Association of Heart Failure Symptoms and Subsequent Heart Failure Hospitalizations and Mortality Among Incident Dialysis Patients

Background: Diagnosing heart failure (HF) in dialysis patients is challenging due to overlapping symptoms of volume overload. The Kansas City Cardiomyopathy Questionnaire Score (KCCQ) is a health status questionnaire measuring HF related symptoms and functional limitations, with lower scores indicating greater burden. KCCQ is also an FDA-validated clinical endpoint. We assessed the association between KCCQ and subsequent HF hospitalization and all-mortality in incident dialysis patients. Methods: We studied 625 Chronic Renal Insufficiency Cohort study (CRIC) participants who developed end-stage kidney disease (ESKD), had KCCQ scores post-ESKD diagnosis and had first dialysis modality of either hemodialysis (HD) or peritoneal dialysis (PD). The primary outcomes were rates of HF hospitalization, 30-day HF readmission and all-cause mortality. KCCQ was modeled dichotomously (<75 vs ≥75) and continuously (per 5-point decrement). Poisson regression with robust standard errors was used to test the association of KCCQ score with each outcome. Results: Among 625 participants (mean age 59 years; 60% male; 87% on HD), median follow-up was 6.8 (IQR 3.2-12.0) years. Those with KCCQ <75 had nearly twice the rate of HF hospitalizations (15.4 vs 8.4/100 person-years) and 30-day readmissions (3.7 vs 1.6/100 person-years) compared to KCCQ ≥75 (Table 1). Each 5-point decrement in KCCQ was associated with 9% higher HF hospitalization risk (aRR 1.09, p=0.0003), 13% higher 30-day HF readmission risk (aRR 1.13, p=0.003), and 6% higher all-cause mortality (aRR 1.06; p<0.0001) (Table 1). Stratified by HF subtype, patients with HFrEF and KCCQ<75 had 81% higher HF hospitalization risk vs those with KCCQ ≥75 (p=0.02). Conclusion: The findings suggest that higher burden of early HF symptoms and functional limitations are associated with subsequent HF hospitalizations, 30-day HF readmissions and mortality. The KCCQ symptom score may be a useful tool to help prognosticate risk of poor clinical outcomes in dialysis patients. Funding: Other NIH Support - NIH NCOR-KUH, 1U2CDK140919-01