Diabetes mellitus and primary hepatocellular carcinoma.

In view of the increasing incidence of primary hepatocellular carcinoma in western Europe and concern that this may in part be related to long-term use of drugs which cause hepatic microsomal enzyme induction, we undertook a comparison of long-term drug use in 105 patients with hepatocellular carcinoma and equal numbers of age and sex-matched patients with colorectal tumours and with fractures of femur. We found no patients with hepatocellular carcinoma who were long-term anticonvulsant users and only one who used oral contraceptives. However, we observed a four-fold excess of diabetic patients among the group with hepatocellular carcinoma. This association did not appear to be due to pre-existing haemochromatosis, alcoholic cirrhosis or viral hepatitis. The association was strongest in patients receiving drug treatment for diabetes, but the data, although suggestive, were insufficient to determine whether any specific anti-diabetic agent could be responsible. Further studies are required to elucidate the nature of this unexpected association. An association of this magnitude with diabetes mellitus could account at least in part for the increasing incidence of hepatocellular carcinoma in western Europe.