By Betty Tran, MD, MSc

Associate Professor of Medicine, Division of Pulmonary and Critical Care Medicine, Northwestern University Feinberg School of Medicine, Chicago

SYNOPSIS: Hypertonic saline given via rapid intermittent bolus therapy was as effective and safe as slow continuous infusion, and was associated with a lower rate of recorrecting treatment and higher efficacy in achieving goal sodium within one hour.

SOURCE: Baek SH, Jo YH, Ahn S, et al. Risk of overcorrection in rapid intermittent bolus vs slow continuous infusion therapies of hypertonic saline for patients with symptomatic hyponatremia. The SALSA Randomized Clinical Trial. JAMA Intern Med 2020;181:81-92.

Limited data are available regarding the best method to correct symptomatic hyponatremia. Baek et al sought to compare the efficacy and safety of administering 3% hypertonic saline via rapid intermittent bolus (RIB) vs. slow continuous infusion (SCI) in patients with moderate to severe symptomatic hyponatremia. SALSA was a multicenter, open-label, randomized clinical trial that enrolled adults with moderate to severe symptoms with serum sodium (sNa) levels < 125 mmol/L. Patients were enrolled through the ED and on the wards. Randomization was stratified by center and hyponatremia symptom severity. Patients in the RIB group received IV 2 mL/kg 3% saline for 20 minutes for moderate symptoms or 4 mL/kg 3% saline over 40 minutes for severe symptoms in the first 24 hours. After the initial treatment, repeat infusion of 2 mL/kg 3% saline over 20 minutes was administered every six hours up until 24 hours until the sNa level increased by 5 mmol/L to 9 mmol/L from baseline and until symptom relief. The same protocol was repeated after 24 hours until 48 hours until sNa increased by 10 mmol/L to 17 mmol/L from baseline or until sNa reached 130 mmol/L and until symptoms improved.

Patients in the SCI group received IV 3% saline at a starting rate of 0.5 mL/kg/hour for moderate symptoms or 1 mL/kg/hour for severe symptoms. The rate was adjusted every six hours up to the first 24 hours. The treatment discontinued if sNa increased by 5 mmol/L to 9 mmol/L with symptom relief. Treatment increased 0.25 mL/kg/hour if the rate of correction was less than 0.5 mmol/hour (or resumed at 0.5 mL/kg/hour if previously discontinued), and maintained if the rate of correction > 0.5 mmol/hour. The infusion rate also was adjusted every six hours depending on sNa level between 24 and 48 hours. It was discontinued if sNa increased by 10 mmol/L to 17 mmol/L or if it reached 130 mmol/L with symptom relief, it increased by 0.25 mL/kg/hour (or resumed at 0.25 mL/kg/hour if discontinued previously) if sNa increased by less than 1.5 mmol/six hours, and maintained if sNa increased by more than 1.5 mmol/six hours.

Given the low incidence of osmotic demyelination syndrome (ODS), the primary outcome chosen to be a surrogate marker of ODS was the incidence of overcorrection, defined as an increase in sNa by more than 12 mmol/L within the first 24 hours or an increase in sNa by more than 18 mmol/L within 48 hours. Multiple secondary outcomes were reported, centering around efficacy and safety. The authors checked overcorrection at every sample time point. They enacted relowering treatment if overcorrection was present: a dextrose 5% infusion of 10 mL/kg over one hour and/or an IV desmopressin of 2 mcg if sNa increase was more than 10 mmol/L within the first 24 hours or more than 18 mmol/L within 48 hours.

The authors performed statistical analyses based on both intention-to-treat (ITT) and per protocol (PP) bases, since the dropout rate was expected to be high given the complexity of the hypertonic saline infusion protocol. Overall, a total of 178 patients with a mean sNa of 118.2 mmol/L (standard deviation, 5.0) were randomized: 87 to the RIB group and 91 to the SCI group, with 72 and 73 patients, respectively, completing the study and included in the PP analysis. The causes of hyponatremia included thiazide diuretics (29.8%), syndrome of inappropriate antidiuretic hormone secretion (29.2%), adrenal insufficiency (16.3%), nonrenal sodium loss (14.0%), and increased extracellular fluid volume (10.7%). There was no significant difference in baseline characteristics between the two groups. For both the ITT and PP analyses, there was no significant difference in the primary outcome of overcorrection incidence: 17.2% in RIB vs. 24.2% in SCI group (absolute risk difference, -6.9%; 95% CI, -18.8% to 4.9%; P = 0.26 for the ITT analysis). In terms of secondary outcomes, the RIB group showed a decreased incidence of needing relowering treatment than the SCI group (41.4% vs. 57.1%, respectively, absolute risk difference, -15.8%; 95% CI, -30.3% to -1.3%; P = 0.04; number needed to treat [NNT], 6.3). Additionally, in terms of a post-hoc ITT analysis, the proportion of patients achieving target correction rate (sNa of 5 mmol/L to 9 mmol/L) within one hour was higher in the RIB group (32.2% vs. 17.6%, respectively; absolute risk difference, 14.6%; 95% CI, 2-27.2%; P = 0.02; NNT, 6.8). Overall, the two groups did not differ in terms of efficacy in increasing sNa after six hours or improving symptoms.


SALSA is the first prospective, multicenter, randomized clinical trial comparing the efficacy and safety of RIB vs. SCI of hypertonic saline in patients with moderate to severe symptomatic hyponatremia. Although both methods were similar in terms of overcorrection risk and cumulative amounts of hypertonic saline administered in 48 hours between the two groups, the trial results suggest that, not surprisingly, RIB achieved the sNa goal within one hour and with an overall reduced incidence of needing a relowering intervention for overcorrection.

Notably, Baek et al did not adjust for secondary and post-hoc outcomes, so false-positives are possible. In addition, although their reasoning in using overcorrection incidence as a surrogate marker is well-delineated, the true outcome of interest when correcting hyponatremia, ODS, was not observed in either group. As such, we cannot define which method is preferable in resolving life-threatening hyponatremic symptoms in a way to prevent ODS.

Regardless, the administration of hypertonic saline to correct symptomatic hyponatremia via RIB appears just as safe as SCI, with the possible advantages of faster efficacy and less need for relowering interventions. In addition, RIB is more user-friendly in that it does not require frequent calculations as seen with infusion rate adjustments. Finally, small, fixed boluses to correct hyponatremia are in keeping with prior American and European guidelines, which were based on smaller randomized trials, case reports, and expert opinions.1,2 


  1. Spasovski G, Vanholder R, Allolio B, et al. Clinical practice guideline on diagnosis and treatment of hyponatraemia. Eur J Endocrinol 2014;170:G1-G47.
  2. Verbalis JG, Goldsmith SR, Greenberg A, et al. Diagnosis, evaluation, and treatment of hyponatremia: Expert panel recommendations. Am J Med 2013;126(10 Suppl 1):S1-S42.