Effect of Heart Rate Variability on the Association Between the Apnea-Hypopnea Index and Cerebral Small Vessel Disease

Background and Purpose- The apnea-hypopnea index (AHI) is associated with cerebral small vessel disease (cSVD), but pathogenesis of this association is elusive. We aimed to assess the effect of nighttime heart rate variability (HRV)-as a proxy of sympathetic upregulation-on the aforementioned association. Methods- Atahualpa residents aged ≥60 years undergoing brain magnetic resonance imaging, polysomnography, and 24-hour Holter monitoring (N=176) were included. The presence of moderate-to-severe white matter hyperintensities, deep cerebral microbleeds, lacunar infarcts, and >10 enlarged basal ganglia perivascular spaces were added for estimating the cSVD score. Interaction models were fitted to assess the effect modification of nighttime HRV in the association between the AHI and the cSVD score, and mediation analysis was utilized to assess the proportion of total effect by nighttime HRV on this association. Results- Generalized linear models showed a significant association between the AHI and the cSVD score (P=0.025), as well as a significant inverse association between nighttime HRV and the cSVD score (P=0.002), but no association between daytime HRV and the cSVD score (P=0.097). Interaction models showed a significant interaction of nighttime HRV on the association between AHI and the cSVD score (P=0.001), and mediation analysis found that the percent of total effect between AHI and cSVD score mediated by HRV was 30.8%. Predictive marginal means of the cSVD score were highly significant when the 10th percentile of nighttime HRV was compared across categories of 10th and 90th percentiles of the AHI (cSVD score margins, 0.61 [95% CI, 0.37-0.86] versus 1.67 [95% CI, 1.26-2.09]). Contour plots showed the effect of nighttime (but not daytime) HRV on the association between AHI and the cSVD score. Conclusions- This study shows an important effect of nighttime HRV on the association between the AHI and the cSVD score and provides further support for the role of sympathetic overactivity on this association.