Olanzapine-Induced Hyponatremia in a Patient with Autism

To The Editor:

Neither olanzapine (Dudeja et al. 2010 ) nor autism (McNally et al. 1988; Hiratani et al. 1997) has been frequently related to hyponatremia in the medical literature. In addition, olanzapine was thought to have a neutral (Goldman and Hussain 2004; Jessani et al. 2006) or even a beneficial effect (Littrell et al. 1997; Kruse et al. 2001; Tényi and Vörös 2006; Phull and Davies 2011) on water and sodium homeostasis in polydipsic patients with schizophrenia. Here, we describe a polydipsic patient with autism who developed hyponatremia during olanzapine treatment, and discuss the possible causative role of olanzapine in this case.

Case Report

A 22-year-old man, with autism and severe mental retardation, had been polydipsic since his childhood. His amount of water consumption had not been formally evaluated, but it had never posed a problem to his caregivers in the past. In the previous 2 years, the patient had been taking olanzapine 5–10 mg and valproate 500 mg every day to control his agitation and occasional self-harming behaviors. There had been no noticeable change in his drinking habits, nor had he shown any signs of dry mouth during olanzapine treatment. Serum glucose, liver enzymes, and creatinine levels measured during this period were all within normal ranges.

We did not find the patient's severe hyponatremia (114 mmol/L), hypokalemia (3.0 mmol/L), hypocalcemia (free calcium 3.89 mg/dL) and hypomagnesemia (1.5 mg/dL), until he presented with generalized tonic–clonic seizure. Serum levels of blood urea nitrogen and creatinine were normal. Urine osmolarity and sodium concentration were not inappropriately elevated (88 mOs/kg and 29 mmol/L, respectively), suggesting that water intoxication caused by primary polydipsia was the principal cause of his multiple electrolyte imbalances. Other possible etiologies of polydipsia, such as diabetes mellitus, diabetes insipidus, syndrome of inappropriate antidiuretic hormone secretion (SIADH), and thyroid or adrenal dysfunction, were excluded. A fasting serum glucose level (115 mg/dL) suggested the presence of pre-diabetes. Nevertheless, the patient did not have any glucosuria. In addition, a computed tomography scan of the brain did not show any intracranial lesion that could explain his polydipsia.

Olanzapine was discontinued because it was a possible cause of hyponatremia, and was replaced by quetiapine 400 mg/day. Valproate was continued. In the following 6 months, the patient received the usual care from his family and caregivers. Polydipsia partially improved, and the patient's serum sodium levels returned to ∼140 mmol/L. He no longer had seizures.

Discussion

Olanzapine-induced hyponatremia is probably more frequent than has been thought. For example, in the World Health Organization (WHO) global individual case safety report database system, olanzapine was the second most frequently reported antipsychotic associated with hyponatremia (Mannesse et al. 2010).In this case, the initiation and resolution of hyponatremia were clearly related to olanzapine exposure status. In addition, the delayed onset of hyponatremia after olanzapine use was similar to that described in a previous report (Dudeja et al. 2010). Taken together, these facts pointed to olanzapine being a “possible” cause of hyponatremia in this case, with a Naranjo score of 4 (Naranjo et al. 1981).

It is likely that olanzapine causes hyponatremia by stimulating the inappropriate release of vasopressin, a mechanism supported by both animal study (Kiss et al. 2010) and human report (Dudeja et al. 2010). Even at low concentrations, inappropriately elevated serum vasopressin can greatly decrease water clearance and worsen hyponatremia (Goldman 2009). In this case, a hint of inappropriate serum vasopressin elevation was the urine osmolarity (88 mOs/kg), which did not reach the maximal diluting capacity of the kidney (∼ 40 mOsm/kg [Goldman 2009]) during severe hyponatremia. However, this patient's urine osmolarity was not high enough (typically ≥100 mOsm/kg) to meet criteria for a formal diagnosis of SIADH.

Autism does not cause elevation of serum vasopressin (Miller et al. 2013). Nevertheless, the stereotype behaviors, a core symptom domain of autism, may have contributed to this patient's polydipsia and predisposed him to the risk of developing hyponatremia. It was, however, difficult to explain the patient's hyponatremia solely by polydipsia, as polydipsia seldom causes hyponatremia unless the water consumption is massive enough (Gillum and Linas 1984). In this patient, it is most likely that olanzapine induced hyponatremia by worsening the patient's preexisting polydipsia and/or by stimulating a sub-threshold increase in serum vasopressin levels.

In conclusion, we suggest that physicians should be alert to olanzapine's possible adverse effects on water and sodium homeostasis, especially in autism patients with stereotypical polydipsia.