Cardiovascular Safety of Current and Emerging Glucose-Lowering Therapies.

Abstract

Cardiovascular disease (CVD) remains the most common major comorbidity in patients with diabetes. Individual short-term studies have failed to show that glycemic control over the short term does not have an impact on CVD. However, until recently, it remained uncertain whether the choice of glycemic drug has a beneficial impact on CVD outcomes. A small study suggesting CVD benefit has resulted in the primacy of metformin for glycemic control. Multiplemeta-analyses of studies with sulphonylurea agents have shown either no CVD benefit or potential harm. Despite the common use of insulin and the potential for both cardiovascular and noncardiovascular harm, only 1 study examining the impact of insulin on cardiovascular outcomes, the Outcome Reduction with Initial Glargine Intervention (ORIGIN) trial, has been undertaken, and the outcome was neutral. The thiazolidinediones (TZDs) have heterogeneous effects; pioglitazone shows potential cardiac benefit and rosiglizone shows possible harm. After a well-publicized meta-analysis of rosiglitazone studies showed potential harm, the US Food and Drug Administration (FDA) issued safety requirements prior to the licensing of new glycemic medications. Since these requirements were instituted, more than 18 studies have been initiated to study dipeptidyl peptidase-4 (DPP4) inhibitors, glucagon-like peptide-1 (GLP-1) agonists and sodium-glucose co-transporter type 2 (SGLT-2) inhibitors. The completed DPP4 and GLP-1 trials meet the safety criteria without improved CVD outcomes. In contrast, the first CVD outcome trial of an SGLT-2 inhibitor, the Empagliflozin Cardiovascular Outcome Event (EMPA-REG) Outcomes trial showed that empagliflozin, in a population of patients with type 2 diabetes and established CVD, reduced CV death and all-cause mortality by 38% and 32%, respectively, and hospitalization for heart failure by 39% in patients with established CVD already treated with metformin. Future guidelines are likely to recommend empagliflozin for patients with type 2 diabetes and CVD. Diabetes confers a substantial risk for the development of atherosclerotic CVD. The life expectancy of a patient with diabetes is 5 to 10 years shorter than that of a person without diabetes (1), largely as a result of the earlier development of coronary heart disease (2). Over the lifetime of a patient with diabetes, there is a greater than 75% chance of developing cardiovascular disease (3), and most patients with diabetes will die as a consequence of CVD. Peripheral vascular disease is common in patients with diabetes, and diabetes is the most common cause for limb amputation. Although patients with diabetes fear blindness, dialysis and amputation, they have a very much greater chance of dying from CVD than suffering one of the feared consequences of diabetes. The risk for CVD increases with the degree of dysglycemia. Both the level of fasting blood glucose and the glucose level after a standard glucose load correlate with the incidence of CVD events (4). The levels of glycated hemoglobin (A1C) predict the incidence of myocardial infarction and mortality (5). However, there is a closer relationship between A1C levels and the development of microvascular events such as nephropathy and retinopathy than is observed for macrovascular CVD events. Over the past 20 years, there has been a marked increase in the number of agents available for glycemic control. Thirty years ago, insulin, first-generation sulphonylureas and phenformin were the only agents available. Subsequently, metformin, glucosidase inhibitors, thiazolidinediones, meglitinides, GLP 1 agonists, DPP4 antagonists and sodium-glucose cotransporter-2 (SGLT2) antagonists have expanded the range of agents for glycemic control. Until recently there was no controlled safety or efficacy data to show whether individual agents had properties beyond glycemic control that were either harmful or beneficial. * St Michael’s Hospital, University of Toronto, 30 Bond Street, Toronto, Ontario M4V 1W5, Canada. E-mail address: fitchettd@smh.ca Can J Diabetes 39 (2015) S176–S182