The QT interval of the resting 12-lead electrocardiogram is normal or borderline in some patients with congenital long QT syndrome (LQTS). Recently, several in vivo experimental studies have shown that the time of maximum dV/dt in the ST-T segment is correlated with the time of local ventricular recovery. The purpose of this study was to examine the value of the body surface recovery time measured by 87-lead body surface mapping for detecting LQTS. Body surface mapping and 12-lead electrocardiography were performed simultaneously in 18 patients with LQTS and 40 controls of similar age and sex. The recovery time (RT), that is, the interval between QRS onset and the time of maximum dV/dt in the ST-T segment, was measured automatically by computer from each of the 87 mapping leads, and the corrected RT (RTc) was calculated by Bazett's method. The QT interval was measured from each of the 12 standard electrocardiographic leads, and the corrected QT (QTc) interval was also calculated. The maximum RT and RTc, the minimum RT and RTc, and the RT and RTc dispersions (difference between maximum and minimum RT and RTc in each patient) were significantly longer in the LQTS group than in the control group. In addition, a maximum RT of 390 msec, a maximum RTc of 430 msec 1/2, an RT dispersion of 170 msec, and an RTc dispersion of 170 msec1/2 separated the 2 groups completely (i.e., no overlap). The maximum QT and QTc, the minimum QT and QTc, and the QT and QTc dispersions (difference between maximum and minimum QT and QTc in each patient) were also significantly longer in the LQTS group than in the control group. However, the maximum QTc was normal (< or = 440 msec1/2) or borderline (< or = 460 msec1/2) in 5 of the 18 LQTS patients, and none of these parameters clearly separated the 2 groups. These results suggest that measurement of RT by 87-lead body surface mapping is useful for diagnosing latent or borderline LQTS.