Confinement, partial sleep deprivation and defined physical activity–influence on cardiorespiratory regulation and capacity

The following inclusion criteria were applied: age range from 26 to 60 years, technical skills demonstrated through professional experience, motivation and work ethics similar to the current astronaut population, psychological qualification, at least bachelor’s degree in engineering, biological science, physical science or mathematics. Ethical approval was obtained from the Institutional Review Board at the NASA Johnson Space Center (Protocol number: Pro2320) as well as the Ethical Committee of the German Sport University Cologne (Protocol number: 074/2016), and the experiments were conducted in accordance with the Declaration of Helsinki (including its amendments until 2013). Written informed consent was available from all participants prior to the experiments.

Overall, the HERA campaign four consisted of five 45 d missions, including four crew members each (N = 20). Due to Hurricane Harvey in August 2017, one mission with four participants had to be terminated early, and it was not possible to analyze these data sets. Further, data of two subjects could not be included for data analysis, because they had to terminate one of the exercise tests during the mission early. Both individuals exceeded the allowed HR or blood pressure values during the moderate exercise test protocol set by NASA medical board. Therefore, data of 14 individuals (9 males, 5 females) were included for statistical analyses. Anthropometric data of the subjects are shown in Table 1.

Every second day, the crew exercised either on a cycle ergometer or they performed toning and stretching exercises over a time period of 30 min, while the allowed HR during exercise was restricted to below 85% of their age adjusted maximum.

The subjects filled in a questionnaire about the general frequency of their exercising habits and their physical activities during the year before the first exercise test prior to the beginning of the mission.

Before, during and after the mission, a standardized exercise test was applied to determine the kinetics responses in the moderate exercise range. The tests were scheduled eight days before the start of the mission (MD-8), on day 22 (MD22) and day 42 (MD42) of the mission, as well as four days after the mission ended (MD + 4). Eight hours ahead of the exercise test, the subjects stopped to consume alcohol and nicotine eight hours before, they stopped drinking caffeine four hours prior to the test and they started fasting two hours ahead of the beginning of the test. Additionally, eight hours before the test, no exercise was permitted and 24 h prior to the test session no maximal exercise was allowed. The exercise tests on MD22 were scheduled on day five of a sleep restriction period and MD42 was day four of a sleep restriction phase.

The exercise test was performed on a cycle ergometer (Lode Corival, Lode BV, Groningen, The Netherlands) and consisted of two parts: the first part started with the resting condition, followed by a 300 s phase at a low constant exercise intensity (30 W), which was followed by two PRBS with dynamic WR changes (30 and 80 W) for 300 s each. Thereafter, a constant phase of the higher WR (80 W) was applied for 300 s. The second part of the WR protocol was only performed on MD-8 and MD + 4. On these days, the WR was further increased to 100 W for one minute and afterwards WR was increased by 25 W every minute until fatigue (Fig. 1A).

During all tests, gas exchange was measured breath-by-breath, using a metabolic cart (Metalyzer 3B, Cortex Biophysik GmbH, Leipzig, Germany), including the corrections of Beaver et al. (1981) for alveolar gas exchange. The metabolic cart was calibrated according to the manufacturer’s guidelines, prior to all tests. HR was measured using an ECG (CustoGuard belt 3, CustoMed, Ottobrunn, Germany) and blood pressure was recorded beat-to-beat via Portpares Model 2 (Finapres Medical Systems, Amsterdam, The Netherlands). Using the blood pressure values, stroke volume (SV) was calculated applying the Modelflow Algorithm of the beatscope software (Finapres Medical Systems, Amsterdam, The Netherlands), considering age, height and body mass.

Breath-by-breath data were interpolated step-wise and beat-to-beat data linearly to 1-s-intervals to obtain homogenous sampling.

Peak data of \({\dot{\text{V}}\text{O}}_{{2}}\), HR and WR before and after the mission are reported as averages over the last 30 s before the termination of the WR protocol. The first VT (VT₁) was determined using the V-slope method in combination with the ventilatory equivalents and endtidal values of O₂ and CO₂, as suggested by Beaver et al. (1986).

To receive the kinetics information of the data, time-series analysis was applied. The PRBS signal was auto-correlated (ACF) and interpreted as a WR impulse (Bennett et al. 1981). The Cross-correlation functions (CCFs) between WR and the respective parameters as i.e., HR and pulmonary \({\dot{\text{V}}\text{O}}_{{2}}\) (\({\dot{\text{V}}\text{O}}_{{{\text{2pulm}}}}\)) were calculated and interpreted as the parameters’ response to this WR impulse. The higher the maximum of the CCF (CCF_max), the faster was the kinetics response. The greater the lag between the CCF_max and the maximum of the ACF (CCF_lag), the more time delayed was the onset of the response (compare Fig. 1B). Muscular \({\dot{\text{V}}\text{O}}_{{2}}\) (\({\dot{\text{V}}\text{O}}_{{{\text{2musc}}}}\)) was estimated from HR and \(\dot{V}\) O_2pulm, using the approach of Hoffmann et al. (2013). In this method, a specific venous volume in combination with \({\dot{\text{V}}\text{O}}_{{2}}\) and perfusion of the remainder of the body is estimated to calculate transit times of the \({\dot{\text{V}}\text{O}}_{{{\text{2musc}}}}\) signal from muscle to mouth.

Normal distribution of the respective parameters was tested applying the Kolmogorov–Smirnov test. If normal distribution could be assumed, analysis of variance (ANOVA) with repeated measures on the factor day (MD-8, MD22, MD42, MD + 4) and Bonferroni post hoc tests were calculated for the CCF_lag and CCF_max values. In case normal distribution could not be assumed, Friedman tests with Wilcoxon post hoc tests were used and the False Discovery Rate correction procedure (Benjamini and Hochberg 1995) for multiple testing was applied. ANOVA with the factors day (MD-8, MD22, MD42, MD + 4) and step [Rest, 30 W, 53.3 W (2xPRBS), 80 W] with repeated measures on both factors were applied to compare the mean values of HR, \({\dot{\text{V}}\text{O}}_{{{\text{2pulm}}}}\), mean arterial blood pressure (mBP), SV and cardiac output (\({\dot{\text{Q}}}\)). In case sphericity could not be assumed, the Huynh–Feldt correction was used to assess the inner subject effects. Bonferroni post hoc tests were applied, if appropriate. The Pearson product-moment correlation coefficient (one-tailed) was calculated for normally distributed data to test for correlations. The level of significance was set to α = 5%. All statistical tests were performed using SPSS 26 (IBM, Amonk, New York, USA).

The \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) values did not change significantly from MD-8 to MD + 4. Therefore, the applied exercise program was sufficient to mitigate the potentially deconditioning effects of lower physical activity throughout the 45 d of confinement in combination with sleep restrictions. However, the small but significant increase in WR of 20 W at VT₁ after the end of the mission (MD + 4), suggests a slightly increased exercise tolerance. Similar results were found for a submarine crew, living in confinement for 70 d. \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) did not change significantly, but VT₁ as a percentage of \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) increased after 8 weeks of endurance exercise training, which is comparable to the effects of the exercise training program applied during HERA C4 (Bennett et al. 1985). However, correlation analyses revealed, that HERA crew members with greater \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) and faster \({\dot{\text{V}}\text{O}}_{{{\text{2musc}}}}\) kinetics before the mission showed greater declines in the respective parameters after the mission. Moraes et al. (2018), analyzing \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) values of differently trained crew members (mountaineers and scientists) before and after an Antarctic field expedition reported comparable findings. Similar results were reported for Astronauts during spaceflight (Moore et al. 2014): the \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) did not decrease in all crew members, but those starting with higher levels of \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) were more prone to losses at the first inflight test, and those who maintained their preflight \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) values after return to Earth, spent about 80% of their inflight aerobic training time at ~ 80% of their maximum HR (Moore et al. 2014). For the HERA crew, training prescriptions should have been more customized for the individual training status of the crew members, to apply individually adequate training stimuli throughout the mission phase. Highly trained individuals need a higher training volume to sustain their personal fitness level, which is also known from longitudinal ageing studies (Katzel et al. 2001).

The training effect, indicated by an increased WR at VT₁ did not result in faster \({\dot{\text{V}}\text{O}}_{{{\text{2musc}}}}\) kinetics. This suggests different timelines for the adjustment of peak exercise capacities (\({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\)), submaximal aerobic capacities (VT₁) and regulatory parameters (kinetics) at muscular level.

In comparison with the pre and post-tests, mean HR values during the constant phases of the WR protocol during the mission phase were lower. However, HR kinetics at MD-8 were slower compared with MD22, MD42 and MD + 4. Hence, changes in the absolute HR values during the WR step and the speed of adjustment of HR (kinetics) in response to increased metabolic demands are not similar. Regarding the altered regulatory behavior of HR, a training effect of the exercise countermeasure can be concluded, which is not evident in the mean values of HR. Therefore, two aspects should be considered: (1) a training effect via the exercise countermeasure and (2) autonomic system changes caused by reduced external stimuli. Regarding the effect of the exercise countermeasure (1), the accelerated HR kinetics starting at MD22 are in accordance with the increased WR at VT1 after the end of the mission (MD + 4). For comparison, after bed rest, HR kinetics were significantly slowed as a result of deconditioning (Koschate et al. 2018) and were correlated with changes in \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\) in astronauts before and after their mission to the International Space Station (Hoffmann et al. 2016). After 12 weeks of endurance exercise training, HR kinetics were accelerated in a group of male type 2 diabetes patients (Koschate et al. 2016a). Hence, HR kinetics seem to be a sensitive parameter to observe changes in cardiovascular regulation in response to exercise training stimuli. These training effects were not visible in the mean values of HR during the exercise test throughout the HERA missions. Considering the potential changes in the autonomic nervous system due to the confinement phase (2), either a higher parasympathetic system activity during the mission or an increased sympathetic system activity before and after the mission (Coote 2010) should be considered, which might (although not statistically significant between MD42 and MD + 4) also slightly dampen the potential training effect, which was observed for HR kinetics at MD42. During Russian confinement experiments with durations of 120 and 520 d (Vigo et al. 2012, 2013), a greater HRV and reduced cortical activation levels were reported (Weber et al. 2020; Jacubowski et al. 2015). Greater HRV was suggested to be influenced by boredom, artificial light or greater parasympathetic system activity during isolation (Vigo et al. 2012, 2013). These circumstances might have altered the exercise HR during the HERA C4 missions as well.

In experiments, applying acute and chronic (partial) sleep deprivation, reduced HR variability, higher HR, increases in sympathovagal activity and decreases in vagal activity in the resting condition and/or during sleep were described (Tobaldini et al. 2017a, b; Dettoni et al. 2012; Glos et al. 2014; Fogt et al. 2011; Liu et al. 2015). Dettoni et al. (2012) found increased sympathetic activity in the resting supine position after only five nights of partial sleep restriction, which is comparable to the HERA campaign (5 h). Using linear mixed-effects models, significant linear relationships were found between fatigue level and HRV (Fogt et al. 2011). This is in contrast to the observed cardiovascular data during exercise in this experiment. The findings of the present experiment, supported by the data of Vigo et al. (2012, 2013) suggest, that the effect of confinement and reduced external stimuli in combination with the applied exercise countermeasure during the simulated mission to an asteroid might have had a greater effect on the cardiovascular system, than partial sleep deprivation. Therefore, partial sleep deprivation, at least when combined with the applied exercise countermeasure does not yield negative effects regarding the regulation of the cardiovascular system at moderate WR intensities. In line with this, no correlations were found between baseline \({\dot{\text{V}}\text{O}}_{{{\text{2peak}}}}\), \({\dot{\text{V}}\text{O}}_{{{\text{2musc}}}}\) or pre to post changes in these parameters with HR kinetics.

No further parameters to explain the lower HRs during the mission compared with the pre and post mission values were assessed. Therefore, it remains unclear whether the sympathetic nervous system activity was increased before or decreased during the mission and vice versa, whether the parasympathetic system activity was higher during or lower before and after the mission. The effect of confinement on cardiorespiratory fitness, cannot be evaluated separately from the effect of sleep restrictions, since no test during the mission was performed after recovery sleep.

Two participants had to terminate the test early, because they exceeded threshold values for HR or blood pressure, which were set by the NASA medical board. Hence, for future confinement experiments, a further reduction of the applied WR intensities should be considered.