Carotid body chemosensitivity is not attenuated during cold water diving

Abstract

Tonic carotid body (CB) activity is reduced during exposure to cold and hyperoxia. We tested the hypotheses that cold water diving lowers CB chemosensitivity and augments CO₂ retention more than thermoneutral diving. Thirteen subjects [age: 26 ± 4 yr; body mass index (BMI): 26 ± 2 kg/m²) completed two 4-h head-out water immersion protocols in a hyperbaric chamber (1.6 ATA) in cold (15°C) and thermoneutral (25°C) water. CB chemosensitivity was assessed with brief hypercapnic ventilatory response ($\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$) and hypoxic ventilatory response ($\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$) tests before dive, 80 and 160 min into the dive (D80 and D160, respectively), and immediately after and 60 min after dive. Data are reported as an absolute mean (SD) change from predive. End-tidal CO₂ pressure increased during both the thermoneutral water dive [D160: +2 (3) mmHg; P = 0.02] and the cold water dive [D160: +1 (2) mmHg; P = 0.03]. Ventilation increased during the cold water dive [D80: 4.13 (4.38) and D160: 7.75 (5.23) L·min⁻¹; both P < 0.01] and was greater than the thermoneutral water dive at both time points (both P < 0.01). $\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$ was unchanged during the dive (P = 0.24) and was not different between conditions (P = 0.23). $\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$ decreased during the thermoneutral water dive [D80: −3.45 (3.61) and D160: −2.76 (4.04) L·min·mmHg⁻¹; P < 0.01 and P = 0.03, respectively] but not the cold water dive. However, $\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$ was not different between conditions (P = 0.17). In conclusion, CB chemosensitivity was not attenuated during the cold stress diving condition and does not appear to contribute to changes in ventilation or CO₂ retention.

INTRODUCTION

In an underwater environment, various stressors lead to altered physiological states that pose health and safety risks to a diver. Carbon dioxide (CO₂) retention occurs during diving (1–3) and can lead to both physical and cognitive impairments (1, 4). CO₂ retention results from reduced alveolar ventilation ($\dot{\text{V}}\text{A}$) due to increased dead space and/or reduced total ventilation [minute ventilation ($\dot{\text{V}}\text{E}$)]. Moreover, it has been proposed that the relative alveolar hypoventilation is mediated in part by reduced peripheral chemoreceptor activity and/or sensitivity that contributes to CO₂ retention by attenuating the hypercapnic ventilatory response (HCVR) (5–8). However, despite extensive review (6, 7), little is known about the mechanisms altering the control of ventilation during diving.

Elevated barometric pressure (i.e., hyperbaria and the resultant hyperoxemia) and/or the hydrostatic pressure (i.e., water immersion) that occur during diving have been identified as potential mediators that contribute to reductions in peripheral and/or central chemosensitivity (6, 7). Hyperoxia attenuates the HCVR at sea level pressure (1 ATA) (9–11), and that observation has been extrapolated to mean that hyperoxia during hyperbaria has similar implications during diving (2, 3, 5–7, 12–15). Moreover, peripheral and central chemosensitivity are integrated, such that when carotid bodies (CBs) are perfused with hyperoxemia the HCVR is drastically reduced (16, 17). However, we recently demonstrated that CB chemosensitivity to hypercapnia during dry conditions did not change from 1 ATA to 1.6 ATA, and it was not altered by breathing gas [i.e., fraction of inspired O₂ ($\mathrm{F}{\mathrm{I}}_{{\mathrm{O}}_2}$) = 0.21 vs. 1.00 at 1.6 ATA] (18).

Additionally, experiments from our laboratory have shown that CB chemosensitivity to hypercapnia (∼35% of the HCVR) and hypoxia is not reduced during thermoneutral head-out water immersion at 1 ATA (19). Our follow-up experiments demonstrated that the HCVR is augmented during thermoneutral head-out water immersion (20, 21). Therefore, it is unlikely that water immersion alone contributes to the reduction in the chemical control of ventilation that has been proposed as a mechanism for CO₂ retention during diving. However, these experiments were completed in thermoneutral water and are not representative of colder ocean and fresh water temperatures across much of the planet. These colder temperatures have been shown to reduce in vitro CB chemoreceptor activity (22, 23).

Consequently, divers exposed to the combined stressors associated with typical diving (i.e., cold stress, hyperoxia, and water immersion) might be at an increased risk of O₂ and CO₂ toxicity due to reductions in CB chemoreceptor activity and subsequent CO₂ retention. Therefore, the primary aim of our study was to determine whether CB chemosensitivity is reduced during thermoneutral and/or cold water dives at 1.6 ATA. We hypothesized that 1) CB chemosensitivity would be attenuated during the thermoneutral water dive, 2) cold water diving would further reduce CB chemosensitivity, 3) both dives would result in CO₂ retention, and 4) there would be a greater CO₂ retention during the cold water dive.

METHODS

Thirteen men [age 26 (4) yr; height 177 (7) cm; body mass 83 (11) kg; body mass index 26 (2) kg·m⁻²] participated in three visits: a screening and familiarization visit and two experimental visits. All subjects were informed of the experimental procedures and risks and provided written informed consent before beginning the study. Ethical approval was granted by the University at Buffalo Institutional Review Board, and the study was performed in accordance with the Declaration of Helsinki. Before the initial visit, subjects completed a health history questionnaire and a diver physical examination with a study physician. Subjects were excluded for any neurological, metabolic, pulmonary, cardiovascular, or gastrointestinal (i.e., specific contraindication for the core temperature capsule) disease and/or surgery. Additionally, subjects self-reported to be nonsmokers and not currently taking any medications that are known to alter pulmonary, cardiovascular, or neurological function. Subjects refrained from alcohol, caffeine, and exercise for 12 h and food for 2 hours before each visit. Additionally, subjects were also asked to abstain from taking any antioxidant supplementation 24 h before each study to prevent the potential interaction of antioxidants and the ventilatory response to hypercapnia (24). The screening and familiarization visit consisted of the subjects breathing through the closed-circuit rebreather (i.e., the mouthpiece, bellows, and pneumatic switching valves), gases, and procedures that were used for assessing CB chemosensitivity during the experimental visits. This experimental protocol was developed in our laboratory and previously published (18). Additionally, subjects completed a brief dive (i.e., descent, a brief bottom time, and ascent) to ensure that subjects were experienced in the diving procedures.

Experimental Protocol

Subjects reported to the laboratory for two experimental visits in a randomized, counterbalanced design: 1) a thermoneutral water (25°C; TN) dive and 2) a cold water (15°C; CW) dive visit. CB chemosensitivity to hypercapnia ($\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$) and CB chemosensitivity to hypoxia ($\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$) were measured at five time points during each visit: predive (PRE), 80 and 160 min during the dive (D80 and D160, respectively), and postdive and 60 min postdive (PD and PD60, respectively).

Predive assessments of CB chemosensitivity were completed at ground level 744 ± 1 mmHg (182 m above sea level) in the unpressurized hyperbaric chamber. After predive assessments, subjects briefly exited the chamber, voided their bladder, and donned a 7-mm wetsuit, gloves, and boots (in both experimental trials) before reentering the dry side of the hyperbaric chamber for descent. The chamber was pressurized to 1,205 ± 9 mmHg to simulate a depth of 1.6 ATA (20 fsw) for 4 h. This experimental protocol was designed for military-relevant external validity and modeled after a maximum operational dive profile (i.e., depth and bottom time) when utilizing a closed-circuit rebreather (CCR). The chamber was pressurized to descend at a rate of 22.8 m/min. However, the descent rate was slowed or the dive paused if the subjects had difficulty equalizing pressure in their ears. Upon reaching a depth of 1.6 ATA, bottom time was started and subjects were transitioned to the wet side of the chamber. Subjects were seated in the wet pod such that the water was at the level of the clavicle, and subjects breathed from a surface-supplied demand regulator (i.e., 21% O₂, 78% N₂, 0.04% CO₂). Periodically, the chamber was vented while chamber pressure was maintained as needed to prevent the subjects from being exposed to a hypercapnic environment. Before decompression, subjects breathed 100% O₂ for 17 min before returning to the surface at a rate of 9.1 m/min (i.e., <1-min decompression). Immediately after exiting the chamber, the subjects were given a neurological assessment to screen for signs and symptoms of decompression illness. After a 15-min surface interval, subjects then reentered the nonpressurized chamber and completed postdive assessments of CB chemosensitivity.

Assessment of Carotid Body Chemosensitivity

CB chemosensitivity to hypercapnia was measured via four separate administrations of a single breath of hypercapnic gas (i.e., 13% CO₂, 21% O₂, and 66% N₂) separated by 2 min of recovery breathing on the CCR. With a pneumatic switching valve (Hans Rudolph, Inc., Shawnee, KS), subjects were rapidly switched between breathing on the CCR and a spirometer (Ohio 822; Ohio Medical Products, Houston, TX) filled with the hypercapnic gas and back to the CCR. All four administrations of hypercapnic gas consisted of a single breath. Subjects were instructed to take normal tidal breaths and were also blinded as to when the gases were administered. CB chemosensitivity to hypercapnia was calculated by plotting the mean of the three highest consecutive ventilations (i.e., single breaths extrapolated to minute values; $\dot{\text{V}}\text{E}$) versus end-tidal CO₂ pressure ($\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$) within 1 min after each hypercapnic gas administration (for more detail see Ref. 18, Fig. 3A). The linear regression line produced from the plot of $\dot{\text{V}}\text{E}$ versus $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ was calculated and used as an index of CB chemosensitivity to hypercapnia (19, 25–27). We also determined the cardiovascular response to CB activation. The peak heart rate (HR) and the peak mean arterial pressure (MAP) were plotted versus $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ within 1 min after each hypercapnic gas administration, with similar methods that have been used during acute hypoxia. Similar to the ventilatory responses to the brief hypercapnic exposures, the linear regression lines produced from the plots of heart rate versus $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ and mean arterial pressure versus $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ were calculated and used as additional indexes of CB chemosensitivity to hypercapnia (19, 28–31).

CB chemosensitivity to hypoxia was measured via four separate nitrogen administrations (i.e., 100% N₂) separated by 2 min of recovery breathing on the CCR. Briefly, with a pneumatic switching valve (Hans Rudolph, Inc., Shawnee, KS), subjects were rapidly switched between breathing on the CCR and a spirometer (Ohio 822; Ohio Medical Products, Houston, TX) filled with 100% nitrogen and back to the CCR. The first two nitrogen administrations consisted of 2 and 4 breaths, respectively, for all subjects. The number of nitrogen breaths for each of the remaining two series of nitrogen administrations (2–6 breaths) was determined based on the arterial oxygen saturation ($\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$) values achieved during the first two nitrogen administrations and kept consistent within a subject during each CB chemosensitivity assessment for both experimental visits. The goal was to achieve a range of nadir $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ values (e.g., 80–100%) after the nitrogen administrations at ground level. CB chemosensitivity to hypoxia was calculated by plotting the mean of the three highest consecutive ventilations (i.e., single breaths extrapolated to minute values; $\dot{\text{V}}\text{E}$) versus the nadir $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ value within 1 min after each nitrogen administration (for more detail see Ref. 18, Fig. 3B) The linear regression line was calculated from these values and used an index of CB chemosensitivity to hypoxia (19, 25–32). Additionally, the brief N₂ exposure infrequently caused a meaningful reduction in $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ at depth; therefore, the means of the three highest consecutive $\dot{\text{V}}\text{E}$ values were also plotted versus the nadir $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ values within 1 min after each nitrogen administration. The linear regression line was calculated from these values and used an index of CB chemosensitivity to hypoxia. Importantly, end-tidal O₂ pressure ($\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$) most likely represents the arterial O₂ content that the CBs are exposed to. We also determined the cardiovascular response to CB stimulation. The peak heart rate and the peak mean arterial pressure values were plotted versus the nadir $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$ values within 1 min after each nitrogen administration. CB chemosensitivity to hypoxia data are reported as the inverse value of the slope of the linear regression line for the ventilatory, heart rate, and blood pressure responses to hypoxia (25, 26, 28–31).

Measurements

Ventilation.

Before each assessment of CB chemosensitivity at depth, subjects were transitioned from breathing on the surface-supplied demand regulator to the custom CCR (see Ref. 18, Fig. 1) for measurement of ventilation and delivery of hypercapnic and hypoxic gases. Subjects breathed from a mouthpiece that was connected to a custom CCR and bellows system to measure inspiratory and expiratory volume. Inspired air passed through a CO₂ scrubber that preceded the bellows in the closed circuit to maintain a constant fraction of inspired CO₂ ($\mathrm{F}{\mathrm{I}}_{\mathrm{C}{\mathrm{O}}_2}$) and prevent hypercapnia. By titrating small amounts of 100% O₂ into the inspiratory line as needed, $\mathrm{F}{\mathrm{I}}_{{\mathrm{O}}_2}$ was maintained at 0.21. Tidal volume (Vt) was measured continuously with a potentiometer in the bellows system. Post hoc measures of Vt and respiratory rate (f_B) were determined breath by breath with data analysis software (AcqKnowledge 4.2, Goleta, CA) for each 300-s baseline and 60-s post-CB assessment duration analyzed, with all sighs, coughs, and talking removed. Minute ventilation ($\dot{\text{V}}\text{E}$) was calculated as the product of Vt and f_B. In all instances, $\dot{\text{V}}\text{E}$ and Vt were converted from ambient temperature and pressure saturated (ATPS) to body temperature and pressure saturated (BTPS) with standard equations. The fractions of expired oxygen ($\mathrm{F}{\text{E}}_{{\mathrm{O}}_2}$; %) and carbon dioxide ($\mathrm{F}{\text{E}}_{\mathrm{C}{\mathrm{O}}_2}$; %) from gases sampled near the mouth were measured with a mass spectrometer located on the outside of the chamber (MGA 1100; MA Tech Services, Inc., St. Louis, MO). Partial pressures of O₂ and CO₂ were calculated by multiplying the fraction of expired gases by the ambient chamber pressure at each time point. End-tidal CO₂ ($\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$) was used as a marker of arterial CO₂ pressure (5, 33, 34), and we have used this approach in previous studies (19–21).

Hemodynamics.

Arterial oxygen saturation ($\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$) was measured from the right index finger with a pulse oximeter (Radical; Masimo Corporation, Irvine, CA). Noninvasive beat-to-beat blood pressure was measured from the middle finger on the right hand, which was supported by a sling at the level of the heart (Finapres NOVA; Finapres Medical Systems, The Netherlands). Heart rate was measured continuously from a three-lead electrocardiogram (Finapres NOVA; Finapres Medical Systems, The Netherlands).

Thermal responses.

Core body temperature (T_C; i.e., intestinal temperature) was measured with an ingestible capsule (CorTemp; HQ Inc., Palmetto, FL). Additionally, subjective thermal [1–7 scale: 1, cold; 4, neutral; 7, hot (35)] and shivering (0–10 scale: 0, not shivering; 10, most shivering ever) perceptions were measured with thermal and shivering perceptual scales.

Data and Statistical Analyses

Data were captured by a data acquisition system for off-line analysis (Biopac MP150, Goleta, CA) at 62.5 Hz for $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ and HR, 1.0 kHz for $\mathrm{F}{\text{E}}_{{\mathrm{O}}_2}$ and $\mathrm{F}{\text{E}}_{\mathrm{C}{\mathrm{O}}_2}$, 250.0 Hz for the bellows (i.e., Vt and f_B), and 125.0 Hz for arterial blood pressure. CB chemosensitivity (i.e., slope of the linear regression for $\dot{\text{V}}\text{E}$ vs. $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ and vs. $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$ for $\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$ and $\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$, respectively) values were multiplied by a factor of 100 for readability. Data are presented in the figures as the change from predive values [mean (SD)]. Two-tailed, paired t tests were used to determine whether there were differences in predive values between conditions. Two-way repeated-measures ANOVAs were used (time and water temperature condition) to examine changes from baseline (i.e., resting) for the outcome variables (ventilatory and hemodynamic data, CB chemosensitivity, thermal responses, and perceptual data). If a significant interaction or main effect was found (36), we used the least squared differences (LSD) post hoc procedure to determine where differences existed. Data were analyzed with Prism software (version 8; GraphPad Software Inc., La Jolla, CA). Significance was set a priori at an α level of 0.05.

RESULTS

Baseline

There were no differences in resting $\dot{\text{V}}\text{E}$, f_B, Vt, $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$, $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$, $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, and HR during predive between CW and TN conditions (all P > 0.05). However, predive MAP was higher in CW than TN (P = 0.05; Table 1).

Table 1.

Predive ventilatory, hemodynamic, and chemosensitivity data

	CW	TN	P Value
$\dot{\mathrm{V}}{\text{E}}_{\mathrm{BTPS}}$, L·min−1	10.59 (1.93)	10.89 (2.45)	0.57
fB, breaths·min−1	12 (3)	13 (4)	0.57
VtBTPS, L	0.81 (0.16)	0.90 (0.34)	0.13
$\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$, mmHg	43 (2)	42 (3)	0.45
$\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$, mmHg	104 (5)	106 (6)	0.32
$\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, %	97 (1)	97 (1)	0.12
HR, beats·min−1	71 (10)	70 (9)	0.77
MAP, mmHg	70 (7)	64 (8)	0.02
$\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$, L·min·mmHg−1	8.29 (6.38)	9.62 (7.71)	0.57
HR response to hypercapnia, beats·min−1·mmHg−1	0.32 (0.22)	0.19 (0.09)	0.02
MAP response to hypercapnia, mmHg·mmHg−1	1.11 (0.21)	0.66 (0.03)	<0.01
$\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$, L·min·mmHg−1	5.30 (2.75)	5.93 (3.65)	0.64
HR response to hypoxia, beats·min−1·mmHg−1	149.2 (53.3)	144.0 (46.4)	0.77
MAP response to hypoxia, mmHg·mmHg−1	362.0 (61.1)	358.9 (60.9)	0.88

Values are presented as mean (SD); n = 13 (all males). Carotid body chemosensitivity to hypercapnia ($\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$) and carotid body chemosensitivity to hypoxia ($\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$) values were multiplied by a factor of 100 for readability. CW, cold water; f_B, respiratory rate; HR, heart rate; MAP, mean arterial pressure; $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$, end-tidal CO₂ pressure; $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$, end-tidal O₂ pressure; $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, arterial oxygen saturation; TN, thermoneutral water; $\dot{\mathrm{V}}{\text{E}}_{\mathrm{BTPS}}$, minute ventilation (body temperature pressure saturated); Vt_BTPS, tidal volume (body temperature pressure saturated). Two-tailed, paired t tests.

Changes in ventilation and hemodynamics from predive are shown in Fig. 1. There was a time effect for $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ (P < 0.01) but no main effect of condition (P = 0.66) or interaction (P = 0.41). $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$ increased from predive at all time points during and PD in both conditions (all P < 0.01). There was a main effect of time on $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$ (P < 0.01), but there was no main effect of condition (P = 0.85) or interaction (P = 0.56). $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$ increased at both time points during the dive in both conditions (all P < 0.01). $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$ returned to predive values PD in both conditions (all P > 0.05). There was a main effect of time for $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ (P < 0.01). $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ increased during the dive in both TN and CW at D160 (P = 0.03 and P = 0.02, respectively) but returned to predive values PD and at PD60 (all P > 0.05) in both conditions. There was no main effect of condition (P = 0.66) or interaction (P = 0.20) for $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$.

Figure 1.

End-tidal CO₂ pressure ($\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$; A), arterial oxygen saturation ($\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$; B), end-tidal O₂ pressure ($\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$; C), minute ventilation ($\dot{\mathrm{V}}\text{E}$; D), tidal volume (Vt; E), respiratory rate (f_B; f), mean arterial pressure (MAP; G), and heart rate (HR; H) values are presented as an absolute change from predive as mean (SD). Baseline ventilatory and hemodynamic data measured predive (PRE), 80 and 160 min into the dive (D80 and D160, respectively), postdive (PD), and 60 min postdive (PD60). Two-way repeated-measures analysis of variance (time × water temperature). If a significant interaction or main effect was found, post hoc multiple comparisons were completed with the least squared differences (LSD) procedure. Cold water (CW), n = 13 and thermoneutral water (TN) n = 13; all males. *P < 0.05 from pre-dive; #P < 0.05 CW vs. TN.

There was an interaction effect for $\dot{\text{V}}\text{E}$ (P < 0.01). $\dot{\text{V}}\text{E}$ increased from predive at D80, D160, and PD during the dive in the CW condition (all P < 0.01) but returned to predive values thereafter (P > 0.05). Additionally, $\dot{\text{V}}\text{E}$ was greater in the CW compared to TN condition at D80 (P < 0.01), D160 (P < 0.01), and PD (P = 0.04). There was an interaction effect. Vt was greater during the CW condition at D80 and D160 (both P < 0.01). There was an interaction effect for f_B (P = 0.034). f_B increased from predive in the CW condition at D80 (P = 0.03) and D160 (P = 0.04). Additionally, f_B was higher in the CW compared to TN condition at D160 (P = 0.02).

There was an interaction of time and condition for HR (p < 0.01). HR decreased from predive in the CW condition at D80 (P < 0.01) and D160 (P < 0.01). Additionally, HR decreased from predive in the TN condition at D80 and D160 (both P < 0.01) and was lower than in the CW condition at D160 (P < 0.01). There was also an interaction of time and condition for MAP (P = 0.05). MAP increased from predive in the CW condition at D80 (P < 0.01) and D160 (P < 0.01) and returned to predive values PD and PD60 (both P > 0.05). MAP increased from predive in the TN condition at D80 and D160 (both P < 0.01) and remained elevated at PD (P < 0.01) and PD60 (P < 0.01). Additionally, MAP was higher in the TN condition PD (P = 0.05).

Carotid Body Chemosensitivity to Hypercapnia

CB chemosensitivity to hypercapnia is shown in Fig. 2 and Table 2. There was no main effect of time (P = 0.24), condition (P = 0.24), or interaction (P = 0.34) for $\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$. However, there was a main effect of time on the HR response to hypercapnia (P = 0.02) but no main effect of condition (P = 0.77) or interaction (P = 0.89). The HR response to hypercapnia was elevated in the TN condition at PD (P = 0.05) and PD60 (P = 0.02). There was an interaction of time and condition on the MAP response to hypercapnia (P < 0.01). The MAP response to hypercapnia decreased from predive at D80 and D160 in the CW condition (both P < 0.01), briefly returned to predive values at PD (P = 0.14), but decreased at PD60 (P < 0.01). Additionally, the MAP response to hypercapnia decreased from predive at D80 (P < 0.01) and D160 (P < 0.01) in the TN condition but increased at PD (P = 0.06) and PD60 (P < 0.01). The MAP response to hypercapnia was lower in CW during the dive (all P < 0.01), PD (P = 0.03), and PD60 (P < 0.01).

Figure 2.

Ventilatory (A), heart rate (HR; B), and mean arterial pressure (MAP; C) values are presented as an absolute change from predive as mean (SD). Carotid body chemosensitivity to hypercapnia ($\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$) measured predive (PRE), 80 and 160 min into the dive (D80 and D160, respectively), postdive (PD), and 60 min postdive (PD60). Two-way repeated-measures analysis of variance (time × water temperature). If a significant interaction or main effect was found, post hoc multiple comparisons were completed with the least squared differences (LSD) procedure. $\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$ values were multiplied by a factor of 100 for readability. bpm, Beats per minute. Cold water (CW) n = 13 and thermoneutral water (TN) n = 13; all males. *P < 0.05 from predive; #P < 0.05 CW vs. TN.

Table 2.

Carotid body chemosensitivity

Parameter	CW					TN
	PRE	D80	D160	PD	PD60	Pre	D80	D160	PD	PD60
$\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$, L·min·mmHg−1	8.29 (6.38)	10.41 (6.66)	11.42 (9.61)	9.88 (11.61)	12.60 (9.40)	9.62 (7.71)	6.56 (2.90)	7.09 (3.65)	10.76 (6.02)	11.77 (4.66)
HR response to hypercapnia, beats·min−1·mmHg−1	0.32 (0.22)	0.40 (0.38)	0.41 (0.43)	0.51 (0.36)	0.66 (0.81)	0.19 (0.09)	0.32 (44)	0.26 (0.18)	0.55 (0.61)	0.51 (0.39)
MAP response to hypercapnia, mmHg·mmHg−1	1.11 (0.21)	0.42 (0.19)*	0.37 (0.13)*	0.86 (0.56)	0.75 (0.17)*	0.66 (0.03)#	0.49 (0.13)*	0.47 (0.16)*	0.84 (0.28)	0.92 (0.19)*#
$\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$, L·min·mmHg−1	5.30 (2.75)	5.17 (4.75)	5.43 (4.29)	7.03 (5.41)	7.80 (4.93)	5.93 (3.65)	2.47 (1.45)	3.17 (1.25)	6.94 (3.36)	7.27 (4.26)
HR response to hypoxia, beats·min−1·mmHg−1	149.2 (53.3)	144.7 (52.4)	105.8 (59.8)*	183.1 (37.2)*	180.4 (63.1)*	144.0 (46.4)	116.6 (44.7)	135.7 (71.9)	167.4 (68.2)	159.2 (60.1)
MAP response to hypoxia, mmHg·mmHg−1	362.0 (61.1)	311.2 (92.3)	228.2 (90.2)	286.8 (66.5)*	240.9 (65.7)*	358.9 (60.9)	284.4 (65.3)*	269.5 (94.8)*	297.4 (88.4)	277.0 (46.3)*

Values are presented as mean (SD); n = 13 (all males). Carotid body chemosensitivity to hypercapnia ($\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$) and carotid body chemosensitivity to hypoxia ($\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$) values were multiplied by a factor of 100 for readability. CW, cold water; D80 and D160, 80 and 160 min into the dive; HR, heart rate; MAP, mean arterial pressure; PD, postdive; PD60, 60 min postdive; PRE, predive; TN, thermoneutral water. Two-way repeated-measures analysis of variance (time × water temperature). If a significant interaction or main effect was found, post hoc multiple comparisons were completed with the least squared differences (LSD) procedure. *P < 0.05 from pre-dive; #P ≤0.05 CW vs. TN.

Carotid Body Chemosensitivity to Hypoxia

CB chemosensitivity to hypoxia is shown in Fig. 3 and Table 2. There was a main effect of time (P < 0.01) but no main effect of condition (P = 0.23) or interaction (P = 0.21) for $\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$. $\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$ decreased from predive during the TN condition at D80 (P < 0.01) and D160 (P = 0.03) but returned to predive values thereafter (both P > 0.05). However, there was a main effect of time on the HR response to hypoxia (P < 0.01) but no main effect of condition (P = 0.97) or interaction (P = 0.58). The HR response to hypoxia was decreased from predive at D80 (P < 0.01) and D160 (P < 0.01) in the CW condition but was increased from predive PD (P = 0.03). Additionally, the HR response to hypoxia was decreased from predive at D80 (P < 0.01) and D160 (P < 0.01) in the TN condition. Finally, there was a main effect of time (P < 0.01) but no main effect of condition (P = 0.47) or interaction (P = 0.33) for the MAP response to hypoxia. The MAP response to hypoxia was decreased from predive at all time points in the CW condition (all P < 0.01) and during the dive in the TN condition (both P < 0.01). Additionally, the MAP response to hypoxia was decreased from predive at PD60 in the TN condition (P < 0.01).

Figure 3.

Ventilatory (A), heart rate (B), and mean arterial pressure (MAP; C) values are presented as an absolute change from predive as mean (SD). Carotid body chemosensitivity to hypoxia ($\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$) measured predive (PRE), 80 and 160 min into the dive (D80 and D160, respectively), postdive (PD), and 60 min postdive (PD60). Two-way repeated-measures analysis of variance (time × water temperature). If a significant interaction or main effect was found, post hoc multiple comparisons were completed with the least squared differences (LSD) procedure. $\mathrm{C}{\mathrm{B}}_{{\mathrm{O}}_2}$ values were multiplied by a factor of 100 for readability. bpm, Beats per minute. Cold water n = 13 and thermoneutral water n = 13; all males. *P < 0.05 from predive.

Thermal Responses and Perceptions

T_C and thermal perceptions are shown in Fig. 4. There was a main effect of time (P < 0.01) but no main effect of condition (P = 0.96) or interaction (P = 0.23) for T_C. However, the post hoc analysis was unable to detect any differences from predive (all P > 0.05). There was an interaction of time and condition for thermal perception (P < 0.01). Thermal perception decreased from predive at all time points during the dive in the CW condition (P < 0.01) and immediately PD (P = 0.05). Additionally, thermal perception was lower than predive at D60 (P = 0.03), D120 (P < 0.01), D180 (P < 0.01), and D240 (P = 0.01) in the TN condition. However, thermal perception was lower at all time points in the CW compared with the TN condition (all P < 0.01). Finally, there was an interaction of time and condition for shivering perception (P < 0.01). Shivering perception during the CW condition increased during the dive at all time points (all P < 0.01) and was greater than TN at all time points (all P < 0.01).

Figure 4.

Core body temperature (T_C; A), thermal perception (B), and shivering perception (C) values are presented as an absolute change from predive as mean (SD). Thermal responses and perceptions measured predive (PRE), at the start of the dive (D0), 60, 120, 180, and 240 min into the dive (D60, D120, D180, and D240, respectively), postdive (PD), and 60 min postdive (PD60). Two-way repeated-measures analysis of variance (time × water temperature). If a significant interaction or main effect was found, post hoc multiple comparisons were completed with the least squared differences (LSD) procedure. a.u., Arbitrary units. Cold water (CW) n = 13 and thermoneutral water (TN) n = 13; all males. *P < 0.05 from predive; #P < 0.05 CW vs. TN.

DISCUSSION

The purpose of our study was to assess CO₂ retention and CB chemosensitivity during thermoneutral and cold water diving at 1.6 ATA. Our main findings are that 1) the brief hypercapnic ventilatory response is unchanged during the thermoneutral and cold water dives and 2) mild CO₂ retention occurred during both the thermoneutral and cold water dives. Contrary to our hypotheses, we show that CB chemosensitivity to hypercapnia (i.e., $\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$) is not attenuated and does not appear to contribute to the CO₂ retention observed in diving.

Carotid Body Chemosensitivity

Our primary finding is that the brief ventilatory response to hypercapnia is unchanged from 1 ATA to 1.6 ATA and was not different between water temperatures (i.e., 25°C vs. 15°C). Contrary to our hypotheses, the combined effects of hyperbaria (and subsequent hyperoxia), cold stress, and the hemodynamic alterations from head-out water immersion did not attenuate ventilatory response to a brief hypercapnic stimulus. This is in line with previous work assessing peripheral chemosensitivity (19) and respiratory control (37, 38) at 1 ATA during thermoneutral head-out water immersion and during dry, hyperbaric exposure at 1.6 ATA (18). Rather, there appeared to be a rightward shift in the respiratory operating point (i.e., elevated $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ for a given $\dot{\text{V}}\text{E}$). Statistically, this was only apparent in the thermoneutral water dive. However, the mean increase in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ during the cold water dive was similar to that of the thermoneutral water dive (i.e., ∼2-mmHg rise in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$). It is unlikely that there are any physiological differences between trials in regard to the magnitude of CO₂ retention. There has long been agreement that the ventilatory response to hypercapnia is attenuated in diving, which was thought to contribute to CO₂ retention (5–8, 39). However, combined with our previous study during dry, hyperbaric conditions at 1.6 ATA, an attenuated ventilatory response to brief hypercapnia does not appear to contribute to CO₂ retention (18). That is, peripheral chemosensitivity to hypercapnia does not appear to be attenuated during shallow water diving. However, we did not assess central chemosensitivity to hypercapnia, and therefore it is still unclear whether central chemosensitivity is attenuated during diving. Importantly, the HCVR (i.e., central chemosensitivity; measured at 1 ATA) has been correlated with CO₂ retention during diving at 4.7 ATA (5) but is a rather poor predictor relative to other factors (e.g., depth, gas density, and inspiratory and expiratory resistance). Rather, it is important to note that the present study was conducted during shallow water diving and carotid body chemosensitivity was assessed. Notably, HCVR slope has been shown to be reduced at far deeper depths (e.g., up to 37 ATA) (40–42), which appears to be associated with increased gas density, a major predictor of CO₂ retention in diving (5, 7). In contrast, our previous work has shown that central chemosensitivity to hypercapnia is augmented during head-out water immersion at 1 ATA (21). Thus, it appears unlikely that a reduction in central chemosensitivity during diving contributes to CO₂ retention. It is also important to note that these data were collected during resting conditions and may not represent physiological responses in exercising divers (7) who exhibit hypoventilation (15).

Although we did not observe changes in the ventilatory responses, we found that the magnitude of the MAP responses to the brief hypercapnic stimuli during diving was attenuated during both dives, with the greatest decrease during the cold dive condition. We previously examined these MAP responses during dry hyperbaric exposures (18) and head-out water immersion in thermoneutral water at 1 ATA (19). The only condition that attenuated the MAP response to brief hypercapnic exposures was in the dry hyperbaric condition while breathing 100% oxygen (18). However, the magnitude of decrease we observed during the cold dive in the present study is substantially lower than other diving conditions we have tested. The underlying mechanisms (i.e., activation of type 1 glomus cells in the carotid body, afferent nerve transmission, brain stem integration, efferent nerve transmission, transduction to the target organ, and/or target organ function) that contribute to this response are currently not clear and require further investigation.

We observed that the ventilatory response to hypoxia was attenuated during the thermoneutral water dive but unchanged during the cold water dive. Interestingly, we observed a reduction in the ventilatory response to hypoxia during a dry, hyperbaric exposure at 1.6 ATA breathing either air (21% O₂) or 100% O₂ (18), both of which were conducted in thermoneutral environments. This is in contrast with previous work from our laboratory that showed that peripheral chemosensitivity to hypoxia is unchanged during thermoneutral head-out water immersion at 1 ATA (19). This results in two primary interpretations: 1) the reductions in the hypoxic ventilatory response are a function of increased barometric pressure, and 2) the maintenance of the hypoxic ventilatory response in the cold water dive is likely a function of cold stress (i.e., sympathetic stimulus) (43). Subjects in the present study and in the air breathing condition in the dry exposure in our previous work (18) were presumably exposed to the same level of hyperoxia and therefore likely similar high arterial oxygen content. We hypothesized that the hyperoxemia resulted in an attenuated relative hypoxia [i.e., acute reduction in arterial partial pressure of oxygen ($\mathrm{P}{\mathrm{a}}_{{\mathrm{O}}_2}$)] and therefore accounted for the reduction in sensitivity from 1 ATA to 1.6 ATA during the dry hyperbaric conditions (18).

Similar to our previous work (18), we found that the magnitude of the MAP responses during brief hypoxic exposures at 1.6 ATA was attenuated. Furthermore, it appears as though the magnitude of the decrease is similar between the environmental conditions of the hyperbaric exposures (i.e., breathing 21% and 100% oxygen during thermoneutral, dry conditions as well as 21% oxygen breathing during thermoneutral and cold, wet conditions). We previously found that the MAP response to hypoxia is not attenuated during head-out water immersion at 1 ATA (19). Therefore, it appears as though it is the hyperbaric condition, and likely the extremely high $\mathrm{P}{\mathrm{a}}_{{\mathrm{O}}_2}$ that can only be achieved at depth, that is the main mediator of this response.

Ventilation and CO₂ Retention

The magnitude of CO₂ retention (i.e., +2–3 mmHg) in the thermoneutral water dive was similar to that of previous work in our laboratory (19–21) and others (38). Statistically, there was no change in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ during the cold water dive. However, to interpret these findings as physiologically different would be in error. Numerically, there was a similar rise in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ during the cold water dive (see Table 3). This similar magnitude of CO₂ retention occurred despite an increase in $\dot{\text{V}}\text{E}$ during the cold water dive. Therefore, the relative contribution of the hydrostatic pressure versus the shivering-induced ventilatory response and increase in CO₂ production may have actually mitigated any reduction in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ under true hyperventilation and ultimately resulted in similar CO₂ retention between trials. These responses have implications for the proposed primary mechanisms for CO₂ retention in diving (i.e., reduced ventilatory drive and alveolar ventilation or an increase in the work of breathing) (6). During diving, there is an increase in physiological dead space due to the rise in breathing frequency that is accompanied by an increase in gas density (5, 7, 14, 34). A failure to adequately increase tidal volume when physiological dead space increases results in a relative alveolar hypoventilation (7). In this context, experienced divers selectively hypoventilate (via a reduction in breathing frequency), which results in a greater CO₂ retention compared with inexperienced divers (5, 15, 44, 45). Although the work of breathing increases at depth, which might result in submissive hypercapnia (46), the work of breathing is not insurmountable. For instance, ventilation increases during exercise (5, 14) and for thermogenesis (39) at depth, which indicates there is still a capacity to increase ventilation at depth. Additionally, respiratory muscle training reduces the work of breathing at depth (47–49) but fails to prevent CO₂ retention (15), which indirectly supports the idea that the increase in the work of breathing at depth likely does not contribute to CO₂ retention during diving.

Table 3.

Baseline ventilatory and hemodynamic data

Parameter	CW					TN
	PRE	D80	D160	PD	PD60	PRE	D80	D160	PD	PD60
$\dot{\text{V}}{\mathrm{E}}_{\mathrm{BTPS}}$, L·min−1	10.59 (1.93)	14.71 (3.27)*#	18.34 (4.76)*#	12.80 (3.75)*	11.37 (2.11)	10.89 (2.45)	10.12 (2.30)	9.84 (2.17)	10.96 (1.50)	10.55 (1.55)
fB, breaths·min−1	12 (3)	15 (5)	16 (5)#	15 (5)	14 (3)	13 (4)	13 (4)	12 (5)	14(4)	13 (3)
VtBTPS, L	0.81 (0.16)	1.12 (0.46)*#	1.24 (0.45)*#	0.87 (0.31)	0.85 (0.24)	0.90 (0.34)	0.83 (0.25)	0.93 (0.33)	0.82 (0.21)	0.82 (0.21)
$\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$, mmHg	43 (2)	43 (3)	44 (3)*	43 (2)	43 (2)	42 (3)	44 (2)	44 (3)*	42 (3)	42 (2)
$\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$, mmHg	104 (5)	197 (12)*	197 (14)*	104 (9)	103 (11)	105	205 (12)*	200(11)*	104(5)	106(7)
$\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, %	97(1)	99(0)*	99(0)*	99(0)*	99(0)*	97(1)	99(0)*	99(0)*	99(2)*	99(2)*
HR, beats·min−1	71(10)	63(10)*	64(10)*	66(11)	68(9)	70(9)	58(9)*	55(8)*#	66(13)*	69(9)
MAP, mmHg	70(7)	81(10)*	87(6)*	70(12)	71(6)	64(8)#	78(7)*	81(8)*#	73(6)*	72(9)*

Values are presented as mean (SD); n = 13 (all males). CW, cold water; D80 and D160, 80 and 160 min into the dive; f_B, respiratory rate; HR, heart rate; MAP, mean arterial pressure; $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$, end-tidal CO₂ pressure; $\mathrm{P}{\mathrm{ET}}_{{\mathrm{O}}_2}$, end-tidal O₂ pressure; PD, postdive; PD60, 60 min postdive; PRE, predive; $\mathrm{S}{\mathrm{p}}_{{\mathrm{O}}_2}$, arterial oxygen saturation; TN, thermoneutral water; $\dot{\mathrm{V}}{\text{E}}_{\mathrm{BTPS}}$, minute ventilation (body temperature pressure saturated); Vt_BTPS, tidal volume (body temperature pressure saturated). Two-way repeated-measures analysis of variance (time × water temperature). If a significant interaction or main effect was found, post hoc multiple comparisons were completed with the least squared differences (LSD) procedure. *P < 0.05 from predive; #P ≤ 0.05 CW vs. TN.

Hemodynamics

We observed bradycardia during both the thermoneutral and cold water dives at 80 min. This is in line with previous work that shows that HR decreases during water immersion secondary to an increase in stroke volume while cardiac output is maintained (50). Although we did not measure peripheral vasoconstriction, it occurs during hyperoxemia and exposure to cold stress causing a rise in MAP (51, 52). Therefore, we also speculate that the attenuated HR response during the cold dive was also mediated by an arterial baroreflex-mediated reduction in HR (53). However, HR was not different from baseline at 160 min during the cold water dive, whereas it remained below baseline during the thermoneutral dive. This is likely due to the increase in metabolic heat production (i.e., shivering) during the cold dive.

Experimental Considerations

There are several experimental considerations that are applicable to our study. First, we hypothesized that cold water diving would result in augmented CO₂ retention from a cold-induced reduction in CB tonic activity. These hypotheses were developed based on in vitro (22) and animal model (e.g., cat) (23) studies. In vitro, chemoreceptor fibers bathed in a solution with a temperature range of 30–40°C resulted in an increase in discharge rate with rising temperatures (i.e., Q₁₀ effect) and decreased discharge rate with decreasing temperatures (22). Additionally, blood pumped from a cat at the common carotid artery to a heat exchanger and back to the carotid bifurcation resulted in a reduction in tonic CB chemoreceptor activity at a blood temperature of 32°C (23). To this end, the methodological differences have important implications for our findings. The dive conditions were selected based on recreational, commercial, and military-relevant water temperatures, thermal protection, and bottom times. Subjects wore a 7-mm wetsuit but were exposed to 15°C water for 4 h. Core body temperature did not decrease in the cold water condition, likely because of cutaneous vasoconstriction and increases in shivering and metabolic rate (i.e., thermogenesis). Therefore, hypothermia may be a requisite to observe reductions in CB tonic activity and sensitivity in vivo. Subjects were also immersed to the clavicle, and thus the carotid arteries were not directly exposed to cold water, which would occur during submersion and open-water diving. Therefore, it is unclear whether direct exposure to cold water would have attenuated CB tonic activity.

Second, we assessed CB chemosensitivity to hypercapnia (i.e., $\mathrm{C}{\mathrm{B}}_{\mathrm{C}{\mathrm{O}}_2}$) utilizing single-breath administrations of 13% CO₂ that resulted in homogeneous increases in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$. This approach is in line with previous work in our laboratory (19) and others (25). Moreover, although this method results in relatively homogeneous $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ values, administering a range of $\mathrm{F}{\mathrm{I}}_{\mathrm{C}{\mathrm{O}}_2}$ values at depth carries a number of contraindications. For instance, 1) a step or ramp increase in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$ values, rather than a single breath, likely would assess central chemosensitivity similar to a CO₂ rebreathe test, and/or 2) this alternative approach would result in administering gases containing greater $\mathrm{F}{\mathrm{I}}_{\mathrm{C}{\mathrm{O}}_2}$ or that could elicit greater increases in $\mathrm{P}{\mathrm{ET}}_{\mathrm{C}{\mathrm{O}}_2}$. Both of these would result in augmented CO₂ exposure and would increase the risk of CO₂ narcosis and/or O₂ toxicity during hyperbaric conditions.

Third, we did not include a normobaric time control trial in our study design, which precluded us from measuring test/retest variability across a multihour protocol within a given day with the present experimental setup. Within- and between-day variability of CB chemosensitivity has been previously reported (27) and should be noted as a consideration when interpreting the results of our study. Moreover, it is possible that such variability within and between trials could have masked any changes in CB chemosensitivity during and following the dive, as such presumed variability would require any changes caused by diving-related stressors (i.e., hyperoxia, water immersion, cold stress) to be substantial.

Finally, subjects breathed 100% O₂ before decompression to mitigate the risk of decompression illness and offload inert gases (e.g., N₂). Although breathing hyperoxia is known to alter chemosensitivity, we think it is unlikely that this safety measure had any impact on postdive measurements. This is particularly clear given that we did not see any impact of dry, hyperbaric exposure breathing either air (21% O₂) or 100% O₂ on postdive ventilatory responses to hypercapnia or hypoxia (18).

Perspectives and Significance

CO₂ retention in diving is a primary concern for military, commercial, and recreational divers. Therefore, identifying the mechanisms contributing to CO₂ retention is pivotal for developing mitigation strategies. Elevated arterial partial pressure of carbon dioxide ($\mathrm{P}{\mathrm{a}}_{\mathrm{C}{\mathrm{O}}_2}$) can result in cerebral vasodilation and increase the risk of central nervous system oxygen toxicity (54–56). Indeed, a 1-mmHg increase in $\mathrm{P}{\mathrm{a}}_{\mathrm{C}{\mathrm{O}}_2}$ results in a 3–5% increase in cerebral blood flow (57). This increase in $\mathrm{P}{\mathrm{a}}_{\mathrm{C}{\mathrm{O}}_2}$ (i.e., CO₂ retention) would be expected to be augmented at greater depths, resulting in an equivalently greater increase in cerebral blood flow. Additionally, CO₂ narcosis can impair both physical and psychological performance and result in confusion, disorientation, or loss of consciousness (6, 7). We have demonstrated that the ventilatory response to hypercapnia that is primarily mediated by the peripheral chemoreceptors does not appear to contribute to CO₂ retention in diving at 1.6 ATA in either thermoneutral or cold water. This is further supported by previous studies observing similar $\mathrm{P}{\mathrm{a}}_{\mathrm{C}{\mathrm{O}}_2}$ values at 4.7 ATA during cold (39) and thermoneutral (5, 8) water submersion.

In conclusion, we observed mild CO₂ retention during both thermoneutral and cold water dives at 1.6 ATA. The magnitude of the CO₂ retention was mild and likely indicates a rightward shift in the respiratory operating point associated with central hypervolemia (i.e., water immersion) and augmented cardiac output. Furthermore, carotid body chemosensitivity to hypercapnia and hypoxia were not attenuated during either dive condition, and therefore reductions in carotid body chemosensitivity do not appear to contribute to CO₂ retention.

GRANTS

This study was funded by Office of Naval Research Award N000141612954. Additionally, B. M. Clemency was supported by National Heart, Lung, and Blood Institute, National Institutes of Health award number K12 HL-138052 to the University at Buffalo and by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number UL1 TR-001412 to the University at Buffalo.

DISCLAIMERS

This article’s contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.

DISCLOSURES

B. M. Clemency is a speaker and consultant for Stryker Corporation. None of the other authors has any conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

H.W.H., D.H., and B.D.J. conceived and designed research; H.W.H., B.M.C., and E.S. performed experiments; H.W.H. and B.D.J. analyzed data; H.W.H., D.H., and B.D.J. interpreted results of experiments; H.W.H. prepared figures; H.W.H. drafted manuscript; H.W.H., D.H., B.M.C., E.S., and B.D.J. edited and revised manuscript; H.W.H., D.H., B.M.C., E.S., and B.D.J. approved final version of manuscript.