New Method for Predicting Neurological Recovery of Heart Attack Patients

A researcher at the Technical Research Center of Finland (VTT, Tampere) conducted a multidisciplinary study that examined EEG readings recorded in 60 patients in hospital operating rooms (ORs) under sevoflurane-induced anesthesia, as well as in 20 patients resuscitated after cardiac arrest in intensive care unit (ICU) environments. The researcher then classified the EEG-derived data to detect epileptiform brain activity. Various quantitative EEG features and conventional biochemical markers were then examined to find out their associations with the patient outcome in a further predictive study of 30 ICU patients admitted for cardiac arrest.

The researcher found that wavelet subband entropy (WSE) was the most important feature for the detection of epileptiform activity. When the outcomes of the 30 ICU patients (survivor or non-survivor) were known, WSE was calculated from the recorded EEG and the average WSE value was obtained for each hour of each recording. The distributions of hourly average values of WSE of the outcome groups were then compared, and a statistically significant difference was found between the distributions, which validated the algorithm that researchers had developed for the classification of EEG waveforms. The study was presented as the doctoral dissertation of Miikka Ermes, M.Sc. in engineering, during May 2009 at the VTT.

"Wavelet subband entropy of EEG is shown to be statistically associated with epileptiform activity both in operating room patients under sevoflurane-induced anesthesia and in intensive care unit patients resuscitated after cardiac arrest," concluded the study author. "The results support the hypothesis that epileptiform activity can be continuously monitored in both clinical settings."

According to the author, the algorithm developed for analyzing human EEG traces could be used to predict neurological recovery, including within the first 24 hours that follow a cardiac arrest, which up till now has been limited due to interpretation difficulties. Thus, correct interpretation almost invariably requires consulting a specialist, which may cause a delay in treatment.

Related Links:
VTT Technical Research Center of Finland