Sertrallne‐induced Hyponatremia in an Older Patient

To the Editor: We report a case that suggests a causal role of sertraline, a selective serotonin reuptake inhibitor (SSRI) drug, in the development of hyponatremia in an older patient. An 87-year-old man was admitted to the hospital in September 1996 for acute changes in mental status. His caretakers reported that he had seemed agitated earlier, then became unresponsive. His had had a hip fracture and total hip replacement in May. Since then he had experienced weight loss of approximately 40 lbs and, despite entry into a rehabilitation program, had not returned to his baseline function either cognitively or physically. He had been seen by a psychiatrist, and mental status testing suggested dementia and possible depression. Antidepressant treatment with sertraline 50 mg per day was initiated in June. His medical condition was otherwise stable. He had a history of coronary artery disease and had had coronary artery bypass graft surgery about 20 years before his most recent admission. Additional medications included trazodone 50 mg at bedtime (added in late August), enteric aspirin, diclofenac, vitamins, and laxatives. Review of the patient's records showed that a serum sodium level of 125 mmol/L had been recorded during a hospital stay in July for the treatment of lower limb and back pain related to his hip replacement. His serum sodium level had increased to 131 mmol/L with the addition of intravenous .9% normal saline; however, it drifted down to 127 mmol/L when intravenous fluids were discontinued. He had also been admitted in August for an “acute confusional state” that was attributed to administration of griseofulvin for a fungal nail infection. His blood glucose, creatinine, and head computed tomogram (CT) were normal, but he was hyponatremic (130 mmol/L). Griseofulvin was discontinued, and intravenous .9% normal saline was infused. His condition improved, and he was discharged shortly thereafter. On admission in September, the patient was noted to be somnolent and unresponsive, cachectic, and with bilateral Babinski signs. His physical examination was otherwise unremarkable, as was a repeat CT exam, but his serum sodium was 129 mmol/L. Glucose, creatinine, and thyroid studies were normal. An electroencephalogram showed only diffuse slowing. Sertraline and trazodone were discontinued, intravenous .9% normal saline was infused, and the patient's serum sodium level returned to normal. He became responsive and oriented, returned to his baseline status, and was discharged to home. Because of continued lethargy and inattention thought to be prohibiting rehabilitation, sertraline 50 mg per day was restarted by his psychiatrist. The patient became increasingly lethargic. Laboratory tests in the outpatient setting showed a progressive decline in his serum sodium. Approximately a week to 10 days after reinitiation of sertraline, his serum sodium was 122 mmol/L. Sertraline was again discontinued, and his serum sodium level returned to normal over the next 2 days. Hyponatremia was considered an infrequent side effect of SSRI drugs at the time of marketing approval, as reported in the Physicians Desk Reference1 and in pharmacology reviews of these agents.2 However, reports of hyponatremia in older patients receiving SSRI's have appeared, with fluoxetine being the most frequently implicated drug.3,4 Accurate data are limited, but a case-control study of hyponatremia in psychiatric inpatients concluded that older patients seemed to develop hyponatremia more often than younger patients. Age was not found to be a significant independent risk factor by multivariate analysis, but fluoxetine was associated with an increased risk.5 There are two reports of patients who became hyponatremic with sertraline and who were rechallenged.6,7 In one, hyponatremia did not recur with rechallenge.6 In the other, fluids were restricted during the rechallenge.7 Because of significant comorbidities in most reported cases, a causal relationship between SSRIs and hyponatremia remains uncertain.8 The patient in this case did not have diabetes, renal disease, pulmonary disease, or a brain lesion. He was not taking diuretics or other medications associated with hyponatremia. Hyponatremia was detected approximately 1 month after the initiation of sertraline and abated a few days after its discontinuation. This is consistent with previous observations of hyponatremia in patients receiving fluoxetine9 as well as other SSRI's.4 Finally, this patient was rechallenged with sertraline and developed hyponatremia again and again it was resolved with drug discontinuation. These data support a causative role of sertraline in the hyponatremia of this patient. Hyponatremia is found frequently in older hospitalized populations, and the etiology is often multifactorial. Depression leads to altered eating patterns and, in some cases, polydipsia. Aging leads to impairment of free water excretion. Polypharmacy is common in older people and may place such patients at increased risk. Although the exact risk of hyponatremia during SSRI therapy in geriatric populations is still unknown,4,8 sufficient data exist to recommend monitoring serum sodium concentrations during the first weeks to months of therapy with an SSRI.