From Wiki Journal Club
Jump to navigation Jump to search
Self WH, et al. "Balanced crystalloids versus saline in noncritically ill adults". The New England Journal of Medicine. 2018. 378(10):819-828.
PubMedFull textPDF

Clinical Question

Among noncritically ill adult patients in the ED, does fluid management with balanced crystalloid result in earlier hospital discharge compared to isotonic crystalloids?

Bottom Line

Among non-critically ill ED patients, initial fluid resuscitation with balanced crystalloids (Lactated Ringer's or Plasma-Lyte) does not reduce duration of hospitalization when compared to the isotonic crystalloid, normal saline. However, balanced crystalloid use is associated with a reduction in major kidney-related events.

Major Points

The supra-physiological chloride concentration of isotonic crystalloid (normal saline) has been hypothesized to offer inferior outcomes compared to nearer physiologic alternatives, such as balanced crystalloids (Plasma-Lyte, Lactated Ringer’s). Although preclinical models have speculated high chloride content contributing to acidosis, renal injury, and mortality, observational and pilot studies have failed to detect clinically significant differences when comparing these two options.[1][2][3]

Published in 2018, the Saline against Lactated Ringer’s or Plasma-Lyte in the Emergency Department (SALT-ED) Trial aimed to compare initial resuscitation of ED patients with balanced crystalloids vs. isotonic crystalloid in a pragmatic setting. A single center enrolled 13,347 adults in the ED thought to need additional fluids after a 500 mL isotonic crystalloid dose (e.g., NS). All patients enrolled on a given month were given either balanced crystalloids (95% use of Lactated Ringer's) or isotonic crystalloids (NS), with the treatment of choice alternating on subsequent months. The intervention fluid was only continued in the ED, not on the floor.

There were no differences between the groups in amount of hospital free days, the primary endpoint. However, ED administration of balanced crystalloid did result in fewer major adverse renal events as compared with isotonic crystalloid. Further, subgroup analysis suggests greater treatment effect in those with preexisting renal disease (serum creatinine ≥1.5mg/dL) and hyperchloremia (>110mmol/L). The SALT-ED trial results are consistent with its critical-care arm, SMART-MED and SMART-SURG (2018), in suggesting that balanced crystalloids have a role in decreasing renal morbidity.


No guidelines have been published that reflect the results of this trial.


  • Pragmatic, open label, cluster-randomized, multiple-crossover, single center trial
  • Single center, type of crystalloid assigned based on calendar month per unit
  • N=13,347
    • N=6,708 Balanced crystalloids (Lactated Ringer’s/Plasma-Lyte)
    • N=6,639 Isotonic crystalloids (0.9% NS)
  • Setting: Vanderbilt University Medical Center
  • Enrollment: 2016-2017
  • Follow-up period: First event of discharge or 30 days
  • Analysis: Intention-to-treat
  • Primary outcome: Hospital-free days to day 28 (Composite)


Inclusion Criteria

Full details are published elsewhere.[4]

  • Age ≥18 years
  • Received ≥500mL isotonic crystalloids in Emergency Department, subsequently hospitalized outside an ICU

Exclusion Criteria

  • <500mL crystalloid received in emergency department
  • Subsequent hospitalization in an ICU
  • Relative contraindication to randomized therapy at discretion of treating physician
    • Balanced crystalloids: hyperkalemia, brain injury

Baseline Characteristics

From the balanced crystalloid group.

  • Demographics: Age 47 years, female sex 52%, white race 77%, black race 20%, other 3%
  • Median Elixhauser Comorbidity Index score (IQR):[5] 7 (3-14)
  • Admission service: General medicine 71%, cardiology 4%, neurology 2%, general surgery 19%, trauma surgery 4%
  • Creatinine:
    • Baseline: 0.84 mg/dL (mean)
    • In ED: 1.32 mg/dL (mean)
      • KDIGO Stage 2+ AKI by creatinine criteria: 10%
  • ED electrolytes: Na 137 mmol/L, K 4.1 mmol/L, Cl 103 mmol/L, bicarb 22.7 mmol/L, BUN 20 mg/dL
  • Total crystalloid volume:
    • Mean (±SD): 1608 mL ±1095
    • Median (IQR): 1089 (1000-2000)
    • ≥2L: 33%
  • Crystalloid choice: Lactated Ringer's 95%, Plasma-Lyte 5%
  • Median volume of balanced crystalloids in mL (IQR): 1000 (1000-2000) -- Note: 0 in saline group.
  • Median volume of saline in mL(IQR): 0 -- Note: 1000 (1000-2000) in the saline group.
  • Percentage of crystalloid volume consistent with assigned group no.(%)
    • 100%: 84%
    • 51-99%: 8%
    • 1-50%: 4%
    • 0%: 5%


  • Emergency department randomized to either isotonic crystalloid (0.9% sodium chloride) or balanced crystalloids (lactated Ringer’s/Plasma-Lyte), with cross-over on a monthly basis
  • No assignment of fluid selection after transfer to a hospital floor


Comparisons are balanced crystalloids vs. saline. Analyses adjusted for age, sex, race, admitting service, and time.

Primary Outcomes

Median hospital-free days to day 28
25 vs. 25 days (OR 0.98; 95% CI 0.92-1.04; P=0.41)

Secondary Outcomes

Major adverse kidney events within 30 days (composite)
4.7% vs. 5.6% (OR 0.82; 95% CI 0.70-0.95; P=0.01)
Death: 1.4% vs. 1.5% (adjusted OR 0.89)
New renal-replacement therapy: 0.3% vs. 0.5% (adjusted OR 0.56)
Final serum creatinine ≥200% baseline creatinine: 3.8% vs 4.5% (adjusted OR 0.84)
KDIGO Stage 2+ AKI by creatinine criteria
KDIGO stages are listed on page 8 of their clinical practice guidelines.[6]
8.0% vs. 8.6% (OR 0.91; 95% CI 0.80-1.03; P=0.14)
In-hospital death
1.4% vs. 1.6% (OR 0.88; 95% CI 0.66-1.16; P=0.36)

Subgroup Analysis

Figure 2 on page 827 presents three forest plots: One for the primary outcome, one for major adverse kidney events at 30 days, and one for AKI (excluding ESRD patients). The authors note that patients who derived the greatest benefit from balanced crystalloids were those with renal dysfunction in the ED (characterized by creatinine ≥1.5 mg/dL [133 umol/L] or hyperchloremia (Cl >110 mmol/L).


  • Single center, open label study
  • No separate analysis of Lactated Ringer’s vs. Plasma-Lyte
    • Only ~5% use of Plasma-Lyte in the balanced crystalloids arm
  • Composite outcomes inherently not patient-centered as each of the outcomes may not be as equally meaningful to patients
    • Composite endpoints are susceptible to confounding[7]
  • Difference in adverse kidney events is a surrogate outcome, a increase in creatinine and not a clinical patient outcome (death, need for dialysis)
  • EHR data is not of the same level of quality as research data collected by trained observers
  • This doesn't address use of fluids beyond initial ED resuscitation
  • The AKI definition did not incorporate rate of urine output, which is included in the KDIGO definitions


  • Funded by the Vanderbilt Institute for Clinical and Translational Research
  • Primary author supported by grants from National Heart, Lung, Blood Institute

Further Reading