 |
 |
 |
 |
 |
 |
 |
|
|
|
|||||||||
|
|||||||||||
Psychiatrist Department of Psychiatry National Cheng Kung University Hospital College of Medicine National Cheng Kung University 138 Sheng Li Road, Tainan, Taiwan FAX 886-6-2759259 E-mail chenps{at}mail.ncku.edu.tw
Psychiatrist and Lecturer Department of Psychiatry National Cheng Kung University Hospital College of Medicine National Cheng Kung University
Psychiatrist and Lecturer Department of Psychiatry National Cheng Kung University Hospital College of Medicine National Cheng Kung University
Pharmacist Institute of Clinical Pharmacy College of Medicine National Cheng Kung University
Resident Department of Psychiatry National Cheng Kung University Hospital College of Medicine National Cheng Kung University
Case Report. A 33-year-old Taiwanese man had a 15-year history of bipolar I disorder. The patient had been taking valproic acid 1000 mg/d, lithium 600 mg/d, and haloperidol 20 mg/d. Olanzapine was prescribed at an initial dosage of 5 mg/d, plus crosstitration of haloperidol, for his poor response.
The patient's medical records showed no history of elevated serum glucose concentrations, and no glucose in his urine. No family history of diabetes was present. His body mass index was 26.4 kg/m2.
Five days after the olanzapine cotreatment, the patient was brought to an emergency department. He reported malaise, polyuria, and polydipsia. There was no abdominal tenderness, nausea, vomiting, or fever. Laboratory analysis revealed: glucose 1121 mg/dL (normal 120–180), osmolality 324 mmol/kg (280–295), and trace of ketonemia; valproic acid 63.4 µg/dL and lithium 0.36 mEq/L. Urinalysis was positive for glucose. No significant findings were noted with other laboratory parameters. No other medications were taken for reasons other than bipolar disorder. The patient was rapidly stabilized through rehydration and an insulin infusion after admission. A 1650-kcal diabetic diet and a regimen of three 10-units regular insulin injections per day were started. Glycosylated hemoglobin was 13.6%. Olanzapine and lithium were then discontinued. Seven days later, the insulin injections were stopped and oral antidiabetic agents (metformin 1 g and glibenclamide 10 mg 3 times/d) were prescribed. The patient was discharged on haloperidol and valproic acid. He remained metabolically stable.
Discussion. This case is similar to that reported by Peterson and Byrd,2 in which hyperglycemia occurred early (5 wk) in the course of clozapine treatment. Olanzapine is an antipsychotic drug similar to clozapine. Mechanisms postulated to account for clozapine-induced hyperglycemia include primary damage to the pancreatic islet cells, sympathetic nervous system dysregulation, and a secondary phenomenon related to weight gain and insulin resistance.3 The same putative mechanisms may also apply to olanzapine. Both clozapine and olanzapine may exacerbate subclinical diabetes or promote glucose metabolic abnormalities.
In fact, concomitant medication, such as valproic acid and lithium, can also increase insulin resistance.4 Recent studies suggest that lithium and valproic acid inhibit a serine/threonine kinase involved in the insulin-signaling pathway.5 However, it has not been documented that either valproic acid, lithium, or both, might cause nonketotic hyperosmolar syndrome.
Use of the Naranjo probability scale indicated a possible relationship between the nonketotic hyperosmolar syndrome and the olanzapine, lithium, and valproic acid combination in this patient.6 Clinicians should be alert for symptoms of increased blood glucose in patients taking olanzapine or clozapine, particularly when used in combination with lithium and/or valproic acid. Therefore, blood glucose levels should be monitored.
Footnotes
Letters are subject to review prior to acceptance. They should address areas related to pharmacy practice, research, or education, or articles recently published. Corrections of previously published material also are accepted. Letters are limited to no more than five authors. In cases where adverse drug effects are described, the Naranjo ADR probability scale should be used to determine the likelihood that the adverse effect was drug-related (Clin Pharmacol Ther 1981;30:239-45). Text: limit 500 words. References: limit 5. Art: limit 1 table or figure.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
 |
 |
 |
 |
