821 Association of Heart Rate at Hospital Discharge With All-Cause Mortality in Patients With Heart Failure

Abstract

With the increasing prevalence of heart failure, there has been interest in identifying potentially modifiable prognostic factors. Recent large-scale clinical trials suggest that elevated heart rate is associated with worse outcomes in cardiovascular disease. However, data from community-based heart failure populations are lacking. To examine the association of discharge heart rate with one-year mortality amongst community-based patients who were discharged after a hospitalization for heart failure. We examined 9097 patients (47.1% men) who were hospitalized with heart failure in Ontario, Canada, in the EFFECT Phase 2 study (2004-2005). Patients aged ≥18 years, who presented with heart failure and were in sinus rhythm were examined. Those with resting heart rate <40 beats/min, diastolic blood pressure <80 or >115 mmHg, atrioventricular block, sick sinus syndrome, and pacemaker implant were excluded. Discharge heart rate was categorized into predefined categories: 40-60 (Q1), 61-70 (Q2), 71-80 (Q3), 81-90 (Q4), and >90 (Q5) beats/min. The associations with one-year mortality risks were determined by multiple logistic regression adjusting for pre-defined variables including baseline characteristics, heart failure etiology, ejection fraction, and medications. Analyses were stratified by ischemic etiology and left ventricular ejection fraction (LVEF) ≤45% vs. >45%. There were 1333, 2170, 2631, 1700, and 1263 HF patients with discharge heart rates 40-60, 61-70, 71-80, 81-90, and >90 beats/min. Compared with category Q1 (referent), adjusted odds ratios for one-year mortality were 1.13 (Q2, 95%CI; 0.95-1.33, p=0.16), 1.15 (Q3, 95%CI; 0.97-1.35, p=0.10), 1.21 (Q4, 95%CI; 1.01-1.45, p=0.036), and 1.50 (Q5, 95%CI; 1.24-1.82, p<0.001). Cubic spline analysis demonstrated an approximately linear effect of discharge heart rate on the log odds of one-year death (Figure). Adjusted odds ratios for death were significantly increased for those with HR >90 beats/min and ischemic (OR 1.50, 95%CI; 1.17-1.93, p=0.001) or non-ischemic (OR 1.46, 95%CI; 1.06-2.00, p=0.019) etiologies. Adjusted odds ratios for mortality were increased in those with HR >90 beats/min who were prescribed a beta-blocker at discharge (OR 1.62, 95%CI; 1.21-2.17, p=0.001) and those with projected LVEF >45% (OR 1.60, 95%CI; 1.26-2.03, p<0.001). Higher discharge heart rates were associated with increased risk of one-year mortality among patients presenting with heart failure. Discharge heart rates >90 beats/min were associated with death overall and amongst patients in key heart failure subgroups.