Abstract 12806: Diabetes Mellitus is a Predictor of Repeat Hospitalization for Heart Failure
Objective: To identify predictors of early re-hospitalization for heart failure (HF).
Methods: We used administrative data to identify patients who were admitted to a single academic hospital for HF during a 1 year period from 9/07 through 8/08 and had at least one re-hospitalization for HF within 180 days of hospital discharge. Hospital records were used to collect patient age, gender, ethnicity, language, admission laboratory data (troponin I, serum creatinine, brain natriuretic peptide), ejection fraction (EF), body mass index, personal history of diabetes mellitus (DM), admission to an ICU, length of stay, admitting attending subspecialty, and HF medications prescribed on hospital discharge. Multivariate regression analysis was used to identify clinical predictors of early HF readmission within 90 days of hospital discharge. Subgroup analysis was used to identify differences for systolic and diastolic HF.
Results: 123 patients were met inclusion criteria. At 90 days, 86 patients (70%) were readmitted at least once for HF. In the multivariate analysis, concurrent diagnosis of DM resulted in an additional 0.36 readmissions (CI 0.04–0.68; p=0.027) and hydralazine use resulted in an additional 0.65 readmissions (CI 0.02–1.29; p=0.04). Associations for all other clinical variables were non-significant. There were 58 admissions for systolic HF (EF≤45%) and 43 patients (74%) were readmitted at 90 days. In patients with systolic HF, the concurrent diagnosis of DM resulted in an additional 0.57 readmissions (CI 0.01–1.52; p=0.046), while other clinical variables were non-significant. 54 admissions were documented for diastolic HF (EF>45%) and 36 patients (67%) were readmitted at 90 days. In this subgroup, hydralazine use resulted in an additional 0.85 readmissions (CI 0.11–1.58; p=0.025), while other clinical variables were non-significant.
Conclusion: Re-hospitalization for HF within 90 days of hospital discharge occurs frequently and was seen in more than two thirds of the patients in the present study. HF patients with DM may be at increased risk for early re-admission, especially in those with systolic dysfunction. Future studies are necessary to better define predictors for HF readmission and to develop interventions to reduce HF admissions.
- © 2010 by American Heart Association, Inc.