The objective of this study was to evaluate in an open population the incidence and risk factors of biguanide related lactic acidosis. All patients currently treated in the Department of Diabetes and Lipid Metabolism of the Instituto Nacional de la Nutrición and their records were reviewed for the present use or history of administration of biguanides. The study was complemented with a revision of all admissions of diabetic patients to the emergency room during 1987-1990. In the outpatient study, 235 cases were included. No case of lactic acidosis was found. A high percentage of the biguanide treated patients had one or more lactic acidosis related risk factors. In the emergency study, 609 admissions of 273 patients were included. In 17 patients a metabolic non-ketotic acidosis was diagnosed. The frequency of non-ketotic acidosis for the different treatments was: 29.4 cases x 1000 emergency admissions for sulphonylurea treated group, 32 for sulphonylurea plus phenformin treated and 47.94 for type II insulin treated patients. All cases had severe precipitant diseases that can cause lactic acidosis with or without associated biguanide administration. No metformin related cases were found. The conclusions of this study are that biguanides in general and metformin in particular are not associated with a high risk of lactic acidosis. Severe systemic dysfunction associated with intercurrent diseases, frequently observed in diabetic patients, is the main determinant for the appearance of lactic acidosis.