Huet 

et al. Critical Care (2023) 27:42

https://doi.org/10.1186/s13054-023-04311-1

RESEARCH Open Access

Impact of continuous hypertonic (NaCl

20%) saline solution on renal outcomes
after traumatic brain injury (TBI): a post hoc
analysis of the COBI trial
Olivier Huet1*†, Xavier Chapalain1†, Véronique Vermeersch1, Jean‑Denis Moyer2, Sigismond Lasocki3,
Benjamin Cohen4, Claire Dahyot‑Fizelier5, Kevin Chalard6, P. Seguin7, Y. Hourmant8, Karim Asehnoune8,
Antoine Roquilly8 and for the Atlanrea Study Group and the Société Française d’Anesthésie Réanimation
(SFAR) Research Network 

Abstract 
Background  To evaluate if the increase in chloride intake during a continuous infusion of 20% hypertonic saline
solution (HSS) is associated with an increase in the incidence of acute kidney injury (AKI) compared to standard of
care in traumatic brain injury patients.
Methods  In this post hoc analysis of the COBI trial, 370 patients admitted for a moderate-to-severe TBI in the 9 par‑
ticipating ICUs were enrolled. The intervention consisted in a continuous infusion of HSS to maintain a blood sodium
level between 150 and 155 mmol/L for at least 48 h. Patients enrolled in the control arm were treated as recom‑
mended by the latest Brain Trauma foundation guidelines. The primary outcome of this study was the occurrence of
AKI within 28 days after enrollment. AKI was defined by stages 2 or 3 according to KDIGO criteria.
Results  After exclusion of missing data, 322 patients were included in this post hoc analysis. The patients rand‑
omized in the intervention arm received a significantly higher amount of chloride during the first 4 days (interven‑
tion group: 97.3 ± 31.6 g vs. control group: 61.3 ± 38.1 g; p < 0.001) and had higher blood chloride levels at day 4
(117.9 ± 10.7 mmol/L vs. 111.6 ± 9 mmol/L, respectively, p < 0.001). The incidence of AKI was not statistically different
between the intervention and the control group (24.5% vs. 28.9%, respectively; p = 0.45).
Conclusions  Despite a significant increase in chloride intake, a continuous infusion of HSS was not associated with
AKI in moderate-to-severe TBI patients. Our study does not confirm the potentially detrimental effect of chloride load
on kidney function in ICU patients.
Trial registration: The COBI trial was registered on clinicaltrial.gov (Trial registration number: NCT03143751, date of
registration: 8 May 2017).
Keywords  Acute kidney injury, Sodium chloride, Fluid therapy, Brain injury

†
Olivier Huet and Xavier Chapalain collaborated equally to this manuscript
*Correspondence:
Olivier Huet
olivier.huet@chu-brest.fr
Full list of author information is available at the end of the article

© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom‑
mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Huet et al. Critical Care (2023) 27:42 Page 2 of 9

Background Methods
Traumatic brain injury (TBI) remains a worldwide Design
health priority [1]. In 2016, 55 million patients suf- This is a post hoc analysis of the COBI trial [26, 27].
fered from TBI and it was responsible for 8.1 million Briefly, the COBI trial is a multicenter, randomized,
years of life with disability [1]. At the initial phase of open-label, controlled trial that has evaluated a continu-
the insult, all medical and surgical interventions are ous infusion of HSS (20% of NaCl) in a population of
focused on avoiding secondary cerebral insults [2]. In moderate-to-severe TBI [27]. Patients aged from 18 to
order to maintain an adequate cerebral perfusion pres- 80  years old and admitted in one of the 9 participating
sure, intravenous fluid administration is one of the ICU for moderate-to-severe TBI patients were eligible.
most common strategies prescribed by physicians [3, Moderate-to-severe TBI was defined as the association of
4]. A recent consensus statement on fluid management a Glasgow Coma Score (GCS) of 12 or lower and trau-
for brain-injured patients suggested the use of crystal- matic abnormal brain CT scan findings (such as: extra-
loids as the first-line choice for maintenance or resus- dural hematoma, subdural hematoma, subarachnoid
citation fluid [5]. The experts also stressed the poor hemorrhage, brain contusion, brain hematoma, brain
level of proof to prefer buffered solution to normal edema or skull fracture) [26]. In the interventional group,
saline solution (0.9% of NaCl) in brain-injured patients TBI patients received a 1-h bolus of 7.5 to 15 g of NaCl
[5]. On the other hand, in case of refractory intracra- immediately after randomization. Then, a continuous
nial hypertension, hyperosmolar fluids (such as HSS) infusion of HSS (1  g/h NaCl) was tapered to maintain
should be considered [5]. However, normal (0.9%) and a blood sodium level between 150 and 155  mmol/L for
hypertonic (3% or 20%) saline solutions are not devoid a minimum of 48 h. After 48 h, HSS (20%) was stopped
of side effects as saline solutions contain a “supra physi- in the absence of intracranial hypertension, and blood
ological” concentration of chloride that may induce sodium level was maintained at a normal range (Na from
metabolic disturbances [6]. Among them, hyperchlo- 140 to 145  mmol/L). In case of a persistent intracranial
remia has been pointed out as an association has been hypertension, HSS infusion was maintained as long as
reported between an increased concentration of extra- necessary. In the control group, the latest Brain Trauma
cellular chloride and an impairment of the vasotone of Foundation guidelines were applied as the standard of
the afferent renal artery or a decrease in glomerular fil- care [28]. The current post hoc analysis uses patient-level
tration rate [7–9]. These reports raised the question of data focused on metabolic disturbances and kidney func-
the impact of hypertonic solutions on kidney function tion to evaluate potential side effects associated with a
in neuro-intensive care patients. continuous infusion of HSS (20%).
Prospective and retrospective clinical studies have
reported an association between chloride intake and Ethical approval
AKI in critically ill patients [10–14]. However, these The current study was approved by the Ethics Commit-
results remain controversial as a recent prospective tee of Ile de France VIII in May 2017 and conducted in
multicenter study did not find any association between accordance with the principles of the Declaration of
a “high dose” of chloride infusion and the renal prog- Helsinki. Written informed consent was provided to all
nosis in critically ill patients with septic shock [15, 16]. eligible patients at inclusion or after they recovered the
Moreover, the detrimental effect of saline solution on ability to consent. In the other cases, the next of kin pro-
kidney function has not been consistently observed in vided informed consent. Consent to participate included
the most recent randomized controlled trials [17–21]. a statement that the current study was carried out in
Furthermore, the side effects of chloride-rich solution accordance with the principles of the Declaration of
remain poorly studied in brain-injured patients, as it Helsinki.
was underlined by a recent meta-analysis [22]. To our
knowledge, only 3 small monocentric studies demon- Data collection
strated metabolic disturbances induced by chloride- We collected some physiological measurements to
rich solution following brain injury, but they didn’t describe metabolic status. From the inclusion to day 4,
study the effect of HSS [23–25]. the following parameters were collected daily: chloride,
This post hoc analysis of the COBI trial aimed at potassium, pH, bicarbonate, lactate, urea and creatinine.
determining whether high dose of chloride delivered by Blood sodium levels were collected from inclusion to
hypertonic saline solution infusion were associated with Day 10. For these metabolic parameters, the worst value
an increased acute kidney injury incidence within the each day was collected. The highest weekly creatinine
28  days after the infusion compared to standard care in levels were collected from inclusion to Day 28. We cal-
TBI patients. culated Glomerular Filtration Rate (GFR) according to
 Huet et al. Critical Care (2023) 27:42 Page 3 of 9

the Modification of Diet in Renal Disease (MDRD)-175 coefficients and estimated p value according to Satterth-
equation. Estimated Glomerular Filtration Rate (eGFR) waite approximation method. Statistical significance was
was calculated daily from inclusion to Day 4, and the set at p < 0.05. All statistical analysis was performed using
worst eGFR each week from Week 1 to Week 4. The pro- R statistical software (version 3.6).
portion of patients exposed to hyperchloremia (chloride
level ≥ 110 mmol/L) was also reported. We defined meta- Results
bolic acidosis using the following parameters: pH < 7.35 Patients
and bicarbonate level < 22  mmol/L. Urine outputs were A total of 370 patients underwent randomization in the
collected from inclusion to Day 4. We also collected the COBI study. Forty-eight (13%) patients were excluded
amount of chloride received daily (g/day) during the first from this post hoc analysis for missing data for the pri-
4 days. mary outcome. Compared to excluded patients (n = 48),
patients included in this post hoc analysis were more
Primary outcome frequently exposed to hypotension, hypoxemia or a drop
The primary outcome was the proportion of patients of hemoglobin level < 9  g/dL before inclusion (Addi-
developing an AKI from inclusion to Day 28. Kidney tional file  2: Table  S2). Included patients had also more
function was evaluated according to the KDIGO crite- frequently a past medical history of chronic kidney dis-
ria such as: Serum creatinine (µmol/L) and Urine output ease. These results are summarized in Additional file  2:
(mL/kg/h) [29]. AKI was defined as a KDIGO stage 2 or Table  S2. Of the 322 remaining patients, 159 (49.4%)
3 [29]. All data were analyzed, even those from patients patients were randomized in the control group and 163
who died before day 28. No censure was applied in the (50.6%) patients in the interventional group. The main
analysis. patients’ characteristics at baseline are described in
Table  1. At baseline, the chloride level was 106.5 in the
Secondary outcomes control group versus 107.3  mmol/L in the intervention
First, we planned to evaluate the association between group (p = 0.18). The proportions of patients with chronic
the use of the continuous infusion of hypertonic saline kidney disease (2.5% vs. 2.5%, respectively; p = 1.00) and
solution and the occurrence of hyperchloremia and baseline eGFR (109.6 vs. 111.5 mL/min/1.73 ­m2; p = 0.65)
metabolic acidosis. Second, we studied the association were also balanced between the two groups.
between the continuous infusion of HSS and the need
for renal replacement therapy (RRT) within 28 days after Chloride intake and metabolic parameters
starting the infusion of HSS. Third, we studied the associ- Patients randomized in the interventional arm received
ation of AKI from inclusion to Day 28 with chloride level a continuous infusion of hypertonic saline solution for a
and cumulative chloride dose infused. Finally, we tested mean of 2.6 (± 1.3) days. The cumulative chloride load
the association between AKI and ICU length of stay, the received from ICU admission to Day 4 was 97.3 ± 38.1 g
need of RRT from inclusion to Day 28, and ICU mortal- in the intervention group and 61.3 ± 31.6 g in the control
ity. As an exploratory analysis, we also compared eGFR group (p < 0.001). The time course of the recorded bio-
and the highest creatinine levels between the two groups. logical parameters is summarized in Table  2. From Day
1 to Day 4, the daily chloride levels were significantly
Statistical analysis higher for patients randomized in the interventional
Continuous variables were expressed as mean (± SD) group (p < 0.001). This univariate analysis was confirmed
or median (IQR). Categorical variables were expressed by a mixed model which found a higher slope in chloride
as percentage. Missing data were identified, and mul- trend in the intervention arm from Day 1 to Day 4 with
tiple imputations (5 iterations) were used for variables a regression coefficient of 5.8 (p < 0.001). From day 1 to
with less than 20% missing data. Continuous variables day 4, the base deficit, arterial pH, and bicarbonate levels
normally distributed were compared with unpaired Stu- were not statistically different between the two groups.
dent t-tests. A Wilcoxon test was used for other continu- Repeated measures were analyzed using a mixed model.
ous variables. Categorical variables were compared with We did not find any between-group differences in terms
the Chi-Square test. In univariate analysis, we identified of base deficit, arterial pH and bicarbonate level. These
the unbalanced variables between the two groups (con- results are summarized in Table  2. The time course of
trol vs. interventional arm) and any variables associated chloride levels and arterial pH in the two study groups
with AKI from inclusion to Day 28. For repeated meas- are shown in Fig. 1 with the results of the mixed model.
ures comparison, we performed a mixed model with The occurrence of metabolic acidosis from Day 1 to Day
a “subject” random effect variable and a “time” fixed 4 was not statistically different between the two groups
effect variable. Results of the mixed model are shown as (Fig. 1).
Huet et al. Critical Care (2023) 27:42 Page 4 of 9

Table 1  Baseline characteristics in the two groups

Control Intervention p
n = 159 n = 163

Age, years mean (SD) 44.1 (17.6) 43.2 (17.7) 0.6

Sex, male n (%) 129 (81.1) 127 (77.9) 0.2
Weight, kg mean (SD) 79 (17.7) 73.6 (15.3) 0.02
Time from trauma to inclusion, Hours mean (SD) 12.7 (6.2) 13 (6.4) 0.67
Severe TBI, n (%) 139 (87.4) 142 (87.1) 1
ICP before inclusion, mmHg mean (SD) 12.9 (9.3) 10.8 (7.1) 0.051
MAP before inclusion, mmHg mean (SD) 86.3 (15.3) 85.2 (14.9) 0.51
Hypotension, n (%) 26 (16.4) 26 (16) 1
Hypoxemia, n (%) 25 (15.7) 26 (16) 1
Hemoglobin level < 9 g/dl, n (%) 14 (8.8) 12 (7.4) 0.79
Chronic kidney disease, n (%) 4 (2.5) 4 (2.5) 1
eGFR, mL/min/1.73 ­m2 109.6 (39) 111.5 (34.8) 0.65
Hyperosmolar therapy, n (%) 94 (59.1) 91 (55.8) 0.63
 Mannitol 52 (55.3) 57 (62.6) 0.39
 Hypertonic saline solution 63 (67.0) 57 (62.6) 0.64
Baseline metabolic parameters, mean (SD)
Sodium, mmol/L 140 (3.9) 140.4 (4.6) 0.42
Chloride, mmol/L 106.5 (5.2) 107.3 (5.4) 0.18
Urea, mmol/L 4.8 (1.9) 4.9 (3.2) 0.87
Creatinine, µmol/L 73.6 (30.4) 72.4 (31.9) 0.74
Acid base status
 Arterial pH 7.34 (0.09) 7.34 (0.09) 0.97
 Base deficit, mmol/L − 3.73 (3.69) − 3.49 (3.70) 0.56
 Bicarbonate, mmol/L 21.67 (3.42) 22.04 (2.78) 0.28
 Arterial ­CO2, mmHg 41.24 (8.07) 42.96 (17.48) 0.26
 Lactate level, mmol/L 2.40 (2.31) 2.25 (1.74) 0.51
  Lactate > 2 mmol/L, n (%) 73 (45.9) 62 (38.0) 0.19
Daily chloride load, g/day mean (SD)
Before inclusion 14.9 (11.2) 22.7 (25.8) 0.001
 Day 1 17.5 (14.8) 34.3 (13.9)  < 0.001
 Day 2 12 (7.3) 19.7 (9.7)  < 0.001
 Day 3 8.9 (6.7) 11.2 (8.7) 0.008
 Day 4 8 (7) 9.4 (7.7) 0.09
Cumulative chloride load at Day 4, g mean (SD) 61.3 (31.6) 97.3 (38.1)  < 0.001
Antidiuretic hormone use during ICU stay, n (%) 26 (16.4) 10 (6.1) 0.006
Outcomes
ICU length of stay, Days mean (SD) 20.4 (17.9) 23.7 (22.1) 0.14
ICU mortality, n (%) 32 (20.1) 26 (16) 0.41
eGFR estimated Glomerular Filtration Rate, ICP Intracranial Pressure, ICU Intensive Care Unit, MAP Mean Arterial Pressure, SD Standard Deviation, TBI: Trauma Brain
Injury

Primary outcome (p = 0.25). Only 3 patients suffered from AKI after day 7,
From inclusion to Day 28, AKI was recorded in 46 these patients were in the intervention arm.
(28.9%) patients in the control group and 40 (24.5%) in
the intervention group (p = 0.45, Table  3). The great
Renal outcomes
majority of AKI event (96.5%) was observed before Day
During the ICU stay, 3 (1.9%) patients in the control
7 without differences between the two groups: 46 (28.9%)
group and 3 (1.8%) in the intervention group received
in control group vs. 37 (22.7%) in intervention group
RRT (p = 1). The main reason for beginning RRT was:
 Huet et al. Critical Care (2023) 27:42 Page 5 of 9

Table 2  Comparison of metabolic parameters between the two analysis confirmed the univariate analysis. In this random
groups effect model, trends of creatinine level and eGFR were
Control Intervention p also comparable (p value > 0.05).
n = 159 n = 163
Association of AKI with TBI patients’ outcomes
Sodium, mmol/L
As an exploratory analysis, we studied the association
Before inclusion 140 (3.9) 140.4 (4.6) 0.42
between AKI onset, defined as a KDIGO stage 2 and
Day 1 145 (5.7) 151.9 (4.9)  < 0.001
3, and several ICU outcomes. The length of ICU stay
Day 2 145.6 (6.3) 154.1 (5.4)  < 0.001
was 19.8 ± 17.4  days in TBI patients without AKI and
Day 3 146.3 (6.5) 153.8 (5.8)  < 0.001
28.2 ± 25.4  days in patients with AKI (p = 0.001). ICU
Day 4 145.9 (6.4) 151.7 (6.2)  < 0.001
mortality was 19.5% for patients without AKI and 14%
Chloride, mmol/L
for patients with AKI (p = 0.33). There was no associa-
Before inclusion 106.5 (5.2) 107.3 (5.4) 0.18
tion between cumulative chloride infusion on Day 4 and
Day 1 113.6 (6) 120.8 (8.1)  < 0.001
the occurrence of AKI: 79.2  g versus 80.4  g (p = 0.81).
Day 2 114.5 (6.4) 122.8 (8.6)  < 0.001 There was no association between hyperchloremia expo-
Day 3 114.6 (8.8) 121 (8.5)  < 0.001 sure from inclusion to Day 4 and the onset of AKI within
Day 4 111.6 (10.7) 117.9 (9)  < 0.001 28  days after randomization. These results are summa-
Arterial pH rized in Additional file 1: Table S1.
Before inclusion 7.34 (0.09) 7.34 (0.09) 0.97
Day 1 7.37 (0.08) 7.36 (0.06) 0.46
Discussion
Day 2 7.40 (0.07) 7.39 (0.07) 0.22
This post hoc analysis of the COBI trial evaluated the
Day 3 7.41 (0.07) 7.42 (0.06) 0.24
impact of a continuous infusion of HSS (20%) on renal
Day 4 7.43 (0.05) 7.42 (0.07) 0.58
function after moderate-to-severe TBI. No association
Base deficit, mmol/L
between the continuous infusion of HSS (20%) and AKI
Before inclusion − 3.7 (3.7) − 3.5 (3.7) 0.56
(KDIGO stage 2 or 3) from inclusion to Day 28 was find.
Day 1 − 2.7 (3.6) − 2.7 (3.1) 0.97
Creatinine levels and eGFR were also similar between the
Day 2 − 0.47 (2.9) − 0.96 (2.9) 0.13
two groups. Chloremia was significantly higher in the
Day 3 1.2 (2.7) 0.9 (2.5) 0.28
HSS group. No difference in acid/base metabolic status
Day 4 2.3 (3.3) 2.4 (3) 0.98
was found between the two groups.
Bicarbonate, mmol/L
Fluid therapy solutions, administered in ICU, con-
Before inclusion 21.7 (3.4) 22.1 (2.8) 0.28
tain different concentrations of electrolytes [3]. Among
Day 1 16.4 (10.8) 17 (10.1) 0.62
these electrolytes, chloride has raised concern about its
Day 2 23.5 (3.2) 23.1 (2.6) 0.25
iniquity [3]. Actually, an acute administration of a large
Day 3 25.1 (3.1) 24.9 (2.3) 0.59
amount of chloride-rich solution may lead to hyperchlo-
Day 4 27 (5) 27.2 (4.3) 0.70
remic metabolic acidosis associated with pathophysiolog-
Lactate, mmol/L
ical consequences such as: coagulopathy, hypotension,
Before inclusion 2.4 (2.3) 2.3 (1.7) 0.51
or splanchnic disorders [30, 31]. Furthermore, chronic
Day 1 1.8 (1.2) 1.6 (0.8) 0.21
hyperchloremia is also theoretically associated with an
Day 2 1.4 (0.8) 1.5 (1) 0.75 excessive vasoconstriction of afferent renal arteries lead-
Day 3 1.3 (0.5) 1.3 (0.6) 0.58 ing to a decrease in glomerular filtration rate [7–9].
Day 4 1.2 (0.49) 1.2 (1) 0.90 In the past decade, three large RCTs compared buff-
Results were expressed as means (± SD) ered crystalloid to normal (0.9%) saline solution in
SD Standard Deviation various settings [17, 19–21]. These RCT’s did show a
statistically significant difference between the groups
severe metabolic acidosis (3 patients), severe hyper- regarding chloride intake or changes in chloremia and
kalemia (1 patient), elevated blood urea nitrogen (1 arterial acid–base balance [17, 19–21]. However, the
patient), and pulmonary edema (1 patient). The time clinical significance of these differences is questionable.
course of the creatinine levels and eGFR during the first Actually, in the SMART trial the difference of chloride
four days were not statistically different between the intake between the two groups was of 0.5 g, and differ-
two groups from inclusion to Day 4. This was also the ences in terms of blood chloride levels were also small
case for the highest weekly creatinine levels and lowest and probably not clinically relevant: 109 mmol/L (saline
eGFR from inclusion to Day 28 (Fig. 2). The mixed model solution) versus 108  mmol/L (balanced crystalloid)
[20]. These limitations are also found in the BaSICS
Huet et al. Critical Care (2023) 27:42 Page 6 of 9

Fig. 1  Trend in main metabolic parameters measured from inclusion to Day 4 in the two groups. A Higher chloride level (mmol/L) from inclusion to
Day 4. B Proportion of patients exposed to hyperchloremia (Cl > 109 mmol/L) at each time from inclusion to Day 4. C Lower pH level from inclusion
to Day 4. D Proportion of patients exposed to metabolic acidosis (pH < 7.35 and bicarbonate level < 22 mmol/L) at each time from inclusion to Day
4. In each boxplot, dots represented outliers. *p < 0.05 for univariate analysis

and the PLUS trials [17, 19, 21]. Therefore, it is difficult

Table 3 Comparison of renal outcomes in the 2 groups from to conclude about the kidney toxicity of high chloride
inclusion to Day 28 intake in ICU patients with these clinical trials [32].
On the other hand, patients enrolled in the COBI Trial
Control Intervention p
received a substantial and prolonged amount of chloride
n = 159 n = 163
in the interventional arm. To our knowledge, this is the
From inclusion to Day 28 first time that the effect of such an important chloride
AKI (KDIGO 2–3), n (%) 46 (28.9) 40 (24.5) 0.45 intake has been reported. Our results are not in line with
KDIGO classification, n (%) 0.62 a potential toxicity of chloride intake on kidney function.
 1 32 (20.1) 30 (18.4) To date, there was only one large RCT comparing HSS
 2 35 (22) 33 (20.2) (3%) to normal saline solution (0.9%) and the propor-
 3 11 (6.9) 7 (4.3) tion of patients who required RRT [33]. In this study, the
RRT, n (%) 3 (1.9) 3 (1.8) 1 needed of RRT between HSS and normal saline solution
AKI Acute Kidney Injury, KDIGO Kidney Disease Improving Global Outcome, RRT​ was not statistically different (38% vs. 33%, p = 0.32) [33].
Renal Replacement Therapy, SD Standard Deviation
 Huet et al. Critical Care (2023) 27:42 Page 7 of 9

Fig. 2  Trend in kidney parameters measured in the two groups. A Creatinine level (µmol/L) measured from inclusion to Day 4. B Higher creatinine
level (µmol/L) measured each week from Week 1 to Week 4. C Daily eGFR (mL/min/1.73 ­m2) from inclusion to Day 4. D Worst eGFR (mL/min/1.73
­m2) measured each week from Week 1 to Week 4. In each boxplot, dots represented outliers. eGFR estimated Glomerular Filtration Rate

The main strength of our study is the design, as it is Our study also has some limitations. First, this is a post
a randomized controlled trial. Second, this is the first hoc analysis, and this ancillary study was not planned
study reporting the effect of the infusion of a highly in the original statistical analysis plan [27]. Second,
concentrated solution in chloride. Third, we report the patients in the control group also received a significant
amount of chloride infused and the difference in blood amount of chloride (61.3 ± 31.6 g in 4 days) and were also
chloride levels which is rarely done in other studies. exposed to hyperchloremia from Day 1 to Day 4. This
Huet et al. Critical Care (2023) 27:42 Page 8 of 9

may have limited the statistical power of this study to Author contributions
OH and XC were involved equally in the conception of the study, hypothesis
demonstrate a difference between the two study groups. generation, writing and revision of the article before submission. OH and XC
However, the amount of chloride infused in the control contributed equally in data analysis. OH, XC, VV, J-DM, SL, BC, CD-F, KC, PS, YH,
group is comparable to the estimated amount reported KA and AR contributed equally in data collection. OH and XC were involved in
the writing of the manuscript before submission. OH, XC, VV and AR contrib‑
in the recent CENTER-TBI cohort [34]. Moreover, in uted equally in revision of manuscript before submission. All authors read and
our study the chloride intake is clinically significantly approved the final manuscript.
higher in the intervention arm. Third, patients included
Funding
in the COBI trial were younger (mean age: 44 years old) This study is an investigator-initiated trial and was supported by a Grant from
and had a lower risk of AKI than patients included in the French Ministry of Health—Programme Hospitalier de Recherche Clinique
previous studies [18, 20]. Moreover, the requirement of Inter-regional 2016 (PHRCI 2016, RC16_0474).
RRT (1.9%) was slightly lower compare to other studies Availability of data and materials
[17, 20]. Thus, our results cannot be generalized to high The dataset supporting the conclusions of this article is fully available. To have
risk patients. Fourth, AKI patients had more frequently an access on it, please contact the corresponding author (O.H.)
a baseline chronic kidney disease but the proportion of
patients who had a chronic kidney disease was well bal- Declarations
anced between control and intervention group. There- Ethics approval and consent to participate
fore, we believe that the analysis of the main results The study protocol was approved by the Ethics Committee of Ile de France VIII
cannot be confounded by chronic kidney disease. in May 2017. Informed consent was given by the patients or their caregivers.

Consent for publication

A written inform consent for publication was obtained.
Conclusion
Competing interest
After moderate-to-severe TBI, the continuous infusion of The authors declare that they have no competing interests.
HSS (20%) was not associated with an increased the risk
of developing AKI from inclusion to Day 28. Our findings Author details
1
 Department of Anesthesiology and Surgical Intensive Care Unit, Brest Uni‑
questioned the potentially detrimental effects of chlo- versity Hospital, Boulevard Tanguy Prigent, 29609 Brest, France. 2 Department
ride-rich solution on kidney function. Further studies are of Anesthesiology and Critical Care, Beaujon Hospital, DMU Parabol, AP-HP
warranted to better evaluate impact of fluid therapy in Nord, Paris, France. 3 Department of Anesthesia and Intensive Care Unit, Angers
Hospital, Angers, France. 4 Department of Anesthesia and Intensive Care Unit,
ICU patients, especially after TBI. Tours Hospital, Tours, France. 5 Department of Anesthesia and Intensive Care
Unit, Poitiers Hospital, Poitiers, France. 6 Department of Anesthesia and Inten‑
sive Care Unit, Montpellier Hospital, Montpellier, France. 7 Department
Abbreviations of Anesthesia and Intensive Care Unit, Rennes Hospital, Rennes, France. 8 Pôle
AKI Acute Kidney Injury Anesthésie Réanimations, Service d’Anesthésie Réanimation Chirurgicale,
eGFR Estimated Glomerular Filtration Rate Hôtel Dieu, Université de Nantes, CHU Nantes, Nantes, France.
GCS Glasgow Coma Scale
GFR Glomerular Filtration Rate Received: 17 October 2022 Accepted: 7 January 2023
ICP Intracranial Pressure
ICU Intensive Care Unit
IQR Inter Quartile Range
KDIGO Kidney Disease Improving Global Outcome
MAP Mean Arterial Pressure
MDRD Modification of Diet in Renal Disease References
RCT​ Randomized Controlled Trial 1. James SL, Theadom A, Ellenbogen RG, Bannick MS, Montjoy-Venning W,
RRT​ Renal Replacement Therapy Lucchesi LR, et al. Global, regional, and national burden of traumatic brain
SD Standard Deviation injury and spinal cord injury, 1990–2016: a systematic analysis for the
TBI Trauma Brain Injury Global Burden of Disease Study 2016. Lancet Neurol. 2019;18:56–87.
2. Maas AIR, Menon DK, Adelson PD, Andelic N, Bell MJ, Belli A, et al. Trau‑
matic brain injury: integrated approaches to improve prevention, clinical
Supplementary Information care, and research. Lancet Neurol. 2017;16:987–1048.
The online version contains supplementary material available at https://​doi.​ 3. Finfer S, Myburgh J, Bellomo R. Intravenous fluid therapy in critically ill
org/​10.​1186/​s13054-​023-​04311-1. adults. Nat Rev Nephrol. 2018;14:541–57.
4. Cecconi M, Hofer C, Teboul J-L, Pettila V, Wilkman E, Molnar Z, et al. Fluid
challenges in intensive care: the FENICE study: a global inception cohort
Additional file 1: Table S1. Comparison between patients according to study. Intensive Care Med. 2015;41:1529–37.
existence of AKI (KDIGO stage 2–3) from inclusion to Day 28. 5. Oddo M, Poole D, Helbok R, Meyfroidt G, Stocchetti N, Bouzat P, et al. Fluid
Additional file 2: Table S2. Comparison of baseline characteristics therapy in neurointensive care patients: ESICM consensus and clinical
between included and non-included patients. practice recommendations. Intensive Care Med. 2018;44:449–63.
6. Kaplan LJ, Frangos S. Clinical review: acid–base abnormalities in the
intensive care unit—part II. Crit Care. 2005;9:198–203.
Acknowledgements 7. Imig JD, Passmore JC, Anderson GL, Jimenez AE. Chloride alters renal
No acknowledgement. blood flow autoregulation in deoxycorticosterone-treated rats. J Lab Clin
Med. 1993;121:608–13.
 Huet et al. Critical Care (2023) 27:42 Page 9 of 9

8. Wilcox CS. Regulation of renal blood flow by chloride. In: Laragh JH, 28. Carney N, Totten AM, O’Reilly C, Ullman JS, Hawryluk GWJ, Bell MJ, et al.
Bühler FR, Seldin DW, editors. Frontiers in Hypertension Research. New Guidelines for the management of severe traumatic brain injury, fourth
York: Springer; 1981. p. 135–8. edition. Neurosurgery. 2017;80:6–15.
9. Hansen PB, Jensen BL, Skøtt O. Chloride regulates afferent arteriolar con‑ 29. KDIGO. Clinical practice guideline update for the diagnosis, evaluation,
traction in response to depolarization. Hypertension. 1998;32:1066–70. prevention, and treatment of chronic kidney disease-mineral and bone
10. Shaw AD, Raghunathan K, Peyerl FW, Munson SH, Paluszkiewicz SM, disorder (CKD-MBD). Kidney Int Suppl. 2017;2017(7):1–59.
Schermer CR. Association between intravenous chloride load during 30. Kellum JA, Song M, Almasri E. Hyperchloremic acidosis increases
resuscitation and in-hospital mortality among patients with SIRS. Inten‑ circulating inflammatory molecules in experimental sepsis. Chest.
sive Care Med. 2014;40:1897–905. 2006;130:962–7.
11. Chowdhury AH, Cox EF, Francis ST, Lobo DN. A randomized, controlled, 31. Tournadre JP, Allaouchiche B, Malbert CH, Chassard D. Metabolic acidosis

and plasma-lyte® 148 on renal blood flow velocity and renal cortical tis‑
double-blind crossover study on the effects of 2-L infusions of 0.9% saline and respiratory acidosis impair gastro-pyloric motility in anesthetized
pigs. Anesth Analg. 2000;90:74–9.
sue perfusion in healthy volunteers. Ann Surg. 2012;256:18–24. 32. Bellomo R, Kellum JA, Ronco C, Wald R, Martensson J, Maiden M, et al.
12. Yunos NM, Bellomo R, Hegarty C, Story D, Ho L, Bailey M. Association Acute kidney injury in sepsis. Intensive Care Med. 2017;43:816–28.
between a chloride-liberal vs chloride-restrictive intravenous fluid 33. Asfar P, Schortgen F, Boisramé-Helms J, Charpentier J, Guérot E, Meg‑
administration strategy and kidney injury in critically ill adults. JAMA. arbane B, et al. Hyperoxia and hypertonic saline in patients with septic
2012;308:1566–72. shock (HYPERS2S): a two-by-two factorial, multicentre, randomised, clini‑
13. Yunos NM, Bellomo R, Glassford N, Sutcliffe H, Lam Q, Bailey M. Chloride- cal trial. Lancet Respir Med. 2017;5:180–90.
liberal vs. chloride-restrictive intravenous fluid administration and acute 34. Wiegers EJA, Lingsma HF, Huijben JA, Cooper DJ, Citerio G, Frisvold S,
kidney injury: an extended analysis. Intensive Care Med. 2015;41:257–64. et al. Fluid balance and outcome in critically ill patients with traumatic
14. Suetrong B, Pisitsak C, Boyd JH, Russell JA, Walley KR. Hyperchloremia and brain injury (CENTER-TBI and OzENTER-TBI): a prospective, multicentre,
moderate increase in serum chloride are associated with acute kidney comparative effectiveness study. Lancet Neurol. 2021;20:627–38.
injury in severe sepsis and septic shock patients. Crit Care. 2016;20:315.
15. Commereuc M, Nevoret C, Radermacher P, Katsahian S, Asfar P, Schortgen
F, et al. Hyperchloremia is not associated with AKI or death in septic shock Publisher’s Note
patients: results of a post hoc analysis of the “HYPER2S” trial. Ann Intensive Springer Nature remains neutral with regard to jurisdictional claims in pub‑
Care. 2019;9:95. lished maps and institutional affiliations.
16. Chapalain X, Huet O, Balzer T, Delbove A, Martino F, Jacquier S, et al. Does
chloride intake at the early phase of septic shock resuscitation impact on
renal outcome? Shock. 2021.
17. Young P, Bailey M, Beasley R, Henderson S, Mackle D, McArthur C, et al.
Effect of a buffered crystalloid solution vs saline on acute kidney injury
among patients in the intensive care unit: the SPLIT randomized clinical
trial. JAMA. 2015;314:1701.
18. Self WH, Semler MW, Wanderer JP, Wang L, Byrne DW, Collins SP, et al.
Balanced crystalloids versus saline in noncritically ill adults. N Engl J Med.
2018;378:819–28.
19. Zampieri FG, Machado FR, Biondi RS, Freitas FGR, Veiga VC, Figueiredo
RC, et al. Effect of intravenous fluid treatment with a balanced solution
vs 0.9% saline solution on mortality in critically ill patients: the BaSICS
randomized clinical trial. JAMA. 2021 [cited 2021 Aug 26]; Available from:
https://​jaman​etwork.​com/​journ​als/​jama/​fulla​rticle/​27830​39
20. Semler MW, Self WH, Wanderer JP, Ehrenfeld JM, Wang L, Byrne DW, et al.
Balanced crystalloids versus saline in critically ill adults. N Engl J Med.
2018;378:829–39.
21. Finfer S, Micallef S, Hammond N, Navarra L, Bellomo R, Billot L, et al. Bal‑
anced multielectrolyte solution versus saline in critically ill adults. N Engl J
Med. 2022;386:815–26.
22. Martín AMA, Mendoza JAB, Muriel A, Sáez I, Chico-Fernández M, Estrada-
Lorenzo JM, et al. Buffered solutions versus 0.9% saline for resuscitation
in critically ill adults and children. Cochrane Database Syst Rev. 2019.
https://​doi.​org/​10.​1002/​14651​858.​CD012​247.​pub2.
23. Roquilly A, Loutrel O, Cinotti R, Rosenczweig E, Flet L, Mahe PJ, et al. Bal‑
anced versus chloride-rich solutions for fluid resuscitation in brain-injured
patients: a randomised double-blind pilot study. Crit Care. 2013;17:R77.
24. Lehmann L, Bendel S, Uehlinger DE, Takala J, Schafer M, Reinert M, et al.
Randomized, double-blind trial of the effect of fluid composition on
electrolyte, acid-base, and fluid homeostasis in patients early after suba‑
rachnoid hemorrhage. Neurocrit Care. 2013;18:5–12.
25. Hassan MH, Hassan WMNW, Zaini RHM, Shukeri WFWM, Abidin HZ, Eu
CS. Balanced fluid versus saline-based fluid in post-operative severe trau‑
matic brain injury patients: acid-base and electrolytes assessment. Malays
J Med Sci. 2017;24:83–93.
26. Roquilly A, Moyer JD, Huet O, Lasocki S, Cohen B, Dahyot-Fizelier C, et al.
Effect of continuous infusion of hypertonic saline vs standard care on
6-month neurological outcomes in patients with traumatic brain injury:
the COBI randomized clinical trial. JAMA. 2021;325:2056–66.
27. Roquilly A, Lasocki S, Moyer JD, Huet O, Perrigault PF, Dahyot-fizelier C,
et al. COBI (COntinuous hyperosmolar therapy for traumatic Brain-Injured
patients) trial protocol: a multicentre randomised open-label trial with
blinded adjudication of primary outcome. BMJ Open. 2017;7:e018035.