M Semler (SALT)

AJRCCM Articles in Press. Published on 17-October-2016 as 10.1164/rccm.
201607-1345OC
Page 1 of 74
Balanced Crystalloids versus Saline in the Intensive Care Unit: The SALT Randomized Trial
Matthew W. Semler, MD, MSc1; Jonathan P. Wanderer, MD, MPhil2,3; Jesse M. Ehrenfeld, MD,
MPH2,3; Joanna L. Stollings, PharmD, FCCM, BCCCP, BCPS4; Wesley H. Self, MD, MPH5; Edward
D. Siew, MD, MSc6; Li Wang, MS7; Daniel W. Byrne, MS7; Andrew D. Shaw, MB, FRCA2; Gordon
R. Bernard, MD1; Todd W. Rice, MD, MSc1 for The SALT Investigators and the Pragmatic Critical
Care Research Group
Affiliations: 1Division of Allergy, Pulmonary, and Critical Care Medicine; 2Department of
Anesthesiology; 3Department of Biomedical Informatics; 4Department of Pharmaceutical
Services; 5Department of Emergency Medicine; 6Division of Nephrology and Hypertension,
Vanderbilt Center for Kidney Disease (VCKD) and Integrated Program for AKI (VIP-AKI);
7
Department of Biostatistics – all at Vanderbilt University Medical Center, Nashville, TN
Running Title: Balanced Crystalloids vs Saline in the medical ICU
Corresponding Author:
Matthew W. Semler, MD
1161 21st Ave S. C-1216 MCN, Nashville, TN 37232-2650
Phone: (615) 322-3412; Fax: (615) 343-7448
Email: matthew.w.semler@vanderbilt.edu
Copyright © 2016 by the American Thoracic Society

AJRCCM Articles in Press. Published on 17-October-2016 as 10.1164/rccm.201607-1345OC
Page 2 of 74
Authors contributions: Study concept and design: M.W.S, A.D.S., G.R.B., and T.W.R.; Acquisition
of data: M.W.S, J.P.W., J.M.E., J.L.S., and T.W.R.; Analysis and interpretation of data: M.W.S,
W.H.S., E.D.S., L.W., D.W.B., and T.W.R.; Drafting of the manuscript: M.W.S, J.P.W., and T.W.R.;
Critical revision of the manuscript for important intellectual content: M.W.S, J.P.W., J.M.E.,
J.L.S., W.H.S., E.D.S., L.W., D.W.B., A.D.S., G.R.B., and T.W.R.; Statistical analysis: M.W.S, W.H.S.,
L.W., D.W.B., and T.W.R.; Study supervision: M.W.S, J.L.S., A.D.S., G.R.B., and T.W.R. M.W.S,
J.P.W., and L.W. had full access to all the data in the study and take responsibility for the
integrity of the data and the accuracy of the data analysis. L.W. and D.W.B. conducted and are
responsible for the data analysis.
Source of Funding: Biostatistical support was provided by the Vanderbilt Institute for Clinical
and Translational Research (UL1 TR000445 from NCATS/NIH). M.W.S. was supported by a
National Heart, Lung, and Blood Institute (NHLBI) T32 award (HL087738 09). W.H.S was
supported in part by the National Institute of General Medical Sciences (K23GM110469). E.D.S.
was supported by the Vanderbilt Center for Kidney Disease (VCKD) and Veterans Affairs HS&RD
IIR 13-073. T.W.R. was supported in part by the NIH (R34HL105869). The funding institutions
had no role in (1) conception, design, or conduct of the study, (2) collection, management,
analysis, interpretation, or presentation of the data, or (3) preparation, review, or approval of
the manuscript.
Conflicts of Interest: All authors completed and submitted the ICMJE Form for Disclosure of
Potential Conflicts of Interest. A.D.S. reported serving as a consultant for Baxter International
Inc. Otherwise, the authors declared no potential conflicts of interest with the current work.
J.L.S. reported receiving fees for travel, lectures, and meeting expenses from the American

Page 3 of 74
College of Chest Physicians, American College of Clinical Pharmacists, and Society of Critical
Care Medicine. W.H.S. reported serving on advisory boards for Venaxis, Inc. and BioFire
Diagnostics Inc., and as a consultant for Abbott Point-of-Care. T.W.R. reported serving on an
advisory board for Avisa Pharma, LLC, as a DSMB member for GlaxoSmithKline PLC, and as the
Director of Medical Affairs for Cumberland Pharmaceuticals, Inc.
Subject Descriptor Number: 4.4 Clinical Trials in Critical Care Medicine
Manuscript Word Count (body only): 3,764
At a Glance Commentary:
Scientific Knowledge on the Subject: Saline is the most commonly administered intravenous
crystalloid globally. Prior studies of critically ill adults have suggested an association between
saline receipt and risk of acute kidney injury and death, but results have been conflicting.
Whether use of balanced crystalloids rather than saline improves patient outcomes requires
evaluation in a large, randomized clinical trial.
What This Study Adds to the Field: This 974-patient pilot study found that a cluster-
randomized, multiple-crossover trial using software tools within the electronic health record to
compare saline with balanced crystalloids among critically ill adults produced well-balanced
study groups and separation in crystalloid receipt, but no overall difference in patient
outcomes.
This article has an online data supplement, which is accessible from this issue's table of
contents online at www.atsjournals.org

Page 4 of 74
ABSTRACT
RATIONALE: Saline is the intravenous fluid most commonly administered to critically ill adults,
but may be associated with acute kidney injury and death. Whether use of balanced
crystalloids rather than saline affects patient outcomes remains unknown.
OBJECTIVES: To pilot a cluster-randomized, multiple-crossover trial using software tools within
the electronic health record to compare saline to balanced crystalloids.
METHODS: Cluster-randomized, multiple-crossover trial among 974 adults admitted to a
tertiary medical intensive care unit from February 3, 2015 through May 31, 2015. The
intravenous crystalloid used in the unit alternated monthly between saline (0.9% sodium
chloride) and balanced crystalloids (Lactated Ringer’s solution or Plasma-Lyte A®). Enrollment,
fluid delivery, and data collection were performed using software tools within the electronic
health record. The primary outcome was the difference between study groups in the
proportion of isotonic crystalloid administered that was saline. The secondary outcome was
the Major Adverse Kidney Events within 30 days (MAKE30) composite of death, dialysis, or
persistent renal dysfunction.
MEASUREMENTS AND MAIN RESULTS: Patients assigned to saline (n=454) and balanced
crystalloids (n=520) were similar at baseline and received similar volumes of crystalloid by 30
days (median [IQR]: 1424 [500–3377] mL vs 1617 [500–3628] mL; P = 0.40). Saline comprised a
larger proportion of the isotonic crystalloid given in the saline group than in the balanced
crystalloid group (91% vs 21%; P < 0.001). MAKE30 did not differ between groups (24.7% vs
24.6%, P = 0.98).

Page 5 of 74
CONCLUSIONS: An electronic health record-embedded, cluster-randomized, multiple-crossover
trial comparing saline with balanced crystalloids can produce well-balanced study groups and
separation in crystalloid receipt.
TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT02345486
Abstract Word Count: 260

Page 6 of 74
INTRODUCTION
Intravenous crystalloid administration is ubiquitous in critical care, yet whether choice
of crystalloid affects patient outcomes remains unknown(1). The most commonly used
crystalloid solution globally is 0.9% sodium chloride (saline)(2), with over 200 million liters
administered each year in the United States alone(1). Saline’s high chloride content, however,
has been hypothesized to contribute to the development of acute kidney injury (AKI) in at-risk
patients(3, 4). Alternatives to saline include crystalloids with electrolyte compositions that
more closely approximate that of plasma (balanced crystalloids), such as Lactated Ringer’s
solution or Plasma-Lyte A®. Emerging data suggest use of balanced crystalloids in critically ill
adults may decrease rates of AKI(3, 4), renal replacement therapy (RRT)(3, 5), and death(6, 7).
To definitively evaluate the effect of balanced crystalloids versus saline on patient-
centered outcomes, a large, randomized trial that carefully accounts for differences in baseline
risk of AKI(8), volume of crystalloid received(7), and underlying pathophysiology(9–12) among
intensive care unit (ICU) patients is required. Such a trial faces numerous logistical challenges
including the need to enroll thousands of patients, deliver the assigned crystalloid in a time-
sensitive manner, and collect detailed data on the exact amounts of crystalloid administered,
physiologic effects (e.g., serum chloride), and outcomes. We hypothesized that these
challenges could be met by assigning crystalloid selection (saline vs balanced crystalloids) at the
ICU level and using software tools built into the electronic health record (EHR) to automatically
enroll patients, steer ordering providers to the assigned crystalloid, and collect data. To test
this hypothesis, we performed a four-month EHR-embedded(13, 14), cluster-randomized,

Page 7 of 74
multiple-crossover trial comparing balanced crystalloids with saline among critically ill adults.
Some of the results of this study have been previously reported in the form of an abstract(15).
METHODS
Study Design and Oversight
The SALT (isotonic Solution Administration Logistical Testing) study was a prospective,
open-label, cluster-randomized, multiple-crossover trial comparing the use of saline versus
balanced crystalloids among adults admitted to a tertiary medical ICU. The study protocol
(available in the online supplement) was approved by the institutional review board at
Vanderbilt University with waiver of informed consent (IRB#141349) and the trial was
registered online prior to initiation (NCT02345486).
Patient Population
From February 3, 2015 to May 31, 2015 all adults (age ≥ 18 years) admitted to the
medical ICU at Vanderbilt University were automatically enrolled at the time of ICU admission
(Figure 1). Enrolled patients who were discharged from the hospital were eligible again if they
were readmitted to the medical ICU during the study period.
Randomization and Allocation
The crystalloid used during the first month of the study (saline or balanced crystalloids)
was selected by computer-generated simple randomization. Thereafter, the crystalloid

Page 8 of 74
assigned to the ICU alternated monthly for the duration of the study, following a cluster-
randomized cluster-crossover design(16).
Study Treatments
Study protocol governed only the choice of intravenous isotonic crystalloid: 0.9%
sodium chloride (saline group) versus the treating clinician’s preference of Lactated Ringer’s
solution or Plasma-Lyte A® (balanced crystalloid group). Composition of the study crystalloids is
given in Table E1 in the online supplement. Decisions regarding frequency, rate, total volume
and additive content of the crystalloids were made by the treating clinician. No restrictions
were placed on other fluids or therapies.
Delivery of the assigned crystalloid to patients was achieved via interventions in
pharmacy supply and clinician order entry. Each month, the dispensing cabinets within the ICU
were stocked with 1000-mL bags of the assigned crystalloid. Additionally, any order for
intravenous crystalloid for a patient located in the ICU triggered an advisor application within
the electronic order entry system. The advisor application informed providers about the study,
asked about relative contraindications to the assigned crystalloid, and (if relative
contraindications were not present) guided providers to order the assigned crystalloid.
Accepted relative contraindications for patients assigned to balanced crystalloid included (1)
hyperkalemia and (2) “brain injury” with a co-existing contraindication to Plasma-Lyte A®. The
severity of hyperkalemia and “brain injury” at which saline was used in favor of balanced
crystalloids was determined by the treating clinician. The non-assigned crystalloid could also be

Page 9 of 74
provided by the pharmacy if a formal statement was submitted that the attending physician felt
the non-assigned crystalloid was required for the safe treatment of a specific patient.
The type of fluid administered before ICU admission, during procedures performed
outside the ICU, and after discharge from the ICU was not controlled by the study. Each day
patients received the crystalloid to which the ICU was currently assigned. Because it was
necessary that an intravenous crystalloid be clinically available at all times, there was no
washout period and patients who remained in the ICU through a crossover (i.e., from one
calendar month to another) were potentially exposed to both types of crystalloid. Patients,
clinicians, and investigators were not blinded to crystalloid assignment as available laboratory
values overtly reflected the crystalloid being used and prior studies showed high levels of
provider awareness of crystalloid assignment despite attempts at blinding(17).
Data Collection
This pragmatic trial used data collected as part of routine clinical care and electronically
extracted from the EHR(18). These data included: measures of pre-study renal function;
demographic characteristics, admitting location and diagnosis, and severity of illness at
enrollment; receipt of intravenous crystalloids, other fluids, and blood products; serum
electrolyte and creatinine values; receipt of RRT, mechanical ventilation, and vasopressors; and
vital status and serum creatinine at hospital discharge. A detailed description of how variables
were electronically collected and classified has been published previously(18) and is available in
the online supplement. For all patients who received new RRT, study personnel performed

Page 10 of 74
manual chart review to confirm the absence of prior RRT and identify indications for RRT
present at the time of RRT initiation.
Study Outcomes
As a pilot study(19), the primary outcome was the proportion of intravenous isotonic
crystalloid administered in the ICU that was saline. This was a continuous variable calculated
for each patient as the volume of saline received divided by volume of saline received plus
volume of balanced crystalloids received with a range from 0.0 (no saline received) to 1.0 (only
saline received). The secondary outcome was the proportion of patients meeting one or more
criteria for Major Adverse Kidney Events within the 30 days after enrollment (MAKE30): in-
hospital mortality, receipt of new RRT, or persistent renal dysfunction defined as a final
inpatient serum creatinine value ≥ 200% of baseline (see definitions of study variables in the
online supplement)(20, 21). Patients who had received RRT prior to enrollment were ineligible
to meet criteria for new RRT or persistent renal dysfunction, but could meet the MAKE30
endpoint if they experienced in-hospital mortality. Additional clinical outcomes included ICU
and in-hospital mortality, ICU-free days, ventilator-free days, vasopressor-free days, and RRT-
free days, all in the 28 days after enrollment. Additional renal outcomes included the incidence
of stage II or greater AKI by Kidney Disease Improving Global Outcomes (KDIGO) creatinine
criteria(22), highest serum creatinine value, change from baseline creatinine to highest
creatinine, and duration of new RRT as an inpatient. Biochemical efficacy and safety outcomes
included the first, highest, and lowest serum value for chloride, bicarbonate, and potassium
each day from enrollment until hospital discharge or 30 days.

Page 11 of 74
Statistical Analysis
The aims of this pilot study were to (1) test the logistics of the EHR-embedded, cluster-
randomized, multiple-crossover design and (2) examine the separation between study groups in
saline administration (19). To observe the logistics of at least one crossover from each
crystalloid assignment to the other while maintaining an equal number of saline and balanced
crystalloid treatment blocks required a minimum study duration of four months. With around
250 ICU admissions per month, we anticipated a total enrollment of around 1000 patients. Our
goal separation between groups was a 60% absolute difference in the percentage of isotonic
crystalloid administered as saline between the saline group and balanced crystalloid group.
Enrollment of 1000 patients would enable detection of a 60% absolute difference between
groups in the percentage of isotonic crystalloid administered as saline with a 95% confidence
interval of 58.5% to 61.5% (additional details available the online supplement).
All analyses were conducted at the level of the individual patient in an intention-to-treat
fashion unless otherwise specified. Continuous variables were reported as mean ± standard
deviation or median and interquartile range; categorical variables as frequencies and
proportions. Between-group comparisons were made with the Mann-Whitney rank sum test
for continuous variables, Fisher’s exact test or chi-square test for categorical variables,
generalized estimating equations for repeatedly measured variables, and multivariable
regression for adjusted analyses and tests of interaction. To facilitate interpretation of the
results, we present univariate comparisons of baseline characteristics, fluid receipt and
laboratory values, and clinical outcomes between the saline and balanced crystalloid groups

Page 12 of 74
among all patients enrolled (intention-to-treat) and all patients who received at least 250 mL of
intravenous crystalloid in the first 72 hours (modified intention-to-treat).
The primary analysis compared the proportion of intravenously administered isotonic
crystalloid that was saline between patients assigned to saline and balanced crystalloids. The
secondary analysis compared the proportion of patients experiencing MAKE30 in the saline and
balanced crystalloid groups, accounting for patients’ overall volume of isotonic crystalloid
received. For this analysis, we constructed a logistic regression model with MAKE30 as the
outcome and independent variables of study group, total isotonic crystalloid received between
enrollment and 30 days, and the interaction between the two (as a cross-product term). Other
analyses included (1) univariate comparison of additional clinical and renal outcomes between
study groups, (2) effect modification by severity of illness and pre-specified subgroups (source
of admission, primary diagnosis, mechanical ventilation, vasopressors, and baseline renal
function), and (3) sensitivity analyses excluding patients admitted in the week prior to a
crossover (‘washout’) and excluding patients who were transferred between ICUs or remained
in the ICU through a crossover (‘per protocol’).
Each patient’s baseline creatinine was considered to be the lowest serum creatinine
value between 12 months and 24 hours prior to hospital admission, when available. For
patients without an available serum creatinine value during this period, the lowest value
between 24 hours prior to hospital admission and enrollment was used. Patients without a
measured serum creatinine value between 12 months prior to hospital admission and
enrollment had baseline creatinine values estimated using a previously-described three-
variable formula(23). Multiple alternative approaches to missing baseline creatinine data were

Page 13 of 74
explored in sensitivity analyses including use of complete cases, multivariable single imputation,
and use of the first creatinine after enrollment or the highest or lowest creatinine during the
study (see the online supplement).
All secondary analyses were considered hypothesis-generating and no statistical
corrections were made for multiple comparisons. All analyses were performed using R version
3.2.0 (R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Enrollment and Baseline Characteristics
All 974 patients admitted to the medical ICU during the study period were enrolled
(Figure 1). Patients assigned to receive saline (n = 454) were similar at baseline to those
assigned to receive balanced crystalloids (n = 520) (Table 1, Tables E2 and E3 in the online
supplement). Patients’ median age was almost 60 years, approximately half were men, and the
majority were admitted from the emergency department. Sepsis and respiratory failure were
the most common admitting diagnoses, with a quarter of patients receiving vasopressors and
more than a third on mechanical ventilation.
Fluid Therapy
The volume of saline and balanced crystalloids received by patients in each group is
displayed in Figure 2. Saline comprised 91.2% of the isotonic crystalloid given in the ICU to
patients in the saline arm compared with 21.2% in the balanced arm (mean proportion of total
isotonic crystalloid: 0.91 vs 0.21; absolute difference 0.70; 95% CI 0.66 – 0.74; P < 0.001).
10

Page 14 of 74
Patients in the saline and balanced groups received a similar total volume of intravenous
crystalloid by 7 days (median [IQR]: 1250 [390–3000] mL vs 1320 [435–3139] mL; P = 0.38) and
30 days (1424 [500–3377] mL vs 1617 [500–3628] mL; P = 0.40) (Table E4; Figure E1 in the
online supplement).
Among 854 orders for crystalloid during months assigned to balanced crystalloids, 788
(92.2%) were for balanced crystalloids. Saline was selected in 45 (5.3%) cases for hyperkalemia,
5 (0.6%) cases for “brain injury” and a contraindication to Plasma-Lyte A®, and 16 (1.9%) cases
because the attending physician felt saline was required for the safe treatment of the patient
(Table E5 in the online supplement). In months assigned to saline, 95.2% of orders were for
saline, while 4.8% were for balanced crystalloid because the attending physician felt balanced
crystalloid was required for safe treatment of the patient (Table E5 in the online supplement).
Receipt of non-study intravenous fluids and blood products did not differ between the saline
and balanced groups (Table E4 in the online supplement).
Electrolytes
Serum electrolyte values are given in Figure 3. The highest serum chloride between
enrollment and day 30 was greater in the saline group than in the balanced group (median 109
mmol/L [IQR 105 – 113 mmol/L] vs 108 mmol/L [104 – 112 mmol/L]; P = 0.03). Serum
potassium values greater than 5.0 mmol/L and less than 3.0 mmol/L appeared to be slightly
more common with balanced crystalloids, but sodium and bicarbonate levels did not differ
between groups (Table E6, Figure E2, and Figure E3 in the online supplement).
11

Page 15 of 74
Renal Function
Daily values for blood urea nitrogen and serum creatinine were similar between groups
(Figure 3). There were no differences in the highest serum creatinine value, change from
baseline to highest value, or final creatinine value before discharge or 30 days (Table 2). The
incidence of stage II or greater AKI by KDIGO creatinine criteria did not differ between groups.
Receipt of new RRT was similar between saline and balanced crystalloids (3.1% versus 4.6%, P =
0.22), with no difference in the indications for RRT (Table E7 in the online supplement).
MAKE30 Outcome
In the saline group, 112 of 454 patients (24.7%) experienced the MAKE30 outcome
compared with 128 of 520 patients (24.6%) in the balanced crystalloid group (P = 0.98) (Table 2,
Table E8 in the online supplement). Results were similar in (1) pre-specified subgroups (Figure
4), (2) modified intention-to-treat analysis of patients who received at least 250 mL of isotonic
crystalloid in the first 72 hours (Table E9 in the online supplement); (3) sensitivity analyses
addressing missing baseline serum creatinine values (n=120) using imputations, extreme
scenarios, or complete cases; (4) analyses excluding patients admitted in the week prior to a
cross-over (‘washout’) or who remained in the ICU through a cross-over (‘per-protocol’); and (5)
multivariable regression analyses accounting for pre-specified covariates (Table E10 in the
online supplement).
Outcomes by Volume of Crystalloid Received
While the proportion of patients who experienced MAKE30 was similar between groups
12

Page 16 of 74
overall, among patients who received larger volumes of isotonic crystalloid, those assigned to
the saline group experienced a higher incidence of MAKE30 (P value for interaction = 0.026)
(Figure 4). Among patients who received larger volumes of isotonic crystalloid, there appeared
to be greater separation between arms in serum chloride concentration and higher peak serum
creatinine, greater incidence of AKI, and more frequent receipt of new RRT for those patients
assigned to saline (Figure E4-8 in the online supplement).
DISCUSSION
Our findings demonstrate that a cluster-randomized, multiple-crossover trial using
software tools within the EHR to compare saline to balanced crystalloids among critically ill
adults can produce well-balanced study groups and separation in crystalloid receipt. Although
we observed no difference between groups in the overall incidence of AKI or major adverse
kidney events, among patients exposed to larger volumes of isotonic crystalloid, those assigned
to saline appeared to experience more major adverse kidney events. These findings have
important implications for future studies of crystalloid choice in critical illness and for the
application of bioinformatics in clinical trials.
Fluid resuscitation with intravenous saline has been shown to cause hyperchloremic
metabolic acidosis(24, 25) and might predispose to AKI via chloride-mediated renal
vasoconstriction(26, 27). Observational studies among critically ill adults have suggested higher
rates of AKI(3, 4), RRT(3, 5), and death(6, 7, 28) with use of saline compared with balanced
crystalloids. Inconsistency across studies(29) and the limitations of observational research have
prompted calls for prospective clinical trials(1).
13

Page 17 of 74
Our findings in the SALT trial add to those of previous observational studies(3, 6, 7) and
the only prior prospective trial, the 0.9% Saline vs Plasma-Lyte 148 (PL-148) for ICU fluid
Therapy (SPLIT) trial(17). SPLIT was a cluster-randomized, double-crossover trial enrolling
nearly all patients admitted to four ICUs. Both SALT and SPLIT were designed as pilot studies to
assist with the planning of definitive trials(30). Both enrolled broad populations of ICU
patients, delivered relatively small overall volumes of intravenous crystalloid (median 1.5L in
SALT vs 2.0L in SPLIT), and observed no overall difference between saline and balanced
crystalloids in rates of AKI, RRT, or death. Despite these similarities, the studies differ in
important ways. Compared to the largely post-operative population of SPLIT, patients in SALT
were admitted predominantly for sepsis or respiratory failure and experienced higher rates of
AKI (27.1% in SALT vs 9.9% in SPLIT) and in-hospital mortality (17.4% in SALT vs 8.0% in SPLIT).
The measures of physiologic effect available in SALT (e.g., daily chloride concentration) permit
clearer characterization of the proposed mechanistic pathway between crystalloid
administration and outcomes. Perhaps SALT’s most intriguing addition to the prior findings is
that the relationship between crystalloid choice and development of major adverse kidney
events differed based on the amount of crystalloid received, suggesting a “dose response”
relationship. Among patients who received small volumes of crystalloid, the incidence of
MAKE30 was similar between groups. In contrast, among patients exposed to larger volumes of
crystalloid, the incidence of MAKE30 was significantly higher in the saline arm. Although
mechanistically plausible, this finding relies on a relatively small number of patients using a
variable that emerges after randomization and should be considered hypothesis-generating.
Future studies may benefit from restricting enrollment to patients predicted to receive large
14

Page 18 of 74
volumes of crystalloid in the ICU (although such prediction may prove challenging) or carefully
accounting for heterogeny in crystalloid exposure during study design and analysis(31).
The current study has important limitations. It was designed as a pilot and was not
powered to detect small, but potentially meaningful, differences in patient outcomes.
Enrollment from a single medical ICU limits generalizability. The study population was
extremely broad and specific subgroups of patients whose underlying physiology might alter
their response to IV crystalloids may have been inadequately represented. Although the impact
of crystalloid composition on laboratory values precluded blinding(17), the study endpoints
were objective and the co-interventions received were similar between groups. Based on the
hypothesized mechanism of chloride-mediated acute kidney injury(26, 27), we evaluated
Lactated Ringer’s solution or Plasma-Lyte A® together as a single ‘balanced crystalloids’ group,
despite differences between these solutions in electrolyte concentration, buffers, and
osmolality. Importantly, treating clinicians in the study ICU chose Lactated Ringer’s for over
90% of the balanced crystalloid administered (Table E4), limiting the conclusions that can be
drawn from our study about Plasma-Lyte A®.
The most important potential concern is whether the difference in exposure to saline
and balanced crystalloids between groups was large enough to influence outcomes. Including
all ICU patients in the intention-to-treat analysis protected against selection bias in a design
where group assignment preceded enrollment, but resulted in (1) inclusion of some patients
who never received any intravenous crystalloid and (2) relatively small volumes of administered
crystalloid overall. It is worth noting, however, that similar average fluid volumes in the
Crystalloid vs Hydroxyethyl Starch Trial (CHEST) were sufficient to demonstrate a difference
15

Page 19 of 74
between groups in the incidence of AKI requiring RRT – albeit in a considerably larger study
population(32). Similarly, the differences between groups in serum chloride concentration and
hyperchloremia in our trial, although small overall, were comparable to those associated with
increased rates of acute kidney injury and RRT in prior observational studies(3, 33). Moreover,
the difference between groups in serum chloride concentration was greatest among patients
receiving large volumes of crystalloids – for whom rates of RRT and MAKE30 also appeared to
be higher in the saline group than the balanced group (Figure E4-8). Contamination (exposure
to the non-assigned crystalloid) is a related concern. This study did not collect or attempt to
control fluid given in the emergency department or operating room prior to ICU admission, and
future trials should address these important sources of fluid exposure. The multiple-crossover
design without a washout period guaranteed that some patients would be exposed to both
types of crystalloid. Because most IV fluid was given shortly after ICU admission, however, the
volume of non-assigned fluid administered as a result of the study structure was relatively small
(less than 125 mL per patient – Table E4), and sensitivity analysis excluding patients who
experienced a crossover did not change the findings in a statistically significant manner.
Despite enrolling a population with high rates of relative contraindications to the study
crystalloids (e.g., 30% incidence of hyperkalemia), almost 95% of provider orders were for the
assigned crystalloid and the separation between groups in the type of crystalloid received (70%)
exceeded our target separation (60%). Despite relatively low overall volumes of intravenous
fluid and some exposure to the non-assigned crystalloid, the separation between groups
appeared to be sufficient to suggest a difference in RRT and MAKE30 among patients who
received large volumes of crystalloid – a key finding for the planning of a definitive trial.
16

Page 20 of 74
Despite these limitations, our study also has important strengths. Group assignment
was random. Large size provides outstanding preliminary data on which to base a definitive
trial. Enrolling all patients admitted to the study ICU precludes selection bias and improves
generalizability. Enrollment immediately upon ICU admission, earlier than many trials of fluid
management in critical illness, captures a peak period of fluid administration and risk for AKI
development. Our study used MAKE30, a patient-centered outcome recommended for phase
III trials by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
workgroup on Clinical Trials in AKI(18, 20, 34). Finally, although prior studies have examined
informatics-based interventions(35–37) or leveraged the EHR to facilitate aspects of trial
conduct(37–39), to our knowledge this is the first large, critical care clinical trial to be
conducted entirely within the EHR. By successfully enrolling nearly 1,000 critically ill adults,
accurately delivering the assigned intervention in a time-sensitive manner, and collecting
detailed data on patient characteristics, fluid receipt, laboratory values, and clinical outcomes,
SALT demonstrates that the EHR-embedded clinical trial(13, 14, 40) is a powerful new tool for
generating evidence to guide clinical practice in critical care.
In summary, an EHR-embedded, cluster-randomized, multiple-crossover trial comparing
saline and balanced crystalloids among critically ill adults can achieve well-balanced study
groups and separation in crystalloid receipt. Larger trials examining patient-centered outcomes
are warranted.
17

Page 21 of 74
ACKNOWLEDGEMENTS
This trial was conducted within the Vanderbilt Learning Healthcare System. The authors
would like to thank the patients, nurses, nurse practitioners, residents, fellows, and attending
physicians of the Vanderbilt Medical Intensive Care Unit for making this study possible. In
particular, we recognize the mentorship of Arthur P. Wheeler, MD of Vanderbilt University.
18

Page 22 of 74
REFERENCES
1. Myburgh JA, Mythen MG. Resuscitation fluids. N Engl J Med 2013;369:1243–1251.
2. Finfer S, Liu B, Taylor C, Bellomo R, Billot L, Cook D, Du B, McArthur C, Myburgh J,
SAFE TRIPS Investigators. Resuscitation fluid use in critically ill adults: an international
cross-sectional study in 391 intensive care units. Crit Care Lond Engl 2010;14:R185.
3. 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 J Am Med Assoc 2012;308:1566–1572.
4. Krajewski ML, Raghunathan K, Paluszkiewicz SM, Schermer CR, Shaw AD. Meta-analysis
of high- versus low-chloride content in perioperative and critical care fluid resuscitation. Br J
Surg 2015;102:24–36.
5. Shaw AD, Bagshaw SM, Goldstein SL, Scherer LA, Duan M, Schermer CR, Kellum JA.
Major complications, mortality, and resource utilization after open abdominal surgery: 0.9%
saline compared to Plasma-Lyte. Ann Surg 2012;255:821–829.
6. Raghunathan K, Shaw A, Nathanson B, Stürmer T, Brookhart A, Stefan MS, Setoguchi S,
Beadles C, Lindenauer PK. Association Between the Choice of IV Crystalloid and In-
Hospital Mortality Among Critically Ill Adults With Sepsis. Crit Care Med
2014;doi:10.1097/CCM.0000000000000305.
7. Shaw AD, Raghunathan K, Peyerl FW, Munson SH, Paluszkiewicz SM, Schermer CR.
Association between intravenous chloride load during resuscitation and in-hospital mortality
among patients with SIRS. Intensive Care Med 2014;40:1897–1905.
8. Iwashyna TJ, Burke JF, Sussman JB, Prescott HC, Hayward RA, Angus DC. Implications of
Heterogeneity of Treatment Effect for Reporting and Analysis of Randomized Trials in
19

Page 23 of 74
Critical Care. Am J Respir Crit Care Med 2015;192:1045–1051.
9. Prescott HC, Calfee CS, Thompson BT, Angus DC, Liu VX. Toward Smarter Lumping and
Smarter Splitting: Rethinking Strategies for Sepsis and Acute Respiratory Distress Syndrome
Clinical Trial Design. Am J Respir Crit Care Med 2016;194:147–155.
10. Famous KR, Delucchi K, Ware LB, Kangelaris KN, Liu KD, Thompson BT, Calfee CS.
ARDS Subphenotypes Respond Differently to Randomized Fluid Management Strategy. Am
J Respir Crit Care Med 2016;doi:10.1164/rccm.201603-0645OC.
11. Semler MW, Wheeler AP, Bernard GR, Thompson BT. Conservative Fluid Management
Decreases Mortality in Acute Respiratory Distress Syndrome Patients with Low Central
Venous Pressure. Am J Respir Crit Care Med 2014;189:A5094.
12. Semler MW, Marney AM, Rice TW, Nian H, Yu C, Wheeler AP, Brown NJ, NIH NHLBI
ARDS Network. B-Type Natriuretic Peptide, Aldosterone, and Fluid Management in ARDS.
Chest 2016;150:102–111.
13. Semler MW, Rice TW, Ehrenfeld JM. Leveraging Clinical Informatics in the Conduct of
Clinical Trials. J Med Syst 2015;39:112.
14. Angus DC. Fusing Randomized Trials With Big Data: The Key to Self-learning Health Care
Systems? JAMA 2015;314:767–768.
15. Semler MW, Noto MJ, Stollings J, Ehrenfeld JM, Wanderer JP, Plante M, Mulherin D,
Strawbridge S, Wheeler AP, Rice TW, The SALT Investigators and the Pragmatic Critical
Care Research Group. Effect Of Saline Versus Balanced Crystalloids On Major Adverse
Kidney Events In The Medical Intensive Care Unit: The Salt Randomized Trial. Am J Respir
Crit Care Med 2016;193:A4290.
16. Crespi CM. Improved Designs for Cluster Randomized Trials. Annu Rev Public Health
20

Page 24 of 74
2016;37:1–16.
17. Young P, Bailey M, Beasley R, Henderson S, Mackle D, McArthur C, McGuinness S,
Mehrtens J, Myburgh J, Psirides A, Reddy S, Bellomo R, SPLIT Investigators and the
ANZICS CTG. 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;1–10.doi:10.1001/jama.2015.12334.
18. Semler MW, Rice TW, Shaw AD, Siew ED, Self WH, Kumar AB, Byrne DW, Ehrenfeld
JM, Wanderer JP. Identification of Major Adverse Kidney Events Within the Electronic
Health Record. J Med Syst 2016;40:167.
19. Kistin C, Silverstein M. Pilot Studies: A Critical but Potentially Misused Component of
Interventional Research. JAMA 2015;314:1561–1562.
20. Palevsky PM, Molitoris BA, Okusa MD, Levin A, Waikar SS, Wald R, Chertow GM,
Murray PT, Parikh CR, Shaw AD, Go AS, Faubel SG, Kellum JA, Chinchilli VM, Liu KD,
Cheung AK, Weisbord SD, Chawla LS, Kaufman JS, Devarajan P, Toto RM, Hsu C, Greene
T, Mehta RL, Stokes JB, Thompson AM, Thompson BT, Westenfelder CS, Tumlin JA, et al.
Design of clinical trials in acute kidney injury: report from an NIDDK workshop on trial
methodology. Clin J Am Soc Nephrol CJASN 2012;7:844–850.
21. Kashani K, Al-Khafaji A, Ardiles T, Artigas A, Bagshaw SM, Bell M, Bihorac A, Birkhahn
R, Cely CM, Chawla LS, Davison DL, Feldkamp T, Forni LG, Gong MN, Gunnerson KJ,
Haase M, Hackett J, Honore PM, Hoste EAJ, Joannes-Boyau O, Joannidis M, Kim P,
Koyner JL, Laskowitz DT, Lissauer ME, Marx G, McCullough PA, Mullaney S, Ostermann
M, et al. Discovery and validation of cell cycle arrest biomarkers in human acute kidney
injury. Crit Care Lond Engl 2013;17:R25.
21

Page 25 of 74
22. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group.
KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney inter 2012;2(Suppl):1–
138.
23. Zavada J, Hoste E, Cartin-Ceba R, Calzavacca P, Gajic O, Clermont G, Bellomo R, Kellum
JA, for the AKI6 investigators. A comparison of three methods to estimate baseline
creatinine for RIFLE classification. Nephrol Dial Transplant 2010;25:3911–3918.
24. Scheingraber S, Rehm M, Sehmisch C, Finsterer U. Rapid saline infusion produces
hyperchloremic acidosis in patients undergoing gynecologic surgery. Anesthesiology
1999;90:1265–1270.
25. Waters JH, Gottlieb A, Schoenwald P, Popovich MJ, Sprung J, Nelson DR. Normal saline
versus lactated Ringer’s solution for intraoperative fluid management in patients undergoing
abdominal aortic aneurysm repair: an outcome study. Anesth Analg 2001;93:817–822.
26. Wilcox CS. Regulation of renal blood flow by plasma chloride. J Clin Invest 1983;71:726–
735.
27. 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;256:18–24.
28. Rochwerg B, Alhazzani W, Sindi A, Heels-Ansdell D, Thabane L, Fox-Robichaud A,
Mbuagbaw L, Szczeklik W, Alshamsi F, Altayyar S, Ip W-C, Li G, Wang M, Wludarczyk A,
Zhou Q, Guyatt GH, Cook DJ, Jaeschke R, Annane D. Fluid Resuscitation in Sepsis: A
Systematic Review and Network Meta-analysis. Ann Intern Med 2014;doi:10.7326/M14-
0178.
22

Page 26 of 74
29. Rochwerg B, Alhazzani W, Gibson A, Ribic CM, Sindi A, Heels-Ansdell D, Thabane L,
Fox-Robichaud A, Mbuagbaw L, Szczeklik W, Alshamsi F, Altayyar S, Ip W, Li G, Wang
M, Włudarczyk A, Zhou Q, Annane D, Cook DJ, Jaeschke R, Guyatt GH, FISSH Group
(Fluids in Sepsis and Septic Shock). Fluid type and the use of renal replacement therapy in
sepsis: a systematic review and network meta-analysis. Intensive Care Med
2015;doi:10.1007/s00134-015-3794-1.
30. Young P, Bailey M, Beasley R, Henderson S, Mackle D, McArthur C, Mehrtens J, Myburgh
J, McGuinness S, Psirides A, Reddy S, Bellomo R. The statistical analysis plan for the
0.9%Saline vs Plasma-Lyte 148 for Intensive care fluid Therapy (SPLIT) study. at
<http://wellingtonicu.com/Data/Trials /SPLITSAP.pdf.>.
31. Semler MW, Rice TW. Saline Is Not the First Choice for Crystalloid Resuscitation Fluids.
Crit Care Med 2016;44:1541–1544.
32. Myburgh JA, Finfer S, Bellomo R, Billot L, Cass A, Gattas D, Glass P, Lipman J, Liu B,
McArthur C, McGuinness S, Rajbhandari D, Taylor CB, Webb SAR, CHEST Investigators,
Australian and New Zealand Intensive Care Society Clinical Trials Group. Hydroxyethyl
starch or saline for fluid resuscitation in intensive care. N Engl J Med 2012;367:1901–1911.
33. Yunos NM, Kim IB, Bellomo R, Bailey M, Ho L, Story D, Gutteridge GA, Hart GK. The
biochemical effects of restricting chloride-rich fluids in intensive care. Crit Care Med
2011;39:2419–2424.
34. Weisbord SD, Gallagher M, Kaufman J, Cass A, Parikh CR, Chertow GM, Shunk KA,
McCullough PA, Fine MJ, Mor MK, Lew RA, Huang GD, Conner TA, Brophy MT, Lee J,
Soliva S, Palevsky PM. Prevention of contrast-induced AKI: a review of published trials and
the design of the prevention of serious adverse events following angiography (PRESERVE)
23

Page 27 of 74
trial. Clin J Am Soc Nephrol CJASN 2013;8:1618–1631.
35. Colpaert K, Hoste EA, Steurbaut K, Benoit D, Van Hoecke S, De Turck F, Decruyenaere J.
Impact of real-time electronic alerting of acute kidney injury on therapeutic intervention and
progression of RIFLE class. Crit Care Med 2012;40:1164–1170.
36. Herasevich V, Yilmaz M, Khan H, Hubmayr RD, Gajic O. Validation of an electronic
surveillance system for acute lung injury. Intensive Care Med 2009;35:1018–1023.
37. Semler MW, Weavind L, Hooper MH, Rice TW, Gowda SS, Nadas A, Song Y, Martin JB,
Bernard GR, Wheeler AP. An Electronic Tool for the Evaluation and Treatment of Sepsis in
the ICU: A Randomized Controlled Trial. Crit Care Med
2015;doi:10.1097/CCM.0000000000001020.
38. Hooper MH, Weavind L, Wheeler AP, Martin JB, Gowda SS, Semler MW, Hayes RM,
Albert DW, Deane NB, Nian H, Mathe JL, Nadas A, Sztipanovits J, Miller A, Bernard GR,
Rice TW. Randomized trial of automated, electronic monitoring to facilitate early detection
of sepsis in the intensive care unit*. Crit Care Med 2012;40:2096–2101.
39. Woodcock A, Bakerly ND, New JP, Gibson JM, Wu W, Vestbo J, Leather D. The Salford
Lung Study protocol: a pragmatic, randomised phase III real-world effectiveness trial in
asthma. BMC Pulm Med 2015;15:160.
40. Fiore LD, Lavori PW. Integrating Randomized Comparative Effectiveness Research with
Patient Care. N Engl J Med 2016;374:2152–2158.
41. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, Feldman HI, Kusek JW, Eggers
P, Van Lente F, Greene T, Coresh J, CKD-EPI (Chronic Kidney Disease Epidemiology
Collaboration). A new equation to estimate glomerular filtration rate. Ann Intern Med
2009;150:604–612.
24

Page 28 of 74
FIGURE LEGENDS
Figure 1. Flow of participants through the trial. All 974 patients admitted to the medical ICU
during the four-month study period were enrolled, assigned to a crystalloid group, followed
through hospital discharge, and included in the intention-to-treat analysis. A total of 51
patients assigned to saline remained in the ICU through a crossover to balanced crystalloids and
53 patients assigned to balanced crystalloids remained in the ICU through a crossover to saline.
Median days spent in the ICU after a crossover in crystalloid assignment were 0 [0-0] overall
and 3.0 [2.3-6.4] among the 104 patients who remained in the ICU through a crossover.
25

Page 29 of 74
Figure 2. Crystalloid receipt in the intensive care unit. For patients assigned to the saline
group (left) and balanced group (right), the cumulative volume of intravenous 0.9% sodium
chloride (diamonds) and balanced crystalloid (circles) is displayed over time. Cumulative
volume (mean and 95% confidence interval) includes fluids given in the intensive care unit as a
bolus, as maintenance, or to accompany medications. Day 0 is the day of study enrollment.
Fluid received before the time of enrollment on day 0 includes fluid received between hospital
admission and intensive care unit admission, but does not include fluid given before hospital
admission by the emergency medical system, emergency department, or transferring facility.
26

Page 30 of 74
Figure 3. Daily laboratory values by group. The first value each day is displayed as mean (95%
confidence interval) for each study group using locally weighted scatterplot smoothing. Values
on Day 0 represent the first available laboratory measurement after enrollment and may not
precede the receipt of the assigned crystalloid. A P value for the difference between groups in
the laboratory value overall (main effect) and the difference between groups in the change in
the laboratory value over time (interaction) were generated using generalized estimating
equations. The number of patients with laboratory values available declined from 882 on Day 1
to 311 on Day 7. Note that the highest and lowest serum values each day are available in
Figures E3 and E4 of the online supplement.
27

Page 31 of 74
Figure 4. Heterogeny of Treatment Effect. The incidence of Major Adverse Kidney Events
within 30 days is compared between patients assigned to the saline (red) and balanced (blue)
groups across the spectrum of exposure to the study crystalloids (upper left), baseline risk of
death (upper right), and pre-specified subgroups (below). Colored vertical bars display a
histogram of the proportion of patients in each group who received a given volume of
crystalloid (upper left) or possessed a given predicted in-hospital mortality (upper right). For
each pre-specified subgroup, the lower panel displays the unadjusted number and percentage
of patients in each study arm who experienced MAKE30, the unadjusted odds ratio for
experiencing MAKE30 with 95% confidence interval (CI), and the P values within the subgroup
and for the test of interaction. Normal kidney function at enrollment is defined as the absence
of acute kidney injury (AKI), chronic kidney disease (CKD), or renal replacement therapy (RRT)
prior to enrollment. AKI refers to patients without CKD whose first creatinine after enrollment
was at least 200% of the baseline value OR both (1) greater than 4.0mg/dL and (2) increased at
least 0.3 mg/dL from the baseline value(22). CKD refers to patents with a glomerular filtration
rate less than 60 ml/min per 1.73 m2 as calculated by the Chronic Kidney Disease Epidemiology
(CKD-EPI) Collaboration equation using the patient’s baseline creatinine value(41). RRT refers
to patients who received any form of renal replacement therapy prior to enrollment.
28

Page 32 of 74
TABLES
Table 1. Patient characteristics at baseline.
Saline Balanced
Patient Characteristics (n = 454) (n = 520)
Age, median [IQR], years 58 [46 – 70] 57 [44 – 68]
Men, No. (%) 246 (54.2) 268 (51.5)
Caucasian, No. (%) 358 (78.9) 376 (72.3)
Weight, median [IQR], kg 77.6 [63.0 – 95.3] 77.1 [63.5 – 95.3]
2
Body mass index, median [IQR], kg/m 26.4 [22.4 – 32.8] 26.7 [22.5 – 32.5]
Renal comorbidities, No. (%)
Chronic kidney disease, stage III or greater * 104 (22.9) 119 (22.9)
Prior renal replacement therapy receipt 44 (9.7) 42 (8.1)
Source of admission to ICU, No. (%)
Emergency department 256 (56.4) 310 (59.6)
Transfer from another hospital 93 (20.5) 106 (20.4)
Hospital ward 80 (17.6) 79 (15.2)
Another ICU within the hospital 15 (3.3) 14 (2.7)
Operating room 6 (1.3) 4 (0.8)
Outpatient 4 (0.9) 7 (1.3)
Admitting diagnosis, No. (%)
Sepsis or septic shock 130 (28.6) 130 (25.0)
Respiratory failure 53 (11.7) 41 (7.9)
Gastrointestinal bleeding 25 (5.5) 19 (3.7)
Liver failure 21 (4.6) 24 (4.6)
Ingestion 22 (4.8) 32 (6.2)
Malignancy 19 (4.2) 27 (5.2)
Diabetic ketoacidosis 20 (4.4) 21 (4.0)
Pneumonia 14 (3.1) 16 (3.1)
Acute kidney injury 5 (1.1) 18 (3.5)
Other 115 (25.3) 150 (28.8)
Mechanical ventilation, No. (%) 155 (34.1) 174 (33.5)
Vasopressors, No. (%) 111 (24.4) 114 (21.9)
UHC expected mortality, mean (95% CI), % † 14.7 (12.7 – 16.7) 13.1 (11.4 – 14.9)
Serum creatinine, median [IQR], mg/dL
Lowest in 12 months prior to hospitalization 0.78 [0.64 – 1.10] 0.76 [0.62 – 1.05]
No. (%) of patients 271 (59.7) 324 (62.3)
Lowest between hospitalization and ICU admission 0.97 [0.76 – 1.51] 0.95 [0.75 -1.61]
No. (%) of patients 122 (26.9) 137 (26.3)
Estimated by three-variable formula ‡ 0.91 [0.88 – 0.96] 0.91 [0.88 – 0.95]
No. (%) of patients 61 (13.4) 59 (11.3)
Study baseline 0.86 [0.69 – 1.12] 0.83 [0.67 – 1.09]
Acute kidney injury, stage II or greater § 87 (19.2) 96 (18.5)
Data are presented as median [25th percentile – 75th percentile] or number (percentage). Comparison of
the saline and balanced crystalloid groups using the Mann-Whitney rank sum test for continuous
29

Page 33 of 74
variables and the Fisher’s exact test or chi-square test for categorical variables found no difference
between groups in any baseline characteristic except Caucasian race (P = 0.02).
* Chronic kidney disease stage III or greater is defined as a glomerular filtration rate less than 60 ml/min
per 1.73 m2 as calculated by the Chronic Kidney Disease Epidemiology (CKD-EPI) Collaboration
equation(41) using the patient’s baseline creatinine value.
† University HealthSystem Consortium (UHC) expected mortality is an estimated probability of death
before hospital discharge generated for each patient based on age, gender, comorbidities, admission
source, race, and principal diagnosis (details at www.uhc.edu).
‡ Definitions for baseline serum creatinine are: “Lowest in 12 months prior to hospitalization” = lowest
available serum creatinine between 12 months and 24 hours prior to hospital admission; “Between
hospitalization and ICU admission” = lowest available serum creatinine between 24 hours prior to
hospital admission and ICU admission; “Estimated by three-variable formula” = creatinine value
calculated using a previously-described three-variable formula [creatinine (mg/dL) = 0.74 − 0.2 (if
female) + 0.08 (if African American) + 0.003 × age (in years)](23); and “Study baseline” = lowest in 12
months prior to hospitalization if available, otherwise between hospitalization and ICU admission, using
the estimated creatinine only for patients without an available creatinine between 12 months prior to
hospitalization and the time of ICU admission.
§ Acute kidney injury, stage II or greater is defined according to Kidney Disease Improving Global
Outcomes (KDIGO) creatinine criteria(22) as a first creatinine value after enrollment at least 200% of the
baseline value OR both (1) greater than 4.0mg/dL and (2) increased at least 0.3 mg/dL from the baseline
value.
30

Page 34 of 74
Table 2. Clinical Outcomes
Saline Balanced
Outcome n (n = 454) (n = 520) P Value
Secondary Outcome
Major Adverse Kidney Event within 30 days, No. (%)* 974 112 (24.7) 128 (24.6) 0.98
Additional Clinical Outcomes

In-hospital mortality, No. (%)
Before ICU discharge 974 44 (9.7) 45 (8.7) 0.57
Before 30 days 974 68 (15.0) 72 (13.8) 0.62
Before 60 days 974 83 (18.3) 87 (16.7) 0.53
ICU-free days, median [IQR] 969 25.1 [22.1 - 26.2] 25.2 [21.8 - 26.4] 0.58
Mean ± SD 21.0 ± 9.3 21.1 ± 9.1
Ventilator-free days, median [IQR] 974 28.0 [25.0 – 28.0] 28.0 [26.0 – 28.0] 0.85
Mean ± SD 22.9 ± 9.9 23.2 ± 9.6
Vasopressor-free days, median [IQR] 974 28.0 [27.0 – 28.0] 28.0 [27.0 – 28.0] 0.78
Mean ± SD 23.5 ± 9.9 23.9 ± 9.6
Renal replacement therapy-free days, median [IQR] 974 28.0 [28.0 – 28.0] 28.0 [28.0 – 28.0] 0.42
Mean ± SD 23.7 ± 10.0 24.0 ± 9.7
Additional Renal Outcomes

Serum creatinine, mg/dL
Highest before discharge or day 30, median [IQR], mg/dL 950 1.19 [0.81 - 2.30] 1.19 [0.80 - 2.62] 0.51
Change from baseline to highest value, median [IQR], mg/dL 950 0.07 [-0.10 - 0.50] 0.07 [-0.10 - 0.50] 0.65
Final value before discharge or 30 days, median [IQR], mg/dL 950 0.89 [0.69 – 1.54] 0.87 [0.70 -1.45] 0.90
Among survivors, median [IQR], mg/dL 808 0.85 [0.68 – 1.40] 0.82 [0.69 – 1.30] 0.48
Final creatinine > 200% baseline, No. (%) 974 59 (13.0) 76 (14.6) 0.47
Among survivors to hospital discharge 834 42 (10.9) 46 (10.3) 0.71
Among survivors to hospital discharge without new RRT 814 39 (10.2) 41 (9.5) 0.39
Acute kidney injury, stage II or greater, No. (%)† 974 129 (28.4) 135 (26.0) 0.39
Developing after enrollment 974 87 (19.2) 97 (18.7) 0.84
Receipt of new renal replacement therapy, No. (%) 974 14 (3.1) 24 (4.6) 0.22
Duration of in-hospital receipt, median [IQR], days 38 5.5 [3.0 – 8.2] 3.0 [0.5 – 4.5] 0.04
Continued receipt after hospital discharge, No. (%) 38 2 (0.4) 5 (1.0) 0.68
Data are presented as median [25th percentile – 75th percentile] or number (percentage). ICU-free,
ventilator-free, vasopressor-free, and renal replacement therapy free days refer to days alive and free
from the specified therapy in the first 28 days after enrollment.
* Major Adverse Kidney Events within 30 days (MAKE30) is the presence of any of the following before
discharge from the hospital or 30 days after enrollment: death, receipt of new renal replacement
31

Page 35 of 74
therapy, or a final serum creatinine value ≥ 200% of baseline.
† Acute kidney injury, stage II or greater is deﬁned according to Kidney Disease Improving Global
Outcomes (KDIGO) creatinine criteria(22) as any creatinine value between enrollment and discharge or
30 days at least 200% of the baseline value OR both (1) greater than 4.0mg/dL and (2) increased at least
0.3 mg/dL from the baseline value. This includes both acute kidney injury present at the time of
enrollment and developing after enrollment.
‡ Stage II or greater acute kidney injury developing aŌer enrollment is defined as any creatinine value
between enrollment and discharge or 30 days that is (1) increased at least 0.3 mg/dL from a preceding
post-enrollment value AND (2) at least 200% of the baseline value, at least 200% of a preceding post-
enrollment value, or at least 4.0 mg/dL.
32

974 patients admitted to the study ICU
Figure 1 of 74
Page 36
454 assigned to saline 520 assigned to balanced crystalloid

207 during February 257 during March
247 during April 263 during May
0 were excluded 0 were excluded
454 were included in 520 were included in
intention-to-treat analysis intention-to-treat
Copyright © 2016 by the American Thoracic Society analysis
Page 37 of 74 Saline Arm Balanced Arm
Figure 2
2000
2000
● Balanced fluid ● Balanced fluid
0.9% Saline 0.9% Saline
1500
1500
Cumulative volume (mL)
Cumulative volume (mL)

● ●
●
●
●
●
1000
1000
●
Before Before
500
500
enrollment enrollment ●
on day 0 on day 0
● ● ● ●
● ●
● ● ●
●
0
0
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Days after enrollment
Copyright © 2016 by the American Thoracic Society Days after enrollment
Page 38 of 74
Figure 3
Main effect = 0.300 Main effect = 0.058 Main effect = 0.311
36
Interaction = 0.744 Interaction = 0.084 Interaction = 0.176
Blood urea nitrogen (mg/dL)

Saline
139
106
34
Chloride (mmol/L)
Sodium (mmol/L)
Saline
32
105
138
Saline
Balanced
30
104
Balanced
137
28
Balanced
103
26
136
24
102
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Day Day Day
Main effect = 0.024 Main effect = 0.109 Main effect = 0.228
4.4
Interaction = 0.009 Interaction = 0.071 Interaction = 0.141

25
2.2
Balanced
Bicarbonate (mmol/L)
4.3
Potassium (mmol/L)
Creatinine (mg/dL)
24
2.0
4.2
Saline
23
1.8
Saline
4.1
Saline
22
1.6
4.0
Balanced
21
1.4
3.9
Balanced
3.8
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Day Day Day

Page 39 of 74
Figure 4
1.0
1.0
P−value for interaction = 0.026 Balanced
Incidence of MAKE30
Incidence of MAKE30
0.8
0.8
Saline
Saline
0.6
0.6
0.4
0.4
Balanced
0.2
0.2
P−value for interaction = 0.449
0.0
0.0
0 2000 4000 6000 8000 10000 0.0 0.2 0.4 0.6 0.8 1.0
Total isotonic crystalloid through day 30 (mL) Predicted in−hospital mortality
P Value for
Saline Balanced Odds Ratio (95% CI) P Value Interaction
Source of admission 0.647
Emergency Department 51/256(19.9) 62/310(20) 1.00(0.66−1.52) 0.982
Hospital ward 26/80(32.5) 25/79(31.6) 0.96(0.49−1.87) 0.908
Transfer from another hospital 30/93(32.3) 32/106(30.2) 0.91(0.50−1.66) 0.753
Other 5/25(20) 9/25(36) 2.25(0.63−8.06) 0.213
Primary diagnosis 0.045
Sepsis or septic shock 53/130(40.8) 36/130(27.7) 0.56(0.33−0.94) 0.027
Respiratory failure 7/53(13.2) 13/41(31.7) 3.05(1.09−8.56) 0.034
Diabetic ketoacidosis 4/20(20) 3/21(14.3) 0.67(0.13−3.44) 0.628
Gastrointestinal bleeding 3/25(12) 1/19(5.3) 0.41(0.04−4.26) 0.453
Liver failure 4/21(19) 9/24(37.5) 2.55(0.65−10.01) 0.180
Maligancy 6/19(31.6) 9/27(33.3) 1.08(0.31−3.80) 0.901
Other 34/185(18.4) 56/254(22) 1.26(0.78−2.02) 0.348
Receipt of mechanical ventilation 0.745
No 58/299(19.4) 65/346(18.8) 0.96(0.65−1.42) 0.844
Yes 54/155(34.8) 63/174(36.2) 1.06(0.68−1.67) 0.796
Receipt of vasopressors 0.030
No 57/343(16.6) 84/406(20.7) 1.31(0.90−1.90) 0.156
Yes 55/111(49.5) 44/114(38.6) 0.64(0.38−1.09) 0.099
Categories of kidney function 0.687
Normal 34/256(13.3) 37/286(12.9) 0.97(0.59−1.60) 0.906
AKI 31/48(64.6) 36/55(65.5) 1.04(0.46−2.34) 0.926
CKD 15/71(21.1) 28/92(30.4) 1.63(0.79−3.36) 0.183
RRT 18/44(40.9) 17/42(40.5) 0.98(0.42−2.32) 0.967
Overall 112/454(24.7) 128/520(24.6) 1.00(0.74−1.34) 0.984
0.1 0.3 1 5 10
Favors Balanced Favors Saline

Page 40 of 74
Balanced Crystalloids versus Saline in the Intensive Care Unit:

The SALT Randomized Trial
Matthew W. Semler, MD, MSc; Jonathan P. Wanderer, MD, MPhil; Jesse M. Ehrenfeld, MD,
MPH; Joanna L. Stollings, PharmD, BCPS; Wesley H. Self, MD, MPH; Edward D. Siew, MD,
MSc; Li Wang, MS; Daniel W. Byrne, MS; Andrew D. Shaw, MB, FRCA; Gordon R. Bernard,
MD; Todd W. Rice, MD, MSc for The SALT Investigators and the Pragmatic Critical Care
Research Group
Online Data Supplement
SALT TRIAL INVESTIGATORS
SUPPLEMENTAL METHODS
A. Definitions of Study Variables
B. Electronic Health Record-based Data Collection
C. Sample Size Determination, Difference between Groups, and Precision
D. Handling of Missing Baseline Creatinine
SUPPLEMENTAL TABLES
Table E1. Composition of the study fluids.
Table E2. Additional admitting diagnoses.
Table E3. Elixhauser comorbidity index.
Table E4. Intravenous fluids and blood products.
Table E5. Orders for study fluid placed through the computerized order entry system.
Table E6. Serum laboratory values.
Table E7. Indications for new renal replacement therapy.
Table E8. Outcomes of patients with and without incident acute kidney injury
Table E9. Modified intention-to-treat analysis.
Table E10. Sensitivity analyses.
SUPPLEMENTAL FIGURES
Figure E1. Crystalloid receipt in the hospital.
Figure E2. Highest serum electrolyte and creatinine values each day.
Figure E3. Lowest serum electrolyte and creatinine values each day
Figure E4. Serum chloride concentration relative to volume of crystalloid received
Figure E5. Serum creatinine concentration relative to volume of crystalloid received.
Figure E6. Development of acute kidney injury relative to volume of crystalloid received.
Figure E7. Renal replacement therapy relative to volume of crystalloid received
Figure E8. Odds of MAKE30 in the balanced crystalloid group compared with the saline group
relative to volume of crystalloid received.
SUPPLEMENTAL REFERENCES

Page 41 of 74
SALT TRIAL INVESTIGATORS

SALT (isotonic Solution Administration Logistical Testing) Trial Investigators include:
Vanderbilt University Medical Center, Nashville, TN – Gordon R. Bernard, Matthew W.
Semler, Michael J. Noto, Todd W. Rice (Division of Allergy, Pulmonary, and Critical Care
Medicine); Daniel W. Byrne, Henry J. Domenico, Li Wang (Department of Biostatistics);
Jonathan P. Wanderer, Jesse M. Ehrenfeld (Department of Biomedical Informatics and
Department of Anesthesiology); Andrew D. Shaw, Antonio Hernandez, Avinash B. Kumar
(Department of Anesthesiology); Wesley H. Self (Department of Emergency Medicine); Edward
D. Siew (Division of Nephrology and Hypertension, Vanderbilt Center for Kidney Disease
(VCKD) and Integrated Program for AKI (VIP-AKI)); Debra F. Dunlap, Joanna L. Stollings,
Mark Sullivan, Molly Knostman (Department of Pharmaceutical Services); David P. Mulherin,
Fred R. Hargrove (Department of Health Information Technology). Louisiana State University
School of Medicine, New Orleans, LA – David R. Janz (Section of Pulmonary/Critical Care
and Allergy/Immunology). American Society of Health-System Pharmacists, Bethesda, MD
– Seth Strawbridge (Clinical Informatics).
Names of writing committee members are italicized.

Page 42 of 74
SUPPLEMENTAL METHODS
A. Definitions of Study Variables
Fluids
Intravenous fluid – For this study, intravenous fluid was defined as the intravenous
administration of any volume at any rate of 0.9% sodium chloride, Lactated Ringer’s, Plasma-
Lyte A®; 0.45% sodium chloride, 0.225% sodium chloride, dextrose in water, 20% or 5%
human albumin solution, gelatins, dextrans, or hydroxyethyl starches. This included fluid given
as a bolus, fluid given as maintenance infusions, fluid given as flushes, fluid given as ‘piggy-
back’ infusions for IV medications, fluid given through pressure-bag systems, fluid given as a
part of thermodilution of pulmonary artery catheters, and fluid given to maintain the patency of
peripheral venous access. This excluded carrier fluid for medications and oral fluids.
Isotonic crystalloid – For the purposes of this study, the term isotonic crystalloid was used to
refer to any of 0.9% sodium chloride, Lactated Ringer’s, or Plasma-Lyte A®. Use of the term
isotonic crystalloid is intended to distinguish these three fluids from colloid solutions and from
significantly hypotonic (0.45% sodium chloride) or hypertonic (3% sodium chloride) crystalloid
solutions, rather than to imply that the tonicity of 0.9% sodium chloride, Lactated Ringer’s, or
Plasma-Lyte A® are precisely comparable to extracellular fluid.
Saline – Saline refers to 0.9% sodium chloride.
Balanced Crystalloid – For this study, Lactated Ringer’s or Plasma-Lyte A® were defined as
balanced crystalloids.
Renal Function
Baseline serum creatinine – The value for baseline serum creatinine was determined in a
hierarchical approach. The lowest serum creatinine between 12 months and 24 h prior to hospital
admission was used when available. If no such creatinine value was available, the lowest
creatinine value between 24 h prior to hospital admission and the time of ICU admission was

Page 43 of 74
used. If no creatinine value was available between 12 months prior to hospital admission and the
time of ICU admission, a baseline creatinine value was estimated using a previously-described
three-variable formula [creatinine = 0.74 − 0.2 (if female) + 0.08 (if African American) + 0.003
× age (in years)](1).
Acute kidney injury, stage II or greater – Stage II or greater acute kidney injury was defined
according to Kidney Disease Improving Global Outcomes (KDIGO) creatinine criteria(2) as a
creatinine value between enrollment and the first of hospital discharge or 30 days at least 200%
of the baseline value OR both (1) greater than 4.0mg/dL and (2) increased at least 0.3 mg/dL
from the baseline value. Patients may have had acute kidney injury present at the time of first
creatinine measurement after enrollment (prevalent AKI) or acute kidney injury developing
during the study (incident AKI). Incident AKI was defined as any creatinine value between
enrollment and discharge or 30 days that was (1) increased at least 0.3 mg/dL from a preceding
post-enrollment value AND (2) at least 200% of the baseline value, at least 200% of a preceding
post-enrollment value, or at least 4.0 mg/dL.
Chronic kidney disease stage III or greater – Chronic kidney disease stage III or greater was
defined as a glomerular filtration rate less than 60 ml/min per 1.73 m2 as calculated by the
Chronic Kidney Disease Epidemiology (CKD-EPI) Collaboration equation(3) using the patient’s
baseline creatinine value.
Outcomes
Major Adverse Kidney Events within 30 days (MAKE30). The MAKE30 composite outcome
was considered to have occurred when patients met one or more of the following criteria in the
30 days after enrollment: in-hospital mortality, receipt of new renal replacement therapy (RRT),
or persistent renal dysfunction(4–7). Patients who had received RRT prior to enrollment were
ineligible to meet the new RRT or persistent renal dysfunction criteria but remained eligible to
meet criteria for in-hospital mortality.

Page 44 of 74
In-hospital mortality – In-hospital mortality was defined as death from any cause prior to
hospital discharge censored at 30 days after ICU admission.
Receipt of new renal replacement therapy – Receipt of new RRT was defined as receipt of any
modality of RRT between ICU admission and the first of (1) hospital discharge or (2) 30 days in
a patient not known to have received RRT prior to ICU admission.
Persistent renal dysfunction – Persistent renal dysfunction was defined as a final serum
creatinine value before hospital discharge (censored at 30 days after enrollment) that was ≥
200% of the baseline creatinine value.
ICU-free days – Intensive care unit-free days to day 28 (ICU-free days) was defined as the
number of days from the time of the patient’s physical transfer out of the ICU until day 28 after
enrollment. Patients who died prior to day 28 after enrollment received a value of 0 for ICU-free
days. Patients who were never transferred out of the ICU prior to day 28 after enrollment
received a value of 0 for ICU-free days. Patients who were transferred out of the ICU, returned
to the ICU, and were not subsequently transferred out of the ICU again before day 28 after
enrollment received a value of 0 for ICU-free days. For patients who were transferred out of the
ICU, were readmitted to the ICU, and were subsequently transferred out of the ICU again prior
to day 28 after enrollment, ICU-free days were awarded based on the time of the final transfer
out of the ICU prior to day 28 after enrollment.
Ventilator-free days – Ventilator-free days to day 28 (VFDs) was defined as the number of days
from the time of initiating unassisted breathing until day 28 after enrollment. Patients who died
prior to day 28 after enrollment received a value of 0 for VFDs. Patients who never achieved
unassisted breathing prior to day 28 after enrollment received a value of 0 for VFDs. Patients
who achieved unassisted breathing, returned to assisted breathing, and did not again achieve
unassisted breathing before day 28 after enrollment received a value of 0 for VFDs. For patients
who achieved unassisted breathing, returned to assisted breathing, and subsequently achieved
assisted breathing again prior to day 28 after enrollment, VFDs were awarded based on the time
of the final initiation of unassisted breathing prior to day 28 after enrollment. Survivors who

Page 45 of 74
never experienced assisted breathing received 28 VFDs.
Vasopressor-free days – Vasopressor-free days to day 28 was defined as the number of days
from the time of vasopressor cessation until day 28 after enrollment. Patients who died prior to
day 28 after enrollment received a value of 0 for vasopressor-free days. Patients who never
ceased to receive vasopressors prior to day 28 after enrollment received a value of 0 for
vasopressor-free days. Patients who achieved vasopressor cessation, returned to receiving
vasopressors, and did not again achieve vasopressor cessation before day 28 after enrollment
received a value of 0 for vasopressor-free days. For patients who achieved vasopressor
cessation, returned to receiving vasopressors, and subsequently achieved cessation of
vasopressors again prior to day 28 after enrollment, vasopressor-free days were awarded based
on the time of the final cessation of vasopressors prior to day 28 after enrollment. Survivors who
never received vasopressors received 28 vasopressor-free days.
Renal replacement therapy-free days – Renal replacement therapy-free days to day 28 (RRT-
free days) was defined as the number of days from the time the final RRT treatment until day 28
after enrollment. Patients who died prior to day 28 after enrollment received a value of 0 for
RRT-free days. Patients who continued to received RRT through day 28 after enrollment
received a value of 0 for RRT-free days. Patients who achieved RRT cessation, returned to
receiving RRT, and did not again achieve RRT cessation before day 28 after enrollment received
a value of 0 for RRT-free days. For patients who achieved RRT cessation, returned to receiving
RRT, and subsequently achieved cessation of RRT again prior to day 28 after enrollment, RRT-
free days were awarded based on the time of the final RRT treatment prior to day 28 after
enrollment. Survivors who never received RRT received 28 RRT-free days.

Page 46 of 74
B. Electronic Health Record-based Data Collection
Electronically-extracted data – Structured data from the study institution’s enterprise EHR was
exported daily to an Enterprise Data Warehouse (EDW), along with data from the patient
registration, billing, and laboratory clinical information systems. Patient identifiers (medical
record number and encounter number) and a timestamp for study enrollment (date and time of
first ICU admission during the hospitalization) were used to extract the pre- and post-enrollment
data elements below(7).
Collection of baseline creatinine – Using all inpatient, outpatient, and emergency department
creatinine values from our institutional laboratory clinical information system, we determined (1)
the lowest serum creatinine value between 12 months and 24 h prior to hospital admission, (2)
the lowest creatinine value between 24 h prior to hospital admission and the time of ICU
admission, and (3) an estimated baseline creatinine value using a previously-described three-
variable formula [creatinine =0.74− 0.2 (if female) + 0.08 (if African American) + 0.003 × age
(in years)]. A baseline creatinine value for each patient was determined using the hierarchical
approach described above.
Collection of demographic characteristics – Gender, age, race, height, weight and body mass
index were extracted directly from the MediPac patient registration system into the EDW.
Collection of admitting location – Admitting location was extracted directly from the MediPac
patient registration system.
Collection of admitting diagnosis – The primary diagnosis prompting ICU admission was
manually collected for a randomly-selected sample of 200 cases using two-physician review of
the medical record and a structured data-collection instrument(7). The physician-adjudicated
diagnoses for these references cases were linked to International Classification of Disease,
Clinical Modification (ICD-9) codes and extrapolated to the full dataset. As an example, patients
classified as having ‘Sepsis or septic shock’ included those with the following ICD-9 codes:
038.0, 038.1, 038.11, 038.12, 038.19, 038.2, 038.3, 038.4, 038.42, 038.43, 038.49, 038.8, 038.9,

Page 47 of 74
042.0, 47.9, 079.89, 482.42, 536.41, 569.61, 599.0, 646.63, 785.5, 785.52, 790.7, 995.91, 995.52,
995.4, 996.61, 996.62, 996.63, 996.64, 996.65, 996.66, 996.68, 996.69, 997.62, 999.32, 999.39.
Collection of severity of illness – Our institution participates in the University HealthSystem

Consortium and routinely obtains their All Hospitals Clinical Database, which provides an
estimated mortality for each inpatient encounter based on coded data for age, gender,
comorbidities, admission source, race, and principal diagnosis. These mortality estimates are
based on Medicare Severity-Diagnosis Related Groupings (details at www.uhc.edu). We
retrieved these mortality estimates via our EDW for each patient in our cohort.
Receipt of intravenous crystalloids – We created a list of intravenous crystalloid fluids by

manual review of all of the fluid administration types administered at our hospital. This list was
used to extract fluid administration data from our nursing flowsheet in Horizon Expert
Documentation, via our EDW. We retrieved data by matching against medical record number
and date of administration.
Receipt of other fluids – A list of other fluids administered was created, and these data were
extracted from our nursing flowsheets using the process described above.
Receipt of blood products – A list of blood products administered was created, and these data
were extracted from our nursing flowsheets using the process described above.
Serum laboratory values – We created a list of applicable labs by manual review of all
laboratory types which matched the values of interest, and extracted laboratory values from our
Cerner laboratory system via our EDW.
Receipt of renal replacement therapy – RRT was identified electronically by the presence of
any one of the following American Medical Association’s Current Procedural Terminology
(CPT) codes (3066 F, 4054 F, 4055 F, 90963, 90964, 90965, 90966, 90967, 90968, 90969,
90970, 90989, 90993, G0257, G8714, G8956, G9013, G9014, G9231, 90935, 90937, 90945,
90947, 90989, 90993, 90921, 90925, 90999) or International Classification of Disease, Clinical

Page 48 of 74
Modification (ICD) codes for ICD-9 (39.95, 54.98) and ICD-10 (5A1D00Z, 5A1D60Z,
3E1M39Z) in the patient registration system or billing system(7).
New receipt of renal replacement therapy – For all patients who were identified as receiving
RRT during the study period, a full text search of the pre-enrollment record was performed using
terms related to receipt of RRT to identify patients who had received RRT prior to enrollment at
an outside facility. Search terms included “renal replacement’, “RRT”, “CRRT”, “dialysis”,
“HD”, “PD”, “end-stage renal”, and “ESRD”. Patients who had not received RRT prior to
enrollment and received RRT between enrollment and hospital discharge, censored at 30 days,
were considered to have met the “new receipt of renal replacement therapy” component of the
MAKE30 endpoint(7).
Receipt of mechanical ventilation – Mechanical ventilation was determined by review of

Medipac technical billing data. The number of calendar days with billing for mechanical
ventilation was retrieved for each continuous period that each patient was admitted to an
intensive care unit.
Receipt of vasopressors – Administration of vasopressors was determined by review of

Medipac technical billing data. The number of calendar days with billing for vasopressors was
retrieved for each continuous period that each patient was admitted to an intensive care unit.
In-hospital mortality – In-hospital mortality was determined by searching for a mortality-

associated discharge disposition in our patient registration system after the date of enrollment.
Patients with a mortality-associated discharge disposition within 30 days of study enrollment
were considered to have met the mortality component of the MAKE30 endpoint(7).

Page 49 of 74
C. Sample Size Determination, Difference between Groups, and Precision
Sample size: The cluster-randomized, multiple-crossover design required us to select a study

duration, from which the anticipated total enrollment could be estimated. We sought to observe
the logistics of the EHR-embedded fluid delivery and provider compliance over each possible
crossover in fluid group assignment (i.e. saline to balanced crystalloids, balanced crystalloids to
saline). To observe each possible crossover in fluid group assignment and maintain an equal
number of saline and balanced crystalloid treatment blocks required a minimum study duration
of four months. The study ICU averages 250 admissions per month. Thus, we anticipated a total
sample size around 1000 patients.
Difference between groups: The primary outcome for the study was the proportion of
intravenous isotonic crystalloid administered in the ICU that was saline, a continuous variable
calculated for each patient as the volume of saline received divided by volume of saline received
plus volume of balanced crystalloids received with a range from 0.0 (no saline received) to 1.0
(only saline received). In the year prior to the study, the mean proportion of intravenous isotonic
crystalloid comprised of saline in the study ICU was 0.85; standard deviation 0.12; range 0.0 to
1.0 – i.e. 85% of the intravenous crystalloid given in the ICU was saline and 15% was balanced
crystalloid. While the ideal separation between groups from the study perspective would be for
all of the intravenous crystalloid in the saline group to be comprised of saline (mean proportion
comprised of saline: 1.0) and none of the intravenous crystalloid in the balanced crystalloid
group to be comprised of saline (mean proportion comprised of saline: 0.0), our study protocol
acknowledged that there were instances in which one fluid type might be relatively
contraindicated. Thus, we aimed for saline to comprise at least 80% (mean proportion comprised
of saline: 0.8) of the isotonic crystalloid administered in the ICU to the saline group compared
with less than 20% (mean proportion comprised of saline: 0.2) administered to the balanced
crystalloid group – an absolute difference between groups of 60% (difference in means of 0.6),
which we felt would be sufficient separation between groups to proceed with a larger trial.
Precision: Using a standard deviation in the proportion of administered intravenous crystalloid

comprised of saline of 0.12 (from the prior year’s data in the same ICU), we calculated that the
10

Page 50 of 74
anticipated sample size of 1000 patients (set by the required minimum study duration of four
months) would allow us to determine the exact difference between groups in the mean proportion
of intravenous crystalloid comprised of saline with a 95% confidence interval of 0.015 in each
direction. For example, 1,000 patients would provide adequate power to determine that an
observed difference between groups of 0.62 (95% CI 0.605 – 0.635) was significantly greater
than our target difference of 0.60, meeting criteria to move forward with a larger trial.
All sample size calculations were performed in nQuery Advisor® version 7.0 (Statistical
Solutions Ltd., Cork, Ireland).
11

Page 51 of 74
D. Handling of Missing Baseline Creatinine
For patients without a measured serum creatinine between 12 months prior to hospital
admission and enrollment, baseline creatinine value for the primary analysis was estimated using
a previously-described three-variable formula [creatinine = 0.74 − 0.2 (if female) + 0.08 (if
African American) + 0.003 × age (in years)](1). Multiple sensitivity analyses employed
alternative approaches to estimating missing baseline creatinine values:
1) A ‘complete cases’ analysis was performed in which patients without a measured
creatinine values between 12 months prior to hospital admission and enrollment were
excluded.
2) Missing baseline serum creatinine values were imputed by multivariable single
imputation using the R function aregImpute in Hmisc package with 5 imputations. The
imputation model included age, gender, race, group assignment, source of admission,
primary diagnosis, receipt of mechanical ventilation, vasopressor receipt, prior
hemodialysis, total fluids received in 30 days, UHC expected mortality, overall mortality,
new RRT received, minimum creatinine value, maximum creatinine value, and final
study creatinine value. Continuous variables were transformed via cubic splines with 3 to
5 knots.
3) Simple imputation was performed in which first serum creatinine value after enrollment
was used as the baseline creatinine.
4) Simple imputation was performed in which the highest serum creatinine value between
enrollment and 30 days was used as the baseline creatinine.
5) Simple imputation was performed in which the lowest serum creatinine value between
enrollment and 30 days was used as the baseline creatinine.
12

Page 52 of 74
SUPPLEMENTAL TABLES
Table E1. Composition of the study fluids.
Sodium Potassium Calcium Magnesium Chloride Acetate Lactate Gluconate Osmolarity

Plasma 135–145 4.5–5.0 2.2–2.6 0.8–1.0 94–111 1–2 275–295
0.9% saline 154 154 308
Lactated Ringer’s 130 4.0 2.7 109 28 273
Plasma-Lyte A® 140 5.0 3.0 98 27 23 294
All values are in mEq/L except calculated osmolarity, which is in mOsm/L. 0.9% saline is “Sodium Chloride
Injection, USP”, Lactated Ringer’s is “Lactated Ringer’s Injection, USP”, and Plasma-Lyte A® is “Multiple
Electrolyte Injection, Type 1, USP”, all from Baxter Healthcare Corporation in Deerfield, IL, USA.
13

Page 53 of 74
Table E2. Additional admitting diagnoses.
Saline Balanced Crystalloid

Admitting diagnosis, No. (%) (n = 454) (n = 520)
Sepsis or septic shock 130 (28.6) 130 (25.0)
Respiratory failure 53 (11.7) 41 (7.9)
Gastrointestinal bleeding 25 (5.5) 19 (3.7)
Liver failure 21 (4.6) 24 (4.6)
Ingestion 22 (4.8) 32 (6.2)
Malignancy 19 (4.2) 27 (5.2)
Diabetic ketoacidosis 20 (4.4) 21 (4.0)
Pneumonia 14 (3.1) 16 (3.1)
Congestive heart failure 6 (1.3) 12 (2.3)
Acute kidney injury 5 (1.1) 18 (3.5)
Pulmonary embolism 4 (0.9) 7 (1.3)
Atrial fibrillation 4 (0.9) 4 (0.8)
Pancreatitis 6 (1.3) 1 (0.2)
Stroke 2 (0.4) 5 (1.0)
Acute coronary syndrome 1 (0.2) 5 (1.0)
Hypotension 3 (0.7) 1 (0.2)
Hypertensive urgency 2 (0.4) 2 (0.4)
Other 117 (25.7) 155 (29.8)
The medical intensive care unit (MICU) at Vanderbilt University Medical Center is a 34-bed unit which admits over
3,000 patients annually. A complete list of all admitting diagnoses cared for in the study ICU over a recent three
year period has been published previously (e-Table 1 in the online supplement of the referenced article) (8).
Additional adult critical care services at the study institution include a 22-bed neurological and neurosurgical ICU, a
27-bed cardiovascular ICU, a 22-bed surgical ICU, a 31-bed trauma ICU, and a 9-bed burn ICU. As a result,
patients with burns, traumatic brain injury, and intracranial hemorrhage are not significantly represented in the
population of the current study.
14

Page 54 of 74
Table E3. Elixhauser comorbidity index.

Comorbidity, No. (%) (n = 454) (n = 520) P value
Congestive heart failure 96 (21.1) 117 (22.5) 0.61
Cardiac arrhythmias 182 (40.1) 179 (34.4) 0.07
Valvular disease 40 (8.8) 50 (9.6) 0.67
Pulmonary circulation disorders 54 (11.9) 67 (12.9) 0.64
Hypertension, uncomplicated 182 (40.0) 211 (40.6) 0.88
Hypertension, complicated 82 (18.1) 84 (16.2) 0.43
Paralysis 29 (6.4) 31 (6.0) 0.78
Other neurological disorders 128 (28.2) 154 (29.6) 0.63
Chronic pulmonary disease 142 (31.3) 182 (35.0) 0.22
Diabetes, uncomplicated 113 (24.9) 136 (26.2) 0.65
Diabetes, complicated 59 (13.0) 48 (9.2) 0.06
Hypothyroidism 70 (15.4) 81 (15.6) 0.95
Renal failure 99 (21.8) 98 (18.8) 0.25
Liver disease 83 (18.3) 105 (20.2) 0.45
Peptic ulcer disease excluding bleeding 6 (1.3) 4 (0.8) 0.39
Acquired immunodeficiency syndrome 11 (2.4) 9 (1.7) 0.45
Lymphoma 13 (2.9) 11 (2.1) 0.45
Metastatic cancer 47 (10.4) 32 (6.2) 0.02
Solid tumor without metastasis 54 (11.9) 63 (12.1) 0.92
Rheumatoid arthritis / collagen vascular disease 28 (0.06) 34 (6.5) 0.81
Coagulopathy 69 (15.2) 78 (15.0) 0.93
Obesity 62 (13.7) 81 (15.6) 0.40
Weight loss 77 (17.0) 96 (18.5) 0.54
Fluid and electrolyte disorders 278 (61.2) 322 (61.9) 0.82
Blood loss anemia 9 (2.0) 9 (1.7) 0.77
Deficiency anemias 121 (26.7) 160 (30.8) 0.16
Alcohol abuse 39 (8.6) 55 (10.6) 0.29
Drug abuse 49 (10.8) 61 (11.7) 0.65
Psychoses 44 (9.7) 66 (12.7) 0.14
Depression 112 (24.7) 137 (26.3) 0.55
The Elixhauser Comorbidity Index is a method for measuring patient comorbidity based on the International
Classification of Diseases (ICD) diagnosis codes (ICD-9-CM and ICD-10) found in administrative data(17, 18).
15

Page 55 of 74
Table E4. Intravenous fluids and blood products.

(n = 454) (n = 520) P value
0.9% sodium chloride, median [IQR]; mean ± SD, mL
Prior to enrollment on Day 0 0 [0 – 0]; 143 ± 507 0 [0 – 0]; 114 ± 401 0.95
Cumulative volume from enrollment through day 3 1000 [150 – 2050]; 1489 ±1730 0 [0 – 175]; 351 ± 916 <0.001
Cumulative volume from enrollment through day 7 1070 [280 – 2440]; 1813 ± 2157 0 [0 – 545]; 687 ± 1967 <0.001
Cumulative volume from enrollment through day 14 1100 [300 – 2800]; 1982 ± 2529 0 [0 – 1000] 965 ± 2611 <0.001
Cumulative volume from enrollment through ICU transfer 1000 [200 – 2170]; 1669 ± 2038 0 [0 – 170]; 411 ± 1229 <0.001
Prior to an ICU crossover in fluid assignment 1000 [150 – 2170]; 1640 ± 2021 0 [0 – 100]; 318 ± 1104 <0.001
After an ICU crossover in fluid assignment 0 [0 – 0]; 28 ± 154 0 [0 – 0]; 93 ± 564 0.23
Cumulative volume from ICU transfer to hospital discharge 0 [0 – 40]; 476 ± 2073 0 [0 – 0]; 762 ± 3000 0.34
Receipt from enrollment through day 3, No. (%) 339 (74.7) 164 (31.5) <0.001
Receipt from enrollment through ICU transfer, No. (%) 339 (74.7) 169 (32.5) <0.001
Receipt from ICU transfer to hospital discharge, No. (%) 104 (22.9) 98 (18.8)) 0.14
Lactated Ringer’s, median [IQR]; mean ± SD, mL

Cumulative volume from enrollment through day 14 0 [0 – 0]; 315 ± 1144 1000 [45 -2122]; 1704 ± 2468 <0.001
Receipt from ICU transfer to hospital discharge, No. (%) 28 (6.2) 45 (8.7) 0.12
Plasma-Lyte A®, median [IQR]; mean ± SD, mL

Prior to enrollment on Day 0 0 [0 – 0]; 5 - 57 0 [0 – 0]; 9 -86 0.94
Cumulative volume from enrollment through day 3 0 [0 – 0]; 47 ± 405 0 [0 – 0]; 116 ± 781 0.007
Cumulative volume from enrollment through ICU transfer 0 [0 – 0]; 59 ± 489 0 [0 – 0]; 97 ± 612 0.007
Receipt from enrollment through day 3, No. (%) 12 (2.6) 32 (6.2) 0.007
16

Page 56 of 74
Receipt from enrollment through ICU transfer, No. (%) 12 (2.6) 32 (6.2) 0.007
Balanced crystalloid, median [IQR]; mean ± SD, mL

Cumulative volume from enrollment through day 14 0 [0 – 0]; 448 ± 1527 1000 [100 - 2348]; 1180 ± 2613 <0.001
Prior to an ICU crossover in fluid assignment 0 [0 – 0]; 166 ± 938 1000 [50 – 2110]; 1644 ± 2359 <0.001
After an ICU crossover in fluid assignment 0 [0 – 0]; 158 ± 1079 0 [0 – 0]; 24 ± 307 0.003
“Isotonic” crystalloid, median [IQR]; mean ± SD, mL

Prior to enrollment on Day 0 0 [0 – 0]; 244 ± 692 0 [0 -0]; 224 ± 777 0.88
“Hypotonic” crystalloid, median [IQR]; mean ± SD, mL

Prior to enrollment on Day 0 [0 – 0]; 23 ± 174 0 [0 – 0]; 16 ± 270 .13
Cumulative volume from enrollment through day 3 0 [0 – 50]; 471 ± 1132 0 [0 -135]; 449 ± 1090 0.75
Cumulative volume from enrollment through day 7 0 [0 -250]; 649 ± 1557 [0 – 425]; 574 ± 1349 0.64
17

Page 57 of 74
Human albumin solutions, median [IQR]; mean ± SD, mL

Blood products, median [IQR]; mean ± SD, mL

Prior to enrollment on Day 0 None None
Cumulative volume from ICU transfer to hospital discharge 0 [0 – 0]; 2.8 ± 55 0 [0 – 0]; 13 ± 205 0.91
Total intravenous fluid, median [IQR]; mean ± SD, mL

Cumulative volume from enrollment through ICU transfer 1625 [500 – 3610]; 2692 ± 3527 1580 [528 – 3465] 2702 ± 3425 0.93
Cumulative volume from ICU transfer to hospital discharge 0 [0 – 300]; 1072 ± 4007 0 [0 -265]; 1294 ± 4074 0.66
18

Page 58 of 74
Day 0 is the day of enrollment. Fluid received before the time of enrollment on day 0 includes fluid received
between hospital admission and intensive care unit admission, but does not include fluid given before hospital
admission by the transferring facility, emergency medical system, or emergency department. During the study
period, 80.5% of all intravenous isotonic crystalloid ordered in the emergency department of the study institution
was saline and 19.5% was balanced crystalloid. Cumulative volume of each fluid from enrollment through days 3,
7, 14, and 30 includes fluid administered both in the intensive care unit (ICU) and after transfer out of the ICU.
Cumulative volume includes fluids given as a bolus, as maintenance infusions, as flushes, as ‘piggy-back’ infusions
for IV medications, through pressure-bag systems, as a part of thermodilution of pulmonary artery catheters, and to
maintain the patency of peripheral venous access. Balanced crystalloid includes Lactated Ringer’s and Plasma-Lyte
A®; “Isotonic Crystalloid” includes 0.9% sodium chloride, Lactated Ringer’s, and Plasma-Lyte A®; “Hypotonic
Crystalloid” includes 0.45% sodium chloride, 0.225% sodium chloride, and dextrose in water; Human albumin
solutions include 20% and 5% albumin; Blood products include packed red blood cells, platelets, and fresh frozen
plasma; total intravenous fluid includes all of the above listed fluids but excludes fluid volume from medications.
No patient in either arm received any volume of semisynthetic colloid (gelatins, dextrans, or hydroxyethyl starches).
19

Page 59 of 74
Table E5. Orders for study fluid placed through the computerized order entry system.
February* March April May

Fluid group assignment Saline Balanced Saline Balanced
Total crystalloid orders, No. 193 375 395 479
Orders for assigned fluid, No. (%) 188 (97.4) 351 (93.6) 372 (94.2) 437 (91.2)
Orders for non-assigned fluid, No. (%) 5 (2.6) 24 (6.4) 23 (5.8) 42 (8.8)
Hyperkalemia, No. -- 15 -- 30
Brain injury, No. -- 3 -- 2
Attending request, No. 5 6 23 10
Reason for attending request for non-assigned fluid

4 orders for balanced crystalloid for a patient with a serum potassium of 5.9 mEq/L, chloride of 117 mEq/L,
and bicarbonate of 15 mEq/L in the context of rhabdomyolysis
February*
1 order for balanced crystalloid for a patient with a serum chloride of 108 mEq/L and bicarbonate of 18
mEq/L in the context of rhabdomyolysis from methamphetamine use
2 orders for saline for a patient with hydrocephalus and increased intracranial pressure from
ventriculoperitoneal shunt malfunction
1 order for saline for a patient with metastatic choriocarcinoma to accompany the administration of cisplatin
and etoposide
March 1 order for saline for a patient with a serum potassium of 6.8 mEq/L in the context of diabetic ketoacidosis
1 order for saline for a patient with a hypochloremic metabolic alkalosis, serum chloride 95 mEq/L and
bicarbonate 33 mEq/L
1 order for saline for a patient with a potassium of 5.2 mEq/L in the context of acute kidney injury requiring
hemodialysis
11 orders for balanced crystalloid for a patient with a serum sodium of 159 mEq/L, chloride of 124 mEq/L,
and bicarbonate of 10 mEq/L
6 orders for balanced crystalloid for a patient with a serum chloride of 111 mEq/L and bicarbonate of 5
mEq/L in the context of diabetic ketoacidosis
2 orders for balanced crystalloid for a patient with a serum chloride of 111 mEq/L and bicarbonate of 13
mEq/L in the context of acute kidney injury requiring hemodialysis
April
2 orders for balanced crystalloid for a patient with a serum bicarbonate of 8 mEq/L in the context of small
bowel ischemia
1 order for balanced crystalloid for a patient with a serum sodium of 153 mEq/L, chloride of 121 mEq/L,
and bicarbonate of 17 mEq/L
1 order for balanced crystalloid for a patient with a serum chloride of 115 mEq/L and bicarbonate of 12
mEq/L in the context of a perforated duodenal ulcer
2 orders for saline for a patient with a serum potassium of 6.6 mEq/L in the context of acute kidney injury
2 orders for saline for a patient with a serum sodium of 109 mEq/L and altered mental status
1 order for saline for a patient with a serum sodium of 120 mEq/L and seizure
1 order for saline for a patient with a serum sodium of 118 mEq/L and altered mental status
1 order for saline for a patient with a serum sodium of 116 mEq/L and altered mental status
May
1 order for saline for a patient with a serum sodium of less than 100 mEq/L
1 order for saline for a patient with a serum potassium of 7.4 mEq/L accompanied by a heart rate of 20
beats per minute requiring transcutaneous pacing
1 order for saline for a patient with adenocarcinoma of unknown primary to accompany the administration
of cisplatin and gemcitabine
* Data collection on fluid orders did not begin until February 13, 2015.
20

Page 60 of 74
Table E6. Serum laboratory values.

Laboratory value (n = 454) (n = 520) P value
Serum sodium, mmol/L
Highest between enrollment and day 30, median [IQR] 141 [139 – 144] 141 [138 – 144] 0.12
Lowest between enrollment and day 30, median [IQR] 136 [132 – 139] 135 [131 – 138] 0.06
>145 between enrollment and day 30, No. (%) 71 (15.6) 74 (14.2) 0.54
< 135 between enrollment and day 30, No. (%) 188 (41.4) 238 (45.8) 0.16
Serum potassium, mmol/L

Highest between enrollment and day 30, median [IQR] 4.6 [4.1 – 5.1] 4.7 [4.1 – 5.4] 0.13
Lowest between enrollment and day 30, median [IQR] 3.5 [3.1 – 3.8] 3.4 [3.1 – 3.9] 0.70
> 5.0 between enrollment and day 30, No. (%) 121 (26.7) 167 (32.1) 0.05
< 3.0 between enrollment and day 30, No. (%) 56 (12.3) 84 (16.2) 0.08
Serum chloride, mmol/L

Highest between enrollment and day 30, median [IQR] 109 [105 – 113] 108 [104 – 112] 0.03
Lowest between enrollment and day 30, median [IQR] 102 [97 – 106] 101 [96 – 105] 0.11
> 110 between enrollment and day 30, No. (%) 171 (37.7) 171 (32.9) 0.12
Serum bicarbonate, mmol/L

Highest between enrollment and day 30, median [IQR] 25.0 [22.7 – 29.0] 26.0 [23.0 – 29.0] 0.29
Lowest between enrollment and day 30, median [IQR] 19.0 [16.0 – 22.0] 19.0 [15.2 – 22.0] 0.78
> 30 between enrollment and day 30, No. (%) 93 (20.5) 108 (20.7) 0.91
21

Page 61 of 74
Table E7. Indications for new renal replacement therapy.
Balanced
Saline Crystalloid
Indication, No. (%) (n = 14) (n = 24) P Value
Oliguria 12 (85.7) 17 (70.8) 0.30
Hyperkalemia with serum potassium > 6.5 mEq/L 5 (35.7) 5 (20.8) 0.32
Acidemia with pH < 7.20 11 (78.6) 12 (50.0) 0.08
Blood urea nitrogen > 70 mg/dL 9 (64.3) 13 (54.2) 0.54
Serum creatinine > 3.39 mg/dL 11 (78.6) 17 (70.8) 0.60
Organ edema 4 (28.6) 8 (33.3) 0.76
Other renal failure–related indication 0 (0.0) 0 (0.0) > 0.99
Other non–renal failure–related indication 2 (14.3) 6 (25.0) 0.43
All potential indications for renal replacement therapy (RRT) present at the time of RRT initiation were identified
by manual chart review. Patients could have more than one indication for RRT. Oliguria is defined as urine output
less than 5ml/kg/hour for at least 6 hours(2). Organ edema was considered present if the clinical team or radiology
reports documented the presence of cerebral edema or pulmonary edema. Other non-renal failure-related indications
for renal replacement therapy included two patients with tumor lysis syndrome in the saline group and two patients
with rhabdomyolysis, one patient with tumor lysis syndrome, one patient with hepatic necrosis, one patient with
ethylene glycol ingestion, and one patient with lithium overdose in the balanced crystalloid group.
22

Page 62 of 74
Table E8. Outcomes of patients with and without incident acute kidney injury
Overall Saline Balanced

Patients with incident acute kidney injury (n = 184) (n = 87) (n = 97) P value
Major Adverse Kidney Event within 30 days, No. (%) 129 (70.0) 55 (63.2) 74 (76.3) 0.11
30-day in-hospital mortality, No. (%) 52 (28.2) 19 (21.8) 33 (34.0) 0.07
Receipt of new renal replacement therapy, No. (%) 38 (20.7) 14 (16.0) 24 (24.7) 0.15
Final creatinine > 200% baseline, No. (%) 95 (51.6) 44 (50.5) 51 (52.6) 0.79
Among survivors to hospital discharge 63/132 (47.7) 34/68 (50.0) 29/64 (45.3) 0.59
Patients without incident acute kidney injury (n = 790) (n = 367) (n = 423) P value
Major Adverse Kidney Event within 30 days, No. (%) 113 (14.3) 57 (15.5) 56 (13.2) 0.36
30-day in-hospital mortality, No. (%) 88 (11.1) 49 (13.4) 39 (9.2) 0.07
Receipt of new renal replacement therapy, No. (%) 0 (0.0) 0 (0.0) 0 (0.0) >0.99
Final creatinine > 200% baseline, No. (%)* 40 (5.0) 15 (4.1) 25 (5.9) 0.24
Among survivors to hospital discharge* 25/702 (3.6) 8/318 (2.5) 17/384 (4.4) 0.17
Data are presented as number (percentage). Incident acute kidney injury refers to stage II or greater acute kidney
injury by Kidney Disease Improving Global Outcomes (KDIGO) creatinine criteria developing after enrollment.
* Patients with acute kidney at the time of enrollment could experience persistent elevations in serum creatinine
concentration relative to their pre-hospital baseline, qualifying for the MAKE30 persistent renal dysfunction criteria
without experiencing incident acute kidney injury between enrollment and hospital discharge or 30 days.
23

Page 63 of 74
Table E9. Modified intention-to-treat analysis.
Saline Balanced
Baseline Characteristics (n = 292) (n = 339) P value
Age, median [IQR], years 58 [46 – 71] 57 [42 – 68] -
Men, No. (%) 166 (56.8) 176 (51.9) -
Caucasian, No. (%) 236 (80.8) 253 (74.6) -
Prior renal replacement therapy receipt, No. (%) 24 (8.2) 20 (6.8) -
Admitted from the emergency department, No. (%) 166 (56.8) 210 (61.9) -
Sepsis or septic shock, No. (%) 101 (34.7) 103 (30.5) -
Mechanical ventilation, No. (%) 102 (34.9) 122 (36.0) -
Vasopressors, No. (%) 86 (29.4) 97 (28.6) -
Baseline creatinine, median [IQR], mg/dL 0.83 [0.69 – 1.10] 0.81 [0.66 – 1.08] -
Fluid Receipt
0.9% sodium chloride, median [IQR], mL
Cumulative volume from enrollment through day 7 1662 [879 – 3111] 0 [0 – 1000] <0.001
Balanced crystalloid, median [IQR], mL
Serum chloride
Highest between enrollment and day 30, median [IQR], mmol/L 110 [106 – 114] 109 [105 – 113] 0.07
Lowest between enrollment and day 30, median [IQR], mmol/L 102 [97 – 106] 102 [97 – 105] 0.38
Clinical Outcomes
Major Adverse Kidney Event within 30 days, No. (%) 77 (26.4) 82 (24.2) 0.53
30-day in-hospital mortality, No. (%) 47 (16.1) 50 (14.7) 0.64
Receipt of new renal replacement therapy, No. (%) 9 (3.1) 14 (4.3) 0.48
Final creatinine > 200% baseline, No. (%) 40 (13.7) 38 (14.2) 0.87
Stage II or greater acute kidney injury by KDIGO criteria, No. (%) 88 (30.1) 87 (25.7) 0.21
Data are presented as median [25th percentile – 75th percentile] or number (percentage). Modified intention-to-treat
analysis includes the 631 patients who received at least 250 mL of intravenous crystalloid in the first 72 hours after
enrollment.
24

Page 64 of 74
Table E10. Sensitivity analyses.
Analysis n Odds Ratio 95% CI P Value

Intention-to-treat 974 1.00 0.74 – 1.34 0.98
Modified intention-to-treat 631 0.89 0.62 – 1.28 0.30
Excluding patients admitted in the week prior to a crossover in
483 1.06 0.70 – 1.62 0.77
fluid group assignment (‘washout’)
Excluding patients transferred between ICUs or who experienced
475 0.78 0.51 – 1.19 0.25
cross-over in fluid group assignment (‘per protocol’)
Alternative approaches to baseline serum creatinine
Only patients with a measured creatinine between 12 months
854 1.03 0.75 – 1.41 0.84
prior to hospital admission and enrollment (‘complete cases’)
Single imputation of baseline creatinine for the 120 patients
without a measured value between 12 months prior to hospital 974 0.99 0.74 – 1.33 0.95
admission and enrollment
Single imputation of baseline creatinine for the 379 patients
without a measured value between 12 months prior to hospital 974 1.01 0.76 – 1.34 0.97
admission and 24 hours prior to hospital admission
Using the first creatinine after enrollment as baseline for the 120
patients without a measured value between 12 months prior to 974 1.02 0.74 – 1.43 0.89
hospital admission and enrollment
Using the highest creatinine between enrollment and 30 days as
baseline for the 120 patients without a measured value between 974 1.05 0.75 – 1.48 0.76
12 months prior to hospital admission and enrollment
Using the lowest creatinine between enrollment and 30 days as
baseline for the 120 patients without a measured value between 974 0.91 0.66 – 1.25 0.56
12 months prior to hospital admission and enrollment
Alternative definitions of the MAKE30 composite endpoint
Death, new renal replacement therapy, or a final creatinine before
974 0.89 0.68 – 1.17 0.41
discharge or 30 days at least 1.5x baseline
Multivariable model 974 1.09 0.75 – 1.61 0.63
Odds of experiencing a Major Adverse Kidney Event within 30 days (MAKE30) are given for patients assigned to
the balanced crystalloid group compared with patients assigned to the saline group (referent). The intention-to-treat
analysis compares balanced crystalloids with saline among all patients enrolled in the trial. The modified intention-
to-treat analysis includes only patients who received at least 250 mL of any isotonic crystalloid between enrollment
and 72 hours after enrollment. The proportion of isotonic crystalloid given in the ICU that was comprised of the
assigned fluid was 81.9% in the intention-to-treat population, 82.1% in the modified intention-to-treat population,
85.7% in the ‘washout’ analysis excluding patients admitted in the week prior to a crossover, and 88.3% in the ‘per
protocol analysis’ excluding all patients who experienced a crossover or were transferred to a non-study ICU.
Multiple alternative approaches to assigning baseline serum creatinine for those without measured values are
presented. Single imputation of missing baseline serum creatinine utilized the following variables: age, gender,
race, group assignment, source of admission, primary diagnosis, mechanical ventilation, vasopressor receipt, receipt
of renal replacement therapy prior to enrollment, total volume of isotonic crystalloids received in the first 30 days,
University HealthSystem Consortium expected mortality, in-hospital 30-day mortality, new receipt of renal
replacement therapy, minimum serum creatinine between enrollment and 30 days, maximum serum creatinine
between enrollment and 30 days, and final serum creatinine between enrollment and 30 days. The multivariable
model compares the outcome of MAKE30 between study groups using generalized estimating equations to account
for time elapsed from the start of the trial and pre-specified baseline covariates of age, source of admission,
predicted in-hospital mortality, receipt of mechanical ventilation, and receipt of vasopressors.
25

Page 65 of 74
SUPPLEMENTAL FIGURES
Figure E1. Crystalloid receipt in the hospital. For patients assigned to the saline group (left)
and balanced group (right), the cumulative volume of intravenous 0.9% sodium chloride
(diamonds) and balanced crystalloid (circles) is displayed over time. Day 0 is the day of study
enrollment. Fluid received before the time of study enrollment on day 0 includes fluid received
after hospital admission before intensive care unit admission, but does not include fluid given
before hospital admission by the emergency medical system, emergency department, or
transferring facility. Cumulative volume (mean and 95% confidence interval) includes bolus
fluids, maintenance fluids, and fluid accompanying medication infusions. In contrast to Figure 2
in the main text which displays fluid given in the intensive care unit, this figure includes both
fluid given in the intensive care unit (during the study intervention) and fluid given after transfer
out of the intensive care unit before hospital discharge (after the study intervention).
26

Page 66 of 74
Figure E2. Highest serum electrolyte and creatinine values each day. The highest value
each day is displayed as mean (95% confidence interval) for each study group using locally
weighted scatterplot smoothing. The highest value after enrollment on Day 0 may not precede
receipt of the assigned crystalloid. A P value for the difference between groups in the laboratory
value overall (main effect) and the difference between groups in the change in the laboratory
value over time (interaction) were generated using generalized estimating equations. The
number of patients with laboratory values available declined from 882 on Day 1 to 311 on Day 7.
Note that the first serum value each day is available in Figure 3 of the main text and the lowest
serum value each day is available in Figure E3.
27

Page 67 of 74
Figure E3. Lowest serum electrolyte and creatinine values each day. The lowest value each
day is displayed as mean (95% confidence interval) for each study group using locally weighted
scatterplot smoothing. The lowest value after enrollment on Day 0 may not precede receipt of
the assigned crystalloid. A P value for the difference between groups in the laboratory value
overall (main effect) and the difference between groups in the change in the laboratory value
over time (interaction) were generated using generalized estimating equations. The number of
patients with laboratory values available declined from 882 on Day 1 to 311 on Day 7. Note that
the first serum value each day is available in Figure 3 of the main text and the lowest serum
value each day is available in Figure E2.
28

Page 68 of 74
Figure E4. Serum chloride concentration relative to volume of crystalloid received. The
highest serum chloride concentration between enrollment and day 30 is compared between
patients assigned to the saline (red) and balanced (blue) groups relative to the total volume of
isotonic crystalloid received between enrollment and Day 30. Colored vertical bars display a
histogram of the proportion of patients in each group who received a given volume of crystalloid.
29

Page 69 of 74
Figure E5. Serum creatinine concentration relative to volume of crystalloid received. The
highest serum creatinine concentration between enrollment and day 30 is compared between
30

Page 70 of 74
Figure E6. Development of acute kidney injury relative to volume of crystalloid received.
The proportion of patients developing stage II or greater acute kidney injury by Kidney Disease
Improving Global Outcomes (KDIGO) creatinine criteria between enrollment and day 30 is
compared between patients assigned to the saline (red) and balanced (blue) groups relative to the
total volume of isotonic crystalloid received between enrollment and Day 30. Colored vertical
bars display a histogram of the proportion of patients in each group who received a given volume
of crystalloid.
31

Page 71 of 74
Figure E7. Renal replacement therapy relative to volume of crystalloid received. The
proportion of patients receiving new renal replacement therapy (RRT) is compared between
32

Page 72 of 74
Figure E8. Odds of MAKE30 in the balanced crystalloid group compared with the saline
group relative to volume of crystalloid received. The odds of experiencing a Major Adverse
Kidney Event within 30 days (MAKE30) in the balanced crystalloid group compared with the
saline group (referent) are displayed relative to the total volume of isotonic crystalloid received
between enrollment and Day 30. The solid black line represents the odds ratio for balanced
crystalloids compared with saline, the dotted black lines represent the 95% confidence interval,
and the horizontal red line represents a reference odds ratio of 1.0 at which the odds of MAKE30
would be equal in both groups. The odds ratio for MAKE30 overall was 1.00 [95% CI 0.74 –
1.34] but among patients who received more than 2 liters of crystalloids the point estimate
favored balanced crystalloids and among patients who received more than 4 liters of crystalloid,
the 95% confidence interval appeared not to include 1.0.
33

Page 73 of 74
SUPPLEMENTAL REFERENCES
1. Závada J, Hoste E, Cartin-Ceba R, Calzavacca P, Gajic O, Clermont G, Bellomo R, Kellum
JA, AKI6 investigators. A comparison of three methods to estimate baseline creatinine for
RIFLE classification. Nephrol Dial Transplant Off Publ Eur Dial Transpl Assoc - Eur Ren
Assoc 2010;25:3911–3918.
2. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group.
KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney inter 2012;2(Suppl):1–
138.
3. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, Feldman HI, Kusek JW, Eggers
P, Van Lente F, Greene T, Coresh J, CKD-EPI (Chronic Kidney Disease Epidemiology
Collaboration). A new equation to estimate glomerular filtration rate. Ann Intern Med
2009;150:604–612.
4. Palevsky PM, Molitoris BA, Okusa MD, Levin A, Waikar SS, Wald R, Chertow GM, Murray
PT, Parikh CR, Shaw AD, Go AS, Faubel SG, Kellum JA, Chinchilli VM, Liu KD, Cheung
AK, Weisbord SD, Chawla LS, Kaufman JS, Devarajan P, Toto RM, Hsu C, Greene T, Mehta
RL, Stokes JB, Thompson AM, Thompson BT, Westenfelder CS, Tumlin JA, et al. Design of
clinical trials in acute kidney injury: report from an NIDDK workshop on trial methodology.
Clin J Am Soc Nephrol CJASN 2012;7:844–850.
5. Kashani K, Al-Khafaji A, Ardiles T, Artigas A, Bagshaw SM, Bell M, Bihorac A, Birkhahn
R, Cely CM, Chawla LS, Davison DL, Feldkamp T, Forni LG, Gong MN, Gunnerson KJ,
Haase M, Hackett J, Honore PM, Hoste EAJ, Joannes-Boyau O, Joannidis M, Kim P, Koyner
JL, Laskowitz DT, Lissauer ME, Marx G, McCullough PA, Mullaney S, Ostermann M, et al.
34

Page 74 of 74
Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Crit
Care Lond Engl 2013;17:R25.
6. Weisbord SD, Gallagher M, Kaufman J, Cass A, Parikh CR, Chertow GM, Shunk KA,
McCullough PA, Fine MJ, Mor MK, Lew RA, Huang GD, Conner TA, Brophy MT, Lee J,
Soliva S, Palevsky PM. Prevention of contrast-induced AKI: a review of published trials and
the design of the prevention of serious adverse events following angiography (PRESERVE)
trial. Clin J Am Soc Nephrol CJASN 2013;8:1618–1631.
7. Semler MW, Rice TW, Shaw AD, Siew ED, Self WH, Kumar AB, Byrne DW, Ehrenfeld JM,
Wanderer JP. Identification of Major Adverse Kidney Events Within the Electronic Health
Record. J Med Syst 2016;40:167.
8. Landsperger JS, Semler MW, Wang L, Byrne DW, Wheeler AP. Outcomes of Nurse
Practitioner-Delivered Critical Care: A Prospective Cohort Study. Chest 2016;149:1146–
1154.
9. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with
administrative data. Med Care 1998;36:8–27.
10. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi J-C, Saunders LD, Beck
CA, Feasby TE, Ghali WA. Coding algorithms for defining comorbidities in ICD-9-CM and
ICD-10 administrative data. Med Care 2005;43:1130–1139.
35

M Semler (SALT)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

M Semler (SALT)

Uploaded by

Copyright:

Available Formats

AJRCCM Articles in Press. Published on 17-October-2016 as 10.1164/rccm.

Care Research Group

Affiliations: 1Division of Allergy, Pulmonary, and Critical Care Medicine; 2Department of

Anesthesiology; 3Department of Biomedical Informatics; 4Department of Pharmaceutical

Services; 5Department of Emergency Medicine; 6Division of Nephrology and Hypertension,

Running Title: Balanced Crystalloids vs Saline in the medical ICU

1161 21st Ave S. C-1216 MCN, Nashville, TN 37232-2650

Phone: (615) 322-3412; Fax: (615) 343-7448

Copyright © 2016 by the American Thoracic Society

responsible for the data analysis.

analysis, interpretation, or presentation of the data, or (3) preparation, review, or approval of

Copyright © 2016 by the American Thoracic Society

Director of Medical Affairs for Cumberland Pharmaceuticals, Inc.

Subject Descriptor Number: 4.4 Clinical Trials in Critical Care Medicine

Manuscript Word Count (body only): 3,764

evaluation in a large, randomized clinical trial.

contents online at www.atsjournals.org

Copyright © 2016 by the American Thoracic Society

crystalloids rather than saline affects patient outcomes remains unknown.

OBJECTIVES: To pilot a cluster-randomized, multiple-crossover trial using software tools within

the electronic health record to compare saline to balanced crystalloids.

METHODS: Cluster-randomized, multiple-crossover trial among 974 adults admitted to a

persistent renal dysfunction.

Copyright © 2016 by the American Thoracic Society

CONCLUSIONS: An electronic health record-embedded, cluster-randomized, multiple-crossover

separation in crystalloid receipt.

TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT02345486

Abstract Word Count: 260

Copyright © 2016 by the American Thoracic Society

Intravenous crystalloid administration is ubiquitous in critical care, yet whether choice

To definitively evaluate the effect of balanced crystalloids versus saline on patient-

risk of AKI(8), volume of crystalloid received(7), and underlying pathophysiology(9–12) among

this hypothesis, we performed a four-month EHR-embedded(13, 14), cluster-randomized,

Copyright © 2016 by the American Thoracic Society

Study Design and Oversight

open-label, cluster-randomized, multiple-crossover trial comparing the use of saline versus

registered online prior to initiation (NCT02345486).

were readmitted to the medical ICU during the study period.

Randomization and Allocation

was selected by computer-generated simple randomization. Thereafter, the crystalloid

Copyright © 2016 by the American Thoracic Society

randomized cluster-crossover design(16).

solution or Plasma-Lyte A® (balanced crystalloid group). Composition of the study crystalloids is

were placed on other fluids or therapies.

Delivery of the assigned crystalloid to patients was achieved via interventions in

Copyright © 2016 by the American Thoracic Society

provider awareness of crystalloid assignment despite attempts at blinding(17).

demographic characteristics, admitting location and diagnosis, and severity of illness at

Copyright © 2016 by the American Thoracic Society

present at the time of RRT initiation.

each day from enrollment until hospital discharge or 30 days.

Copyright © 2016 by the American Thoracic Society

interval of 58.5% to 61.5% (additional details available the online supplement).

deviation or median and interquartile range; categorical variables as frequencies and

generalized estimating equations for repeatedly measured variables, and multivariable

results, we present univariate comparisons of baseline characteristics, fluid receipt and

Copyright © 2016 by the American Thoracic Society

intravenous crystalloid in the first 72 hours (modified intention-to-treat).

The primary analysis compared the proportion of intravenously administered isotonic

of admission, primary diagnosis, mechanical ventilation, vasopressors, and baseline renal

in the ICU through a crossover (‘per protocol’).

enrollment had baseline creatinine values estimated using a previously-described three-

Copyright © 2016 by the American Thoracic Society