Large U.K. database shows sulfonylurea mortality link

View on the News

Potential confounding clouds observational analyses

Results from clinical trials that included patients treated with sulfonylurea drugs have not shown evidence of harm from these drugs. Major trials such as ACCORD, ADVANCE, and UKPDS ran subgroup analyses to try to look at whether certain drugs seemed to drive poor outcomes, and nothing has been reported for sulfonylureas.

In contrast, observational studies like the ones reported by Dr. Currie and Mr. Morgan have shown links between sulfonylurea treatment and worse outcomes, but these analyses face the problem that physicians prescribe various diabetes drugs to very different groups of patients. We can use fancy statistical models to try to adjust for this, but it’s not clear that they adjust for all possible confounding factors.

For example, one of the big differences between sulfonylureas and other possible second-line drugs that patients might get is cost. So socioeconomic differences might exist between the patients who received a sulfonylurea and those who received a DPP-4 inhibitor that were not included in the adjustments. Many factors enter into the decision for a physician to prescribe one drug or another.

A major limitation of the sulfonylurea drugs is that they have never been the subject of a large, well-designed study to test their cardiovascular safety, such as the studies begun a few years ago for several newer drugs following suggestions of a problem with rosiglitazone. One problem we know for certain about sulfonylureas is that they pose more of a risk for causing hypoglycemic events than do other oral drugs, and there is increasing evidence that severe hypoglycemic events can affect the rates of cardiovascular outcomes. There is also good evidence that the glycemic control exerted by sulfonylureas is not as durable as that of other drugs, and they also cause weight gain, so for several reasons, the sulfonylurea class is gradually being bypassed in favor of other options.

However, many of the alternative agents have their own limitations, especially higher cost. Clinicians need to weigh the suggestions of harm caused by sulfonylurea drugs against their better affordability. If patients can’t afford their diabetes medications, they won’t take them, and undertreated diabetes may pose an even greater risk.

Much more definitive data on the efficacy and safety of sulfonylurea drugs should come from the Cardiovascular Outcome Study of Linagliptin versus Glimeprimide in Patients With Type 2 Diabetes (CAROLINA). These results should be the make or break for sulfonylureas, but data from this study are not expected until 2018.

Dean T. Eurich, Ph.D., is a pharmacoepidemiologist at the University of Alberta in Edmonton. He made these comments in an interview. He said he had no relevant financial disclosures.



BARCELONA – Treatment of type 2 diabetes with a sulfonylurea drug was tied to a significantly increased risk of death in two large U.K. epidemiology studies with a total of more than 100,000 patients.

These and other recent findings that raised questions about the safety and efficacy of sulfonylurea drugs for type 2 diabetes highlight the need for regulatory reassessment of the sulfonylurea drug class, Craig J. Currie, Ph.D., said at the annual meeting of the European Association for the Study of Diabetes.

"The safety of sulfonylureas needs urgent evaluation," said Dr. Currie, professor of applied pharmacoepidemiology at Cardiff (Wales) University. "Regulatory agencies have to take this seriously." He noted that current U.K. recommendations from the U.K. National Institute for Health and Care Excellence (NICE) cite sulfonylurea drugs as the top second-line drug treatment for patients with type 2 diabetes if monotherapy with metformin fails. Last year, management guidelines from the American Diabetes Association and the EASD listed sulfonylurea drugs as candidate second-line agents after metformin along with several other drug classes (Diabetes Care 2012;35:1364-79).

Mitchel L. Zoler/IMNG Medical Media

Dr. Craig J. Currie

But others cautioned of the risk for unrecognized confounding when using observational data to gauge drug safety. "I’m not sure they can get rid of all the confounders; the patients [treated with a sulfonylurea] may just be fundamentally different. In most clinical trials we did not see harm" from sulfonylureas, commented Dean T. Eurich, Ph.D., a pharmacoepidemiologist at the University of Alberta in Edmonton.

Both studies reported by Dr. Currie and his associates used data collected by the Clinical Practice Research Datalink , an observational database of the U.K. National Health Service that routinely collects data from a representative sample of British patients. In one study, they focused on patients who began initial therapy for type 2 diabetes during 2000-2012 with metformin monotherapy, 76,811 patients, or sulfonylurea monotherapy, 15.687 patients. During an average 3-year follow-up, the all-cause mortality rate was roughly 14 deaths/1,000 patient-years for those on metformin, and about 45/1,000 patient-years for those on a sulfonylurea. After adjustment for many potential confounders, including age, sex, body mass index, duration of diabetes, serum creatinine, and hemoglobin A1c level, the analysis found a relative 58% increased risk for death in the sulfonylurea patients compared with those taking metformin, a statistically significant difference, reported Christopher L. Morgan, a researcher at Cardiff University.

Mitchel L. Zoler/IMNG Medical Media

Christopher L. Morgan

The second study compared patients who began a two-drug combination regimen during 2007-2013, comprising 33,983 patients who began metformin plus a sulfonylurea and 7,864 who began metformin plus a dipeptidyl peptidase-4 inhibitor. The adjusted rate of all-cause mortality was a relative 36% higher among patients treated with a sulfonylurea, compared with those treated with a DPP-4 inhibitor, Dr. Currie reported.

Metformin plus a sulfonylurea is the most commonly used drug combination for treating type 2 diabetes worldwide, Dr. Currie noted.

He speculated that sulfonylureas may boost mortality by increasing insulin levels and thereby causing more episodes of severe hypoglycemia than other oral diabetes drugs.

The study was funded by Astra Zeneca and Bristol-Myers Squibb. Dr. Currie said he owns a drug-assessment company that has done work for several drug companies. Mr. Morgan and Dr. Eurich said they had no relevant financial disclosures.

On Twitter @mitchelzoler

Next Article: