Abstract 11414: Ventricular Size and Function Measured by Cardiac MRI Improve Prediction of Major Adverse Clinical Outcomes Independent of Prolonged QRS Duration in Patients with Repaired Tetralogy of Fallot
Background: Major adverse cardiac outcomes in patients with repaired tetralogy of Fallot (rTOF) occur at increasing frequency once patients reach adulthood. Identifying predictors for these outcomes remains difficult as event rates are low and most reports lack statistical power. One large multicenter cohort identified QRS duration ≥180 ms as a predictor of death but did not investigate quantitative measures of ventricular size and function. The goal of this study is to determine whether cardiac MR (CMR)-measured ventricular size and function predict adverse outcomes independent of QRS duration in rTOF.
Methods: Data from 4 large congenital cardiac centers in the US, Canada, and Europe participating in the INDICATOR cohort were analyzed. Inclusion criteria were 1. rTOF; 2. CMR examination; 3. ECG performed ≤1 year from CMR; and 4. clinical follow-up data available. Exclusion criteria included pulmonary valve replacement before CMR. The primary outcomes were death or sustained VT. Multivariable analyses adjusting for length of f/u were performed using logistic regression models. The C statistic, which examines the power of a risk factor to predict an outcome, and the Nagelkerke R2 value, which describes the proportion of variability in a model explained by the variables were calculated.
Results: Of the 1265 subjects enrolled in the registry, 871 were eligible for analysis. Median age at TOF repair was 2.8 years, at 1st CMR 22.8 years, and at last follow-up 26.9 years. There were 31 outcomes: 26 deaths and 5 sustained VT. Prolonged QRS duration alone was a modest predictor of outcome (OR for 10 ms increase 1.28; 95% CI 1.11, 1.48; C= 0.676; R2= 0.054). The addition of CMR-measured right ventricular (RV) mass-to-volume ratio (OR for 0.1 increase 1.95; 95% CI 1.57, 2.41) and ejection fraction (OR for 5% decrease 1.27; 95% CI 1.02, 1.59), to a model with prolonged QRS duration substantially improved outcome prediction (C= 0.833; R2= 0.23).
Conclusions: The addition of CMR-measured RV ejection fraction and mass-to-volume ratio substantially improves prediction of major adverse outcomes in patients with rTOF. Taking advantage of this large cohort, future analyses from the INDICATOR registry will further refine outcome prediction in specific subpopulations of rTOF.
- © 2011 by American Heart Association, Inc.