THE INCIDENCE OF ETHANOL-INDUCED HYPOGLYCEMIA

The purpose of the study was to determine the incidence of ethanol-induced hypoglycemia. All nondiabetic patients who had blood alcohol levels above 0.10% and a random serum chemistry drawn (including glucose) were included. Over a 6-month study period, 378 patients were included in the study. Fifteen patients (4%) presented with hypoglycemia (glucose < 67 mg%). Of these, four (1%) were profoundly hypoglycemic (glucose < 50 mg%). Two hundred and fifty three patients (67%) patients were normoglycemic (glucose 67-106 mg%) and 110 patients (29%) were hyperglycemic (glucose > 106 mg%). Nonfasting glucose measurements from a control group of 96 nondiabetic, nonintoxicated patients were compared with those of the study group. Two patients (2%) presented with hypoglycemia and none (0%) presented with profound hypoglycemia. Forty-five patients (47%) presented with normoglycemia, and 49 (51%) with hyperglycemia. There were no statistically significant differences found in the hypoglycemia rate between the intoxicated patients and nonintoxicated control population (odds ratio of 0.75; P > .05). Four of 378 intoxicated patients (1%) had profound hypoglycemia (< 50 mg/dL) compared with none (0%) of the 96 nonintoxicated patients. Intoxicated patients had a statistically significant lower rate of hyperglycemia in comparison with the nonintoxicated control group (29% versus 51%; P < .0001). Age, sex, race, ethanol level, and serum electrolyte measurement had no predictive value for hypoglycemia in intoxicated patients. The anion gap was consistently elevated in hypoglycemic patients in comparison with normoglycemic patients (P < .05). Hypoglycemia appears to occur as frequently in an ethanol-intoxicated population as in a population without elevated ethanol levels. Demographic factors were not useful in predicting ethanol-induced hypoglycemia. Altered mental status protocol should be followed in both of these populations as indicated by the clinical situation. © 1992.