External validation of a risk score model for predicting major clinical events in adults after atrial switch

Abstract

A risk model has been proposed to provide a patient individualized estimation of risk for major clinical events (heart failure events, ventricular arrhythmia events, all-cause mortality) in patients with transposition of the great arteries corrected by an atrial switch operation (D-TGA). The aim of this study was to externally validate the model. A retrospective, multicentric, longitudinal cohort of 417 patients with D-TGA (median age 24 years [interquartile range 18–30], 63% male) independent of the model development and internal validation cohort was studied. Data on risk model predictors (age > 30 years, prior ventricular arrhythmia, age > 1 year at repair, moderate or greater right ventricular dysfunction, severe tricuspid regurgitation, and mild or greater left ventricular dysfunction) were collected from the time of baseline clinical evaluation. The performance of the prediction model in predicting risk at 5 years was assessed. Twenty-eight patients (6.7%) met the major clinical events endpoint within 5 years, with an overall incidence rate of 1.42 per 100 patient-years [95% confidence interval (CI) 0.94–2.05]. Model validation showed a good discrimination between high and low 5-year risk patients (Harrell's C-index of 0.710 (95% CI 0.66 to 0.75)) but tended to overestimate this risk (calibration slope of 0.42 (95% 0.061–0.78); Fig. 1 ). We reported the first external validation of a clinical events risk model in a large D-TGA patient population. Although a good discrimination, the model tends to overestimate the 5-year risk. The development of new risk models is needed to individualize risk predictions in D-TGA patients.