322-OR: ECG-Derived QRS-T Angle Is an Independent Predictor of Sudden Cardiac Death in Patients with Type 2 Diabetes: From the SURDIAGENE Prospective Study

Diabetes is associated with a high risk of sudden cardiac death (SCD) but the risk of non-SCD is even higher. We aimed to assess cardiac electrical markers for predicting SCD in patients with T2D.

SURDIAGENE is a French prospective cohort of T2D patients. Baseline 12-lead ECG of 1285 (64±years; 43% female) patients were concordantly evaluated by 2 cardiologists to measure 15 ECG-derived markers. Causes of death were adjudicated by an independent committee. The primary outcome was SCD which was defined as a sudden, natural, and unexpected death, with preceding duration of symptoms <24 hours for witnessed cases and <1 hour witnessed cases. Bonferroni-adjusted cause-specific Cox proportional hazards and Fine and Gray competing-risk models were performed to identify independent predictors (adjustment for age, sex, HbA1c, NT-proBNP, eGFR, ACR, history of myocardial infarction, diuretic, and insulin) . At baseline, mean diabetes duration was 14±years and mean HbA1c 7.8±1.6%. During a median follow up of 7.4 years, 446 deaths occurred including 83 (18.6%) SCD. In univariate analysis, R wave voltage ≥mm in aVL lead, corrected QT interval and QRS-T angle ≥90° were associated with SCD risk at the Bonferroni-adjusted level of significance (P < 0.0033) . Using competing-risk multivariate analysis, QRS-T angle ≥90° only was significantly associated with a higher risk of SCD (sHR 1.63 [1.01-2.62]; P=0.045) but interestingly not with non-SCD (sHR 1.02 [0.78-1.33], P=0.87) . Discrimination of SCD prediction model was improved by QRS-T angle beyond multivariate Cox model parameters (C-statistic improved from 0.798 to 0.805, P<0.0001) .

ECG-derived QRS-T angle is an independent predictor of SCD in patients with T2D and improves SCD-risk model prediction accuracy.