Effects of Valsalva on breath-holding D. BARTLETT, JR. Department of Physiology,
and Mueller time Dartmouth
Medical
maneuvers
School, Hanover,
New Hampshire
03755
BARTLETT, D., JR. Effects of Valsalva and Mueller maneuvers on breath-holding time. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. U(5): 717-721, 1977. -Breath holding to the breaking point was studied at FRC in six healthy subjects in the sitting position. Breath-holding time increased with successive trials within experimental sessions in all subjects. To study the influence of Valsalva and Mueller maneuvers on breath-holding performance, sustained inspiratory or expiratory effort against an occluded mouthpiece was initiated 5 s before the anticipated breaking point, determined in previous trials. The subject tried to maintain a target mouth pressure of +20 or - 20 cmH,O, displayed on an oscilloscope, for the remainder of the breath hold. Both types of maneuver consistently prolonged breath-holding time in all subjects. However, a control maneuver, in which the subjects squeezed rubber bulbs with their hands, was equally effective in prolonging breath-holding time. The results demonstrate the important influence of psychological factors on breath-holding performance and emphasize the need for caution in the interpretation of effects of “relieving maneuvers” on breath-holding time.
five breaths. More recently, Rigg, Rebuck, and Campbell (17) demonstrated that small breaths provided very nearly as much relief as large breaths, and that transient inspiratory efforts against a closed airway (Mueller maneuvers) were also effective in prolonging breathholding time. In discussing this last finding, Rigg and co-workers suggested that relief of discomfort may have derived from afferents in inspiratory muscles or from a change in the pattern of vagal afferent information owing to intrapulmonary shifts of regional lung volume. A third possible explanation is suggested by the recent demonstration of large numbers of vagally innervated, slowly adapting stretch receptors in the extrathoracic trachea of the dog (1). These receptors respond to both distending and collapsing transmural pressures, and would be maximally activated during Mueller maneuvers such as those studied by Rigg et al. (17). If these airway receptors are responsible for the relief provided by inspiratory efforts, then expiratory efforts against a closed control of breathing; asphyxia airway (Valsalva maneuvers) should be equally effective. On the other hand, if inspiratory muscle receptors VOLUNTARY BREATH HOLDING is one of the oldest and are responsible, expiratory efforts should provide little or no relief. Regional shifts in intrapulmonary gas simplest methods for studying respiratory control. should be quite different during inspiratory and expiraEarly experiments established, and more recent studies tory efforts, and would not be expected to provide simihave confirmed that hypoxia and hypercapnia are imlar degrees of relief in the two circumstances. portant determinants of the breaking point (3, 11, 16). This paper describes experiments carried out to deterLung volume also influences the breaking point (14, 15), mine the effects of Valsalva and Mueller maneuvers, presumably via afferents from pulmonary mechanoreinstituted near the breaking point, on breath-holding ceptors (9), but the precise effect of the volume stimulus time. The results show clearly that both types of maneuis not easily defined (7, 8). Models of breath-holding ver are effective in prolonging breath-holding time. control proposed by Kobayasi and Sasaki (12) and by However, the results of control experiments, which apGodfrey and Campbell (6) emphasize that the breaking pear not to have been carried out in previous breathpoint is influenced by the duration as well as the intenholding studies, suggest that the effects of Valsalva and sity of afferent stimulation. Mueller maneuvers, and of other respiratory acts as In 1954 Fowler (4) reported that subjects who took well, may be nonspecific. several breaths of an asphyxiant gas mixture at the breaking point could resume breath holding for a few METHODS seconds, even though this maneuver did not diminish chemical stimulation. This finding, together with the Subjects. Two male and four female volunteers, aged old observation that rebreathing can be continued in the 21-30 yr, served as experimental subjects. All were in good health. Two of the subjects had a general knowlface of chemical stimuli that are intolerable during breath holding (ll), was taken to indicate that some edge of respiratory physiology, but none knew the speneural or mechanical feature of breathing provides re- cific hypothesis of the present study, and none had any lief from the discomfort of progressive asphyxia. This previous experience with breath-holding studies or with phenomenon was explored by Godfrey and Campbell (7), diving. None knew his own or any other subject’s results who found that a single relieving breath was as effective until all of the experiments were completed. Procedure. The experiments were carried out with in prolonging breath-holding time as was a cluster of Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
718
D. BARTLETT,
the subject seated in a dentist’s chair. After resting quietly in the chair for 10 min, the subject, wearing a noseclip, began to breathe room air through a rubber mouthpiece and a large threeway stopcock. End-tidal CO, concentration was monitored by continuous infrared analysis of gas sampled from the distal limb of the stopcock. Respiratory movements were registered by means of a bellows pneumograph, fastened around the lower part of the chest. The CO, and pneumogram signals were displayed on a multichannel recorder. When a resting steady state existed, as judged by the stability of these two signals, the operator started the breath hold without warning by turning the stopcock at the end of a norma l expiration. This connected the subject to the other limb of the stopcock, which was occluded by a rubber stopper, pierced by an l&gauge hypodermic needle connected to a pressure transducer. The signal from this transducer, representing mouth pressure during the breath hold, was registered on the recorder and also displayed on a large oscilloscope screen in the subject’s view. The sweep speed of the oscilloscope was very fast, so the beam appeared as a horizontal line and did not provide the subject with any time cues. When the subject was unable to continue the breath hold, he returned the stopcock to its original position and attempted to exhale to residual volume to provide a record of alveolar CO, concentration at the breaking point. The duration of the breath hold was noted by the operator, using a stopclock controlled by a foot switch. The effects of Valsalva and Mueller maneuvers were tested during breath holds initiated as described above. At a command from the operator given 5 s before the anticipated breaking point, as determined in two preceding control breath holds, the subject made a sustained inspiratory or expiratory effort against the closed airway, trying to reach and maintain a “target” mouth pressure of +20 or -20 cmH,O while watching the tracing of mouth pressure on the oscilloscope. This effort was maintained for the duration of the breath hold, which was terminated in the usual way. The subject did not know how the time of starting the relieving maneuvers was chosen, or that this time was equivalent in all trials. Experimental design. Preliminary experiments, in which the subjects performed repeated control breath holds, without relieving maneuvers, showed considerable improvement in performance from trial to trial within each l- to 2-h session (Fig. 1). This “training
c13L-L~
IlLI
3 4 5 6 TRIALS FIG. 1. Control breath-holding times in successive trials within experimental sessions. Each point denotes the average duration of six breath holds by each of 6 subjects. Vertical bars show + 1 SE. I
2
JR.
effect” did not diminish with experience, even with as many as 10 sessions. Thus the experiments had to be designed in such a way as to minimize the influence of this effect on the results. To this end, each breath hold with a Valsalva or Mueller maneuver was preceded and followed by control breath holds, and the analysis was made by comparison with the average of these two control observations (Fig. 2). In each session, following two initial control trials, breath holds with relieving maneuvers were alternated with controls, with rest periods of 5 min or more between trials. Four relieving maneuvers were studied in each of six experimental sessions. In each session the maneuvers were studied in one of six randomly selected, predetermined testing orders, as follows: Valsalva (V), V, V, V; Mueller (M), M, M, M; V, M, M, V; M, V, V, M; V, M, V, M; M, V, M, V. Thus in the main series of experiments, each type of maneuver was studied in 12 trials in each subject. Less extensive studies, with similar designs, were carried out in some of the subjects to determine the effects of less energetic man .euvers (+ ,lO cmH,O) and of Valsalva maneuvers alternating with relaxation at l- to 2-s intervals to approximate rhythmic breathing efforts. ControZ studies. Because the results of these studies, and some of those in the literature, suggested the possibility that the relief provided by certain respiratory maneuvers might not depend on specific receptor mechanisms, control experiments were carried out in each subject to see whether breath-holding time could be influenced by a maneuver not involving the respiratory system. A similar approach was proposed by Dornhorst (2). Two rubber sphygmomanometer bulbs were attached by rubber tubing and a Y junction to a pressure transducer, the signal from which was registered on the recorder and displayed on the oscilloscope. The control maneuver consisted of manually squeezing the rubber bulbs to approach and maintain a “target” bulb pressure 70zi.i -m g60I=
FIG. 2. Results of a representative experimental session. Points connected to each other by lines represent durations of control breath holds. Points marked V denote durations of breath holds with Valsalva maneuvers (+ 20 cmH,O). Prolongation of breath-holding time by each maneuver was assessed by comparison with average duration of the preceding and following control breath holds; these comparisons are shown graphically by vertical lines.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
VALSALVA
AND
MUELLER
MANEUVERS
ON
BREATH-HOLD
of 150 Torr. The subjects were cautioned not to perform a Valsalva maneuver during this effort, and mouth pressure was monitored to ensure that they did not. At least six trials were carried out in each subject. In some subjects additional trials were run, in which bulb squeezing and relaxation were alternated at intervals of 1-2 s. RESULTS
Base-line observations. The average durations of sequential control breath holds for all subjects in six different sessions are shown in Fig. 1. Each point in the figure represents the overall mean value for six breath holds by each subject, or 36 breath holds. As noted under METHODS, there was a definite tendency for breath-holding time to increase with successive trials within experimental sessions. A concomitant small increase in breaking point end-tidal Pco2 (1 Torr) was demonstrable in one of the four subjects who were able to exhale fully enough at the breaking point to provide technically acceptable data. This “training effect” was not due to a decrease in pre-breath-hold end-tidal PCO~ in the course of each session. The overall mean Pco* was 35.4 t 3.2 (SD) Torr before the first breath holds and 34.8 * 3.3 Torr before the last breath holds of the experimental sessions, and analysis of variance within each subject’s data showed no effect of trial sequence within sessions on initial Pco~. The “training effect” showed no tendency to diminish with experience (10 sessions), and no systematic change in performance from session to session could be discerned. Effects of Valsalva and Mueller manuevers. The effects of Valsalva and Mueller manuevers on breathholding time were assessed by comparing the duration of each test breath hold with the average duration of the control breath holds that immediately preceded and followed it (Fig. 2). Each subject’s data were analyzed separately. Analysis of variance showed no significant effect of trial sequence on the results, expressed either as absolute prolongation of breath-holding time in seconds, or as percent prolongation. In other words, Valsalva and Mueller maneuvers performed near the start of an experimental session were as effective in prolonging breath-holding time as those performed toward the end of the session, whether the data were expressed in seconds of prolongation or in percent prolongation. The results are reported in terms of percent prolongation, but the findings and conclusions are essentially the same if absolute prolongation times are used. The subjects’ performance in the main series of experiments is summarized in Table 1. Valsalva and Mueller maneuvers prolonged breath-holding time by an average of 14.8% and 11.7%, respectively. These results were very consistent and were statistically significant (P < 0.05, t-test for paired data) in every subject. The relative effectiveness of the two maneuvers varied among the subjects. Valsalva maneuvers were significantly more effective in two subjects (P < 0.05, t-test for unpaired data); in the other four subjects the degrees of prolongation by the two maneuvers were not significantly different (P > 0.05).
719
TIME
1. Percent prolongation of breath-holding time in six subjects by Valsalva maneuvers (12 trials per subject), Mueller maneuvers (12 trials), and bulb squeezing (6 trials) ~ __.--
TABLE
Percent
Prolongation
of Breath-Holding
Time by:
Subj Valsalva
_____~
CB JB MJL SC ss TH
maneuvers
16 + 8+2 15 + 4+1 32 + 14 +
3 1 4 1
Averages 14.8 + 3.9 --______Values listed for each subject values + 1 SE. Averages shown for all six subjects.
Mueller
maneuvers
Bulb
squeezing
B&2 12 2 2 16 k 2 422 22 2 2 8+2
17 11 30 13 31 11
11.7 + 2.6
18.8 t 3.8 ~------__~ percent prolongation mean values + 1 SE
are the mean are the overall
2 It It + + 2
4 3 1 4 9 1
Valsalva maneuvers with mouth pressure maintained at + 10 rather than +20 cmH,O were studied in three subjects with four trials each. This maneuver prolonged breath holding time significantly in these subjects (P < 0.05), the average prolongation being 13% compared with 15% with 20 cmH,O Valsalva maneuvers in the same subjects. Valsalva maneuvers alternating with relaxation at l- to 2-s intervals were tested with 46 trials in each of three subjects. This maneuver also resulted in significant prolongation (P < 0.05), and in one subject, the mean prolongation obtained (69%) was much greater than that found with sustained Valsalva maneuvers (32%). To determine whether the subjects tended to hyperventilate in anticipation of breath holds with test maneuvers as opposed to control breath holds, the prebreath-hold end-tidal Pco2 data were examined. No subject showed a significant difference in initial Pco2 between test and control breath holds (P > 0.20, t-test for paired data). As noted above, only four of the subjects were able to exhale fully enough at the breaking point to provide technically acceptable final Pco2 values. Valsalva maneuvers significantly increased breaking-point Pcoz in one of these subjects (P < 0.05), but did not significantly influence this variable in the other three. Similarly, final PCO~ was significantly increased by Mueller maneuvers in one subject, but unchanged in three. Effects of control maneuvers. As illustrated in Fig. 3 and documented in Table 1, squeezing the rubber bulbs prolonged breath-holding time significantly in all six subjects. The average degree of prolongation was 18.8%, or slightly more than that found with Valsalva and Mueller maneuvers. However, the data with control maneuvers were obtained in different experimental sessions, and though no source of bias is apparent, quantitative comparison with the Valsalva and Mueller maneuver data obtained in the main series of experiments may not be justified. Breath holds with bulb squeezing did not differ from their control breath holds with respect to pre-breath-hold Pco2 (P > 0.10, t-test for paired data). Breaking point Pco2 values obtained in this series of experiments were not adequate for analysis. A limited study of bulb squeezing alternating with relaxation at 1- to 2-s intervals indicated that this maneuver
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
720
D. BARTLETT,
v BULB SQUEEZE 9
t
BULB SQUEEZE
1 , .
SUBJECT M.J.L.
TRIALS FIG. 3. Results of an experimental session showing salva maneuvers ( +20 cmH,O) and of bulb squeezing. conventions are as in Fig. 2.
effects of ValSymbols and
was about as effective as sustained bulb squeezing in prolonging breath-holding time in four subjects, and definitely more effective in one subject. DISCUSSION
“Training effect .” Improvement in breath-holding performance with successive trials has been documented in previous investigations. Hill and Flack (11) commented on this phenomenon, and Gaensler and associates (5) rejected breath holding as a clinical test of pulmonary function partly because of this “training factor.” The effect was documented systematically by Heath and Irwin (lo), who were unable to find a physiological explanation for it. In the experiments reported here, the subjects increased their control breath-holding times during the sessions by an average of 12.5 s (Fig. 1). One of the subjects also showed a small but significant increase in breaking point Pco.,. The absence of a large increase in breaking point Pco., is expected in view of the virtual cessation of CO, evolution into the lungs late in the course of a breath hold (13). Breaking point PO, was not measured, but probably decreased from about 60-65 Torr to about 50-55 Torr during a typical experimental session ( 16). As noted above, the “traini n g” effect cannot be attributed to progressive hyperventilation during the sessions. Pre-breath-hold end-tidal Pco,! remained stable in the present experiments, and Heath and Irwin (10) found little progressive change in pre-breath-hold endtidal Pco., or end-tidal PO., in their subjects. A decreasing metabolic-rate during the course of an experimental session could theoretically bring about an increasing breath-holding time with a constant chemical breaking point. If a subject arrived in the laboratory in a mildly hypermetabolic state, he might approach resting metabolism during the course of the session, thus increasing his breath-holding time. No metabolic measurements were made to exclude this possibility, but two subjects showed typical progressive increases in breath-holding time in sessions preceded by 30 min of rest in the dentist’s chair.
JR.
A progressive increase in FRC also might account for seems unlikely the “training effect.” This possibility since data gathered by Mithoefer (15) from several sources indicate that an increase in FRC of about 1 liter would be required to account for the observed 35% increase in breath-holding time (Fig. 1). In summary, the “training effect” on breath-holding time, documented in these and previous studies, remains unexplained. In the absence of an apparent physiological basis, and in view of the importance of psychological factors in breath-holding performance (see below), consideration must be given to the possibility that the effect does not depend on specific receptor mechanisms. Effects of relieving maneuvers. The results confirm the report of Rigg and associates (17) that breath-holding time can be prolonged by Mueller maneuvers, performed near the breaking point. However, the roughly equal effectiveness of Valsalva maneuvers (see also (18)) argues against the proposal that afferents from inspiratory muscles or from intrapulmonary airways are responsible. Viewed by themselves, these findings are consistent with the hypothesis that stimulation of stretch receptors in the extrathoracic trachea can prolong breath-holding time. The results of the control experiments with bulb squeezing, however, vitiate such a mechanistic interpretation. These findings show that a maneuver that does not involve the respiratory system can consistently prolong breath-holding time, possibly by providing distraction from the discomfort of the breath hold. Whether transient bulb squeezing, followed by relaxation, would also provide relief was not determined, but would be of interest for comparison with the transient Mueller maneuvers studied by Rigg et al. (17). Other data suggesting the lack of specificity of relieving maneuvers include the comparable effectiveness of 10 cmH,O Valsalva maneuvers in the present study and previous reports indicating the uniform effectiveness of rebreathing near the breaking point, regardless of the number (7) or size (17) of the relieving breaths. Evidence that breath-holding performance is greatly influenced by suggestion was provided by Gaensler and associates (5), who prolonged their subjects’ breathholding times by giving them false time cues. On two occasions in the present study, the operator neglected to tell the subject when to start a Valsalva maneuver, and the subject continued to hold his breath for many seconds, waiting for the command. It therefore appears that breath holding behavior is importantly influenced by psychological as well as physiological factors. Investigations of breath-holding performance should include appropriate control procedures to take these factors into account in order to avoid erroneous conclusions about the influence of specific neural mechanisms on breath-holding time. I am grateful to the subjects for their cooperation and patience and to Mrs. Jennie Areson for expert technical assitance. This study was supported by Research Career Development Award HL-70129 and Research Grant HL-19827 from the National Heart and Lung Institute. Received
for publication
10 September
1976.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
VALSALVA
AND
MUELLER
MANEUVERS
ON
BREATH-HOLD
TIME
721
REFERENCES 1. BARTLETT, D., JR., P. JEFFERY, G. SANT’AMBROGIO, AND J. C. M. WISE. Location of stretch receptors in the trachea and bronchi of the dog. J. Physiol., London 258: 409-420, 1976. 2. DORNHORST, A. C. Discussion. In: Breathing: Hering-Breuer Centenary Symposium, edited by R. Porter. London: Churchill, 1970. 3. DOUGLAS, C. G., AND J. S. HALDANE. The regulation of normal breathing. J. PhysioZ., London 38: 420-440, 1909. 4. FOWLER, W. S. Breaking point of breath holding. J. Appl. PhysioZ. 6: 539-545, 1954. 5. GAENSLER, E. A., D. F. RAYL, AND D. M. DONNELLY. The breathholding test in pulmonary insufficiency. Surg. Gynecol. Obstet. 92: 81-90, 1951. 6. GODFREY, S., AND E. J. M. CAMPBELL. The control of breath holding. Respiration Physiol. 5: 385-400, 1968. 7. GODFREY, S., AND E. J. M. CAMPBELL. Mechanical and chemical control of breath holding. Quart. J. Ecptl. Physiol. 54: 117-128, 1969. 8. GODFREY, S., R. H. T. EDWARDS, AND D. A. WARRELL. The influence of lung shrinkage on breath holding time. Quart. J. Exptl. PhysioZ. 54: 129-140, 1969. 9. Guz, A., M. I. M. NOBLE, J. G. WIDDICOMBE, D. TRENCHARD, W. W. MUSKIN, AND A. R. MAKEY. The role of vagal and glossopharyuged afferent nerves in respiratory sensation, control of breathing and arterial pressure regulation in conscious man. CZin. Sci.
30: 161-170, 1966. 10. HEATH, J. R., AND C. J. IRWIN. An increase in breath-hold time appearing after breath-holding. Respiration PhysioZ. 4: 73-77, 1968. The effect of excess of carbon dioxide and 11. HILL, L, AND M. FLACK. of want of oxygen upon the respiration and circulation. J. PhysioZ., London 37: 77-111, 1908. 12. KOBAYASI, S., AND C. SASAKI. Breaking point of breath holding and tolerance time in rebreathing. Japan. J. Physiol. 17: 43-56, 1967. 13. LANPHIER, E. H., AND H. RAHN. Alveolar gas exchange during breath holding with air. J. AppZ. Physiol. 18: 478-482, 1963. 14. MITHOEFER, J. C. Lung volume restriction as a ventilatory stimulus during breath holding. J. AppZ. Physiol. 14: 701-705, 1959. 15. MITHOEFER, J. C. Breath holding. In: Handbook of Physiology. Respiration. Washington, D.C.: Am. Physiol. Sot., 1965, sect. 3, vol. II, chapt. 38, 1011-1025. 16. OTIS, A. B., H. RAHN, AND W. 0. FENN. Alveolar gas changes during breath holding. Am. J. Physiol. 152: 674-686, 1948. 17. RIGG, J. R. A., A. S. REBUCK, AND E. J. M. CAMPBELL. A study of factors influencing relief of discomfort in breath holding in normal subjects. CZin. Sic. MOL. Med. 47: 193-199, 1974. 18. SEARS, T. A. Discussion. In: Breathing: Hering-Breuer Centenary Symposium, edited by R. Porter. London: Churchill, 1970, p. 214.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
and Mueller time Dartmouth
Medical
maneuvers
School, Hanover,
New Hampshire
03755
BARTLETT, D., JR. Effects of Valsalva and Mueller maneuvers on breath-holding time. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. U(5): 717-721, 1977. -Breath holding to the breaking point was studied at FRC in six healthy subjects in the sitting position. Breath-holding time increased with successive trials within experimental sessions in all subjects. To study the influence of Valsalva and Mueller maneuvers on breath-holding performance, sustained inspiratory or expiratory effort against an occluded mouthpiece was initiated 5 s before the anticipated breaking point, determined in previous trials. The subject tried to maintain a target mouth pressure of +20 or - 20 cmH,O, displayed on an oscilloscope, for the remainder of the breath hold. Both types of maneuver consistently prolonged breath-holding time in all subjects. However, a control maneuver, in which the subjects squeezed rubber bulbs with their hands, was equally effective in prolonging breath-holding time. The results demonstrate the important influence of psychological factors on breath-holding performance and emphasize the need for caution in the interpretation of effects of “relieving maneuvers” on breath-holding time.
five breaths. More recently, Rigg, Rebuck, and Campbell (17) demonstrated that small breaths provided very nearly as much relief as large breaths, and that transient inspiratory efforts against a closed airway (Mueller maneuvers) were also effective in prolonging breathholding time. In discussing this last finding, Rigg and co-workers suggested that relief of discomfort may have derived from afferents in inspiratory muscles or from a change in the pattern of vagal afferent information owing to intrapulmonary shifts of regional lung volume. A third possible explanation is suggested by the recent demonstration of large numbers of vagally innervated, slowly adapting stretch receptors in the extrathoracic trachea of the dog (1). These receptors respond to both distending and collapsing transmural pressures, and would be maximally activated during Mueller maneuvers such as those studied by Rigg et al. (17). If these airway receptors are responsible for the relief provided by inspiratory efforts, then expiratory efforts against a closed control of breathing; asphyxia airway (Valsalva maneuvers) should be equally effective. On the other hand, if inspiratory muscle receptors VOLUNTARY BREATH HOLDING is one of the oldest and are responsible, expiratory efforts should provide little or no relief. Regional shifts in intrapulmonary gas simplest methods for studying respiratory control. should be quite different during inspiratory and expiraEarly experiments established, and more recent studies tory efforts, and would not be expected to provide simihave confirmed that hypoxia and hypercapnia are imlar degrees of relief in the two circumstances. portant determinants of the breaking point (3, 11, 16). This paper describes experiments carried out to deterLung volume also influences the breaking point (14, 15), mine the effects of Valsalva and Mueller maneuvers, presumably via afferents from pulmonary mechanoreinstituted near the breaking point, on breath-holding ceptors (9), but the precise effect of the volume stimulus time. The results show clearly that both types of maneuis not easily defined (7, 8). Models of breath-holding ver are effective in prolonging breath-holding time. control proposed by Kobayasi and Sasaki (12) and by However, the results of control experiments, which apGodfrey and Campbell (6) emphasize that the breaking pear not to have been carried out in previous breathpoint is influenced by the duration as well as the intenholding studies, suggest that the effects of Valsalva and sity of afferent stimulation. Mueller maneuvers, and of other respiratory acts as In 1954 Fowler (4) reported that subjects who took well, may be nonspecific. several breaths of an asphyxiant gas mixture at the breaking point could resume breath holding for a few METHODS seconds, even though this maneuver did not diminish chemical stimulation. This finding, together with the Subjects. Two male and four female volunteers, aged old observation that rebreathing can be continued in the 21-30 yr, served as experimental subjects. All were in good health. Two of the subjects had a general knowlface of chemical stimuli that are intolerable during breath holding (ll), was taken to indicate that some edge of respiratory physiology, but none knew the speneural or mechanical feature of breathing provides re- cific hypothesis of the present study, and none had any lief from the discomfort of progressive asphyxia. This previous experience with breath-holding studies or with phenomenon was explored by Godfrey and Campbell (7), diving. None knew his own or any other subject’s results who found that a single relieving breath was as effective until all of the experiments were completed. Procedure. The experiments were carried out with in prolonging breath-holding time as was a cluster of Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
718
D. BARTLETT,
the subject seated in a dentist’s chair. After resting quietly in the chair for 10 min, the subject, wearing a noseclip, began to breathe room air through a rubber mouthpiece and a large threeway stopcock. End-tidal CO, concentration was monitored by continuous infrared analysis of gas sampled from the distal limb of the stopcock. Respiratory movements were registered by means of a bellows pneumograph, fastened around the lower part of the chest. The CO, and pneumogram signals were displayed on a multichannel recorder. When a resting steady state existed, as judged by the stability of these two signals, the operator started the breath hold without warning by turning the stopcock at the end of a norma l expiration. This connected the subject to the other limb of the stopcock, which was occluded by a rubber stopper, pierced by an l&gauge hypodermic needle connected to a pressure transducer. The signal from this transducer, representing mouth pressure during the breath hold, was registered on the recorder and also displayed on a large oscilloscope screen in the subject’s view. The sweep speed of the oscilloscope was very fast, so the beam appeared as a horizontal line and did not provide the subject with any time cues. When the subject was unable to continue the breath hold, he returned the stopcock to its original position and attempted to exhale to residual volume to provide a record of alveolar CO, concentration at the breaking point. The duration of the breath hold was noted by the operator, using a stopclock controlled by a foot switch. The effects of Valsalva and Mueller maneuvers were tested during breath holds initiated as described above. At a command from the operator given 5 s before the anticipated breaking point, as determined in two preceding control breath holds, the subject made a sustained inspiratory or expiratory effort against the closed airway, trying to reach and maintain a “target” mouth pressure of +20 or -20 cmH,O while watching the tracing of mouth pressure on the oscilloscope. This effort was maintained for the duration of the breath hold, which was terminated in the usual way. The subject did not know how the time of starting the relieving maneuvers was chosen, or that this time was equivalent in all trials. Experimental design. Preliminary experiments, in which the subjects performed repeated control breath holds, without relieving maneuvers, showed considerable improvement in performance from trial to trial within each l- to 2-h session (Fig. 1). This “training
c13L-L~
IlLI
3 4 5 6 TRIALS FIG. 1. Control breath-holding times in successive trials within experimental sessions. Each point denotes the average duration of six breath holds by each of 6 subjects. Vertical bars show + 1 SE. I
2
JR.
effect” did not diminish with experience, even with as many as 10 sessions. Thus the experiments had to be designed in such a way as to minimize the influence of this effect on the results. To this end, each breath hold with a Valsalva or Mueller maneuver was preceded and followed by control breath holds, and the analysis was made by comparison with the average of these two control observations (Fig. 2). In each session, following two initial control trials, breath holds with relieving maneuvers were alternated with controls, with rest periods of 5 min or more between trials. Four relieving maneuvers were studied in each of six experimental sessions. In each session the maneuvers were studied in one of six randomly selected, predetermined testing orders, as follows: Valsalva (V), V, V, V; Mueller (M), M, M, M; V, M, M, V; M, V, V, M; V, M, V, M; M, V, M, V. Thus in the main series of experiments, each type of maneuver was studied in 12 trials in each subject. Less extensive studies, with similar designs, were carried out in some of the subjects to determine the effects of less energetic man .euvers (+ ,lO cmH,O) and of Valsalva maneuvers alternating with relaxation at l- to 2-s intervals to approximate rhythmic breathing efforts. ControZ studies. Because the results of these studies, and some of those in the literature, suggested the possibility that the relief provided by certain respiratory maneuvers might not depend on specific receptor mechanisms, control experiments were carried out in each subject to see whether breath-holding time could be influenced by a maneuver not involving the respiratory system. A similar approach was proposed by Dornhorst (2). Two rubber sphygmomanometer bulbs were attached by rubber tubing and a Y junction to a pressure transducer, the signal from which was registered on the recorder and displayed on the oscilloscope. The control maneuver consisted of manually squeezing the rubber bulbs to approach and maintain a “target” bulb pressure 70zi.i -m g60I=
FIG. 2. Results of a representative experimental session. Points connected to each other by lines represent durations of control breath holds. Points marked V denote durations of breath holds with Valsalva maneuvers (+ 20 cmH,O). Prolongation of breath-holding time by each maneuver was assessed by comparison with average duration of the preceding and following control breath holds; these comparisons are shown graphically by vertical lines.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
VALSALVA
AND
MUELLER
MANEUVERS
ON
BREATH-HOLD
of 150 Torr. The subjects were cautioned not to perform a Valsalva maneuver during this effort, and mouth pressure was monitored to ensure that they did not. At least six trials were carried out in each subject. In some subjects additional trials were run, in which bulb squeezing and relaxation were alternated at intervals of 1-2 s. RESULTS
Base-line observations. The average durations of sequential control breath holds for all subjects in six different sessions are shown in Fig. 1. Each point in the figure represents the overall mean value for six breath holds by each subject, or 36 breath holds. As noted under METHODS, there was a definite tendency for breath-holding time to increase with successive trials within experimental sessions. A concomitant small increase in breaking point end-tidal Pco2 (1 Torr) was demonstrable in one of the four subjects who were able to exhale fully enough at the breaking point to provide technically acceptable data. This “training effect” was not due to a decrease in pre-breath-hold end-tidal PCO~ in the course of each session. The overall mean Pco* was 35.4 t 3.2 (SD) Torr before the first breath holds and 34.8 * 3.3 Torr before the last breath holds of the experimental sessions, and analysis of variance within each subject’s data showed no effect of trial sequence within sessions on initial Pco~. The “training effect” showed no tendency to diminish with experience (10 sessions), and no systematic change in performance from session to session could be discerned. Effects of Valsalva and Mueller manuevers. The effects of Valsalva and Mueller manuevers on breathholding time were assessed by comparing the duration of each test breath hold with the average duration of the control breath holds that immediately preceded and followed it (Fig. 2). Each subject’s data were analyzed separately. Analysis of variance showed no significant effect of trial sequence on the results, expressed either as absolute prolongation of breath-holding time in seconds, or as percent prolongation. In other words, Valsalva and Mueller maneuvers performed near the start of an experimental session were as effective in prolonging breath-holding time as those performed toward the end of the session, whether the data were expressed in seconds of prolongation or in percent prolongation. The results are reported in terms of percent prolongation, but the findings and conclusions are essentially the same if absolute prolongation times are used. The subjects’ performance in the main series of experiments is summarized in Table 1. Valsalva and Mueller maneuvers prolonged breath-holding time by an average of 14.8% and 11.7%, respectively. These results were very consistent and were statistically significant (P < 0.05, t-test for paired data) in every subject. The relative effectiveness of the two maneuvers varied among the subjects. Valsalva maneuvers were significantly more effective in two subjects (P < 0.05, t-test for unpaired data); in the other four subjects the degrees of prolongation by the two maneuvers were not significantly different (P > 0.05).
719
TIME
1. Percent prolongation of breath-holding time in six subjects by Valsalva maneuvers (12 trials per subject), Mueller maneuvers (12 trials), and bulb squeezing (6 trials) ~ __.--
TABLE
Percent
Prolongation
of Breath-Holding
Time by:
Subj Valsalva
_____~
CB JB MJL SC ss TH
maneuvers
16 + 8+2 15 + 4+1 32 + 14 +
3 1 4 1
Averages 14.8 + 3.9 --______Values listed for each subject values + 1 SE. Averages shown for all six subjects.
Mueller
maneuvers
Bulb
squeezing
B&2 12 2 2 16 k 2 422 22 2 2 8+2
17 11 30 13 31 11
11.7 + 2.6
18.8 t 3.8 ~------__~ percent prolongation mean values + 1 SE
are the mean are the overall
2 It It + + 2
4 3 1 4 9 1
Valsalva maneuvers with mouth pressure maintained at + 10 rather than +20 cmH,O were studied in three subjects with four trials each. This maneuver prolonged breath holding time significantly in these subjects (P < 0.05), the average prolongation being 13% compared with 15% with 20 cmH,O Valsalva maneuvers in the same subjects. Valsalva maneuvers alternating with relaxation at l- to 2-s intervals were tested with 46 trials in each of three subjects. This maneuver also resulted in significant prolongation (P < 0.05), and in one subject, the mean prolongation obtained (69%) was much greater than that found with sustained Valsalva maneuvers (32%). To determine whether the subjects tended to hyperventilate in anticipation of breath holds with test maneuvers as opposed to control breath holds, the prebreath-hold end-tidal Pco2 data were examined. No subject showed a significant difference in initial Pco2 between test and control breath holds (P > 0.20, t-test for paired data). As noted above, only four of the subjects were able to exhale fully enough at the breaking point to provide technically acceptable final Pco2 values. Valsalva maneuvers significantly increased breaking-point Pcoz in one of these subjects (P < 0.05), but did not significantly influence this variable in the other three. Similarly, final PCO~ was significantly increased by Mueller maneuvers in one subject, but unchanged in three. Effects of control maneuvers. As illustrated in Fig. 3 and documented in Table 1, squeezing the rubber bulbs prolonged breath-holding time significantly in all six subjects. The average degree of prolongation was 18.8%, or slightly more than that found with Valsalva and Mueller maneuvers. However, the data with control maneuvers were obtained in different experimental sessions, and though no source of bias is apparent, quantitative comparison with the Valsalva and Mueller maneuver data obtained in the main series of experiments may not be justified. Breath holds with bulb squeezing did not differ from their control breath holds with respect to pre-breath-hold Pco2 (P > 0.10, t-test for paired data). Breaking point Pco2 values obtained in this series of experiments were not adequate for analysis. A limited study of bulb squeezing alternating with relaxation at 1- to 2-s intervals indicated that this maneuver
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
720
D. BARTLETT,
v BULB SQUEEZE 9
t
BULB SQUEEZE
1 , .
SUBJECT M.J.L.
TRIALS FIG. 3. Results of an experimental session showing salva maneuvers ( +20 cmH,O) and of bulb squeezing. conventions are as in Fig. 2.
effects of ValSymbols and
was about as effective as sustained bulb squeezing in prolonging breath-holding time in four subjects, and definitely more effective in one subject. DISCUSSION
“Training effect .” Improvement in breath-holding performance with successive trials has been documented in previous investigations. Hill and Flack (11) commented on this phenomenon, and Gaensler and associates (5) rejected breath holding as a clinical test of pulmonary function partly because of this “training factor.” The effect was documented systematically by Heath and Irwin (lo), who were unable to find a physiological explanation for it. In the experiments reported here, the subjects increased their control breath-holding times during the sessions by an average of 12.5 s (Fig. 1). One of the subjects also showed a small but significant increase in breaking point Pco.,. The absence of a large increase in breaking point Pco., is expected in view of the virtual cessation of CO, evolution into the lungs late in the course of a breath hold (13). Breaking point PO, was not measured, but probably decreased from about 60-65 Torr to about 50-55 Torr during a typical experimental session ( 16). As noted above, the “traini n g” effect cannot be attributed to progressive hyperventilation during the sessions. Pre-breath-hold end-tidal Pco,! remained stable in the present experiments, and Heath and Irwin (10) found little progressive change in pre-breath-hold endtidal Pco., or end-tidal PO., in their subjects. A decreasing metabolic-rate during the course of an experimental session could theoretically bring about an increasing breath-holding time with a constant chemical breaking point. If a subject arrived in the laboratory in a mildly hypermetabolic state, he might approach resting metabolism during the course of the session, thus increasing his breath-holding time. No metabolic measurements were made to exclude this possibility, but two subjects showed typical progressive increases in breath-holding time in sessions preceded by 30 min of rest in the dentist’s chair.
JR.
A progressive increase in FRC also might account for seems unlikely the “training effect.” This possibility since data gathered by Mithoefer (15) from several sources indicate that an increase in FRC of about 1 liter would be required to account for the observed 35% increase in breath-holding time (Fig. 1). In summary, the “training effect” on breath-holding time, documented in these and previous studies, remains unexplained. In the absence of an apparent physiological basis, and in view of the importance of psychological factors in breath-holding performance (see below), consideration must be given to the possibility that the effect does not depend on specific receptor mechanisms. Effects of relieving maneuvers. The results confirm the report of Rigg and associates (17) that breath-holding time can be prolonged by Mueller maneuvers, performed near the breaking point. However, the roughly equal effectiveness of Valsalva maneuvers (see also (18)) argues against the proposal that afferents from inspiratory muscles or from intrapulmonary airways are responsible. Viewed by themselves, these findings are consistent with the hypothesis that stimulation of stretch receptors in the extrathoracic trachea can prolong breath-holding time. The results of the control experiments with bulb squeezing, however, vitiate such a mechanistic interpretation. These findings show that a maneuver that does not involve the respiratory system can consistently prolong breath-holding time, possibly by providing distraction from the discomfort of the breath hold. Whether transient bulb squeezing, followed by relaxation, would also provide relief was not determined, but would be of interest for comparison with the transient Mueller maneuvers studied by Rigg et al. (17). Other data suggesting the lack of specificity of relieving maneuvers include the comparable effectiveness of 10 cmH,O Valsalva maneuvers in the present study and previous reports indicating the uniform effectiveness of rebreathing near the breaking point, regardless of the number (7) or size (17) of the relieving breaths. Evidence that breath-holding performance is greatly influenced by suggestion was provided by Gaensler and associates (5), who prolonged their subjects’ breathholding times by giving them false time cues. On two occasions in the present study, the operator neglected to tell the subject when to start a Valsalva maneuver, and the subject continued to hold his breath for many seconds, waiting for the command. It therefore appears that breath holding behavior is importantly influenced by psychological as well as physiological factors. Investigations of breath-holding performance should include appropriate control procedures to take these factors into account in order to avoid erroneous conclusions about the influence of specific neural mechanisms on breath-holding time. I am grateful to the subjects for their cooperation and patience and to Mrs. Jennie Areson for expert technical assitance. This study was supported by Research Career Development Award HL-70129 and Research Grant HL-19827 from the National Heart and Lung Institute. Received
for publication
10 September
1976.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
VALSALVA
AND
MUELLER
MANEUVERS
ON
BREATH-HOLD
TIME
721
REFERENCES 1. BARTLETT, D., JR., P. JEFFERY, G. SANT’AMBROGIO, AND J. C. M. WISE. Location of stretch receptors in the trachea and bronchi of the dog. J. Physiol., London 258: 409-420, 1976. 2. DORNHORST, A. C. Discussion. In: Breathing: Hering-Breuer Centenary Symposium, edited by R. Porter. London: Churchill, 1970. 3. DOUGLAS, C. G., AND J. S. HALDANE. The regulation of normal breathing. J. PhysioZ., London 38: 420-440, 1909. 4. FOWLER, W. S. Breaking point of breath holding. J. Appl. PhysioZ. 6: 539-545, 1954. 5. GAENSLER, E. A., D. F. RAYL, AND D. M. DONNELLY. The breathholding test in pulmonary insufficiency. Surg. Gynecol. Obstet. 92: 81-90, 1951. 6. GODFREY, S., AND E. J. M. CAMPBELL. The control of breath holding. Respiration Physiol. 5: 385-400, 1968. 7. GODFREY, S., AND E. J. M. CAMPBELL. Mechanical and chemical control of breath holding. Quart. J. Ecptl. Physiol. 54: 117-128, 1969. 8. GODFREY, S., R. H. T. EDWARDS, AND D. A. WARRELL. The influence of lung shrinkage on breath holding time. Quart. J. Exptl. PhysioZ. 54: 129-140, 1969. 9. Guz, A., M. I. M. NOBLE, J. G. WIDDICOMBE, D. TRENCHARD, W. W. MUSKIN, AND A. R. MAKEY. The role of vagal and glossopharyuged afferent nerves in respiratory sensation, control of breathing and arterial pressure regulation in conscious man. CZin. Sci.
30: 161-170, 1966. 10. HEATH, J. R., AND C. J. IRWIN. An increase in breath-hold time appearing after breath-holding. Respiration PhysioZ. 4: 73-77, 1968. The effect of excess of carbon dioxide and 11. HILL, L, AND M. FLACK. of want of oxygen upon the respiration and circulation. J. PhysioZ., London 37: 77-111, 1908. 12. KOBAYASI, S., AND C. SASAKI. Breaking point of breath holding and tolerance time in rebreathing. Japan. J. Physiol. 17: 43-56, 1967. 13. LANPHIER, E. H., AND H. RAHN. Alveolar gas exchange during breath holding with air. J. AppZ. Physiol. 18: 478-482, 1963. 14. MITHOEFER, J. C. Lung volume restriction as a ventilatory stimulus during breath holding. J. AppZ. Physiol. 14: 701-705, 1959. 15. MITHOEFER, J. C. Breath holding. In: Handbook of Physiology. Respiration. Washington, D.C.: Am. Physiol. Sot., 1965, sect. 3, vol. II, chapt. 38, 1011-1025. 16. OTIS, A. B., H. RAHN, AND W. 0. FENN. Alveolar gas changes during breath holding. Am. J. Physiol. 152: 674-686, 1948. 17. RIGG, J. R. A., A. S. REBUCK, AND E. J. M. CAMPBELL. A study of factors influencing relief of discomfort in breath holding in normal subjects. CZin. Sic. MOL. Med. 47: 193-199, 1974. 18. SEARS, T. A. Discussion. In: Breathing: Hering-Breuer Centenary Symposium, edited by R. Porter. London: Churchill, 1970, p. 214.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 21, 2018. Copyright © 1977 American Physiological Society. All rights reserved.
