Skip Navigation

European Journal of Heart Failure 2006 8(4):404-408; doi:10.1016/j.ejheart.2005.10.005
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kamalesh, M.
Right arrow Articles by Tierney, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kamalesh, M.
Right arrow Articles by Tierney, W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2006 European Society of Cardiology

Decreased survival in diabetic patients with heart failure due to systolic dysfunction{star}

Masoor Kamalesh*, Usha Subramanian, Stephen Sawada, George Eckert, M'Hamed Temkit and William Tierney

Krannert Institute of Cardiology and VA Medical Center, Indiana University School of Medicine Indianapolis, IN 46202, United States

* Corresponding author. 1481 W 10th St. VAMC, Indianapolis, IN 46202, United States. Tel.: +1 317 554 0000X2081; fax: +1 317 554 0230. E-mail address: masoor.kamalesh{at}med.va.gov


   Abstract
Top
 Notes
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Background: Prognosis of patients with heart failure (HF) has improved in recent years due to advances in therapy. Whether this is also true for diabetic subjects with HF in clinical practice has not been studied in a prospective manner.

Methods: All patients with HF and left ventricular systolic dysfunction attending the outpatient clinic at our Veteran's Hospital between October 1999 and November 2000 were enrolled in our study and followed prospectively. Electronic medical records were accessed for data on comorbid conditions, medications, echocardiogram results and mortality information. Mean follow-up was 2.7 years.

Results: Of 495 patients with HF due to systolic dysfunction enrolled in the study, 293 (59%) had diabetes. Prevalence of hypertension, diuretic use and angiotensin converting enzyme inhibitor use was higher among diabetics. Beta-blocker usage was equal and high in both groups (60%). On follow-up, 109/273 (37%) patients in the diabetic group died, compared with 49/202 (24%) in the non-diabetic group. Independent predictors of death were diabetes (p<0.005, OR=1.73), age at enrollment (p<0.0001, OR=1.06), serum creatinine (p<0.01, OR=1.44) and diuretic use (p=0.038, OR=1.85). Beta-blocker use was associated with a decreased risk of death on univariate analysis only.

Conclusions: Our results show that diabetic patients with HF continue to have higher mortality than non-diabetic patients with HF despite advances in therapy.

Key Words: Diabetes • Heart failure • Prognosis

Received January 6, 2005; Revised July 12, 2005; Accepted October 5, 2005


   1. Introduction
Top
 Notes
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
The prevalence of diabetes is increasing in the general population [1] and among patient with diagnosis of heart failure due to systolic dysfunction [2]. The presence of diabetes increases mortality in patients with and without established heart disease [3,4]. Acceleration of atherosclerosis, autonomic dysfunction and development of diabetic cardiomyopathy contribute to the increase in heart disease and mortality in diabetics [3,4]. Despite this, HF in patients with diabetes appears to be under-studied and warrants further investigation [5,6]. Retrospective analyses of clinical trials enrolling selected populations in a randomised controlled trial setting have shown that survival of diabetic patients with HF may be poorer than those with HF but no diabetes [7-10]. In recent years, there have been a number of advances in medical therapy that have improved the prognosis of patients with HF [2]. The impact of these advances on the outcome of diabetics with HF treated in a tertiary care facility, outside of the large clinical trials, is largely unknown.

We therefore sought to investigate the effect of diabetes on survival in patients with HF prospectively at a big university teaching hospital.


   2. Methods
Top
 Notes
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
The study was approved by the Institutional Review Boards of Indianapolis Veterans Affairs Medical Center. All subjects with a diagnosis of HF and LVSD attending the outpatient clinics at our Veteran's Affairs Medical Center between October 1999 and November 2000 were enrolled.

2.1. Inclusion criteria
Patients were recruited during scheduled general medicine practice visits and were eligible if they had both an active diagnosis of heart failure (as determined by the treating primary care physician) and evidence of left ventricular systolic dysfunction on echocardiogram, nuclear cardiac scan, or cardiac catheterization. Patients with a heart failure diagnosis without confirmatory objective evidence of systolic dysfunction underwent an echocardiogram. Left ventricular systolic dysfunction was estimated from the cardiologist's visual impression of the ejection fraction plus the fractional shortening as measured by the sonographer. In case of discrepancy between the two, the cardiologist's impression was taken as the final estimate. Patients with objective evidence but no recorded diagnosis of heart failure were eligible if their primary care physicians confirmed the presence of active heart failure.

2.2. Exclusion criteria
We excluded patients who were not expected to survive for 1 year or who could not provide subjective information due to psychosis, cognitive impairment, hearing loss, or no telephone access. Diabetes was diagnosed if subjects were on oral hypoglycemics or insulin therapy.

For data sources, we used the Department of Veterans Affairs national electronic medical record for comorbid conditions, echocardiogram results and pharmacy data [10a]. For outcomes we used the Beneficiary Information and Resource Locator (BIRLS) to determine whether a date of death existed [10b]. BIRLS contains date of death but not cause of death, hence we assessed all-cause mortality.

2.3. Statistical analyses
Survival was compared between patients with and without diabetes. Student's t test was used to compare the means between groups when the variable was continuous and normally distributed. If the departure from normality was significant, Wilcoxon's rank sum test was used to compare the medians between the groups. For categorical variables, proportions were compared using Pearson's {chi}2 test when the samples were large and Fischer's Exact Test when n was less than 100. Cox regression analysis was used to determine predictors of mortality. Kaplan-Meier curves were drawn to assess long-term trends in mortality and p-values from the log rank test calculated to compare the curves. A p-value <0.05 was considered statistically significant.


   3. Results
Top
 Notes
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
We screened 1090 consecutive patients with a diagnosis of HF during the study enrollment period. Of these, 595 were excluded and were as follows: not able to meet enrollment criteria (n=401), refusal to participate (n=149), withdrawal of consent (n=6), incomplete baseline assessments (n=6) and other reasons, including unavailability of echo reports and death (n=33).

A total of 495 patients with HF due to systolic dysfunction were enrolled in the study, of these, 293 (59%) had diabetes. Almost all the patients (98%) were male and the majority were white (84%). The baseline characteristics of all patients according to the presence or absence of diabetes are shown in Table 1. Prevalence of hypertension, diuretic use and angiotensin converting enzyme use were higher in the diabetic group. There was no difference in beta-blocker usage between the groups (approximately 60%). Systolic function measured by fractional shortening was higher in the diabetic group.


View this table:
[in this window]
[in a new window]

  		Table 1 Baseline characteristics

 
Mean follow-up time was 2.7 years and maximum follow-up was 3.6 years. During follow-up 109/293 (37%) patients in the diabetic group died, while 49/202 (24%) patients died in the non-diabetic group (p=0.003). The combined end point of death and first readmission for HF was also significantly higher for the diabetic group (168/293 patients) compared with 93/202 for non-diabetics (p=0.02). Independent predictors of death hazards were diabetes (p<0.005, Odds ratio (OR) =1.73), age at enrollment (p<0.0001, OR 1.06) and serum creatinine (p<0.01, OR=1.44) (Tables 2 and 3). Diuretic intake doubled the risk of death in the diabetic group (p=0.038, OR=1.85). Beta-blocker usage was associated with decreased odds of death only on univariate analysis but was no longer significant when assessed in multivariate analysis (Tables 2 and 3). Within the diabetic group, a higher age at enrollment and higher serum creatinine at baseline was more commonly associated with increased mortality (Table 4). Unadjusted Kaplan-Meier mortality curves are shown in Fig. 1 along with log rank p-values. The mean glycosylated hemoglobin (HbA1c) values for the diabetic cohort was 7.027 (S.D.=1.633). Among the diabetics, 40% were on insulin treatment during the year prior to study enrollment. There was no ‘time to death’ difference between diabetics not on insulin treatment and diabetics on insulin treatment (hazard ratio 1.96, p-value 0.16).


View this table:
[in this window]
[in a new window]

  		Table 2 Univariate Cox proportional hazards survival model for time of death

 


View this table:
[in this window]
[in a new window]

  		Table 3 Multivariate Cox proportional hazards survival modelfor time of death including all the covariates

 


View this table:
[in this window]
[in a new window]

  		Table 4 Baseline characteristics according to outcome for diabetic patients

 


Figure 1
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Kaplan-Meier survival curves for diabetic and non-diabetic patients. Log rank p=0.003.

 

   4. Discussion
Top
 Notes
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
Our study prospectively looked at a cohort of patients with HF being treated at a University affiliated teaching hospital. It differs from previous reports which are based on selected populations from randomized clinical trials [7-10] and as such are not truly representative of the patients seen and treated in daily clinical practice. We investigated the effect of diabetes on mortality in HF patients with systolic left ventricular dysfunction. We found that despite modern management, the presence of diabetes significantly worsens prognosis in patients with HF and increases the risk of death by 30% compared to subjects without diabetes.

At baseline, the incidence of coronary disease was similar. Higher incidence of hypertension in patients with diabetes is well known. Fractional shortening was marginally but significantly higher in the diabetic group. The majority of patients in both groups were on angiotensin converting enzyme inhibitors (or receptor blockers) and beta-blockers, in keeping with the evidence from trials showing their unequivocal benefit in reducing mortality in HF. Compared to recent Medicare data on the frequency of use of beta-blockers and angiotensin converting enzyme inhibitors, our patients were more optimally treated, in accordance with recent guidelines [15,16]. It is heartening to note the high usage of beta-blockers in the diabetic groups, as for a long time, beta-blockers were thought to be relatively contraindicated in patients with diabetes.

Our results show that in HF, diabetic patients have a 30% higher mortality compared to non-diabetic patients. These results are consistent with prior reports from randomized clinical trials which show increased mortality from HF in diabetics [7-10]. Shindler and coworkers reporting on the diabetic cohort from the SOLVD trials and registry [10] showed 50% higher all-cause mortality in the diabetic group. While the difference in all-cause hospitalization was of borderline significance in this study (p=0.05), death or HF hospitalization was significantly higher in the diabetic group (p<0.001, OR=1.37). However, this study was done before the widespread use of beta-blockers for the management of HF. Dries and coworkers also retrospectively analyzed the SOLVD data and reported a strong association of diabetes with all-cause mortality (relative risk of 1.37, confidence interval=1.21 to 1.55; p<0.0001) in those with ischaemic cardiomyopathy. More recently, Vaur and coworkers reported on outcomes in HF subjects with diabetes from the DIABHYCAR (type 2 diabetes, hypertension, cardiovascular events and ramipril) study [8]. They prospectively followed 4912 subjects with type 2 diabetes for 3-6 years and found a 12 times higher annual mortality among diabetics who developed HF on follow-up than in those diabetics who did not develop HF. In this era of beta-blockers, a recent retrospective analysis of the BEST (Beta-blocker evaluation of survival trial) reported [7] that diabetes was independently associated with increased mortality (adjusted hazard ratio=1.33, 95% CI=1.12-1.58, p=0.001), but only in the group with ischaemic aetiology. At 12 months follow-up, there was relative risk reduction of 24.5% and 20% in the diabetes and non-diabetes groups for the combined endpoint of death or HF hospitalization. Thus, there appeared to be a greater reduction in the combined endpoint with beta-blockers in the diabetic group.

All these trials suffer from the drawback that they are retrospective analyses and based on patients enrolled in clinical trials, thus they differ from ‘real life’ clinical practice. The vast majority of patients in our cohort had a history of coronary artery disease, thus making it comparable to real life clinical practice. Our finding of a 30% increased risk of death in diabetic subjects with HF is slightly better than the percentages reported in the above-mentioned studies. This could be due to the higher use of beta-blockers in our study, reflecting current HF management guidelines. Also, lack of a significant difference in the combined endpoint in our study could be due to a greater effect of beta-blockers in reducing HF re-admissions in the diabetic group, similar to the results of the BEST trial.

A recent report by De Groote et al. [18] concluded that prognosis of heart failure in subjects with diabetes is determined by the underlying aetiology, and the presence of ischaemic heart disease makes it worse. In their study, diabetes was not an independent predictor of mortality in non-ischaemic patients. It is interesting to note that approximately half their patients overall had diabetes and among the diabetics, 60% had ischaemic heart disease. In our population, the incidence of ischaemic heart disease was much higher in both the diabetic and non-diabetic groups (77% and 84%, respectively), making it mostly a comparison of heart failure with ischaemic etiology in subjects with and without diabetes. In the De Groote study, like ours, the ischaemic cohort was predominantly male (over 90%) and overall mortality was significantly higher in the diabetic group. Beta-blocker use was much higher in our study (60% compared to 40%) while beta-blocker use in the De Groote non-ischaemic cohort was particularly low (16% in diabetic and 23% in non-diabetic). Further, while age was comparable between the groups in our study, in the De Groote study there was a significant difference, the non-ischaemic cohort was significantly younger, both among diabetics and non-diabetics. These differences may have accounted for some of their findings in the overall group, while their group with ischaemic etiology showed results which were similar to ours.

Acceleration of atherosclerosis, autonomic dysfunction and development of diabetic cardiomyopathy contribute to the increase in heart disease and mortality in diabetics [3,4]. Diabetic cardiomyopathy in the absence of epicardial coronary artery disease, has been the subject of intense investigation [5,17]. Studies in 1970s and 80s reported on the occurrence of left ventricular hypertrophy, fibrosis and myocardial enlargement in diabetics in the absence of hypertension. Other studies point to a diabetic vasculopathy which predominantly involves smaller intramyocardial vessels. Thus, it appears that diabetes may be associated with a specific cardiomyopathy that may contribute to increased mortality despite adequate modern medical management.

It must also be mentioned that diabetes is a metabolic disease and control of blood sugar is important in improving prognosis. Although this was not the aim of our study, we feel that involving a nutrition and metabolism expert in the care of diabetic subjects in general and those with coronary disease in particular may be beneficial in improving long-term prognosis.

Some of the limitations of our study are as follows. The data was collected predominantly in men, and hence should be extrapolated to women with caution. We have no data on the use of aldosterone receptor antagonists, which have been shown to reduce mortality in HF due to systolic dysfunction [11]. Further, our study provides no information regarding prognosis of HF due to diastolic dysfunction in diabetic patients. Diastolic dysfunction is very common in diabetes, the incidence may be as high as 60% even in well-controlled diabetics when examined carefully [12]. The role of glycemic control was not specifically addressed in our study. There is evidence that better glycemic control can improve myocardial metabolism and cardiac function [13,14].

In summary, despite advances in therapy, patients with diabetes have a poorer prognosis from HF compared to those without diabetes. Aggressive preventive programs should be directed at subjects with diabetes to improve outcome.


   Notes
Top
 Notes
Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
{star} Presented at the 8th annual scientific meeting of the Heart Failure Society of America in Toronto, Canada, Sept. 12–15, 2004. Back


   References
Top
 Notes
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. King H., Aubert R.E., Herman W.H. Global burden of diabetes, 1995-2025. Prevalence, numerical estimates and projections. Diabetes Care (1998) 21:1414–1431.[Abstract]
  2. Kamalesh M., Nair G. Disproportionate increase in prevalence of diabetes among patients with congestive heart failure due to systolic dysfunction. Int J Cardiol (2005) 99(1):125–127. Mar 10 [Review].[CrossRef][Web of Science][Medline]
  3. Nichols G.A., Hillier T.A., Erbey J.R., Brown J.B. Congestive heart failure in type 2 diabetes: prevalence, incidence and risk factors. Diabetes Care (2001) 24:1614–1619.[Abstract/Free Full Text]
  4. Aronow W.S., Ahn C. Incidence of heart failure in 2737 older persons with and without diabetes mellitus. Chest (1999) 115:867–868.[CrossRef][Web of Science][Medline]
  5. Bell D.S.H. Heart failure: the frequent, forgotten and often fatal complication of diabetes. Diabetes Care (2003) 26:2433–2441.[Abstract/Free Full Text]
  6. Ryden L. Diabetes mellitus and congestive heart failure: further knowledge needed. Eur Heart J (1999) 20:789–795.[Free Full Text]
  7. Domanski M., Krause-Steinrauf H., Deedwania P., et al. The effect of diabetes on outcomes of patients with advanced heart failure in the BEST trial. J Am Coll Cardiol (2003) 42:914–922.[Abstract/Free Full Text]
  8. Vaur L., Gueret P., Lievre M., Chabaud S. Passa; DIABHYCAR Study Group. Development of congestive heart failure in type 2 diabetic patients with microalbuminuria or proteinuria. Diabetes Care (2003) 26:855–860.[Abstract/Free Full Text]
  9. Dries D.L., Sweitzer N.K., Drazner M.H., Stevenson L.W., Gersh B.J. Prognostic impact of diabetes mellitus in patients with heart failure according to the etiology of left ventricular dysfunction. J Am Coll Cardiol (2001) 38:421–428.[Abstract/Free Full Text]
  10. Shindler D.M., Kostis J.B., Yusuf S., et al. Diabetes mellitus, a predictor of morbidity and mortality in the studies of left ventricular dysfunction (SOLVD) trials and registry. Am J Cardiol (1996) 1017–1020. Martin D.K. Making the connection: the VA-Regenstrief project. MD Comput (1992) 9:91–96.[Medline] Fisher S.G., Weber L., Goldberg J., Davis F. Mortality ascertainment in veteran population: alternatives to the National Death Index. Am J Epidemiol (1995) 141:242–250.[Abstract/Free Full Text]
  11. Pitt B., Zannad F., Remme W.J., et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med (1999) 341:709–717.[Abstract/Free Full Text]
  12. Poirier P., Bogaty P., Garneau C., Marois L., Dumesnil J.G. Diastolic dysfunction in normotensive men with well controlled type 2 diabetes. Diabetes Care (2001) 24:5–10.[Abstract/Free Full Text]
  13. Malmberg K., Ryden L., Efendic S., et al. Randomized trial of insulin glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI) study. Effects on mortality at 1 year. J Am Coll Cardiol (1995) 26:57–65.[Abstract]
  14. Stratton I.M., Adler A.I., Neil H.A., et al. Association of glycemia with macrovascular and microvascular complications of type 2 diabetes(UKPDS 35):prospective observational study. BMJ (2000) 321:405–412.[Abstract/Free Full Text]
  15. Krumholz H.M., Baker D.W., Ashton C.M., et al. Evaluating quality of care for patients with heart failure. Circulation (2000) 101:e122.[Web of Science][Medline]
  16. Packer M. The impossible task of developing a new treatment for heart failure. J Card Fail (2002) 8:193–196.[CrossRef][Web of Science][Medline]
  17. Fein F.S., Sonnenblick E.H. Diabetic cardiomyopathy. Prog Cardiovasc Dis (1985) 27:255–270.[CrossRef][Web of Science][Medline]
  18. De Groote P., Lamblin N., Mouquet F., et al. Impact of diabetes mellitus on long-term survival in patients with congestive heart failure. Eur Heart J (2004) 25:656–662.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Eur J Heart FailHome page
N. M. Hawkins, M. C. Petrie, P. S. Jhund, G. W. Chalmers, F. G. Dunn, and J. J.V. McMurray
Heart failure and chronic obstructive pulmonary disease: diagnostic pitfalls and epidemiology
Eur J Heart Fail, February 1, 2009; 11(2): 130 - 139.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
M. R. MacDonald, M. C. Petrie, N. M. Hawkins, J. R. Petrie, M. Fisher, R. McKelvie, D. Aguilar, H. Krum, and J. J.V. McMurray
Diabetes, left ventricular systolic dysfunction, and chronic heart failure
Eur. Heart J., May 2, 2008; 29(10): 1224 - 1240.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kamalesh, M.
Right arrow Articles by Tierney, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kamalesh, M.
Right arrow Articles by Tierney, W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?