Prognosis in myocardial infarction survivors with left ventricular dysfunction is predicted by electrocardiographic RR interval but not QT dispersion

Objectives. The aim of the study was to assess if QT dispersion and RR inte rval on the standard 12-lead electrocardiogram (ECG) predict cardiac death and late arrhythmic events in postinfarction patients with low left ventric ular ejection fraction (LVEF). QT dispersion on a standard electrocardiogra m (ECG) is a measure of repolarization inhomogeneity, but its prognostic me aning in myocardial infarction (MI) survivors is unclear, especially in pat ients with left ventricular dysfunction. RR interval has been shown to pred ict mortality in post-MI patients, but its prognostic power has not been co mpared with other noninvasive risk factors. Methods. Retrospective cohort study. Ninety patients were identified, from a series of 547 consecutive postinfarction patients admitted to our institu tion for phase II cardiac rehabilitation, as having a LVEF of <0.40 at two- dimensional echocardiography (mean LVEF 0.35+/-0.04; range 0.20-0.39). QT d ispersion and RR interval were analyzed on the admission 12-lead electrocar diogram, 20+/-10 (range 8-45) days after MI, using specially designed softw are. Additional risk markers were collected from clinical variables, signal -averaged ECG and Holter recording. Results. During 24+/-18 (range 1-63) months of follow-up, 10 of 90 patients (11%) died, all from cardiac causes, and there were 18 late arrhythmic eve nts, defined as sudden death or the occurrence of a sustained ventricular a rrhythmia greater than or equal to 5 days after the index MI. QT interval a nd dispersion were not significantly prolonged in patients who died compare d to survivors and not significantly different between patients with and wi thout arrhythmic events. Mean RR interval from standard ECG was significant ly shorter in patients with both cardiac death (682+/-99 vs. 811+/-134 ms; P=0.004) and arrhythmic events (720+/-100 vs. 818+/-139 ms; P=0.006). A Cox proportional hazards model identified RR interval from standard ECG (P<0.0 01) and a history of more than one MI (P=0.002) as significant predictors o f cardiac death independent of thrombolytic therapy, LVEF, filtered QRS com plex duration at signal-averaged EGG, mean RR and its standard deviation at 24-h Holter monitoring. Conclusions. Measurement of QT interval and dispersion 3 weeks after MI has no prognostic power in patients with LV dysfunction after a recent MI. RR interval on standard 12-lead ECG is as good a prognostic indicator as other , more expensive, noninvasive markers. These findings may be relevant in th is era of limited health care resources. (C) 1999 Elsevier Science Ireland Ltd. All rights reserved.