San Diego, CA - A recent study has indicated that any concerns over the use of thiazolidinediones (TZDs)—often referred to as the "glitazones"—in patients with type 2 diabetes and mild heart failure do not appear to be warranted [1]. Prospective data have shown that the use of rosiglitazone (Avandia, GlaxoSmithKline) in type 2 diabetic patients with NYHA class 1-2 heart failure did not adversely alter echocardiographic structure or function, and most fluid-related changes associated with the insulin-sensitizing agent were not due to a worsening of congestive heart failure.

"This study has shown that rosiglitazone had no adverse effect on ejection fraction, left ventricular volumes, and cardiac structure in subjects with type 2 diabetes and preexisting New York Heart Association class 2 heart failure," said lead investigator Dr John Wilding (University Hospital Aintree, Liverpool, UK). "There was a higher incidence of fluid-related end points in the rosiglitazone group, but a majority of these were not associated with a possible or definite worsening of heart failure and were managed in the clinic with an increase in diuretics."

The results of the study were presented recently here at the American Diabetes Association 2005 Scientific Sessions.

Concern about worsening congestive heart failure

Fluid retention and edema are recognized side effects associated with the peroxisome proliferator-activated receptor-gamma (PPAR-) agonists. Because congestive heart failure is common in patients with type 2 diabetes, there has been concern that treatment with the drugs might lead to worsening of heart failure. The US Food and Drug Administration cautions against the use of TZDs in patients with NYHA class 3-4 heart failure.

However, the benefit of TZDs is also well documented in type 2 diabetic patients, with the American Diabetes Association and the American Heart Association having issued a consensus statement supporting the cautious use of TZDs in patients with class 1-2 heart failure. Wilding explained, however, that the statement is based more on observational data than prospective clinical trials. The purpose of the present study was to confirm the safety of the drugs by measuring changes in ejection fraction after one year of treatment in patients with mild heart failure.

Patients included in the European study were type 2 diabetics treated with diet or oral antidiabetic agents and with an ejection fraction <45%. All patients had a three-month history of NYHA class 1-2 congestive heart failure and were currently being treated with ACE-inhibitor or angiotensin-receptor-blocker therapy.

In total, 224 patients were randomized to rosiglitazone 4 to 8 mg (n=110) or placebo (n=114) plus background antidiabetic agents for 52 weeks. Patients were well matched in terms of age, BMI, and gender and had similar lengths of diabetes duration. Slightly more patients in the rosiglitazone-treated study arm had more severe class 2 heart failure, but baseline ejection fraction and echocardiographic function were similar between the two groups.

At 52 weeks, glycemic control was significantly better with rosiglitazone compared with placebo, with hemoglobin A1_c (HbA1_c) levels reduced 0.65% from baseline compared with no changes in HbA1_c in the placebo-treated patients. There was no significant difference in ejection fractions or any difference in left ventricular end diastolic and left ventricular end systolic volume indices.

Differences between rosiglitazone and placebo at 52 weeks

End point	Rosiglitazone (n=110) (%)	Placebo (n=114) (%)	p
Baseline ejection fraction	35.3	35.7	NS
Ejection fraction at 52 weeks	37.8	36.8	NS

Adjudication of protocol-defined cardiovascular mortality and morbidity end points revealed few events of worsening or possible worsening of congestive heart failure but more edema and dyspnea in the rosiglitazone-treated patients compared with placebo.

"There were quite a large number of cases of edema—28 cases in the rosiglitazone study arm compared with 10 cases in the placebo arm—but a majority of these cases were not adjudicated by the end point committee to be related to worsening heart failure," said Wilding. "Similar results were seen for dyspnea. Looking at the changes in medication, it appears that what was happening here was that the increased symptoms were being managed predominantly by an increase in diuretics."