Low Body Temperature and Poor Outcomes in Heart Failure
Low Body Temperature and Poor Outcomes in Heart Failure
A total of 4133 patients were enrolled; they were mostly male (74%), white (85%), and had a mean age of 66 ± 12 years. The aetiology of HF was primarily ischaemic (65%), and the mean EF was 27 ± 8%. Their body temperature ranged from 32.2 to 38.8°C and had a mean of 36.4 ± 0.5°C. The median duration of follow-up was 9.9 months, during the course of which 1541 (38%) patients reached the composite endpoint of CV death or HF rehospitalization, and 893 (21%) patients died.
Baseline characteristics of patients in the three groups of body temperature are presented in Table 1. There were 476 patients in the lowest group of body temperature (<36°C) and they were more likely to have a family history of heart disease, lower body mass index, diabetes mellitus, chronic renal insufficiency, longer HF duration, non-ischaemic cardiomyopathy, mitral valvular heart disease, previous pacemaker implantation, history of ventricular arrhythmias, previous defibrillator implantation, and lower EF. Upon presentation, they were more likely to have severe congestion, elevated jugular venous pressure >10 cm, higher respiratory rate, lower heart rate, lower SBP, elevated BUN >26 mg/dL, and hyponatraemia <135 mEq/L.
In unadjusted analysis, body temperature as a continuous variable was associated with all-cause mortality, as well as with the composite endpoint of CV death or HF rehospitalization. For every 1°C decrease in body temperature, compared with patients with body temperature 1°C higher, the risk of all-cause mortality increased by 27% [hazard ratio (HR) 1.27, 95% confidence interval (CI) 1.12–1.44; P = 0.003]; and the risk of CV death or HF rehospitalization by 29% (HR 1.29, 95% CI 1.17–1.42; P < 0.001). The relationship between body temperature and the composite endpoint appeared to be linear (Figure 1), and no evidence of non-linearity was found after testing the restricted cubic spline term in the model (P = 0.20).
(Enlarge Image)
Figure 1.
The incidence of all-cause mortality (cyan bars) and the composite endpoint of cardiovascular death or heart failure rehospitalization (purple bars) plotted against 0.5°C increments of body temperature at randomization.
Body temperature categorized into three groups of <36°C, 36–36.5°C, and >36.5°C was associated with all-cause mortality, as well as with the composite endpoint of CV death or HF rehospitalization in unadjusted analyses. The distribution of all-cause mortality and time to CV death or HF rehospitalization in the three groups of body temperature is shown in Table 2, and the Kaplan–Meier curves of time to events are illustrated in Figure 2. The separation of Kaplan–Meier curves appear to occur early, i.e. in the first month for the <36°C group and in the third month for the 36–36.5°C group. Kaplan–Meier estimate of all-cause mortality at 1 year was 31% in the group with lowest body temperatures, and 27% and 22% in the other groups, respectively (log-rank χ 13.5, P = 0.002); and Kaplan–Meier estimate of CV death or HF rehospitalization at 1 year was 53% in the group with lowest body temperatures, and 45% and 38% in the other groups. respectively (log-rank χ 30.7, P < 0.0001).
(Enlarge Image)
Figure 2.
Unadjusted Kaplan–Meier plots for time to events including all-cause mortality (top) and the composite endpoint of cardiovascular (CV) death and heart failure (HF) rehospitalization (bottom) plotted amongst the three groups of body temperature
Body temperature as a continuous variable was associated with the composite endpoint of death from CV causes or hospitalization for HF in adjusted analyses; for every 1°C decrease in body temperature, compared with patients with body temperature 1°C higher, the risk of adverse outcomes increased by 16% (HR 1.16, 95% CI 1.04–1.28; P = 0.005; after adjustment for age, gender, race, SBP, EF, BUN, and serum sodium) (Figure 3). However, it was not independently associated with all-cause mortality after multivariable adjustment (HR 1.13, 95% CI 0.99–1.29; P = 0.08).
(Enlarge Image)
Figure 3.
Forest plot of hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) for association with the composite endpoint of cardiovascular death and heart failure hospitalization in adjusted analysis for body temperature as a continuous variable. BP, blood pressure; BUN, blood urea nitrogen.
The three categories of body temperature were associated with the composite endpoint of death from CV causes or hospitalization for HF in adjusted analyses. After adjustment for age, gender, race, SBP, EF, BUN, and sodium, the adjusted HR of mortality was 1.35 (95% CI 1.15–1.58; P = 0.0003) with lowest temperatures (<36°C); and 1.14 (95% CI 1.02–1.29, P = 0.02) for the group with body temperatures between 36 and 36.5°C (Figure 4). However, it was not independently associated with all-cause mortality after multivariable adjustment, with HR for the group <36°C = 1.22, 95% CI 0.99–1.51; P = 0.06 and HR for the group 36–36.5°C = 1.13, 95% CI 0.97–1.31; P = 0.11.
(Enlarge Image)
Figure 4.
Forest plot of hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) for association with the composite endpoint of cardiovascular death and heart failure hospitalization in adjusted analysis among the three groups of body temperature. BP, blood pressure; BUN, blood urea nitrogen.
Results
Patient Characteristics
A total of 4133 patients were enrolled; they were mostly male (74%), white (85%), and had a mean age of 66 ± 12 years. The aetiology of HF was primarily ischaemic (65%), and the mean EF was 27 ± 8%. Their body temperature ranged from 32.2 to 38.8°C and had a mean of 36.4 ± 0.5°C. The median duration of follow-up was 9.9 months, during the course of which 1541 (38%) patients reached the composite endpoint of CV death or HF rehospitalization, and 893 (21%) patients died.
Baseline characteristics of patients in the three groups of body temperature are presented in Table 1. There were 476 patients in the lowest group of body temperature (<36°C) and they were more likely to have a family history of heart disease, lower body mass index, diabetes mellitus, chronic renal insufficiency, longer HF duration, non-ischaemic cardiomyopathy, mitral valvular heart disease, previous pacemaker implantation, history of ventricular arrhythmias, previous defibrillator implantation, and lower EF. Upon presentation, they were more likely to have severe congestion, elevated jugular venous pressure >10 cm, higher respiratory rate, lower heart rate, lower SBP, elevated BUN >26 mg/dL, and hyponatraemia <135 mEq/L.
Unadjusted Association of Body Temperature With Outcomes
In unadjusted analysis, body temperature as a continuous variable was associated with all-cause mortality, as well as with the composite endpoint of CV death or HF rehospitalization. For every 1°C decrease in body temperature, compared with patients with body temperature 1°C higher, the risk of all-cause mortality increased by 27% [hazard ratio (HR) 1.27, 95% confidence interval (CI) 1.12–1.44; P = 0.003]; and the risk of CV death or HF rehospitalization by 29% (HR 1.29, 95% CI 1.17–1.42; P < 0.001). The relationship between body temperature and the composite endpoint appeared to be linear (Figure 1), and no evidence of non-linearity was found after testing the restricted cubic spline term in the model (P = 0.20).
(Enlarge Image)
Figure 1.
The incidence of all-cause mortality (cyan bars) and the composite endpoint of cardiovascular death or heart failure rehospitalization (purple bars) plotted against 0.5°C increments of body temperature at randomization.
Body temperature categorized into three groups of <36°C, 36–36.5°C, and >36.5°C was associated with all-cause mortality, as well as with the composite endpoint of CV death or HF rehospitalization in unadjusted analyses. The distribution of all-cause mortality and time to CV death or HF rehospitalization in the three groups of body temperature is shown in Table 2, and the Kaplan–Meier curves of time to events are illustrated in Figure 2. The separation of Kaplan–Meier curves appear to occur early, i.e. in the first month for the <36°C group and in the third month for the 36–36.5°C group. Kaplan–Meier estimate of all-cause mortality at 1 year was 31% in the group with lowest body temperatures, and 27% and 22% in the other groups, respectively (log-rank χ 13.5, P = 0.002); and Kaplan–Meier estimate of CV death or HF rehospitalization at 1 year was 53% in the group with lowest body temperatures, and 45% and 38% in the other groups. respectively (log-rank χ 30.7, P < 0.0001).
(Enlarge Image)
Figure 2.
Unadjusted Kaplan–Meier plots for time to events including all-cause mortality (top) and the composite endpoint of cardiovascular (CV) death and heart failure (HF) rehospitalization (bottom) plotted amongst the three groups of body temperature
Independent Association of Body Temperature With Outcomes
Body temperature as a continuous variable was associated with the composite endpoint of death from CV causes or hospitalization for HF in adjusted analyses; for every 1°C decrease in body temperature, compared with patients with body temperature 1°C higher, the risk of adverse outcomes increased by 16% (HR 1.16, 95% CI 1.04–1.28; P = 0.005; after adjustment for age, gender, race, SBP, EF, BUN, and serum sodium) (Figure 3). However, it was not independently associated with all-cause mortality after multivariable adjustment (HR 1.13, 95% CI 0.99–1.29; P = 0.08).
(Enlarge Image)
Figure 3.
Forest plot of hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) for association with the composite endpoint of cardiovascular death and heart failure hospitalization in adjusted analysis for body temperature as a continuous variable. BP, blood pressure; BUN, blood urea nitrogen.
The three categories of body temperature were associated with the composite endpoint of death from CV causes or hospitalization for HF in adjusted analyses. After adjustment for age, gender, race, SBP, EF, BUN, and sodium, the adjusted HR of mortality was 1.35 (95% CI 1.15–1.58; P = 0.0003) with lowest temperatures (<36°C); and 1.14 (95% CI 1.02–1.29, P = 0.02) for the group with body temperatures between 36 and 36.5°C (Figure 4). However, it was not independently associated with all-cause mortality after multivariable adjustment, with HR for the group <36°C = 1.22, 95% CI 0.99–1.51; P = 0.06 and HR for the group 36–36.5°C = 1.13, 95% CI 0.97–1.31; P = 0.11.
(Enlarge Image)
Figure 4.
Forest plot of hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) for association with the composite endpoint of cardiovascular death and heart failure hospitalization in adjusted analysis among the three groups of body temperature. BP, blood pressure; BUN, blood urea nitrogen.