QT dispersion as a predictor of long-term mortality in patients with acute myocardial infarction and clinical evidence of heart failure

BACKGROUND: QT interval dispersion is a marker of inhomogeneous ventricular repolarization, and therefore has the potential to predict re-entry arrhythmias. Following acute myocardial infarction, increased QT dispersion has been associated with a higher risk of ventricular arrhythmias. However, whether or not QT dispersion predicts prognosis post-acute myocardial infarction is not clear. We addressed this issue by analysing the AIREX study registry. METHODS: AIREX was a follow-up study of 603 post-acute myocardial infarction patients who exhibited clinical signs of heart failure and were randomly allocated to ramipril or placebo. An interpretable 12-lead ECG obtained between day 0 and day 9 after the index infarction (median time 2 days) was available in 501 patients. We examined whether QT dispersion was a predictor of all-cause mortality in the AIREX study registry (mean follow-up 6 years). RESULTS: QT dispersion measurements were significantly increased in patients who subsequently died (QT dispersion: 92.0 +/- 38.5 ms vs 82.7 +/- 34.3 ins. P=0.005; rate corrected QT dispersion: 105.7 +/- 42.7 ms vs 93.1 +/- 35.9 ms, P<0.001). Univariate analysis showed that QT dispersion as a predictor of all-cause mortality risk (QT dispersion: hazard ratio per l0 ms 1.05, 95% CI 1.02 to 1.09. P= 0.004; rate corrected QT dispersion: 1-07 1.03 to 1.10, P<0.001): an increase of 10 ms added a 5-7%, relative risk of death. QT dispersion remained an independent predictor of all-cause mortality risk on multivariate analysis (QT dispersion: 1.05 1.01 to 1.09, P=0.027; rate corrected QT dispersion: 1.05 1.01 to 1.09. P=0.022). CONCLUSION: QT dispersion. measured from Li routine 12-lead ECG following acute myocardial infarction complicated by heart failure provides independent information regarding the probability of long-term survival. However. the low sensitivity of this electrocardiographic marker limits its usefulness for risk stratification if used in isolation.