Abstract
AIMS: Efficient use of preventive cardiac therapies is often limited by inefficient risk prediction, calling for new prediction tools. Ventricular premature complexes (VPCs) elicit electrocardiographic changes in the repolarization of the first post-extrasystolic normal beat. The aim of this study was to assess whether this post-extrasystolic ST segment and T wave variation (PEST) conveys prognostic information regarding the mortality risk of cardiac patients. METHODS: PEST was calculated from 30-min ECGs obtained from 941 survivors of acute myocardial infarction (AMI) as mean difference between the sum of squared voltages from three orthogonal leads (XYZ) of the first (post-extrasystolic) and second (reference) beat after each VPC, in a time window between the limits ϕ(1) and ϕ(2). Optimal limits yielding a maximum area under the receiver-operating characteristics (ROC) curve were determined by systematic testing, covering the time window from the J point to the end of the T wave. A strong association was found with ϕ(1)/ϕ(2) encompassing 40-230 ms after the J point, which was used to calculate PEST in the analysis. Kaplan-Meier curves and univariable/multivariable Cox proportional hazards models were used to study mortality prediction by PEST. The findings were validated in an independent cohort of 1.788 general population subjects aged 60 years or older. RESULTS: The area under the ROC curve for PEST was 0.72, with an optimum cutoff at ≤ -6.69 mV(2). The 88 patients with PEST values below this cutoff had a considerably higher mortality than the remainder of the patients (25% vs. 5.8%, p < 0.0001; univariable hazard ratio 4.7, 95% CI 2.4-12.0, p < 0.001). In a multivariable Cox regression analysis considering left-ventricular ejection fraction, presence of diabetes mellitus, and Global Registry of Acute Coronary Events (GRACE) score, PEST remained significantly associated with mortality (hazard ratio 3.6, 95% CI 1.9-6.9, p < 0.0001). In the validation cohort, abnormal PEST was also associated with significantly increased 4-year mortality (11.9% vs. 4.3%, p = 0.00095). CONCLUSION: PEST is a strong independent predictor of all-cause mortality in AMI survivors and elderly subjects from the general population. While the pathophysiology of this association remains to be investigated, PEST may complement current risk prediction tools in various clinical settings.