Breath-Hold Diving: Respiratory Function on the Longer Term : Medicine & Science in Sports & Exercise

Journal Logo


Breath-Hold Diving

Respiratory Function on the Longer Term


Author Information
Medicine & Science in Sports & Exercise 43(7):p 1214-1219, July 2011. | DOI: 10.1249/MSS.0b013e31820a4e0c
  • Free


Glossopharyngeal insufflation (GI) is a breathing technique that allows pumping air in to the lungs by contractions of the upper pharyngeal muscles. It was initially adopted by patients experiencing respiratory insufficiency due to poliomyelitis (6,7) to increase off-respirator time. It has been reported more recently that competitive breath-hold divers (BHD) use GI to add air on top of their fully inflated lungs to reach greater depths, store surplus oxygen, and add space for CO2 storage (21). However, information about the short and possible long-term (patho)physiological effects of GI (21) is scarce. So far, there are only single reports on acute lung damage due to BHD and GI (17,18,22). In a recent study by Chung et al. (4), asymptomatic signs of pneumomediastinum could be detected in five of six BHD after performing GI in computed tomographic scans. No prospective lung function data have been reported yet in elite competitive BHD.

In the recent years, the boundaries of breath-hold durations (current world record = 11 min 35 s) and diving depths (current word record = 214 m) have been continuously pushed further (1), exposing BHD to both severe hypoxia and hypercapnia during these challenges. Conflicting data exist whether excelling CO2 tolerance may be an inherited phenomenon (2,15) or an adaptation to frequent short-term hypercapnia as observed in other sports (8). Suggestions have been made that elite performance in BHD may be hereditary, as shown in three BHD family members with diminished respiratory drive under hypercapnia (13,15); an altered respiratory drive was also detected and postulated in later studies (12,34).

We hypothesized that aging and constant performance of GI may alter pulmonary distensibility and therefore possibly induce long-term damage. In a previous study, we were able to show a transient change in pulmonary static lung compliance due to GI (34). Here, we performed a 3-yr follow-up measurement in four divers from our preliminary study group. Further, we measured pulmonary compliance during 5-min intervals in three consecutive sessions in a greater group of 12 experienced elite divers. To assess further adaptive effects to hypercapnia during continuation of competitive breath-hold diving, we also investigated the respiratory drive using the steady-state method (5) in this unrelated group of divers in adherence to our and other previous studies.


The study protocol was approved by the Ethics Committee of the Albert-Ludwigs-University at Freiburg, Germany, and the study was performed in accordance with the ethical standards set by the latest Declaration of Helsinki. All subjects signed their informed consent before the study. For eligibility, the BHD had to present actual training records with a minimum annual diving record exceeding of 100 dives and participation in three competitions per year. Only male subjects were investigated (Table 1) and recruited via Internet and the local diving communities.

Divers' characteristics.

Baseline spirometry was assessed using a constant volume body plethysmograph (MasterLab®; Viasys, Höchberg, Germany) according to current ATS/ERS guidelines (25,35). Reference values for forced expiratory flows and volumes as well as static volumes were calculated according to the technique described by Matthys et al. (24). Maximal inspiratory vital capacity (VC) was considered as a baseline measure for the comparison of lung volumes before and during GI (VCGI).

The esophageal pressure method was used for the measurement of pulmonary compliance. Esophageal pressure was used as a determinant of transpulmonary pressure. All measurements were performed in a sitting position with the subjects breathing quietly through the spirometer. Pressure curves were recorded using a conventional balloon catheter (nSpire® GmbH, Oberthulba, Germany) placed in the mid third part of the esophagus. The esophageal balloon was filled with 1.5 mL of air over a three-way tap and a syringe for optimal pressure transduction; for compliance measurements during GI, the balloon was filled with 0.5 mL of air to avoid signal disturbances. No recordings were done during esophageal contractions. For correct assessment of pulmonary compliance, a standardization of volume history has been calculated according to the resting lung volume measurements. Dynamic compliance (Cdyn) was registered during quiet breathing at a respiratory rate between 10 min−1 and 5 min−1. At least 10 closed curves with clearly determined points of reversal at the end of inspiration and expiration were registered, and the mean value was calculated. Static compliance (Cstat) was carried out during gentle passive noninterrupted expiration from total lung capacity (Cstatrest) or after GI (CstatGI). At least three technically acceptable curves had to be achieved for calculation of Cstatrest. For determination of Cstat, the slope of the curve was manually adjusted for perfect fit at 50%-80% of the calculated lung volume (MasterLab®; Viasys) (14).

Subsequent measurements of Cstat were done as follows: Cstat was measured before (Cstatrest), at GImax (CstatGI), and each minute for 5 min (Cstat1-5min) in three sessions after GI. GI was always performed without previous hyperventilation. These measurements were performed according to the measurement of Cstatrest.

Mouth occlusion pressure at 0.1 s (P0.1), tidal breathing ventilation per minute (E), and breathing frequency (fb) were measured as an equivalent of respiratory drive. For P0.1 resting measurements (P0.1 rest), the subjects were asked to breathe freely through a mouthpiece with a nose clip. Ten airway occlusions occurred at random order 0.1 s after the onset of inspiration in a computer-controlled fashion (MasterLab®; Viasys), with the mean P0.1 being automatically calculated (3). The effect of CO2 on respiratory drive was measured using mixtures of 6% CO2 (P0.1 6%) and 9% CO2 (P0.1 9%) in ambient air using the steady-state method (5). Drive measurements under 6% and 9% CO2 were started after reaching a steady state in E and fb. P0.1 measurements were performed analogous to the resting measurements. Additionally, blood gas samples were drawn from the arterialized ear lobe (cobas b221; Roche Diagnostics, Grenzach-Wyhlen, Germany), and body surface area (BSA) was calculated according to Mosteller's (26) method.

Data were analyzed using SigmaPlot® 11.2 for Windows® (SSI, San Jose, CA). Results are expressed as means ± SD or as median (range) if data were not normally distributed. All variables were tested for normal distribution and equal variance. Paired t-tests (Student's t-test or Mann-Whitney rank sum test) and one-way repeated-measures ANOVA (RM-ANOVA) for repeated measurements were used to test statistical differences between measurements (Student-Newman-Keuls test). P < 0.05 was considered statistically significant.


Mean BHD training in the group of 12 BHD averaged 6.6 yr. Personal records are presented in Table 1. Forced expiratory flows and volumes as well as body plethysmographic measurements of static lung volumes yielded normal values (Table 2). In the group of four BHD who had presented to our facility previously, mean forced expiratory volumes and inspiratory VC rather had increased during the 3-yr interval (Table 3).

Respiratory characteristics.
Individual diver's follow-up after 3 yr.

One BHD declined the esophageal catheter, and in one BHD, data are missing because of technical problems. Compliance was therefore assessed in 10 divers, of whom 4 had participated in the previous study (Fig. 1). A statistically significant difference was found in the current measurement for CstatGI and all other Cstat measurements (all P < 0.001) using RM-ANOVA. No significant difference was observed between previous follow-up measurements and the current experiments for Cstatrest, CstatGI, Cstat1min, and Cstat3min (Table 4).

Compliance follow-up measurements for five consecutive minutes. Rest, resting measurement; GI, glossopharnygeal insufflation; Cstat, static compliance. C statGI is significantly different from all other measurements (P < 0.001).
Blood gas samples.

There was a statistically significant increase in fb between 6% and 9% CO2 (all P < 0.05) when compared with free breathing. A significant increase in E and P0.1 between free breathing and 6% CO2 as well as between 6% CO2 and 9% CO2 (all P < 0.001) was detected. For normalization according to body size, E/BSA was calculated, which was significantly different between measurements under 9% CO2 and 6% CO2 as well as between 9% CO2 and ambient air breathing (all P < 0.001; Fig. 2). Blood gas samples were drawn from the arterialized ear lobe at each experiment (Table 4). The missing values are due to technical problems. No statistically significant difference between former measurements of BHD using 6% CO2 rebreathing could be detected (Table 3).

Steady-state measurement of respiratory drive using 6% and 9% CO2 in ambient air (fb, breathing frequency; VE, tidal volume; BSA, body surface area; P 0.1, mouth occlusion pressure). For statistical significance levels, please refer to the results section.


Thus far, little is known about the (patho)physiological and especially the long-term effects of GI to the pulmonary system (21), causing transient barotrauma to the lungs while hyperinflating up to 50% above total lung capacity (23). Our study in 12 elite BHD for the first time shows that persistent competitive breath-hold diving using the GI technique does not impose deleterious long-term effects onto the pulmonary parenchyma. The increase seen in ventilatory volumes may rather indicate a training effect of GI. Further, the investigation of respiratory drive regarding CO2 response supported the evidence that increased tolerance of high-level CO2 resembles a training effect rather than an inherited phenomenon.

In a preceding study, we showed a transient change in pulmonary static lung compliance due to GI indicating high stress to the lung tissue during hyperinflation (34). Extreme transpulmonary pressures during GI up to 109 cm H2O (23) and hyperinflation may lead to parenchymal changes in the lung tissue as seen in pulmonary barotraumas during mechanical ventilation on the intensive care unit (28). The current findings in the follow-up measurements of four divers constantly performing BHD with GI revealed no changes to their lungs' distensibility (Table 3). All resting values of Cstat and Cdyn were within the normal range; also, pulmonary compliance has been shown not to be age dependent so that significant aging effects can be neglected (14).

In magnetic resonance imaging during GI maneuvers, full reversibility of the transient hyperinflation with herniation of the lung beneath the sternum and enlargement of the costodiaphragmatic angle had been shown previously (11). This is in line with the findings of velocity encoding cine magnetic resonance imaging of the main pulmonary artery showing a complete reversibility of hemodynamic parameters mimicking pulmonary hypertension during GI (10).

In contrast to our previous findings on the time course of pulmonary static compliance after GI (n = 5 divers) (34), here we did not observe a statistically significant difference between measurements of Cstat at baseline and Cstat after 1 min (Cstat1min). Looking at absolute values in the current group of 12 divers, a higher mean value of Cstat1min was observed (7.41 L·kPa−1 (n = 12) vs 5.36 L·kPa−1 (n = 5), all P > 0.05). The lack of difference between Cstat at GI and Cstat1min may therefore be due to the higher sample size in the current study. The absolute value for Cstat at GI, however, was close to our previous data (13.75 L·kPa−1 (n = 12) vs 13.21 L·kPa−1 (n = 5)).

In the group of four divers who were available for a 3-yr follow-up pulmonary function testing, mean ventilatory volumes had increased over time (Table 3). This somewhat unexpected finding was in line with earlier studies that reported positive effects on lung volumes from GI after a 5-wk training in competitive swimmers and a cross-sectional study comparing lung volumes between novice and more experienced BHD (20,27). We had anticipated to find a decline in forced expiratory volumes, should GI cause deleterious effects to the airways and lung tissue on the longer term. Indeed, in scuba divers who were of comparable age to our BHD, a decrease in forced expiratory volume in 1 s of about 100 mL was found after a 3-yr follow-up that was attributable to the effects of diving exposure (33). Although the sample size of the group of BHD who were available for follow-up was rather small, the lack of a deleterious effect was in line with the findings in the larger group of BHD we investigated here. The mean breath-hold diving performance of 6.6 yr, with the maximum being 10 yr, does further support the evidence that the stress on the pulmonary parenchyma is transient and reversible. It should be emphasized that long-term effects may occur in divers with more extreme exposures such as very deep diving, which causes numerous other effects on the body and the lungs (21). In our BHD, no subject except for diver 6 reached depth levels beyond 70 m (Table 1). However, the current results do indicate that extensive use of GI for breath-hold diving at more shallow depths can be considered a safe procedure regarding possible long-term damage to the pulmonary parenchyma.

A reduced respiratory drive has been shown previously in studies measuring ventilatory response in hyperoxia and hypercapnia (16) and during immersion accompanied by both hyperoxia and hypercapnia (9). For the first time and in line with findings of Grassi et al. (15) in three elite BHD family members, we could show a blunted response to CO2 using a steady-state CO2 challenge in ambient air in a greater group of 12 unrelated elite BHD (5). On the contrary, ventilatory response to inhaled CO2 was shown to be significantly different when monozygous and dizygous twins were studied, revealing that personality factors seemed to influence the frequency response to CO2 (2). In this even larger group, we could now show a blunted response to 6% and 9% CO2 in ambient air with a characteristic breathing pattern focusing on increased E with steady fb (Fig. 2). This was also observed in a study by Ivancev et al. (16) using a ramp protocol for inducing hypercapnia with hyperoxia (29). Our unrelated BHD revealed the same blunted response to CO2 when normalized for BSA as shown by Grassi et al. (15) with an average of 9 L·min−1·m−2 for divers at 52 torr PETCO2 versus 12 L·min−1·m−2 in control subjects and in the study by Lambertsen (19) with 31 L·min−1 at 50 mm Hg PETCO2. Schaefer (31) demonstrated two different ventilatory response types classifying a high-response group (higher tidal volume, higher breathing frequency) and a low-response group for resting breathing and under various CO2 concentrations in normal subjects. The low-response breathing pattern was also seen in tank instructors and divers and may therefore serve as an adaptation marker to elevated CO2 concentrations (32). The same type of low response is seen in BHD. We therefore postulate that breath-hold diving results in a trainable adaptation to increased CO2 levels; genetics cannot be the only determination of a successful BHD. However, low-response subjects might be prone to excel in breath-hold diving or related sports. This is in line with findings of reduced ventilatory responses in other sports such as underwater hockey (8) or tank instructors (30). Whether GI has additional enhancing effects on CO2 tolerance may be speculated but is not supported by the current study.

A limitation of the present study is the lack of a control group. However, data from our previous study may serve as control because the study setup was identical except for the measurement of respiratory drive under 9% CO2 and normal data may be considered from the studies by Grassi et al. (15) and Lambertsen (19) using the same technique. For pulmonary compliance, current reference values exist and may be referred to (14). Moreover, the follow-up was only possible in a limited number of four subjects who were willing to undergo all of the demanding procedures again. The lack of an impairment in lung function as an indicator of damage due to GI was in accordance with the cross-sectional data of samples of BHD with long-lasting GI experience published by us and others. However, this was the first study with a longitudinal design to provide long-term pulmonary data in individual elite BHD subjects. We cannot exclude, however, that there are deleterious effects in susceptible subjects who may give up BHD or escape clinical investigation. Further, a selection bias cannot be ruled out because of the recruitment method of the divers via Internet and the local diving communities.

In conclusion, the present study is the first prospective follow-up on BHD lung function. The main finding was that repeated GI in elite BHD does not permanently alter pulmonary distensibility because all changes to the pulmonary compliance remain transient and lung function parameters remain stable. Also, respiratory drive measurements revealed a distinct pattern of blunted ventilatory response to elevated CO2 concentrations indicating this phenomenon being associated with BH training rather than being inherited.

There was no special funding supporting this study.

All authors declare no conflict of interest regarding this study.

The authors thank all divers for their participation.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.


1. AIDA - Worldwide Federation for breath-hold diving [Internet]. Lausanne (Switzerland): AIDA; [cited 2010 Nov 12]. Available from:
2. Arkinstall WW, Nirmel K, Klissouras V, Milic-Emili J. Genetic differences in the ventilatory response to inhaled CO2. J Appl Physiol. 1974;36(1):6-11.
3. ATS/ERS. Statement on respiratory muscle testing. Am J Respir Crit Care Med. 2002;166(4):518-624.
4. Chung SC, Seccombe LM, Jenkins CR, Frater CJ, Ridley LJ, Peters MJ. Glossopharyngeal insufflation causes lung injury in trained breath-hold divers. Respirology. 2010;15(5):813-7.
5. Clark TJ. The measurement of ventilatory response to CO2. Br J Dis Chest. 1968;62(3):113-25.
6. Dail CW. 'Glossopharyngeal breathing' by paralyzed patients: a preliminary report. Calif Med. 1951;75(3):217-8.
7. Dail CW, Affeldt JE, Collier CR. Clinical aspects of glossopharyngeal breathing; report of use by one hundred postpoliomyelitic patients. J Am Med Assoc. 1955;158(6):445-9.
8. Davis FM, Graves MP, Guy HJ, Prisk GK, Tanner TE. Carbon dioxide response and breath-hold times in underwater hockey players. Undersea Biomed Res. 1987;14(6):527-34.
9. Delapille P, Verin E, Tourny-Chollet C, Pasquis P. Ventilatory responses to hypercapnia in divers and non-divers: effects of posture and immersion. Eur J Appl Physiol. 2001;86(1):97-103.
10. Eichinger M, Puderbach M, Smith H, et al. Magnetic resonance-compatible-spirometry: principle, technical evaluation and application. Eur Respir J. 2007;30(5):972-9.
11. Eichinger M, Walterspacher S, Scholz T, et al. Lung hyperinflation: foe or friend? Eur Respir J. 2008;32(4):1113-6.
12. Ferretti G. Extreme human breath-hold diving. Eur J Appl Physiol. 2001;84(4):254-71.
13. Ferretti G, Costa M, Ferrigno M, et al. Alveolar gas composition and exchange during deep breath-hold diving and dry breath holds in elite divers. J Appl Physiol. 1991;70(2):794-802.
14. Galetke W, Feier C, Muth T, Ruehle K, Borsch-Galetke E, Randerath W. Reference values for dynamic and static pulmonary compliance in men. Respir Med. 2007;101(8):1783-9.
15. Grassi B, Ferretti G, Costa M, et al. Ventilatory responses to hypercapnia and hypoxia in elite breath-hold divers. Respir Physiol. 1994;97(3):323-32.
16. Ivancev V, Palada I, Valic Z, et al. Cerebrovascular reactivity to hypercapnia is unimpaired in breath-hold divers. J Physiol. 2007;582(Pt 2):723-30.
17. Jacobson FL, Loring SH, Ferrigno M. Pneumomediastinum after lung packing. Undersea Hyperb Med. 2006;33(5):313-6.
18. Kiyan E, Aktas S, Toklu AS. Hemoptysis provoked by voluntary diaphragmatic contractions in breath-hold divers. Chest. 2001;120(6):2098-100.
19. Lambertsen CJ. Carbon dioxide and respiration in acid-base homeostasis. Anesthesiology. 1960;21642-51.
20. Lemaitre F, Clua E, Andreani B, Castres I, Chollet D. Ventilatory function in breath-hold divers: effect of glossopharyngeal insufflation. Eur J Appl Physiol. 2010;108(4):741-7.
21. Lindholm P, Lundgren CEG. The physiology and pathophysiology of human breath-hold diving. J Appl Physiol. 2009;106(1):284-92.
22. Liner MH, Andersson JPA. Pulmonary edema after competitive breath-hold diving. J Appl Physiol. 2008;104(4):986-90.
23. Loring SH, O'Donnell CR, Butler JP, Lindholm P, Jacobson F, Ferrigno M. Transpulmonary pressures and lung mechanics with glossopharyngeal insufflation and exsufflation beyond normal lung volumes in competitive breath-hold divers. J Appl Physiol. 2007;102(3):841-6.
24. Matthys H, Zaiss A, Theissen J, Virchow J, Werner P. Definitions, predicted values and measures for diagnosis of obstructive, restrictive as well as combined ventilatory disorders for clinical lung function diagnostics. Atemw-Lungenkrkh. 1995;21(3):130-8.
25. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J. 2005;26(2):319-38.
26. Mosteller RD. Simplified calculation of body-surface area. N Engl J Med. 1987;317(17):1098.
27. Nygren-Bonnier M, Gullstrand L, Klefbeck B, Lindholm P. Effects of glossopharyngeal pistoning for lung insufflation in elite swimmers. Med Sci Sports Exerc. 2007;39(5):836-41.
28. Petersen GW, Baier H. Incidence of pulmonary barotrauma in a medical ICU. Crit Care Med. 1983;11(2):67-9.
29. Read DJ. A clinical method for assessing the ventilatory response to carbon dioxide. Australas Ann Med. 1967;16(1):20-32.
30. Schaefer KE. The Role of Carbon Dioxide in the Physiology of Human Diving. Washington (DC): National Academy of Science; 1955. p. 131-9.
31. Schaefer KE. Respiratory pattern and respiratory response to CO2. J Appl Physiol. 1958;13(1):1-4.
32. Schaefer KE. Effect of prolonged diving training. In: Lambertsen, CJ, Greenbaum, LJ, editors. Proceedings Second Symposium on Underwater Physiology. Washington (DC): National Academy of Sciences; 1963. p. 271-8.
33. Skogstad M, Thorsen E, Haldorsen T. Lung function over the first 3 years of a professional diving career. Occup Environ Med. 2000;57(6):390-5.
34. Tetzlaff K, Scholz T, Walterspacher S, et al. Characteristics of the respiratory mechanical and muscle function of competitive breath-hold divers. Eur J Appl Physiol. 2008;103(4):469-75.
35. Wanger J, Clausen JL, Coates A, et al. Standardisation of the measurement of lung volumes. Eur Respir J. 2005;26(3):511-22.


© 2011 American College of Sports Medicine