Individual Sulfonylureas and Serious Hypoglycemia in Older People

OBJECTIVE: To compare the risk of serious hypoglycemia associated with the use of individual sulfonylureas in older people. DESIGN: A retrospective cohort study. SETTING: The Tennessee Medicaid Program. PATIENTS: A total of 13,963 Medicaid enrollees, aged 65 years or older, who were prescribed one of six sulfonylureas from 1985 to 1989. MAIN OUTCOME MEASURE: Hospitalization, emergency room admission, or death associated with neuroglycopenic or autonomic symptoms, myocardial infarction, stroke, or injury, with a concomitant blood glucose determination of less than 2.8 mmol/L (50 mg/dL). RESULTS: We identified 255 persons with a first episode of serious hypoglycemia during 20,715 person-years of sulfonylurea use. The crude rate (per 1000 person-years) of serious hypoglycemia was highest in glyburide users, 16.6 (95% confidence interval CI, 13.2 to 19.9 and lowest among users of tolbutamide, 3.5 (95% CI, 1.2 to 5.9). Users of tolbutamide, tolazamide, and glipizide had lower risks of serious hypoglycemia than users of chlorpropamide, whereas the risk of serious hypoglycemia among glyburide users did not differ from that of chlorpropamide users. Among second generation sulfonylureas, the adjusted relative risk of severe hypoglycemia among glyburide users, compared with glipizide users, was 1.9 (95% CI, 1.2 to 2.9). An increased risk of serious hypoglycemia associated with use of glyburide compared with glipizide occurred in all strata, including those defined by gender, race, nursing home residence, dose, and duration of use. CONCLUSIONS: Significant differences in risk of serious hypoglycemia were observed among users of individual agents. This may be explained by duration, timing, or potency of hypoglycemic action. These data confirm previous findings that chlorpropamide use is associated with high risk of hypoglycemia and indicate that among second generation sulfonylureas, glipizide is less associated with hypoglycemia than is glyburide. More information comparing the effectiveness of glycemic control among individual sulfonylureas is needed to assist prescribers in selecting a specific agent for use in clinical practice.