Beat-to-beat blood pressure variability in patients with obstructive sleep apnea

Abstract

Study objectives: Cardiovascular comorbidities in obstructive sleep apnea (OSA) are difficult to treat, perhaps due to autonomic dysfunction. We assessed beat-to-beat blood pressure (BP) variability (BPV) in OSA while considering other markers derived from electrocardiogram and continuous BP signals.

Methods: We studied 66 participants (33 participants with OSA: respiratory event index [mean ± SEM]: 21.1 ± 2.7 events/h; 12 females, aged 51.5 ± 2.4 years; body mass index: 32.8 ± 1.4 kg/m²; 33 healthy controls: 20 females; aged 45.3 ± 2.4 years; body mass index: 26.3 ± 0.7 kg/m²). We collected 5-minute resting noninvasive beat-to-beat BP and electrocardiogram values. From BP, we derived systolic, diastolic, and mean BP values, and calculated variability as standard deviations (systolic BPV, diastolic BPV, BPV). We also calculated diastole-to-systole time (time to peak). From the electrocardiogram, we derived QRS markers and calculated heart rate and heart rate variability. We performed a multivariate analysis of variance based on sex and group (OSA vs control), with Bonferroni-corrected post hoc comparisons (P ≤ .05) between groups. We calculated correlations of BPV with biological variables.

Results: Multivariate analysis of variance showed effects of diastolic BPV and BPV in OSA; post hoc comparisons revealed high diastolic BPV and BPV only in female participants with OSA vs controls. QRS duration was higher in OSA, with post hoc comparisons showing the effect only in males. BPV correlated positively with heart rate variability in controls but not in participants with OSA. BPV correlated positively with time to peak in females with OSA and OSA combined, whereas there was no BPV-time-to-peak correlation in healthy participants.

Conclusions: The findings show sex-specific autonomic dysfunction reflected in beat-to-beat BP in OSA. The higher BPV may reflect poor baroreflex control or vascular damage in OSA, which are potential precursors to cardiovascular complications.

Keywords: autonomic nervous system; baroreflex; cardiovascular disease; sleep-disordered breathing.

Conflict of interest statement

All authors have seen and approved the manuscript. Work for this study was performed at the University of California, Los Angeles. This study was funded by the National Institute of Nursing Research NR-017435 and National Institute of Heart, Lung, and Blood Institute HL-135562. The authors report no conflicts of interest.

© 2021 American Academy of Sleep Medicine.

Figures

Figure 1. Protocol timeline.

Sequence, participant position, and duration of procedures. Green indicates the 5-minute-rest recording period. Participants had back support in a “chair” but not at a “desk.” BP = blood pressure; CNAP = continuous noninvasive arterial pressure; ECG = electrocardiogram.

Figure 2. Box plots of demographics and biomarkers.

Box plots indicating median, interquartile range, range, and outliers (see Methods). Demographic and biomarkers from Table 1 for OSA and control in the combined-sex population and separately in males and in females. *Significant OSA vs control difference by t test. (A) Age, BMI, and OSA sleep parameters. (B) Brachial BP measures. AHI = apnea-hypopnea index; BMI = body mass index; BP = blood pressure; DBP = diastolic blood pressure; MBP = mean blood pressure; OSA = obstructive sleep apnea; SBP = systolic blood pressure; SpO2 = saturation of oxygen obtained from pulse oximeter.

Figure 3. Physiological variables.

Box plots indicating median, interquartile range, range, and outliers (see Methods). (A) BPV, BP, CO, and TTPK obtained from the continuous BP. MBP obtained from the brachial cuff (y axis) and the average MBP from 5 minutes of CNAP (x axis) are shown. (B) HR, QRS duration and HRV, and sympathetic-to-vagal balance from the ECG are shown. *Significant OSA vs control difference in the MANOVA model (refer to Table 2). Regression lines in A are thick for significant correlations. BP = blood pressure; BPV = mean arterial blood pressure variability; CNAP = continuous noninvasive arterial pressure; CO = cardiac output; DBP = diastolic blood pressure; ECG = electrocardiogram; HR = heart rate; HRV = heart rate variability; MBP = mean arterial blood pressure; OSA = obstructive sleep apnea; RRI = RR interval; SBP = systolic blood pressure; TTPK = time to peak.

Figure 4. Scatterplots of BPV with demographic and biological measures.

BPV plotted against (A) age, BMI, brachial MBP; (B) averaged BP and TTPK, CO; (C) heart rate, QRS duration, HRV (sympathetic to vagal balance, RR-I variability). The regressions in each group of OSA (in blue) or control (in red) are shown as solid where significant. Correlations significantly different between OSA and control groups are highlighted in gray. The correlation results are shown in Table 3. BMI = body mass index; BP = blood pressure; BPV = mean arterial blood pressure variability; CNAP = continuous noninvasive arterial pressure; CO = cardiac output; HR = heart rate; HRV = heart rate variability; MBP = mean arterial blood pressure; OSA = obstructive sleep apnea; QRS = QRS duration; +RR-I var = RR interval variability; Symp = sympathetic; TTPK = time to peak.