QT variability index on 24-hour Holter independently predicts mortality in patients with heart failure: analysis of Gruppo Italiano per lo Studio della Sopravvivenza nell'Insufficienza Cardiaca (GISSI-HF) trial

BackgroundIncreased temporal variability of repolarization, as reflected by QT interval variability measured over 10–15 minutes, predicted spontaneous ventricular arrhythmias and death in implantable cardioverter-defibrillator patients in mild to moderate heart failure (HF).ObjectiveThe purpose of this study was to test our hypothesis that increased mean QT variability over 24 hours would be associated with increased cardiovascular (CV) mortality in a heterogeneous HF population.MethodsThe Gruppo Italiano per lo Studio della Sopravvivenza nell'Insufficienza Cardiaca-Heart Failure trial prospectively enrolled subjects with HF of any cause. Twenty-four-hour Holter recordings from 268 subjects were analyzed using a template-matching, semiautomatic algorithm to measure QT and heart rate time series in sequential 5-minute epochs over 24 hours. The QT variability index (QTVI) was expressed as the log ratio of the normalized QT variance over normalized heart rate variance. Total and CV mortality were assessed as a function of continuous and dichotomous QTVI (>−0.84) in univariate and multivariable Cox proportional hazards models, adjusting for significant clinical predictors.ResultsAfter a median of 47 months, there were 53 deaths, of which 44 were from CV causes. A significant association with the outcome was found for QTVI both as continuous and dichotomous variables after adjustment for clinical covariates (age >70, New York Heart Association class III–IV, left ventricular ejection fraction, nonsustained ventricular tachycardia, creatinine): QTVI hazard ratio (HR) 4.0 (confidence interval [CI] 1.8–88; P = .008) for total and 4.4 (CI 1.9–10.1; P = .0006) for CV mortality; QTVI >−0.84 HR 2.0 (CI 1.1–3.6; P = .02) for total and 2.1 (CI 1.1–3.8; P = .02) for CV mortality.ConclusionIncreased repolarization lability, as reflected in QTVI measured over 24 hours, is associated with increased risk for total and CV mortality in a heterogeneous population with chronic HF.