Department of Biostatistics Seminar/Workshop Series
Multivariate Reference Regions for ECG Parameters
Kimberly Crimin, PhD
Senior Principal Biostatistician II
Wyeth Research and Development
Tuesday, May 12, 10:30-11:30am, VCH Theater 2210
Intended Audience: Persons interested in applied statistics, statistical theory, epidemiology, health services research, clinical trials methodology, statistical computing, statistical graphics, R users or potential users
Increasing the time for cardiac repolarization (QT prolongation) following drug administration has been responsible for the withdrawal and re-labeling of numerous marketed compounds and the outright rejection of potentially important compounds still in discovery. QT prolongation, associated with an increased occurrence of Torsades de Pointes (a fatal ventricular tachyarrhythmia), has an inverse relationship with heart rate (1/RR). Traditional methods to correct the QT interval (QTc) for heart rate are an attempt to reduce a bivariate observation (QT, RR) to a single value. These traditional methods often disagree with each other and do not take into account changes in the autonomic state of an individual. This can completely confound distinguishing between increased QT intervals caused by drug-induced delayed repolarization and autonomic-mediated response. The development of automatic ECG signal acquisition systems have made it possible to collect continuous ECG data. Using this beat-to-beat ECG data, it is potentially possible to extract information on the autonomic tone and compare this to a baseline ECG region. In this presentation, the traditional methods for assessing arrhythmia liability are reviewed and several of the issues discussed. I then present the development of reference regions for ECG data and discuss methods for assessing the agreement between two regions. Data from two clinical studies will be used to illustrate the methods.