Abstract W P239: Multiscale Entropy Analysis of Heart Rate Variability Predicts Functional Outcome in ICU-Admitted Acute Stroke Patients
Objectives: Heart rate variability (HRV) has been proposed as a predictor of acute stroke outcome. The present study aimed to apply a novel non-linear method, multiscale entropy (MSE) analysis to investigate the association between the complexity of HRV and outcome in intensive care unit (ICU) admitted acute stroke patients.
Methods: Continuous EKG signals were recorded for one hour in non-atrial fibrillation (AF) stroke patients within 48 hours after admission and controls. The complexity index (CI) was defined as the area under the MSE curve from scale 1 to scale 20. The values of CI related to outcome of functional independence (modified Rankin Scale 0-2) and death at 3 months were analyzed.
Results: From February, 2012 to February, 2013, a total of 109 non-AF acute stroke patients (mean age 61.7±15.0 years, female 45.9%) and 60 age-sex matched controls were recruited. The CI of the MSE was significantly lower in stroke patients than the controls (32.3 ±4.25 versus 25.2 ±6.84 p<0.001). After adjustment for age, gender, and known clinical predictors including NIH stroke scales and glucose levels at admission, the values of CI had significant associations with outcomes of both functional independence and death at 3 months (adjusted odds ratio=1.186, p=0.004 and 0.784, p=0.039, respectively).
Conclusions: In ICU admitted non-AF acute stroke patients, early assessing the complexity of HRV via MSE can help in predicting long-term functional outcome.
Author Disclosures: S. Tang: None. H. Jen: None. Y. Lin: None. C. Hung: None. D. Lai: None. A. Wu: None. J. Jeng: None. M. Chen: None.
- © 2014 by American Heart Association, Inc.