#NephJC Chat
Tuesday September 7 at 9 pm Eastern
Wednesday September 8 at 9 pm IST and BST
JAMA. 2021 Aug 10;e2111684. doi: 10.1001/jama.2021.11684. Online ahead of print.
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
Fernando G Zampieri, Flávia R Machado, Rodrigo S Biondi, Flávio G R Freitas, Viviane C Veiga, Rodrigo C Figueiredo, Wilson J Lovato, Cristina P Amêndola, Ary Serpa-Neto, Jorge L R Paranhos, Marco A V Guedes, Eraldo A Lúcio, Lúcio C Oliveira-Júnior, Thiago C Lisboa, Fábio H Lacerda, Israel S Maia, Cintia M C Grion, Murillo S C Assunção, Airton L O Manoel, João M Silva-Junior, Péricles Duarte, Rafael M Soares, Tamiris A Miranda, Lucas M de Lima, Rodrigo M Gurgel, Denise M Paisani, Thiago D Corrêa, Luciano C P Azevedo, John A Kellum, Lucas P Damiani, Nilton Brandão da Silva, Alexandre B Cavalcanti, BaSICS investigators and the BRICNet members.
PMID: 34375394
Introduction
Intravenous fluids remain one of the most ubiquitous interventions in medicine. The ideal resuscitation fluid is one that (Myburgh and Mythen 2013):
Closely resembles physiologic body fluid
Is effective in increasing intravascular volume
Completely metabolized and excreted
Devoid of adverse effects
Cost-effective
The search for the utopian fluid has recently fueled studies pitting balanced crystalloids against saline solutions.
The opening salvo to the fluid wars came with the prospective, open-label sequential period study by Yunos et al. in 2012. They found that use of a chloride-restrictive regimen with balanced crystalloids resulted in a significant decrease in AKI and KRT. Three years later came the SPLIT trial (Young et al. 2015). This was the first randomized and blinded controlled trial and did not report any significant difference in the incidence of AKI, RRT-use and in-hospital mortality between balanced and saline solutions in the ICU setting. The next phase in the war came with two more trials in 2018. The SALT-ED trial (Self et al. 2018) was a single center, cluster randomized trial which included non-critically ill patients. The primary outcome, hospital-free days, was not significantly different. However, major adverse kidney events (composite of all-cause mortality, RRT and persistent renal dysfunction) was lower in the balanced solution group. Its companion trial, SMART, was also a single center cluster RCT (Semler et al. 2018) that dealt with ICU patients. The primary outcome, major adverse kidney events, was significantly lower in the balanced solution arm. See the NephJC summary for a detailed discussion of SPLIT and SMART/SALTeD.
If you consider only the 2 trials that handled critically-ill patients, that’s 1 point for non-superiority and 1 point for superiority of balanced crystalloids. This stalemate did not last long with the arrival of the BaSICS trial this year.
The Study
Methods
Study Design
Multicenter, 75 Brazilian ICUs
Randomized controlled trial
2x2 factorial design (saline vs balanced crystalloid and slow vs. fast infusion) to answer if either factor affected outcomes.
Blinding was achieved by putting fluids in identical 500 mL bags labeled only by a code
Physicians, patients and outcome assessors were all blinded.
Patient management including fluid challenges were left to the discretion of the attending physician.
Inclusion Criteria
ICU patients needing:
At least 1 fluid expansion
Admitted for more than 1 day after enrollment
Met at least 1 of the following criteria:
Age >65 years
Hypotension (MAP <65 mmHg, SBP <90 mmHg, or vasopressor use)
Sepsis
Requiring mechanical / non-invasive ventilation for at least 12 hours
Oliguria (<0.5 mL/kg/hr for at least 3 hours) or azotemia (sCr >1.2 mg/d for women, >1.4 mg/d for men)
Liver cirrhosis or acute liver failure
Exclusion Criteria
AKI requiring RRT within 6 hours of admission
Severe electrolyte disturbance (serum Na ≤120 mmol/L or ≥160 mmol/L)
Imminent death within 24 hours
Suspected or confirmed brain death
On palliative or comfort care
Previously enrolled in the trial
Serum hyperkalemia >5.5 mEq/L (added after second interim analysis)
Outcomes
90-day survival (primary outcome)
Secondary outcomes:
Need for KRT up to 90 days after enrollment
Occurrence of AKI (KDIGO stage 2 or 3) at days 3 and 7
SOFA score (total and individual components) at days 3 and 7
Mechanical ventilation-free days within 28 days
Tertiary outcomes:
ICU and hospital mortality
ICU and hospital length of stay
Statistical Analysis
Study was designed to enroll 11,000 patients (based on 35% mortality within 90 days in control group with 89% power to detect hazard ratio for mortality of 0.9 or less with an a level of 0.05
Three interim analysis were done at 25%, 50% and 75% of study cohort
Primary outcome was tested using mixed-effects Cox proportional hazards models. Multiple imputation chains were used for patients with missing data.
Pre-specified subgroup analysis for the primary outcome was performed
Funding
The trial was supported by the Brazilian Ministry of Health through the Programa de Desenvolvimento Institucional do Sistema Único de Saúde. Fluids and logistics were supplied by Baxter Hospitalar.
Results
A total of 11,052 patients were randomized between May 29, 2017 and March 2, 2020 at 75 ICUs. There were 532 patients who were excluded from the analysis (486 patients subsequently refused to provide consent and 46 were duplicate patients, leaving 10,520 patients for the analysis (5230 patients in balanced solution and 5290 in saline solution).
Figure 1. Flow of Patients in the Trial Comparing a Balanced Solution vs Saline Solution (0.9% Sodium Chloride)
Baseline Characteristics
Baseline characteristics are shown in Table 1. The mean age of patients was 61.1 years and 44% were women. 48% of patients were admitted to the ICU after elective surgery. Majority of the patients were hypotensive (~61%) and 44% required mechanical ventilation. Mean serum creatinine level was 1.2 mg/dL. 80% of patients had a serum creatinine level of <1.5 mg/dL. 68% received crystalloid fluid bolus before ICU admission.
Table 1. Baseline Characteristics
Patients in both groups received a median of 1.5 L of fluid during the first day. Accumulated median total fluid received (study and non-study fluid) during the first 3 days was 4.1 L. During the same period, median study fluid received was 2.9 L. The amount of total fluid infused to both groups were similar all throughout the protocol adherence check time points (Figure 2A).
Figure 2A. Mean infused fluids on Days 1, 2, 3 and 7
Primary Outcome
There was no significant difference in the 90-day mortality between the two groups. Within 90 days, 1,381 of 5,230 patients (26.4%) assigned to a balanced solution died versus 1,439 of 5,290 patients (27.2%) assigned to saline solution (adjusted HR, 0.97 [95% CI, 0.90-1.05]; P = .47).
Figure 3. Cumulative Incidence of the Primary Outcome of 90-Day Survival for a Balanced Solution vs Saline Solution (0.9% Sodium Chloride)
There was no significant interaction between the 2 interventions (fluid type and infusion speed; P = .98) or between groups for the primary outcome (see eTable 3).
eTable 3
Secondary Outcomes
Among the secondary outcomes, only the total SOFA score at day 7 was significantly different for the balanced solution group (median difference, 0.27 [95% CI, 0.08-0.45]). This was mainly due to a higher neurological SOFA score (>2) at day 7 in the balanced group (32% vs 26% for the saline solution group; odds ratio, 1.40 [95% CI, 1.18-1.66]).
Table 2: outcomes from BaSICs trial
Subgroup Analysis
90-day mortality was significantly higher in patients with traumatic brain injury in the balanced solution group (31.3% for the balanced solution group vs 21.1% for the saline solution group [HR, 1.48; 95% CI, 1.03-2.12]; P=.02) (Figure 4). There were no other significant interactions for the predefined subgroups.
Figure 4 from BaSICs trial
Tertiary Outcomes, Exploratory and Post-Hoc Analyses
There was no significant difference between groups in terms of ICU/hospital mortality and length of stay (Table 2B). Stratification by baseline serum chloride did not significantly affect primary outcomes between groups (eTable 4).
Table 2B. Tertiary Outcomes
eTable 4. Results for primary endpoint stratified by baseline chloride values
The Bayesian network analysis was consistent with a high probability that use of the balanced solution is associated with a lower Glasgow Coma Scale score (≤12) for patients that required mechanical ventilation at day 7 (eTable 5).
eTable 5. Results for queries in the Bayesian Network
There was no significant difference between groups in the composite outcome of mortality and use of KRT both in patients who did and those who did not receive fluids within the 24 hours prior to enrollment (eTable 6). There was no difference in the primary outcome between groups at any KDIGO stage of AKI at enrollment (eTable 7)
eTable 6. Composite endpoint of mortality and use of renal replacement therapy during hospital stay according to use of resuscitation fluid before enrollment
eTable 7. Primary endpoint analysis according to KDIGO at enrollment
There was no difference between groups in the occurrence of creatinine doubling or incident renal failure (eTable 8).
eTable 8. Sensitivity based on creatinine analyses
Patients randomized to the balanced solution group had lower chloride levels than patients randomized to the saline solution group (P<0.001) (eFigure 7). After excluding patients with traumatic brain injury, results for primary, secondary and tertiary outcomes remained unchanged.
eFigure 7. Chloride levels over time displayed at mean and standard deviation
Adverse Events
There were no reported treatment-related severe adverse events in both groups.
Discussion
The BaSICS trial showed that in critically ill patients, the use of balanced solution did not change 90-day mortality compared to saline solution. Stratifying according to infusion rate did not affect the primary outcome in both groups. None of the secondary or subgroup analyses showed benefit with the use of balanced solution. The use of a balanced solution suggested potential harm, as seen with a higher total SOFA score at day 7, an increased 90-day mortality among patients with traumatic brain injury, and a lower Glasgow coma score for patients needing mechanical ventilation at day 7. Does it matter that SPLIT and BaSICS used Plasmalyte and SMART/SALTeD and the original Yunos mostly used Ringer's/Hartmann’s (both of which have lactate buffer compared to acetate/gluconate in Plasmalyte)?
Table from Kwong and Liu, AJKD 2018
Strengths
Large blinded randomized controlled trial which investigated the effect of varying infusion rates to 90-day mortality and various secondary and tertiary outcomes
Multicenter study with data from 75 ICUs!
Large percentage of enrolled patients were included in the primary outcome analysis.
Adherence to the study fluids was good, with ~80% of all fluids received during the first days being the study fluid (balanced or saline solution)
Limitations
Allowed use of non-study fluids led to some degree of contamination
Fluid use prior to ICU stay may have affected the effect of fluid type on outcomes
Post randomization exclusions occurred (but mostly due to lack of consent)
Almost half of participants in both groups were elective surgical patients, which may have reduced overall mortality
Trial was designed with a higher expected 90-day mortality than what was observed in the results. This may have caused the study to have a lower power to detect significant differences in the outcomes.
Conclusion
Unlike in the similarly ICU-centered SMART trial, the BaSICS trial failed to demonstrate a lower mortality outcome, need for KRT, or incidence of AKI with the use of balanced solutions compared to saline solutions. The difference in outcomes in both trials was about 1% but the differences in their study designs meant that there was a significant P value in SMART and a non-significant P-value in BaSICS. Its findings closely resemble the smaller but similarly blinded SPLIT trial. Being conducted in a huge number of centers (unlike the single-center SMART trial) increases the generalizability of the BaSICS trial results. Because the initial resuscitation fluid was not part of the protocol and there was a lower than expected 90-day mortality, it diluted a lot of the study’s power. The balanced solution proponents point to the significant changes in chloride levels affecting outcomes. The BaSICS trial demonstrated otherwise; with serum chloride having no significant effect on the outcomes. Given that the balanced solution was not harmful (except in patients with traumatic brain injury) and there is a suggestion of benefit from previous studies, one might suggest using the balanced solution anyways. What is the downside? Well, these are about $4.50 per litre, compared to $2.00 for 0.9% saline (Kwong and Liu, AJKD 2018). The dollars will add up quickly given the amount of IV fluids we use in practice.
We may have to bank on the ongoing PLUS trial (N = 8,800 critically ill patients) and the BEST FLUIDS (N = 800 kidney transplant recipients, with delayed graft function as the primary outcome) to be able to enroll more higher risk patients and finally break the balanced-saline solution deadlock.