Predicting arrhythmia-free survival using spectral and modified-moving average analyses of T-wave alternans

Background: T-wave alternans (TWA) is a promising electrocardiogram (ECG) predictor of sudden cardiac arrest, yet needs specialized recordings for conventional spectral analysis. Modified moving average (MMA) analysis is a new approach that can measure TWA from routine ECGs, thus widening its applicability. However, MMA-TWA has not been calibrated against spectral TWA nor outcome in high risk patients. We hypothesized that spectral and MMA-TWA would both predict arrhythmia-free survival on long-term prospective follow-up. Methods and Results: In 41 patients with left ventricular systolic dysfunction (ejection fraction 31 ± 13%), we studied TWA simultaneously using spectral and MMA during pacing (< 110 beats/min). MMA amplified TWA over spectral analyses (13.0 ± 8.28 μV vs 1.96 ± 5.15 μV, P < 0.001). On 542 ± 311 days' follow-up, from clinic visits, telephonic interviews, and device interrogations, there were 11 deaths or sustained ventricular arrhythmias ('events'). Positive spectral TWA (≥1.9 μV) identified patients with from those without events (P = 0.02). Receiver-operating characteristics for MMA-TWA showed that the cutpoint ≥ 10.75 μV was optimal for the combined endpoint. Kaplan-Meier analysis using this MMA-TWA cutpoint trended to predict events (P = 0.06), while MMA combined with spectral TWA identified events (P = 0.01). Conclusions: MMA amplifies TWA compared to traditional spectral analyses, but both likely reflect similar pathophysiology. Validation in larger populations will enable MMA-TWA to be widely applied to stratify risk for sudden cardiac arrest.