Assessing the impact of menstrual cycle phase on respiration rate, resting heart rate, heart rate variability, nocturnal oxygenation and sleep time using wearable devices

Background Prior studies have observed changes in parameters such as respiration rate, heart rate variability, and resting heart rate over the course of the menstrual cycle. However, prior studies have been limited in scale and have not considered concurrent variations in sleep duration or nocturnal oxygenation. Objectives To determine the variations in nocturnal respiration rate, heart rate variability, resting heart, main sleep duration and average nocturnal SpO 2 , associated with underlying menstrual cycle timing. Design This was a retrospective observational study based on wearable data and self-reported menstrual cycle timing and birth control usage. Methods Data was collected over 24 months in women using wearable devices (Fitbit and Pixel Watch) who also self-recorded their menstrual cycles and birth control usage. The study considered 1754 women on hormonal birth control (HBC) pills, and 7279 naturally cycling (NC) women. Self-reported menstrual cycle parameters were consistent with normal physiology, and participants recorded wearable data on greater than 70% of the nights. Results By averaging parameters relative to their day within a menstrual cycle, variations were seen in respiration rate (RR), resting heart rate (RHR) and heart rate variability (HRV), dependent on cycle phase. For the NC cohort, RR varied between the assumed follicular and luteal phase by approximately 0.6 breaths/min (∼4% variation, p <0.05), RHR by 3.0 bpm (∼4.4%, p <0.05), and HRV by 6 ms (∼14%, p <0.05). RR was highest in the assumed premenstrual phase, and lowest near the assumed ovulation date. There were also variations in main sleep time (7 minutes longer in menses than pre-menstrual, p <0.05), and in average nocturnal SpO 2 (absolute increase of ∼0.1% in assumed late luteal phase, p <0.05). Similar patterns were seen in the HBC group with reduced amplitude of variation. At a cohort level, increased respiration rate was observed in the HBC group (about 0.2 breaths/min, p <0.05), as was increased main sleep time by approximately 7 minutes. Conclusion The menstrual cycle has a small but statistically significant effect on a number of cardiorespiratory variables over a typical cycle. The effect size is largest for changes in resting heart rate, followed by HRV and respiration rate, with effects on main sleep time and SpO 2 being negligible.