Cumulative effects of repeated exposure to po 2 = 200 kpa (2 atm)

UHM 2014, Vol. 41, No. 4 Repeated Chamber po 2 = 200 kpa Cumulative effects of repeated exposure to po 2 = 200 kpa (2 atm) Barbara E. Shykoff Navy Experimental Diving Unit, Panama City, Florida USA EMAIL: Dr. Barbara Shykoff barbara.shykoff.ca@navy.mil ABSTRACT Even asymptomatic exposures to elevated oxygen partial pressure (po 2 ) can influence subsequent exposures. Dry chamber dives of three hours duration at po 2 of 200 kpa were conducted to examine cumulative effects. Experiments were single (n=27), or paired exposures with surface intervals (SIs) 15 to 17 hours (n=30), six hours (n=33), or three hours (n=36). Flow-volume loops, diffusing capacity, and symptoms were recorded before and after exposures. Immediately after surfacing from second exposures, some significant (p<0.05) mean changes from baseline in pulmonary function indices occurred for all SIs and persisted for two days for three-hour SIs and for one day for 15-hour SIs. Incidences of symptoms were 15% immediately after one exposure and 28%, 38% and 31% immediately after a second exposure following 15-, six-, or three-hour SIs, respectively. Incidences of changes in pulmonary function indices (ΔPF) were 5%, 11%, 12% and 14% for single exposures or two with 15-, six-, or three-hour SIs, respectively. Two days following the second exposure, resolution of symptoms was incomplete after six- or 15-hour SIs, as was resolution of ΔPF after 15-hour SIs. The incidences indicate non-linear superposition of effects of a first and second exposure, and are not readily explained by delayed-onset injury. INTRODUCTION Any exposure to elevated oxygen partial pressure, even one that is asymptomatic, causes oxygen stress which takes time to resolve. Although slow accumulation may not be evident without many exposures, a new dive or chamber exposure will be complicated by a previous exposure unless the surface interval is long enough to ensure that the effects have resolved completely. Others have seen evidence of cumulative effects after hyperbaric oxygen treatments. In one study, after daily standard hyperbaric oxygen (HBO 2 ) treatments of 90 minutes at oxygen partial pressure (po 2 ) 240 kpa with air breaks every 30 minutes, no significant changes in pulmonary function were found after seven days, but after 14 and 21 days, progressive decreases were noted in parameters of forced expiration, and some patients developed coughs [1]. In another study with a different sequence of air breaks, 10 days of 90-minute HBO 2 treatments at 250 kpa provoked changes in forced expiratory flow indices associated with small airway function [2]. However, a third study with HBO 2 treatments at 240 kpa continuously for 90 minutes a day, five days per week for six weeks, did not provoke changes in measured pulmonary function indices [3]. Knowledge of recovery time and of the residual pulmonary effects during recovery after exposures to elevated po 2 would help in planning multiple exposures while simultaneously controlling for the risk of pulmonary oxygen injury. No current models permit the estimation of recovery times after exposure to more than one po 2. The unit pulmonary toxicity dose (UPTD) model, which tries to permit comparison of exposures to different po 2, does not include provisions for recovery and does not match data very well at the po 2 levels used therapeutically [4]. We have previously described recovery from inwater dives with po 2 between 130 and 140 kpa [5], but those results are specific to that range of oxygen partial pressure. This study addressed recovery after exposures to po 2 = 200 kpa, particularly in connection with any cumulative effects. Residual effects for earlier dives were assessed by examination of effects after second dives following different surface intervals (SIs). For this study, participants were exposed in a dry hyperbaric chamber for a low risk of central nervous system (CNS) oxygen toxicity; after more than 20,000 hyperbaric treatments at one facility, seizure rate was 0.03% [6]. Pulmonary oxygen toxicity result- Copyright 2014 Undersea & Hyperbaric Medical Society, Inc. 291

ing from exposures to 200 kpa of oxygen and its time course for recovery, particularly in connection with any cumulative effects, are described. METHODS Participants were active-duty Navy divers from Navy Experimental Diving Unit (NEDU), from other commands in the geographic vicinity of NEDU, and from NEDU reserves Great Lakes. A number participated in more than one dive series but did not dive at all within two weeks prior to any exposure. All were medically qualified to dive, in good health and within body composition standards for the diving Navy; none suffered from asthma, emphysema or chronic bronchitis. Diver characteristics, including known smoking history, reported respiratory allergy and relevant medications, are summarized in Tables 1 a-c. The protocol was approved by the Navy Experimental Diving Unit Institutional Review Board, and divers gave written informed consent. The experiments were single exposures or pairs of exposures. Surface intervals (SIs) were chosen to be equal to the dive duration, twice the dive duration, and five to six times the dive duration. Exposure times were selected to be long enough to provoke some symptoms and some measurable decrease in pulmonary function. Preliminary dives were single two-hour exposures, and subsequent exposures were three hours in length. The exposure and testing scheme for the three-hour dives is shown in Figure 1. Pulmonary function was measured within the week before diving, immediately before divers entered the chamber for each exposure, immediately after surfacing, two hours after the end of UHM 2014, Vol. 41, No. 4 Repeated Chamber po 2 = 200 kpa single exposures, and on the two days following each dive evolution. The instruments used, CPL raptor and CPL pf (nspire Medical, Longmont, Colorado), measure volume and flow using a spirometer rather than a pneumotachograph. Table 1a. Diver characteristics by dive set: median (range) Dive Number Age (yrs) Height (cm) Weight (kg) one 2-hour 12 37 (20 62) 175 (170 193) 93 (75 116) one 3-hour 27 36 (20 51) 175 (145 191) 84 (67 113) two 3-hour 36 39 (20 62) 178 (168 188) 86 (67 109) SI=3 hours two 3-hour 33 35 (19 56) 178 (168 188) 91 (71 111) SI=6 hours two 3-hour 30 37 (20 50) 178 (168 191) 91 (71 111) w/ overnight SI (15-hr, n=18; 17-hr, n=12) Table 1b. Smoking history by dive set Dive Current Former smokers Never smokers non-smoking years smoked < 5 5-10 >10 one 2-hour 3 (a,b, c) 1 (d) 0 1 7 one 3-hour 4 (b, 3c) 1 (c x 20 yrs) 0 3 19 two 3-hour 4 (2a, 2b) 3 (2c, a x 1yr) 2c 4 23 SI=3 hours two 3-hour 1 (c) 3 (2c, a x 2 5 23 SI=6 hours 3 months) two 3-hour 3 (2a, b) 3 (3c) 0 3 21 overnight SI a =1 pack/day; b = ½ pack/day; c < ½ pack/day; d = unknown Table 1c. Allergy and medication by dive set Dive Reported Number on routine or frequent medication respiratory NSAIDS Anti- Antiallergies * hypertensive dyslipidemia one 2-hour 5 2 2 0 one 3-hour 4 4 1 2 two 3-hour 4 5 1 1 SI=3 hours two 3-hour 3 1 0 1 SI=6 hours two 3-hour 6 5 3 0 overnight SI * pollens (including those out of season during the study), cats, dogs, dust proton pump inhibitors, eye drops, and vitamin supplements are not listed Forced flow-volume loops were measured at every session. Single breath diffusing capacity of the lungs for carbon monoxide (D L CO) was recorded in the week before diving, after the last exposure of the day, and on the two follow-up days (Figure 1). The goal for every pulmonary function measurement session was 292

UHM 2014, Vol. 41, No. 4 Repeated CHaMbeR po 2 = 200 kpa Figure 1. Time scheme of three-hour exposures and measurement times relative to the fi rst chamber dive Gray blocks indicate three-hour chamber exposures. Distance on the figure is not proportional to time. Legend: SI: surface interval; F-V: fl ow-volume measurements; DLCO: Measurement of diffusing capacity of the lung for carbon monoxide; Pre 1, post 1 are pre- and post-dive 1 of the pair, etc. 293

UHM 2014, Vol. 41, No. 4 Repeated Chamber po 2 = 200 kpa three measurements that satisfied the criteria set by the American Thoracic Society (ATS) [7], but no more than eight attempts were made for flow-volume loops, and no more than three for D L CO. Valid test results were averaged after acceptability was confirmed by both numerical criteria and by visual assessment of the reproducibility and acceptability of the curves. Baseline values of D L CO came from the measurements made in the week before diving, and those for flow-volume parameters were the mean of usually six measurements, three from the week before diving and three from immediately before the first exposure of the series. Differences from baseline were assessed. Flow-volume variables of interest were forced vital capacity (FVC), forced expired volume in one second (FEV 1 ), and forced expired flow between 25% and 75% of volume expired (FEF 25 75 ). D L CO was corrected for hemoglobin and carboxyhemoglobin concentrations [8]. A one-liter gas sample was used for calculations, and sample timing was chosen to ensure measurement during steady-state portions of the gas concentration record. Divers were asked to fill out a symptoms questionnaire every time pulmonary function was measured. Respiratory symptoms of interest were cough, inspiratory burning, chest tightness and dyspnea. To rule out hyperoxic myopia, visual refraction was checked (Humphrey model 599, Carl Zeiss Meditec, Dublin, Calif.) at any session that included D L CO. Repeated measures ANOVA was used individually to assess the effects of time on average changes from baseline of FVC, FEV 1, FEF 25 75 and D L CO; variables were treated as independent, and thus no correction for multiple comparisons was made. Individual comparisons within subjects were made using orthogonal contrasts which, as independent, uncorrelated measures, do not increase the Type I error rate. Additionally, pulmonary function measurements related to a first exposure were combined across different SI series when the measurements were made at the same time after surfacing (Figure 1). The effects of time after a single exposure were assessed using one-way ANOVA for each variable individually with Bonferroni corrections for individual time-point comparisons. Results were considered significant if p<0.05. Individuals may show significant changes in pulmonary function even when mean values for the measurement group do not differ from baseline. Any measure of pulmonary function was considered to show a decrement from baseline (ΔPF) if it decreased by more than the established normal variability: 7.7% for FVC, 8.4% for FEV 1, 17% for FEF 25-75, and 14.2% for D L CO [9]. A subject who had one or more ΔPF at any measurement session or who reported one or more symptoms was considered to manifest pulmonary oxygen toxicity. Incidences of symptoms and of ΔPF at different times were compared using Fisher s exact test. Again, p<0.05 was considered significant. RESULTS Mean changes No significant changes in the selected indices of pulmonary function were seen immediately after single two-hour exposures, though mean FVC and FEV 1 were depressed on the first day after the two-hour exposures (Figure 2 a,b). However, as none of 12 subjects reported symptoms or showed measurable ΔPF, two-hour exposures with po 2 = 200 kpa were considered too short for estimation of residual effects. Further work used three-hour exposures. Immediately after single three-hour exposures, mean FVC, FEV 1, and FEF 25-75 were slightly but significantly depressed (Figure 2a-c). Mean FVC remained low (p < 0.02) two hours after surfacing (n = 63). On the average, no significant differences from baseline remained five or 15 to 17 hours post diving, or on either of the days (Day +1 or Day +2) following a single threehour exposure, and changes in DLCO were not significant. As noted below, the changes from baseline of FVC and FEV 1 after one dive were significant when all threehour first dives and single dives were combined (n = 127). However, they were not significant for the sixor 17-hour SI dive series taken alone (n = 33, n = 12, respectively). Conversely, the change from baseline of FEF 25-75 after a single dive was not significant when all first and single exposures were combined (n = 127), but was significant when the series with three- and 15-hour SIs (n = 36, n = 18, respectively) were considered individually. Facing page Figure 2. Changes from baseline after single exposures, means and standard errors: a) FVC; b) FEV 1 ; c) FEF 25-75 ; d) DLCO. Dashed line: two-hour exposure (n=12), solid line: three-hour exposure, including measurements immediately after all first dives and before second dives from all dive series (n=127, 63, 33, 30, 26 and 26 for the times immediate, +2 hours, +5 hours, overnight, Day +1 and Day +2, respectively). Overnight is the combination of measurements before second dives of the 15- and 18-hour surface interval series. Significant (p<0.05) differences from baseline are marked: *: two-hour dive : three-hour dive. 294

UHM 2014, Vol. 41, No. 4 Repeated CHaMbeR po 2 = 200 kpa FIGURE 2 FIGURE 3 Above Figure 3. Changes from baseline (means and standard errors) after pairs of exposures to po 2 = 200 kpa: a) three-hour surface interval (n=36), b) six-hour surface interval (n=33), c) overnight surface interval (n=30). ΔFVC: ΔFEV 1 : n ΔFEF 25-75 : ΔD L CO: In panel c, the closed symbols and black lines show values for a 15-hour SI (n=18), while open symbols and gray lines indicate those for a 17-hour SI (n=12). Signifi cant (p<0.05) differences from baseline are marked at the top of each panel: * = FVC = FEF 25-75 = FEV 1 = D L CO. 295

UHM 2014, Vol. 41, No. 4 Repeated CHaMbeR po 2 = 200 kpa For the three-hour SI dive series (Figure 3a), mean FVC, FeV 1 and FeF 25-75 after the first dive differed from baseline (p<0.01, p<0.01, p<0.02, respectively) (Figures 3a-3c), but before the second dive, only FVC remained low (p<0.04). after the second dive, mean FVC, FeV 1 and FeF 25-75 were depressed (p<0.01). The deficits from baseline after the second dive were not statistically different from those immediately after the first dive, but the depressed values persisted to day +1 (p<0.01). on day +2, although both FeV 1 and FeF 25-75 still differed from baseline (p<0.02), they showed recovery, at that time differing also from the lower value immediately after the second dive (FeV 1 : p<0.01, FeF 25-75 : p<0.03). For the six-hour SI series of exposures (Figure 3b) none of the values differed from baseline after the first dive. Mean FVC did not differ from baseline at any measurement, and FeV 1 and FeF 25-75 were depressed only immediately after the second dive (both p<0.02). Mean dlco was depressed on day +2 (p<0.05). When all overnight surface interval dives (both 15- and 17-hour) were considered as one dive series (n=30), FVC, FeV 1 and FeF 25-75 differed from baseline after the first dive (p<0.01, p<0.01, p<0.02, respectively). FVC and FeV1 did not differ from baseline at any other measurement but FeF 25-75 was depressed after the second dive and on day +1. If only data from the 15-hour SI group were considered (Figure 3c), FeV 1 was below baseline after the second dive (p<0.03) and on day +1 (p<0.2), as was FeF 25-75 immediately after the second exposure and on both follow-up days (p<0.02). None of the indices was significantly different from baseline with the 17-hour SI (Figure 3c), probably because of the small number (n=12) of participants. Incidences of pulmonary oxygen toxicity the incidence of pulmonary oxygen toxicity at a measurement time is given by the fraction of the divers who have any symptoms or ΔPF at that point. Values obtained immediately after the second exposures are shown in Figures 4a (symptoms) and 4b (ΔPF) along with Agresti Coull 95% binomial confidence intervals. Symptoms and ΔPF did not generally coincide, and symptoms were approximately three times as frequent as ΔPF. The numbers of participants who reported symptoms after each evolution and the times at which they reported them are shown in Figure 5. FIGURE 4 a b Figure 4. Probability of noticeable pulmonary oxygen toxicity immediately after exposure: a) symptoms and b) ΔPF. Measured incidences and Agresti Coull binomial 95% confi dence intervals are shown. Numerals on the fi gures indicate the number of measurements for each condition. Time course of recovery, single dive Figure 6a shows the change with time of the probabilities of symptoms or ΔPF after single three-hour exposures to po 2 of 200 kpa. Immediately after exposure, the incidence of symptoms was 15%, significantly different from zero (p<0.01) when all 127 exposures were considered. It was no longer different from zero after two hours when data from 63 exposures were available. However, incidence was again significantly different from zero after five hours (n=33), then not different from zero at 14 hours (n=30) or on days +1 and +2 (n=26). 296

UHM 2014, Vol. 41, No. 4 Repeated CHaMbeR po 2 = 200 kpa Figure 5. Numbers of subjects reporting symptoms at any time. The solid bar indicates the fi rst report, while the shaded bar indicates that the participant had reported symptoms previously, though not necessarily without relief between reports. Distances on the x-axis are not scaled by number of hours. Single three-hour dives, n=27; overnight surface interval (SI), n=30; six-hour SI, n=33; three-hour SI, n=36. No symptoms were reported at any time after two-hour single exposures, n=12 FIGURE 5 The 5% incidence of ΔPF immediately after a single exposure became significant when all 127 exposures were considered. although it was essentially unchanged numerically two and five hours after surfacing, it was no longer different from zero with the smaller number of measurements (n = 63 and 33, respectively). Time course of recovery, dive pairs the incidence of symptoms (Figure 6b) after second three-hour exposures differed significantly from zero immediately after diving and on day +1 for all SIs, and on day +2 for six- and 15-hour SIs. Immediately after second dives with three- or six-hour SIs the incidence of symptoms was greater than that after single threehour dives. With the six-hour SI it remained significantly greater on day +1 than it had been after one dive. The incidences of ΔPFs (Figure 6c) were significantly greater than zero immediately after surfacing from second exposures with three-hour SI, on day +1 after six-hour SI, and at all times assessed after 15-hour SI. Any ΔPF immediately after diving was in a flowvolume parameter; d l Co was depressed only on days +1 and +2. Other effects No changes in visual refraction or acuity were observed across these exposures. one subject aborted his oxygen exposure with symptoms of hypercapnia. because his hood had been overinflating before the onset of symptoms, he had adjusted the flow to try to correct the problem. He reported feeling hot and flushed, was aware that his heart rate was high, and had difficulty concentrating on the card game he was playing. Symptoms abated when he removed the hood to breathe chamber air. In a followup unmanned test with a breathing simulator that exhales Co 2, inspired Co 2 in a hood on a manikin head was found to be elevated under all conditions, highly variable, and a strong function of inlet oxygen flow [10]. 297

UHM 2014, Vol. 41, No. 4 Repeated CHaMbeR po 2 = 200 kpa FIGURE 6 Figure 6. Measured incidences of pulmonary oxygen toxicity with three-hour exposures to po 2 = 200 kpa. Error bars omitted for clarity. a) Following a single dive, symptoms n ; ΔPF - - - - Numerals on the fi gure indicate the number of measurements included in each pair of points. These include measurements immediately after all fi rst dives and before second dives from all dive series. The numbers for two hours after a single dive are combined with those before second exposures with a three-hour surface interval (SI). b) symptoms after pairs of exposures and c) ΔPF after pairs of exposures. Post 1 is after the fi rst exposure. SI: three-hour (n=36): -- --; six-hour (n=33): n ; 15-hour (n=18): ; 17-hour (n=12): Δ. Composite single exposure for comparison (n=127, 26, 26 for immediate, Day +1, and Day +2, respectively): Signifi cant differences from a single exposure are marked as *: three-hour SI : six-hour SI. For differences from zero, see text. one subject reported tingling and numbness of his head and face after two hours, 52 minutes of oxygen at 200 kpa, symptoms that probably represented central nervous system (CNS) oxygen toxicity. the symptoms resolved when he removed the hood to breathe chamber air. another two subjects reported symptoms consistent with CNS oxygen toxicity, but, because they did not resolve for more than 20 minutes after cessation of oxygen breathing or recurred later in the day, they were deemed to be from other causes. DISCUSSION Immediately after three-hour exposures to po 2 of 200 kpa, pulmonary oxygen toxicity is mild but measurable, with a 15% probability of symptoms, and 5% probability of ΔPF, and with an average 1.2% decrease in FVC, 1.6% in FeV 1, and 2.6% in FeF 25-75. the single exposure results are in concordance with those of Clark and lambertsen [11], who reported that during exposure to po 2 = 200 kpa symptoms began after three to eight hours and vital capacity was measurably decreased in 9 of 11 subjects by the fourth hour. However, based on recovery from exposures to po 2 = 130 kpa, the time course of symptom reports is slightly surprising. at the lower po 2, no recovery would be expected in the first few hours but symptoms would be expected to be fully resolved by day +2 [5]. Inspection of the average changes from baseline of indices of pulmonary function does not give a clear indication of recovery from pulmonary oxygen exposure. 298

UHM 2014, Vol. 41, No. 4 Repeated Chamber po 2 = 200 kpa With the exception of FVC in the three-hour SI series, average values of the indices did not differ from baseline after the surface intervals ( pre-dive 2 ), yet the second dives for both three- and six-hour SI caused numerically greater average decrement than the first (Figure 3 a,b), and second dives after three- and six-hour SIs were associated with greater incidence of symptoms than first were dives, a greater incidence that persisted to Day +1 (Figure 5). Second exposures following various SIs had been planned to measure rate of recovery in a manner similar to that previously used for po 2 = 130 kpa [5]. Tacit assumptions were: 1) that exposure effects combine through linear superposition; and 2) that recovery begins at the end of exposure and is monotonic. In other words, the assumptions were: 1) that the immediate effects of a second three-hour exposure would be the same as the immediate effects of the first three-hour exposure, implying that three-hour exposure effects could be subtracted to yield residual effects from the first exposure; and 2) that the incidence of symptoms and ΔPF seen after the second exposure would decrease or remain unchanged as the intermediate recovery time, the SI, increased. The second assumption could have been modified to include a delay. Inconsistent with these assumptions, the incidence of symptoms was highest after the intermediate, sixhour SI (Figure 5b), which was also the only SI for which a decrement was seen in average D L CO. Further, the slowest recovery seen here in mean FEF 25-75 occurred after the 15-hour SI. Potentially delayed injury manifests itself after exposure, beginning sometime between six and nine hours into recovery. Clark and Lambertsen [11] saw continued decrement in vital capacity through the first two to four hours after a 10- to 12-hour exposure to 200 kpa. The data presented here show a trend toward lower mean FEF 25-75 in the period five to 22 hours (Day +1) after a single threehour exposure (Figure 2), depression of FVC and FEV 1 on Day +1 after single two-hour exposures (Figure 2), and D L CO decrements only on Days +1 and +2. However, after single three-hour dives, mean values of pulmonary function variables (Figure 2) and the probability of ΔPF or of symptoms (Figure 5a) do not indicate a large post-exposure effect. Further, mean decreases in FVC after the second dive were greatest after the three-hour SI, while those after the six-hour SI and 15-hour SI were similar to each other. All other changes in flow-volume indices after second dives were similar across SI. That evidence appears to exclude a simple delayed injury effect. The interactions of later exposures with earlier ones appear not to be simply additive, and effects differ for symptoms and ΔPF. Our data do not allow for more than speculation as to cause, but one possibility is that airway inflammation in specific phases of recovery might increase local sensitivity to oxygen injury. Different groups of people participated in the various phases, and thus different individual sensitivity can be considered. However, after their first dives, the cohort for the six-hour SI dives showed the fewest changes in mean pulmonary function indices of any group, no ΔPF, and incidence of symptoms not different from that of any other group. It is thus hard to argue that the high incidence of symptoms after a second dive following a six-hour SI were the result of unusual sensitivity to O 2. In fact, the three-hour SI group was the one with the largest number of smokers and recent smokers (Table 1b). LIMITATIONS AND CONCLUSION This series of exposures does not include an air control. The doubled gas density of these experiments will have altered gas flow dynamics and may have promoted some airway irritation. However, even if some of the symptoms relate to gas density, the increased density is inescapably connected to breathing 100% O 2 at po 2 = 200 kpa. Further, flow-volume parameters have been shown to be unaltered at about twice the density encountered in these experiments after 14 to 30 days breathing a helium-oxygen mixture with po 2 = 40 to 60 kpa at total pressures of 3100 to 4600 kpa [13]. Some after-effects of a three-hour exposure to 200 kpa can linger for 15 hours: a second three-hour exposure causes a significantly greater average decrease in FEF 25-75 than that immediately after a single exposure (Figure 3c). Further, after a second threehour exposure following a 15-hour SI, the incidence of symptoms (11%) remains significant two days later (Figure 6b). Because of the inherent non-linearities we are unable to derive a model to generalize from these exposures. 299

UHM 2014, Vol. 41, No. 4 Repeated Chamber po 2 = 200 kpa We cannot differentiate either between a six-hour SI and an SI twice the length of the exposure. Many more exposures with more SIs and more specific measures of inflammation would be needed to generate a model. Because the effect of a second dive is greater for six-hour than for three-hour SIs, we cannot assign meaningful equivalent times at po 2 = 130 kpa to exposures to 200 kpa or expect a UPTD model to make sense across multiple exposures However, we can caution that repeated exposures to po 2 = 200 kpa can be more than additive in their deleterious effects. Acknowledgments The author would like to thank the many divers (participants, inside tenders and chamber operators) and hospital corpsmen for making this study possible. Funding was provided by Naval Sea System Command Deep Submergence Biomedical Program. Conflict of interest The author has declared that no conflict of interest exists with this submission. n REFERENCES 1. Thorsen E, Aanderud L, Aasen TB. Effects of a standard hyperbaric oxygen treatment protocol on pulmonary function. Eur Respir J 1998; 12:1442-1445. 2. Mialon P, Barthélémy L, Michaud A, Lacour JM. Pulmonary function in men after repeated sessions of oxygen breathing at 0.25 MPa for 90 min. Aviat Space Environ Med 2001; 72(1):215-218. 3. Pott F, Westergaard P, Mortensen J, Jansen EC. Hyperbaric oxygen treatment and pulmonary function. Undersea Hyperb Med 1999; 26(4):225-228. 4. Shykoff BE. Performance of various models in predicting vital capacity changes caused by breathing high oxygen partial pressures. NEDU TR 07-13, Navy Experimental Diving Unit, Panama City, FL, 2007. Available from http://archive.rubicon-foundation.org/6867. Accessed 10 Jan 2011. 5. Shykoff BE. Residual oxygen time model for oxygen partial pressure near 130 kpa. (Manuscript under review, Undersea Hyperbar Med). 6. Hampson NB, Atik DA. Central nervous system toxicity during routine hyperbaric ixygen therapy. Undersea Hyperbar Med. 2003; 30(2): 147-153. 7. American Thoracic Society. Standardization of spirometry 1994 Update. Am J Respir Critical Care Med 1995; 152:1107-1136. 8. Collins Medical. Instruction manual for the Collins Comprehensive Pulmonary Laboratory (CPL). Braintree, MA: Collins Medical, 2000. 9. Shykoff BE. Pulmonary effects of submerged oxygen breathing: 4-, 6-, and 8-hour dives at 140 kpa. Undersea Hyperbar Med 2005; 32(5):351-361. 10. Warkander DE. Unmanned determination of CO 2 values in an oxygen delivery hood (abstract). UHMS 45th Annual Scientific Meeting, Phoenix AZ, June 2012, p133. 11. Clark JM, Lambertsen CJ. Rate of development of pulmonary O 2 toxicity in man during O 2 breathing at 2.0 ATA. J Appl Physiol 1971; 30(5):739-752. 12. Arieli R, Yalov A,. Goldenshluger A. Modeling pulmonary and CNS O 2 toxicity and estimation of parameters for humans. J Appl Physiol 2003; 92:248-256. 13. Thorsen E, Segedal K, Myrseth E, Påsche A, Gulsvik A. Pulmonary mechanical function and diffusion capacity after deep saturation dives. Br J Industrial Med 1990; 47:242-247. 300