Abstract MP28: Heart Rate At Hospital Discharge Is Associated With Adverse Long Term Outcomes In Patients With Heart Failure

Abstract

Background: Heart rate (HR) has been identified as a risk factor for adverse long-term outcomes in patients with and without heart disease. However, the relationship between discharge HR and outcomes in patients receiving guideline based therapy for HF remains unknown. Hypothesis: Heart rate at hospital discharge in patients with HF is an independent predictor of mortality and readmission rates at 1 year. Methods: Using the American Heart Association Get With The Guidelines Heart Failure national quality assurance registry linked with Medicare data, long term outcomes for patients ≥65 years of age hospitalized from 2005-2011 were obtained. Univariable and multivariable analyses were performed to examine the relationship between 1-year outcomes (death or all cause readmission) and discharge HR. Heart rate was analyzed as a categorical (tertiles) or continuous variable. Crude and adjusted odds ratios and 95% confidence intervals for the association between discharge HR and the risk of death or hospitalization by 1-year were obtained from logistic regression modeling. Results: The analysis sample was comprised of 45,672 patients with a valid link to Medicare data and 1 year outcome status. The mean age was 80 ±8 years; % male was 45.7; and the mean ejection fraction was 43.5 ±16. Median discharge HRs for the first, second and third tertiles significantly differed (61, 72 and 87 bpm, respectively, p <0.0001). In univariable and multivariable analyses adjusted for clinical and hospital level variables there was a statistically significant relationship between 1 year outcomes and discharge HR (Table). Conclusion: In hospitalized patients with HF, HR at the time of discharge is an independent predictor of mortality. Despite optimal adherence to current guideline based therapy, the odds for all-cause mortality increased by 13% for every 10 beat increases in HR. The use of HR both as a prognostic variable, as well as potentially modifiable risk factor, may be of value in risk-stratification algorithms at discharge.