Electrocardiogram-based predictors of clinical outcomes: A meta-analysis of the prognostic value of ventricular repolarization

Objectives: To estimate age- and sex-specific prognostic values of eight electrocardiographic repolarization descriptors to predict various mortality endpoints. Background: Using electrocardiographic markers for risk stratification is well studied; however, the prognostic value of many markers is controversial, and their clinical utility remains debatable. No meta-analyses exist that address the prognostic value of ECG markers. Methods: Data were synthesized from 106 primary studies using a random-effect variance model. Age and sex subgroups were analyzed using sensitivity analysis. Results: Four classic (i.e., duration, amplitude, inversion, and ST-T changes) and four novel (i.e., axis, loop, wavefront direction, and waveform complexity) repolarization descriptors were studied. These novel descriptors were particularly useful in predicting sudden death. Abnormal repolarization duration, vectors, and loops have greater impact on negative cardiovascular outcomes in women compared to men; additionally, ischemic repolarization changes have greater impact on negative cardiovascular outcomes in younger versus older adults. Conclusions: Assessing repolarization abnormalities is particularly helpful in women and younger adults. Researchers need to further explore the clinical utility of these abnormalities in management algorithms. (C) 2014 Elsevier Inc. All rights reserved.