A cohort study of the incidence of serious acute liver injury in diabetic patients treated with hypoglycemic agents

Background: The incidence of acute liver failure or serious liver injury in diabetic patients is needed to evaluate the safety of hypoglycemic drug therapy. Methods: We conducted a retrospective cohort study of 5 health maintenance organizations. Study patients were 171264 health plan members 19 years or older when they received oral hypoglycemic drugs or insulin between April 1, 1997, and June 30, 1999. We searched for hospital discharge diagnoses and procedures potentially indicative of acute liver injury and reviewed the full-text medical records. Acute liver failure was defined as acute liver disease and (1) hepatic encephalopathy, (2) prothrombin time prolongation greater than 3 seconds or international normalized ratio greater than 1.5, and (3) a total bilirubin level greater than 3.0 mg/dL (>51 mumol/L). Acute liver injury was diagnosed in individuals who did not meet I or more of the criteria for acute liver failure but had alanine transaminase or aspartate transaminase levels greater than 500 U/L. Results: We identified 35 cases of acute liver failure or injury not clearly attributable to a known cause other than use of hypoglycemic agents. The age- and sex-standardized incidence per 1000 person-years was 0.15 for insulin users, 0.08 for sulfonylurea users, 0.12 for metformin users, and 0.10 for troglitazone users. The incidence-was higher (on the order of 0.3 per. 1000) during the first 6 months of exposure to all hypoglycemic agents. Conclusions: Acute liver failure or injury not clearly attributable to other known causes occurred on the order of 1 per 10000 person-years among diabetic patients treated with oral hypoglycemic drugs or insulin.