By Elaine Chen, MD

Assistant Professor, Department of Internal Medicine, Division of Pulmonary and Critical Care Medicine, Section of Palliative Medicine, Rush University Medical Center, Chicago

Dr. Chen reports no financial relationships relevant to this field of study.

SYNOPSIS: Using balanced crystalloids rather than normal saline for intravenous fluid administration in critically ill adults leads to statistically significant lower rates of major adverse kidney events, including death from any cause, new renal replacement therapy, and persistent renal dysfunction, compared to normal saline in critically ill adults. Clinical judgment should be applied when selecting fluid.

SOURCE: Semler MW, Self WH, Wanderer JP, et al. Balanced crystalloids versus saline in critically ill adults. N Engl J Med 2018;378:829-839.

Intravenous crystalloid solutions are used commonly for fluid resuscitation in the ICU. Normal saline is used most often, but can be associated with hyperchloremic metabolic acidosis or acute kidney injury. In this study, Isotonic Solutions and Major Adverse Renal Events Trial (SMART), the authors hypothesized that use of “balanced” crystalloids would result in lower overall incidence of death, new renal replacement therapy, and persistent renal dysfunction compared to saline.

The study was a single-center, pragmatic, unblinded, cluster-randomized, multiple-crossover trial that took place from June 2015 through April 2017. All adults admitted to a participating ICU during the study time frame were included. Each ICU was randomized, by month, to use either saline or balanced crystalloid (lactated Ringer’s or PlasmaLyte, at the clinician’s preference) whenever intravenous fluids were administered. Fluid therapy administered in the ED and operating room were coordinated with the admitting ICU when possible.

Over the course of 23 months, 15,802 patients from five ICUs were enrolled in the trial. Baseline characteristics were similar between the groups. Approximately 5% of patients in each group received any volume of unassigned crystalloid due to crossing over of months. In the saline group, there was a higher incidence of measured plasma chloride > 110 mmol/L (23.5% vs. 35.6%; P < 0.001) and plasma bicarbonate < 20 mmol/L (35.2% vs. 42.1%; P < 0.001), with greater differences among those patients receiving larger fluid volumes.

The primary outcome, the presence of a major adverse kidney event within 30 days, was the composite of death, new renal replacement therapy, and persistent renal dysfunction (defined as final inpatient creatinine value ≥ 200% of baseline value). This outcome was noted in 1,139 patients in the balanced crystalloids group and 1,211 patients in the saline group, with a marginal odds ratio (OR) of 0.91 (95% confidence interval [CI], 0.84-0.99; P = 0.04). The difference in rate of primary outcome between groups was greater among patients who received larger fluid volumes.

Among patients with sepsis, 30-day in-hospital mortality was 25.2% in the balanced crystalloid group and 29.4% in the saline group (adjusted OR, 0.80; 95% CI, 0.67-0.97; P = 0.02). Regarding secondary outcomes, overall 30-day mortality was observed in 818 patients in the balanced crystalloid group compared with 875 patients in the saline group (P = 0.06). New renal replacement therapy occurred in 189 patients in the balanced crystalloid group and 220 patients in the saline group (P = 0.08). Renal dysfunction during hospitalization did not differ between groups.

The authors proposed that while the absolute differences are modest, if the findings were extrapolated to the 5 million patients admitted to ICUs each year, the reduction in death, new renal replacement therapy, or persistent renal dysfunction could be substantial. Strengths of this trial included large sample size, trial design that allowed early delivery of assigned fluid, and minimal selection bias. Limitations included the single-center setting and the unblinded nature of the study. PlasmaLyte and lactated Ringer’s were studied together, so differences between the two types of balanced crystalloids cannot be evaluated in this study.


The authors of SMART reported that the use of balanced crystalloids for intravenous fluid administration resulted in a lower rate of major adverse kidney events, the composite outcome of death from any cause, new renal replacement therapy, or persistent renal dysfunction than the use of saline. While this was statistically significant, none of the three individual outcomes reached statistical significance. Additionally, the differences were modest, with an absolute difference of 1.1% in the composite measure.

Interestingly, the authors of the SALT-ED trial1 (the results of which were published in the same issue of The New England Journal of Medicine as SMART) studied the same fluids in a non-critically ill population. Those authors reported no significant difference in the primary outcome of hospital-free days to day 28, but, statistically, observed a similar outcome regarding major adverse kidney events within 30 days (a secondary outcome), with balanced crystalloids resulting in a lower incidence of major adverse kidney events within 30 days than saline (4.7% vs. 5.6%, adjusted OR, 0.82; 95% CI, 0.70-0.95, P = 0.01).

Caution is recommended in interpreting the results of this study. After albumin and hydroxyethyl starch were shown to produce adverse patient-centered outcomes, saline has been the crystalloid of choice for resuscitation despite a lack of evidence for its safety and efficacy. Large-volume saline administration is associated with hyperchloremic metabolic acidosis and acute kidney injury. Prior observational trials suggested lower rates of acute kidney injury and lower mortality when balanced solutions were used. However, none of the current commercially available balanced crystalloid solutions, which substitute anions such as lactate, acetate, gluconate, or bicarbonate in place of chloride to partially buffer or balance the solution, are truly “balanced”; they are all relatively hypotonic and are associated with generation of a metabolic alkalosis.2

Should this study inform a change in practice? In my own practice, I had already selected balanced crystalloids frequently prior to these trial results. Several years ago, our hospital pharmacy encouraged us to avoid PlasmaLyte in favor of normal saline or lactated Ringer’s because of cost, but now the cost differential is less substantial. When teaching my trainees, I will encourage them to consider a balanced crystalloid in the ICU rather than saline, which seems to be the default. We should be thoughtful about our choice of fluid rather than automatically reaching for what is most comfortable and familiar. Each patient in whom fluid administration is considered warrants this thought: Is fluid administration necessary? If so, which fluid will provide the most benefit and the least harm?


  1. Self WH, Semler MW, Wanderer JP, et al. Balanced crystalloids versus saline in noncritically ill adults. N Engl J Med 2018; 378:819-828.
  2. Myburgh J. Patient-centered outcomes and resuscitation fluids. N Engl J Med 2018;378:862-863.