Objectives

1. Understand basic mechanisms of respiratory system adaptation to high altitude. ƒ
2. Describe mechanisms in the development of high-altitude pulmonary edema and its prophylaxis and treatment.
3. Describe the effects of increased environmental pressure on lung volume.
4. Describe the mechanism of shallow-water blackouts using Boyle’s law.
5. Describe the diving reflex and its components.

Key words

breath-hold diving; diving reflex; high-altitude pulmonary edema; hypoxic vasoconstriction; nitric oxide; shallow-water blackout

Abbreviations

AMS = acute mountain sickness; CO = cardiac output; ED = elite breath-hold diver; HAPE = high-altitude pulmonary edema; NO = nitric oxide; PAN₂ = alveolar partial pressure of nitrogen; PAO₂ = alveolar partial pressure of oxygen; PETCO₂ = end-tidal CO₂ pressure; PETO₂ = end-tidal O₂ pressure; PT = total pressure; PVR = peripheral vascular resistance; RV = residual volume; SaO₂ = arterial oxygen saturation; SV = stroke volume; TLC = total lung capacity

The human respiratory system has a remarkable ability to adapt to many different environmental demands. In this review, we will discuss how the respiratory system is affected by the changes in barometric pressure that occur with ascent to high altitude and during breath-hold diving.

Low-Pressure Adaptations

On ascent to high altitude, barometric pressure falls, resulting in a decrease in the partial pressure of oxygen (Fig 1). Gas exchange in the lung is therefore immediately affected by the reduced pressure gradient for oxygen between alveolus and capillary. Other adverse effects of altitude on gas exchange include (1) reduced affinity of hemoglobin for oxygen at lower arterial partial pressure of O₂ owing to the shape of the oxyhemoglobin dissociation curve; (2) reduced time for equilibration of oxygen with hemoglobin because of increased cardiac output (CO); and (3) worsened ventilation-perfusion matching likely secondary to the development of interstitial edema from increased pulmonary vascular pressures as a result of hypoxic vasoconstriction.¹

Figure 1. Gas pressures related to oxygenation at various altitudes. Actual pressures will also vary with latitude, season, and local weather. Notice that the SaO₂ at 29,028 feet (summit of Mt. Everest) is higher than at 24,000 feet because of the effect of respiratory alkalosis on the oxyhemoglobin dissociation curve. PiO₂ = partial pressure of inspired oxygen; PaO₂ = partial pressure of arterial oxygen; SaO₂ = arterial oxyhemoglobin saturation. Adapted from Hultgren HN. High-altitude medical problems. In: Rubenstein E, Federman DD, eds. Scientific American medicine. New York, NY: Scientific American, 1992; CTM IX:1.

In order to continue to effect adequate respiration at high altitude, the respiratory system first responds with an increase in minute ventilation. This process enhances the partial pressure of alveolar oxygen by the increased alveolar clearance of CO₂. In addition, the fall in arterial partial pressure of CO₂ causes a respiratory alkalosis, which shifts the oxyhemoglobin dissociation curve to the left and therefore enhances the binding of oxygen to hemoglobin. The body responds further by increasing the amount of ^2,3 diphosphoglycerate, which binds hemoglobin and improves the release of oxygen at the tissue level. Although the energy required to move the less dense air at altitude is lower than that at sea level for an equivalent degree of activity, the increased work of breathing at altitude results in a net, overall increased energy requirement to perform physical activity at altitude.¹

With chronic adaptation, minute ventilation remains elevated, but the kidneys respond to the disturbed acid-base status by eliminating bicarbonate, thus leaving a more chronic state of compensated respiratory alkalosis. The lower ambient oxygen over time also causes an increase in the hemoglobin concentration, resulting in erythrocytosis. At the tissue level, increased capillary density, muscle myoglobin concentration, and mitochondrial oxidative enzyme activity all may contribute to enhanced oxygen utilization and the body’s ability to adapt to chronic hypoxia. However, one potential maladaptation is chronic mountain sickness, or Monge’s disease, which is typically seen in natives of the Andes and is characterized by severe pulmonary hypertension, erythrocytosis, and right heart failure.

Abnormal Effects of High Altitude

The responses described above are the processes by which the body adapts to the effects of high altitude through acclimatization. When healthy individuals ascend without sufficient time for acclimatization, various manifestations of highaltitude illness may be seen. The most common effect is that of acute mountain sickness (AMS), which occurs in about 25% of sea-level dwellers who ascend rapidly to altitudes >8,000 feet.² AMS is characterized by headache, anorexia, nausea, vomiting, insomnia, fatigue, or dizziness. The etiology of AMS is thought to be related to mild edema of the brain, which may become life-threatening if it progresses to high-altitude cerebral edema. Prophylaxis of AMS is best provided by proper acclimatization (see below), although acetazolamide and dexamethasone are also effective.² Both AMS and high-altitude cerebral edema respond to oxygen, descent, and treatment with acetazolamide and dexamethasone.² This review will not discuss these aspects of high-altitude illness further, but instead will focus on high-altitude pulmonary edema (HAPE).

The most common, serious adverse effect of high altitude is HAPE (Fig 2). HAPE is a form of noncardiogenic pulmonary edema that is characterized by shortness of breath, cough, chest discomfort, and severe hypoxemia. HAPE occurs in 0.1 to 15% of visitors to high altitude, who typically develop symptoms in association with physical exertion after 2 to 3 days at altitude.²

Figure 2. Schematic diagram illustrating current thinking about the characteristics of individuals susceptible to HAPE on exposure to high altitude. ET-1 = endothelin-1; ANP = atrial natriuretic peptide.

Although the pathophysiology of HAPE is unknown, a key factor is the development of excessive pulmonary capillary pressure. Individuals susceptible to HAPE appear to have an enhanced pulmonary vasoconstrictive response to hypoxia.² This response occurs in an uneven fashion throughout the lung, resulting in areas of intense vasoconstriction and other areas of overperfusion. This overperfusion leads to increased pulmonary capillary pressure,³ which ultimately causes stress failure of the capillaries and leak of plasma and cells into the interstitial space and alveoli. Enhanced pulmonary arteriolar and venous constriction may also lead to increased vascular leak.⁴ In addition, HAPE subjects may also have increased vascular permeability and fluid overload from increased levels of atrial natriuretic peptide and vasopressin.⁵ Individuals susceptible to HAPE may be identified by their abnormal pulmonary vascular responses to hypoxia or exercise as measured by Doppler echocardiography.⁶

But why do only some subjects develop such excessive pulmonary pressures? First, many studies have shown that subjects susceptible to HAPE may have a reduced hypoxic ventilatory response,2 and are thus unable to adapt their minute ventilation and acid-base status as described above. This may lead to worsened hypoxemia and oxygen delivery. However, variability in the human hypoxic ventilatory response does not allow this factor to accurately predict the development of HAPE. Another recent theory is that such subjects have less ability to synthesize nitric oxide (NO), a potent vasodilator.⁷ Patients with HAPE produce less exhaled NO than healthy people, and exogeneous NO is known to lower pulmonary artery pressures and improve oxygenation in HAPE patients.⁸ Recent studies also suggest that those susceptible to HAPE may have higher plasma levels of the potent vasoconstrictor endothelin-1, or increased uptake of endothelin-1 by the pulmonary circulation.⁴

Patients with HAPE may also be less able to clear extravasated fluid from the alveolar space.⁹ Sartori and colleagues⁹ demonstrated that inhaled salmeterol, a long-acting beta-agonist, protects against the development of HAPE. In this study, the benefits of salmeterol were postulated to result from enhanced clearance of alveolar fluid as a result of increased sodium channel activity, as the drug had no effect on pulmonary pressures. In further support of this notion, HAPE-susceptible individuals had hyperpolarization of their nasal potential difference at sea level, a measure thought to possibly reflect reduced sodium and water transport by the alveolar epithelium.

Early studies on the BAL fluid from HAPE patients found that such fluid was rich in protein, complement, and eicosanoids, suggesting a role for inflammation in the pathogenesis of the disease.¹⁰ High urinary leukotriene E4 levels have also been found in patients with HAPE.¹¹ However, more recent studies have been unable to confirm the presence of inflammation in the early stages of the development of HAPE,^12,13 suggesting that the findings of the previous studies reflect a secondary inflammatory response to capillary injury. Inflammatory mediators may be generated from exposure of antigenic components of the epithelial basement membrane, which occurs during stress failure of the capillaries. These mediators then initiate an inflammatory response that potentiates the edema formation.¹⁴

The best treatment of HAPE is preventive.² Adequate acclimatization with gradual ascent (<1,000 feet per day once above 14,000 ft) and avoidance of excessive physical activity early during stays at altitude is recommended. For climbers who must ascend quickly, nifedipine (20 to 30 mg extended release every 12 h) has been shown to be prophylactic,² and recent data with salmeterol (125 µg inhaled every 12 h) also suggest this as a prophylactic agent.⁹ Theoretically, sildenafil, a selective phosphodiesterase inhibitor, may also be useful in the prophylaxis of HAPE.¹⁵

Treatment of HAPE is primarily with supplemental oxygen and descent. Descent can be mimicked by use of a portable hyperbaric chamber or positive endexpiratory pressure mask until definitive descent can take place.² Nifedipine may also be used for treatment. As discussed earlier, inhaled NO improves oxygenation and pulmonary pressures in HAPE.

High-Pressure Adaptations

Now we turn to the opposite end of the atmospheric-pressure spectrum with breath-hold diving. Specifically, we will review changes in lung physiology at depth, respiratory drive in response to breath-hold training, and evidence that a diving reflex exists in humans.

History of Breath-hold Diving

Breath-hold diving began 1,500 years ago with the Ama divers off the coasts of Korea and Japan, where divers foraged the sea bottom for shellfish and edible seaweed.¹⁶ The divers, mostly women, performed repeated dives up to 80 feet in depth with breath-hold times as long as 2 min. To support their families, they worked year-round and often through pregnancy. Because of their unique training and large numbers, they have been the subjects of many research studies in lung physiology, respiratory drive, and the diving reflex.^16,17 Divers using selfcontained underwater breathing apparatus (scuba) technology have replaced many of the Ama divers, but traditional breath-hold diving remains a source of livelihood for many in this region even today.

More recently, extreme breath-hold diving has become a popular sport worldwide (http://www.redefineyourlimits.com) and the currently recognized world record depth of 525 feet (Tanya Streeter, August 2002) is proof of the human lungs’ ability to rapidly adapt to extreme changes of pressure. The continued tragedy associated with this sport underscores the extreme physiology of such attempts.¹⁸

Changes in Lung Volume With Diving

Up until the 1960s, it was generally believed that the depth to which a breathhold diver could descend was limited by the ratio of their residual volume (RV) to total lung capacity (TLC).¹⁹ When a diver descends during a dive, the lung gas volume is compressed according to Boyle’s law,

P₁V₁=P₂V₂

where P₁ and V₁ represent the pressure and volume at one condition, and P₂ and V₂, another condition. For every 10 m of depth, an additional 1 atm of pressure is placed on the diver. The lungs must be uniformly compressed in order to prevent the development of pressure gradients and “thoracic squeeze.” If the lung is not uniformly compressed, as may be the case when TLC is less than RV during a dive, then a pressure gradient occurs between the compressed and noncompressed areas, resulting in thoracic squeeze. If a typical diver has a TLC of 7 L and a RV of 1.8 L (25% of TLC), according to Boyle’s law he or she would be able to dive 116 feet (38.8 m) (Fig 3) before TLC equaled RV. This depth is far shallower than the current world record of 525 feet, where, using Boyle’s law, a compression of lung volume to 440 mL should result, a value much below estimated RV. Therefore, in order to achieve such a depth, alternate mechanisms of pulmonary gas volume compression must exist. Leading theories suggest that changes in circulating blood volume are important.

Figure 3. Changes in lung volume with depth based on Boyle’s law (P₁V₁ = P₂V₂) for a hypothetical diver with a TLC of 7 L on dry land. The theoretical limit is defined as TLC at depth = RV (~25% of TLC). The current world record is in marked excess of this theoretical limit. Adapted from Craig.²⁰

Physiologists have shown that a transthoracic pressure gradient does not occur in a diver diving at RV, despite changes in depth.²⁰ To account for this, a further decrease in intrathoracic gas volume must occur after the lung has reached RV. The exact mechanism of this continued fall in lung volume is incompletely understood, but is postulated to result from changes in circulating blood volume. As the body is subjected to increased pressure, peripheral blood vessels become compressed, forcing additional blood volume from the periphery to the central circulation.²⁰ Consequently, central vessels, including the pulmonary vascular bed, become engorged. The volume of blood displaced into the thoracic cavity is proportional to the depth of the dive. At a depth of 30 m, as much as 1 L of blood volume is shifted to the thorax.²¹ The resulting alveolar capillary engorgement compresses the surrounding airspace, effectively decreasing the lung volume by intrinsic alveolar compression. Data from diving seals²² further support this hypothesis.

Radiographic examination of the chest during breath-hold diving shows engorgement of pulmonary vessels and increased transverse diameter of the heart.¹⁹ Taken together, these pieces of evidence support increased thoracic blood volume as a contributor to thoracic gas compression.

Overdistention of pulmonary capillary vessels can result in hemorrhage into the alveolar space and hemoptysis. Accordingly, case reports of transient hemoptysis occurring in elite divers after deep breath-hold dives^21,23 implicates a similar mechanism of lung compression. In all reported cases, the divers were healthy, did not aspirate sea water, and recovered quickly without complications.^21,23 The high resistance of lung capillaries to mechanical stress²⁴ may explain why hemoptysis is not a more significant problem in breath-hold divers.

Changes in Alveolar Gas Content With Depth

To understand the changes that occur in alveolar gas composition with diving, it is necessary to first review Dalton’s law of partial pressures. Dalton’s law states that the total pressure exerted by a gaseous mixture is the sum of the partial pressures of all the component gases. Stated mathematically,

PT = P₁ + P₂ + P₃ + … P_n

where PT is the total pressure for the mixture and P₁ to P_n are the partial pressures of each individual gas in the mixture. Significant dynamic changes occur in the partial pressure of oxygen (O₂), carbon dioxide (CO₂), and nitrogen (N₂) in the alveolus during a breath-hold dive.

As a diver descends, the PT applied to the lung increases, as do the alveolar partial pressures of O₂, CO₂, and N₂ in the alveoli (PAO₂, PACO₂ and PAN₂ respectively). At maximal depth, PAO₂ often exceeds PAO₂ at the surface, despite the time elapsed since breath-hold initiation. The opposite is true as a diver ascends. PT decreases, as do PAO₂, PACO₂ and PAN₂. A “steal” phenomenon can occur¹⁶ in which O₂ diffuses from the blood back into the alveoli when PAO₂ exceeds PAO₂ as a result of rapid changes in alveolar partial pressures (Fig 4).

Figure 4. Changes in alveolar gas content with breath-hold diving demonstrating the “steal” phenomenon that can occur on emersion when PaO₂ exceeds PAO₂ and results in accentuated tissue hypoxemia. If tissue hypoxemia is profound, shallow-water blackout may result.

Thus, the diver is at the greatest risk of hypoxemia just prior to emersion from the dive when alveolar oxygen is at its lowest. This condition can result in the socalled “shallow-water blackout,” which typically occurs at depths of ≤ 10 feet and is often implicated in breath-hold diving deaths from drowning. Many drownings of recreational swimmers are attributed to this cause, as well.¹⁹

Changes in Respiratory Drive With Breath-Hold Training

Divers frequently hyperventilate before performing a dive. The purpose of hyperventilation is to decrease PACO₂ and therefore increase the time to the breath-hold breaking point. This is defined as the levels of PAO₂ and PACO₂ past which the divers can no longer hold their breath. The drive to breathe at PAO₂ and PACO₂ beyond the breath-hold breaking point increases 10-fold in normal, untrained individuals. Although the breath-hold breaking point is dependent on both PAO₂ and PACO₂, a rising PACO₂ is the primary respiratory stimulus.¹⁹ By hyperventilating and effectively decreasing PACO₂ before a dive, breath-hold time can be prolonged. By prolonging the breath-hold time, the degree of alveolar and tissue hypoxia on emersion is also increased, which raises the risk of shallow-water blackout.¹⁶ Divers most at risk of shallow-water blackouts are novices diving to moderate depths. With training and experience, tolerance of hypoxia and hypercapnia can be prolonged and the body can adapt to these conditions.¹⁹

A number of investigators have looked at end-tidal O₂ and CO₂ pressure (PETO₂ and PETCO₂, respectively) during breath-hold dives, and some have attempted to predict the maximum depth one might attain with conscious emersion after a single breath hold.²⁵ Researchers predicted that a dive to >100 m with conscious emersion would not be possible.^19,25 In 1988, the first elite breath-hold diver exceeded 100 m (Enzo Majorca), and the current recognized world record is well beyond this depth (Tanya Streeter, 160 m).

In an attempt to better understand the physiology of elite breath-hold divers (EDs), Ferretti and colleagues²⁵ studied three EDs and a group of nine healthy, untrained control individuals. The goal of this study was to compare alveolar gas exchange and alveolar gas composition at the end of deep dives and dry-land breath-holds. This study yielded several interesting results. First, EDs and control subjects experience the same, linear change in PETO₂ and PETCO₂ with breath hold, but the EDs held their breath longer to achieve the same degree of hypoxia and hypercapnia. Second, the EDs could tolerate a greater degree of hypoxia (lowest arterial oxygen saturation [SaO₂], 38%) than could control subjects (lowest SaO₂, 83%). Others have shown that the response to hypercapnia in divers is also blunted¹⁹ such that trained divers have smaller increases in minute ventilation following equal increases in PETCO₂. Increased anaerobic metabolism and blunted response to hypoxia and hypercapnia may represent acquired mechanisms by which EDs, with training, are able to hold their breath longer and dive deeper. Additionally, a genetic bias may contribute to performance differences.²⁵

The Diving Reflex

The diving reflex, comprising peripheral vasoconstriction, bradycardia, and decreased CO, has been postulated to be an oxygen-conserving mechanism.

There is clear evidence of a diving reflex in birds and diving mammals,^22,26 but definitive proof of one in humans remains elusive.

Bradycardia is frequently found in breath-hold divers, and its mechanism is not well understood. Water temperature, lung volume at breath hold, and depth of dive are all important factors contributing to the magnitude of the bradycardic response.^27,28 Many believe it is a compensatory response to concomitant peripheral vasoconstriction. Bradycardia is more pronounced with face immersion in cold water than with dry breath hold.²⁸ The magnitude of the bradycardia can be extreme and heart rates as low as 20 beats/min during dives have been recorded in elite breath-hold divers.¹⁹

The effect of breath-hold diving on CO has been reported inconsistently within the literature, possibly because of differences in lung volume at the time of breath-hold dive.²⁶ Ferrigno et al²⁶ addressed this issue by investigating changes in CO at both small (15% of vital capacity) and large (85% of vital capacity) lung volumes. Large lung volumes were of particular interest, as most breath-hold divers (elite and Ama) perform dives at high lung volume. In their study, they found a decrease in CO only with dives at large lung volume. There are many factors resulting in this fall in CO, most importantly stroke volume (SV) and heart rate. During typical breath-hold diving, SV is simultaneously decreased by increased intrathoracic pressures resulting from large lung volumes and increased by shunting of peripheral blood to the central circulation from increased pressure on the extremities. The net effect of this is a slight increase in SV, but not enough to compensate for the often profound bradycardia. Consequently, a fall in CO is typically observed during breath-hold diving at large lung volumes.

With diving, an increase in peripheral vascular resistance (PVR) occurs from increased pressure on the extremities. Increased PVR increases arterial blood pressure and afterload, promoting bradycardia and decreased CO. Increased PVR also effectively shunts blood away from the extremities and into the central circulation. Given the energy constraints of prolonged breath hold, decreasing blood flow to the extremities has the advantage of decreasing oxygen consumption by skeletal muscle. This theory is supported by evidence that trained divers are able to undergo more efficient anaerobic metabolism6 and often emerge from a dive with sore muscles.²⁹ The key components of the diving reflex are interdependent (Fig 5) and may represent an adaptive mechanism of altered oxygen consumption that allows humans to survive the extreme hypoxia of the breath-hold diving environment.

Figure 5. Schematic of the key components of the diving reflex and the means by which they promote oxygen conservation. HR= heart rate.

Summary

As we have seen, the respiratory system has the potential to adapt to extreme variations in atmospheric pressure, whether it is a climber adapting to the rarefied atmosphere of a high mountain peak such as Mt. Everest (0.34 atm) or a diver who performs breath-hold dives to 160 m (17 atm). In fact, the ability of some elite athletes to perform at these extremes of atmospheric pressure is nothing short of miraculous, and it exceeds our current scientific understanding of the mechanisms involved in performing these feats. The studies into these mechanisms not only advance science but hold important lessons for our patients as well. By observing the functioning of the human body at its limits, we learn important lessons in physiology that yield insights into tolerance of stressed hypoxia, which have important implications for our patients with lung disease.

References

1. Schoene RB. Limits of human lung function at high altitude. J Exp Biol 2001; 204(pt 18):3121–3127
2. Hackett PH, Roach RC. High-altitude illness. N Engl J Med 2001; 345:107–114
3. Maggiorini M, Mélot C, Pierre S, et al. High-altitude pulmonary edema is initially caused by an increase in capillary pressure. Circulation 2001; 103:2078–2083
4. Bärtsch P, Swenson ER, Maggiorini M. Update: high altitude pulmonary edema. Adv Exp Med Biol 2001; 502:89–106
5. Cosby RL, Sophocles AM, Durr JA, et al. Elevated plasma atrial natriuretic factor and vasopressin in high-altitude pulmonary edema. Ann Intern Med 1988; 109:796–799
6. Grünig E, Mereles D, Hildebrandt W, et al. Stress Doppler echocardiography for identification of susceptibility to high altitude pulmonary edema. J Am Coll Cardiol 2000; 35;980–987
7. Duplain H, Sartori C, Lepori M, et al. Exhaled nitric oxide in high-altitude pulmonary edema: role in the regulation of pulmonary vascular tone and evidence for a role against inflammation. Am J Respir Crit Care Med 2000; 162:221–224
8. Scherrer U, Vollenweider L, Delabays A, et al. Inhaled nitric oxide for high-altitude pulmonary edema. N Engl J Med 1996; 334:624–629
9. Sartori C, Allemann Y, Duplain H, et al. Salmeterol for the prevention of high-altitude pulmonary edema. N Engl J Med 2002; 346:1631–1636
10. Schoene RB, Swenson ER, Pizzo CJ, et al. The lung at high altitude: bronchoalveolar lavage in acute mountain sickness and pulmonary edema. J Appl Physiol 1988; 64:2605–2613
11. Kaminsky DA, Jones K, Schoene RB, et al. Urinary leukotriene E4 levels in high-altitude pulmonary edema: a possible role for inflammation. Chest 1996; 110:939–945
12. Swenson ER, Maggiorini M, Mongovin S, et al. Pathogenesis of highaltitude pulmonary edema: inflammation is not an etiologic factor. JAMA 2002; 287:2228–2235
13. Bärtsch P, Eichenberger U, Ballmer PE, et al. Urinary leukotriene E4 levels are not increased prior to high-altitude pulmonary edema. Chest 2000; 117:1393–1398
14. Tsukimoto K, Yoshimura N, Ichioka M, et al. Protein, cell, and LTB4 concentrations of lung edema fluid produced by high capillary pressures in rabbit. J Appl Physiol 1994; 76:321–327
15. Kleinsasser A, Loeckinger A. Are sildenafil and theophylline effective in the prevention of high-altitude pulmonary edema? Med Hypotheses 2002; 59:223–225
16. Hong SK, Rahn H. The diving women of Korea and Japan. Sci Am 1967; 216(5):34–43
17. Honda Y, Hayashi F, Yoshida A, et al. Relative contributions of chemical and non-chemical drives to the breath-holding time in breath-hold divers (Ama). Jpn J Physiol 1981; 31:181–186
18. Smith G. The deadly dive. Sports Illustrated 2003; 98(June 16):62–73
19. Ferretti G. Extreme human breath-hold diving. Eur J Appl Physiol 2001; 84:254–271
20. Craig AB Jr. Depth limits of breath hold diving (an example of Fennology). Respir Physiol 1968; 5:14–22
21. Boussuges A, Pinet C, Thomas P, et al. Haemoptysis after breath-hold diving. Eur Respir J 1999; 13:697–699
22. Elsne, RW, Scholander AB, Craig E, et al. A venous blood oxygen reservoir in the diving elephant seal. Physiologist 1964; 7:124–130
23. Kiyan E, Aktas S, Toklu AS. Hemoptysis provoked by voluntary diaphragmatic contractions in breath-hold divers. Chest 2001; 120:2098–2100
24. West JB, Tsukimoto K, Mathieu-Costello O, et al. Stress failure in pulmonary capillaries. J Appl Physiol 1991; 70:1731–1742
25. 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; 70:794–802
26. Ferrigno M, Hickey DD, Liner MH, et al. Cardiac performance in humans during breath holding. J Appl Physiol 1986; 60:1871–1877
27. Ferrigno M, Grassi B, Ferretti G, et al. Electrocardiogram during deep breath-hold dives by elite divers. Undersea Biomed Res 1991; 18:81–91
28. Butler PJ, Woakes AJ. Heart rate in humans during underwater swimming with and without breath-hold. Respir Physiol 1987; 69:387–399
29. Andersson JP, Liner MH, Runow E, et al. Diving response and arterial oxygen saturation during apnea and exercise in breath-hold divers. J Appl Physiol 2002; 93:882–886