Background: Clozapine has been demonstrated to be superior to typical neuroleptics in reducing refractory symptoms in patients with schizophrenia, but it has also been associated with hyperglycemia and diabetes mellitus. This study was designed to investigate the proportion of undiagnosed impaired fasting glucose and diabetes mellitus in patients prescribed clozapine at 8 Department of Veterans Affairs (VA) medical centers.

Method: All patients diagnosed by the VA in New England with ICD-9 schizophrenia from Oct. 1, 1999, to Sept. 30, 2000, who received a prescription for clozapine were identified, and an attempt was made to obtain a fasting plasma glucose (FPG) test. All patients were also characterized as to whether they were diagnosed as diabetic prior to the screening FPG. Patients not previously diagnosed as diabetic were divided into 2 groups: normal FPG (< 110 mg/dL) and elevated FPG (>= 110 mg/dL). Clinical and sociodemographic characteristics of the 2 groups were compared using chi-square and t tests.

Results: Overall, 121 patients were not previously diagnosed as diabetic and received an FPG. Ninety-three (77%) had a normal FPG, and 28 (23%) had an elevated plasma glucose-including 17% with impaired fasting glucose and 6% with diabetes. Patients with hyperglycemia were significantly older (p = .007) and more commonly codiagnosed with bipolar disorder (p = .04).

Conclusion: Hyperglycemia was common in patients receiving clozapine who had not been previously diagnosed as diabetic. These patients should be considered a group at high risk to develop diabetes mellitus and deserve both close monitoring and early intervention at the first sign of the onset of either diabetes or impaired glucose tolerance.

J Clin Psychiatry 2003;64(5):605-608