| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Israel Naval Medical Institute, Israel Defense Forces Medical Corps, Haifa, Israel
Submitted 25 June 2004 ; accepted in final form 10 December 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
0.001]. One year of shallow oxygen diving activity with a closed-circuit apparatus does not affect the ability to detect CO2 nor does it lead to increased CO2 retention; rather, it may even bring about a decrease in this trait. This finding suggests that acquiring experience in oxygen diving with a closed-circuit apparatus at shallow depths does not place the diver at a greater risk of central nervous system oxygen toxicity due to CO2 retention. hypercapnia; carbon dioxide detection; carbon dioxide retention; hyperbaric oxygen
) and increased risk of CNS-OT has been well established in animal models (1–3, 11, 15). Moreover, a recent study (3) has shown that even mild levels of hypercapnia can increase the risk of CNS-OT when hyperoxic gas mixtures are breathed. The correlation has also been reported in divers with a low ventilatory response to inhaled CO2 (15) and in CO2 retainers who may convulse while apparently within the safety limits for hyperoxic exposure (13). Some authors suggest that experienced divers, whether they use self-contained underwater breathing apparatus or a closed-circuit apparatus, tend to retain CO2 more than nondivers (5, 12). Kerem et al. (12) found that end-tidal PCO2 (PETCO2) was higher in ex-divers and active divers compared with a nondiving population. It still remains an open question whether the tendency to retain CO2 is an inherent trait that causes CO2-sensitive divers to give up diving, leaving only nonsensitive divers active, or whether insensitivity to CO2 increases over the years as the diver gains experience. It is important to emphasize that the majority of studies dealing with the tendency to retain CO2 were conducted on deep air diving and not diving with oxygen at shallow depths (7, 19).
At the Israel Naval Medical Institute, candidates for oxygen diving are examined in a routine physiological training procedure, during which we test them for CO2 retention and the ability to detect CO2. Our laboratory has previously shown (8) that a CO2-detection training session improves the diver's ability to detect CO2 (the training session and the test were conducted on the same day). We also found that a diver who is both a poor CO2 detector and a CO2 retainer is much more prone to CNS-OT than a diver with only one or neither of these two traits. We are unaware of any published data on the effect of a period of active oxygen diving on CO2 retention and the ability of divers to detect CO2.
In the present study, we tested the ability to detect CO2 and the retention trait in novice oxygen divers and after 1 year of extensive diving activity with a closed-circuit apparatus.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
150 dives in the course of the year) using closed-circuit apparatus at depths of 3–5 m, the divers performed the second test. This was termed the experienced oxygen diving phase (ED). Thirty-six percent of subjects in the ND phase and 41% in the ED phase were smokers. The mean (with SD in parentheses) age, height, weight, vital capacity, and forced expiratory volume in 1 s at the time of the first and second tests are presented in Table 1. All physical examinations were normal.
|
In the latter situation, the tender could control the level of CO2 in the rebreathed gas by operating a fine-control stopcock (screened from the subject) to direct part of the expirate through a soda lime CO2-absorption unit. The absorption unit was immersed in a cool- water tank (18°C) to avoid overheating of the rebreathed gas. From the bag, the gas passed through a three-way breathing valve (Hans Rudolph) and then back to the mouthpiece. In the open-circuit mode, the valve could be switched via demand regulators (Drager, Lubeck, Germany) to a high-pressure source of either air or oxygen. The breathing resistance of the closed circuit was adjusted to simulate that in the operational rebreather (inspiratory and expiratory pressures of +5 and –5 cmH2O, respectively, at a ventilatory rate of 30 l/min).
Gas samples were drawn from the first and second sampling ports, respectively, for either breath-by-breath recording of inspired and end-tidal gas tensions or for continuous recording of mixed expired tensions for the computation of gas exchange. Oxygen and CO2 concentrations were measured using gas analyzers (Servomex model 571, Sussex, UK, and CD-3A, Ametek, Applied Electrochemistry, Pittsburgh, PA, respectively). The gas analyzers were calibrated before each test with premixed gases of known composition. The sampled gas was directed back into the system. The temperature of the expired gas was continuously measured by a thermistor (YSI model 4002, Yellow Springs Instruments, Yellow Springs, OH). After amplification (Gould, Cleveland, OH), all signals were fed into a computer for online analysis using in-house software. Sampled data were averaged every 15 s. Gas was sampled from the mixing box to calculate oxygen consumption and CO2 production (
CO2). Samples were taken from the mouthpiece to calculate inspired PO2, PI, end-tidal PO2, and PETCO2. The temperature of the expired gas, expired minute ventilation (E), tidal volume, and breathing frequency were calculated at both sampling locations.
The individual's ability to detect hypercarbia was determined using an electric scale board with a matrix of push buttons related to five subjective symptoms (hyperpnea, air hunger, headache, dizziness, and a warm sensation), each of which could be scored according to five degrees of intensity (very slight, slight, moderate, severe, and extreme). A push on any of the buttons would flash a corresponding light on a similar electric scale board on the tender's side of the screen and would also be recorded by the computer. Subjects were informed that they might experience any or all of the symptoms appearing on the board.
Experimental Protocol Before the first and second tests, the subjects received an oral lecture on the role of CO2 in closed-circuit diving, the signs and symptoms of hypercarbia, and the nature of the test.
Training session.
The training session was performed before the first test. The subject was fitted with a nose clip and was instructed to start pedaling at a rate of 80 W and to continue for 5 min while breathing air in the open-circuit mode. The trainee could watch his mouthpiece CO2 fluctuations as well as his expiratory flow signal on the computer screen. Because we chose to keep the rate of CO2 production between 800 and 1,000 ml/min STPD (we kept track of this value by sampling the gases from the mixing box), the ergometer workload was adjusted to place the CO2 within the desired range. The breathing gas was then switched to 100% oxygen, and when the expired fraction of oxygen exceeded 75%, the system was switched to the closed-circuit mode with full CO2 absorption and gases were sampled from the mouthpiece. The expired fraction of oxygen was maintained above 75% throughout rebreathing by the demand valve that supplied the breathing circuit with 100% oxygen. After a few minutes to adjust to oxygen rebreathing, the subject was instructed to stop exercising and observe or sense his recovery from exercise hyperpnea with no CO2 in the inspirate.
Exercise was then resumed, and after another few minutes, the inspiratory fraction of CO2 was allowed to increase by partial bypassing of the absorbent. The CO2 level in the inspired gas was arbitrarily cycled within a range of 0–5.6 kPa (0–42 Torr), while the subject was provided with a digital display of the inspired CO2 concentration and also informed verbally, until he could sense the presence of CO2 in his inspired gas without hints and signal accordingly. The training session lasted between 20 and 30 min, with each subject being allowed as long as he needed to assimilate the information and thus learn to detect CO2. The subject was repeatedly asked to focus on the type and intensity of the symptoms he was experiencing. In the last training stage, the PI was gradually increased (at a rate of 1 kPa/min, 7.5 Torr/min) to 5.6 kPa (42 Torr) and maintained at that level for 1 min. This value was chosen as being high enough to stimulate most subjects without inducing severe symptoms of CO2 toxicity in that time span.
At that point, the subject was again instructed to stop exercising and compare his recovery from hyperpnea while PI was maintained at 5.6 kPa (42 Torr) for a further 1 min. The breathing source was then abruptly changed to air in the open mode so that the inspired CO2 immediately returned to normal levels. The session ended a few minutes later when the subject signaled that his rate and depth of breathing had returned to resting levels.
All subjects were instructed to make use of their training experience to try to detect CO2 in the subsequent first test.
Test session (ND).
The test session was started 15–30 min after training, when subjects' PETCO2 had returned to baseline (pretraining) levels and any of the symptoms (such as headache or dizziness) that had occurred during the training session had disappeared. It comprised the same procedure as the training session, but the gradual elevation of PI began immediately after the oxygen rebreathing. The CO2 level in the inspired gas was arbitrarily cycled two or three times within a range of 0–5.6 kPa. Subjects were not informed of the inspired fraction of CO2 other than that it would at some point start to rise and that they should indicate on the electric board the moment they detected it. This first detection was unsolicited. From then on, subjects were asked to grade their sensations every 30 s for as long as the rebreathing continued. The subject was informed of any false-positive detections and was then instructed to concentrate and only signal when he was certain of a new sensation. The number of false-positive events for all subjects on their first test was 10%. False detections (at apparently too low a level of inspired CO2 for detection) resulted in switching back and forth from pure oxygen to the CO2 level in question. The minimum level of inspired CO2 for which we considered a response to be a true detection was found for every subject individually according to his detection repeatability during the test session.
In our efforts to improve detection in poor detectors, we found previously that repeated training and tests during the week following the initial test did not change the detection threshold (8). On these grounds, we believe that after the first training session the diver exhibits his innate threshold, and a single measurement will therefore suffice.
Test session after 1 year (ED).
After a short reminder, subjects performed the same test as they had in the ND phase, with the CO2 once again being kept within the range of 800–1,000 ml/min but without any training session beforehand. This was to fulfill one of the goals of our experiment, namely to assess whether the effect of the initial training was maintained over 1 year of diving activity.
Data Analysis
The threshold for hypercarbia detection (the lowest PCO2 detected repeatedly), the extent of CO2 retention [PETCO2 at a PI of 5.6 kPa (42 Torr)] and ventilatory response to CO2 (assessed from the slope of the linear regression of E against PETCO2 during CO2 elevation) were determined for each subject in the ND and ED phases. The ventilatory response to CO2 does not represent the complete response, which required more time at each PI. However, it can be used to compare the ND and ED phases.
Statistical analysis.
All group values are reported as means (with SD in parentheses). Significant differences in detection threshold and CO2 retention between the ND and ED phases were determined using a paired t-test. The differences in the distribution of PI and PETCO2 at detection thresholds and the differences in the distribution of PETCO2 while pure oxygen or pure oxygen with 5.6 kPa (42 Torr) CO2 was breathed in ND and ED were determined using the F-test. A nonparametric Kruskal-Wallis test was used to compare differences in the intensity of each subjective symptom for ascending values of PI in the ND and ED phases. A P value of <0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
at the detection threshold in the ND and ED phases was not significantly different [2.9 kPa (SD 0.7), 21 Torr (SD 5) and 3.0 kPa (SD 0.9), 22 Torr (SD 7), respectively]. A subject whose detection threshold was above 3.6 kPa (27 Torr) was defined as a "poor detector" (8). Only two subjects in the ND phase were poor detectors, and six subjects were found to be poor detectors in the ED phase. Moreover, the distribution of PI detection thresholds in ND and ED was not significant (F-test, P = 0.07; Fig. 1). Mean PETCO2 at the detection threshold in the ND and ED phases was not significantly different [5.9 kPa (SD 0.7), 39 Torr (SD 5) and 6.0 kPa (SD 0.7), 44 Torr (SD 5), respectively]. The distribution of PETCO2 at the detection thresholds in ND and ED is shown in Fig. 2.
|
|
E normalized to body weight at the point of CO2 detection in the ND and ED phases was significantly different [0.45 (SD 0.15) and 0.39 l·min–1·kg–1 (SD 0.11), respectively; P < 0.03].
Subjective Sensation
The frequency with which the five CO2-related symptoms were indicated and their perceived intensity as a function of PI are shown in Fig. 3. PI detection threshold was reported in the ND phase mainly by the sensation of hyperpnea (70%), whereas only 40% of subjects in the ED phase reported this sensation (P < 0.003). In the ED phase, PI detection threshold was reported mainly by a warm sensation, whereas no subject in the ND phase reported a warm sensation at his PI detection threshold (P < 0.02). Throughout the whole of the test period, a significant difference was found between the ND and ED phases for reporting of two symptoms: hyperpnea was reported more in the ND phase (P < 0.003), whereas a warm sensation was reported more in the ED phase (P < 0.02). There was no significant difference in reporting of dizziness, air hunger, or headache between the ND and ED phases. The symptoms related to ventilation (hyperpnea and air hunger) became more intense as the level of CO2 inhalation increased.
|
|
E normalized to body weight while breathing pure oxygen with 5.6 kPa (42 Torr) CO2 was significantly different in the ND and ED phases [0.79 (SD 0.22) and 0.62 l·min–1·kg–1 (SD 0.14), respectively; P < 0.001]. We also evaluated the effect of 1 year of diving activity on the ventilatory response to CO2 (
E/PETCO2; in l·min–1·kPa–1) obtained by linear regression analysis and found that 1 year of diving with oxygen at shallow depths has no effect on this parameter. The slopes for the ND and ED phases were 1.56 l·min–1·kPa–1 (SD 0.67) [r2 = 0.83 (SD 0.13)] and 1.25 l·min–1·kPa–1 (SD 0.47) [r2 = 0.87 (SD 0.10)], respectively (P > 0.05). |
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Hypercarbia Detection Threshold
It was found in the present study that 1 year of oxygen diving activity with closed-circuit apparatus does not alter a subject's ability to detect CO2 compared with the ND phase. Our laboratory previously reported that a training session before the test improved a diver's ability to detect CO2 (8). Then, mean PI at the detection threshold in 231 trained subjects was 2.9 kPa (SD 0.7). In the present study, mean PI at the detection threshold in the 40 novice oxygen divers [3.0 kPa (SD 0.7)] was close to that value. All the novice divers had performed the detection test 15–30 min after CO2 detection training. Although no training session preceded the second test, this did not have an adverse effect on the average ability to detect CO2. The ability to detect CO2 did not worsen dramatically in any of the experimented subjects, except for two whose CO2 detection threshold increased from 3.2 kPa in the ND phase to 4.2 and 4.8 kPa in the ED phase. However, these two subjects did not become CO2 retainers in the ED phase, and so their safety was probably not compromised.
In our laboratory's previous study (8), we wondered how long the training effect might persist. We may now postulate that the ability to detect CO2 is maintained for at least 1 year after training. It is possible that the experience the divers gained during that year might unintentionally have included the occasional need to sense increased CO2. This could have helped maintain their ability to detect CO2.
Subjective Sensation
Our laboratory demonstrated previously (8) that training made subjects more attuned to their sensations, and they learned to recognize them at a lower PI. This is in contrast with Pisculli et al. (18), who did not find improvement in CO2 detection after hypercarbia training. This conflict may be explained by the different methods employed in the two studies. Here and in our laboratory's previous study (8), the training session lasted for an unlimited time until the subjects learned to detect CO2. In the study of Pisculli et al. (18), the training session was restricted to only a few minutes, with 3.7% as the highest CO2. Perhaps this time limitation and the maximum CO2 concentration were not sufficient for the divers to improve their CO2 detection. Further comparison is difficult due to the limited information available in the abstract describing this study of 12 subjects, compared with our previous study of 213 divers.
In the present study, we showed once again that the respiratory symptom hyperpnea was most often reported at the time of PI elevation in the first test (Fig. 3). However, in the second test a year later, the nonrespiratory symptom of a warm sensation plays an important role in the perception of PI at the detection threshold. The shift from the sensation of hyperpnea in the ND phase to a warm sensation in the ED phase might stem from the significant reduction in E at the CO2 detection point that was found in the ED phase. This might be due to a reduction in the ventilatory response to CO2, although E/PETCO2 was not significantly different.
These two categories of sensations, hyperpnea and a warm sensation, are not conditioned by the same drive. Whereas respiratory sensations arise from chemoreceptor and proprioceptive input and are affected by hyperoxia, gas density (17), pH changes, and exercise, the nonrespiratory symptoms may be partly related to the vasoactive effect of CO2. During diving, it may be difficult to distinguish whether the sensation of hyperpnea stems from exercise or inspired CO2. Therefore, nonrespiratory signs may be of advantage to the experienced oxygen diver engaged in underwater activity. Fothergill et al. (9) also suggested that nonrespiratory symptoms may be of particular importance to divers for detecting hypercarbia during underwater work. However, the subjects in that study performed head-out immersion. There is a chance that sweating and flushing (nonrespiratory symptoms) were sensed simply because the subject's head was out of the water. During diving, the whole body is submerged so that, when diving in cool water, the diver would be less likely to be aware of a warm sensation.
From our experience, with more than a dozen records of an accidental rise in CO2 during actual diving, the symptoms described by the divers are similar to those reported during the CO2 detection test (excluding a warm sensation). Therefore, the dry test is good training for underwater activity. Cool water may neutralize the warm sensation induced by CO2, and this symptom may therefore be of no significance in actual diving.
CO2 Retention
Most of the studies dealing with CO2 retention and diving claim that divers (mainly open-circuit air and breath-hold) tend to retain CO2 (5, 6, 12). However, the divers in the present study were engaged only in oxygen diving with closed-circuit apparatus, which is limited to shallow water to prevent CNS-OT. In the present study, we addressed the question of whether CO2 retention is inherent or acquired as the diver gains experience with closed-circuit apparatus on an oxygen diving profile. The divers in the present study did not tend to retain CO2 to a greater degree after 1 year's experience using this apparatus, whether they breathed air or inhaled 5.6 kPa CO2 in oxygen. On the contrary, their tendency to retain CO2 was significantly reduced. We may, therefore, conclude that, for this kind of diving profile, CO2 retention is not a trait that is acquired as experience is gained. However, for other diving profiles, the question remains as to whether CO2 retention is inherent or acquired during diving. Although the tendency to retain CO2 decreased significantly after 1 year of extensive diving, E was also reduced. We speculate that the reduction in E may stem from better energetic efficiency of the divers in the ED phase. However, we are unable to prove this on the basis of the data obtained from the present study because, although we kept the CO2 at a fixed level, we did not record the subjects' mechanical workloads during any phase of the investigation.
Despite our finding that the tendency to retain CO2 significantly decreased, we nevertheless continue to discourage CO2 retainers who are also poor CO2 detectors from taking part in closed-circuit oxygen diving because of the high risk in the year after diagnosis.
The mean PETCO2 when breathing pure oxygen was 5.9 kPa (SD 0.7) [39 Torr (SD 5)] and 6.0 kPa (SD 0.7) [44 Torr (SD 5)] in the ND and ED phases, respectively. The PETCO2 obtained while breathing air or pure oxygen is not a good indicator as to which diver might be a CO2 retainer when actually breathing moderate levels of CO2 (8). The major risk of being a CO2 retainer [PETCO2 > 9.2 kPa (69 Torr) at a PI of 5.6 kPa (42 Torr)] (8) is the possibility of being involved in an oxygen toxicity incident when breathing hyperoxic gas mixtures. One mechanism of this increased susceptibility to hyperbaric hyperoxia is CO2-induced cerebral vasodilatation, which increases the blood flow through neural tissues and, in turn, increases the transfer of oxygen to these tissues. Even at moderate hyperbaric pressure, CO2 retention may be presumed to increase the production of nitric oxide, which also results in cerebral vasodilatation (6). Another mechanism for increased susceptibility to hyperbaric hyperoxia is the possibility of increased production of reactive oxygen species in the presence of H+ and CO2 (7). However, the depth limitations on the oxygen divers in the present study preclude this possibility.
Our finding of a significant reduction in the tendency to retain CO2 after 1 year differs from those of other reports in the literature. In those studies, experienced air and breath-hold divers typically exhibited higher levels of PETCO2, probably due to a reduction in their ventilatory chemosensitivity to inspired CO2 (6, 12). Schaefer (19) found blunted responses to hypoxia and hypercapnia in submarine escape training instructors during a period of intensive underwater work compared with a 3-mo layoff period. Delapille et al. (7) also found lower sensitivity of the respiratory center to hypercapnia for trained breath-hold divers compared with nondivers. Air and breath-hold diving are activities mainly practiced in deep water, which differs from the diving activity in the present study. Thus these conflicting observations could be attributed to the diving profile.
Diving at shallow depths has little effect on the work of breathing. The peak inspiratory and expiratory pressure in the diver's rebreather does not exceed 10 cmH2O during moderate exercise, which implies that there is low resistance. However, there is an increase in the work of breathing in deep diving (16). At depth, the higher density of the inspired gas leads to an increase in the work of breathing and a subsequent reduction in ventilation (6). We therefore suggest that optimizing ventilation on deep dives saves on some of the work of breathing. This reduction in the work of breathing may result in an increase in arterial PCO2. Hypoventilation when a dense gas is breathed reduces the work of breathing without affecting oxygen transport. Therefore, relinquishing a certain amount of CO2 regulation might be beneficial in reducing the physiological cost of the work of breathing. In contrast, in shallow-water oxygen diving, due to the low energy cost of breathing, no adjustment of the arterial PCO2-E balance is required. Thus CO2 retention may develop in deep- but not shallow-water dives.
In addition to depth, another parameter that is different in our study is the inspired gas (100% oxygen vs. air, i.e., 79% nitrogen). However, the difference in density between oxygen and nitrogen is insignificant.
It might be expected that the tendency to retain CO2 will be greater in hyperoxia than in hypoxia because the peripheral chemoreceptors are not stimulated. However, the oxygen chemoreceptors might undergo adaptation, which may be another reason why the divers in the present study did not became greater CO2 retainers after 1 year of oxygen diving activity.
In summary, 1 year of oxygen diving using closed-circuit apparatus at shallow depths does not alter the tendency to retain CO2 nor does it induce changes in the PI detection threshold. Longer follow-up is required to examine whether this trait is preserved further into the diver's career.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The opinions and assertions contained herein are the private ones of the authors and are not to be construed as official or as reflecting the views of the Israel Naval Medical Institute.
|
FOOTNOTES |
|---|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
