Method for Predicting of Acute Hypotensive Episodes

Abstract

Hypotension is type of secondary insult and it is related to poor outcome. The ability to predict adverse hypotensive events, where a patient's arterial blood pressure drops to abnormally low levels, would be of major benefit to the fields of primary and secondary health care. The aim of the paper is to present the novel method for predicting of acute hypotensive episodes, based on ECG analysis by the complex system theory approach. 45 patients (in four neurointensive care facilities throughout Europe) data were selected for the analysis. 11 patients had EUSIG-defined hypotensive events. The method includes determining of time varying biomarkers corresponding to plurality of physiological processes in patient's organism as a non-linear dynamic complex system and generating an acute hypotension prediction classifier. The calculations of biomarkers are based on complex system approach and algebraic matrix analysis of ECG parameters. The classifier is based on the comparison of biomarkers behaviour in 3D images. It is demonstrated that the presented method allows us to predict arterial hypotension events 40-50 minutes ahead with a sensitivity of 81 %, specificity 94 %. This result was obtained from prospective real-time data collection in a live clinical intensive care environment.