Analysis of T-wave morphology from the 12-lead electrocardiogram for prediction of long-term prognosis in male US veterans.

Abstract

The aim of the present study was to assess the prognostic value of novel repolarization descriptors from the 12-lead ECG in a large cohort of US veterans. Male US veterans (n=813) with cardiovascular disease had digital 12-lead ECGs recorded at the VA Medical Center, Washington, DC, between 1984 and 1991. The patient series was retrospectively compiled in 1991; follow-up was prospectively assessed until 2000. Novel ECG variables characterizing repolarization and the T-wave loop were automatically analyzed. Of 772 patients with technically analyzable data, 252 patients (32.6%) died after a mean follow-up of 10.4 +/- 3.8 years. Direct comparison between dead and alive patients showed that the so-called T-wave residua (the absolute and relative amount of nondipolar contents within the T wave) predicted mortality (111 900 +/- 164 700 versus 85 600 +/- 144 800 between dead and alive patients, P<0.0002; and 0.43 +/- 0.62% versus 0.33 +/- 0.56%, P<0.0005 for the absolute and relative T-wave residuum, respectively). On Cox regression analysis entering age, left ventricular ejection fraction, echocardiographic left ventricular hypertrophy, and either of the T-wave residua, risk prediction was independent for the absolute (P=0.022) and for the relative (P=0.006) T-wave residuum, respectively, with age (P<0.0001), presence of left ventricular hypertrophy (P=0.002), and left ventricular ejection fraction (P=0.004) also being predictors of survival. The heterogeneity of myocardial repolarization, measured by the so-called T-wave residuum in the ECG, confers long-term independent prognostic information in US veterans with cardiovascular disease.