Both chronic heart failure (CHF) and diabetes mellitus are associated with high mortality and morbidity, and have increasingly become a growing public health issue.1 Approximately 15%—35% of patients with CHF had diabetes,2 highlighting an important role of diabetes in the pathogenesis of CHF.2,3 Interestingly, the presence of CHF also appears to independently increase the risk for developing diabetes.3 Patients with both CHF and diabetes had a 1.5-2.0-fold higher risk of mortality, compared with those with CHF alone.4 Despite complex interplay between CHF and diabetes, one potential risk factor associated with adverse outcomes is poor glycemic control. Hemoglobin A1c (HbA1c) is a measure of the average blood glucose levels over 2 months,5 and after adjustment for age and sex, each 1% increase in HbA1c was associated with a 12% increased risk of hospitalization for CHF or death.1,2
Currently, spare data are available regarding the prevalence of diabetes in patients hospitalized for CHF, and studies examining the prognosis of CHF patients and the relationship between glucose control and survival in diabetic patients with CHF have been limited with discrepant results.6,7 During the past 2 decades, the incidence of cardiovascular disease and diabetes is increasing in China, and many novel anti-CHF and anti-diabetes therapies have been applied to routine clinical practice, which could have a significant impact on the overall outcome of patients with CHF and diabetes. Under such circumstances, it is pertinent to examine the changes in prevalence of CHF with diabetes and to assess the effects of modern therapy on prognosis of these patients.
This large multi-center cohort study consisted of 1119 patients with CHF who were admitted to 3 hospitals affiliated to Shanghai Jiao Tong University School of Medicine from January 1995 to May 2009. All patients met the diagnostic criteria of ACC/AHA 2005 guidelines for the diagnosis and management of heart failure in adults,8,9 and had symptoms and signs of CHF (NYHA functional class II-IV) and left ventricular ejection fraction (LVEF) <45% by echocardiography. Diabetes was diagnosed based on fasting plasma glucose (FPG) ≥7 mmol/L, 2-hour oral glucose tolerance tests or a random glucose ≥11.1 mmol/L, or receiving hypoglycemic agents or parental insulin10.
Detailed clinical information was collected from review of hospital medical records by two experienced cardiologists. Data concerning the patient's demographics, biochemical and echocardiographic measurements, and anti-CHF and anti-diabetes treatment including β-blockers, angiotensin- converting enzyme inhibitors (ACEI) or angiotensin II receptor blockers (ARB), insulin and hypoglycemic agents were acquired using a special data sheet. The abnormal diastolic filling patterns assessed by Doppler echocardiography included E wave (which reflects early diastolic filling)/A wave (which reflects late atrial contraction) ratio <1, pseudonormal E/A ratio, or an increased E/A ratio >2.11 Follow-up was completed in all patients by chart review and telephone conversation with patients or/and their relatives. Composite major cardiac events (MCE) were defined as death, heart transplantation, and refractory heart failure requiring multiple hospitalizations. The follow-up time was defined as the period from first CHF-related hospitalization to occurrence of composite MCE or to the end of the study (June 2009).
Data are presented as mean ± standard deviation (SD) for continuous variables and percentage of total for categorical variables, respectively. Independent-sample t test and Pearson chi-square test were used for comparison between groups where appropriate. MCE-free survival rate was calculated by the Kaplan-Meier method and tested for significance with the log-rank statistic. Multivariable Logistic regression was carried out to determine risk factors for MCE in total CHF patients by incorporating multiple covariates. Univariate analysis and multivariable Logistic models were used to assess the relationship between outcomes (risk of MCE) and admission HbA1c (using HbA1c <7% as the reference). Logistic model for MCE was constructed using backward stepwise selection with variables with P ≤0.1 being entered into the model after adjustment for gender, age, NYHA class, etiology of CHF, history of cerebrovascular disease or hypertension, diabetes, left atrial diameter, LVEF, hemoglobin, total cholesterol (TC), triglyceride (TG), creatinine (Cr) and uric acid (UA). Statistical Package for Social Sciences (SPSS) for Windows, Version 15.0 was used for all analyses. A 2-sided P value <0.05 was considered statistically significant.
Of the total of 1119 patients with CHF, 798 were (71.3%) men and 321 women, with a mean age of (65±15) years and a mean weight of (65±12) kg. CHF was caused by ischemic disease in 503 (45.0%) patients, dilated cardiomyopathy in 477 (42.6%), and hypertension in 139 (12.4%). Clinically, 858 (76.7%) were in NYHA class III—IV, and 269 (24%) had diabetes. The proportion of CHF patients with diabetes was progressively increased with time (16.9% in 1995-1999, 20.4% in 2000-2004, and 29.1% in 2005-2009, P <0.01) and age (18.5% in <60 years, 26.6% in 60-80 years, and 26.6% in >80 years, P <0.05).
Comparison between diabetic versus non-diabetic patients with CHF
Patients with CHF and diabetes were older, and had higher body weight and systolic blood pressure, lower hemoglobin and LVEF compared with their non-diabetic counterparts. Not unexpectedly, fasting and postprandial serum glucose, HbA1c and C-reactive protein (CRP) levels were elevated in CHF patients with diabetes, but pro-brain natriuretic peptide (pro-BNP) level and glomerular filtration rate (GFR) were similar. Abnormal left ventricular diastolic filling was more frequently observed in CHF patients with diabetes than those without diabetes. The rate of composite MCE was higher (Table 1) but MCE-free survival was lower in CHF patients with than in those without diabetes (Figure).
In patients with CHF and diabetes, measurement of HbA1c was available in 228 patients. Left ventricular end-systolic diameter (LVESD) was greater and LVEF was lower in those with HbA1c ≥7% (all P <0.05 compared to patients with HbA1c <7%) (Table 2). CHF patients with HbA1c ≥7% had poorer MCE-free survival than those with HbA1c <7% (Figure).
Effects of medical treatment
Despite high fasting blood glucose levels ((7.72±3.62) mmol/L vs. (7.39±3.04) mmol/L, P <0.01), patients with CHF and diabetes who received glucose-lowering treatment had significantly lower left atrial diameter ((44±7) mm vs. (47±7) mm, P <0.01), higher LVEF ((40±8)% vs. (36±7)%, P <0.01), and reduced rate of composite MCE (6.6% vs. 17.1%, P <0.05). Similarly, the use of β-blockers (7.1% vs. 19.5%, P <0.01) and ACEI/ARB (9.3% vs. 12.3%, P >0.05) was associated with less MCE in patients with CHF and diabetes.
Multivariable regression analysis
During follow-up (mean (4.57±3.32) years), MCE occurred in 75 patients (cardiac death 77.3%, non-cardiac death 13.3%, heart transplantation 2.7%, and refractory heart failure 6.7%). Multivariable regression analysis revealed that NYHA functional class, diabetes, and LVEF were independently associated with composite MCE in patients with CHF, and NYHA functional class and HbA1c level ≥7% predict adverse outcomes in CHF patients with diabetes (Table 3).
In this study, the prevalence of diabetes in patients with CHF was 24%, consistent with previous reports in clinical trials (15%—35%).2 Importantly, we found that prevalence of CHF with diabetes was dramatically changing over the past decades, with 16.9% in the late 1990s rising up to 29.1% within recent years, and grew with age from 18.5% in patients <60 years to 26.6% for those 60-80 years or older. This phenomenon may be well explained by the current situations with rapid development of national economy and health care. At present, nearly 40 million individuals have diabetes, and almost 15% of total population aged beyond 60 years in China. Although the incidence of cardiovascular diseases is increasing especially for coronary artery disease, evidence-based medical therapies including anti-platelet agents, beta-blocker or ACEI/ARB, and myocardial revascularization have been evolved considerably, which reduced significantly the total mortality during acute phase of cardiac diseases (e.g., myocardial infarction). However, many survivors still had cardiac dysfunction or symptoms of CHF even on aggressive medications. Diabetes contributes to CHF by promoting micro- and macro-vasculopathies as well as through engendering independent diabetes-induced cardiomyopathy.1 Similarly, many elderly patients developed CHF because of increased degenerative or atherosclerotic process or elevated incidence of cardiac (e.g., coronary artery or valvular disease) or non-cardiac diseases (e.g., diabetes).
Our study showed that patients with CHF and diabetes manifested a cluster of abnormal metabolic alterations, which were similar to the findings in diabetic patients without CHF.12 Besides fasting and postprandial hyperglycemia, these patients often had abnormal lipid profiles. In diabetes, lipolysis augments and the heart utilizes exclusively fat acids for ATP generation, leading to high triglyceride in the blood.13-15 Anemia was more common in patients with CHF and diabetes, and anemia itself could contribute to a raised HbA1c and worsen outcome of these patients.12 In this study, all participants were selected specifically as they had at least moderately reduced left ventricular systolic function (LVEF <45%) and more than three-quarters of them were in NYHA class III—IV. Interestingly, abnormal left ventricular diastolic filling also occurred more frequently in patients with CHF and diabetes than in those with CHF alone. Many animal experiments have shown that hyperglycemia and insulin resistance induce abnormal myocardial contractile state and endothelial dysfunction at the cellular level, and diabetes causes interstitial fibrosis and myocardial deposition of advanced glycated end-products, leading to an increase in left ventricular chamber stiffness and impairment in diastolic filling.13,16 From et al17 found that duration of diabetes greater than four years was correlated with significant left ventricular diastolic dysfunction, and the latter was predictive of all-cause mortality in patients with diabetes independent of coronary disease or hypertension. HbA1c is recommended for monitoring glycemic control in patients with diabetes.12,18 Although the pathophysiological mechanisms of HbA1c in aggravation of CHF may be multifactorial, including neurohormonal activation, endothelial dysfunction, oxidative stress, and accelerated atherosclerosis,7 HbA1c was considered as a risk factor for cardiovascular events in patients with symptomatic CHF after adjustment for other confounders.12,18 In this study, patients with CHF and diabetes experienced more composite MCE and had worse survival during follow-up, particularly when serum HbA1c levels were ≥7%. These observations support the notion that diabetes served as an independent risk factor for progression from asymptomatic left ventricular dysfunction to symptomatic CHF, and was associated with an increased risk for either death or composite MCE in patients with CHF, and further substantiate the concept that diabetes has emerged as an important deleterious prognostic factor for patients with CHF.2,19,20
In general, glucose-lowering treatment favorably affects left ventricular function and outcomes for patients with CHF and diabetes. It should be noted that although thiazolidinediones are likely to be of cardiovascular benefit in CHF patients with diabetes in terms of improvement of blood pressure control, endothelial function and anti-atherosclerotic effects, these agents may have the potential to cause edema or weight gain as a result of fluid retention and fat accumulation, which could accelerate the development of CHF.21 Likewise, the effects of metformin remain controversial in the setting of CHF and diabetes.21-23 Consistent with previous reports,16,23 we observed that the use of β-blockers and ACEI/ARB decreases the occurrence of MCE in diabetic patients with CHF. Noticeably, β-blockers could blunt adrenaline release inducing hypoglycemia and lipid abnormality, and decrease insulin sensitivity. Thus, the dosage must be adjusted individually, especially for CHF patients with diabetes.15,19,24
This study has several limitations. First, this is an observational study and subjected to the inherent bias of retrospective nature. However, given the overall large number of patients observed, some differences in this study may be still clinically relevant. Second, certain patients with diabetes may have not been diagnosed at the study entry, which could further increase the already high proportion of patients with CHF and diabetes. Third, all patients were recruited from three hospitals, and there existed some differences in medical treatments of CHF and diabetes. Likewise, the use of anti-diabetes medications was not adjusted based on the degree of glycemic control, and certain other confounders were not measured, which may have influenced clinical outcomes in multivariable analysis. Finally, all patients in this study were Chinese and highly selected, thus they may not be representative of other patient populations.
In summary, the present study demonstrates a significant increase in prevalence of CHF with diabetes over the past decades in China. Patients with CHF and diabetes are often characterized by a cluster of abnormal biochemical and hemodynamic alterations, and have poor clinical outcomes. Glucose-lowering treatment and use of β-blockers could favorably improve the prognosis of these patients.
1. Fonarow GC. An approach to heart failure and diabetes mellitus. Am J Cardiol 2005; 96: 47E-52E.
2. Cohen-Solar A, Beauvais F, Loqeart D. Heart failure and diabetes mellitus: epidemiology and management of an alarming association. J Card Fail 2008; 14: 615-625.
3. Erdmann E, Wilcox RG. Weighing up the cardiovascular benefits of thiazolidinedione therapy: the impact of increased risk of heart failure. Eur Heart J 2008; 29:12-20.
4. Choy CK, Rodgers JE, Nappi JM, Haines ST. Type 2 diabetes mellitus and heart failure. Pharmacotherapy 2008; 28: 170-192.
5. Pazin-Filho P, Kottgen A, Bertoni AG, Russel SD, Selvin F, Rosamond WD, et al. HbA1c as a risk factor for heart failure in persons with diabetes: the Atherosclerosis Risk in Communities (ARIC) study. Diabetologia 2008; 51: 2197-2204.
6. Dai SM, Zhang S, Chen KP, Hua W, Wang FZ, Chen X, et al. Prognostic factors affecting the all-cause death and sudden cardiac death rates of post myocardial infarction patients with low left ventricular ejection fraction. Chin Med J 2009; 122: 802-806.
7. Aguilar D, Bozkurt B, Ramasubbu K, Deswal A. Relationship of hemoglobin A1C and mortality in heart failure patients with diabetes. J Am Coll Cardiol 2009; 54: 422-428.
8. Hunt SA, Abranham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG, et al. 2009 Focused update incorporated into the ACC/AHA 2005 guidelines for the diagnosis and management of heart failure in adults. J Am Coll Cardiol 2009; 53: e1-e90.
9. Zheng ZL, Mcissner A, Hausmann B, Alexander H, Simon R. Prognostic value of Doppler transmitral filling patterns in patients with chronic heart failure. Chin Med J 2004; 117: 176-182.
10. Preiss D, Zetterstrand S, McMurray JJ, Ostergren J, Michelson EL, Granger CB, et al. Predictors of development of diabetes in patients with chronic heart failure in the Candesartan in heart failure assessment of reduction in mortality and morbidity (CHARM) program. Diabetes Care 2009; 32: 915-920.
11. Gary R, Davis L. Diastolic heart failure. Heart Lung 2008; 37: 405-406.
12. Goode KM, John J, Rigby AS, Kilpatrick ES, Atkin SL, Braqadeesh T, et al. Elevated glycated haemoglobin is a strong predictor of mortality in patients with left ventricular systolic dysfunction who are not receiving treatment for diabetes mellitus. Heart 2009; 95: 917-923.
13. An D, Rodrigues B. Role of changes in cardiac metabolism in development of diabetic cardiomyopathy. Am J Physiol Heart Circ Physiol 2006; 291: 1489-1506.
14. Jagasia D, McNulty PH. Diabetes mellitus and heart failure. Congest Heart Fail 2003; 9: 133-139.
15. Stratmann B, Tschoepe D. Sweet heart-contributions of metabolism in the development of heart failure in diabetes mellitus. Exp Clin Endocrinol Diabetes 2008; 116 Suppl 1: S40-S45.
16. Skouri HN, Wilson Tang WH. The impact of diabetes on heart failure: opportunities for intervention. Curr Heart Fail Rep 2007; 4: 70-77.
17. From AM, Scott CG, Chen HH. Changes in diastolic dysfunction in diabetes mellitus over time. Am J Cardiol 2009; 103: 1463-1466.
18. Eshaghian S, Horwich TB, Fonarow GC. An unexpected inverse relationship between HbA1c levels and mortality in patients with diabetes and advanced systolic heart failure. Am Heart J 2006; 151: e1-e6.
19. Burger AJ, Tsao L, Aronson D. Prognostic impact of diabetes mellitus in patients with acute decompensated heart failure. Am J Cardiol 2005; 95: 1117-1119.
20. Berry C, Brett M, Stevenson K, McMurray JJ, Norrie J. Nature and prognostic importance of abnormal glucose tolerance and diabetes in actue heart failure. Heart 2008; 94: 296-304.
21. Nichols GA, Koro CE, Gullion CM, Ephross SA, Brown JB. The incidence of congestive heart failure associated with antidiabetic therapies. Diabetes Metab Res Rev 2005; 21: 51-57.
22. Inzucchi SE. Metformin and heart failure: innocent until proven guilty. Diabetes Care 2005; 28: 2585-2587.
23. Giles TD. The patient with diabetes mellitus and heart failure: at-risk issues. Am J Med 2003; 115: 107S-110S.
24. Lukas MA. Beta blockade in diabetic heart failure. Heart Fail Clin 2006; 2: 89-99.