Abstract
Helen Smith, J. Kenney-Herbert and L. Knowles, Reaside Clinic, Birmingham, United Kingdom:
We would like to report a case of diabetic ketoacidosis in a 40-year-old Afro-Caribbean man with a diagnosis of paranoid schizophrenia after commencing clozapine therapy. He was admitted to a Regional Secure Unit in August 1996 under Section 3 of the Mental Health Act 1983. He was born in Jamaica. He came to the United Kingdom as a child and has a 20-year history of mental illness. When detained, he was actively psychotic and was restarted on antipsychotic medication (flupenthixol decanoate 400 mg weekly and chlorpromazine 200 mg four times a day). This led to some improvement in his mental state. However, it remained unstable, and attempts to move the patient to less supportive and more stimulating environments resulted in a marked deterioration in his condition on a number of occasions. In view of the failure of conventional medication, a trial of clozapine was discussed with the patient. No contraindications were noted and he was started on a standard treatment regime. This had to be modified to a slower increase as the patient was experiencing excessive sedation. He also complained of nocturnal incontinence and other symptoms consistent with postural hypertension. These failed to settle and by day 16 he was confused and smelt strongly of ketones. His blood glucose was 55 mmol/L, sodium 131 mmol/L, potassium 7 mmol/L. He was diagnosed as having diabetic ketoacidosis and transferred to a general hospital for treatment. His ketoacidosis settled uneventfully on a reducing regime of insulin and he was transferred back to a psychiatric unit requiring 30 units twice daily of human mixtard.
Over the following 4 months, his insulin requirements decreased, although he has continued to require maintenance insulin therapy in the community.
The patient involved had few risk factors for the development of diabetes. He was mildly obese but had no family history of diabetes and no previous evidence of glucose intolerance. His symptoms were initially in keeping with the common recognised side effects of clozapine. It was not until he was ketotic that it became evident that more serious pathology was present.
We found reference to eight similar cases in the literature, four associated with diabetic ketoacidosis [1, 3, 4, 5] and four with hyperglycaemia [2, 3]. In cases associated with the former, three involved males and one female, all had a family history of either Type I or Type II insulin-dependent diabetes and one individual had an impaired glucose tolerance before starting clozapine. Two individuals were already taking lithium and had been doing so for some time. A third was taking risperidone concurrently. Only one individual took clozapine and nothing else. The time of onset of the problem was only identified in four cases and these were 5 and 6 weeks after commencing clozapine therapy. Two patients continued to require insulin after clozapine had been withdrawn.
Of the four cases of hyperglycaemia, all involved males and although none had a family history of insulin-dependent diabetes, two of the individuals themselves had impaired glucose tolerance before starting clozapine. Four of the individuals were black and one white and all were taking other drugs concurrently. These included lithium, rantidine, benzhexol and hydrochlorothiazide, although these had been taken by the individuals for some time prior to the introduction of clozapine with no problems. Only two reports commented on continuing insulin requirements after stopping clozapine, one individual requiring long-term therapy and the other requiring no treatment at all. A further case continued on clozapine despite requiring insulin to control concurrent hyperglycaemia. For these cases, the average onset of symptoms was between 2 weeks and 4 months.
These findings would suggest that clozapine may cause markedly impaired glucose tolerance in some individuals. Although in the case studies described above many patients took other drugs, some of which are known to impair glucose tolerance in their own right, there is at least one example of clozapine causing these problems alone. We would recommend that patients starting on this treatment have their glucose measured prior to the beginning and during the first 6 weeks of therapy. Problems with impaired glucose tolerance may be rare but are potentially life threatening. Simple testing would not only identify this problem quickly but would also be more cost effective than patients being transferred and treated in general hospital settings.