PACE-PACING AND CLINICAL ELECTROPHYSIOLOGY, vol.26, no.2, pp.544-550, 2003 (SCI-Expanded)
This study is designed to assess the value of heart rate turbulence (HRT) in the acute phase of MI for prediction of long-term mortality risk. The study included 128 consecutive acute MI patients with 24-hour Halter recordings to evaluate HRT (turbulence onset and slope), SDNN, mean RR interval, and ventricular premature beat frequency. LVEF was evaluated by two-dimensional echocardiography. Data from 117 patients (mean age 58 +/- 11 years) were available for further analysis. Twelve patients died during follow-up (mean 312 +/- 78 days). Although SDNN < 70 ms was the most powerful predictor of mortality among all presumed risk factors (hazard ratio 20 [95% CI 2.6-158]; P = 0.004) in univariate Cox regression analysis, in multivariate analysis LVEF less than or equal to 0.40 and turbulence slope less than or equal to 2.5 ms/RR interval were the only independent predictors of mortality (hazard ratio 6.9 [95% CI 1.8-26]; P = 0.006, hazard ratio 7.3 [95% CI 1.4-37]; P = 0.016, respectively). Addition of HRT parameters for LVEF increased remarkably the positive predictive value (60%) without any decrease in the negative predictive value (92%). Blunted HRT reaction within the first 24 hours of acute MI is an independent predictor of long-term mortality. Furthermore, its predictive power is comparable and also additive to that of LVEF.