Autoimmune mechanisms are involved in the pathogenesis of type 1 diabetes mellitus [1]. Organ- or non-organ-specific autoimmune events have been reported to occur after interferon therapy [2]. We report a case of type 1 diabetes mellitus associated with ketoacidosis during interferon- therapy.

A 66-year-old Japanese man (body mass index, 24.2 kg/m²) had a left nephrectomy for renal cell carcinoma on 22 June 1993. He had no overt or family history of diabetes mellitus. One month earlier, his fasting and postprandial blood glucose levels were 5.7 mmol/L and 6.5 mmol/L, respectively. He began receiving interferon-, 5 MU 3 times per week, beginning on 27 July 1993. Interferon- therapy was discontinued on 2 May 1994 because of bone marrow suppression and was restarted on 16 May. The patient reported general fatigue on 8 June and abdominal pain on 13 June. He visited Gunma University Hospital the next day. At that time, his blood glucose level was 22.0 mmol/L, his urinary acetone level was [4], and his arterial blood gas measurements were the following: pH, 7.275; PCO₂, 22.8 mm Hg; PO₂, 99.2 mm Hg; HCO₃^-, 10.3 mmol/L; base excess, –15.3 mmol/L. Diabetic ketoacidosis was diagnosed, interferon- therapy was discontinued, and intravenous insulin was started. Urinary acetone and metabolic acidosis disappeared by 16 June. On admission, the patient's hemoglobin A_1C level was 9.8%, his serum acetone level was 91 mg/L, and his urinary C peptide level was 13.6 µg/d. Islet cell antibodies were absent, but insulin autoantibodies were present (22%; normal, 7%) before insulin therapy. The patient continues to require 20 U of insulin per day to maintain his hemoglobin A_1C level at 7.7%. Renal cell carcinoma has not recurred.

Development of type 1 diabetes mellitus during interferon- therapy [3] and in transgenic mice in which the ß cells expressed interferon- [4] has been reported. However, interferon-induced diabetic ketoacidosis, a condition characteristic of the onset of type 1 diabetes mellitus, has not been reported. Our patient had both ketoacidosis and insulin autoantibodies [5], both of which were compatible with the onset of classic type 1 diabetes mellitus. This case highlights the possible development of type 1 diabetes mellitus associated with ketoacidosis during interferon- therapy.