Original Article Association of Heart Rate at Hospital Discharge With Mortality and Hospitalizations in Patients With Heart Failure Marlena V. Habal, MD; Peter P. Liu, MD; Peter C. Austin, PhD; Heather J. Ross, MD; Gary E. Newton, MD; Xuesong Wang, MSc; Jack V. Tu, MD, PhD; Douglas S. Lee, MD, PhD Background—Heart failure (HF) is associated with a high burden of morbidity and mortality. Hospital discharge is an opportunity for identification of modifiable prognostic factors in the transition to chronic HF. Methods and Results—We examined the association of discharge heart rate with 30-day and 1-year mortality and hospitalization outcomes in a cohort of 9097 patients with HF discharged from hospital. Discharge heart rate was categorized into predefined groups: 40 to 60 (n=1333), 61 to 70 (n=2170), 71 to 80 (n=2631), 81 to 90 (n=1700), and >90 bpm (n=1263). There was a significant increase in all-cause 30-day mortality with adjusted odds ratios of 1.59 (95% confidence interval [CI], 1.18–2.14; P=0.003) for discharge heart rates 81 to 90 bpm and 1.56 (95% CI, 1.13–2.16; P=0.007) for heart rates >90 bpm when compared with the reference group (heart rates, 61–70 bpm). Cardiovascular death risk at 30 days was also higher with adjusted odds ratio 1.59 (discharge heart rates, 81–90 bpm; 95% CI, 1.09–2.33; P=0.017) and 1.65 (discharge heart rates, >90 bpm; 95% CI, 1.09–2.48; P=0.017). One-year all-cause mortality (adjusted odds ratio, 1.41; 95% CI, 1.16–1.72; P90 bpm when compared with the reference group (heart rates, 40–60 bpm). Readmissions for HF (adjusted hazard ratio, 1.26; 95% CI, 1.04–1.54; P=0.021) and cardiovascular disease (adjusted hazard ratio, 1.29; 95% CI, 1.08–1.54; P=0.004) within 30 days were also higher with discharge heart rates >90 bpm. Conclusions—Higher discharge heart rates were associated with greater risk of all-cause and cardiovascular mortality ≤1-year follow-up and an elevated risk of 30-day readmission for HF and cardiovascular disease. (Circ Heart Fail. 2014;7:12-20.) Key Words: heart failure ◼ heart rate ◼ hospitalization ◼ mortality ◼ prognosis
H
eart failure (HF) is a major cause of morbidity and mortality affecting >5 million North Americans,1 who experience poor prognosis and reduced life expectancy.2 Many patients are treated, stabilized, and discharged from hospital after an acute decompensation, transitioning to the phase of chronic HF. The time of discharge from hospital represents the beginning of chronic HF management. Although the predischarge phase is an opportunity for initiation of potentially beneficial drug therapies, modifiable prognostic factors must first be identified before outcomes can be improved.3
Clinical Perspective on p 20 We previously demonstrated that predischarge blood pressure was a significant predictor of death, with a U-shaped association with mortality.4 A natural extension of the predischarge evaluation is the examination of heart rate, which has long been a variable of interest as a prognostic marker in cardiovascular disease. Early studies suggested that higher
heart rates were associated with a greater risk of death in otherwise healthy men5–7 and those with stable coronary artery disease.8–10 The role of heart rate in chronic HF has become an area of recent interest, which has been further accentuated by the introduction of the If current inhibitor ivabradine that selectively blocks the sinus node to lower heart rate.11 Despite the growing interest in the association of heart rate with outcomes among patients with HF, there is a paucity of literature on the role of heart rate in the community-based population and in important clinical HF subgroups. These clinical subgroups include those with new versus prior HF admissions, reduced versus preserved left ventricular ejection fraction (LVEF), ischemic versus nonischemic HF, and those who are or are not prescribed concomitant β-adrenoreceptor antagonists. In this study, we examined the association of discharge heart rate with all-cause mortality, cardiovascular death, and hospitalizations at 30-day and 1-year follow-up among community-based patients with HF.
Received April 21, 2013; accepted November 13, 2013. From the Institute for Clinical Evaluative Sciences, Toronto, Canada (M.V.H., P.C.A., X.W., J.V.T., D.S.L.); Department of Medicine, Ottawa Heart Institute, Ottawa, Canada (P.P.L.); Institute for Health Policy, Management, and Evaluation, Toronto, Canada (P.C.A., J.V.T., D.S.L.); Peter Munk Cardiac Centre, University Health Network, Toronto, Canada (H.J.R., G.E.N., D.S.L.); Mt. Sinai Hospital, Toronto, Canada (G.E.N.); Sunnybrook Health Sciences Centre, Toronto, Canada (J.V.T.); and Department of Medicine, University of Toronto, Toronto, Canada (M.V.H., P.C.A., H.J.R., G.E.N., J.V.T., D.S.L.). Correspondence to Douglas S. Lee, MD, PhD, Institute for Clinical Evaluative Sciences, University of Toronto, Rm G-106, 2075 Bayview Ave, Toronto, Ontario, M4N 3M5, Canada. E-mail [email protected] © 2013 American Heart Association, Inc. Circ Heart Fail is available at http://circheartfailure.ahajournals.org
DOI: 10.1161/CIRCHEARTFAILURE.113.000429
Downloaded from http://circheartfailure.ahajournals.org/ 12 at CONS CALIFORNIA DIG LIB on April 13, 2015
Habal et al Heart Rate and Heart Failure Mortality 13
Methods
stay. Institutional review board approval was obtained from all participating institutions before data linkage.
Patients We examined patients aged ≥18 years who were admitted with HF in the first and second phases of the Enhanced Feedback For Effective Cardiac Treatment (EFFECT-HF) cohort, which has been described in detail elsewhere.12,13 Briefly, this was a 2-phase population-based retrospective chart review of a sample of patients presenting to acute care hospitals in Ontario, Canada, between 1999 and 2001 (phase I) and between 2004 and 2005 (phase II). To qualify for inclusion in this study, patients were required to meet the modified Framingham HF criteria at presentation and have a primary diagnosis of HF in the discharge abstract.14 Clinical data, including admission vital signs, baseline investigations obtained in the emergency department, medications, and comorbidities, were abstracted from hospital records. Chart abstraction was performed by highly experienced cardiology nurses, with high data reliability.13 To be included, patients were required to be in normal sinus rhythm throughout the duration of the hospital stay. Patients who were diagnosed with HF after admission, transferred from another acute care facility, pregnant, nonresidents, and those with invalid health card numbers were excluded. Patients with resting heart rate 90 bpm. LVEF was determined by echocardiogram, radionuclide angiogram, or cardiac catheterization performed during hospitalization. Patients were classified as newly hospitalized if they had not been hospitalized for HF (International Classification of Diseases [ICD] Code 428.x, 9th revision, or I50.x, 10th revision) within 3 years before the index EFFECT hospital admission, determined by linking the clinical chart data with the Canadian Institute for Health Information Discharge Abstract Database.
Outcomes The primary outcomes were all-cause mortality at 30 days and 1 year. In addition, we examined the effect of heart rate on cardiovascular death and hospital readmission for HF, ischemic heart disease, and cardiovascular disease. Mortality data were obtained through linkages to the Registered Persons Database and the Canadian Institute for Health Information database using the patients’ encrypted health card number. Causes of death were identified from the Ontario Registrar
Table 1. Baseline Characteristics Heart Rate Category, bpm N
40–60
61–70
71–80
81–90
>90
1333
2170
2631
1700
1263
80 (74–85)
79 (74–85)
80 (74–85)
79 (73–84)
P Value
Age, median (25th–75th percentile)
79 (74–85)
Men, n (%)
681 (51.1)
1063 (49.0)
1197 (45.5)
785 (46.2)
Ischemic HF, n (%)
891 (66.8)
1395 (64.3)
1595 (60.6)
1001 (58.9)
681 (53.9)
H
eart failure (HF) is a major cause of morbidity and mortality affecting >5 million North Americans,1 who experience poor prognosis and reduced life expectancy.2 Many patients are treated, stabilized, and discharged from hospital after an acute decompensation, transitioning to the phase of chronic HF. The time of discharge from hospital represents the beginning of chronic HF management. Although the predischarge phase is an opportunity for initiation of potentially beneficial drug therapies, modifiable prognostic factors must first be identified before outcomes can be improved.3
Clinical Perspective on p 20 We previously demonstrated that predischarge blood pressure was a significant predictor of death, with a U-shaped association with mortality.4 A natural extension of the predischarge evaluation is the examination of heart rate, which has long been a variable of interest as a prognostic marker in cardiovascular disease. Early studies suggested that higher
heart rates were associated with a greater risk of death in otherwise healthy men5–7 and those with stable coronary artery disease.8–10 The role of heart rate in chronic HF has become an area of recent interest, which has been further accentuated by the introduction of the If current inhibitor ivabradine that selectively blocks the sinus node to lower heart rate.11 Despite the growing interest in the association of heart rate with outcomes among patients with HF, there is a paucity of literature on the role of heart rate in the community-based population and in important clinical HF subgroups. These clinical subgroups include those with new versus prior HF admissions, reduced versus preserved left ventricular ejection fraction (LVEF), ischemic versus nonischemic HF, and those who are or are not prescribed concomitant β-adrenoreceptor antagonists. In this study, we examined the association of discharge heart rate with all-cause mortality, cardiovascular death, and hospitalizations at 30-day and 1-year follow-up among community-based patients with HF.
Received April 21, 2013; accepted November 13, 2013. From the Institute for Clinical Evaluative Sciences, Toronto, Canada (M.V.H., P.C.A., X.W., J.V.T., D.S.L.); Department of Medicine, Ottawa Heart Institute, Ottawa, Canada (P.P.L.); Institute for Health Policy, Management, and Evaluation, Toronto, Canada (P.C.A., J.V.T., D.S.L.); Peter Munk Cardiac Centre, University Health Network, Toronto, Canada (H.J.R., G.E.N., D.S.L.); Mt. Sinai Hospital, Toronto, Canada (G.E.N.); Sunnybrook Health Sciences Centre, Toronto, Canada (J.V.T.); and Department of Medicine, University of Toronto, Toronto, Canada (M.V.H., P.C.A., H.J.R., G.E.N., J.V.T., D.S.L.). Correspondence to Douglas S. Lee, MD, PhD, Institute for Clinical Evaluative Sciences, University of Toronto, Rm G-106, 2075 Bayview Ave, Toronto, Ontario, M4N 3M5, Canada. E-mail [email protected] © 2013 American Heart Association, Inc. Circ Heart Fail is available at http://circheartfailure.ahajournals.org
DOI: 10.1161/CIRCHEARTFAILURE.113.000429
Downloaded from http://circheartfailure.ahajournals.org/ 12 at CONS CALIFORNIA DIG LIB on April 13, 2015
Habal et al Heart Rate and Heart Failure Mortality 13
Methods
stay. Institutional review board approval was obtained from all participating institutions before data linkage.
Patients We examined patients aged ≥18 years who were admitted with HF in the first and second phases of the Enhanced Feedback For Effective Cardiac Treatment (EFFECT-HF) cohort, which has been described in detail elsewhere.12,13 Briefly, this was a 2-phase population-based retrospective chart review of a sample of patients presenting to acute care hospitals in Ontario, Canada, between 1999 and 2001 (phase I) and between 2004 and 2005 (phase II). To qualify for inclusion in this study, patients were required to meet the modified Framingham HF criteria at presentation and have a primary diagnosis of HF in the discharge abstract.14 Clinical data, including admission vital signs, baseline investigations obtained in the emergency department, medications, and comorbidities, were abstracted from hospital records. Chart abstraction was performed by highly experienced cardiology nurses, with high data reliability.13 To be included, patients were required to be in normal sinus rhythm throughout the duration of the hospital stay. Patients who were diagnosed with HF after admission, transferred from another acute care facility, pregnant, nonresidents, and those with invalid health card numbers were excluded. Patients with resting heart rate 90 bpm. LVEF was determined by echocardiogram, radionuclide angiogram, or cardiac catheterization performed during hospitalization. Patients were classified as newly hospitalized if they had not been hospitalized for HF (International Classification of Diseases [ICD] Code 428.x, 9th revision, or I50.x, 10th revision) within 3 years before the index EFFECT hospital admission, determined by linking the clinical chart data with the Canadian Institute for Health Information Discharge Abstract Database.
Outcomes The primary outcomes were all-cause mortality at 30 days and 1 year. In addition, we examined the effect of heart rate on cardiovascular death and hospital readmission for HF, ischemic heart disease, and cardiovascular disease. Mortality data were obtained through linkages to the Registered Persons Database and the Canadian Institute for Health Information database using the patients’ encrypted health card number. Causes of death were identified from the Ontario Registrar
Table 1. Baseline Characteristics Heart Rate Category, bpm N
40–60
61–70
71–80
81–90
>90
1333
2170
2631
1700
1263
80 (74–85)
79 (74–85)
80 (74–85)
79 (73–84)
P Value
Age, median (25th–75th percentile)
79 (74–85)
Men, n (%)
681 (51.1)
1063 (49.0)
1197 (45.5)
785 (46.2)
Ischemic HF, n (%)
891 (66.8)
1395 (64.3)
1595 (60.6)
1001 (58.9)
681 (53.9)
