The Influence of Sleep Duration on Cerebrovascular Reactivity

Abstract

The impaired ability of the cerebrovasculature to dilate in response to a vasoactive stimulus (i.e., cerebrovascular reactivity, CVR) is associated with an increased risk of cognitive impairment. The association between sleep duration and dementia risk is “U‐shaped”, where adults who achieve seven hours of sleep have the lowest risk of cognitive impairment, compared to adults who achieve greater than or less than seven hours of sleep. In otherwise healthy adults, poor sleep quality and duration impairs peripheral vascular function, but its effect on CVR in generally healthy adults has remained largely been unexplored. The purpose of this study was to assess the relation between subjective sleep quality and CVR in healthy adults. Because the relation between sleep duration and cognitive impairment is an inverted U‐shape, we hypothesized that CVR would be lower in adults who slept longer than or shorter than the recommended 7‐8 hours of sleep. Data from thirty‐five individuals across the adult lifespan (21 males / 14 females, age: 44 ± 17 years; BMI: 26.8 ± 4.8 kg/m2; blood pressure, BP: 114 ± 12 / 71 ± 10 mmHg; heart rate, HR: 62 ± 8 bpm) were included in this analysis. CVR was assessed by measuring the change in middle cerebral artery blood flow velocity (MCAv) at baseline and following a 3‐minute hypercapnic stimulus (+9 mmHg increase in the end‐tidal partial pressure of CO2) using transcranial Doppler ultrasound. MCAv was normalized to beat‐to‐beat BP (MCACVC), and CVR was calculated as the percentage change in MCACVC and normalized to the absolute change in end‐tidal CO2.Sleep duration was negatively predictive of lower CVR (R2=0.17, β1± SE = ‐0.45 ± 0.18 %/mmHg/hour, P=0.014). A correlation matrix was performed to determine if any variables other than sleep duration (e.g., age, sex, etc) might also impact CVR. Age (P=0.53), BMI (P=0.93), systolic BP (P=0.22), diastolic BP (P=0.84), HR (P=0.071), and mean arterial pressure (P=0.57) were not associated with CVR. Only sex was significantly associated with CVR (R= ‐0.36, P=0.036), where men had augmented CVR compared to women (2.20 ± 0.73 %/mmHg vs. 1.59 ± 0.93 %/mmHg, P=0.036). Importantly, sex (β1± SE= ‐0.18 ± 0.27 %/mmHg, P=0.51) did not influence the relation between CVR (β1± SE= ‐0.41 ± 0.16 %/mmHg/hour, P=0.013) and sleep duration (model adjusted for sex: R2= 0.17, P=0.042). In conclusion, longer sleep duration was independently associated with lower CVR in healthy adults. Future studies should include more individuals with shorter sleep duration to test the possibility of an inverted U‐shaped association with CVR.