Evaluating Liver Enzyme Abnormalities in Clinical Trials Involving Diabetic Patients

Abstract

The frequent occurrence of elevated liver enzymes in otherwise healthy diabetics confounds evaluation of drug-induced hepatotoxicity in clinical trials involving diabetic patients. In a recent single-dose tolerance study of an aldose-reductase inhibitor, we found abnormal liver enzyme tests occurring in 19 of 34 (56%) “healthy” diabetic patients. Alkaline phosphatase (AP) deviations occurred in 35%, and serum gluatamic-oxaloacetic transminase (SGOT) elevations occurred in 32%. Of these elevations, 91% were less than 30% above the upper limit of normal in the nondiabetic population. The enzyme elevation patterns varied both between and within patients and were not typical of a specific liver abnormality. We recommend guidelines to perform valid clinical trials in diabetics without either inappropriately excluding potential participants who are healthy or unnecessarily dropping them from the study. The guidelines also allow for maintenance of good safety monitoring practices. We recommend (1) increasing the upper limit of “normal” (acceptable) for serum enzyme tests in diabetic patients, (2) increasing the number of placebo participants, (3) obtaining baseline data at the time of administration of a test dose, and (4) establishing an algorithm to differentiate diabetic liver abnormalities from drug induced hepatitis.