The effect of the diving response on alveolar gas exchange was investigated in 15 subjects. During steady-state exercise (80 W) on a cycle ergometer, the subjects performed 40-s apneas in air and 40-s apneas with face immersion in cold (10°C) water. Heart rate decreased and blood pressure increased during apneas, and the responses were augmented by face immersion. Oxygen uptake from the lungs decreased during apnea in air (-22% compared with eupneic control) and was further reduced during apnea with face immersion (-25% compared with eupneic control). The plasma lactate concentration increased from control (11%) after apnea in air and even more after apnea with face immersion (20%), suggesting an increased anaerobic metabolism during apneas. The lung oxygen store was depleted more slowly during apnea with face immersion because of the augmented diving response, probably including a decrease in cardiac output. Venous oxygen stores were probably reduced by the cardiovascular responses. The turnover times of these gas stores would have been prolonged, reducing their effect on the oxygen uptake in the lungs. Thus the human diving response has an oxygen-conserving effect.
- diving reflex
- oxygen and carbon dioxide stores
- breath holding
the human diving response is mainly characterized by brady-cardia, decreased cardiac output, peripheral vasoconstriction, and increased arterial blood pressure (8, 12, 13). Both the cardiac and vascular components of the human diving response are initiated by apnea and augmented by several factors, among which face immersion in cold water is especially recognized (13, 31). During apnea, the cardiac output decreases because of concurrent reductions in stroke volume and heart rate (5, 11, 39). Peripheral blood flow is reduced because of arterial vasoconstriction (14, 39), whereas blood flow in the carotid artery increases (18, 33), suggesting a redistribution of blood flow toward vital organs during apnea. The human diving response is also initiated by apnea and augmented by face immersion during exercise (3, 4, 37, 40).
Although the diving response of some diving animals has been shown to be O2 conserving (6), the potential O2-conserving effect of the human diving response has been questioned (22, 38, 39). However, recent observations indicate that the human diving response may also have an O2-conserving effect. In three elite breath-hold divers, apneas and breath-hold dives were associated with blood lactate accumulation and a reduced O2 uptake that was not observed during apneas in untrained control subjects (9, 10). A more pronounced diving response resulted in less arterial O2 desaturation, in both resting and exercising subjects (1, 2, 25). All of these results support the view that the human diving response has an O2-conserving effect.
The alveolar gas exchange during apnea has previously been studied extensively (e.g., Refs. 16, 21, 23, 28, 29, 39). However, no study has shown whether the apneic alveolar gas exchange can be attenuated by face immersion, which has been shown for the arterial O2 desaturation (1, 2). The effects of the diving response on the alveolar gas exchange can be studied by eliciting responses of different magnitudes in the same individuals under otherwise identical experimental conditions. Therefore, to elucidate a possible conserving effect of the human diving response on the pulmonary O2 store, we recorded cardiovascular responses and alveolar gas exchange during apneas with and without face immersion in cold water in exercising men. It was hypothesized that the pulmonary O2 uptake would be reduced and plasma lactate concentration increased during apneas compared with eupneic control. Furthermore, we hypothesized that a more pronounced diving response during apneas with face immersion would be associated with a lower pulmonary O2 uptake and higher plasma lactate concentration compared with during apneas in air.
The research ethics committee at Lund University had approved the experimental protocol. After receiving a description of the procedures and an explanation of potential risks involved, all subjects gave their informed consent.
Subjects. A group of 15 healthy, male subjects volunteered for the study. Their mean age was 25 yr (range: 19-33 yr), height was 186 cm (179-195 cm), weight was 85 kg (72-108 kg), vital capacity was 6.3 liters (5.0-7.1 liters), and residual volume was 1.9 liters (1.4-2.5 liters). They were active breath-hold divers (10 subjects), training breath-hold diving at least 2 h/wk to daily, Class B according to Schagatay and Andersson (36), or recruited among subjects having performed long apneas in previous studies at our laboratory (5 subjects; Classes C and D). In addition to their breath-hold diving training, their physical training averaged 6 h/wk (0-11 h/wk). All subjects were nonsmokers, and they reported to the laboratory after at least 2 h without any heavy meal or caffeine-containing beverages.
Experimental protocol. First, after local infiltration with <1 ml of lidocaine (Xylocaine, 10 mg/ml, Astra, Södertälje, Sweden), an arterial catheter was inserted in the radial artery at the right wrist. A venous catheter was inserted in a superficial vein in the right elbow. Then, the subject mounted a cycle ergometer (Monark 829 E, Monark Exercise, Varberg, Sweden), and the probes of the noninvasive instruments were attached, after which the vital capacity was measured. A water container used for face immersion was positioned on a shelf in front of the ergometer. The subject's arms could rest on the shelf on both sides of the container. When stable cardiovascular values were observed, recordings began. After an additional 2 min of rest on the cycle ergometer, the subject performed upright, steady-state, dynamic leg exercise at a workload of 80 W for ∼50 min.
During the exercise, the subject performed a total of eight apneas alternating between apnea with face immersion and apnea with the face held above the water surface. The starting order alternated between apnea with face immersion and apnea in air among the subjects. During apneas with face immersion, the entire face was immersed with a maintained body position by just flexing the neck. Also, during the apneas in air, the neck was flexed, to maintain the body in virtually the same position during all apneas. Exercise began 5 min before the first apnea and continued until 5 min after the last apnea. The breath-holding time was kept at 40 s in both conditions, and apneas were spaced by 5 min. The water temperature was maintained between 10 and 11°C, and the ambient air temperature was 22-26°C.
Before each apnea, the subject exhaled to his residual volume through an open-circuit spirometry mouthpiece and inhaled, from a rubber bag, a volume of air equal to 80% of the individual sitting vital capacity. On command from the experimenter, the face was lifted and the apnea terminated after 40 s with a maximal expiration through the mouthpiece. After the exercise, the residual volume was determined with a nitrogen-dilution technique (35).
Measurements. Before the test, an electrocardiogram was recorded and checked for anomalies (Cardisuny 501, Fukuda ME Kogoyo, Tokyo, Japan). The vital capacity was measured with a spirometer (Micro Plus, Micro Medical, Rochester, UK). During the experiment, the heart rate was continuously recorded with a heart rate monitor (Polar Vantage, Polar Electro Oy, Kempele, Finland). The arterial blood pressure was continuously recorded with a photoplethysmometer with the cuff on the left middle finger (Finapres 2300, Ohmeda, Madison, WI). It has previously been reported that it is possible to accurately record changes in mean arterial blood pressure with the Finapres during both exercise and apnea (17, 34). The left hand was positioned at the same level relative to the heart throughout the whole experiment. A relative index of skin capillary blood flow in the left thumb was continuously recorded in 10 of the subjects with a laser-Doppler flowmeter (Advanced Laser Flowmeter 21, Advance, Tokyo, Japan). Arterial hemoglobin O2 saturation (SaO2) was continuously recorded with an earlobe pulse oximeter (Biox 3700e, Ohmeda). Respiratory flow and volume as well as expiratory O2 and CO2 fractions for determination of alveolar gas exchanges were recorded with an open-circuit spirometry system (CPX/D Cardiopulmonary Exercise System, Medical Graphics, Minneapolis, MN). Temperature, barometric pressure, and humidity were measured in the laboratory just before each experimental session. All of the cardiovascular and respiratory parameters were recorded from 2 min before exercise until 2 min after exercise.
Arterial and venous blood samples for blood-gas analysis were collected 1 min before apnea and simultaneously with the expiration ending apnea. In two of the subjects, it was not possible to collect any venous samples, and the blood gas and pH results are, therefore, based on 13 subjects. Blood gases and pH were immediately analyzed with a blood-gas analyzer (AVL Compact 3, AVL List, Graz, Austria). Arterial blood samples for analysis of the arterial plasma lactate concentration were collected 1 min before and 1 min after each apnea in all subjects. Blood samples were kept on ice and cold centrifuged within 60 min. The separated plasma was then stored at -18°C until it was analyzed at the Laboratory of Clinical Chemistry at Lund University Hospital (Modular P, Roche Diagnostics/Hitachi Instruments).
Data analysis. Average values for heart rate, mean arterial blood pressure, skin blood flow, SaO2, and alveolar gas exchanges were calculated for the period 90-30 s before each apnea and used as control. Apneic values for heart rate, mean arterial blood pressure, and skin blood flow were calculated as average values for the last 10 s of each apnea. For SaO2, an average value was calculated for the 10-s period ending with the nadir SaO2 value. The relative change from control for these apneic 10-s periods was calculated. As an estimate of the myocardial O2 demand (32), the rate-pressure product (heart rate × systolic arterial blood pressure: dimensionless parameter) during the control and apneic periods was calculated.
The O2 and CO2 exchange between the lungs and blood and the respiratory exchange ratio (RER) during apneas were calculated from the differences between volumes of O2 and CO2 in the lungs at the beginning of and at the end of apnea. The volume of O2 in the lungs at the beginning of apnea was calculated by adding the volume of O2 in the rubber bag to the volume of O2 in the residual volume (using the end-tidal fraction of O2 in the last, maximal expiration before each apnea). The same calculations were done for the volumes of CO2 and inert gases. For the determination of the lung volume at the end of apnea, it was assumed that the volume of inert gases in the lungs was constant during apnea (16). The O2 and CO2 volumes in the lungs at the end of apnea were thus calculated by using the end-apnea lung volume and the end-tidal fractions in the maximal expiration after each apnea. The time-averaged alveolar gas exchange during apneas was calculated as the change in volumes of O2 and CO2 in the lungs during apneas divided by the corresponding breath-holding time, including the time required for inhalation from and exhalation to residual volume.
For each subject, individual mean values from the three last apneas of each condition were calculated for all parameters. From the 15 individual means, a group mean ± SE was calculated. The apneic values were compared with eupneic control, and the apnea in air was compared with apnea with face immersion by using paired t-test. The level used for accepting significance was P < 0.05.
All reported data are group means ± SE. The eupneic control values for the cardiovascular and respiratory parameters during steady-state exercise did not differ before apnea in air and apnea with face immersion, except the rate-pressure product, which was higher before apnea with face immersion (Table 1).
The diving response, noticeable as a heart rate reduction from the control level, was initiated by apneas and augmented by face immersion (Fig. 1). The skin blood flow was also reduced from control during apneas, whereas the mean arterial blood pressure increased during apneas (Fig. 2). The increase in mean arterial blood pressure was augmented during apnea with face immersion compared with during apneas in air. With the changes in heart rate and arterial blood pressure, the rate-pressure product did not change during apnea in air but was reduced from the control level during apnea with face immersion (Fig. 2).
The SaO2 was reduced during apneas. From the pulse oximeter recordings, it could be calculated that the decrease from the control level during the last 10 s before the nadir SaO2 value was attenuated during apnea with face immersion, 6 ± 1%, compared with during apnea in air, 8 ± 1% (P < 0.001). Because of an averaging algorithm by the pulse oximeter, the nadir pulse oximeter values after apneas were higher (apnea in air: 88 ± 1%; apnea with face immersion: 90 ± 1%) than the SaO2 determined in the end-apneic arterial blood samples (apnea in air: 80 ± 2%; apnea with face immersion: 84 ± 1%).
The alveolar Po2, as represented by end-tidal Po2 (PetO2), was reduced from the preapnea level during apneas. In the last, maximal expiration just before apneas, the PetO2 did not differ before apnea in air and apnea with face immersion, 98 ± 2 vs. 99 ± 2 Torr. After apnea in air the PetO2 was lower, 50 ± 1 Torr, than after apnea with face immersion, 53 ± 1 Torr (P < 0.001). The O2 uptake from the lungs to the blood during the apneic period was reduced compared with the eupneic control level during both apnea in air and apnea with face immersion (Fig. 3). During apnea in air, the O2 uptake fell to 78 ± 1% of the eupneic control level and during apnea with face immersion to 75 ± 1% (P < 0.001, air vs. face immersion).
The alveolar Pco2 increased during apneas. In the last, maximal expiration just before apneas, the end-tidal Pco2 (PetCO2) did not differ before apnea in air and apnea with face immersion, 45 ± 1 vs. 44 ± 1 Torr. The PetCO2 was higher after apnea in air, 62 ± 1 Torr, than after apnea with face immersion, 61 ± 1 Torr (P < 0.001). The blood-to-lung CO2 elimination during apneas was reduced compared with the eupneic control level (Fig. 3). During apnea in air and apnea with face immersion, the CO2 elimination fell to 42 ± 1% and 41 ± 1% of the control level, respectively. With the changes in CO2 elimination and O2 uptake, the RER was reduced during apneas compared with the eupneic control level, to 0.52 ± 0.01 and 0.53 ± 0.01 during apnea in air and apnea with face immersion, respectively (P < 0.001, air vs. face immersion).
During the first 15 s of the recovery from apneas, the O2 uptake fell markedly (Fig. 4). The initial fall was followed by a transient increase in O2 uptake. This increase in O2 uptake lasted for ∼30 s, after which the O2 uptake slowly returned to the control level. There was no difference in O2 uptake, calculated as the total volume of O2 uptake during the period 15-195 s after apnea in air and apnea with face immersion. The CO2 elimination followed a similar, although slower, pattern of changes in the early recovery period.
The arterial and venous Po2, Pco2, and pH at 1 min before and at the end of apneas are shown in Tables 2, 3, 4. Notable is that the arterial Po2 is higher and arterial Pco2 lower at the end of apnea with face immersion compared with at the end of apnea in air, and that the venous Pco2 is lower than the arterial Pco2 at the end of apneas (P < 0.001).
The arterial plasma lactate concentration after apneas was higher compared with the concentration before apneas (Fig. 5). The increase in plasma lactate concentration was augmented during apnea with face immersion compared with during apnea in air.
In the present study with steady-state exercise, the alveolar gas exchange was reduced from eupneic control during apnea in air, confirming earlier observations (9, 23, 29). The alveolar gas exchange was further reduced during apnea with cold-water face immersion, showing that the lung O2 store is depleted at a slower rate during apnea with face immersion. This is probably due to the augmented diving response and associated decrease in cardiac output. The increase in plasma lactate concentration after apneas indicates that the reduced O2 uptake is associated with a shift toward more anaerobic metabolism during apnea. Thus we confirm earlier suggestions that the human diving response has an O2-conserving effect during apnea, similar to the diving response of some diving mammals.
It has previously been demonstrated that the diving response is augmented by cold-water face immersion, in both resting and exercising humans (4, 19, 39, 40), and our results are in accordance with these observations. The cardiac output is reduced in relation to the degree of the heart rate reduction during apnea, and there is a concurrent increase in systemic vascular resistance, which reduces peripheral blood flow (5, 24, 39). Furthermore, during apnea with a lung volume above the functional residual capacity and with relaxed respiratory muscles, the increased intrathoracic pressure reduces the venous return and, consequently, stroke volume and cardiac output (11, 27). This latter effect on cardiac output is most likely identical, however, in the two experimental conditions of this study. Cardiac output would, therefore, be lower during apnea with face immersion than during apnea in air because of the more pronounced heart rate reduction. The increase in mean arterial blood pressure in the present study supports earlier observations indicating that there is a pronounced arterial vasoconstriction during both apnea in air and apnea with face immersion during exercise and that the vasoconstriction is more pronounced during apnea with face immersion (39).
The myocardial O2 demand, estimated by the rate-pressure product, was reduced during apnea with face immersion, as observed earlier by Bjertnaes et al. (5). Although this is an indirect measure of the myocardial workload, the reduction is noteworthy, because it has been claimed that the increasing arterial blood pressure during apneas would offset any O2-conserving effects of the bradycardia, because of an increase in cardiac afterload (15, 22).
Because the cardiac output is reduced during apnea, the lung-to-blood O2 uptake will be reduced as a result of a reduced pulmonary blood flow, i.e., changes in alveolar gas exchange are largely determined by changes in cardiac output (23, 26, 27, 29). Therefore, the reduced O2 uptake and consequent higher alveolar and arterial Po2 and SaO2 during apnea with face immersion could partly be a result of an augmented decrease in cardiac output during apnea with face immersion compared with during apnea in air. In addition to reducing O2 uptake in the lungs, a decrease in cardiac output will reduce venous O2 stores if the tissue O2 consumption is maintained (7, 26, 27, 29). However, the reduced cardiac output will simultaneously slow the flow of deeply desaturated venous blood to the lungs, i.e., prolong the turnover time of the venous O2 stores, thereby reducing their influence on the pulmonary O2 uptake. The diving response will in this way affect the utilization of lung and tissue O2 stores during apneas, and the lung O2 store will be preserved at the expense of the venous O2 stores.
The redistributions of blood flow and venous blood volume during apneas will further reduce the O2 uptake and the venous O2 stores. The peripheral vasoconstriction will increase the arterial-to-venous difference in O2 content in the peripheral tissues (22). Furthermore, venous blood is displaced peripherally because of high intrathoracic pressure. The result is a reduced O2 content of the blood draining into the increased peripheral venous blood pool (29, 30). Simultaneously, the turnover time for these peripheral venous O2 stores will be prolonged by the peripheral vasoconstriction and peripheralization of venous blood volume (26). Hence, the reduction in venous O2 stores contributes to the observed reduction in O2 uptake from the lungs. With the decrease in cardiac output and the peripheral vasoconstriction during apnea, the lung O2 store will be preserved for use by mainly the heart and the brain. Therefore, with a pronounced diving response, the lung O2 store will be preserved for a longer period, thereby delaying the time of development of hypoxic levels that threatens the integrity of the heart and the brain during apnea. Taken together, the results thus indicate that the human diving response has an O2-conserving effect during apnea.
The transient increase in O2 uptake after 15 s of breathing in the recovery period after apneas supports the conclusion of reduced peripheral venous O2 stores. It is suggested that this increase reflects the circulatory delay of the peripheral venous O2 stores. When more desaturated venous blood enters the lung circulation after some 15-45 s of resumed breathing, the O2 uptake is increased.
It should be emphasized that, in addition to just reducing the lung-to-blood O2 uptake, the human diving response could also affect the aerobic rate of peripheral tissues by limiting the O2 delivery. If the peripheral O2 delivery is reduced, affected tissues are restricted to derive energy from aerobic metabolism with tissue O2 stores or from anaerobic energy production with lactic acid formation and breakdown of high-energy phosphates (9, 10). We provide additional support for these observations because we also found an increase in arterial plasma lactate concentration after apneas compared with the concentration before apneas. It is noteworthy that this occurred at a workload that is normally not associated with lactic acid accumulation during eupnea. Furthermore, the increase was greater during apnea with face immersion, when the diving response was augmented. This increase in arterial lactate concentration could be influenced by a number of factors. First, the samples were collected from the radial artery. The plasma lactate concentration would thus probably have been higher if the samples had been collected from the femoral vein. Second, we only analyzed the lactate concentration in blood samples collected 1 min before and 1 min after apneas. A higher peak in arterial plasma lactate concentration after apneas might have been present, before or after the collection of the postapnea samples. Furthermore, the plasma lactate concentration is not a direct measure of the lactic acid formation but rather the summation of formation, release, and clearance. Nevertheless, the increase in plasma lactate concentration supports the view of an increased anaerobic metabolic rate during apneas, correlated to the magnitude of the diving response.
When normal CO2 elimination from the body is blocked during apnea, the Pco2 values of the lungs and tissues increase (16). The blood-to-lung CO2 elimination is continuing during the beginning of apnea. Later during the apnea, CO2 delivery to the lungs almost ceases and may even be reversed (16, 21, 23, 28). The reversal in CO2 exchange is explained by an increase in alveolar and arterial Pco2 above that of the mixed-venous Pco2, caused by lung volume shrinkage. This, in turn, is attributed to the continuous lung-to-blood O2 uptake (16, 21). During apneas of relatively short durations, as in the present study, the lung and blood CO2 stores will be of major importance for storing the CO2 produced by the tissues (9). The reduced RER during apneas compared with eupneic control in the present study supports the idea that CO2 is stored in the blood and tissues during apneas. Also, it is reasonable to assume that a portion of the CO2 will be stored in the muscles where it is produced, because of reduced blood flow during apnea. Actually, the time course of CO2 elimination during the recovery period, which had a similar increase after some 15 s of breathing just as the O2 uptake, supports this assumption. The lower CO2 elimination during apnea with face immersion compared with during apnea in air may be a result of an augmented decrease in cardiac output and prolonged turnover times for tissue CO2 stores in the former condition.
The reduced Pco2 of the venous blood from the arm compared with that of the arterial blood at the end of apneas is of interest. Even considering the Haldane effect, this marked fall in Pco2 can be calculated (20) to correspond to a decrease in blood CO2 content from 52-53 ml/100 ml in the arterial blood to 49-50 ml/100 ml in the venous blood. This indicates that some CO2 is stored in the resting arm.
In conclusion, the human diving response is confirmed to have an O2-conserving effect, probably through reductions in cardiac output and peripheral blood flow during apnea. The apneic lung-to-blood O2 uptake is reduced compared with eupneic control, because of the circulatory adjustments, thereby preserving the lung O2 store for vital organs during apnea. In addition, increased plasma lactate concentration after apnea indicates that the anaerobic metabolic rate may increase as a consequence of reduced peripheral O2 delivery because of the diving response. The changes observed during apnea in air were augmented during apnea with face immersion, further substantiating the conclusions about the O2-conserving effect.
We thank Assoc. Prof. Boris Holm, Elisabeth Rünow, Camilla Olin, Lilian Qvist, and Hanna Wallin for assistance during the experiments and the volunteers for participation.
The study was supported by grants from the Crafoord Society, the Royal Physiographic Society in Lund, the Swedish National Centre for Research in Sports, AGA AB Medical Research Fund, and the Lars Hierta's Memory Foundation.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society