Abstract 17458: Use of Galectin-3 to Create a Simplified Heart Failure Rehospitalization Risk Model
Introduction: Heart failure (HF) patients are at high risk for short-term rehospitalization and death but there are no simple risk models for the prediction of these outcomes. We hypothesized that a simple risk score, composed of clinical variables and galectin-3, a marker of myocardial fibrosis, predicts HF rehospitalization or death at 60-days post-discharge.
Methods: We analyzed patient-level data from 3 studies of hospitalized HF patients: COACH, PRIDE, and UMD; total N=798. We used Cox models to determine the independent association of galectin-3 with 60-day outcomes. We used ROC curve, net reclassification improvement (NRI) and integrated discrimination index (IDI) analyses to determine the utility of adding galectin-3 to conventional risk factors. We also developed a risk equation.
Results: At baseline, mean age was 70±12 yrs, EF was 37±16% (27% had preserved EF>50%), and 38% were female. At 60-days post-discharge, there were 62 HF rehospitalizations and 34 deaths. NYHA class, EF, diabetes, GFR, and galectin-3 were each associated with outcomes on univariate Cox regression (P<0.05 for each predictor). On multivariable analysis, galectin-3 remained independently associated with HF hospitalization or death (HR 1.4 [95% CI 1.1-1.8], P=0.003). C-statistics for the model with galectin-3 alone vs. the full model with NYHA class, EF, diabetes, and GFR were 0.62 and 0.68, respectively. The addition of galectin-3 to conventional risk factors resulted in an NRI of 7% (P=0.041) and absolute IDI of 0.014 (P=0.008), indicating improved model performance. A risk equation (0.034*galectin-3 + 0.19*NYHA - 0.2*EF - 0.008*GFR + 1 if diabetic) score ≥0 was associated with a ∼3x increased risk of 60-day events (16.4% vs. 5.5%, P<0.001).
Conclusions: A simple HF risk model using galectin-3 predicts risk of HF rehospitalization or death. Use of the simple galectin-3 based risk score to identify high risk patients may be helpful in improving outcomes and reducing healthcare costs.
- © 2012 by American Heart Association, Inc.