Increased Insulin Requirement in a Patient with Type 1 Diabetes on Rifampicin

Infection may have a major effect on diabetes control, and patients with diabetes have an increased susceptibility to some infections, including tuberculosis.’ Pulmonary tuberculosis is reported to be increasing in inner city areas,2 thus more patients may be seen in diabetic clinics on anti-tuberculous chemotherapy. We report a case of a patient with Type 1 diabetes mellitus whose insulin requirements increased during rifampicin therapy for pulmonary tuberculosis. A 54-year-old woman with Type 1 diabetes, free of diabetes-related complications, was diagnosed as having pulmonary tuberculosis and commenced on triple therapy with Rifinah 300 (rifampicin 300 mg and isoniazid 150 mg) two tablets daily, and pyrazinarnide 1500 mg daily. She was maintained on a total of 36 units of human actrapid and monotard insulin on a twice daily regime which was not changed on starting anti-tuberculous chemotherapy. She was on no other medication. Two weeks later she was admitted to hospital with nausea and vomiting and a blood glucose of 28 mmol I-’, but with no evidence of ketoacidosis. Her nausea and vomiting settled after stopping the pyrazinamide, and she was discharged 5 days later on Rifinah 30, two tablets daily. Glycaemic control was only achieved after increasing her insulin to 48 units per day. During the remaining 8 months of her antituberculous chemotherapy her diabetes remained well controlled on 44-48 units of insulin per day, and her weight remained steady at 79 kg. She experienced only one or two minor hypoglycaemic episodes per month, usually precipitated by exercise. Immediately on discontinuing her anti-tuberculous chemotherapy, she developed frequent hypoglycaemic attacks; over a 2-month period her weight increased by 12 kg. These attacks persisted until her insulin was reduced to 36 units per day, the dosage preceding treatment with rifampicin. She remains well, with no evidence of recurrence, 9 months after discontinuation of her anti-tuberculous chemotherapy, and her weight i s returning to its pretreatment level. The temporal relationship between the increase in her insulin requirement on initiating treatment, and the sudden decrease on stopping treatment, suggest that her anti-tuberculous chemotherapy was responsible for the change in her diabetes control. Infection alone may affect diabetes control, and the changes of insulin dosage reported here could be accredited to the underlying disease of pulmonary tuberculosis. However, following the increase in her insulin dose, she was well throughout the remainder of her anti-tuberculous treatment. Her insulin requirement decreased sharply on stopping treatment, at a time when her tuberculosis would be considered to be cured. Therefore, it i s unlikely that tuberculous infection per se was the cause of her relative insulin-resistant state. Rifampicin is a potent hepatic enzymeinducing agent that may accelerate the metabolism of oral hypoglycaemic agents in patients with Type 2 d iabe te~ .~ However, we have been unable to find any reports of rifampicin alone affecting glycaemic control in patients with Type 1 diabetes . Rifampicin has been shown to cause early hyperglycaemia in non-diabetic patients with and without pulmonary tuberculosis, associated with enhanced insulin and C-peptide ~ecre t ion .~ No effects were seen with isoniazid, ethambutol, streptomycin, and/or para aminosalicylic acid on blood glucose, serum insulin or Cpeptide level^.^ This would suggest that the isoniazid in Rifinah taken by the patient was not responsible for the change in her insulin requirements. It i s unlikely that the pyrazinamide was responsible as there was no decrease in her insulin requirement when this was discontinued (due to nausea was still taking that the effect alone. It has been may augment and vomiting) while she rifampicin. This suggests was due to rifampicin