In this work, we aimed to study HRV during arousal-free sleep periods in a homogenous patient sample (free from medications, co-morbidities and any other sleep disorders) to analyze acute, very-short term, autonomic effects of CPAP treatment on severe OSA (AHI>30), and its differences depending on gender and obesity.
CPAP treatment has shown to improve HRV in OSA patients 
and HRV is considered a more sensitive parameter than HRM to detect changes in ANS. Our results showed that, even in the first night with CPAP, the HRM was similar to the basal night but the HRV decreased significantly, considering all patients and sleep stages. Only patients with total normalization of the apnea index and oxygen values were included in the study; therefore, this result may be regarded as highly significant.
Given that, according to previous research, HRV is a good indicator of ANS activity, it is likely that CPAP treatment is able to reduce cardiac autonomic dysfunction in a very short time span. The results of this study also suggest that it may not be completely necessary for researchers and clinicians to wait months in order to see significant improvements in HRV.
Another of our aims was to see if the improvement of HRV was limited to either REM or NREM. According to our results, significant improvements in HRV were more relevant in NREM, even though some improvements were seen in REM sleep after CPAP treatment. A previous study suggests that increased PNS activity during NREM may be a compensating mechanism to SatO2
fluctuations, which REM sleep disrupts 
REM is a unique sleep stage from a physiological point of view when compared to NREM sleep stages because the heart rate and breathing rate are similar to WAKE. However, the results of this study did not support this vision, although our findings may be a consequence of the high severity of OSA and obesity seen in the patient group. The lack of significant changes in HRV in REM after CPAP treatment could be explained by the REM interference theory in PNS activity. The clinical implication of these results may suggest that patients who suffer from OSA with apneas occurring predominantly during REM sleep, may not enjoy the same cardiovascular benefits compared to OSA patients whose apneic episodes are scattered between REM and NREM. However, further studies are required to determine whether the unique properties of REM sleep influence autonomic function in patients with OSA.
The heart rate mean did not decrease significantly from NREM to REM during CPAP studies. This result is similar to previous studies in healthy individuals 
. The decrease HRV from NREM to REM was seen in basal studies, which suggests that OSA may influence the variance of the heart rate, but does not vary the natural change of overall beats per minute from NREM to REM. The lack of change in HRM during the CPAP study suggests that CPAP treatment lowers the variance of heart rate during NREM, as well as eliminates the change in the heart rate between NREM and REM.
Previous studies showed obese patients to have lower variations in total HRV 
. Our results also showed significant differences in HRV in obese patients during REM when compared to the non-obese group in both CPAP and basal studies. From these results, it seems plausible that CPAP treatment may improve HRV in obese patients, sufficiently so as to resemble the improved HRV seen in non-obese patients with CPAP treatment. In our study, another aim was to study the effects of obesity on CPAP treatment in severe OSA (AHI>30) patients, as CPAP treatment has shown to lower night time blood pressure and increase SNS activity in OSA patients 
. The results from our study showed that obesity does not influence HRV or HRM during CPAP treatment when compared to non-obese patients. Based on these results, obesity does seem to influence heart rate mean in WAKE and REM, but not enough to be statistically significant. Therefore, from a clinical perspective, additional or differential forms of treatment do not seem necessary when combating ANS dysfunction in obese patients.
The final aim of our study was to determine the role of gender in HRV during short-term CPAP treatment of severe OSA (AHI>30) patients, as women with breathing disorders have been shown to have increased SNS activity during NREM, and men with breathing disorders have been shown to have lower PNS activity during wake 
, and women with high AHI had low SNS activation during REM 
. Our results did show a difference in HRV between men and women in wakefulness. In addition, HRM was different in all the sleep stages. Our results showed a significant difference in HRV and heart rate mean between our male and female groups during wake. However, the changes were similar in the basal and in the CPAP nights, which suggests that gender does not influence the improvements seen in ANS activity from CPAP treatment. Nevertheless, the absence of statistical significance when assessing the effects of gender and obesity on HRV may be a consequence of the small sample size of our study, which is one of its limitations.
The main limitation of our study is a consequence of its transversal nature. In these regard, there is a possibility that changes in HRV during acute CPAP treatment may reflect a normalization of the respiratory pattern or may be due to changes in venous return secondary to the dramatic changes of intrathoracic pressure that occur during the apneic event rather than changes in cardiovascular control mechanisms. Thus, caution must be used when analyzing our results, as the effect of OSA in autonomic activity may perhaps more likely to be detected by comparing HRV on the first night of treatment with a sleep study on CPAP after weeks of treatment, and not with a single-night study, such as the present one. We neither studied whether the changes in HRV were associated with the severity of OSA, as we only selected patients with an AHI>30. Moreover, we do not provide data regarding the percentage of stage 2 and stage 3 sleep in the NREM segments, in spite of the fact that several studies show that ANS balance differs between these 2 stages. Finally, we did not study frequency-domain measures (i.e. HF, LF, and VLF); hence, further studies may be needed to understand our findings in depth.
Nonetheless, all these limitations do not hamper the fact that cardiac variability improves as an acute effect, independently of gender or weight, in the first night of CPAP use in severe OSA patients. Thus, we think that the CPAP treatment should not be delayed. Severe OSA patients should be advised that even a single night without CPAP has changes in the cardiac rate, which are corrected with CPAP. Hence, to use or not to use the CPAP for a single night does matter.