Abstract
The purpose of this study was to provide a meta-analysis to estimate the performance of 12 commonly reported risk stratification tests as predictors of arrhythmic events in patients with nonischemic dilated cardiomyopathy.
Multiple techniques have been assessed as predictors of death due to ventricular tachyarrhythmias/sudden death in patients with nonischemic dilated cardiomyopathy.
Forty-five studies enrolling 6,088 patients evaluating the association between arrhythmic events and predictive tests (baroreflex sensitivity, heart rate turbulence, heart rate variability, left ventricular end-diastolic dimension, left ventricular ejection fraction, electrophysiology study, nonsustained ventricular tachycardia, left bundle branch block, signal-averaged electrocardiogram, fragmented QRS, QRS-T angle, and T-wave alternans) were included. Raw event rates were extracted, and meta-analysis was performed using mixed effects methodology. We also used the trim-and-fill method to estimate the influence of missing studies on the results.
Patients were 52.8 ± 14.5 years of age, and 77% were male. Left ventricular ejection fraction was 30.6 ± 11.4%. Test sensitivities ranged from 28.8% to 91.0%, specificities from 36.2% to 87.1%, and odds ratios from 1.5 to 6.7. Odds ratio was highest for fragmented QRS and TWA (odds ratios: 6.73 and 4.66, 95% confidence intervals: 3.85 to 11.76 and 2.55 to 8.53, respectively) and lowest for QRS duration (odds ratio: 1.51, 95% confidence interval: 1.13 to 2.01). None of the autonomic tests (heart rate variability, heart rate turbulence, baroreflex sensitivity) were significant predictors of arrhythmic outcomes. Accounting for publication bias reduced the odds ratios for the various predictors but did not eliminate the predictive association.
Techniques incorporating functional parameters, depolarization abnormalities, repolarization abnormalities, and arrhythmic markers provide only modest risk stratification for sudden cardiac death in patients with nonischemic dilated cardiomyopathy. It is likely that combinations of tests will be required to optimize risk stratification in this population.
