Microvolt T-wave alternans as a predictor of mortality and severe arrhythmias in patients with left-ventricular dysfunction: a systematic review and meta-analysis

BACKGROUND: Studies have demonstrated that the use of implantable cardioverter defibrillators (ICDs) is effective for the primary prevention of arrhythmic events but due to imposing costs, there remains a need to identify which patients will derive the greatest benefit. Microvolt T-wave alternans (MTWA) has been proposed to assist in this stratification. METHODS: We systematically searched the literature using MEDLINE, EMBASE, Current Contents, the Cochrane Library, INAHTA, and the Web of Science to identify all primary prevention randomized controlled trials and prospective cohort studies with at least 12 months of follow-up examining MTWA as a predictor of mortality and severe arrhythmic events in patients with severe left-ventricular dysfunction. The search was limited to full-text English publications between January 1990 and May 2007. The primary outcome was a composite of mortality and severe arrhythmias. Data were synthesized using Bayesian hierarchical models. RESULTS: We identified no trials and 8 published cohort studies involving a total of 1,946 patients, including 332 positive, 656 negative, 84 indeterminate, and 874 non-negative (which includes both positive and indeterminate tests) MTWA test results. The risk of mortality or severe arrhythmic events was higher in patients with a positive MTWA compared to a negative test (RR = 2.7, 95% credible interval (CrI) = 1.4, 6.1). Similar results were obtained when comparing non-negative MTWA to a negative test. CONCLUSION: A positive MTWA test predicts mortality or severe arrhythmic events in a population of individuals with severe left ventricular dysfunction. However, the wide credible interval suggests the clinical utility of this test remains incompletely defined, ranging from very modest to substantial. Additional high quality studies are required to better refine the role of MTWA in the decision making process for ICD implantation.