Implications of methodological differences in digital electrocardiogram interval measurement

Abstract

Well-specified recommendations have yet to be established on how electrocardiogram (ECG) interval measurement should be performed by digital on-screen caliper systems to assess drug-induced effect on cardiac repolarization in pharmaceutical clinical trials with adequate precision and reproducibility. Since 1997, the industry has followed the European Committee for Proprietary Medicinal Products Points to Consider by using fully manual measurement of 3 consecutive sinus rhythm PQRST complexes in 1 lead only (typically limb lead II). More recently, semiautomatic measurement performed on representative (median) beats and based on the global leads has been considered. The International Conference on Harmonization E14 guidance (June 2005) advocates development of quality standards for centralized ECG interval measurement and allows all methods "whether or not assisted by computer" but includes no recommendations on how to perform the measurement. We provide an overview of the currently available methods for digital ECG interval measurement and the implications of between-method differences on quality of ECG interval measurements. We applied 4 methods most commonly used to assess QT prolongation (applied on 3 raw beats in limb lead II or by global measurement on 1 or 12 superimposed representative beats). QT, QTc Fridericia, and RR interval durations were measured on resting 12-lead digital ECGs obtained in 26 healthy volunteers predose and at 1, 2, and 3 hours after dosing with a single 160 mg oral dose of sotalol. Absolute interval durations and changes from baseline were compared between the 4 measurement methods. A better understanding of the implications from different measurement methodologies will facilitate more informed choice of the appropriate method for ECG interval measurement on clinical trials.