Abstract 13340: 10-Year Risk Prediction Equations for Heart Failure in Chronic Kidney Disease: The Chronic Renal Insufficiency Cohort Study

Introduction: Heart failure (HF) is a leading contributor of cardiovascular morbidity and mortality in the chronic kidney disease (CKD) population. While risk equations such as the pooled cohort equations to prevent HF (PCP-HF) have been validated in the general population, no such tools have been evaluated in CKD-specific cohorts. Strategies for identification of individuals with CKD at higher risk of HF are needed. Methods: Among black and white participants with CKD stages 2-4 enrolled in the CRIC Study, we included individuals between the ages of 30-79 without baseline cardiovascular disease. We assessed model performance of the PCP-HF to predict incident HF and recalibrated the PCP-HF in the CKD population (PCP-HF CKD ). Given that indices of kidney function are associated with HF, we examined the contribution of estimated glomerular filtration rate (eGFR) and urinary albumin-to-creatinine ratio (UACR) to PCP-HF CKD by change in C-statistic, net reclassification index (NRI) and integrated discrimination index (IDI). Results: Among 2328 participants, mean (SD) baseline eGFR was 46 (15) ml/min/1.73m2 and median (IQR) UACR was 37 (7, 381) mg/g. Over 9.5 years of follow-up, 336 incident HF events occurred. The PCP-HF equations systematically under-predicted HF risk in participants with CKD (C-statistic <0.71 for all sex-race groups). The recalibrated PCP-HF CKD had good discrimination and calibration (C-statistic 0.75-0.78), except in black women (0.61). Addition of UACR, but not eGFR, to the PCP-HF CKD improved model calibration (Figure) and performance in all race-sex groups (delta C-statistic 0.03-0.11, NRI and IDI values p<0.001). Conclusions: Routinely available clinical data in patients with CKD can reliably predict incident HF. Further study is needed to examine whether implementation of the modified PCP-HF CKD risk prediction equation into clinical practice can improve HF prevention and patient outcomes.