Isocapnic CO₂ administration stabilizes breathing and eliminates apneas during sleep in obese mice exposed to hypoxia

Acute hypoxic exposure can precipitate central sleep apnea by inducing respiratory alkalosis and reducing carbon dioxide (CO₂) tension in the arterial blood toward apneic threshold [1–3]. Exposure to hypoxia at high altitude exacerbates sleep disordered breathing (SDB) with emerging central events superimposed on obstructive, which were present at low altitude [4, 5]. CO₂ may stabilize breathing in patients with central apneas [6]. However, the effect of isocapnic CO₂ administration on hypoxia-induced respiratory instability during sleep has not been systematically studied in obesity-related SDB. Our group has shown that diet-induced obese (DIO) mice develop SDB manifested by obesity hypoventilation and obstructive hypopneas, which are similar to that observed in obese humans [7]. We hypothesized that exposure to hypoxia will destabilize breathing during sleep in our mouse model of SDB, and the isocapnic administration of CO₂ during severe hypoxia will improve SDB.

Eight male C57BL/6J mice were fed with high-fat diet (60% kcal of fat) until they reached 20 weeks old (body weight: 45.1 ± 0.8 g). Mice were implanted with EEG/EMG electrodes and the sleep–wake stages were scored based on standard criteria of frequency and amplitude of EEG/EMG signals [7, 8]. Full polysomnography was performed using a whole-body plethysmography chamber as previously described [7, 8]. All mice underwent 2 days of sleep studies, separated by 1 day for recovery. In the first sleep study, mice were exposed to periods of isocapnic hypoxia by the continuous infusion of a gas mixture of 10% O₂ + 3% CO₂. We have shown that the addition of 3% CO₂ during hypoxia maintain a stable partial pressure of arterial CO₂ (PaCO₂) in mice [8]. In the second sleep study, mice were exposed to periods of poikilocapnic hypoxia (10% O₂ gas infusion). Mice were exposed to cycles of normoxia and hypoxia (1 h/each) from 10 AM to 4 PM. The order of normoxic and hypoxic episodes was randomized. Apneas were scored as reductions of ≥90% in the airflow for ≥0.5 s and classified as either obstructive or central based on the presence or absence of effort during these events. We quantified minute ventilation (V_E) across all epochs of sleep from the rectified moving average of flow. To examine whether fluctuations in V_E were driven by changes in tidal volume (V_T) or respiratory rate (RR), we performed breath detection analysis as previously reported [7, 8]. Respiratory stability was estimated using Poincaré analysis. We averaged V_E every 0.5 s. V_E data points (V_E(n+1)) were plotted against the preceding V_E value (V_E(n)) using scatterplots. In brief, the dispersion of the points perpendicular (SD1, short-term respiratory variability) and in parallel (SD2, long-term respiratory variability) to the line of identity was calculated. Short-term and long-term respiratory variabilities reflect rapid changes in ventilation (i.e. breath-to-breath variability) and oscillations across the entire sleep, respectively. All data were averaged for each sleep study and for each type of gas exposure in each mouse. Mice were euthanized at the end of the protocol by anesthetic overdose and cervical dislocation. The study was approved by the institutional Animal Care and Use Committee (#MO21M165). Data were analyzed in R (version 4.2.1, https://www.R-project.org/). A mixed-effects model with random intercepts was fitted to the data to account for repeated within-mouse measurements and differing baseline outcomes (P < 0.05 statistically significant).

Figure 1, A depicts the respiratory data analysis during NREM sleep. Isocapnic hypoxia, but not poikilocapnic hypoxia, significantly augmented V_E due to increases in both RR (Figure 1, B) and V_T (Figure 1, C). Isocapnic hypoxia improved SDB observed during poikilocapnic hypoxia abolishing an increase in the apnea index and attenuating hypoxemia (Figure 1, D and E). There was no evidence of obstructive apneas at any condition. Figure 1, F shows V_E-based Poincaré plots during NREM sleep from all mice. Poikilocapnic hypoxia destabilized breathing, which was evidenced from a greater short-term respiratory variability (Figure 1, G). In contrast, isocapnic hypoxia showed stable respiration with SD1 values similar to normoxia. Isocapnic hypoxia increased SD2 (Figure 1, H) and suppressed NREM and REM sleep (Figure 1, I-J). Poikilocapnic hypoxia did not affect NREM sleep duration, but it completely abolished REM sleep.

Figure 1.

Breathing during sleep and sleep architecture in diet-induced obesity (DIO) mice (n = 8) under normoxia (20.9% O₂), poikilocapnic hypoxia (10% O₂), and isocapnic hypoxia (10% O₂ + 3% CO₂). Isocapnic hypoxia, but not poikilocapnic hypoxia, significantly increased (A) minute ventilation (V_E), (B) respiratory rate (RR), and (C) tidal volume (V_T). Isocapnic hypoxia (D) reduced the apnea severity and (E) maintained oxyhemoglobin saturation (SpO₂) higher than poikilocapnic hypoxia. (F) Representative Poincaré plots combining all V_E data points from all mice during NREM sleep. (G) Poikilocapnic hypoxia destabilized breathing, indicated by increased SD1 value (short-term respiratory variability). Isocapnic hypoxia showed similar SD1 value to normoxia, but increased (H) SD2 (long-term respiratory variability). Isocapnic hypoxia suppressed (I) NREM and (J) REM sleep. Poikilocapnic hypoxia did not affect NREM sleep duration, but it completely abolished REM sleep. Respiratory data was analyzed only during NREM sleep. Data presented as median ± 1.5*IQR. Mixed effects model.

We showed that isocapnic administration of 3% CO₂ eliminated apneas, improved oxygen saturation, and mitigated hypoxia-related sleep architecture alterations in obese mice exposed to 10% O₂. Our study may have clinical implications, for instance during ascent to the high altitude. We exposed obese mice to 10% O₂, which corresponds to approximately 70–75 mmHg of partial pressure. Similar levels of hypoxic exposure are observed at the Mount Kilimanjaro (5971 m above sea level). Hypoxia at high altitude is associated with major physiological distress. Obesity increases the risk of acute high altitude illness and nocturnal desaturation [9]. Unlike human SDB, in which high altitude leads to central apneas superimposed on pre-existing obstructive apneas, in DIO mice, apneas were exclusively central during all gas challenges. Reduction in central apneas by CO₂ supplementation is likely attributable to an increase in PaCO₂ above the apnea threshold. CO₂ administration in patients corrects CO₂ oscillations and stabilizes breathing [6]. Acute exposure to high altitude is also associated with reduced sleep duration. In animal models, poikilocapnic hypoxia decreases REM sleep, which is partially restored with the addition of CO₂ [10]. In our study, poikilocapnic hypoxia abolished REM sleep in DIO mice and isocapnic administration of CO₂ restored REM sleep to approximately 15% of normoxia values. Our hypoxia protocol in DIO model reproduces some key features of acute altitude exposure in humans and shows beneficial effects of isocapnic CO₂ administration on short-term breathing stability, SDB reduction, and sleep architecture. Increased long-term breathing variability during isocapnic hypoxia is probably attributable to changes in sleep architecture with reduction in NREM sleep and an increase in wakefulness, when respiratory pattern is variable. In conclusion, rebreathing is a simple technique to increase inspired CO₂ levels. Calibrated rebreathing techniques may stabilize short-term breathing and improve hypoxemia diminishing severity of altitude sickness.

Data Availability

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Disclosure Statement

None declared.

Funding

This research was supported by American Heart Association Postdoctoral Fellowship Award 828142 (to LJK), NHLBI grant NIH R01 HL128970, R01 HL133100, and R01 HL13892 (to VYP), American Academy of Sleep Medicine Foundation 238-BS-20, and American Thoracic Society Unrestricted Award (to LVP).