Available via license: CC BY-NC
Content may be subject to copyright.
Copyright © 2023 The Korean Society of Critical
Care Medicine
This is an Open Access article distributed under
the terms of Creative Attributions Non-Com-
mercial License (https://creativecommons.org/
li-censes/by-nc/4.0/) which permits unrestricted
noncommercial use, distribution, and reproduc-
tion in any medium, provided the original work
is properly cited.
95
https://www.accjournal.org
| pISSN 2586-6052 | eISSN 2586-6060
INTRODUCTION
Acute kidney injury (AKI) is defined as a sudden decline in kidney function. AKI is an inde-
pendent risk factor for increased mortality in patients with severe trauma admitted to the
intensive care unit (ICU) [1-5]. In trauma patients, the incidence of AKI ranges widely from 1%
Background: In patients with severe trauma, the diagnosis of acute kidney injury (AKI) is import-
ant because it is a predictive factor for poor prognosis and can affect patient care. The diagnosis
and staging of AKI are based on change in serum creatinine (SCr) levels from baseline. However,
baseline creatinine levels in patients with traumatic injuries are often unknown, making the diag-
nosis of AKI in trauma patients difficult. This study aimed to enhance the accuracy of AKI diagno-
sis in trauma patients by presenting an appropriate reference creatinine estimate (RCE).
Methods: We reviewed adult patients with severe trauma requiring intensive care unit admission
between 2015 and 2019 (n=3,228) at a single regional trauma center in South Korea. AKI was di-
agnosed based on the current guideline published by the Kidney Disease: Improving Global Out-
comes organization. AKI was determined using the following RCEs: estimated SCr75-modification
of diet in renal disease (MDRD), trauma MDRD (TMDRD), admission creatinine level, and first-day
creatinine nadir. We assessed inclusivity, prognostic ability, and incrementality using the different
RCEs.
Results: The incidence of AKI varied from 15% to 46% according to the RCE used. The receiver
operating characteristic curve of TMDRD used to predict mortality and the need for renal replace-
ment therapy (RRT) had the highest value and was statistically significant (0.797, P<0.001; 0.890,
P=0.002, respectively). In addition, the use of TMDRD resulted in a mortality prognostic ability
and the need for RRT was incremental with AKI stage.
Conclusions: In this study, TMDRD was feasible as a RCE, resulting in optimal post-traumatic AKI
diagnosis and prognosis.
Key Words: acute kidney injury; intensive care unit; mortality; multiple trauma; reference values
Selection of appropriate reference creatinine estimate for
acute kidney injury diagnosis in patients with severe
trauma
Kangho Lee1,2, Dongyeon Ryu1,2, Hohyun Kim1,2, Sungjin Park1,2, Sangbong Lee1,2, Chanik Park1,2, Gilhwan Kim1,2,
Sunhyun Kim1,2, Nahyeon Lee1,2
1Department of Trauma and Surgical Critical Care, Pusan National University Hospital, Busan, Korea
2Biomedical Research Institute, Pusan National University Hospital, Busan, Korea
Received: August 12, 2022
Revised: October 5, 2022
Accepted: October 21, 2022
Corresponding author
Hohyun Kim
Department of Trauma and Surgical
Critical Care, Pusan National University
Hospital, 179 Gudeok-ro, Seo-gu,
Busan 49241, Korea
Tel: +82-51-240-7369
Fax: +82-51-240-7719
Email: gskhh@naver.com
Original Article
Acute and Critical Care 2023 February 38(1):95-103
https://doi.org/10.4266/acc.2022.01046
96 https://www.accjournal.org Acute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
to 50%, seemingly due to inconsistent diagnostic criteria and
unclear estimates of reference creatinine estimates [6].
Accurate diagnosis and management of AKI can help im-
prove survival in severely injured patients [7]. The diagnostic
criteria for AKI have been developed over the past few de-
cades. Recently, the Kidney Disease: Improving Global Out-
comes (KDIGO) group proposed AKI diagnostic criteria based
on urine output (UO) and serum creatinine (SCr) levels [8].
These criteria are used to diagnose and stage AKI according to
the degree of increase in creatinine levels from baseline SCr or
according to UO. However, baseline SCr is often unknown in
trauma patients, making AKI diagnosis difficult.
There are no clear guidelines on how to estimate the refer-
ence creatinine in trauma patients; however, several methods
have been proposed to estimate baseline SCr for AKI diagnosis
in patients with severe trauma. This study aimed to determine
an appropriate reference creatinine estimate for post-traumat-
ic AKI diagnosis. We hypothesized that a more appropriate
reference creatinine estimate is more strongly related to the
diagnosis of AKI and its clinical outcomes, such as mortality.
We assessed the ability of reference creatinine estimates to di-
agnose AKI based on incidence, prognosis, and incrementality
by AKI stage and existing verified reports.
MATERIALS AND METHODS
This study was reviewed and approved by the Institutional Re-
view Board of Pusan National University Hospital (No. H-2004-
020-090). Owing to its retrospective design, informed consent
was waived.
Study Population
In this retrospective cohort study, we reviewed adult patients
with severe trauma from the Korea Trauma Database (KTDB)
between January 1, 2015, and December 31, 2019 at Pusan
National University Hospital, a high-volume regional trauma
center in Busan, South Korea. A total of 4,293 patients had
severe trauma (Injury Severity Score [ISS] ≥16). The exclusion
criteria were age <16 years, ICU admission time <12 hours,
known chronic kidney disease, and unclear laboratory results
or medical records. Accordingly, 3,228 patients were enrolled
in the study (Figure 1). Data on demographic characteristics,
initial vital signs, medical history, injury type, length of ICU
admission, and mortality were obtained from the KTDB. In
addition, laboratory results for seven days were extracted from
the medical records.
Definition and Outcome Measures
AKI was defined and staged based on the current KDIGO
guidelines (Table 1) [8]. Initiation of renal replacement therapy
(RRT) corresponds to stage 3 AKI in the KDIGO guidelines [8].
However, this study did not include RRT as a diagnostic crite-
rion. RRT was included as an outcome variable rather than a
diagnostic criterion in our study. In addition, the KDIGO 2012
guidelines include the UO criterion; however, this study did
not apply it because the accuracy of the available data could
not be verified.
■ Acute kidney injury (AKI) in patients with severe trauma is
an important prognostic factor; however, diagnosing AKI
is challenging because the baseline creatinine level for a
definitive diagnosis is not known.
■ Currently, there is no clear consensus on the estimated
baseline creatinine level that is diagnostic of AKI in pa-
tients with severe trauma.
■ In this study, we compared the diagnostic ability of several
reference creatinine estimates (RCEs) and found trauma
modification of diet in renal disease to be the most suit-
able RCE for AKI diagnosis.
KEY MESSAGES
Figure 1. Flowchart of the study. ISS: Injury Severity Score; ICU:
intensive care unit; CKD: chronic kidney disease.
4,293 Severe traumatic injury
(ISS>16)
Unclear laboratory results
and/or medical record
717 Excluded
3,228 Included
Age ≥16 yr
ICU admission (time >12 hr)
No CKD and/or dialysis history
YesNo
3,576 Included
97
https://www.accjournal.orgAcute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
We identified some reference creatinine estimates that
were used when the patient's baseline creatinine level was
unknown in the trauma literature. The modification of diet in
renal disease (MDRD) equation used for back-calculation was
developed for estimating the glomerular filtration rate (GFR)
and is widely accepted [9]. The equation is as follows:
GFR = 175 × SCr−1.154 × age−0.203 × 0.742 (if female) × 1.212 (if
black).
The estimated serum creatinine 75 (eSCr75)-MDRD back-
calculates the MDRD equation to estimate unidentified SCr,
assuming a lower normal GFR (75 ml/min/1.73 m2), according
to international recommendations [8]. The trauma MDRD
(TMDRD) was also designed to estimate creatinine in the
young and generally healthy trauma population; however, it
uses the highest median GFR (121 ml/min/1.73 m2) demon-
strated by trauma patients during the first week of admission
[10]. The admission creatinine level was defined as the first SCr
measurement after arrival at the hospital [11], and the first-day
nadir was defined as the lowest SCr measured within 24 hours
of arrival [12].
This study assessed four reference creatinine estimates
(eScr75-MDRD, TMDRD, admission creatinine, and first-
day nadir) with the following primary outcomes: inclusivity,
prognostic ability, and incrementality. Inclusivity was assessed
based on AKI incidence. Prognostic ability was assessed using
the estimated odds ratio (OR) of mortality, initiation of RRT,
and area under the curve (AUC). Incrementality was assessed
by evaluating whether mortality and RRT also increased as the
AKI stage increased.
Statistical Analysis
Continuous variables were presented as median and inter-
quartile ranges (IQRs), and categorical variables were pre-
sented as numbers and percentages. The categorical variables
were compared using the chi-square test when appropriate;
otherwise, Fisher’s exact test was used. Continuous variables
were compared using the Wilcoxon rank-sum test based on
the distribution. Modified Poisson regression analysis was per-
formed to estimate the OR. This method has been proposed
as an alternative to log binomial models when convergence is
a problem (as was the case in this study). We used the receiver
operating characteristic (ROC) curve and AUC to evaluate the
prognostic factors predicting mortality. Statistical significance
was defined as P≤0.05. All statistical analyses were performed
using IBM SPSS ver. 20.0. (IBM Corp.) and Stata ver. 14.2
(StataCorp.).
RESULT
Patient Demographics
In total, 3,228 patients were included in the analysis. The pro-
portion of men was higher (77%), and the median age was 57
years (IQR, 44–68 years). Most of the damage mechanisms
were blunt injuries (97%), and the median ISS was 25 (IQR,
19–29). The reference creatinine estimate ranged widely from
0.67 mg/dl (when TMDRD was used) to 1.00 mg/dl (when
eScr75-MDRD was used). The participant characteristics are
presented in Table 2.
Inclusivity
In univariate analysis, patients with AKI diagnosed using
any of the reference creatinine estimates were more severely
injured, older, and had a larger arrival base deficit, higher
mortality rate, and higher rate of RRT requirement compared
with those without AKI. This trend was the same regardless of
Table 1. KDIGO acute kidney injury definition and stage
KDIGO acute kidney injury
Definition • Increase in SCr by ≥0.3 mg/dl within 48 hr or
• Increase in SCr to ≥1.5 times baseline, which is known or
presumed to have occurred within the prior 7 days or
• Urine volume <0.5 ml/kg/hr for 6 hr
Stage
1Scr
• 1.5–1.9 Times baseline or ≥0.3 mg/dl increase
Urine output
• <0.5 ml/kg/hr for 6–12 hr
2Scr
• 2.0–2.9 Times baseline
Urine output
• <0.5 ml/kg/hr for ≥12 hr
3Scr
• 3.0 Times baseline or
• Increase in serum creatinine to ≥4.0 mg/dl or
• Initiation of renal replacement therapy or
• In patients <18 yr, decrease in eGFR <35 ml/min/1.73 m2
Urine output
• <0.3 ml/kg/hr for ≥24 hr or
• Anuria for ≥12 hr
KDIGO: kidney disease: improving global outcomes; SCr: serum creatinine;
eGFR: estimated glomerular filtration rate.
98 https://www.accjournal.org Acute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
the reference creatinine estimate used. Patients with AKI di-
agnosed using admission creatinine levels had no statistically
significant difference in arrival systolic blood pressure (SBP).
The incidence of AKI varied among the reference creatinine
estimates, from 15% (when the admission creatinine level was
used) to 46% (when the TMDRD was used) (Table 3).
Prognostic Ability
The estimated OR of increased mortality and RRT showed
associations with AKI diagnosed using all reference creatinine
estimates after adjusting for age, arrival SBP, ISS, and arrival
base deficit (Table 4). This result is similar to the significant
increase in mortality and RRT according to AKI diagnosis pre-
viously identified in the univariate analysis (P<0.001). A ROC
curve analysis according to each reference creatinine estimate
was performed to determine the sensitivity, specificity, and
positive and negative predictive values of mortality and RRT
by AKI stage. The results are presented in Table 5. The AUCs of
TMDRD in predicting the mortality and initiation of RRT were
Table 2. Characteristics of study participants
Variable Value (n=3,228)
Age (yr) 57 (44 to 68)
Year
2015 363 (11)
2016 707 (22)
2017 679 (21)
2018 771 (24)
2019 708 (22)
Male 2,479 (77)
Blunt mechanism of injury 3,142 (97)
Injury Severity Score 25 (19 to 29)
Arrival systolic blood pressure (mm
Hg)
110 (90 to 140)
Arrival base deficit (mmol/L) 1.5 (−0.9 to 5.1)
Base creatinine (mg/dl)
eSCr75-MDRD 1.00 (0.96 to 1.05)
TMDRD 0.67 (0.64 to 0.70)
Admission creatinine 0.86 (0.72 to 1.05)
First-day nadir 0.75 (0.63 to 0.91)
30-Day ICU-free day 25 (16 to 28)
30-Day hospital-free day 6 (0 to 17)
Length of stay 24 (13 to 44)
Length of ICU 5 (2 to 14)
In-hospital mortality 459 (14)
Values are presented as median (interquartile range) or number (%).
eSCr75: estimated serum creatinine 75; MDRD: modification of diet in renal
disease; TMDRD: trauma MDRD; ICU: intensive care unit.
Table 3. Characteristics of patients with and without acute kidney injury, by reference creatinine estimate
Variable
eSCr75-MDRD Trauma MDRD Admission creatinine First-day nadir
No AKI
(n=2,550, 79%)
AKI
(n=678, 21%) P-value No AKI
(n=1,740, 54%)
AKI
(n=1,488, 46%) P-value No AKI
(n=2,746, 85%)
AKI
(n=482, 15%) P-value No AKI
(n=2,260, 70%)
AKI
(n=968, 30%) P-value
Age (yr) 56 (42–66) 64 (51–75) <0.001 54 (40–64) 61 (49–73) <0.001 56 (42–67) 62 (51–74) <0.001 57 (43–67) 58 (46–70) 0.004
Male 1,963 (77) 516 (76.1) 0.632 1,328 (76) 1,151 (77) 0.489 2,096 (76) 383 (80) 0.133 1,729 (77) 750 (77) 0.548
ISS 24 (19–29) 27 (22–34) <0.001 22 (18–27) 26 (21–33) <0.001 24 (19–29) 27 (24–35) <0.001 22 (18–29) 26 (22–34) <0.001
SBP (mm Hg) 110 (100–140) 100 (70–140) <0.001 120 (100–140) 100 (80–140) <0.001 110 (90–140) 119 (80–150) 0.731 120 (100–140) 100 (70–140) <0.001
Arrival base deficit (mmol/L) 1.0 (−1.2 to 4.0) 4.5 (0.6–8.8) <0.001 0.5 (−1.5 to 3.2) 3.1 (0–7.0) <0.001 1.3 (−1 to 4.6) 3.3 (0–7.4) <0.001 0.8 (−1.3 to 3.6) 4.0 (0.4–7.9) <0.001
Mortality 183 (7.2) 276 (40.7) <0.001 76 (4.4) 383 (25.7) <0.001 211 (8) 248 (51) <0.001 152 (7) 307 (32) <0.001
RRT 20 (0.8) 90 (13.3) <0.001 6 (0.3) 104 (7.0) <0.001 32 (1) 78 (16) <0.001 17(0.8) 93 (10) <0.001
Values are presented as median (interquartile range) or number (%).
eSCr75: estimated serum creatinine 75; MDRD: modification of diet in renal disease; AKI: acute kidney injury; ISS: Injury Severity Score; SBP: systolic blood pressure; RRT: renal replacement therapy.
99
https://www.accjournal.orgAcute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
Table 4. Estimated odds ratio of mortality and renal replacement
therapy by acute kidney injury diagnosis, adjusted for age, arrival
systolic blood pressure, Injury Severity Score, and arrival base deficit
Variable Mortality Renal replacement therapy
OR (95% CI) P-value OR (95% CI) P-value
eSCr75-MDRD 5.4 (4.2–6.8) <0.001 14.5 (8.5–24.7) <0.001
TMDRD 4.6 (3.5–6.1) <0.001 13.8 (5.9–32.2) <0.001
Admission
creatinine
9.0 (7.1–11.5) <0.001 13.1 (8.3–20.6) <0.001
First-day nadir 4.3 (3.4–5.4) <0.001 9.5 (5.5–16.3) <0.001
OR: odds ratio; CI: confidence interval; eSCr75: estimated serum creatinine
75; MDRD: modification of diet in renal disease; TMDRD: trauma MDRD.
higher than those of the other estimates, and this difference
was statistically significant (AUC: 0.797, P<0.001; AUC: 0.890,
P=0.002) (Figure 2).
Table 5. Sensitivity, specificity, PPV, and NPV value of mortality and renal replacement therapy by acute kidney injury stage
Variable Mortality Renal replacement therapy
Sensitivity Specificity PPV NPV Sensitivity Specificity PPV NPV
eSCr75-MDRD 25.3 (21.4–29.5) 98.1 (97.5–98.6) 69.0 (61.5–75.9) 88.8 (87.6–89.9) 64.5 (54.9–73.4) 96.9 (96.2–97.5) 42.3 (34.7–50.1) 98.7 (98.3–99.1)
TMDRD 41.6 (37.1–46.3) 95.2 (94.3–96.0) 59.0 (53.4–64.4) 90.8 (89.7–91.8) 75.5 (66.3–83.2) 92.3 (91.3–93.2) 25.6 (21.0–30.7) 99.1 (98.7–99.4)
Admission
creatinine
22.9 (19.1–27.0) 98.5 (98.0–98.9) 71.9 (63.9–79.0) 88.5 (87.3–89.6) 57.3 (47.5–66.7) 97.3 (96.7–97.9) 43.2 (35.0–51.6) 98.5 (98.0–98.9)
First-day nadir 27.9 (23.8–32.2) 98.3 (97.7–98.7) 72.7 (65.5–79.2) 89.2 (88.0–90.2) 65.5 (55.8–74.3) 96.7 (96.0–97.3) 40.9 (33.6–48.6) 98.8 (98.3–99.1)
Values are presented as percent (95% confidence interval).
PPV: positive predictive value; NPV: negative predictive value; eSCr75: estimated serum creatinine 75; MDRD: modification of diet in renal disease; TMDRD:
trauma MDRD.
Figure 2. Comparison of estimated serum creatinine 75 (eSCr75) modification of diet in renal disease (MDRD), trauma MDRD (TMDRD), admission
creatinine, and first-day nadir creatinine area under the curve (AUC) for predicting mortality (A) and the need for renal replacement therapy (B).
Values are presented as AUC (95% confidence interval).
1.00
0.75
0.50
0.25
0
P<0.001 P=0.002
1.00
0.75
0.50
0.25
0
Sensitivity
Sensitivity
1-Specificity
eScr75-MDRD (0.746 [0.721–0.770])
Admission creatinine (0.736 [0.712–0.760])
Reference
eScr75-MDRD (0.863 [0.822–0.905])
Admission creatinine (0.818 [0.771–0.864])
Reference
TMDRD (0.890 [0.859–0.922])
First-day nadir (0.864 [0.822–0.906])
TMDRD (0.797 [0.774–0.820])
First-day nadir (0.755 [0.730–0.780])
1-Specificity
1.00 1.000.75 0.750.25 0.250.50 0.500 0
AABB
Incrementality
The estimated OR of mortality and RRT by AKI stage showed
that it increased with each increase in AKI stage, and all were
statistically significant after adjusting for age, arrival SBP, ISS,
and arrival base deficit (Table 6). The mortality rate and re-
quirement of RRT according to AKI stage was calculated for
each reference creatinine estimate (Figure 3). The association
of the increase in mortality and RRT with the increasing stage
only occurred with TMDRD, and not in the other reference
creatinine estimates.
DISCUSSION
In this study, there were variations in the strength of associ-
ation between AKI diagnosis and mortality based on the ref-
erence creatinine estimate used. The TMDRD reference cre-
100 https://www.accjournal.org Acute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
atinine estimate was the most relevant in terms of incidence,
prognosis, and incrementality, and it was the most appropriate
for AKI diagnosis according to our hypothesis. AKI in patients
with severe trauma admitted to the ICU is a common compli-
cation with considerable mortality. Moreover, severe injury is a
strong risk factor for AKI [1-5]. Indeed, trauma patients experi-
ence single or multiple exposures to AKI risk factors, including
severe trauma, hypovolemic shock, rhabdomyolysis, and ab-
dominal compartment syndrome [2,13-16].
According to several guidelines, a diagnosis of AKI should
be made based on an increase in SCr from the reference val-
ue. However, the choice of reference creatinine estimate for
post-traumatic AKI diagnosis remains controversial. Various
approaches have been used to define reference creatinine
Table 6. Estimated odds ratio of mortality and renal replacement therapy by acute kidney injury stage, adjusted for age, arrival systolic blood
pressure, Injury Severity Score, and arrival base deficit
Variable Mortality Renal replacement therapy
OR (95% CI) P-value OR (95% CI) P-value
eSCr75-MDRD
Stage 1 2.6 (1.9–3.6) <0.001 3.5 (1.7–7.4) 0.001
Stage 2 7.1 (4.8–10.4) <0.001 4.1 (1.5–11.0) 0.005
Stage 3 17.3 (11.8–25.4) <0.001 81.1 (44.4–150.0) <0.001
TMDRD
Stage 1 2.3 (1.6–3.2) <0.001 3.0 (1.1–8.6) 0.037
Stage 2 4.6 (3.2–6.6) <0.001 8.2 (3.0–22.8) <0.001
Stage 3 18.5 (13.0–26.2) <0.001 93.8 (38.3–229.4) <0.001
Admission creatinine
Stage 1 4.1 (2.9–5.7) <0.001 2.4 (1.1–5.3) 0.027
Stage 2 15.5 (9.7–24.6) <0.001 4.0 (1.5–10.2) 0.004
Stage 3 23.2 (15.4–35.0) <0.001 56.3 (33.4–94.8) <0.001
First-day nadir
Stage 1 1.8 (1.3–2.4) <0.001 2.0 (1.0–4.3) 0.053
Stage 2 8.7 (5.9–12.9) <0.001 4.4 (1.7–11.3) 0.002
Stage 3 25.1 (16.9–37.2) <0.001 71.9 (39.3–131.6) <0.001
OR: odds ratio; CI: confidence interval; eSCr75: estimated serum creatinine 75; MDRD: modification of diet in renal disease; TMDRD: trauma MDRD.
Figure 3. Mortality (A) and the need for renal replacement therapy (B) by acute kidney injury (AKI) stage. eSCr75: estimated serum creatinine 75;
MDRD: modification of diet in renal disease.
80
70
60
50
40
30
20
10
0
80
70
60
50
40
30
20
10
0
Mortality (%)
Renal replacement therapy (%)
eSCr75-
MDRD
25.3
15.2
19.6
39.9
eSCr75-
MDRD
64.5
5.4
11.8
18.2
Trauma
MDRD
41.6
21.3
20.5
16.6
Trauma
MDRD
75.4
10.9
8.2
5.4
Admission
creatinine
22.9
14.2
17.0
46.0
Admission
creatinine
57.3
5.4
8.2
29.1
First-day
Nadir
27.9
15.9
23.1
33.1
First-day
Nadir
65.4
6.4
12.7
15.4
■ No AKI ■ Stage 1 ■ Stage 2 ■ Stage 3 ■ No AKI ■ Stage 1 ■ Stage 2 ■ Stage 3
AABB
101
https://www.accjournal.orgAcute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
estimates [17,18]. Common approaches include the use of
admission creatinine, the lowest inpatient creatinine, or
other surrogates, such as MDRD. Without reliable baseline
SCr, the KDIGO 2012 guideline recommends that SCr be esti-
mated using the back-calculated MDRD equation assuming
a lower normal GFR of 75 mL/min/1.73 m2 [8]. Although a
recent study questioned the reliability of estimated creatinine
clearance, several studies have reported that eSCr75-MDRD
appears applicable in critical care [17,19,20]. In contrast, TM-
DRD may be more appropriate than eScr75-MDRD for the di-
agnosis and prognosis of AKI in the trauma population [10,18].
In our study, TMDRD appeared to be more useful than other
estimates (eScr75-MDRD, admission creatinine, and first-day
nadir) as a reference creatinine estimate for the diagnosis of
AKI in patients with severe trauma.
The incidence of AKI in trauma patients is reportedly in the
range of 1%–76% because of the different AKI criteria, refer-
ence creatinine estimates, levels of trauma severity, and length
of follow-up used [6,21-23]. In our study, there was a wide
range in AKI incidence, from 15% (when admission creatinine
was used) to 46% (when TMDRD was used). This result is
comparable to that previously reported. In a recent systematic
review and meta-analysis of trauma patients admitted to criti-
cal care, the overall incidence of AKI was 20.4% without a clear
reference creatinine estimate. The incidence of AKI increased
by 31.9% in studies that focused mainly on blunt injuries [21].
In another systematic review and meta-analysis of trauma pa-
tients in the ICU, the overall mean incidence of post-traumatic
AKI was 24% (95% confidence interval [CI], 20%– 29%), with
considerable heterogeneity in significant outcomes [22].
According to a large cohort study, the in-hospital mortality
of patients with AKI was 27% after adjusting for the KDIGO
stage and differences in age, sex, and severity of illness (OR,
1.13–2.20; P<0.001) in various patients admitted to the ICU
[24]. A meta-analysis of trauma patients found that the mor-
tality rate in patients with AKI was 27% (95% CI, 20%–35%)
[22]. Haines et al. [21] reported that the pooled relative risk
of death from AKI was 3.6 (95% CI, 2.4–5.3). In our study, the
estimated OR of mortality ranged from 4.3 (when the first-day
creatinine nadir was used) to 9.0 (when admission creatinine
was used). The estimated OR of mortality of TMDRD was 4.6
(95% CI, 3.5–6.1). In particular, the AUC value of TMDRD was
higher than that of the other estimates, and this difference was
statistically significant (Figure 2A). The TMDRD reference cre-
atinine estimate resulted in a diagnosis that was prognostic of
both mortality and incrementality for each AKI stage (Figure
3A). Although there were some differences in patient char-
acteristics and diagnostic criteria between our study and the
others, the mortality rate of TMDRD (25.7%) was comparable
to that in the other observations [21,22,24].
RRT is the only supportive measure in patients with severe
AKI. The KDIGO guidelines recommend immediate initiation
of RRT if an absolute indication exists [8]. The initiation of
RRT before the onset of major complications has conceivable
advantages for patients with severe AKI [25]. In our study, the
AUC value of TMDRD used to predict the requirement of RRT
was higher than that of the other estimates, and this difference
was statistically significant (Figure 2B). The association of the
increase in RRT with the increasing stage only occurred with
TMDRD and not in the other reference creatinine estimates
(Figure 3B).
There are three main differences between our study and
other studies that have suggested methods of approximating
reference creatinine estimates to diagnose AKI in patients with
trauma. First, our study was based on a relatively large number
of patients and compared the adequacy of previously known
reference creatinine estimates in patients with severe trauma.
Saour et al. [10] assessed MDRD performance in predicting
SCr in a severe trauma population of 775 patients. In contrast,
3,228 patients were included in our study. Second, the severity
of trauma in patients included in the present study was more
than that in patients included in the other studies. Saour et
al. [10] reported a mean ISS of 19. Hatton et al. [23] reported
a median ISS of 20. In contrast, the median ISS in this study
was 25. We believe that the difference in results between our
study and others may have been influenced by the severity of
trauma. Third, in contrast to the diversity in population com-
position and trauma mechanisms in other studies, all patients
in our study were Asian (100%), and 97% had suffered blunt in-
juries. In the study by Hatton et al. [23], most participants were
Caucasian (52%) or Hispanic (22%), while African Americans
comprised only 17% of the population and Asians only 2%. In
addition, participants with a blunt injury accounted for 85%
of the cohort [23]. Those authors reported that eSCr75-MDRD
may be more useful than TMDRD for the diagnosis and prog-
nosis of AKI [23]. In contrast, TMDRD appeared to be a more
appropriate reference creatinine estimate than eSCr75-MDRD
in our study. This is thought to be on account of the differ-
ences in race composition and injury mechanism. Although
the role of race adjustment in estimating kidney function is
controversial [26,27], we believe that racial differences must be
considered when selecting an appropriate reference value for
102 https://www.accjournal.org Acute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
creatinine.
This study had several limitations. First, UO could not be
applied as a diagnostic criterion because of our inability to ac-
curately verify UO. UO records before admission to the ICU (via
the emergency, operating, or angiography room) were miss-
ing. In studies where both SCr and UO were used as diagnostic
criteria, the incidence of AKI increased from 24% based on SCr
alone to 52% when UO was added as a diagnostic criterion.
Moreover, the risk of death was greatest when patients met
both SCr levels and UO criteria for AKI [28,29]. Thus, our study
may have underestimated the incidence and mortality of AKI.
Second, mortality or initiation of RRT is one of the strongest in-
dicators of prognosis in patients with AKI. However, there are
several other indicators of prognosis, such as complications,
duration of RRT, and renal disease progression to chronic kid-
ney disease. However, investigation of this aspect was limited
by model structure. Finally, this was a single-center retrospec-
tive study, and the results were insufficient for drawing conclu-
sions. Additional multicenter, prospective, randomized con-
trolled trials are necessary to confirm the validity of TMDRD as
a reference creatinine estimate.
In the current study, TMDRD was the most appropriate
reference creatinine estimate (in terms of inclusivity, prog-
nostic ability, and incrementality) for diagnosing and staging
post-traumatic AKI. When using TMDRD as a diagnostic crite-
rion, the incidence of AKI in patients with severe trauma was
approximately 46%. If complementary prospective random-
ized controlled, multicenter comparative studies are conduct-
ed in the future, we believe a more definitive conclusion can
be reached.
CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was re-
ported.
FUNDING
This work was supported by a clinical research grant from Pu-
san National University Hospital in 2022.
ACKNOWLEDGMENTS
None.
ORCID
Kangho Lee https://orcid.org/0000-0003-4089-8831
Dongyeon Ryu https://orcid.org/0000-0003-2832-9572
Hohyun Kim https://orcid.org/0000-0001-9434-8654
Sungjin Park https://orcid.org/0000-0002-4852-5442
Sangbong Lee https://orcid.org/0000-0001-9396-1000
Chanik Park https://orcid.org/0000-0001-8788-2381
Gilhwan Kim https://orcid.org/0000-0002-5449-5681
Sunhyun Kim https://orcid.org/0000-0003-4868-0319
Nahyeon Lee https://orcid.org/0000-0001-5242-1454
AUTHOR CONTRIBUTIONS
Conceptualization: HK. Data curation: KL, DR, GK. Formal
analysis: HK, KL, SP. Methodology: HK, KL, CP. Project ad-
ministration: HK, KL. Visualization: HK, KL, SK, NL. Writing–
original draft: KL. Writing–review & editing: all authors.
REFERENCES
1. Bihorac A, Delano MJ, Schold JD, Lopez MC, Nathens AB, Maier
RV, et al. Incidence, clinical predictors, genomics, and out-
come of acute kidney injury among trauma patients. Ann Surg
2010;252:158-65.
2. Eriksson M, Brattström O, Mårtensson J, Larsson E, Oldner A.
Acute kidney injury following severe trauma: Risk factors and
long-term outcome. J Trauma Acute Care Surg 2015;79:407-12.
3. Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu H, Morg-
era S, et al. Acute renal failure in critically ill patients: a multina-
tional, multicenter study. JAMA 2005;294:813-8.
4. Lai WH, Rau CS, Wu SC, Chen YC, Kuo PJ, Hsu SY, et al.
Post-traumatic acute kidney injury: a cross-sectional study
of trauma patients. Scand J Trauma Resusc Emerg Med
2016;24:136.
5. Harrois A, Soyer B, Gauss T, Hamada S, Raux M, Duranteau
J, et al. Prevalence and risk factors for acute kidney injury
among trauma patients: a multicenter cohort study. Crit Care
2018;22:344.
6. Harrois A, Libert N, Duranteau J. Acute kidney injury in trauma
patients. Curr Opin Crit Care 2017;23:447-56.
7. National Institute for Health and Care Excellence. Acute kidney
injury: prevention, detection and management of acute kidney
injury up to the point of renal replacement therapy [Internet].
National Institute for Health and Care Excellence; 2019 [cited
2022 Jul 28]. Available from: https://www.nice.org.uk/guid-
103
https://www.accjournal.orgAcute and Critical Care 2023 February 38(1):95-103
Lee K, et al. Diagnosis of AKI in severe trauma patients
ance/ng148
8. Khwaja A. KDIGO clinical practice guidelines for acute kidney
injury. Nephron Clin Pract 2012;120:c179-84.
9. Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney
function: measured and estimated glomerular filtration rate. N
Engl J Med 2006;354:2473-83.
10. Saour M, Klouche K, Deras P, Damou A, Capdevila X, Charbit
J. Assessment of modification of diet in renal disease equation
to predict reference serum creatinine value in severe trauma
patients: lessons from an observational study of 775 cases. Ann
Surg 2016;263:814-20.
11. Brandt MM, Falvo AJ, Rubinfeld IS, Blyden D, Durrani NK, Horst
HM. Renal dysfunction in trauma: even a little costs a lot. J
Trauma 2007;62:1362-4.
12. Podoll AS, Kozar R, Holcomb JB, Finkel KW. Incidence and
outcome of early acute kidney injury in critically-ill trauma pa-
tients. PLoS One 2013;8:e77376.
13. Hultström M. Neurohormonal interactions on the renal oxygen
delivery and consumption in haemorrhagic shock-induced
acute kidney injury. Acta Physiol (Oxf) 2013;209:11-25.
14. Elterman J, Zonies D, Stewart I, Fang R, Schreiber M. Rhabdo-
myolysis and acute kidney injury in the injured war fighter. J
Trauma Acute Care Surg 2015;79(4 Suppl 2):S171-4.
15. Bosch X, Poch E, Grau JM. Rhabdomyolysis and acute kidney
injury. N Engl J Med 2009;361:62-72.
16. Mohmand H, Goldfarb S. Renal dysfunction associated with in-
tra-abdominal hypertension and the abdominal compartment
syndrome. J Am Soc Nephrol 2011;22:615-21.
17. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC,
De Bacquer D, et al. RIFLE criteria for acute kidney injury are
associated with hospital mortality in critically ill patients: a co-
hort analysis. Crit Care 2006;10:R73.
18. Bagshaw SM, George C, Gibney RT, Bellomo R. A multi-center
evaluation of early acute kidney injury in critically ill trauma
patients. Ren Fail 2008;30:581-9.
19. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P, Acute
Dialysis Quality Initiative workgroup. Acute renal failure: defini-
tion, outcome measures, animal models, fluid therapy and in-
formation technology needs: the Second International Consen-
sus Conference of the Acute Dialysis Quality Initiative (ADQI)
Group. Crit Care 2004;8:R204-12.
20. Kadivarian S, Heydarpour F, Karimpour H, Shahbazi F. Mea-
sured versus estimated creatinine clearance in critically ill pa-
tients with acute kidney injury: an observational study. Acute
Crit Care 2022;37:185-92.
21. Haines RW, Fowler AJ, Kirwan CJ, Prowle JR. The incidence and
associations of acute kidney injury in trauma patients admitted
to critical care: a systematic review and meta-analysis. J Trauma
Acute Care Surg 2019;86:141-7.
22. Søvik S, Isachsen MS, Nordhuus KM, Tveiten CK, Eken T, Sunde
K, et al. Acute kidney injury in trauma patients admitted to the
ICU: a systematic review and meta-analysis. Intensive Care Med
2019;45:407-19.
23. Hatton GE, Du RE, Pedroza C, Wei S, Harvin JA, Finkel KW, et al.
Choice of reference creatinine for post-traumatic acute kidney
injury diagnosis. J Am Coll Surg 2019;229:580-8.
24. Srisawat N, Sileanu FE, Murugan R, Bellomod R, Calzavacca
P, Cartin-Ceba R, et al. Variation in risk and mortality of acute
kidney injury in critically ill patients: a multicenter study. Am J
Nephrol 2015;41:81-8.
25. STARRT-AKI Investigators; Canadian Critical Care Trials Group;
Australian and New Zealand Intensive Care Society Clinical
Trials Group; United Kingdom Critical Care Research Group;
Canadian Nephrology Trials Network; Irish Critical Care Trials
Group; et al. Timing of initiation of renal-replacement therapy
in acute kidney injury. N Engl J Med 2020;383:240-51.
26. Lee K, Kim H, Ryu D. Choice of reference value of creatinine for
diagnosing post-traumatic acute kidney injury: consider ethnic
and racial disparities. J Am Coll Surg 2020;231:497.
27. Eneanya ND, Yang W, Reese PP. Reconsidering the conse-
quences of using race to estimate kidney function. JAMA
2019;322:113-4.
28. Macedo E, Malhotra R, Bouchard J, Wynn SK, Mehta RL. Oli-
guria is an early predictor of higher mortality in critically ill pa-
tients. Kidney Int 2011;80:760-7.
29. Kellum JA, Sileanu FE, Murugan R, Lucko N, Shaw AD, Clermont
G. Classifying AKI by urine output versus serum creatinine lev-
el. J Am Soc Nephrol 2015;26:2231-8.
