Tran Thong1, Merritt H Raitt. 1. Department of Biomedical Engineering, Oregon Health and Science University, Beaverton, Oregon 97006-8921, USA. trant@bme.ogi.edu
Abstract
BACKGROUND: A reliable predictor of an imminent episode of ventricular tachyarrhythmia that could be incorporated in an implantable defibrillator capable of preventive therapy would have important clinical utility. METHOD: A test set of 208 R-R records saved by defibrillators spanning a mean of 1.6 hours before sustained tachyarrhythmia were used to derive criteria that would improve the specificity of the previously identified monotonic heart rate acceleration predictor. Additional criteria were used, namely two such patterns need to occur within a period of 1.8 hour and the heart rate during these accelerations exceeds 86 bpm (700 ms). The specificity was tested using R-R records matched in duration from 26 control patients with defibrillators during normal periods. RESULTS: The basic acceleration pattern was found during sinus rhythm in the 1.8-hour period prior to 83% of episodes of ventricular tachyarrhythmia. It was also found in 43% of the matched set of non-arrhythmic records, corresponding to a specificity of 57%. With the two extra requirement of multiplicity within 1.8 hour and peak heart rate, the sensitivity of the proposed predictor is reduced to 53%, but the specificity is increased to 91%, which corresponds to an average false positive rate of 0.8 event/day across the patient population. CONCLUSION: A ventricular tachyarrhythmia predictor based on a pattern of heart rate acceleration has been proposed that can yield sensitivity from 53% to 69%, with specificity up to 91%. Instances of this predictor increase significantly prior to an episode of tachyarrhythmia.
BACKGROUND: A reliable predictor of an imminent episode of ventricular tachyarrhythmia that could be incorporated in an implantable defibrillator capable of preventive therapy would have important clinical utility. METHOD: A test set of 208 R-R records saved by defibrillators spanning a mean of 1.6 hours before sustained tachyarrhythmia were used to derive criteria that would improve the specificity of the previously identified monotonic heart rate acceleration predictor. Additional criteria were used, namely two such patterns need to occur within a period of 1.8 hour and the heart rate during these accelerations exceeds 86 bpm (700 ms). The specificity was tested using R-R records matched in duration from 26 control patients with defibrillators during normal periods. RESULTS: The basic acceleration pattern was found during sinus rhythm in the 1.8-hour period prior to 83% of episodes of ventricular tachyarrhythmia. It was also found in 43% of the matched set of non-arrhythmic records, corresponding to a specificity of 57%. With the two extra requirement of multiplicity within 1.8 hour and peak heart rate, the sensitivity of the proposed predictor is reduced to 53%, but the specificity is increased to 91%, which corresponds to an average false positive rate of 0.8 event/day across the patient population. CONCLUSION: A ventricular tachyarrhythmia predictor based on a pattern of heart rate acceleration has been proposed that can yield sensitivity from 53% to 69%, with specificity up to 91%. Instances of this predictor increase significantly prior to an episode of tachyarrhythmia.