A Novel method for quantifying fluctuations in wearable derived daily cardiovascular parameters across the menstrual cycle

Abstract

Currently, knowledge of changes in cardiovascular function across the menstrual cycle and how these changes may inform upon underlying health is limited. Utilizing wrist-worn biometric data we developed a novel measure to quantify and investigate the cardiovascular fluctuation (i.e. cardiovascular amplitude) within resting heart rate (RHR) and heart rate variability (RMSSD) across 11,590 participants and 45,811 menstrual cycles. Within participants, RHR and RMSSD fluctuated in a regular pattern throughout the menstrual cycle, with population RHR_min and RMSSD_max at cycle day 5, RHR_max at day 26, and RMSSD_min at day 27. Cardiovascular amplitude was attenuated (p < 0.05) in older participants and participants using birth control, suggesting the novel metric may mirror differences in hormonal fluctuations in these cohorts. Longitudinal tracking of cardiovascular amplitude may offer accessible non-invasive monitoring of female physiology and underlying health across the menstrual cycle.

Fig. 1. Diagram of included participants.

Diagram of participant exclusion criteria illustrating the number of participants excluding at each step of data refinement. Two pathways are shown illustrating the naturally cycling cohort on the left side and the birth control pills cohort on the right side. In total 23,395 participants were assessed for eligibility; 9968 participants were included in the naturally cycling cohort and 1661 participants were included in the birth control pill cohort.

Fig. 2. Population-Level GAMM of RMSSD HRV offset and RHR offset during the menstrual cycle.

The partial dependence plot of the day of the menstrual cycle (D) from the population-level GAMM modeling the offset from the cycle-mean RHR and RMSSD in the naturally cycling cohort. Adjusted models are represented by the black dashed line, unadjusted models by the grey dotted line. The) models for RHR offset values are displayed in (a), models for RMSSD offset are displayed in (b). A red dashed line indicates the mean end of menstrual bleeding. Boundary lines indicate the 95% CI for estimates.

Fig. 3. Results of GLM for RHR_amp and RMSSD_amp.

The partial dependence of age is plotted for (a) RHR_amp and (d) RMSSD_amp and the partial dependence of the four spline terms of baseline RHR is plotted for RHR_amp (b) and the four spline terms of baseline RMSSD is plotted for RMSSD_amp (e). The partial dependence of BMI is plotted for (c) RHR_amp and (f) RMSSD_amp. Boundary lines indicate the 95% confidence interval for estimations.

Fig. 4. Distribution of novel cardiovascular amplitude metric for participants using a birth control pill and naturally cycling participants.

Histogram plots for the distributions of RHRamp values are shown in (a) and the distributions of RMSSDamp values are shown in (b). In both (a) and (b), the naturally cycling cohort is displayed in light gray with the cohort mean represented by the dashed (---) line, while the birth control pill cohort is displayed in dark gray with the cohort mean represented by the dotted (…) line.

Fig. 5. Visual methodology.

Visual exploration of the assignment of individual menstrual cycle measures of cardiovascular amplitude. The ‘Population Model’ frame illustrates the population GAMM trend of Resting Heart Rate offset across the menstrual cycle. The population minimum and maximum are identified, and the surrounding days are shaded, with the grey shaded region encompassing the range of the expected minimum values and the blue shaded range encompassing the range of the expected maximum values. The second frame, ‘Individual Cycle’ illustrates the application to a single menstrual cycle. The individual resting heart rate offset values are represented by open circles. The dashed lines in each shaded region correspond to the mean values of resting heart rate offsets in the grey and blue shaded regions respectively. In the final pane, ‘Cycle Cardiovascular Amplitude’, the values from the prior panel are subtracted from one another, with the mean offset of the expected maximum minus the mean offset of the expected minimum representing the final cardiovascular amplitude metric.