To assess the effects of measurement methodology on QT determinations and to define the spectrum of QT values, including interlead variability, on the body surface, we measured QT in each of 120 simultaneously-recorded, signal-averaged ECG leads in 10 normal subjects and 14 patients with QT prolongation (lead II QTc greater than 440). Two separate, but related, methods of QT measurement were utilized. Method A was a relatively conventional technique in which ST-T offset was defined as the time instant of return of the T wave to a P-P baseline, or as the point of U-on-T intersection. Method B was a more rigorous method, which defined ST-T offset in a similar manner, and in addition discarded from analysis all QT values from leads with monophasic ST-T waveform in which the QT values were greater than the longest QT from leads with definite U waves. Method B was utilized to minimize factitious prolongation of QT by inapparent U-on-T. By both methods the mean body surface QTc values were significantly greater (p less than 0.001) in the patient group (482 +/- 65 [S.D.] msec, method A; 447 +/- 43 msec, method B), than in the normal subject group (399 +/- 14 msec, method A; 396 +/- 12, method B). Interlead QTc variability (difference between the longest and shortest QT) was considerable with both methods and in both study groups. Expressed as percent of average body surface values, the mean interlead QTc variability in normal subjects averaged 22 percent with method A and 19 percent with method B; in the patient group, however, it averaged 32 percent with method A and only 18 percent with method B. In absolute terms, the mean variability in the patient group with method A (155 +/- 62 msec) was significantly greater (p less than 0.001) than that of the normal group (89 +/- 33 msec); with method B, interlead variability was the same (p = NS) in the normal (76 +/0 27 msec) and patient groups (80 +/- 44 msec). This latter finding suggests the possibility that the repolarization abnormality in patients with QT prolongation may occur relatively uniformly throughout the ventricular myocardium. Thus, measurement techniques are important in multiple-lead QT determinations. Although reduced by techniques designed to minimize factitious QT prolongation, interlead QT variation is considerable over the torso surface, in both normal subjects and patients with repolarization abnormalities.(ABSTRACT TRUNCATED AT 400 WORDS)