Does serum neutrophil gelatinase–associated lipocalin level predict acute kidney injury in patients with acute rhabdomyolysis in the emergency department? A multicentre prospective study

Abstract

Objectives

The major complication of rhabdomyolysis is acute kidney injury (AKI), which requires prompt treatment. Currently, few biomarkers are available for the early detection of AKI. Serum neutrophil gelatinase–associated lipocalin (NGAL) has been suggested as an early biomarker for renal ischemia. However, its capacity to predict AKI in patients presenting with rhabdomyolysis in the emergency department (ED) remains unclear. The aim of this study was to evaluate the ability of NGAL to predict 48-hour AKI.

Design

Prospective, multicentre study.

Setting

Five adult EDs in France from August 2013 to December 2015.

Participants

NGAL levels were measured on ED admission in patients with rhabdomyolysis. A total of 197 patients were enrolled, and 189 (96%) were analysed, of whom 89 (47%) were women. Patients were included if they presented to the ED with rhabdomyolysis and a creatine phosphokinase (CPK) level above 1000 IU/L. Exclusion criteria were pregnancy, presentation with acute coronary syndrome, the need for iodinated contrast, chronic dialysis or recent use of nephrotoxic drugs (within 72 hours prior to the ED visit). Patients who withdrew consent or had AKI due to other causes were also excluded.

Primary and secondary outcome measures

The primary outcome was AKI at 48 hours, defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) criteria. Secondary outcomes included in-hospital mortality, length of hospital stay, admission to intensive care and the need for renal replacement therapy.

Results

Overall, 54 (29%) patients developed AKI by day 2. The area under the ROC curve (AUC-ROC) for NGAL in predicting AKI on day 2 was 0.60 (95% CI 0.51 to 0.70), with an optimal cut-off of 129 ng/mL. The sensitivity was 0.65, and specificity was 0.50. After adjustment for CPK levels, age, sex and oxygen saturation, the AUC-ROC for predicting AKI on day 2 increased slightly to 0.64 (95% CI 0.54 to 0.74).

Conclusion

NGAL has limited ability to predict day 2 AKI in patients presenting with acute rhabdomyolysis in the ED.

Trail registration number

NCT01544231.

Comité de Protection des Personnes Sud Méditerranée III n°2011-A01059-32.

Correspondence to Dr Fabien Coisy; [email protected]

STRENGTHS AND LIMITATIONS OF THIS STUDY

• The prospective and multicentre design enhanced the generalisability of the findings across different hospital settings and patient populations.

• A standardised treatment protocol was implemented to ensure uniform patient care, which may have the effect of reducing variability in outcomes related to acute kidney injury (AKI).

• The low creatine phosphokinase threshold for inclusion may lead to a lower risk population for (AKI), which could affect the study’s ability to detect neutrophil gelatinase–associated lipocalin predictive power for more severe cases of rhabdomyolysis.

• The use of convenience sampling may have introduced a degree of selection bias, as not all patients were systematically included, which could affect the representativeness of the sample.

• The delay in data analysis may limit the applicability of the findings to current clinical practices.

Introduction

Rhabdomyolysis is defined as the destruction of striated muscle cells, as evidenced by a rise in creatinine phosphokinase (CPK) levels exceeding 1000 UI/L.¹ Rhabdomyolysis may be the result of muscle trauma or metabolic disturbances that affect cellular metabolism. The release of CPK into the bloodstream can result in hydroelectrolytic imbalances and may contribute to the development of acute kidney injury (AKI) by reducing renal blood flow and causing direct myoglobin toxicity.² The prevalence of rhabdomyolysis remains unclear. It is estimated that there are 26 000 cases of rhabdomyolysis in the USA each year.³ It is thought to be responsible for 5 to 25% of cases of acute kidney injury (AKI), with renal replacement therapy required in approximately half of these cases.⁴ It would appear that AKI is associated with mortality in intensive care unit (ICU) patients.⁵ The prevention of AKI is of paramount importance in the management of patients with rhabdomyolysis, necessitating an expedient diagnosis and treatment plan.⁶

The definition of AKI has been based, since 2012, on diuresis and serum creatinine level, according to the Kidney Disease: Improving Global Outcome (KDIGO) scale.⁷ This definition necessitates a 48-hour follow-up to assess urine output and serum creatinine levels. The use of creatinine levels in acute settings is limited by several factors. First, its elevation may be delayed by up to 24 hours. Second, creatinine levels are related to muscle mass and do not reflect tissue damage. Research is ongoing to identify reliable biomarkers that can predict AKI. Potential biomarkers include soluble urokinase plasminogen activator receptor and neutrophil gelatinase–associated lipocalin (NGAL).⁸ NGAL is a lipocalin-type protein, expressed by polynuclear cells in a number of different tissues, including the kidney, lungs, trachea, stomach and colon.⁹ This protein plays a regulatory role in cells, including those in the renal epitheliums. Animal experimental models have shown that NGAL is secreted in response to renal ischemia, as part of the kidney’s repair mechanism.¹⁰ NGAL seems to be an early biomarker of renal epithelial dysfunction.

Clinical studies have shown that urinary or serum NGAL level can predict AKI in cardiac surgery, interventional radiology or patients in ICU.¹¹ In the emergency department (ED), urinary NGAL could be predictive of AKI.¹² However, urinary NGAL levels are influenced by chronic kidney disease (CKD) and urinary infection, making its use less relevant in ED.¹³ Moreover, serum testing appears more feasible in the context of ED, as urinary samples may be difficult to collect.¹⁴ Serum NGAL could therefore be an early predictor of kidney dysfunction in rhabdomyolysis and could permit an earlier therapeutic management, potentially reducing AKI-related morbidity and mortality.¹⁵ This study aimed to evaluate if serum NGAL levels could predict AKI in ED patients presenting with rhabdomyolysis.

Methods

Study design and setting

This study was designed as a prospective, multicenter diagnostic study in five French university hospitals. Patients were sampled for convenience, due to the variation of patient inflow and physician availability in ED, which did not allow for the inclusion of all patients. Written consent was obtained after providing information about the protocol and before inclusion. If a patient could not provide consent, an emergency procedure allowed their inclusion. If patients later declined to provide consent, the collected data were deleted. This study was approved by an ethics committee (Comité de Protection des Personnes Sud Méditerranée III n°2011-A01059-32) and registered at ClinicalTrials.gov (NCT01544231). The entire study was conducted in accordance with the Declaration of Helsinki. Full study protocol is available here: https://clinicaltrials.gov/study/NCT01544231?cond=neutrophilgelatinaserhabdomyolysis&rank=1

Participant selection

Inclusion criteria were patients presenting to ED with rhabdomyolysis and a CPK level over 1000 UI/L. This threshold was chosen because the risk of AKI increases above this level.³ Patients were also required to have health insurance and be hospitalised with a 48-hour follow-up. Non-inclusion criteria were participation in another clinical study or the inability to provide consent. Exclusion criteria included pregnancy, acute coronary syndrome, long-term dialysis or nephrotoxic drug intake within 72 hours prior to the ED visit. If patient withdrew consent or had AKI that could be explained by anything else than rhabdomyolysis, patient was excluded.

Procedures

After inclusion day (D0) and before any treatment, venous blood sample were collected to measure NGAL, CPK, creatinine, myoglobin and biochemistry parameters (natraemia, kaliemia, calcaemia and chloraemia). NGAL was measured by immunofluorescence, with a minimum detection cut-off of 15 ng/mL (Alere Triage NGAL test, Abbott, Chicago, USA). Blood samples for measuring CPK, creatinine, myoglobin and other biochemistry parameters were also collected at 24 hours (D1) and 48 hours (D2). A standardised therapy was early initiated, in accordance with Bosch et al’s recommendations for treating rhabdomyolysis complications.¹ Vascular filling with 0.9% sodium chloride (NaCl) was administered at a rate of 200 and 1000 mL/h, targeting a urine output of 3 mL/kg/h. Potassium levels were regularly monitored, symptomatic hypocalcaemia was treated as needed, and 200 mL of sodium bicarbonate was given if urinary pH was belCreatininaemiaow 6.5. The experimental protocol is presented in figure 1.

Figure 1. Experimental protocol schema. CPK, creatinine phosphokinase; D0, inclusion day; D1, day 1; D2, day 2; NGAL, neutrophil gelatinase-associated lipocalin.

Outcomes

The primary aim of the study was to predict in-hospital AKI at 48-hour (D2) following ED admission for rhabdomyolysis, based on initial serum NGAL levels. The primary endpoint was D2 AKI defined by the KDIGO criteria.⁷ We evaluated KDIGO scale comparing D0 and D2 levels of creatinine levels. Secondary objective included assessing the predictive value of NGAL for mortality, hospitalisation length, ICU admission and the need for renal replacement therapy. ICU admission was required when patient had at least one organ failure, as defined by local procedures. Renal replacement therapy was initiated according to the recommendations at the time of the study.^{16 17} Two follow-up visits on D1 and D2 allowed investigators to assess clinical, biological and therapeutic parameters. A long hospital stay was defined as hospitalisation longer than the cohort median stay.

Analysis

Qualitative data are expressed as numbers and percentages. Quantitative data are expressed as the mean and SD if the distribution was normal or as median assorted of first and third quartiles (Q1; Q3) if the distribution was not normal. Qualitative data were compared using χ² test if there were more than 5 patients in theorical groups or otherwise using Fisher’s exact test. Quantitative data were compared using Student t-test ifor normal distribution and the Mann-Whitney U-test for non-normal distributions. The ability of serum NGAL to predict AKI was analysed using receiver operating characteristic (ROC) curve, with the aera under curve (AUC) and its 95% CI calculated by the DeLong method. The optimal NGAL level was determined by evaluating sensitivity and specificity for every ROC curve threshold.

To control for potential confounding factors related to AKI, multivariate logistic regression was performed, including variables with a p value <0.2 in the univariate model and less than 10% missing data. Serum NGAL levels for predicting mortality, ICU admission or length of stay were also analysed using ROC curves with AUC and 95% CI. AUC was interpreted as excellent if ≥0.9, good if between 0.8 and 0.9, fair if between 0.7 and 0.8, poor if between 0.6 and 0.7, and as failed if between 0.5 and 0.6.¹⁸

All patients included in the study were analysed, and the statistical difference threshold for statistical significance was set at 0.05. Statistical analysis was performed by the laboratory of biostatistics, clinical epidemiology, public health, innovation and methodology (BESPIM) of Nîmes University, using R 3.5.1 software (R Development Core Team (2018); R Foundation for Statistical Computing, Vienna, Austria).

Sample size calculation

The required sample size was 171 patients, calculated based on an optimal AUC-ROC of 0.90, with a 13% risk of AKI, a 5% SD and 5% first-type error. Considering a 15% risk of missing data, a total of 197 patients were needed to meet the study protocol requirements.

Patient and public involvement

None.

Results

From August 2013 to December 2015, a total of 194 patients were included, of whom 189 (96%) had 48-hour data that were analysed. The study flowchart is presented in figure 2. The median NGAL level at admission was 130 (84; 227) ng/mL. The median time between the onset of rhabdomyolysis and NGAL measurement was 12 (8; 24) hours. The main aetiology of rhabdomyolysis was prolonged immobility (ie, lying on the floor for extended periods) in 134 (69%) patients. The median length of hospital stay was 11 (6; 18) days. A total of 20 (11%) patients died during hospitalisation. Patients’ initial characteristics and their outcome following ED discharge are presented in table 1. The evolution of patients at D1 and D2 is presented in online supplemental table S1.

Figure 2. Study flowchart. AKI, acute kidney injury; NGAL, neutrophil gelatinase-Associated lipocalin. *Details of included patients per centre: university hospital of Nîmes, n=88 (45%); university hospital of Montpellier, n=41 (21%); university hospital of Nice, n=34 (18%); military university hospital of Toulon, n=29 (15%); university hospital of La Pitié Salpétrière, n=2 (1%).

Table 1

Patients’ initial characteristics and discharge after emergency department visit

Variable	Total n=189	48 hours AKI n=54	No 48 hours AKI n=135	P value
Women, no (%)	89 (47)	23 (43)	66 (49)	0.53
Age, mean±SD, years	77±15	79±13	76±16	0.30
Mass, mean (SD), kilograms	71±16	70±16	71±16	0.74
Size, mean (SD), metres	1.7±0.1	1.7±0.1	1.7±0.1	0.82
Previous disease
High blood pressure, no (%)	105 (56)	32 (59)	73 (54)	–
Cardiac arrythmia, no (%)	33 (17)	12 (22)	20 (15)	–
Ischaemic cardiopathy, no (%)	20 (11)	4 (7)	16 (12)	–
Diabetes, no (%)	44 (23)	14 (26)	30 (22)	–
Dyslipidaemia, no (%)	39 (21)	13 (24)	26 (19)	–
Chronic kidney disease, no (%)	9 (5)	5 (9)	4 (3)	–
Cerebrovascular disease, no (%)	11 (6)	3 (6)	8 (6)	–
COPD, no (%)	19 (10)	8 (15)	11 (8)	–
Neoplasia, no (%)	19 (10)	3 (6)	16 (12)	–
Kidney graft, no (%)	2 (1)	0 (0)	2 (1)	–
Physiologic parametres
SBP, mean±SD, mm Hg	137±27	135±33	137±24	0.51
DBP, mean±SD, mm Hg	73±15	72±16	73±15	0.86
MBP, mean±SD, mm Hg	94±18	92±21	95±16	0.55
Heart rate, mean (SD), beats per minute	87±18	89±19	86±17	0.80
Pulse oxygen saturation, mean±SD, %	96±3	94±8	96±3	0.12
Biological parametres
CPK, median (Q1; Q3), UI.L−1	2848 (1692 ; 5360)	3377 (1805; 8682)	2665 (1686; 5000)	0.15
Creatininaemia, median (Q1; Q3), µmol/1	95 (68; 141)	128 (95; 188)	85 (61; 124)	< 0.05
Myoglobin, median (Q1; Q3), µg/L	1057 (42; 2727)*	2796 (1291; 7542)	738 (0; 2075)	< 0.05
Natremia, median (Q1; Q3), mmol/L	139 (136; 142)	140 (136; 142)	139 (135; 142)	0.45
Kaliemia, median (Q1; Q3), mmol/L	4.2 (3.8; 4.6)	4.2 (4.0; 4.8)	4.1 (3.8; 4.5)	0.11
Phosphoraemia, median (Q1; Q3), mmol/L	1.06 (0.88; 1.31)*	1.21 (0.97; 1.57)	1.01 (0.81; 1.21)	< 0.05
SGOT, median (Q1; Q3), UI/L	96 (64; 144)*	123 (80; 270)	92 (60; 139)	< 0.05
SGPT, median (Q1; Q3), UI/L	36 (26; 59)*	48 (32; 91)	33 (24; 49)	< 0.05
Urea, median (Q1; Q3), mmol/L	10.1 (6.9; 13.8)	13 (10; 23)	9 (6; 12)	< 0.05
Venous lactate, median (Q1; Q3), mmol/L	1.5 (1.1; 2.2)*	1.6 (1.1; 2.5)	1.5 (1.2; 2.0)	0.38
CRP, median (Q1; Q3), mg/L	60 (25; 118)	60 (34; 132)	61 (24; 118)	0.58
NGAL, median (Q1; Q3), ng/mL	130 (84; 227)	170 (109; 332)	125 (80; 220)	< 0.05
Bicarbonate, median (Q1; Q3), mmol/L	23.6 (21; 25.6)*	22 (19.5; 24.4)	24 (22; 25.8)	< 0.05
Hospitalisation ward
Medical ward, no (%)	171 (90)	45 (83)	126 (93)	–
Intensive care unit, no (%)	11 (6)	6 (11)	5 (4)	–
Surgical ward, no (%)	7 (4)	3 (5)	4 (3)	–
Need for renal replacement therapy, no (%)	2 (1)	2 (4)	0 (0)
ICU admission at 24 hours	12 (6)	6 (11)	6 (4)
ICU admission at 48 hours	11 (6)	5 (9)	6 (4)
In-hospital LOS, median (Q1; Q3), days	11 (6; 18)	11 (6; 18)	11 (6; 17)
In-hospital death, no (%)	20 (11)	11 (20)	9 (7)
Time before death, median (Q1; Q3), days	7 (4; 23)	5 (4; 14)	18 (8; 27)

*more than 10% missing values.

AKI, acute kidney injury; COPD, chronic obstructive pulmonary disease; CPK, creatinine phosphokinase; CRP, C-reactive protein; DBP, diastolic blood pressure; ICU, intensive care unit; LOS, length of stay; MBP, mean blood pressure; Q1, first quartile; Q3, third quartile; SBP, systolic blood pressure; SGOT, serum glutamic-oxaloacetic transaminase; SGPT, serum glutamic-pyruvic transaminase .

A total of 54 (28%) patients developed AKI within 48 hours. Twenty-four patients (44%) had stage 1 AKI, 10 (19%) stage 2 AKI, 19 (35%) stage 3 AKI and 1 (2%) had undetermined AKI stage. The results of AUC, sensitivity, specificity, negative and positive predictive value, and optimal threshold are presented in table 2. The ROC curve to predict AKI based on NGAL level is presented in figure 3a. The AUC was 0.60 (95% CI 0.51; 0.70). The optimal threshold of NGAL was 128.5 ng/mL for sensitivity of 0.65 and specificity of 0.54. After adjustment on spO2, sex and CPK, the AUC was 0.64 (95% CI 0.54; 0.74) with a sensitivity of 0.44 and specificity of 0.83. Global AUC to predict KDIGO stage was 0.59 (95% CI 0.50; 0.68). ROC curve showing the predictive ability of NGAL for long hospital stays, ICU admission and in-hospital mortality are presented in figure 3b–d respectively. Only two (1%) patients required renal replacement therapy, and thus a ROC curve for this outcome could not be generated.

Figure 3. Receiver operating characteristic curve of the ability of neutrophil gelatinase-associated lipocalin to predict: (a) 48 hours acute kidney injury, (b) hospitalisation over 11 days, (c) intensive care unit admission and (d) in-hospital death. AUC, area under curve.

Table 2

Characteristics of receptor operating channels concerning different endpoints

	AUC (95% CI)	Se	Sp	NPV	PPV	Optimal cut-off
NGAL and 48 hours AKI	0.60 (0.51; 0.70)	0.65	0.54	0.80	0.36	129
Adjusted model * and 48 hours AKI	0.64 (0.54; 0.74)	0.44	0.83	0.80	0.49	–
NGAL and length of stay >11 days	0.63 (0.54; 0.72)	0.92	0.34	0.82	0.57	75
NGAL and ICU admission	0.72 (0.57; 0.87)	0.73	0.66	0.97	0.13	194
NGAL and mortality	0.75 (0.62; 0.87)	0.60	0.87	0.94	0.38	327

*The adjusted model was NGAL + SpO2 + CPK + sex + CPK * sex.

AKI, acute kidney injury; AUC, area under curve; CPK, creatinine phosphokinase; ICU, intensive care unit; NGAL, neutrophil gelatinase–associated lipocalin; NPV, negative predictive value; PPV, positive predictive value; Se, sensitivity; Sp, specificity; SpO2, pulsed oxygen saturation.

Discussion

The ability for NGAL to predict AKI in ED patients admitted for rhabdomyolysis with CPK levels over 1000 UI/L was poor, even after adjustment on SpO2, CPK and sex. Its performances in predicting long hospitalisation were also poor. However, NGAL showed fair performance in predicting ICU admission and in-hospital mortality.

There is a high correlation between pre- and postexercise changes in serum creatinine and serum NGAL, but this does not reliably predict AKI.¹⁹ A meta-analysis on AKI biomarkers in athletes found no clear evidence of a correlation between NGAL elevation and AKI following sport-induced rhabdomyolysis.²⁰ A study on methamphetamine-induced rhabdomyolysis reported that patients with a median CPK level of 653 IU/L on admission only showed a urinary NGAL increase in 10% of those who developed AKI.²¹ In trauma patients with injury severity score above 25, higher serum NGAL levels at admission were observed in those who developed AKI within the first 8 days of hospitalisation.²² The correlation between NGAL and AKI was moderate on different days post-trauma. However, no assessment of trauma severity was conducted in our study for trauma patients. Thus, serum NGAL may be more useful as a monitoring tool for AKI development rather than as a diagnostic tool in patients with acute rhabdomyolysis.

The prevalence of AKI in our study was 28%, which is lower than in other observational studies on ED patients with rhabdomyolysis, where AKI rates ranged from of 42% to 75% during hospitalisation.^{23 24} Patients with CPK levels exceeding 15 000 UI/L have a higher risk of developing AKI during hospitalisation. Patients with CPK levels over 15 000 UI/L have a higher risk of developing AKI during hospitalisation. The relatively low CPK levels in our cohort may explain the lower prevalence of 48-hour AKI. On the other hand, the study protocol mandated intensive hydration based on patients’ diuresis, which may have reduced the incidence of AKI. The evidence supporting hydration strategies for preventing AKI after acute rhabdomyolysis remains weak and needs further investigation.²⁵

In the meta-analysis by Haase et al, it had better AKI prediction from serum NGAL measurements compared with adults (AUC-ROC=0.93 vs AUC-ROC=0.78).²⁶ Age seems to influence the diagnostic performance of NGAL which might explain our results. Moreover, the NGAL cut-off values in the meta-analysis ranged from 100 ng/mL to 278 ng/mL, while the standard cut-off appears to be 150 ng/mL. In another meta-analysis, NGAL’s accuracy for predicting AKI in non-ICU patients had a diagnostic OR of 17.1 (7.8; 37.5), irrespective of AKI aetiology.²⁷ Among patients older than 65 years old admitted to the ED, NGAL had an AUC of 0.78 to predict AKI, with a cut-off of 140 ng/mL.⁸ However, patients in that study were retrospectively included for any medical issue, and NGAL was measured from stored blood samples. In the present study, optimal cut-off to predict 48-hour AKI was 129 ng/mL and 436 ng/mL to predict ICU admission, consistent with existing literature. The median initial serum creatinine level in our study was high (95 (68; 141) µmol/L), and it is possible that some patients already had AKI at inclusion. These patients were not considered having developed AKI during hospitalisation, which could introduce a selection bias.

In this study, AUC-ROC for NGAL in predicting ICU admission and in-hospital death was fair. NGAL has been shown to be a reliable predictor of ICU admission for patients with infections and sepsis.²⁸ It may also serve as a useful biomarker for predicting AKI in ICU patients, especially those with associated sepsis.^{29 30} NGAL’S ability to predict 28-day mortality in septic patients has also been demonstrated.^{31 32} In our study, patients’ illness severity was not assessed using standard scores such as the Acute Physiology and Chronic Health Evaluation II score. An assessment of initial patient severity could have been useful in determining whether high NGAL levels are better predictors of ICU admission than traditional scores in patients with rhabdomyolysis. Further study should explore the potential of NGAL to predict in-hospital death and ICU admission in rhabdomyolysis patients.

Limitations

This study has several limitations that must be highlighted. Patients received aggressive intravenous hydration, aimed at achieving a diuresis >0.3 mL/kg/h, with regular monitoring of urinary pH and administration of bicarbonates—practices not typically standard in ED care. This likely helped prevent myoglobin precipitation in the renal tubules, thereby reducing the incidence of rhabdomyolysis-induced AKI.^{1 25} Additionally, patients with CKD were included, which could have lowered accuracy of the test, since NGAL levels can be elevated in CKD patients.³³ Furthermore, the CPK threshold of 1000 UI/L might have been too low to induce AKI,⁵ although AKI risk increases even with minor CPK elevations.³ The severity of medical and trauma patients was not assessed, and NGAL measurement might be more relevant in severe trauma patients.²² Although the sample size was pre-calculated, it is possible that it was insufficient to detect a strong predictive value for severe AKI. Moreover, convenience sampling could introduce selection bias.³⁴ However, this method is frequently used in ED studies due to the random nature of patient inclusion. Finally, this study was conducted nearly 10 years ago, with initial data analysis delays and multiple writing postponements due to increased ED activity. Nevertheless, the results remain relevant, as few studies on the utility of NGAL in ED patients with rhabdomyolysis have been published since.

Conclusion

In summary, this study is, to our knowledge, the first to prospectively evaluate the diagnostic accuracy of serum NGAL level in patients admitted to the ED for acute rhabdomyolysis regardless of aetiology. The ability for NGAL to predict AKI in patients with rhabdomyolysis was found to be poor. However, its performance in predicting ICU admission and in-hospital mortality may warrant further investigation.

The authors thanks Isabelle Burgos, Julien Bordes, Ludivine Tendron and Yonathan Freund for their help in patients’ inclusion.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).

Ethics approval

This study involves human participants and was approved by the Comité de Protection des Personnes Sud Méditerranée III n°2011-A01059-32. Participants gave informed consent to participate in the study before taking part.

Footnote

Presented at This study has been presented to congress 'Urgences 2024', of the French Emergency Medicine Society in June 2024.

Contributors Funding and protocol redaction: SP, XB and RGG. Manuscript conception: SP, FC and LB. Critical review: LG-M, TM and XB. Statistical analysis: CD. Biological analysis: D-PdB. Guarantor: SP. Corresponding author: FC.

Funding This study was grant by French ministry of Health and Solidarity, PHRC-I/2011/SP-03.

Competing interests None declared.

Patient and public involvement Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Bosch X, Poch E, Grau JM. Rhabdomyolysis and Acute Kidney Injury. N Engl J Med 2009; 361: 62–72. doi:10.1056/NEJMra0801327

² Harrois A, Libert N, Duranteau J. Acute kidney injury in trauma patients. Curr Opin Crit Care 2017; 23: 447–56. doi:10.1097/MCC.0000000000000463

³ Fernandez WG, Hung O, Bruno GR, et al. Factors predictive of acute renal failure and need for hemodialysis among ED patients with rhabdomyolysis. Am J Emerg Med 2005; 23: 1–7. doi:10.1016/j.ajem.2004.09.025

⁴ Holt SG, Moore KP. Pathogenesis and treatment of renal dysfunction in rhabdomyolysis. Intensive Care Med 2001; 27: 803–11. doi:10.1007/s001340100878

⁵ de Meijer AR, Fikkers BG, de Keijzer MH, et al. Serum creatine kinase as predictor of clinical course in rhabdomyolysis: a 5-year intensive care survey. Intensive Care Med 2003; 29: 1121–5. doi:10.1007/s00134-003-1800-5

⁶ Long B, Koyfman A, Gottlieb M. An evidence-based narrative review of the emergency department evaluation and management of rhabdomyolysis. Am J Emerg Med 2019; 37: 518–23. doi:10.1016/j.ajem.2018.12.061

⁷ Kellum JA, Lameire N, for the KDIGO AKI Guideline Work Group. Diagnosis, evaluation, and management of acute kidney injury: a KDIGO summary (Part 1). Crit Care 2013; 17: 204. doi:10.1186/cc11454

⁸ Walls AB, Bengaard AK, Iversen E, et al. Utility of suPAR and NGAL for AKI Risk Stratification and Early Optimization of Renal Risk Medications among Older Patients in the Emergency Department. Pharmaceuticals (Basel) 2021; 14: 843. doi:10.3390/ph14090843

⁹ Supavekin S, Zhang W, Kucherlapati R, et al. Differential gene expression following early renal ischemia/reperfusion. Kidney Int 2003; 63: 1714–24. doi:10.1046/j.1523-1755.2003.00928.x

¹⁰ Mishra J, Dent C, Tarabishi R, et al. Neutrophil gelatinase-associated lipocalin (NGAL) as a biomarker for acute renal injury after cardiac surgery. The Lancet 2005; 365: 1231–8. doi:10.1016/S0140-6736(05)74811-X

¹¹ Constantin J-M, Futier E, Perbet S, et al. Plasma neutrophil gelatinase-associated lipocalin is an early marker of acute kidney injury in adult critically ill patients: a prospective study. J Crit Care 2010; 25: 176. doi:10.1016/j.jcrc.2009.05.010

¹² Nickolas TL, O’Rourke MJ, Yang J, et al. Sensitivity and specificity of a single emergency department measurement of urinary neutrophil gelatinase-associated lipocalin for diagnosing acute kidney injury. Ann Intern Med 2008; 148: 810–9. doi:10.7326/0003-4819-148-11-200806030-00003

¹³ Stevens JS, Xu K, Corker A, et al. Rule Out Acute Kidney Injury in the Emergency Department With a Urinary Dipstick. Kidney Int Rep 2020; 5: 1982–92. doi:10.1016/j.ekir.2020.09.006

¹⁴ Frazee B, Frausto K, Cisse B, et al. Urine Collection in the Emergency Department: What Really Happens. WestJEM 2011; 13: 401–5. doi:10.5811/westjem.2012.1.6855

¹⁵ Bellomo R, Kellum JA, Ronco C. Acute kidney injury. The Lancet 2012; 380: 756–66. doi:10.1016/S0140-6736(11)61454-2

¹⁶ Karvellas CJ, Farhat MR, Sajjad I, et al. A comparison of early versus late initiation of renal replacement therapy in critically ill patients with acute kidney injury: a systematic review and meta-analysis. Crit Care 2011; 15: R72. doi:10.1186/cc10061

¹⁷ Hoste EAJ, De Corte W. Implementing the Kidney Disease. Curr Opin Crit Care 2013; 1. doi:10.1097/MCC.0000000000000039

¹⁸ Nahm FS. Receiver operating characteristic curve: overview and practical use for clinicians. Korean J Anesthesiol 2022; 75: 25–36. doi:10.4097/kja.21209

¹⁹ Lippi G, Sanchis-Gomar F, Salvagno GL, et al. Variation of serum and urinary neutrophil gelatinase associated lipocalin (NGAL) after strenuous physical exercise. Clin Chem Lab Med 2012; 50. doi:10.1515/cclm-2011-0954

²⁰ Wołyniec W, Ratkowski W, Renke J, et al. Changes in Novel AKI Biomarkers after Exercise. A Systematic Review. IJMS 2020; 21: 5673. doi:10.3390/ijms21165673

²¹ Isoardi KZ, Mudge DW, Harris K, et al. Methamphetamine intoxication and acute kidney injury: A prospective observational case series. Nephrology (Carlton) 2020; 25: 758–64. doi:10.1111/nep.13762

²² Sklienka P, Maca J, Neiser J, et al. Physiologic risk factors for early acute kidney injury in severely injured patients. BLL 2020; 121: 779–85. doi:10.4149/BLL_2020_127

²³ Zhu D, Li W, Zhang J, et al. Rhabdomyolysis‐associated acute kidney injury: clinical characteristics and intensive care unit transfer analysis. Intern Med J 2022; 52: 1251–7. doi:10.1111/imj.15308

²⁴ Nielsen FE, Cordtz JJ, Rasmussen TB, et al. The Association Between Rhabdomyolysis, Acute Kidney Injury, Renal Replacement Therapy, and Mortality

²⁵ Michelsen J, Cordtz J, Liboriussen L, et al. Prevention of rhabdomyolysis‐induced acute kidney injury–A DASAIM/DSIT clinical practice guideline. Acta Anaesthesiol Scand 2019; 63: 576–86. doi:10.1111/aas.13308

²⁶ Haase M, Bellomo R, Devarajan P, et al. Accuracy of Neutrophil Gelatinase-Associated Lipocalin (NGAL) in Diagnosis and Prognosis in Acute Kidney Injury: A Systematic Review and Meta-analysis. Am J Kidney Dis 2009; 54: 1012–24. doi:10.1053/j.ajkd.2009.07.020

²⁷ Pan H-C, Yang S-Y, Chiou TT-Y, et al. Comparative accuracy of biomarkers for the prediction of hospital-acquired acute kidney injury: a systematic review and meta-analysis. Crit Care 2022; 26: 349. doi:10.1186/s13054-022-04223-6

²⁸ Min J-H, Lee H, Chung SJ, et al. Neutrophil Gelatinase-Associated Lipocalin for Predicting Intensive Care Unit Admission and Mortality in Patients with Pneumonia. Tohoku J Exp Med 2020; 250: 243–51. doi:10.1620/tjem.250.243

²⁹ Huang C-Y, Shih C-C, Chung K, et al. Predictive value of plasma neutrophil gelatinase-associated lipocalin for acute renal failure in patients with severe sepsis. J Chin Med Assoc 2016; 79: 428–34. doi:10.1016/j.jcma.2016.03.006

³⁰ Imoto Y, Wakasaki A, Izumida K, et al. Analysis of the diagnostic capabilities of urinary neutrophil gelatinase‐associated lipocalin and serum procalcitonin for acute kidney injury at the early stage of critical care intensive care unit admission. Clinical Laboratory Analysis 2021; 35: e23852. doi:10.1002/jcla.23852

³¹ Liu Y, Bu L, Chao Y, et al. Combined Serum NGAL and Fetuin-A to Predict 28-Day Mortality in Patients with Sepsis and Risk Prediction Model Construction. Cell Mol Biol 2022; 68: 47–52. doi:10.14715/cmb/2022.68.11.9

³² Wu Y, Yu C, Zhou Y, et al. Risk stratification and prognostic value of serum neutrophil gelatinase-associated lipocalin (sNGAL) in sepsis patients. Acta Biochim Pol 2022; 69: 113–7. doi:10.18388/abp.2020_5755

³³ Rysz J, Gluba-Brzózka A, Franczyk B, et al. Novel Biomarkers in the Diagnosis of Chronic Kidney Disease and the Prediction of Its Outcome. Int J Mol Sci 2017; 18: 1702. doi:10.3390/ijms18081702

³⁴ Lines T, Burdick C, Dewez X, et al. Nature and extent of selection bias resulting from convenience sampling in the emergency department. Emerg Med J 2022; 39: 325–30. doi:10.1136/emermed-2021-211390