Heart failure hospitalization in adults with congenital heart diseases: What predicts it and how does it affect mortality?

Abstract

Adults with congenital heart disease (ACHD) are not cured and residual abnormalities predispose them to heart failure (HF). To calculate the cumulative incidence of HF and assess the impact of HF on mortality. To identify predictors of one-year risk of HF. This population-based retrospective cohort of 27,975 ACHD aged 18–65 in 1995–2010 was based on the Quebec CHD database. We calculated the cumulative probability of HF hospitalization using the practical incidence estimator macro to adjust for the competing risk of death. To assess the impact of HF on mortality, we first used propensity score matching to select random controls for each HF hospitalized patient. We then compared the mortality rates between cases and their matched controls. Finally, we applied nested-case control study and conducted logistic regression analyses to identify the predictors of one-year risk of HF hospitalization. We further used the regression model to construct a risk scoring system (RAAID-HF) for HF hospitalization to identify patients at high-risk of 1-year HF hospitalization. The lifetime cumulative risk of HF hospitalization by age 65 was 33.2%. HF hospitalization was associated with a 5-fold increase in mortality risk (HR = 5.4, 95% CI: 3.5, 8.3). Age, sex, CHD severity, HF hospitalization history and comorbidities (arrhythmia, pulmonary hypertension, coronary heart disease, diabetes, hypertension) were significant predictors of one-year HF hospitalization. The RAAID-HF had excellent predictive performance for HF hospitalization (C-statistics = 0.92). HF is a common comorbid condition in ACHD patients and is strongly associated with an risk of death in ACHD population. We developed a convenient clinical risk score for predicting the risk of HF hospitalization within a year. These data enable targeting patients at high risk for 1-year HF hospitalization for accelerated referral to CHD centers.