Stroke Outcome Prediction by Blood Pressure Variability, Heart Rate Variability, and Baroreflex Sensitivity

Abstract

Background and Purpose— Poststroke autonomic dysfunction portended an unfavorable prognosis. We investigated whether blood pressure variability (BPV), heart rate variability, and baroreflex sensitivity might predict stroke functional outcome. Methods— We calculated BPV, heart rate variability, baroreflex slope, and baroreflex effectiveness index from a 5-minute beat-to-beat blood pressure and heart rate monitoring within 7 days from the stroke onset. We compared the parameters between patients with a good outcome (modified Rankin Scale score, 0–2) and those with a poor outcome. Results— Among 142 patients (mean age, 63.9±10.2 years; 88.0% men), functional outcome was good in 112 (78.9%) and poor in 30 (21.1%). There were significant differences in admission National Institutes of Health Stroke Scale, prior stroke, high-frequency systolic BPV, low/high-frequency ratio of BPV, baroreflex sensitivity-up, and baroreflex sensitivity-total between the 2 groups (all P <0.05). In multivariate analysis, National Institutes of Health Stroke Scale (OR, 1.672 [95% CI, 1.316–2.125]; P <0.001), low/high-frequency ratio of systolic BPV (OR, 0.493 [95% CI, 0.250–0.973]; P =0.041), and baroreflex effectiveness index-down (OR, 0.958 [95% CI, 0.924–0.992]; P =0.017) independently predicted a poor functional outcome. Conclusions— A decreased low/high-frequency ratio of systolic BPV and impaired baroreflex sensitivity predicted an unfavorable stroke outcome, in addition to the established prognostic factor such as the National Institutes of Health Stroke Scale.