Baseline characteristics
Among the study population, 2,321 patients with AHF had DM (43.0%) (Table 1). Patients with DM frequently had other risk factors including old age, obesity, hypertension as well as ischemic heart disease, chronic kidney disease, and cerebrovascular disease. Moreover, patients with DM had a higher proportion of BNP ≥ 500 pg/mL or NT-proBNP ≥ 1000 pg/mL, NYHA class III-IV on admission acute pulmonary edema on chest x-ray, higher level of systolic blood pressure, C-reactive protein, potassium, creatinine concentration, lower level of sodium, and lower LVEF compared to those without DM. As a result, patients with DM were prescribed more parenteral diuretics, inotropic agents, and vasodilators. However, aldosterone antagonists were prescribed less frequently in patients with DM.
Table 1
Baseline clinical characteristics according to diabetes mellitus (DM)
Variables | All patients (N = 5,394) | Non-DM (N = 3,073) | DM (N = 2,321) | P-value |
Age | 68.5 ± 14.5 | 67.6 ± 15.9 | 69.6 ± 12.3 | < 0.001 |
Body mass index (kg/m2) | 23.0 ± 3.9 | 23.0 ± 3.9 | 23.7 ± 3.8 | < 0.001 |
Male, N (%) | 2,872 (53.2) | 1,596 (51.9) | 1,277 (55.0) | 0.023 |
Current smoker, N (%) | 961 (17.8) | 546 (17.8) | 415 (17.9) | 0.086 |
Risk factors, N (%) | | | | |
Hypertension | 3,183 (59.0) | 1,554 (50.6) | 1,629 (70.2) | < 0.001 |
Ischemic heart disease | 1,501 (27.8) | 636 (20.7) | 865 (37.2) | < 0.001 |
Atrial fibrillation | 1,523 (28.2) | 921 (30.0) | 602 (25.9) | 0.001 |
Chronic lung disease | 608 (11.3) | 350 (11.4) | 258 (11.1) | 0.492 |
Chronic kidney disease | 756 (14.0) | 277 (9.0) | 479 (20.6) | < 0.001 |
Cerebrovascular disease | 807 (15.0) | 405 (13.2) | 402 (17.3) | < 0.001 |
Previous heart failure | 2,539 (47.1) | 1,380 (44.9) | 1,159 (49.9) | < 0.001 |
Physical & laboratory Findings | | | | |
SBP, mmHg | 131.4 ± 30.1 | 130.4 ± 29.4 | 132.8 ± 30.9 | 0.003 |
DBP, mmHg | 78.7 ± 18.7 | 79.2 ± 18.8 | 78.1 ± 18.6 | 0.028 |
Heart rate, beats/minute | 92.8 ± 25.9 | 92.5 ± 26.4 | 93.1 ± 25.2 | 0.379 |
Glucose, mg/dL | 155.3 ± 76.7 | 129.6 ± 47.8 | 189.1 ± 94.1 | < 0.001 |
Total Cholesterol, mg/dL | 151.8 ± 43.2 | 153.9 ± 42.2 | 149.2 ± 44.4 | < 0.001 |
BNP ≥ 500 pg/mL or NT-proBNP ≥ 1000 pg/mL | 4,047 (75.0) | 2,267 (73.8) | 1,780 (76.7) | 0.014 |
CRP, mg/dL | 2.4 ± 4.3 | 2.1 ± 3.5 | 2.9 ± 5.0 | < 0.001 |
hsCRP, mg/dL | 2.3 ± 4.2 | 2.0 ± 3.8 | 2.6 ± 4.6 | < 0.001 |
Sodium, mmol/L | 137.5 ± 4.8 | 138.0 ± 4.6 | 136.8 ± 5.0 | < 0.001 |
Potassium, mmol/L | 4.4 ± 0.7 | 4.3 ± 0.6 | 4.5 ± 0.8 | < 0.001 |
BUN, mg/dL | 26.1 ± 16.3 | 23.7 ± 14.3 | 29.2 ± 18.3 | < 0.001 |
Creatinine, mg/dL | 1.5 ± 1.5 | 1.3 ± 1.3 | 1.7 ± 1.6 | < 0.001 |
NYHA class III-IV, N (%) | 4,582 (84.9) | 2,558 (83.2) | 2,024 (87.2) | < 0.001 |
Acute pulmonary edema on chest X-ray, N (%) | 1,039 (19.3) | 502 (16.3) | 537 (23.1) | < 0.001 |
Echocardiographic Findings | | | | |
LVEDD, mm | 57.4 ± 10.1 | 57.5 ± 10.6 | 57.4 ± 9.3 | 0.863 |
LVESD, mm | 45.2 ± 12.3 | 45.1 ± 12.8 | 45.4 ± 11.7 | 0.302 |
LVEF (%) | 37.8 ± 15.6 | 38.5 ± 15.9 | 36.7 ± 15.0 | < 0.001 |
LA volume index, mL/m2 | 63.8 ± 42.1 | 66.7 ± 41.9 | 59.6 ± 42.0 | < 0.001 |
E’, cm/sec | 5.0 ± 2.3 | 5.2 ± 2.1 | 4.8 ± 2.5 | < 0.001 |
S’, cm/sec | 5.1 ± 2.0 | 5.1 ± 2.1 | 5.0 ± 1.9 | 0.026 |
E/E’ | 21.2 ± 11.5 | 20.1 ± 10.8 | 22.7 ± 12.2 | < 0.001 |
RVSP | 43.9 ± 15.1 | 43.2 ± 14.9 | 44.9 ± 15.4 | < 0.001 |
Management, N (%) | | | | |
Parenteral diuretics | 4,062 (75.3) | 2,222 (72.3) | 1,840 (79.3) | < 0.001 |
Parenteral inotropics | 1,672 (31.0) | 760 (24.7) | 912 (39.3) | < 0.001 |
Parenteral vasodilators | 2,231 (41.4) | 1,105 (36.0) | 1,126 (48.5) | < 0.001 |
ACEIs/ARBs at admission | 3,383 (62.7) | 1,977 (64.3) | 1,406 (60.6) | 0.001 |
ACEIs/ARBs at discharge | 3,601 (66.8) | 2,117 (68.9) | 1,484 (63.9) | < 0.001 |
Beta-blockers at admission | 2,054 (38.1) | 1,183 (38.5) | 871 (37.5) | 0.001 |
Beta-blockers at discharge | 2,725 (50.5) | 1,533 (49.9) | 1,192 (51.4) | 0.285 |
AAs at admission | 2,206 (40.9) | 1,379 (44.9) | 827 (35.6) | < 0.001 |
AAs at discharge | 2,443 (45.3) | 1,472 (47.9) | 971 (41.8) | < 0.001 |
Warfarin at discharge | 1,531 (28.4) | 965 (31.4) | 566 (24.4) | < 0.001 |
Heart transplantation | 69 (1.3) | 13 (0.4) | 56 (2.4) | < 0.001 |
Values are presented as mean ± standard deviation, or n (%). |
Abbreviations: DM, diabetes mellitus; SBP, systolic blood pressure; DBP, diastolic blood pressure; BNP, brain natriuretic peptides; NT-proBNP, N-terminal pro-brain natriuretic peptides; hsCRP, high sensitivity C-reactive protein; CRP, C-reactive protein; BUN, blood urea nitrogen; LVEDD, left ventricular end-diastolic dimension; LVEDV, left ventricular end-diastolic volume; LVEF, left ventricular ejection fraction; LA, left atrium; RVSP, right ventricular systolic pressure; ACEIs, angiotensin converting enzyme inhibitors; ARBs, angiotensin receptor blockers; AAs, aldosterone antagonists |
All patients underwent echocardiography during the index admission (Table 1). There were no significant differences in LV end-diastolic dimension (LVEDD) and LV end-systolic dimension (LVESD) between DM group and non-DM group. However, there was significant difference in LVEF (38.5 ± 15.9% vs. 36.7 ± 15.0%, p < 0.001). Furthermore, parameters such as E/e’ (20.1 ± 10.8 vs. 22.7 ± 12.2, p < 0.001), and right ventricular (RV) systolic pressure (43.2 ± 14.9 mmHg vs. 44.9 ± 15.4 mmHg, p < 0.001) reflecting LV diastolic function were more impaired in patients with DM. Conversely, patients without DM had larger LA volume index (66.7 ± 41.9 mL/m2 vs. 59.6 ± 42.0 mL/m2, p < 0.001).
In-hospital And Long-term Mortality According To DM
The median follow-up duration was 3.5 years. During the follow-up period, 235 (4.4%) patients died in during the index hospitalization and 2,500 (46.3%) patients died in overall during follow-up. Patients with DM had a higher incidence of in-hospital mortality and long-term mortality compared to patients without DM (Fig. 2). After adjusting for potential confounders including age, sex, BMI, etiology of heart failure (ischemic vs. non-ischemic), prior admission history due to HF, parenteral inotropics usage, creatinine concentration, and elevated BNP/NT-proBNP (≥ 1000), DM was still independently associated with long-term mortality (adjusted HR 1.12, 95% CI 1.03–1.22).
Independent Predictors Of In-hospital And Long-term Mortality
Results of multivariable Cox proportional hazard regression on in-hospital and long-term all-cause mortality are reported in Table 2. DM, by itself, was not significantly associated with an increased in-hospital mortality (HR 0.81; 95% CI 0.61–1.06, p = 0.127). Parenteral inotropic usage, age, ischemic origin, and higher level of serum creatinine were other strong independent predictors for in-hospital mortality.
Table 2
Independent predictors of in-hospital and long-term mortality on multivariable Cox proportional hazard regression model
Variables | Adjusted HR1 | p value |
In-hospital mortality | | |
Diabetes | 0.81 (0.61–1.06) | 0.127 |
Age (years) | 1.03 (1.02–1.04) | < 0.001 |
Ischemic cause (vs non-ischemic cause) | 1.40 (1.06–1.85) | 0.017 |
Parenteral inotropics usage | 5.16 (3.45–7.72) | < 0.001 |
Serum creatinine ≥ 2.0 (vs < 2.0 mg/dL) | 1.54 (1.15–2.06) | 0.015 |
Long-term mortality | | |
Diabetes | 1.12 (1.03–1.22) | 0.009 |
Age (years) | 1.04 (1.04–1.05) | < 0.001 |
Sex (male) | 1.24 (1.14–1.34) | < 0.001 |
Body mass index (kg/m2) | | |
Underweight vs. Normal | 1.66 (1.47–1.87) | < 0.001 |
Overweight or obese vs. Normal | 0.81 (0.73–0.89) | < 0.001 |
Ischemic cause (vs non-ischemic cause) | 1.16 (1.07–1.26) | < 0.001 |
Prior admission history due to HF | 1.52 (1.40–1.65) | < 0.001 |
Parenteral inotropics usage | 1.42 (1.30–1.55) | < 0.001 |
Serum creatinine ≥ 2.0 (vs < 2.0 mg/dL) | 1.66 (1.50–1.83) | < 0.001 |
Higher BNP (≥ 500), or NT-proBNP (≥ 1000) during index hospitalization | 1.35 (1.21–1.49) | < 0.001 |
1. Adjusted for age, sex, body mass index, etiology of heart failure (ischemic vs. non-ischemic), prior admission history due to HF, parenteral inotropics usage, creatinine concentration (< 2.0 vs. ≥2.0 mg/dL), and elevated BNP (≥ 500) or NT-proBNP (≥ 1000) |
Of note, DM was an independent predictor for long-term mortality (HR 1.12; 95% CI 1.03–1.22, p = 0.009). Other variables, such as old age, male, high BMI, ischemic origin, acute decompensated HF, parenteral inotropic usage, and high concentrations of serum creatinine and higher BNP or NT-proBNP during index hospitalization were also independent predictors for high long-term mortality.
In-hospital and long-term mortality according to DM in subgroup by LVEF
Patients with DM had a higher in-hospital mortality rate in all the LVEF subgroups compared to patients without DM (3.4% vs. 7.1%; 3.2% vs. 4.3%, 2.7% vs. 3.8%, respectively), However, there were no significant associations after adjusting for potential confounders (HFrEF, adjusted HR 0.96, 95% CI 0.68–1.34; HFmrEF, adjusted HR 0.71, 95% CI 0.33–1.53; HFpEF, adjusted HR 0.78, 95% CI 0.41–1.49) (Table 3).
Table 3
In-hospital and long-term mortality according to diabetes in 3 subtypes of HF
Diabetes mellitus (DM) | Unadjusted HR (95% CI) | Adjusted HR (95% CI)1 |
In-hospital mortality | | |
LVEF < 40% | | |
Non-diabetes | 1.00 | 1.00 |
Diabetes | 1.28 (0.92–1.77) | 0.96 (0.68–1.34) |
40%≤LVEF < 50% | | |
Non-diabetes | 1.00 | 1.00 |
Diabetes | 0.83 (0.41–1.68) | 0.71 (0.33–1.53) |
LVEF ≥ 50% | | | | |
Non-diabetes | 1.00 | 1.00 |
Diabetes | 0.94 (0.50–1.77) | 0.78 (0.41–1.49) |
Long-term mortality | | |
LVEF < 40% | | |
Non-diabetes | 1.00 | 1.00 |
Diabetes | 1.48 (1.33–1.64) | 1.14 (1.02–1.27) |
40%≤LVEF < 50% |
Non-diabetes | 1.00 | 1.00 |
Diabetes | 1.19 (0.98–1.44) | 0.99 (0.80–1.23) |
LVEF ≥ 50% |
Non-diabetes | 1.00 | 1.00 |
Diabetes | 1.15 (0.98–1.35) | 1.16 (0.98–1.38) |
1. Adjusted for age, sex, body mass index, etiology of heart failure (ischemic vs. non-ischemic), prior admission history due to HF, parenteral inotropics usage, creatinine concentration (< 2.0 vs. ≥2.0 mg/dL), and elevated BNP (≥ 500) or NTproBNP (≥ 1000) |
In contrast, during follow-up, the presence of DM had different impacts on long-term mortality according to LVEF. In HFrEF, DM was significantly associated with increased long-term mortality even after adjusting for potential confounders (adjusted HR 1.14; 95% CI 1.02–1.27). However, DM did not have significant associations with long-term mortality in patients with HFmrEF (adjusted HR 0.99; 95% CI 0.80–1.23), or HFpEF (adjusted HR 1.16; 95% CI 0.98–1.38) (Table 3). The Kaplan-Meier analysis also revealed significant worse long-term mortality in patients with HFrEF and DM compared with patients with HFrEF but without DM (40.2% vs. 52.7%; Log-rank p < 0.001) (Fig. 3).
Long-term Mortality, According To Prespecified Subgroup And Glycemic Control
Figure 4 presents the association between DM and long-term mortality in stratified group according to potential confounders including age, sex, ischemic origin, hypertension, chronic kidney disease, and de novo HF. The impact of DM on the long-term mortality was generally consistent across stratified subgroups (p for interaction ≥ 0.05). However, there was a significant difference in the impact of DM on long-term mortality between patients with LVEF ≥ 40% (adjusted HR 1.09; 95% CI 0.95–1.25) and LVEF < 40% (adjusted HR 1.27; 95% CI 1.13–1.43) (p for interaction < 0.001).
Figure 5 demonstrated that patients with uncontrolled DM (HbA1c ≥ 7.0%) had significantly higher long-term mortality compared to patients with well-controlled DM (HbA1c < 7.0%) by the Kaplan-Meier analysis (44.0% vs. 36.8%; Log-rank p = 0.016).