Endocrine Abstracts

Introduction: Admission hyponatremia, frequently detected in patients hospitalized for COVID-19, has been associated with increased mortality. However, although SIADH-induced euvolemic hyponatremia is the single most-common cause of hyponatremia in community-acquired pneumonia, repercussions of admission-hyponatremia volemic classification on COVID-19 hospitalizations have yet to be described. We sought to identify factors contributing to mortality and hospital length-of-stay (LOS) in hospitalized patients admitted with hyponatremia, taking volemia into account.

Method: Retrospective study of 247 patients admitted with COVID-19 to a tertiary hospital in Madrid, Spain from March 1st through March 30th, 2020, with a glycemia-corrected serum sodium level (SNa) < 135 mmol/l. SNa, creatinine, urea, glycemia, O2 Saturation were recorded at admission, at day 2nd–3rd of hospitalization, and ensuing days when hyponatremia persisted (>3 days). Volemia was determined according to: maximum height of internal jugular vein pulse (HIYP), and 2 of following: presence/absence of thirst, orthostatic symptoms, blood pressure ≤ 90/60 mmHg, heart rate ≥90 bpm, urinary sodium ≤ 30 mmol/l and/or a rise in serum creatinine (SC) and/or serum urea (SU) accompanying the descent in SNa. In absence of registered HIJP, 3 of the latter criteria were used. Variables studied included demographics, comorbidities, therapy of hyponatremia, LOS and in-hospital mortality rate.

Results: Age: 68 years [56–81]; 99/247 (39.9%) female. Median admission SNa: 133 mmol/l [131–134]. In the majority (188/247:76%), hyponatremia was mild (SNa 131–134 mmol/l). Volemia in 208/247 patients: 119/208 (57.2%) were euvolemic, and 89/208 (42,8%) were hypovolemic. The median LOS was 11 days [6–18], with a mortality rate of 21.5%. LOS was correlated with total hospitalization days with hyponatremia (r=0.474, P <0.001), and negatively correlated with admission SNa (r=−0.224, P <0.001). Patients who died had a higher median rate of hospitalization days with hyponatremia (25% vs. 17%, P=0.007) than survivors, with a shorter LOS: 7 [4–12] vs. 12 [7–20] days, P <0.001. Higher 02 saturation levels were independently associated with reduced mortality (OR: 0.88, 95%CI 0.78 to 0.99), and an oncological history with increased mortality (OR: 13.17, 95%CI 1.79 to 96.84). Neither the degree of hyponatremia, nor volemia were independently associated with mortality. However, initial therapy of hyponatremia was: euvolemic patients inadequately treated at admission with iv isotonic saline showed an increased mortality rate (36.6% vs 20.2% P=0.02).