Do Sulfonylureas Increase Mortality?

Older readers might recall the controversial University Group Diabetes Program study (published in 1970), in which the sulfonylurea tolbutamide was associated with increased cardiovascular mortality rates. The unexpected results triggered years of debate, but the study was largely forgotten as tolbutamide disappeared from use. Now, in a retrospective cohort study using administrative data from Saskatchewan, Canadian researchers have examined mortality associated with the sulfonylurea glyburide and with metformin. From 1991 through 1996, 4138 diabetic patients (mean age, about 65) began glyburide monotherapy, and 1537 began metformin monotherapy.

During an average follow-up of about 5 years, 61 deaths per 1000 person-years occurred among glyburide users, and 40 deaths per 1000 person-years occurred among metformin users. Users of high-dose glyburide had greater risk for all-cause mortality (hazard ratio, 1.3) and death from cardiovascular ischemia (HR, 1.4) than did users of low-dose glyburide. In contrast, users of high-dose metformin did not have greater risk for all-cause or cardiovascular mortality than did low-dose users. These analyses were adjusted for age, chronic disease score, nitrate use, and numbers of physician visits and hospital admissions.

Although results from this retrospective study fall short of proving that metformin is safer than glyburide, they are provocative and should stimulate additional research on comparative outcomes with various oral antidiabetic agents. An editorialist discusses mechanisms at the myocardial cellular level that might explain the association between sulfonylurea use and cardiac complications. At the time of publication, the full text of the original article was available free of charge.

Reference:

1. Simpson SH et al. Dose-response relation between sulfonylurea drugs and mortality in type 2 diabetes mellitus: A population-based cohort study. CMAJ 2006 Jan 17; 174:169-74.

2. Bell DSH. Do sulfonylurea drugs increase the risk of cardiac events? CMAJ 2006 Jan 17; 174:185-6.