Bio-Monitor Detects Reduced Obstructive Sleep Apnea and Susceptibility to Arrhythmia in Spaceflight
BACKGROUND: Attenuation of sleep disordered breathing in astronauts entering microgravity was previously reported during shuttle missions. However, the impacts of microgravity on obstructive sleep apnea (OSA) have not previously been quantified.
CASE REPORT: An astronaut with OSA participated in the noninvasive assessment of multiple physiological variables using the Bio-Monitor biometric shirt before, twice during, and after a 6-mo mission to the International Space Station. Cardiorespiratory variables, including electrocardiogram, respiratory rate and tidal volume, oxygen saturation, and activity, were measured and allowed for the assessment of apneic events and cardiac arrhythmias before, during, and after the mission.
DISCUSSION: While in the microgravity environment, apneic and desaturation events completely resolved, reoccurring immediately following landing. Associated electrophysiological features, including complete heart block, were not temporally associated with apneic or desaturation events when present, persisting even when OSA was absent. The participant’s use of hypnotic medication in the evening prior to data collection toward the end of flight may have contributed to sporadic apneic episodes, and to longer QT intervals exhibited at that specific timepoint.
Mastrandrea CJ, Rabineau J, Greaves D, Hughson RL. Bio-monitor detects reduced obstructive sleep apnea and susceptibility to arrhythmia in spaceflight. Aerosp Med Hum Perform. 2025; 96(12):1084–1089.
Obstructive sleep apnea (OSA) results from transmural pressures falling to a critical point within the compliant upper airways, causing their collapse and pauses in breathing (hypopneas and apneas). Disturbances in upper airway neuromuscular control and structure both predispose individuals in developing OSA; simultaneously respiratory pressures can attenuate or augment this collapsibility.1 Spaceflight provides a novel environment in which some of these airway deficiencies that lead to OSA can be investigated; microgravity offloads the soft tissues within the upper airway,2 significantly reducing tissue pressures. Importantly, atmospheric pressure aboard the International Space Station is maintained at sea-level pressures, with similar gas compositions, eliminating potential effects of changes in ambient pressure on OSA pathogenesis.
Previously, five shuttle astronauts exhibited a reduction in the number of apnea-hypopnea events and elimination of snoring during sleep in microgravity, with an associated attenuation in the number of night arousals.2 These findings confirmed the effectiveness of microgravity in reducing the number of obstructive respiratory events; however, it did so in participants with mild apnea-hypopnea indices (AHI, 8.3/h during preflight data collection) and “high” minimum oxygen saturation of 95% during sleep. Quantification of AHI defines mild levels at 5–14 events/h, moderate as 15–30, and severe as >30.3,4 Here we report the effects of microgravity on OSA in one individual with moderate OSA and far greater reductions in oxygen saturation during sleep. We also report on cardiac rhythm disturbances during sleep.
CASE REPORT
As included in their science complement during a 6-mo mission to the International Space Station, the astronaut participated in continual 48-h recording of physiological signals at five timepoints: 6 mo before launch (Pre); early in flight (flight-day 30, Early); late in flight (landing −30 d, Late); immediately following landing (recovery day 0, R0); and 6 mo following landing (R180). Two sleep periods of data were collected for each time point and are designated with the suffix 1 and 2, respectively. Data were recorded with the Bio-Monitor biometric garment (manufactured for the Canadian Space Agency by Carré Technologies, Montreal, Canada; Fig. 1). Respiratory inductance plethysmography signals were calibrated in the upright and supine positions via a two-step procedure that included isovolumetric and bag-rebreathing maneuvers.5 At recruitment, the astronaut participant was in good health and certified to fly, having cleared all regular medical and physical checks and requirements. The astronaut subsequently developed OSA. The study protocol was approved in advance by five boards: the University of Waterloo Office of Research Ethics (ORE #31453), Johnson Space Center Committee for the Protection of Human Subjects, NASA Human Research Medical Review Board, the European Space Agency Medical Review Board, and the Japanese Space Agency Research Ethics Board (NASA IRB Pro2887), in accordance with the Declaration of Helsinki. Voluntary written informed consent was collected prior to commencement of data collection.
Citation: Aerospace Medicine and Human Performance 96, 12; 10.3357/AMHP.6745.2025
Unprocessed data from each sensor were analyzed in house using custom Matlab® algorithms (Matlab version 2024b, Mathworks Inc., Natick, MA, United States). Periods of sleep were identified by time of day and comparison of activity, heart rate, and breathing patterns. Derived signals included R-R timings, heart rate, breathing rate, tidal volume, oxygen saturation, and activity. Apneic events longer than 10 s were summated and divided by the duration of sleep, with the quotient providing the number of apneic events per hour. The start and end point of each apneic event was identified manually, using calibrated lung volume signals, as the period between cessation of breathing and recommencement of breathing. Desaturation events for entire sleep periods were recorded as the number of times oxygen concentrations fell by greater than 3% and returned to at least 1% below baseline, or 3% above the minimum oxygen saturation during the event, and lasted for between 10–60 s.6 The total number of events that achieved levels below 90% saturation were additionally highlighted. Total time below an Spo2 of 90% (T90) was recorded as a proportion of total sleep time (Table I).
Fig. 2 provides an example of the repeated apneic patterns exhibited by our participant. These patterns included:
Paroxysmal breathing with antiphasic thoracic and abdominal respiratory signals.
Repeated cycles of complete cessation of air movement (lack of change in lung volume), followed by sudden onset of deep breathing where tidal volumes reached 2–3 L. Following this, gradual diminution of volume continued until complete cessation of breathing occurred again.
Capillary oxygen desaturation that commenced shortly after cessation of breathing and only resolved once breathing recommenced.
Heart rate elevations associated with the sudden onset of deep breathing, with return to baseline during each cycle.
Increased activity that synchronized with the commencement of the deep breathing following each apneic event.
Citation: Aerospace Medicine and Human Performance 96, 12; 10.3357/AMHP.6745.2025
Continuous electrocardiogram signals were also analyzed and we identified electrophysiological abnormalities during sleep before, during, and after spaceflight that included: premature supraventricular and premature ventricular ectopics, bigeminy, and complete heart block. Visible P-waves without subsequent QRS complexes were noted for one or two consecutive cycles, identifying presence of complete heart block episodes that occurred in all data collections except R180. The largest occurrence of such arrhythmias was noted at Late 1, with seven complete heart blocks during one night (Table I). The frequencies of apneic and complete heart block events are shown in Fig. 3A. The QT timings were measured based on the mean lead II ECG waveform computed on successive windows of 5 min with Kubios HRV Scientific (v. 4.1.1, Kubios Oy, Kuopio, Finland). The QT intervals were compared to the threshold suggested by Chan et al.7 The participant exhibited prolonged QT above this threshold for risk of bradyarrhythmia and torsades de pointes during multiple sleep periods throughout the mission (Fig. 3B).
Citation: Aerospace Medicine and Human Performance 96, 12; 10.3357/AMHP.6745.2025
Work, activity, and drug logs were available for Early, Late, and R0 timepoints. On both days prior to sleep at Early 1 and Early 2, the astronaut performed two exercise sessions each day, including resistive and cycle exercises. No medications were prescribed or used during this period. Prior to sleep at Late 1 and Late 2, the astronaut performed two exercise sessions each day; these included resistive, cycle, and treadmill exercises. It was noted that the astronaut took 10 mg of modafinil the morning preceding the Late 1 data collection to aid with alertness and undertaking activities requiring a high level of concentration. Subsequently in the evening of the Late 1 timepoint, the astronaut took 10 mg of zaleplon, a sedative and hypnotic, to aid with sleep. Prior to landing (several hours prior to the R0 collection), the astronaut underwent a fluid loading regimen that included salt tablets and oral and intravenous fluids and took 25 mg meclizine for motion sickness. We did not have access to activity or treatment/medication records for the Pre or R180 timepoint.
Obstructive sleep apnea was not present during Early in-flight sleep and occurred rarely at the Late in-flight timepoint. In contrast, apneas were present during all terrestrial collections (Table I).
DISCUSSION
This study used the Bio-Monitor garment to obtain high-fidelity, continuous, and synchronous recordings from complementary sensors in an unintrusive and wireless method; the combination of ECG, respiration, oxygen saturation, and activity provide a rich dataset for the interrogation of physiological variables. Here we provide results from data obtained using the hardware on the ground and aboard the International Space Station, supporting its use as an important research tool for ongoing physiological studies in current and future spaceflight missions.
In support of previous findings in astronauts with mild pre-existing sleep-disordered breathing,2 microgravity significantly attenuated features of OSA in our participant. At the preflight timepoint, our participant exhibited over 19 apneic events lasting at least 10 s/h and 13.4 oxygen desaturations of >3% Spo2/h. Gravity clearly plays a major role in the pathogenesis of OSA in terrestrial environments, as during sleep periods of two consecutive nights at the Early in-flight timepoint, there was complete resolution of apneic and oxygen desaturation episodes, with reduction in activity counts from an accelerometer located near the hip, indicating reduced arousal events. Importantly, at the Late in-flight timepoint, the re-establishment of mild apneic and sporadic oxygen desaturation events could be the result of our participant’s use of hypnotic medications (zaleplon), as these do exhibit muscle relaxant properties. Therefore, while we present evidence of gravity’s major role in causing OSA, drug-induced alterations to myogenic activity within the upper airways may also contribute to the development of OSA.
Immediately following landing, re-establishment of our participant’s pre-existing OSA was apparent, with the postflight sleep periods exhibiting the greatest number of desaturation events where Spo2 fell below 90% compared to other timepoints; the participant also experienced the lowest oxygen saturation (75%) during R0 1, and greatest T90 (17.2%) during R0 2. By R180, it appeared that the prevalence of OSA features had returned to preflight levels.
While it is known that complete heart block is associated with OSA, there appears to be no identifiable correlation between the severity of OSA, as defined by AHI, and the predisposition for complete heart block.8 However, discourse concerning the magnitude of desaturation induced by OSA-induced desaturation and likelihood of arrhythmic events, including complete heart block, exists.8,9 The literature suggests that approximately 60% of all third-degree AV blocks occur when arterial oxygen concentrations fall below 72%.9 Our participant experienced complete heart block at all timepoints other than R180, exhibiting the highest frequency of these episodes during spaceflight, when oxygen saturation was more stable and generally much higher than during terrestrial recordings. The temporal relationships between these cardiac events and apneic/hypoxic episodes were also not readily evident. In terrestrial settings, resolution of bradyarrhythmia following treatment of OSA with positive airway pressure10 might reduce and/or eliminate vagal activation during desaturation events that would otherwise lead to bradyarrhythmia. The persistence of third-degree heart block in our participant during flight, despite the microgravity-induced resolution of hypoxemia, may indicate additional effects of spaceflight that prevent complete resolution of heart block. For example, autonomic control in spaceflight is altered, including changes to vagally mediated reflexes,11 but further work would be required to assess this robustly. Importantly, zaleplon was used in the evening preceding Late 1, when the frequency of complete heart block was greatest; QT assessment identified both Late 1 and Late 2 as above the threshold for significant risk of bradyarrhythmia and torsades de pointes,7 which could indicate that zaleplon may have contributed to the cardiac changes we identified. Furthering this avenue of investigation in astronauts is important given the frequent use of such hypnotic medications during spaceflight.
Previously, concerns about the levels of carbon dioxide aboard the International Space Station were raised in relation to the increased frequency of headaches during flight,12 with the suggestion that pockets of carbon dioxide may form around sleeping astronauts. As loop-gain plays an important role in the propagation of cycles of hypopnea and apnea in those with OSA, it would be interesting to evaluate the role of carbon dioxide concentration in the immediate environment around sleeping astronauts. Unfortunately, given the nature of the unexpected findings described in this case study, we did not foresee the need to measure carbon dioxide levels. Nevertheless, it is reported that the crew quarters aboard the International Space Station incorporate adequate air flow in the immediate vicinity of sleeping astronauts, and accumulation of carbon dioxide should be minimal.
With the increasing frequency of private travel to microgravity and space environments, investigating the effects of this novel environment on chronic medical conditions such as OSA ethically demands additional support. However, findings from this case report and previous shuttle investigations that OSA events diminish or resolve completely upon entering microgravity is reassuring. Our findings also serve as support for professional astronauts and their selection due to the diminished risk of OSA related to anatomical upper airway predisposition. Finally, although risk for arrhythmia caused by microgravity has been questioned,13 we identified persistent arrhythmic risk that continued for some time after resolution of OSA and may indicate the need for ongoing monitoring of arrhythmic events in individuals with OSA and identified arrhythmic sequelae.14
Components of the Bio-Monitor hardware. Customized and gender-specific shirts (female left, male right) are fitted for each crewmember. Both versions incorporate five sensors as shown. Data are recorded to flash memory located inside the Data Unit via a wired connection. The Data Unit also houses the haptic alarm, batteries, and the LED indicator lights. Figure created by authors using publicly available information.
Periods of identified sleep apnea at the preflight timepoint. Top: thoracic (black) and abdominal (red) respiratory inductance plethysmography signals showing in phase patterns associated with air movement into and out of the lung and periods of apnea where thoracic and abdominal signals were antiphasic, resulting in no net air movement as indicated directly below in the middle panel. Middle: lung respiratory volume change computed from combined thoracic and abdominal signals (left axis, black), and oxygen saturation (right axis, red). Bottom: heart rate (left axis, black) and movement (right axis, red) showing activity and cardiac acceleration associated with apnea.
A) Rate of apneic events (left y-axis, solid black circles) and number of complete heart block events (right y-axis, empty red squares) occurring during each sleep period. B) Median QT and interquartile range measurements based on the mean lead II ECG waveform computed on successive windows of 5 min. The quality of the lead II ECG signals was poor and prevented accurate QT assessment on R0 2. The results during each sleep period of the five study timepoints are superimposed against the threshold for risk (dashed blue line) of torsades de pointes described by Chan et al.7 QT values above the line indicate increased risk of arrhythmia.
Contributor Notes
