Ostensible day-night difference of QT prolongation during long-term treatment with antiarrhythmic drugs: reappraisal of the law of "regression to the mean".

This study was designed to test whether the law of regression to the mean explains the diurnal variation in the modulation of electrocardiographic variables during the treatment with antiarrhythmic agents. In part 1, in 34 subjects, ambulatory ECG monitorings were repeated twice, and the corrected QT interval (QTc) at a heart rate of 60 beats/min was calculated separately for the daytime and night. The individual diurnal QTc variation (day-night difference) of the first recording (4.4 +/- 3.3%) was significantly correlated with that of the second recording (5.0 +/- 3.1%; r = 0.61; p < 0.0001), and naturally, the second measurement tended to be lower than the first value in those with relatively greater baseline diurnal QTc variation and vice versa (p < 0.005). In part 2, 30 subjects undertook ambulatory ECG recordings before and during treatment with class Ia antiarrhythmic drugs. Mean QTc changes in the daytime and in the night with the drugs were comparable (18 +/- 17 ms and 19 +/- 15 ms). However, the day-night difference of postdrug QTc changes in each subject was inversely correlated with baseline diurnal QTc variation (r = -0.64; p < 0.0001). These observations in part 2 were comparable with those in part 1, and individual day-night difference in QT prolongation with antiarrhythmic drugs seemed to be a chance occurrence. It was suggested that the law of regression to the mean is appreciably reflected in the ostensible intraday variation of pharmacologic modulation of electrocardiographic variables.