Background: Although heart failure is a common comorbidity in patients with diabetes, the optimal therapy is highly controversial. Survival in patients with heart failure has been linked to individual antidiabetic agents and to tight glycemic control, regardless of the agent used. Eurich and colleagues evaluated the association between diabetic therapy and outcomes in patients with diabetes and heart failure.

The Study: The meta-analysis included searches of several electronic databases and the reference lists of relevant studies and review articles. Two reviewers independently reviewed each citation to determine whether it met the inclusion criteria; principally, selection of participants with diabetes and heart failure, use of appropriate out-come measures, and use of a control group. Authors were contacted for additional data if necessary.

Results: Of more than 10,000 citations, eight studies met the inclusion criteria for the analysis. Three of the studies had more than two comparison groups. Overall, four studies evaluated insulin therapy, three evaluated metformin (Glucophage), four evaluated thiazolidinediones, and two compared sulfonylureas with other medications.

Insulin was associated with worse outcomes in at least three large studies. In the Survival and Ventricular Enlargement (SAVE) trial, 168 patients treated with insulin had significantly increased rates of all-cause mortality (hazard ratio [HR] = 1.66) and cardiovascular morbidity and mortality compared with 328 patients treated with diet and a sulfonylurea or metformin. Similarly, data from the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) study showed an increased risk of all-cause mortality, cardiovascular mortality, or hospital admission for heart failure in patients treated with insulin compared with patients treated with other antidiabetic therapies. Data from a third study of patients referred to a university medical center showed an unadjusted risk ratio for all-cause mortality of 3.42 at one year in patients treated with insulin compared with those not treated with insulin. Conversely, a retrospective cohort study of Medicare patients with heart failure discharged from the hospital found no excessive mortality risk associated with insulin therapy (adjusted HR = 0.96).

The pooled effect of the metformin studies suggests a statistically significant reduction in hospital admission rates (odds ratio [OR] = 0.85). A large retrospective cohort study of hospitalized Medicare patients and a study of administrative records reported significantly lower all-cause mortality and readmission for heart failure in patients treated with metformin.

The pooled effect of the thiazolidinedione studies suggests a reduced all-cause mortality rate (OR = 0.83), but an increased risk of hospital readmission for heart failure (OR = 1.13).

Conclusion: The authors conclude that, based on evidence mainly from observational studies, metformin is the only therapy studied that was not associated with harm in patients with heart failure and diabetes. They stress the urgent need for more research in this area because of the growing number of patients with diabetes and heart failure.