesting unchanged peripheral vasoconstriction and implying the beneficial influence of the IBM on the cardiac output
recovery (primarily by restoration of the stroke volume). The IBMinduced short-lasting, sudden increases in mean arterial pressure were
followed by similar oscillations in bHbO2. These results suggest that
an increase in the cerebral blood volume observed during the struggle
phase of dry apnea is most likely caused by the IBM at the time of the
hypercapnia-induced cerebral vasodilatation and peripheral vasoconstriction.
CEREBRAL OXYGENATION DURING APNEA STRUGGLE PHASE IN HUMANS
METHODS
Table 1. Demographic and baseline lung function in eight
breath-hold divers
Age, yr
Height, cm
BMI, kg/m2
Body fat index, %body fat/kg
FEV1, %predicted
FVC, %predicted
FEV1/FVC ratio, %predicted
Mean ⫾ SD
Range
28⫾4
184⫾7
25⫾2
18⫾11
115⫾13
131⫾15
90⫾9
22–33
170–190
22–28
5–36
92–131
112–154
74–100
BMI, body mass index; FEV1, forced expiratory volume in first second;
FVC, forced vital capacity. Spirometric values were estimated by Cosmed
software (Quark PFT, Cosmed, Rome, Italy). Body fat index was calculated by
Jackson and Pollock three-site method measurement.
J Appl Physiol • VOL
pair was attached over the participant’s left calf muscle. An interoptode spacing of 4 cm was used to separate the emitting from the
detecting sensor. The optodes were held within a black rubber housing
that maintained constant optode spacing. The housing was attached to
the skin and covered with an optically dense vinyl sheet that was
affixed with tape. This minimized the loss of the near-infrared light
from the recording field, as well as the intrusion of light from the
environment. The full assembly was further secured with an elastic
bandage that encircled the head and lower leg.
Levels of oxygenated hemoglobin, deoxygenated hemoglobin, and
total hemoglobin were continuously sampled by the NIRS unit at a
rate of 6 Hz. These parameters are expressed in micromoles as a
change from zero.
Data acquisition and analysis. Analog signals were sampled at
1,000 Hz and stored on a personal computer using a PowerLab 16S
data acquisition system (ADInstruments, Castle Hill, Australia). The
time points of interest were the periods between 30 and 60 s before the
onset of apnea and last 15 s of an easy-going phase of the breath hold.
During the apnea struggle phase, we averaged three IBM at the start,
in the middle, and at the end. In addition, data were collected during
60 s of postapnea recovery period.
Throughout the struggle phase, duration, amplitude, and frequency
of IBM were analyzed. Duration was calculated as the base width of
the IBM curve (in seconds). The amplitude was derived from the
height of the IBM curve and was expressed as arbitrary units. Average
frequency of the IBM was calculated from the total duration of three
consecutive IBM for the observed time period and extrapolated as
numbers of IBM per minute.
Latencies between the onset of the IBM and the peaks in MAP and
change (⌬) in bTHb tracings were determined at the beginning,
middle, and the end of the struggle phase. Time delays were measured
for three consecutive IBM for each part of the struggle phase and
averaged. The beginning of the IBM was defined as the onset of
deflection in the respiratory belt recording.
Statistics. Using Statistica 7.0 software (Statsoft, Tulsa, OK),
comparisons between changes of variables from the control value
were first tested with nonparametric Friedman analysis of variance. In
case of a significant difference, the Wilcoxon signed rank test was
applied for the particular comparison. Differences between time
delays from onset of IBM until the rise in MAP vs. onset in IBM and
increase in ⌬bTHb were compared using Mann-Whitney U-test. The
level of significance was set at P ⬍ 0.05.
RESULTS
Apnea performed by the eight participants lasted, on average, 240 ⫾ 51 s. The struggle phase, during which the IBM can
be observed, occupied 47.0 ⫾ 14.1% of total apnea duration.
During this phase, on average, 27.1 ⫾ 8.1 IBM per diver were
detected.
Analysis of the IBM properties is shown in Table 2. During
the struggle phase, the IBM duration significantly decreased,
while their frequency progressively increased. The amplitude
of IBM remained unchanged throughout the struggle phase.
Analysis of the individual responses during apnea (Fig. 1)
revealed that the fluctuations caused by the movements of
diaphragm muscle during IBM precede fluctuations of MAP,
HR, and SV. Moreover, fluctuations in brain hemoglobin levels
also had the same pattern as the fluctuations of IBM, but that
was not observed in hemoglobin measured in the calf muscle.
Analysis of time latency between the onset of individual IBM
and fluctuations in MAP and brain hemoglobin levels revealed
that IBM occurs first and increase in MAP is second and is
followed by the change in the level of brain hemoglobin (Fig. 2,
Table 3, individual recordings shown in data supplement; the
107 • DECEMBER 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on April 27, 2017
Subjects. Experimental group consisted of eight breath-hold divers
(6 men and 2 women). At the time of the study, they were all
apparently healthy. The demographic and lung function data of the
study subjects are shown in Table 1.
Experimental procedures. All experimental procedures were performed in accordance with the Declaration of Helsinki on the treatment of human subjects and were approved by the Ethical Committee
of the University of Split School of Medicine. Informed, written
consent was obtained from each subject. All experiments were carried
out in a climatized room in the morning hours. The participants
arrived at the laboratory 30 – 45 min before the start of the experiments for instrumentation and explanation of the procedures. They
had abstained from caffeine for at least 12 h and from food for at least
4 h. In each subject, the dynamic spirometry (Quark PFT, Cosmed,
Rome, Italy) was evaluated in the upright posture. Measurement of
body fat index, height, and weight was performed. After emptying
their bladders, the subjects rested in the supine position for 30 min to
ensure stabilization of the cardiovascular parameters. During testing,
each participant performed one maximal breath hold. The subjects
were instructed not to hyperventilate before the apnea.
AP and heart rate (HR) were assessed using a pneumatic cuff
placed around the middle finger of the nondominant hand (Finometer,
Finapress Medical Systems, Arnhem, the Netherlands). SaO2 was
monitored continuously by pulse oximetry (Poet II, Criticare Systems,
Waukesha, WI), with the probe placed on the middle finger of the
dominant hand.
From the continuous blood pressure measurement, the arterial pulse
wave was analyzed, and changes in left ventricular SV were computed
from the pulsatile systolic area. We employed the improved method of
Wesseling using the Modelflow program (model-based measurement
method based on a nonlinear, three-element model of the input
impedance of the aorta) (13). This methodology detects rapid changes
in SV accurately (compared with inert-gas rebreathing) during leg
crossing, with and without muscle tension (24). CO was computed as
SV times HR, and total peripheral resistance (TPR) was calculated as
the quotient of mean AP (MAP) and CO.
A pneumatic respiratory belt, located around the chest at the level
of the xiphoid process, was coupled to a differential pressure transducer (Prignitz Mikrosystemtechnik, Wittenberge, Germany) and was
used to monitor the IBM. An additional reservoir was connected to the
pneumatic system, which avoided any constriction of the thoracic
cage during chest expansion. This method for IBM assessment was
superior to the approach utilized in our laboratory’s previous study
(17), and it enabled more detailed analysis of the IBM properties.
Local tissue oxygenation levels were recorded from the muscle and
brain using a NIRS unit NIRO-200 (Hamamatsu Photonics KK,
Tokyo, Japan). The first pair of emitting and detecting optodes of the
NIRS unit was placed to the left side of the forehead, while the second
1841
1842
CEREBRAL OXYGENATION DURING APNEA STRUGGLE PHASE IN HUMANS
Table 2. IBM characteristics during the struggle phase
of apnea
IBM duration, s
IBM amplitude, AU
IBM frequency, min⫺1
Beginning
Middle
End
3.9⫾1.8
101.9⫾8.9
10.7⫾3.7
3.6⫾1.0
100.0⫾4.1
13.7⫾4.6*
2.6⫾0.8*
99.7⫾5.7
19.9⫾7.1*
Values are means ⫾ SD for all subjects completing the study. IBM,
involuntary breathing movements; AU, arbitrary units. *P ⬍ 0.05 vs. the
preceding stage of the struggle phase.
DISCUSSION
This study is the first to report that IBM during the struggle
phase of the dry breath hold are followed by the simultaneous
phasic fluctuations of the cerebral oxygenated hemoglobin.
The brain oxygenation fluctuations were also paralleled with
the fluctuations of MAP, HR, and SV, indicating that IBM may
influence the central hemodynamic by maintaining the CO.
This suggests that IBM, together with peripheral vasoconstriction-mediated centralization of the blood volume and progressive hypercapnia-induced cerebral vasodilatation, likely act to
Fig. 1. Individual response of various measured parameters during apnea in diver 1. AP, arterial blood pressure; HR, heart rate; SV, stroke volume; TPR, total
peripheral resistance; bHbO2, change in concentration of oxygenated hemoglobin in brain; bdHb, change in concentration of deoxygenated hemoglobin in brain;
bTHb, change in concentration of total hemoglobin in brain; mHbO2, change in concentration of oxygenated hemoglobin in calf muscle; mdHb, change in
concentration of deoxygenated hemoglobin in calf muscle; mTHb, change in concentration of total hemoglobin in calf muscle; Resp., movements of the chest
wall recorded by respiratory belt.
J Appl Physiol • VOL
107 • DECEMBER 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on April 27, 2017
online version of this article contains supplemental data).
There were no significant changes in the time delays between
different parts of the struggle phase of the apnea.
The magnitudes of responses of hemodynamic parameters
and blood SaO2 are summarized in Table 4. During the breath
hold, the arterial blood saturation was reduced to 81.5 ⫾
12.7%. MAP increased continuously throughout apnea, with
the progressive rise occurring during the struggle phase. The
SV fell down to ⬃64% of its baseline value at the end of the
easy-going phase, but started to normalize during the struggle
phase, and reached ⬃90% of its initial value at the end of
apnea. The HR remained unchanged until the middle of the
struggle phase, when it reduced discretely toward the end of
apnea. These almost opposite trends of changes in the HR and
SV caused the CO, after the initial drop, to remain constant
throughout the apnea and stabilized ⬃70% of the baseline
value. The TPR increased rapidly after the onset of apnea and
remained almost doubled throughout the breath hold.
The levels of oxygenated hemoglobin in the brain showed an
increase in concentration during the major part of apnea (Fig.
1 and Table 5). Decline in bHbO2 concentration can be seen in
the last part of the struggle phase, while the bdHb concentration increased continuously throughout apnea. bTHb concentration also rose during the breath hold, especially in the late
struggle phase, when ⬎60% of the entire bTHb increase
occurred, indicating an increase in CBV. Concentration of
oxygenated hemoglobin in the calf muscle decreased progressively during apnea, in contrast to the similar magnitude of
increase in deoxygenated hemoglobin concentration, resulting
in an unchanged concentration of the total hemoglobin in calf
muscle throughout the apnea (Table 5).
CEREBRAL OXYGENATION DURING APNEA STRUGGLE PHASE IN HUMANS
1843
maintain the cerebral oxygenation throughout the struggle
phase, thus prolonging the maximal apneic time.
The impact of IBM on cerebral oxygenation and blood
volume. In this report, we investigated the influence of IBM on
cerebral oxygenation during the struggle phase of apnea. We
used NIRS to monitor the cerebral tissue oxygenation, which
depends on changes in blood flow, SaO2, cerebral metabolism,
and arterial/venous vascular beds partitioning. In our laboratory’s
previous study, we found that, during the apnea period, skeletal
muscle oxygenation decreased earlier than the cerebral oxygenation (18). This observation was also confirmed in the
present study, supporting the hypothesis of centralization of
blood volume toward the brain and heart under such conditions. Dynamic cerebral autoregulation is a mechanism that
maintains CBF relatively constant, despite the changes in MAP
in the range of 60 –150 mmHg (21). However, the nature of this
mechanism has recently been challenged when the spontaneous
fluctuations of CBF that corresponded to simultaneous changes
in MAP were found (1, 19, 27). In the present study, our data
Table 3. Time delay between onset of IBM and peaks in
MAP and ⌬bTHb for three parts of the struggle phase
Beginning
Middle
End
Time to MAP peak, s
Time to ⌬bTHb peak, s
P
2.3 (1.6–3.0)
2.8 (1.7–3.8)
2.3 (1.6–3.0)
4.1 (3.4–4.8)
5.3 (3.9–6.7)
4.3 (3.4–5.2)
0.009
0.041
0.017
Values are means with 95% confidence intervals in parentheses. MAP, mean
arterial pressure; ⌬bTHb, change in concentration of total hemoglobin in the
brain. P values present the differences between time delays from onset of IBM
until the rise in MAP vs. onset of IBM and increase in ⌬bTHb.
J Appl Physiol • VOL
imply that the IBM-induced sudden oscillations in MAP were
paralleled by the oscillations in brain oxygenation and blood
volume, suggesting incomplete cerebral autoregulation under
these conditions. Our analysis of time relationship between
these parameters revealed that IBM occurs first and is followed
by the increase in MAP and, last, by the change in brain
hemoglobin levels (Fig. 2, Table 3), suggesting a causal relationship between the three parameters.
The most distinct increase in bTHb occurred during the
struggle phase (⬃60% of the total increase detected during
apnea), which is not proportional to the average duration of this
phase (⬃45% of total apnea time). Since the IBM are the
hallmark of the struggle phase, it can be inferred that they are
responsible for the additional increase in CBV observed during
this period. In addition, a significant increase in bHbO2 level
was observed until the middle of the struggle phase, and it may
help to prolong the total apnea time. Therefore, the IBM
impose as the most likely mechanism that, in addition to
redistribution of the blood to the central nervous system caused
by the massive sympathetic activation (12) and CO2-induced
cerebral vasodilation, works to raise the CBV and to maintain
the cerebral oxygenation during the struggle phase of apnea.
Afterwards, close to the end of the breath hold, an evident drop
in bHbO2 occurs, possibly due to the prevailing regional
cerebral desaturation, despite the compensatory mechanisms.
The cerebral oxygen desaturation, together with intense hypercapnia, finally determines the end of the breath hold.
IBM characteristics and breath-hold phases. This is the first
study that analyzed the characteristics of IBM in elite breathhold divers during the maximal dry attempts, lasting around 4
107 • DECEMBER 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on April 27, 2017
Fig. 2. Individual response of various measured parameters during struggle phase of apnea in diver 1. MAP, mean arterial pressure; ⌬, change. Vertical lines
represent onset of the involuntary breathing movements (IBM) (beginning of downward deflection). Arrows point at the peaks in MAP, ⌬bHbO2, and ⌬bTHb
tracings, which follow certain IBM (labeled by vertical lines). Numbers mark consecutive IBMs and respective oscillations in MAP, ⌬bHbO2, and ⌬bTHb.
1844
CEREBRAL OXYGENATION DURING APNEA STRUGGLE PHASE IN HUMANS
Table 4. Hemodynamic parameters and arterial blood oxygen saturation before, during, and following apnea attempts
SaO2, %
MAP, mmHg
HR, beats/min
SV, ml
CO, l ⫻ min⫺1
TPR, AU
Baseline
End of Easy-Going Phase
Beginning of the Struggle Phase
Middle of the Struggle Phase
End of Apnea
Recovery
98.4⫾0.9
100.6⫾22.4
75.1⫾7.1
97.4⫾22.1
7.3⫾1.9
14.9⫾6.0
97.6⫾2.1*
117.3⫾41.3*
79.7⫾14.0
62.7⫾22.9*
5.0⫾2.0*
29.4⫾24.0*
97.0⫾2.4*
121.8⫾41.7*†
74.6⫾9.6†
71.9⫾23.4*
5.4⫾2.0*
28.6⫾23.7*
93.4⫾3.8*†
134.3⫾53.5*†
70.0⫾11.5†
80.1⫾21.4*
5.7⫾1.9*
29.3⫾25.0*
81.5⫾12.7*†
140.9⫾54.7*†
65.3⫾13.6†
87.2⫾32.7†
5.6⫾2.0*
31.4⫾26.1*
97.8⫾1.7
106.4⫾28.8*
78.4⫾7.4
111.7⫾23.3*
8.8⫾2.1*
13.0⫾5.9*
Values are means ⫾ SD for all subjects completing the study. SaO2, arterial oxygen saturation; HR, heart rate; SV, stroke volume; CO, cardiac output; TPR,
total peripheral resistance. *Differences are reported between baseline values vs. intra-apneic changes and recovery (P ⬍ 0.05). †Differences between the end
of an easy-going phase and various parts of struggle phase (P ⬍ 0.05).
activity during the breath holding in our laboratory’s recent
study (12) revealed a fivefold increase in muscle sympathetic
neural activity during maximal dry apnea in elite divers.
However, in the later stage of the breath holding, this increase
in sympathetic activity was not associated with a concomitant
increase in TPR. Indeed, in the present study, we found no
further increase in TPR during the struggle phase. It is very
likely that the sympathetically mediated vasoconstriction has
been opposed by the direct vasodilator effect of hypoxia (14)
and/or hypercapnia (22). Therefore, the gradual rise in the AP,
which was, after the initial drop, observed throughout the
apnea, is probably due to the partial restoration of the CO
[possibly via IBM-induced venous return (17)] that is backed
by the sympathetically maintained peripheral resistance. This
mechanism is further supported by the observation that fluctuations in the AP appeared soon after the onset of IBM and
stopped after the end of apnea. The HR can be either unchanged during dry breath holds or reduced as a part of the
diving response. Here, we observed a reduction in the HR,
which also contributed to diminished CO. This suggests that
either parasympathetic input to the heart prevails over the
sympathetic input under this condition, or that sympathetic
efferent nerve activity can be differently expressed in different
tissues.
Methodological considerations. Introduction of an additional group of divers with respiratory muscle paralysis would
further confirm the causal relationship between the IBM and
improved brain perfusion. However, due to ethical issues and
technical difficulties, we were unable to perform such a procedure. Additionally, apnea attained under those circumstances
would likely elicit different physiological responses than the
“normal” apnea.
Table 5. Changes of tissue hemoglobin levels measured by near-infrared spectroscopy
Baseline
End of Easy-Going Phase
Beginning of the Struggle Phase
Middle of the Struggle Phase
End of Apnea
Recovery
7.2⫾5.0*
7.6⫾4.0*†
14.8⫾8.0*†
3.9⫾6.3
16.5⫾12.2*†
20.4⫾10.5*†
11.2⫾5.8*
2.8⫾2.1*
13.3⫾7.0*
Head
⌬HbO2
⌬dHb
⌬THb
0
0
0
3.7⫾3.4*
3.7⫾2.3*
7.4⫾4.7*
4.4⫾4.2*
4.2⫾2.2*†
8.6⫾5.2*
Leg
⌬HbO2
⌬dHb
⌬THb
0
0
0
⫺5.5⫾4.1*
5.4⫾4.2*
⫺0.1⫾4.8
⫺6.1⫾4.3*
6.0⫾4.9*
0.0⫾5.2
⫺6.5⫾5.2*
6.4⫾5.0*
⫺0.1⫾5.0
⫺8.6⫾6.1*
8.2⫾5.7*
⫺0.4⫾5.2
0.2⫾4.8
0.7⫾2.4
0.9⫾4.6
Values are means ⫾ SD in ␮M for all subjects completing the study. ⌬HbO2, change in concentration of oxygenated hemoglobin; ⌬dHb, change in
concentration of deoxygenated hemoglobin; ⌬THb, change in concentration of total hemoglobin. *Differences are reported between baseline values vs.
intraapneic changes and recovery (P ⬍ 0.05). †Differences between the end of an easy-going phase and various parts of struggle phase (P ⬍ 0.05).
J Appl Physiol • VOL
107 • DECEMBER 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on April 27, 2017
min. The respiratory contractions were also analyzed previously in untrained or moderately trained subjects during breath
holding that lasted only ⬃1 min (26). The authors reported that
diaphragmatic electromyography showed little or no muscle
activity during the initial easy-oing phase of the breath hold,
but IBM became more frequent and intense in the latter,
struggle, phase until the breath-old breaking point (26). In the
present study, a similar trend was found, with an increase in the
number of IBM from 10.4/min in the first part of the struggle
phase to 19.5/min in the last part and with reduction in their
duration for ⬃32%. However, the average IBM activity
throughout the struggle phase was unchanged (assessed from
the area under curve). Previously, the IBM were related only to
the sense of discomfort felt during the longer breath holding
and were thought to be elicited mostly through stimulation of
the central chemoreceptors and respiratory neurons by elevated
arterial blood CO2 and O2 levels, thus evoking the contractions
of inspiratory muscles (11). However, later studies revealed
that breath holding per se produces additional stimulation of
central respiratory neurons by an unknown mechanism, which
acts to increase the amplitude of the IBM (26).
In the present study, large interindividual differences in
duration of the struggle phase were noted (ranged from 28 to
68% of the total breath-hold duration), but this did not affect
the maximal duration of the breath hold. Different proportions
of struggle phase in total apnea time may be explained by an
interindividual variability in achieving the long breath holds.
IBM-induced cardiovascular changes. As anticipated, the
cardiovascular changes observed in the present study were
typical for dry breath holds (Fig. 1). Initially, AP dropped due
to an increased intrathoracic pressure with subsequent reduction in venous return, SV, and CO, as shown previously by
others (9, 20). Measurements of the muscle sympathetic neural
CEREBRAL OXYGENATION DURING APNEA STRUGGLE PHASE IN HUMANS
ACKNOWLEDGMENTS
The authors thank the apnea divers for enthusiastic participation in the
study.
GRANTS
This study was supported by Croatian Ministry of Science, Education and
Sports Grant No. 216-2160133-0130.
J Appl Physiol • VOL
DISCLOSURES
No conflicts of interest are declared by the author(s).
REFERENCES
1. Ainslie PN, Barach A, Murrell C, Hamlin M, Hellemans J, Ogoh S.
Alterations in cerebral autoregulation and cerebral blood flow velocity
during acute hypoxia: rest and exercise. Am J Physiol Heart Circ Physiol
292: H976 –H983, 2007.
2. Brazy JE. Near-infrared spectroscopy. Clin Perinatol 18: 519 –534,
1991.
3. Brazy JE, Lewis DV, Mitnick MH, Jobsis-Vander Vliet FF. Monitoring
of cerebral oxygenation in the intensive care nursery. Adv Exp Med Biol
191: 843– 848, 1985.
4. Bucher HU, Edwards AD, Lipp AE, Duc G. Comparison between
near infrared spectroscopy and 133 Xenon clearance for estimation of
cerebral blood flow in critically ill preterm infants. Pediatr Res 33:
56 – 60, 1993.
5. Dejours P. Hazards of hypoxia during diving. In: Physiology of breathhold diving and the Ama of Japan papers, edited by Rahn H. Washington,
DC: National Academy of Sciences-National Research Council, 1965, p.
183–193.
6. Dujic Z, Ivancev V, Heusser K, Dzamonja G, Palada I, Valic Z, Tank
J, Obad A, Bakovic D, Diedrich A, Joyner MJ, Jordan J. Central
chemoreflex sensitivity and sympathetic neural outflow in elite breath-hold
divers. J Appl Physiol 104: 205–211, 2008.
7. Elwell CE, Cope M, Edwards AD, Wyatt JS, Delpy DT, Reynolds EO.
Quantification of adult cerebral hemodynamics by near-infrared spectroscopy. J Appl Physiol 77: 2753–2760, 1994.
8. Fagius J, Sundlof G. The diving response in man: effects on sympathetic activity in muscle and skin nerve fascicles. J Physiol 377:
429 – 443, 1986.
9. Ferrigno M, Hickey DD, Liner MH, Lundgren CE. Cardiac performance in humans during breath holding. J Appl Physiol 60: 1871–1877,
1986.
10. Finley JP, Bonet JF, Waxman MB. Autonomic pathways responsible for
bradycardia on facial immersion. J Appl Physiol 47: 1218 –1222, 1979.
11. Godfrey S, Campbell EJ. The control of breath holding. Respir Physiol
5: 385– 400, 1968.
12. Heusser K, Dzamonja G, Tank J, Palada I, Valic Z, Bakovic D, Obad
A, Ivancev V, Breskovic T, Diedrich A, Joyner MJ, Luft FC, Jordan
J, Dujic Z. Cardiovascular regulation during apnea in elite divers. Hypertension 53: 719 –724, 2009.
13. Jellema WT, Wesseling KH, Groeneveld AB, Stoutenbeek CP, Thijs
LG, van Lieshout JJ. Continuous cardiac output in septic shock by
simulating a model of the aortic input impedance: a comparison with bolus
injection thermodilution. Anesthesiology 90: 1317–1328, 1999.
14. Lahana A, Costantopoulos S, Nakos G. The local component of the
acute cardiovascular response to simulated apneas in brain-dead humans.
Chest 128: 634 – 639, 2005.
15. Lin YC. Breath-hold diving in terrestrial mammals. Exerc Sport Sci Rev
10: 270 –307, 1982.
16. Liner MH. Tissue gas stores of the body and head-out immersion in
humans. J Appl Physiol 75: 1285–1293, 1993.
17. Palada I, Bakovic D, Valic Z, Obad A, Ivancev V, Eterovic D,
Shoemaker JK, Dujic Z. Restoration of hemodynamics in apnea struggle
phase in association with involuntary breathing movements. Respir
Physiol Neurobiol 161: 174 –181, 2008.
18. Palada I, Obad A, Bakovic D, Valic Z, Ivancev V, Dujic Z. Cerebral
and peripheral hemodynamics and oxygenation during maximal dry
breath-holds. Respir Physiol Neurobiol 157: 374 –381, 2007.
19. Panerai RB, Dawson SL, Eames PJ, Potter JF. Cerebral blood flow
velocity response to induced and spontaneous sudden changes in arterial
blood pressure. Am J Physiol Heart Circ Physiol 280: H2162–H2174,
2001.
20. Paulev PE, Honda Y, Sakakibara Y, Morikawa T, Tanaka Y, Nakamura W. Brady- and tachycardia in light of the Valsalva and the Mueller
maneuver (apnea). Jpn J Physiol 38: 507–517, 1988.
21. Paulson OB, Strandgaard S, Edvinsson L. Cerebral autoregulation.
Cerebrovasc Brain Metab Rev 2: 161–192, 1990.
22. Simmons GH, Minson CT, Cracowski JL, Halliwill JR. Systemic
hypoxia causes cutaneous vasodilation in healthy humans. J Appl Physiol
103: 608 – 615, 2007.
107 • DECEMBER 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on April 27, 2017
One limitation that most techniques that monitor the
cardiovascular parameters, such as the CO, have in common
is that they do not provide the continuous assessment of
cardiovascular changes in a beat-to-beat manner (for example, dye or thermodilution). The continuous monitoring of
the SV can be performed noninvasively by impedance
cardiography, ultrasound, and by pulse-wave analysis. However, with pulmonary hyperinflation occurring after maximal inspiration, the use of ultrasound is complicated by
locating the appropriate recording window, while the motion artifacts remain a problem in impedance cardiography.
Therefore, we used the Modelflow method that is based on
computation of an aortic flow waveform from finger by
simulating the nonlinear three-element model of the aortic
input impedance (13). However, even this method is somewhat limited, since the model assumes a normal aortic valve
and unaffected transmural aortic pressure that may change
by the maximal pulmonary hyperinflation. During maximal
apneas, pulmonary volume is slightly below the total lung
capacity, but it remains nearly unchanged throughout the
breath hold. Nevertheless, van Lieshout et al. (25) have
shown very reproducible continuous measurement of SV by
the noninvasive Modelflow method and ultrasound in patients with instantaneous fluctuations in SV during arrhythmias. Our laboratory has used this method for assessment of
the relative SV changes in several recent studies (6, 17, 18).
In our previous report investigating IBM, we have used the
respiratory belt that enabled a reliable detection of the respiratory movements, but a detailed analysis of individual IBM in
terms of duration, amplitude, and area under curve was not
possible (17). This was significantly improved in the present
study with the use of the pneumatic respiratory belt that
allowed for such analyses.
Two methods that allow for the continuous monitoring of the
CBF are the TCD and NIRS. Previously, we have described
changes in TCD (determined by the middle cerebral artery
blood velocity) and NIRS during the breath hold (18). In this
study, we chose to measure relative changes in the CBV by
NIRS, since it has been demonstrated that changes in cerebral
oxygenation assessed by NIRS correlate well with the changes
in CBF (4, 7, 23).
In conclusion, the present study is the first to report that
IBM may increase cerebral oxygenation during the struggle
phase of the dry breath hold. It is a result of the beneficial
impact of IBM on central hemodynamics via partial restoration of the SV and CO. This effect of IBM on cerebral
oxygenation may act, concomitantly with the peripheral
vasoconstriction-mediated centralization of the blood volume and progressive hypercapnia-induced cerebral vasodilatation, as a cofactor in prolonging maximal apnea time in
these individuals exposed to extreme hypoxia/hypercapnia.
Further studies are required to differentiate the relative
contribution of each of these factors.
1845
1846
CEREBRAL OXYGENATION DURING APNEA STRUGGLE PHASE IN HUMANS
23. Terborg C, Gora F, Weiller C, Rother J. Reduced vasomotor reactivity
in cerebral microangiopathy: a study with near-infrared spectroscopy and
transcranial Doppler sonography. Stroke 31: 924 –929, 2000.
24. van Dijk N, de Bruin IG, Gisolf J, de Bruin-Bon HA, Linzer M, van
Lieshout JJ, Wieling W. Hemodynamic effects of leg crossing and
skeletal muscle tensing during free standing in patients with vasovagal
syncope. J Appl Physiol 98: 584 –590, 2005.
25. van Lieshout JJ, Toska K, van Lieshout EJ, Eriksen M, Walloe L,
Wesseling KH. Beat-to-beat noninvasive stroke volume from arterial
pressure and Doppler ultrasound. Eur J Appl Physiol 90: 131–137,
2003.
26. Whitelaw WA, McBride B, Amar J, Corbet K. Respiratory neuromuscular output during breath holding. J Appl Physiol 50: 435– 443,
1981.
27. Zhang R, Zuckerman JH, Levine BD. Spontaneous fluctuations
in cerebral blood flow: insights from extended-duration recordings
in humans. Am J Physiol Heart Circ Physiol 278: H1848 –H1855,
2000.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on April 27, 2017
J Appl Physiol • VOL
107 • DECEMBER 2009 •
www.jap.org