In a cross-sectional study the frequency of insulin-induced lipohypertrophy at injection sites was assessed in 223 type 1 and 56 type 2 diabetic patients. 64 (28.7%) of the subjects with type 1 diabetes, but only 2 (3.6%) of those with type 2 diabetes presented clinical evidence of lipohypertrophy. In every second affected type 1 diabetic patient lipohypertrophy developed within 2 years after starting insulin therapy. The occurrence of lipohypertrophy was independent of the insulin source and mode of therapy. In a multivariate logistic regression analysis young age, low body mass index, abdominal injection site and, particularly, missing rotation of injection site were significant independent risk factors for the presence of insulin-induced lipohypertrophy. Avoidance of such areas led to a partial or full remission of tissue swellings in 6 of 11 cases under observation for one year. In conclusion, lipohypertrophy is still a frequent complication of insulin therapy. To prevent such local skin reactions insulin-treated patients should be more intensively trained to regularly change injection sites.