Review: Sulfonylureas are associated with overall mortality and CV events vs other antihyperglycemics

A review looked at randomized trials that compared mortality rates among patients taking sulfonylureas versus several other classes of hypoglycemic drugs.


A review of randomized trials found higher rates of overall and cardiovascular (CV)-related mortality and some CV events with sulfonylureas (SUs) compared to several other classes of hypoglycemic drugs, including glucagon-like peptide-1 (GLP-1) agonists and dipeptidyl peptidase 4 (DPP-4) inhibitors. The review included 82 randomized controlled trials with 35,624 patients.

The review was published in the March Diabetes, Obesity and Metabolism. The following commentary by Donald A. Smith, MD, MPH, FACP, was published in the ACP Journal Club section of the April 18 Annals of Internal Medicine.

The meta-analysis by Bain and colleagues found that SUs increase both overall and CV mortality compared with other hypoglycemic drugs, but not with placebo. Although SUs were found to increase CV mortality compared with insulin, the hazard ratio (HR) (1.30) was similar in value to the nonsignificant HRs for placebo (1.25) and biguanides (1.38). More interesting are the much-higher HRs of SUs for CV mortality than with DPP-4 inhibitors (4.42), thiazolidinediones (3.05), GLP-1 agonists (45.4), and sodium–glucose co-transporter-2 (SGLT-2) inhibitors (42.6), suggesting that—rather than harm—SUs may not have the preventive effect for atherosclerotic CV (ASCVD) events of other glucose-lowering agents, especially GLP-1 agonists and SGLT-2 inhibitors.

The ACCORD trial found that intensive glucose-lowering with insulin, SUs, pioglitazone, or metformin increased all-cause mortality and did not reduce ASCVD events (nonfatal MI, stroke, and CV death). The ADVANCE trial, however, found that achieving HbA1c levels of 6.5%, vs 7.3%, more slowly with SUs, insulin, or metformin reduced diabetic nephropathy and did not increase mortality or ASCVD events.

Two major mechanisms are postulated for a possible increase in CV events with SUs. A preceding transient episode of ischemia can induce tolerance to myocardial ischemia–reperfusion injury, resulting in reduced infarct size and ischemic arrhythmias. SUs block the ATP-sensitive K+ channel necessary for producing this ischemic-protective preconditioning. Hypoglycemia caused by SUs may also increase CV events by raising catecholamines and thus, blood pressure, heart rate, platelet activation, and possibly fatal arrhythmias.

Since 2007, all new glucose-lowering agents are required by the US Food and Drug Administration to demonstrate CV safety. In 2 placebo-controlled trials, liraglutide (GLP-1) and empagliflozin (SGLT-2) have reduced CV- and all-cause deaths, and empagliflozin decreased hospitalization for congestive heart failure. Neither significantly reduced MI or cerebrovascular accidents. Two large ASCVD SU comparison studies comparing linagliptin with glimepiride (ClinicalTrials.gov NCT01243424), and pioglitazone with 3 SUs (ClinicalTrials.gov NCT00700856), are expected to be completed in 2018.