The objective is to define the clinical echocardiographic characteristics and cardiovascular outcome in patients with acute heart failure (HF) with versus without diabetes mellitus (DM). Demographic, clinical, laboratory, and echocardiographic data were collected in Olmsted County adults hospitalized for acute HF between 2005 and 2008. Analyses were performed for mortality and acute HF hospitalization outcomes stratified by diabetic status, systolic function, and diastolic function. There were 912 subjects who met inclusion criteria, and mean age was 79 (SD 13.1) years with 53% women. Prevalence of DM was 42% in the study population, and those with DM had worse diastolic function and increased mortality and HF rehospitalization. Among those with DM and acute HF, reduced left ventricular ejection fraction and worse diastolic function conferred increased HF rehospitalization (p = 0.010 and p = 0.022, respectively). In conclusion, DM is common in those hospitalized for acute HF and is associated with worse long-term clinical outcomes. The subgroup of DM with acute HF and left ventricular systolic dysfunction or diastolic dysfunction had worse HF rehospitalization outcomes.
The prevalence of acute heart failure (HF) and diabetes mellitus (DM) in the aging US population is high and anticipated to grow. An estimated 5.4 million Americans over the age of 20 have HF, with prevalence increasing, and over 1 million annual HF hospitalizations. The Framingham Heart Study revealed that the risk of HF is up to 5 times higher in patients with DM than in patients without DM when controlling for other risk factors. The close correlation between DM and HF has been demonstrated in previous studies which reported that HF is the most common cardiovascular complication in DM, and that patient with DM have a higher risk of being admitted to hospital for HF than patients without DM. Although DM contributes to the progression of chronic HF through atherosclerotic coronary artery disease (CAD) and hypertension, direct effects on cardiac structure also contribute to diabetic cardiomyopathy and HF. Although DM is prevalent and associated with worse outcomes in HF, there is incomplete characterization of outcomes in the subgroups of patients with acute HF and DM with and without systolic or diastolic dysfunction. The objective of this study is to define the clinical and echocardiographic characteristics and cardiovascular outcomes of those who are hospitalized for acute HF and DM. Understanding the clinical outcomes and echocardiographic characteristics among those with acute HF and DM will lead to greater recognition of a unique subset of the population whose conditions have fundamentally different pathophysiology, outcomes, and management strategies. Furthermore, it will allow identification of those at highest risk for adverse clinical outcomes.
Methods
The Mayo Clinic and Olmsted Medical Center institutional review boards approved the study protocol with data obtained from the Rochester Epidemiology Project database of residents in Olmsted County, Minnesota.
We collected demographic, clinical, and laboratory data on adults in Olmsted County, Minnesota who were hospitalized for acute HF between January 2005 and December 2008, with follow-up until 2020. Inclusion criteria were male and female adults, 20 years of age or older, who were hospitalized with a main discharge diagnosis of acute HF identified with International Classification of Diseases, Ninth Revision codes. Those with a recent myocardial infarction, unstable angina, and end-stage renal disease were excluded. Patients with multiple HF hospitalizations were included once they were in the demographic assessments with the index hospitalization for enrollment.
Baseline patient demographics collected included age, gender, ethnicity, body mass index, hypertension history, tobacco history, CAD history, HF history, and chronic kidney disease history. Clinical characteristics include cardiac biomarkers, diabetic labs, metabolic panels, vital signs and echocardiographic data, including ejection fraction, mitral tissue Doppler, and transmitral inflow velocities. Outcomes collected include mortality (and all-cause), and cardiac and HF rehospitalization. Mortality and rehospitalization data were collected from the electronic Mayo Clinic and Olmsted Medical Center medical records.
Continuous variables are presented as mean ± SD and discrete variables as frequency (percentage). Comparisons between patient groups (by DM status, preserved and reduced ejection fraction, and by diastolic function) were made using the t test for continuous variables and the chi-square test for discrete variables. The occurrence of outcomes was estimated using the Kaplan-Meier method, with individual comparisons of the Kaplan-Meier estimates completed using the log-rank test. Multivariate Cox proportional hazard models were constructed for outcomes of interest, adjusting for selected covariates. Statistical significance is defined as a 2-tailed p <0.05. Statistical analyses were performed using SAS Version 9.4 (SAS Institute Inc., Cary, North Carolina).
Results
There were 912 subjects between January 2005 and December 2008 who had a primary hospitalization diagnosis of HF. The mean age of the study population was 79 years and 53% were women. The prevalence of DM in this population was 42% (381/912). Those with DM had higher prevalence of co-morbidities including history of hypertension, CAD, myocardial infarction, and renal diseases. Table 1 demonstrates the baseline characteristics of the study population including demographic data and cardiovascular co-morbidities, stratified by diabetic status.
| Diabetes mellitus | |||||||
|---|---|---|---|---|---|---|---|
| Variable | Overall (n = 912) | Yes (n = 381) | No (n = 531) | P-value | |||
| Age (years) (SD) | 79.3 | (13.1) | 77.3 | (11.8) | 80.7 | (13.8) | <.001 |
| Male | 427 | (47%) | 185 | (49%) | 242 | (46%) | 0.37 |
| Black | 14 | (2%) | 10 | (3%) | 4 | (1%) | |
| Asian | 8 | (1%) | 7 | (2%) | 1 | (0%) | |
| White | 870 | (95%) | 351 | (92%) | 519 | (98%) | |
| Body mass index (kg/m 2 ) | 29.8 | (8.0) | 32.9 | (8.5) | 27.8 | (7.0) | <.001 |
| Smoker | 473 | (62%) | 201 | (62%) | 272 | (62%) | 1.00 |
| Hypertension | 812 | (89%) | 367 | (96%) | 445 | (84%) | <.001 |
| CAD | 695 | (76%) | 314 | (82%) | 381 | (72%) | <.001 |
| Prior MI | 389 | (43%) | 192 | (50%) | 197 | (37%) | <.001 |
| Prior renal disease | 273 | (30%) | 136 | (36%) | 137 | (26%) | 0.001 |
| Hemoglobin (A1c), mean (SD) | 6.6 | (1.3) | 6.8 | (1.3) | 5.8 | (1.1) | <.001 |
| Estimated GFR (ml/min/1.73m 2 ), mean (SD) | 55.7 | (25.8) | 53.7 | (26.4) | 57.1 | (25.4) | 0.08 |
| GFR < 60 | 448 | (62%) | 199 | (67%) | 249 | (59%) | 0.016 |
| BNP (pg/ml) (SD) | 970.1 | (801.2) | 910.0 | (782.0) | 1010.5 | (812.4) | 0.11 |
| Troponin T (ng/ml) (SD) | 0.1 | (0.3) | 0.1 | (0.2) | 0.1 | (0.4) | 0.25 |
| LV ejection fraction (%) (SD) | 45.8 | (17.7) | 46.8 | (16.0) | 45.1 | (18.8) | 0.19 |
| E/A ratio | 1.5 | (0.9) | 1.6 | (0.9) | 1.4 | (0.9) | 0.023 |
| E/e’ ratio | 21.1 | (10.3) | 22.2 | (10.6) | 20.4 | (10.0) | 0.035 |
Echocardiographic analyses showed that among those hospitalized with acute HF, those with DM versus those without had similar left ventricular ejection fraction (46.8% vs 45.1%, p = 0.19). However, those with DM versus those without had worse diastolic function as shown by a higher ratio of peak mitral inflow velocity in early diastole to atrial contraction (E/A) (1.6 ± 0.9 vs 1.4 ± 0.9, p = 0.023) and E/e′ (22.2 ± 10.6 vs 20.4 ± 10.0, p = 0.035).
Among those hospitalized for acute HF, there was increased mortality in the group with DM versus those without DM (99 vs 95%, p <0.001 adjusted for age and gender) ( Figure 1 ). Those with DM versus those without DM had increased risk for acute HF rehospitalizations (75 vs 69%, p = 0.002 adjusted for age and gender) ( Figure 1 ).
Cox proportional hazards model for the patient population with and without DM shows that DM adjusted for age and gender confers a 1.336 (1.163 to 1.533) hazard ratio for mortality. Cox proportional hazards model for the patient population with and without DM shows that DM adjusted for age and gender confers a 1.364 (1.122 to 1.658) hazard ratio for HF hospitalization.
Among 775 subjects hospitalized with HF and DM, 168 had reduced ejection fraction (52%) and 153 had preserved ejection fraction (48%). Among those with DM, those with reduced versus preserved ejection fraction had similar ages (76 vs 78 year, p = 0.12), higher proportion of men (58% vs 37%, p <0.01), lower prevalence of hypertension (95% vs 99%, p = 0.046), and higher prevalence of CAD (89% vs 75%, p <0.001) and myocardial infarction (65% vs 33%, p <0.001). Table 2 demonstrates characteristics for hospitalized patients with acute HF among those with and without DM and with preserved versus reduced ejection fraction.
| Variable | DM rEF (n = 168) | DM pEF (n = 153) | P-value | No DM rEF (n = 254) | No DM pEF (n = 200) | P-value | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Age (years) | 76.3 | (11.7) | 78.3 | (11.2) | 0.12 | 78.1 | (14.3) | 81.2 | (14.2) | 0.022 |
| Male | 97 | (58%) | 57 | (37%) | <.001 | 150 | (59%) | 62 | (31%) | <.001 |
| Ethnicity | 0.37 | 0.72 | ||||||||
| . Black | 3 | (2%) | 4 | (3%) | 2 | (1%) | 1 | (1%) | ||
| . Asian | 1 | (1%) | 3 | (2%) | 0 | (0%) | 1 | (1%) | ||
| . White | 160 | (95%) | 140 | (92%) | 249 | (98%) | 195 | (98%) | ||
| Body mass index (kg/m 2 ) | 32.4 | (7.6) | 33.5 | (9.3) | 0.29 | 27.7 | (6.2) | 28.2 | (8.2) | 0.55 |
| Smoking | 98 | (66%) | 75 | (58%) | 0.17 | 141 | (67%) | 94 | (57%) | 0.042 |
| Hypertension | 159 | (95%) | 151 | (99%) | 0.046 | 197 | (78%) | 177 | (89%) | 0.002 |
| CAD | 150 | (89%) | 114 | (75%) | <.001 | 204 | (80%) | 121 | (61%) | <.001 |
| Prior MI | 110 | (65%) | 50 | (33%) | <.001 | 117 | (46%) | 53 | (27%) | <.001 |
| Renal disease (n) | 60 | (36%) | 53 | (35%) | 0.84 | 71 | (28%) | 39 | (20%) | 0.037 |
| Hemoglobin A1c (%) (SD) | 6.8 | (1.3) | 6.7 | (1.1) | 0.54 | 6.0 | (1.4) | 5.7 | (0.5) | 0.30 |
| Estimated GFR, (mL/min/1.73m 2 ), (SD) | 51.9 | (24.1) | 55.5 | (27.6) | 0.27 | 54.3 | (23.3) | 60.0 | (27.7) | 0.030 |
| GFR < 60 | 96 | (71%) | 73 | (62%) | 0.14 | 123 | (64%) | 97 | (55%) | 0.07 |
| BNP (pg/ml) (SD) | 1155 | (905.5) | 675.9 | (567.5) | <.001 | 1292.4 | (934.9) | 748.8 | (579.4) | <.001 |
| Troponin T (ng/ml) (SD) | 0.1 | (0.2) | 0.0 | (0.1) | 0.10 | 0.1 | (0.5) | 0.0 | (0.1) | 0.014 |
| LV ejection fraction (%) | 33.8 | (10.5) | 61.0 | (5.4) | <.001 | 30.7 | (11.1) | 63.3 | (6.9) | <.001 |
| E/A ratio | 1.6 | (0.9) | 1.6 | (1.0) | 0.54 | 1.5 | (1.0) | 1.3 | (0.7) | 0.12 |
| E/e’ ratio | 22.6 | (10.4) | 21.7 | (10.9) | 0.49 | 21.2 | (10.7) | 19.3 | (9.0) | 0.08 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
