Issue 110 Item 2 ADA: Thiazolidinediones May Pose Heart Failure Risk

Physicians should thus remain vigilant to CHF symptoms in patients taking these drugs Although thiazolidinediones (TZDs) have a broad spectrum of anti-diabetic and potential vasculoprotective properties, they may increase the risk of congestive heart failure (CHF), researchers reported at the 62nd Scientific Sessions of the American Diabetes Association (ADA).

Thomas E. Delea, MBA, with Policy Analysis, Inc. in Brookline, Massachusetts, United States, and his team used a health insurance claims database with information on roughly 17 million patients annually to examine the risk of CHF in patients treated with TZD. They selected all patients with at least one diagnosis for type 2 diabetes and at least one prescription for an oral anti-diabetic drug during a recent six -year period.

Patients with at least one claim for a TZD were designated "exposed". The date of the first TZD claim for each such patient was designated the "index date". Five unexposed controls were randomly selected for each TZD patient and assigned the same index date. The trial excluded all patients who were not continuously enrolled, those who had a diagnosis of CHF, or those who had a prescription for digoxin or a diuretic in the year before the index date.

A total of 5,445 patients were considered exposed to TZD and had a mean age of 57 years, while the 28,137 controls had a mean age of 59 years.

Those on TZD were more likely to have the following characteristics: coronary artery disease/stroke/transient ischemic attack; diabetes complications; a recent prescription for an angiotensin-converting enzyme inhibitor or a beta blocker or insulin; and higher pre-index medical costs.

The primary outcome measure was the incidence of CHF, as defined by the occurrence of at least one claim with a primary or secondary diagnosis of CHF.

Results showed that TZDs were independently associated with an increased risk of CHF after controlling for differences in baseline clinical characteristics (hazard ratio 1.7, p<0.001). Adjusted Kaplan-Meier estimates of CHF risk at 36 months were 8.2 percent for patients on TZD compared to 5.3 percent for controls.

Delea cautioned that interpretation of the results may be limited by the fact that treatment was not based on random assignment, suggesting that the observed difference in risk may be due to differences in unobserved factors such as hemoglobin A1c levels and body mass index. Also, clinicians were not blinded to treatment. Thus, the observed difference in risk may be due to heightened vigilance to signs and symptoms of heart failure in patients taking a TZD, he said.

The investigators concluded that the results suggest use of TZDs is associated with an increased risk of CHF, and physicians should thus remain vigilant to CHF symptoms in patients taking these drugs. Alternate therapies should be considered in those who develop shortness of breath. The study was sponsored by Novartis Pharmaceuticals. American Diabetes Association’s 62nd Annual Scientific Sessions