Diabetes and Heart Failure: Which Drug Is Best?

Abstract & Commentary

By Allan J. Wilke, MD, Residency Program Director, Associate Professor of Family Medicine,University of Alabama at Birmingham School of Medicine—Huntsville Regional Medical Campus, Huntsville; Dr. Wilke reports no financial relationship to this field of study.

Synopsis: In patients with heart failure and diabetes, metformin is not associated with harm and may reduce all cause mortality.

Source: Eurich DT, et al. Benefits and harms of antidiabetic agents in patients with diabetes and heart failure: systematic review. BMJ. 2007;335(7618):497.

Diabetes mellitus and congestive heart failure frequently afflict patients simultaneously. More ominously, diabetes potentiates the harm of congestive heart failure. The number of medications used to treat diabetes has grown. In recent months we've learned that some of those medications may exacerbate congestive heart failure.1 This summer the Food and Drug Administration (FDA) issued "black box" warnings regarding the thiazolidinediones.2,3 To help put this information in perspective, Eurich and colleagues performed a systematic review. They identified eight studies, all published within the last two years, which met their inclusion criteria (randomized controlled trial [RCT] or cohort study). Only one study was an RCT; two were post-hoc subgroup analyses from RCTs, four were retrospective cohort studies, and one was a prospective cohort study. Three studies had more than one comparison group. The antidiabetic drugs involved in these studies included insulin (4 studies, n = 9104), metformin (3 studies, n = 3327), thiazolidinediones (4 studies, n = 3409), and sulfonylureas (2 studies, n = 8918). When insulin was compared to treatment with diet, sulfonylureas, or metformin, the patients had a significantly increased risk of all cause mortality and cardiovascular morbidity, including heart failure. Looking specifically at patients with comorbidities of heart failure and diabetes, those treated with insulin had a threefold increase in the risk of all cause mortality. Metformin was risk-neutral for all cause mortality in patients with diabetes and moderate-to-severe impaired left ventricular systolic function in one study and had a significantly lower all cause mortality in another. Pooling the results of four studies of thiazolidinediones, treatment with them was associated with reduction in all cause mortality. However, there also was an increased risk for hospital admission for heart failure. Sulfonylureas were not associated with increased risk of mortality at one year when compared with other antidiabetic agents.

Commentary

The proliferation of therapeutic options in the treatment of diabetes reminds me of where we were in the early 1980s in the treatment of hypertension. The first angiotensin-converting enzyme inhibitor, captopril, had just been released. It joined a slew of diuretics, beta-blockers, calcium channel blockers, and other centrally-acting and peripherally acting antihypertensives. We found ourselves in the situation of having to look at our patients and be cautious not to prescribe an antihypertensive that exacerbated another illness. If we were lucky, we could prescribe one that targeted the high blood pressure and had a beneficial effect on the second condition, ie, "killing two birds with one stone." We may be entering a similar situation in the treatment of diabetes, at least as it pertains to heart failure.

This review runs counter to several beliefs about the drugs we prescribed to diabetics with heart failure. For instance, it was not that long ago that metformin carried a heart failure contraindication. A recent review4 argues that not only does metformin not cause heart failure, the concern of developing lactic acidosis from its use is just so much hooey, too. Thiazolidinediones are associated with fluid retention, and intuitively, we've avoided them in patients with these two conditions. I have difficulty getting my arms around the concept that thiazolidinediones can be associated with both a decrease in all cause mortality and an increase in hospital admission for heart failure. A systematic review5 suggests the reason is that CHF in patients given thiazolidinediones is somehow different than CHF caused by progressive systolic or diastolic dysfunction. Insulin comes out looking bad, but because only one study in this review was an RCT, it's possible that insulin was prescribed to patients physicians perceived to be sicker, thus inserting bias into patient selection.

How should you incorporate these findings into clinical practice? Carefully! What we desperately need is a large trial that randomly assigns diabetic patients with heart failure to each of these drug classes. Until then, what I suggest is this:

1. Metformin is the first choice.

2. If glycemic control is insufficient at maximally tolerated doses of metformin, add a sulfonylurea.

3. If a third drug is needed, either insulin or a thiazolidinedione can be considered, but you will need to monitor the patient closely for exacerbation of heart failure.

References

1. Home PD, et al. N Engl J Med. 2007;357:28-38.

2. http://www.fda.gov/cder/drug/InfoSheets/HCP/pioglitazoneHCP.htm (assessed October 29, 2007)

3. http://www.fda.gov/cder/drug/InfoSheets/HCP/rosiglitazone200707HCP.htm (assessed October 29, 2007)

4. Tahrani AA, et al. Metformin, heart failure, and lactic acidosis; is metformin absolutely contraindicated? BMJ. 2007;335:508-512.

5. Lago RM, et al. Lancet. 2007;370:1129-1136.