Editor’s Note: Check out Medscape’s for the latest COVID-19 news and guides Coronavirus Resource Center.
Both high and low serum sodium levels have been linked to adverse outcomes for hospitalized patients with COVID-19, new research shows.
In the retrospective study of 488 patients hospitalized in one of two London hospitals with COVID-19 between February and May 2020, hypernatremia (defined as serum sodium levels> 145 mmol / L) was associated with a threefold increase associated with inpatient mortality.
Hyponatremia (Serum sodium level <135 mmol / L) was associated with twice the likelihood of requiring advanced ventilation assistance. In-hospital mortality was also increased in patients with hypovolemic hyponatremia.
“Serum sodium levels could be used in clinical practice to identify patients with COVID-19 at high risk for poor outcomes who would benefit from more intensive monitoring and sensible rehydration,” say Dr. med. Ploutarchos Tzoulis and colleagues was in their article published online on February 24th in The Journal of Clinical Endocrinology and Metabolism.
The results will be presented at the upcoming virtual annual meeting of the Endocrine Society (ENDO 2021).
Should sodium be included in a risk calculator for COVID-19?
Tzoulis, professor of endocrinology at University College London Medical School in London, UK, said Medscape Medical News, “Sodium could be in risk calculators for other routine biomarkers like white blood cell count, lymphocytes, and CRP [C-reactive protein]to provide a tool for dynamic risk stratification throughout the clinical course of COVID-19 and to support clinical decision-making. “
In addition, he said, “We should adopt less conservative strategies regarding the rate and amount of fluid resuscitation to prevent hypernatremia, which is caused by negative fluid balance and which can often be iatrogenic.”
Steven Q. Simpson, a professor of medicine in the Department of Lung, Intensive Care, and Sleep Medicine at the University of Kansas, Kansas City, said the article lacks key findings that would be helpful in interpreting the results.
“Data on diuretic use and fluid sparing are not included in the publication. It is simply not possible to say whether serum sodium is a ‘predictor’ … or whether it is a side effect of other problems or actions taken by doctors at Patients who are getting on poorly.
“To say that sodium needs to be put on a risk calculator subtly means that there is a causal relationship with mortality, and that clearly has not been established,” pointed out Simpson, who is president of the American College of Chest Physicians, did not speak for the organisation.
He added, “The data is interesting but not actionable. In critical care medicine, it is common practice to adjust water and salt intake to keep serum sodium in the normal range so that the paper really doesn’t change behavior.”
Tzoulis told Medscape Medical News Although our acute care physicians and intensive care physicians at both study sites do not have electronic medical records on diuretic use or fluid intake and output, they firmly believe that they have not routinely used diuretics in COVID-19 patients. Diuretics were used sparingly. In our cohort, the incidence of pulmonary edema was reported to be less than 5%. “
Regarding the volume of fluid intake, Tzoulis noted, “At our hospital locations, the strategy was to carefully resuscitate the fluid. In fact, the amount of fluid administered was deliberately indicated by our doctors and critical care physicians to be much more conservative than that indicated common in patients with community-acquired pneumonia Risk of respiratory failure. ‘””
Hyper- and Hyponatremia Associated with Adverse COVID-19 Outcomes
In the study, of the 488 patients presented at the hospital, 5.3% had hypernatremia and 24.6% had hyponatremia. Notably, only 19% of patients with hyponatremia underwent a laboratory test to determine the etiology. Of these, three quarters had hypovolemic hyponatremia, which was determined based on a urinary sodium limit of 30 mmol / L.
The overall in-hospital mortality was 31.1%. There was a strong, though not significant, trend in higher mortality related to sodium status at ingestion. The mortality rates were 28.4%, 30.8% and 46.1% for those who were normonatraemic, hyponatraemic and hypernatraemic, respectively (P. = 0.07). The sodium levels in the basic serum did not differ between survivors (137 mmol / l) and non-survivors (138 mmol / l).
In the multivariable analysis, the occurrence of hypernatremia at any point in time during the first 5 days in the hospital was one of the three independent risk factors for higher mortality in the hospital (adjusted hazard ratio, 2.74; P. = 0.02). The other risk factors were older age and a higher CRP value.
Overall, hyponatremia was not associated with death (P. = .41).
During the hospital stay, 37.9% of the patients remained normonatremic; 36.9% had hyponatremia; 10.9% had hypernatremia; and 14.3% had both conditions at some point during their stay.
In-hospital mortality was 21% in patients with normonatremia compared with 56.6% in patients with hypernatremia (odds ratio) [OR]3.05; P. = 0.0038) and 45.7% for those with both (OR, 2.25; P. <.0001).
The death rate of 28.3% in the total group with hyponatremia did not differ significantly from 21.1% in the normonatremic group (OR, 1.34; P. = 0.16). However, the mortality rate in the subgroup that developed hypovolemic hyponatremia was 40.9%, which was significantly higher than in the normonatremic group (OR, 2.59, P. = 0.0017).
The incidence of hyponatremia decreased 5 days later from 24.6% at admission to 14.1%, while the incidence of hypernatremia increased from 5.3% to 13.8%.
Key Finding: Relationship Between Hospital Acquired Hypernatremia and Death
“The main new finding of our study was that hospital-acquired hypernatremia, rather than admission hypernatremia, was a predictor of hospital mortality, with the worst prognosis reported in patients with the greatest increase in serum sodium in the first 5 days Hospitalization, “note Tzoulis and colleagues.
Hypernatremia was present in 29.6% of the non-survivors compared with 5.2% in the survivors.
Of 120 patients with hyponatremia on admission, 31.7% received advanced respiratory support, compared with 17.5% and 7.7% of patients with normonatremia and hypernatremia, respectively (OR, 2.18; P. = 0.0011).
In contrast, there was no difference in the proportions requiring ventilation support between those with hypernatremia and those with normonatremia (16.7% versus 12.4%; OR 1.44; P. = 0.39).
Acute kidney injury occurred in 181 patients (37.1%). It was at no time associated with serum sodium concentration.
Tzoulis and Simpson have not disclosed any relevant financial relationships.
J Clin Endocrinol Metab. Published online February 24, 2021. abstract