Prevalence, Risk Factors, and Mortality of Patients Presenting with Moderate and Severe Hyponatremia in Emergency Departments

Abstract

Background. Hyponatremia is among the most common electrolyte disturbances encountered in clinical practice and is associated with a high rate of morbidity and mortality. However, there are very limited data on adult cases presenting to emergency departments with hyponatremia. Objectives. This study aimed to evaluate the frequency, clinical characteristics, and outcomes in hyponatremic patients presenting to emergency departments. Methods. This retrospective study analyzed all patients older than 18 years who visited our institution’s emergency department between October 2018 and October 2019 and has a serum sodium (Na) level <130 mmol/L. Results. Among 24,982 patients who visited the emergency department and had a documented serum sodium level, 284 were included. Patients’ median age was 67.13 ± 14.8 years. Younger patients are less likely to develop severe hyponatremia compared to older patients (adjusted odds ratio (AOR): 0.415; 95% confidence interval (CI): 0.231–0.743; ). Asymptomatic hyponatremia and gastrointestinal manifestations were the most common presenting hyponatremia symptoms (33.7% and 24.2%, respectively). Proton pump inhibitor (PPI) use, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker (ACE/ARB) use, and spironolactone use (OR = 2.6 and 3.9, 2.3 with a , 0.03, and 0.05, respectively) were associated with increased odds of severe hyponatremia. There is no difference in the overall mortality rate within 6 months of presentation between severe and moderate hyponatremia groups (11.1% versus 16.2%, ). Conclusion. Moderate and severe hyponatremia are not uncommon among patients presenting to emergency departments. Moderate hyponatremia can be asymptomatic with clinical significance. Older patients, use of PPI, use of ACEi/ARBs, and spironolactone use were associated with an increased risk of severe hyponatremia compared to moderate. Further prospective analysis of a larger population is needed to confirm our findings.