Comparing methods of measurement for detecting drug-induced changes in the QT interval: implications for thoroughly...

The aim of this study was to compare the reproducibility and sensitivity of four commonly used methods for QT interval assessment when applied to ECG data obtained after infusion of ibutilide.

Four methods were compared: (1) 12-lead simultaneous ECG (12-SIM), (2) lead II ECG (LEAD II), both measured on a digitizing board, (3) 3-LEAD ECG using a manual tangential method, and (4) a computer-based, proprietary algorithm, 12SL trade mark ECG Analysis software (AUT). QT intervals were measured in 10 healthy volunteers at multiple time points during 24 hours at baseline and after single intravenous doses of ibutilide 0.25 and 0.5 mg. Changes in QT interval from baseline were calculated and compared across ECG methods, using Bland-Altman plots. Variability was studied using a mixed linear model.

Baseline QT values differed between methods (range 376-395 ms), mainly based on the number of leads incorporated into the measurement, with LEAD II and 3-LEAD providing the shortest intervals. The 3-LEAD generated the largest QT change from baseline, whereas LEAD II and 12-SIM generated essentially identical result within narrow limits of agreement (0.4 ms mean difference, 95% confidence interval +/- 20.5 ms). Variability with AUT (standard deviation 15.8 ms for within-subject values) was clearly larger than with 3-LEAD, LEAD II, and 12-SIM (9.6, 10.0, and 11.3 ms).

This study demonstrated significant differences among four commonly used methods for QT interval measurement after pharmacological prolongation of cardiac repolarization. Observed large differences in variability of measurements will have a substantial impact on the sample size required to detect QT prolongation in the range that is currently advised in regulatory guidance.