One of the mainstays of shock management is fluid resuscitation, and there has been tremendous interest in the choice of intravenous fluids. At present, isotonic crystalloid solutions are favored over colloid solutions. Among isotonic crystalloids, 0.9% “normal” saline is perhaps the most widely prescribed medication in the United States. In clinical practice, alternatives to 0.9% saline are balanced salt solutions that contain a physiologic amount of chloride and lactate or acetate as the base equivalent (Table 1A). Two commonly used balanced salt solutions are PlasmaLyte A and Lactated Ringers. Animal studies as far back as the 1980s1,2 suggest that renal blood flow is reduced with the administration of chloride-rich crystalloid solutions compared to more physiological solutions. A crossover trial in healthy humans3 supports the animal findings, with decreased renal blood flow and cortical tissue perfusion after 0.9% saline administration. However, more recent studies4,5 have been equivocal in demonstrating differential outcomes in using 0.9% saline versus balanced salt solutions.
Table 1:
Properties of 0.9% Saline, Lactated Ringer’s, and Plasma-Lyte Solutions
| 0.9% Saline | Lactated Ringer’s | Plasma-Lyte A | |
|---|---|---|---|
| Sodium (mEq/l) | 154 | 130 | 140 |
| Potassium (mEq/l) | 0 | 4 | 5 |
| Calcium (mEq/l) | 0 | 3 | 0 |
| Magnesium (mEq/l) | 0 | 0 | 3 |
| Chloride (mEq/l) | 154 | 109 | 98 |
| Lactate (mEq/l) | 0 | 28 | 0 |
| Acetate (mEq/L) | 0 | 0 | 27 |
| Gluconate (mEq/l) | 0 | 0 | 23 |
| Osmolarity (mOsm/L) | 308 | 273 | 294 |
| pH | 5.5 | 6.5 | 7.4 |
What do these important studies show?
In two pragmatic randomized clinical trials, Semler et al. (for the SMART Investigators)6 and Self et al. (for the SALT-ED investigators)7 compared outcomes after 0.9% saline or balanced salt administration (Lactated Ringers or PlasmaLyte A) in critically ill and non-critically ill populations, respectively. Given an anticipated small relative effect size, to be adequately powered, the SMART and SALT-ED trials needed to be large (in fact, the SMART trial was 15,802 patients, and to the best of our knowledge, the largest clinical trial conducted in the ICU setting outside of cardiology and stroke).
The SMART trial originally planned to enroll 8,000 patients during 60 unit-months to detect a 12% relative difference in the primary outcome of Major Adverse Kidney Events within 30 days (MAKE-30), with 90% power and p <0.05 and assuming an incidence of MAKE-30 of 22% in the 0.9% saline arm. MAKE-30 is a composite endpoint composed of death, new initiation of renal replacement therapy (RRT) and persistent renal dysfunction (200% increase in creatinine from baseline) at hospital discharge or 30 days. This endpoint is similar to what has been used in a number of large cardiology studies. At an interim analysis, the target study population was adjusted to 14,000 patients during 82 unit-months by the Data Safety Monitoring Board, as observational data suggested that the incidence of MAKE-30 would be lower than anticipated (approximately 15%).
Using historical data from the Emergency Department (ED) at Vanderbilt University, the SALT-ED trial planned to enroll 14,000 subjects who received at least ≥500cc of crystalloids in the ED over 16 months and would be hospitalized outside an ICU. The primary endpoint for the SALT-ED trial was hospital-free days to day 28, a composite of death and time out of the hospital for survivors, and this sample size provided 90% power to detect a 0.5-day difference in hospital-free days, p < 0.05.
There are several reasons that it would have been impractical to conduct these studies as conventional randomized clinical trials. First, the cost of performing such studies would have been prohibitive. Second, given the acute nature of the intervention, it would be impractical to obtain consent and randomize a subject before fluids were administered. Although there are alternative approaches that could be used to overcome this barrier (waiver of initial consent for research in an emergency setting, as is allowed by FDA/Office of Human Research Protections provided very clear guidelines are met, or randomization after initial fluids were administered), these would each introduce additional complexities or potentially contaminate the intervention. Thus, both SMART and SALT-ED were embedded into clinical care at Vanderbilt University: after a pilot study demonstrated the feasibility of enrolling patients in the medical ICU and achieving separation of the two treatment arms,4 the ED and ICUs were cluster randomized to alternating months where balanced salt solutions (either PlasmaLyte or Lactated Ringers) or 0.9% saline were the default solution recommended by the electronic health record unless the subject had relative contraindications or a physician chose to override the designated fluid. Possible contraindications for balanced crystalloids included hyperkalemia and brain injury (given concern for increased intracranial pressure risk from the relative hypotonicity of balanced crystalloids compared to 0.9% saline).
The baseline characteristics of the study population across the two treatment arms were as follows in SMART: median age of 58, ~58% male, 80% white, median baseline Cr ~0.9, ~17% baseline stage 3 CKD or higher, and ~8.5% AKI at enrollment. In the SALT-ED trial, participants had a median age of ~54, ~50% male, 75% white, median baseline Cr of 0.85, ~18% with Serum Cr ≥ 1.5 mg/dL and ~10% stage 2 or higher AKI. Using an opt-out design, adherence to protocol was high. In the SMART trial, of 13,085 orders for crystalloid during the balanced salt administration periods, the fluid allocation was overruled 348 times for hyperkalemia, 278 for brain injury, and 232 per attending request (adherence of 93.%). During the 0.9% saline months, 270 of 12,261 orders were overruled (adherence of 97.8%). In SALT-ED, adherence was 83.8% in the balanced crystalloid group and 92.8% in the 0.9% saline group.
Both studies demonstrated a reduction in MAKE-30 with the use of balanced salt solutions: 1.1% (14.3% vs 15.4%, p = 0.04) in SMART and 0.9% (4.7% vs 5.6) in SALT-ED, p = 0.01). Subgroup analyses suggest that the difference in outcomes was greater among those who received larger volumes of fluids and those with sepsis (30 day in hospital mortality 25.2% with balanced crystalloids versus 29.4% with 0.9% saline, p = 0.02) within the critically ill population. Interestingly, however, the impact of the intervention on the subcomponents of MAKE-30 differed in the two trials. While the major contributor to MAKE-30 in SMART was a difference in mortality, the major contributor for SALT-ED was the difference in those with persistent renal dysfunction (Table 1B). Given the pragmatic nature of the study, a potential criticism of the MAKE-30 endpoint is that mortality, being initiated on RRT and persistent renal dysfunction on discharge are outcomes with very different clinical significance for patients. Furthermore, baseline creatinine had to be imputed for 10.7% of study population in SMART and 35% of those in SALT-ED, which may impact the reliability of the creatinine change component of the MAKE-30 outcome. Data was also censored post hospital discharge, so the incidence of renal dysfunction at 30 days may be overestimated, in particular in SALT-ED, where the median number of hospital-free days to day 28 was 25, for median hospital length of stay of 3 days. Ultimately, no difference in hospital-free days was identified in SALT-ED between the two groups.
How do these studies compare with prior studies?
A number of observational studies8,9 have suggested benefit with the administration of balanced salt solutions but are challenging to interpret because of the potential for residual confounding. In a sequential period study design, Yunos9 and colleagues examined the impact of balanced salt solutions and chloride rich solutions for resuscitation; in this analysis, both colloids and crystalloids were modified. During the balanced salt period, the incidence of AKI defined by the RIFLE criteria and of RRT decreased significantly.
In contrast, the SPLIT trial,5 a double blind, cluster randomized crossover trial of 0.9% saline versus PlasmaLyte failed to show a difference in the rate of AKI (defined as RIFLE “Injury” or “Failure” between groups). In this analysis of 2278 patients, the incidence of AKI was 9.6% vs 9.2% in those who received balanced salt solutions versus 0.9% saline (absolute difference 0.4%, [95%CI −2.1%to 2.9]). The incidence of RRT was 3.3% vs 3.4% (absolute difference −0.1%, [95%CI −1.6%to 1.4%]). Although not statistically significantly different, the incidence of death was lower with balanced salt administration 7.6% vs 8.6% (absolute difference −1.0%, [95%CI, −3.3%to 1.2%]). This difference is similar in magnitude (approximately 1%) to that observed in the SMART trial and raises the possibility that although very large in size for a critical care trial, the SPLIT trial was not sufficiently powered to observe small outcome differences. In addition, although the SPLIT trial was a blinded intervention, by the end of the study period two thirds of clinicians were able to correctly guess the treatment arm (likely due to the development of hyperchloremic metabolic acidosis in the 0.9% saline treated patients), suggesting that there is a limited role for blinding in such fluid trials.
What are the implications for nephrologists?
The SMART trial suggests potential benefit to the routine use of balanced crystalloids in critically ill patients, especially those with sepsis. SALT-ED, although not positive for its primary endpoint of hospital-free days, demonstrated a lower incidence of MAKE-30 with balanced crystalloids. Guidance on fluid management in non-critically ill patients is perhaps more limited as the intervention was applied only in the ED setting and was not maintained throughout the subjects’ entire non-ICU hospitalization. The pragmatic design of both studies allowed for rapid recruitment with limited contamination of the designated intervention and high adherence rate. Potential bias may have been introduced as hyperkalemia was listed as a relative contraindication for balanced crystalloids only when physiologically, 0.9% saline infusions may also be associated with hyperkalemia as hyperchloremic acidosis from the infusion may leads to extracellular shifting of potassium. This bias is evidenced by the decreased adherence observed in the balanced crystalloid arm of both studies. However, such deviations from allocation to balanced salts could only diminish the observed difference in outcome between the two treatment arms. Both trials are also limited in differentiating outcomes between types of balanced solutions; in SALT-ED, 95.3% of the balanced crystalloids administered were Lactated Ringers, so generalizability of the findings to PlasmaLyte A is unknown. In SMART, more PlasmaLyte A was administered than Lactated Ringers, but the exact proportion and the number of participants who had either versus both is unknown. Based on the results of these studies, nephrologists are likely to see shifts in favor of balanced crystalloids in fluid resuscitation. In clinical practice, the potential benefit of balanced crystalloids must be weighed against the potential increase in cost. The current estimated cost of PlasmaLyte and Lactated Ringers is about $4.50 compared to $2 per liter for 0.9% saline. The PLUS study,10 an upcoming multicenter trial, with 90-day mortality is likely to further inform the ideal choice of resuscitation fluids for the critically ill. Similar trials in a multicenter setting are also needed in the ED and outside the ICU.
Table 2:
Selected Results of the SMART and SALT-ED Trials
| ICU SMART | Non ICU SALT-ED | |||||
|---|---|---|---|---|---|---|
| Components of Primary Outcome |
Balanced Crystalloid % |
Saline % |
% Difference | Balanced Crystalloid % |
Saline % | % Difference |
| In Hospital Death Before 30 Days |
10.3% | 11.1% | 0.8% | 1.4% | 1.5% | 0.1% |
| New RRT | 2.5% | 2.9% | 0.4% | 0.3% | 0.5% | 0.2% |
| Final Serum Creatinine ≥ 200% of Baseline |
6.4% | 6.6% | 0.2% | 3.8% | 4.5% | 0.8% |
|
Major Adverse
Kidney Events within 30 days |
14.3% | 15.4% | 1.1% | 4.7% | 5.6% | 0.9% |
Acknowledgements
KDL has been a consultant for Theravance, Potrero Medical, Quark, and Durect; none of these companies make products relevant to this commentary. She holds stock in Amgen. She will be a speaker at a symposium held by Baxter, and her talk will focus on Fluid Overload and Acute Kidney Injury. She will present data but she will not specifically endorse any products.
References
- 1.Wilcox CS. Regulation of renal blood flow by plasma chloride. J Clin Invest. 1983;71(3):726–735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bullivant EM, Wilcox CS, Welch WJ. Intrarenal vasoconstriction during hyperchloremia: role of thromboxane. Am J Physiol. 1989;256(1 Pt 2):F152–157. doi: 10.1152/ajprenal.1989.256.1.F152 [DOI] [PubMed] [Google Scholar]
- 3.Chowdhury AH, Cox EF, Francis ST, Lobo DN. A randomized, controlled, double-blind crossover study on the effects of 2-L infusions of 0.9% saline and plasma-lyte® 148 on renal blood flow velocity and renal cortical tissue perfusion in healthy volunteers. Ann Surg. 2012;256(1):18–24. doi: 10.1097/SLA.0b013e318256be72 [DOI] [PubMed] [Google Scholar]
- 4.Semler MW, Wanderer JP, Ehrenfeld JM, et al. Balanced Crystalloids versus Saline in the Intensive Care Unit. The SALT Randomized Trial. Am J Respir Crit Care Med. 2017;195(10):1362–1372. doi: 10.1164/rccm.201607-1345OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Young P, Bailey M, Beasley R, et al. Effect of a Buffered Crystalloid Solution vs Saline on Acute Kidney Injury Among Patients in the Intensive Care Unit: The SPLIT Randomized Clinical Trial. JAMA. 2015;314(16):1701–1710. doi: 10.1001/jama.2015.12334 [DOI] [PubMed] [Google Scholar]
- 6.Semler MW, Self WH, Wanderer JP, et al. Balanced Crystalloids versus Saline in Critically Ill Adults. N Engl J Med. 2018;378(9):829–839. doi: 10.1056/NEJMoa1711584 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Self WH, Semler MW, Wanderer JP, et al. Balanced Crystalloids versus Saline in Noncritically Ill Adults. N Engl J Med. 2018;378(9):819–828. doi: 10.1056/NEJMoa1711586 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Shaw AD, Bagshaw SM, Goldstein SL, et al. Major complications, mortality, and resource utilization after open abdominal surgery: 0.9% saline compared to Plasma-Lyte. Ann Surg. 2012;255(5):821–829. doi: 10.1097/SLA.0b013e31825074f5 [DOI] [PubMed] [Google Scholar]
- 9.Yunos NM, Bellomo R, Hegarty C, Story D, Ho L, Bailey M. Association between a chloride-liberal vs chloride-restrictive intravenous fluid administration strategy and kidney injury in critically ill adults. JAMA. 2012;308(15):1566–1572. doi: 10.1001/jama.2012.13356 [DOI] [PubMed] [Google Scholar]
- 10.Hammond NE, Bellomo R, Gallagher M, et al. The Plasma-Lyte 148 v Saline (PLUS) study protocol: a multicentre, randomised controlled trial of the effect of intensive care fluid therapy on mortality. Crit Care Resusc. 2017;19(3):239–246. [PubMed] [Google Scholar]