To the Editor: We report the case of a patient with a history of borderline personality disorder who developed polyuria after an olanzapine overdose.

The constellation of polyuria, hyposmolar urine (166 mosmol/kg H₂O), normosmolar plasma, and an increasing level of serum sodium supports the diagnosis of diabetes insipidus in our patient. The rapid resolution of polyuria after administration of a small dose of desmopressin (total of 6 μg in 12 hours) supports the central origin of diabetes insipidus because desmopressin decreases urinary output in central but not in nephrogenic diabetes insipidus (1). Additionally, urine osmolalities lower than 200 mosmol/kg H₂O are uncommonly seen in nephrogenic diabetes insipidus (2). The low normal level of ADH in this patient probably represents early or partial central diabetes insipidus.

Although olanzapine is considered to be a safe agent, there are several reports associating it with the development of hyperglycemia, diabetes mellitus (3), and weight gain (4). To our knowledge, olanzapine has not been associated with the development of central or nephrogenic diabetes insipidus. Although schizophrenia has been reported to be associated with polyuria, diabetes insipidus is usually related to lithium therapy and is then of the nephrogenic type. Of note is that olanzapine is chemically related to clozapine, which has been linked to the occurrence of nephrogenic diabetes insipidus (5). We conclude that diabetes insipidus can possibly occur with high doses of olanzapine.