Development of a Real-Time Risk Tracking Model for Neurological Deterioration in Ischemic Stroke Patients Based on Blood Pressure Dynamics

Abstract

AbstractObjectivesUsing the significant link between blood pressure fluctuations and neurological deterioration (ND) in patients with ischemic stroke, this study aims to develop a predictive model capable of real-time tracking of ND risk, enabling timely detection of high-risk periods.MethodsA total of 3,906 consecutive ischemic stroke patients were recruited. As an initial predictive model, a polytomous logistic regression model, incorporating clinical parameters to estimate a probability of ND occurring within and beyond 12 hours post-stroke onset, was developed. To refine ND risk assessments over time, we subsequently introduced an iterative risk-tracking model that utilizes continuously updated blood pressure measurements. We endeavored to integrate these models, assessing their combined discriminative capacity and clinical utility, with a particular emphasis on pinpointing time periods of increased ND risk.ResultsND rates were observed at 6.1% within the first 12 hours and 7.3% during the following 60 hours. We noted variations in incidence rates and their distribution over time across predefined patient groups. Significant predictors of ND varied among these subgroups and across different time periods. The iterative risk-tracking model maintained a consistent relationship between blood pressure variables and ND risk across different patient groups, successfully forecasting ND within a 12-hour window. The integrated models achieved an area under the receiver operating characteristic curve (AUC) ranging from 0.68 to 0.76. This performance effectively narrowed down the critical window for ND risk identification without sacrificing predictive accuracy across diverse patient groups. At 90% and 70% sensitivity settings, the combined model precisely identified the periods of highest ND risk, showing slightly higher or comparable specificity and positive predictive values relative to other models.ConclusionThis study presents a novel approach for real-time monitoring of ND risk in ischemic stroke patients, utilizing BP trends to identify critical periods for potential intervention.