Disturbed glucose metabolism among patients taking olanzapine and typical antipsychotics

Concern about the potentially diabetogenic effect of antipsychotics has existed since the introduction of chlorpromazine in the 1950s. However, there is controversy as to whether the newer "atypical" antipsychotics are more diabetogenic than older "typical" agents. We undertook a retrospective cohort analysis of 484 patients receiving outpatient prescriptions for either olanzapine (n=312) or one of four typical antipsychotics (n=172) at the Atlanta Veterans Affairs Medical Center from October 1, 1996 (when olanzapine first became available at our institution) through December 31, 2000. The four typical antipsychotics were haloperidol, fluphenazine, chlorpromazine, and perphenazine.Random plasma glucose values collected over the 4.25-year period were classified as to whether they had been drawn before, during, or after exposure to an antipsychotic. Primary outcome measures were the development of a random plasma glucose > or =160 mg/dl or > or 200 mg/dl while on an antipsychotic. Significant differences were found in the percentages of antipsychotic-treated patients developing a random plasma glucose > or =160 mg/dl (olanzapine 12.5% versus typical antipsychotics 5.2%, P<.01, adjusted odds ratio 3.6) and > or =200 mg/dl (olanzapine 5.4% versus typical antipsychotics 1.7%, P=.02, adjusted odds ratio 5.1). For patients younger than 60 years of age, 10.5% of those taking olanzapine developed a random plasma glucose > or =160 mg/dl compared with 0% of those taking typical antipsychotics, P=.0003. In conclusion, we found that among individuals without a plasma glucose value > or =160 mg/dl at baseline, use of olanzapine was associated with an increased risk of elevation in plasma glucose compared with use of typical antipsychotics.