11. Balanced Crystalloids vs. Saline in Critical Illness

Quick Recap

Cross-cutting protocol — companion to Fluid Responsiveness & Hemodynamic Assessment protocol, which addresses how much fluid to give; this protocol addresses the separate, more basic question of which crystalloid to give. Follows a now-familiar pattern in this library: a large, influential single-center trial found a real signal favoring an intervention, only for two much larger, better-resourced, multicenter trials to fail to replicate it — though with a genuinely interesting, still-relevant twist regarding what fluid a patient received before the study even began.

1. Definition

Balanced crystalloids: intravenous fluids with electrolyte composition closer to physiologic plasma levels — principally lactated Ringer's solution (LR) and Plasma-Lyte A/148 — typically containing lower chloride concentrations and a buffer (lactate, acetate, or gluconate) compared to normal saline.

Normal saline (0.9% sodium chloride): the most commonly administered crystalloid historically, containing a supraphysiologic chloride concentration (154 mEq/L, compared to plasma's ~100–106 mEq/L) — the specific chemical property proposed to drive the theoretical harms discussed below. As one widely-cited commentary on this literature puts it succinctly: "there is nothing normal about normal saline."

The proposed mechanism of saline-associated harm: large-volume saline administration can produce hyperchloremic metabolic acidosis, and observational studies and experimental models have suggested this may cause renal cortical vasoconstriction, reduced renal blood flow, and consequently increased risk of acute kidney injury and death — a biologically coherent, mechanistically plausible hypothesis that motivated this entire line of trial evidence, distinct from the volume-status question addressed in the Fluid Responsiveness protocol.

2. Pathophysiology

The proposed causal chain is direct: supraphysiologic chloride load from saline administration → hyperchloremia → metabolic acidosis and/or direct chloride-mediated renal vasoconstriction → reduced renal cortical blood flow → increased AKI risk and, downstream, increased mortality. This mechanism has genuine experimental and observational support, and is not simply a discredited hypothesis — the question this evidence base actually resolves is not "is the mechanism real" but "does this mechanism translate into a clinically detectable difference in patient-centered outcomes at the fluid volumes typically administered in contemporary ICU practice" — a genuinely different and, as the evidence below shows, harder question to answer definitively than the underlying biochemistry alone would suggest.

3. Immediate Stabilization (ABCDE) — Not Applicable in the Traditional Sense

This protocol addresses routine crystalloid fluid selection rather than acute stabilization; cross-reference Fluid Responsiveness & Hemodynamic Assessment and Vasopressor & Inotrope Selection & Titration protocols for the broader resuscitation framework this fluid-choice decision sits within.

Checklist:

4. Focused History

• Type and volume of IV fluid already received before the current care episode/admission — a specifically relevant historical detail per the BaSICS secondary analysis (Section 11)
• Renal function and potassium status (relevant to balanced crystalloid contraindication screening)
• Traumatic brain injury or other condition where osmolality/tonicity considerations are specifically relevant
• Anticipated volume of fluid resuscitation required (relevant given the dose-response relationship identified in SMART's subgroup analysis, Section 11)

5. Comprehensive System-wise Examination — Not a Primary Component

Not a primary examination-driven protocol; relevant assessment is standard volume status and renal function monitoring, addressed more fully in the Fluid Responsiveness & Hemodynamic Assessment protocol.

6. Syndrome Identification — Reframed as Fluid-Choice Applicability Classification

• General critically ill patient, anticipated moderate-to-large volume resuscitation, no specific contraindication: balanced crystalloid remains a reasonable default per SMART's subgroup findings and per the broader safety profile, even though the larger confirmatory trials (BaSICS, PLUS) did not replicate SMART's overall mortality benefit
• Patient who has already received primarily saline before this specific care episode: a genuinely important, specific consideration per the BaSICS secondary analysis (Section 11) — switching to balanced crystalloids mid-course in a patient already saline-loaded may not confer the same benefit as consistent balanced-crystalloid use from the outset
• Hyperkalemia or renal failure with elevated potassium: relative contraindication to potassium-containing balanced crystalloids (LR, Plasma-Lyte); saline or a potassium-free balanced alternative may be preferred
• Traumatic brain injury with cerebral edema concern: LR's mild hypotonicity relative to plasma has raised theoretical concern in this specific population, though this represents a narrower, population-specific consideration distinct from the general critically ill population's evidence base

7. Differential Diagnosis — Not Applicable

8. Severity/Risk Assessment

Volume of crystalloid administered: SMART's own subgroup analysis found the difference in the primary outcome was more pronounced among patients receiving larger volumes of crystalloid — a genuine, biologically plausible dose-response signal suggesting that if a true effect exists, it is likely concentrated in patients receiving substantial fluid volumes rather than uniformly present regardless of total volume administered; median volumes in SMART itself were relatively modest (~1000–1020 mL), which some commentators have noted may partly explain why the overall effect size, while statistically significant for the composite outcome, was not dramatic.

Sepsis as a specific subgroup of interest: SMART's sepsis subgroup (and a dedicated secondary analysis) showed a larger absolute mortality difference (33.8% vs. 38.9%, a 4.2% absolute risk reduction) than the trial's overall population — though this remains a subgroup, not the trial's primary, pre-specified population-wide finding, and should be weighed with the corresponding degree of caution.

9. Investigations

Routine labs: serum chloride and bicarbonate/acid-base status as relevant markers of the proposed underlying mechanism (hyperchloremic acidosis), though not required to inform the routine, prospective fluid-choice decision itself.

10. Point-of-Care Ultrasound — Not Applicable

11. Evidence-Based Management — The Full Evidence Arc

Early, Single-Center Signal — SMART (2018)

• SMART trial (Semler et al., NEJM 2018): a pragmatic, unblinded, cluster-randomized, multiple-crossover trial conducted in 5 ICUs at a single academic medical center (Vanderbilt), n=15,802, comparing balanced crystalloids (LR or Plasma-Lyte A) vs. normal saline
• Primary outcome (major adverse kidney event within 30 days [MAKE-30]: composite of death, new RRT, or persistent renal dysfunction): 14.3% (balanced) vs. 15.4% (saline), p=0.04 — a statistically significant, though modest, absolute difference, corresponding to a number needed to treat of 94
• 30-day in-hospital mortality specifically: 10.3% (balanced) vs. 11.1% (saline), p=0.06 — not independently statistically significant, with the composite primary outcome's significance being primarily driven by the mortality component even though mortality alone did not cross the conventional threshold — a nuance worth holding alongside the trial's headline positive result
• A specific, quantified dose-response relationship: the difference in outcomes was more pronounced among patients receiving larger volumes of crystalloid, and among patients with sepsis specifically (a secondary analysis found 33.8% vs. 38.9% mortality in the sepsis subgroup, an even larger absolute difference than the trial's overall population)
• Explicit, acknowledged generalizability limitation: SMART was conducted at a single academic center — a genuine constraint on external validity that the trial's own authors and subsequent commentary specifically flagged, directly motivating the design of the two larger, multicenter confirmatory trials discussed below
• The companion SALT-ED trial (non-critically-ill ED patients) found broadly similar directional results in a distinct, lower-acuity population, extending interest in this question beyond the ICU specifically, though this protocol focuses on the critically ill population

The Larger, Multicenter Confirmatory Trials — A Genuinely Different Result

• BaSICS trial (Zampieri et al., JAMA 2021): a multicenter, double-blind, factorial RCT across 75 ICUs in Brazil, n=10,520, specifically enrolling patients with at least one AKI risk factor — by a wide margin, a larger, more methodologically rigorous (double-blind, vs. SMART's unblinded design) test of the same underlying question
• Primary outcome (90-day mortality): no significant difference between balanced solution (Plasma-Lyte 148) and saline (hazard ratio 0.97, 95% CI 0.90–1.05) — a clear, well-powered null result, directly at odds with SMART's headline finding
• PLUS trial (Finfer et al., NEJM 2022): a multicenter, double-blind RCT across ICUs in Australia and New Zealand, n>5,000, comparing Plasma-Lyte 148 vs. saline — similarly found no significant difference in 90-day mortality or other major clinical outcomes, though saline did produce significantly higher serum chloride and lower pH than balanced solution, confirming the underlying biochemical mechanism occurs as expected, without this biochemical difference translating into a detectable difference in patient-centered clinical outcomes
• A pooled systematic review/meta-analysis incorporating SMART, BaSICS, and PLUS (Hammond et al., NEJM Evidence 2022): the definitive, most current synthesis of this literature — findings broadly consistent with the larger trials' null result rather than SMART's original, single-center positive signal

A Genuinely Interesting, Still-Relevant Twist — Pre-Enrollment Fluid Exposure as an Effect Modifier

• A specific secondary analysis of BaSICS examined whether the type of fluid patients had received in the 24 hours before enrollment (i.e., before the study's own randomized fluid assignment took effect) modified the relationship between randomized fluid assignment and mortality — a substantial majority (68%) of BaSICS patients had already received saline, balanced crystalloids, or both before enrollment, since most patients were enrolled within 1 day of ICU admission but had typically already received some initial resuscitation fluid via standard, non-randomized care
• Finding: for patients who had received only balanced crystalloids before enrollment, continuing balanced crystalloids in the ICU appeared to decrease mortality; for patients who had already received saline before enrollment, the choice of fluid after enrollment did not appear to affect outcomes
• Practical significance: this is a genuinely important, mechanistically plausible refinement — it suggests that consistency of fluid strategy from the earliest point of resuscitation, rather than simply the fluid type chosen at any single, later decision point, may be the more relevant variable, and that switching a patient's fluid type partway through their resuscitation course (after they have already received a substantial saline load) may not "undo" whatever chloride/acid-base perturbation has already occurred — a nuance that the overall, population-wide null results of BaSICS and PLUS do not by themselves fully capture, though this remains a secondary, hypothesis-generating analysis rather than a primary, confirmed trial finding

Practical Synthesis

The weight of evidence, particularly from the two larger, more rigorously designed multicenter trials (BaSICS and PLUS) that specifically followed and were motivated by SMART's single-center signal, does not support a broad, population-wide mortality benefit from balanced crystalloids over saline in critically ill adults — a genuine reversal pattern consistent with several other topics in this library (cross-reference the Glycemic Control protocol's Leuven-to-NICE-SUGAR arc, and the ACURASYS-to-ROSE pattern in Neuromuscular Blockade in ARDS, as structurally similar single-center-to-multicenter evidence evolutions). This does not mean balanced crystalloids are harmful or that saline should be preferred — both fluids remain reasonable choices for the general critically ill patient given the overall null findings, and balanced crystalloids retain a plausible, if unconfirmed at the population level, advantage in patients receiving large fluid volumes or with sepsis specifically (per SMART's dose-response and sepsis-subgroup signals), and in patients who have received primarily balanced fluids from the earliest point of their resuscitation course (per the BaSICS pre-exposure analysis). No major guideline had been published reflecting SMART's specific findings even before the larger confirmatory trials' null results emerged — reflecting that this question, even at its most enthusiastic single-center-trial stage, had not yet crossed into confident, guideline-endorsed practice change, and the subsequent null confirmatory trials further support a position of reasonable equipoise between the two fluid choices for the general population, with fluid choice reasonably individualized based on the specific factors above (volume anticipated, sepsis status, prior fluid exposure, contraindications) rather than a uniform mandate in either direction.

12. Organ Support

Interacts directly with Fluid Responsiveness & Hemodynamic Assessment protocol (the volume/timing question this protocol's fluid-type question is distinct from but complementary to) and with relevant AKI/Renal System protocols given this evidence base's original AKI-focused rationale.

13. Disease-Specific Therapy

• Balanced crystalloids (LR or Plasma-Lyte): a reasonable default choice, particularly for patients anticipated to require larger fluid volumes, with sepsis, or who have already received predominantly balanced fluids earlier in their resuscitation course
• Normal saline: remains a reasonable, non-inferior choice per the larger confirmatory trials; specifically preferred in hyperkalemia/renal failure with elevated potassium (avoiding LR/Plasma-Lyte's potassium content) and in situations where balanced crystalloid's mild hypotonicity is a specific theoretical concern (e.g., some traumatic brain injury contexts)
• Consistency of fluid strategy from early resuscitation onward may matter more than the specific fluid chosen at any single later decision point, per the BaSICS pre-exposure secondary analysis — a genuinely practical, actionable nuance for institutional fluid-ordering protocols and early resuscitation practice

14. Consultation Matrix

15. Monitoring Framework

• Laboratory: serum chloride and acid-base status as markers of the underlying proposed mechanism, though not required to guide routine fluid-choice decisions given the null population-level clinical outcome findings
• Clinical: standard volume status and renal function monitoring per Fluid Responsiveness & Hemodynamic Assessment protocol

16. ICU Bundle Checklist

17. Complications

Saline-specific (proposed, though not confirmed to translate into clinical outcome differences at typical resuscitation volumes): hyperchloremic metabolic acidosis — confirmed as a real biochemical effect in PLUS (significantly higher chloride, lower pH with saline), without this translating into a detectable difference in kidney function or mortality in that trial

Balanced crystalloid-specific: hyperkalemia risk in patients with renal failure or other hyperkalemia-prone states, given LR/Plasma-Lyte's potassium content; theoretical hypotonicity-related concern in specific populations (e.g., some TBI contexts)

Prevention: appropriate contraindication screening; individualized fluid selection per Section 6/13

18. Escalation & De-escalation — Not Applicable

19. ICU Discharge Criteria — Not Applicable

20. Documentation & Medicolegal Checklist

• Fluid type selected and rationale documented, particularly for any specific contraindication-driven choice
• Prior fluid exposure (if relevant/known) considered and documented where it informs the decision

21. Key Guidelines

• No major guideline has issued a strong, universal recommendation favoring balanced crystalloids over saline (or vice versa) for the general critically ill population, consistent with the genuine equipoise reflected in the larger confirmatory trials' null findings

22. Landmark Trials

23. Controversies

• This is a structurally familiar pattern in this library, worth naming explicitly: an influential single-center trial (SMART) generating enthusiasm and a statistically significant, if modest, composite finding, followed by larger, better-resourced, methodologically stronger multicenter trials failing to replicate the result — directly analogous to the Leuven-to-NICE-SUGAR glycemic control arc and the ACURASYS-to-ROSE neuromuscular blockade arc discussed elsewhere in this library. This protocol treats the recurrence of this specific pattern across multiple, unrelated clinical questions as itself a meaningful, generalizable lesson about evidence hierarchy — single-center pragmatic trials, however large their internal sample size, carry a structurally different level of external validity than adequately powered multicenter replication.
• The BaSICS pre-enrollment fluid exposure finding deserves more attention than it typically receives: because it is a secondary, hypothesis-generating analysis, it has not driven a corresponding change in guideline recommendations or the overall "balanced vs. saline is roughly equivalent" conclusion — but it offers a genuinely plausible, mechanistically coherent explanation for why the field's simple "which fluid is better" framing may be somewhat mis-specified, and that "was the patient's entire resuscitation course consistent" may be the more clinically relevant question that no trial to date has been specifically designed to test as its primary hypothesis.
• SMART's composite outcome vs. its individually non-significant mortality component: similar to several other examples in this library (the Doig refeeding syndrome trial, HFNC vs. NIV's FLORALI discussion), SMART's positive result was driven by a composite outcome whose most clinically important individual component (mortality) did not independently reach significance — a recurring pattern worth recognizing as a general principle in evidence interpretation across this entire library, not unique to this specific topic.
• Whether a true, smaller effect concentrated in specific subgroups (large-volume resuscitation, sepsis) remains genuinely unconfirmed or has been adequately excluded by the larger trials: BaSICS and PLUS's overall null results do not definitively rule out a real, if modest, benefit in these specific subgroups, since neither trial was specifically powered or designed around SMART's own subgroup-derived hypotheses — this represents a genuine, still-open area rather than a fully closed question, even as the overall population-level conclusion has clearly shifted toward equipoise.

24. References

1. Semler MW, Self WH, Wanderer JP, et al; SMART Investigators and the Pragmatic Critical Care Research Group. Balanced crystalloids versus saline in critically ill adults. N Engl J Med. 2018;378(9):829-839.
2. Self WH, Semler MW, Wanderer JP, et al. Balanced crystalloids versus saline in noncritically ill adults (SALT-ED). N Engl J Med. 2018;378:819-828.
3. Brown RM, Wang L, Coston TD, et al. Balanced crystalloids versus saline in sepsis: a secondary analysis of the SMART trial. Am J Respir Crit Care Med. 2019;200(12):1487-1495.
4. Zampieri FG, Machado FR, Biondi RS, et al; BaSICS Investigators and the BRICNet. Effect of intravenous fluid treatment with a balanced solution vs 0.9% saline solution on mortality in critically ill patients: the BaSICS randomized clinical trial. JAMA. 2021;326(9):818-829.
5. Finfer S, Micallef S, Hammond N, et al; PLUS Study Investigators. Balanced multielectrolyte solution versus saline in critically ill adults. N Engl J Med. 2022;386(9):815-826.
6. Hammond NE, Zampieri FG, Di Tanna GL, et al. Balanced crystalloids versus saline in critically ill adults — a systematic review with meta-analysis. NEJM Evid. 2022;1(2).
7. Zampieri FG, et al. Back to BaSICS: early treatments matter in critical illness [secondary analysis on pre-enrollment fluid exposure]. Am J Respir Crit Care Med. 2022.
8. Young P, Bailey M, Beasley R, et al; SPLIT Investigators. 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;314:1701-1710.
9. The Washington Manual of Critical Care, 4th ed. 2025 — fluid therapy chapter.
10. ICU Protocols: A Step-wise Approach, 2nd ed. — fluid management content.