LETTER TO THE EDITOR doi: 10.1111/j.1463-1326.2008.00928.x SUR receptor activity vs. incidence of hypoglycaemia and cardiovascular mortality with sulphonylurea therapy for diabetics To the Editor: Evans et al. recently tested their hypothesis if sulphonylur- eas’ ability to bind to the Sulphonylurea Receptor-2 (SUR2) receptor alters cardiac function and survival [1]. Unfortu- nately, their hypothesis was not substantiated with their data analysis. An alternative explanation of their associ- ated cardiovascular mortality may relate to hypoglycaemia. Recently, one half of the Action to Control Cardiovas- cular Risk in Diabetics (ACCORD) study was stopped because of an increase in cardiovascular mortality (11/ 1000 vs. 14/1000 patient years, p < 0.05) in diabetic patients with a goal A1c of <6.0% [2]. This more ag- gressive target almost surely increased the incidence of hypoglycaemia given median A1c of 6.4% in the treat- ment arm vs. 7.5% in the conventional arm. Sulfonylureas have properties in addition to their SUR binding characteristics. For example, glyburide (gliben- clamide) continues to stimulate insulin secretion to a greater extent than glimepiride in the setting of pro- found hypoglycaemia [3]. In addition, glyburide is known to accumulate in the islet beta cells where is it can prolong insulin secretion, whereas other sulfonylureas do not accumulate in the beta cell [4]. Furthermore, glyburide has been recently associated with twice the cardiovascu- lar events (8.8 vs. 4.4%, p < 0.05) in a prospective ran- domized double-blind clinical 1-year trial when compared with pioglitazone treatment [5]. Jain et al. [5] data support the university-related diabetes project where cardiovascular mortality was more than doubled (12.7 vs. 4.9%) in the tolbutamide monotherapy group compared with the placebo-treated diabetic group [6]. Glyburide has recently been identified as the most com- mon agent associated with documented hypoglycaemia [7]. Hypoglycaemia was 1.8 times more likely than with other sulphonylureas including glimepiride [7]. Severe hypoglycaemia has been shown to be eightfold higher compared with glimepiride [8]. A review of several articles has provided data summarized in table 1 listing relative hypoglycemic potential rates and severe (<50 mg/dl) hypoglycemic event rates [7–10]. If the increased mortality seen in the ACCORD trial was because of hypoglycaemia, then is the hypoglycemic potential of each medicine associated with an increased cardiovascular mortality. While each agent may not be associated with an increased cardiovascular risk, those with less hypoglycaemia such as gliclazide, glipizide and glimepiride should be compared with those with a higher rate of hypoglycaemia or more sustained hypo- glycaemia. I challenge Dr Evans and her authors to retest the hypoglycaemia hypothesis with her current data (both relative and severe). Evans results may have been due in part to in vivo human data not supported by the SUR binding data. Both the non-specific SUR receptor–binding agents paradox- ically have different effects on ischaemic preconditioning when tested in humans. Intravenous infusion of glime- piride led to the same time angina and ST depression as placebo [11]. In contrast, glyburide had a greater effect on ST depression and shorter time to angina [11]. These data support the notion that glyburide and glimepiride act dif- ferently in the cardiac tissues than would be expected based on their similar SUR2 binding capacities. Recent data suggest that glimepiride and glyburide have con- trasting effects on cardiovascular mortality. The gly- buride–metformin combination has a 22-fold increase in mortality (8.7 vs. 0.4%) when compared with a glime- piride–metformin combination [12]. Interestingly, there is a 19-fold difference in the rate of severe hypo- glycaemia between these two agents (table 1). This dis- crepancy seen between glyburide and glimepiride may be based on the potential for hypoglycaemia that may explain why cardiac risk is higher with some sulphony- lureas despite similar SUR binding. J. Tayek Internal Medicine, Harbor-UCLA Medical Center, 1000W, Carson Street, Torrance, USA Email: jtakey@ladhs.org Table 1 Rates of Hypoglycemia with Sulphonylurea Therapy Relative rates of hypoglycaemia Severe hypoglycaemia 4.6% Gliclazide 0.85/1000 person years 8.0% Glipizide 8.70/1000 person years 11.5% Glimepiride 0.86/1000 person years 12.3% Tolbutamide 3.50/1000 person years 12.3% Chlorpropamide 16.00/1000 person years 24.0% Glyburide (glibenclamide) 16.00/1000 person years 1128 j Diabetes, Obesity and Metabolism, 10, 2008, 1128–1130 # 2008 The Author Journal Compilation # 2008 Blackwell Publishing Ltd