Skip Navigation


NDT Advance Access originally published online on October 1, 2007
Nephrology Dialysis Transplantation 2008 23(2):566-572; doi:10.1093/ndt/gfm638
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
23/2/566    most recent
gfm638v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in NDT
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (8)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Washburn, K. K.
Right arrow Articles by Goldstein, S. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Washburn, K. K.
Right arrow Articles by Goldstein, S. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The author [2007]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org


Urinary interleukin-18 is an acute kidney injury biomarker in critically ill children

Kimberly K. Washburn1,5, Michael Zappitelli1,5, Ayse A. Arikan1, Laura Loftis2, Rajesh Yalavarthy3, Chirag R. Parikh4, Charles L. Edelstein3 and Stuart L. Goldstein1

1Pediatrics, Renal Section, Baylor College of Medicine, Houston, TX, USA, 2Pediatrics, Critical Care Medicine, Baylor College of Medicine, Houston, TX, USA, 3Nephrology and Hypertension, University of Colorado Health Sciences Center, CO, USA and 4Medicine, Nephrology, Yale University School of Medicine, CT, USA

Correspondence and offprint requests to: Stuart Leonard Goldstein, 6621 Fannin Street, MC 3-2482, Houston, TX 77030, USA. Tel: +1-832-824-3800; Fax: +1-832-825-3889. E-mail: stuartg{at}bcm.edu



   Abstract
Top
 Abstract
Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
Background. Urinary interleukin-18 (uIL-18) is an earlier acute kidney injury (AKI) biomarker than serum creatinine (SCr) in specific populations. In the present study, the relationship between uIL-18 and AKI was determined in a heterogeneous group of critically ill children.

Methods. We studied critically ill children to determine whether uIL-18 was an early predictor of AKI. SCr was determined daily for up to 14 days from mechanical ventilation initiation and up to four serial urine specimens were collected for the uIL-18 measurement. AKI was graded by paediatric modified risk, injury, failure, loss, end-stage kidney disease (pRIFLE) criteria. Day 0 was defined as the day of attaining pRIFLE AKI.

Results. One hundred thirty-seven children aged 6.5 ± 6.4 years (53% male) were studied. The peak levels of IL-18 correlated with the severity of AKI by pRIFLE classification (P < 0.05). In non-septic AKI patients, uIL-18 rose to a level higher than control levels 2 days prior to a significant rise in SCr. Urinary IL-18 concentration from the first urine specimen was associated with AKI development within 48 h (odds ratio = 3.5, P < 0.05) independent of the paediatric risk of mortality (PRISM II) score. Urinary IL-18 concentration ≥100 pg/ml had a specificity and negative predictive value of 81 and 83% to predict AKI development within 24 h. Urinary IL-18 ≥200 pg/ml collected within 24 h of Day 0 had a specificity and positive predictive value of 93 and 88% respectively to predict the AKI duration ≥48 h. Urinary IL-18 was associated with mortality (odds ratio = 1.29, P < 0.05), independent of the PRISM II score.

Conclusions. Urinary IL-18 rises prior to SCr in non-septic critically ill children, predicts severity of AKI and is an independent predictor of mortality.

Keywords: acute kidney injury; biomarkers; critically ill children; diagnostic test; sepsis



   Introduction
Top
 Abstract
 Introduction
Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
Acute kidney injury (AKI) is an independent risk factor for mortality in adult and paediatric patients [4,5]. Animal models strongly suggest that early treatment is the key to preventing AKI [6], yet SCr is a late marker of AKI; therefore current translational AKI research is aimed at identifying earlier AKI biomarkers [7–11].

Interleukin-18 (IL-18) is an 18 kDa pro-inflammatory cytokine upregulated during endogenous inflammatory processes and plays an important role in the pathophysiology of sepsis [12]. Urinary IL-18 (uIL-18) of tubular epithelial cell origin mediates acute tubular necrosis in mice, providing the rationale to evaluate uIL-18 as an AKI biomarker [13,14]. Urinary IL-18 was an early AKI marker in critically ill adults participating in the acute respiratory distress syndrome (ARDS) network trial and in children undergoing cardiopulmonary bypass surgery [11,15]. In each of these studies, uIL-18 was also an independent predictor of mortality. It is unknown to what extent uIL-18 serves as an early AKI marker in a heterogeneous group of critically ill children. Given the inter-relationships between AKI, mortality and sepsis in critically ill patients, it is necessary to evaluate the utility of uIL-18 as an AKI biomarker in critically ill patients with a wide variety of diagnoses and varying degrees of underlying infection.

The goals of this study were to determine the association between uIL-18 and AKI in critically ill children, to determine the influence of sepsis and to evaluate whether uIL-18 could serve as an early predictor of AKI in this population.



   Subjects and methods
Top
 Abstract
 Introduction
 Subjects and methods
Results
 Discussion
 Acknowledgements
 References
 
Study design and subject selection
This study was performed concurrently with a prospective study that validated paediatric modified RIFLE (pRIFLE) criteria for defining AKI in critically ill children [2]. Patients aged 1 month to 21 years, admitted to the paediatric intensive care unit (PICU) who received mechanical ventilation and indwelling bladder catheterization, were eligible for enrollment. Patients with end-stage renal disease (ESRD) and immediately following renal transplantation were excluded. Patients with less than two SCr levels or those with no urine specimens were excluded from the present study. Patients were enrolled within 48 h of mechanical ventilation initiation and followed for up to 14 days from enrollment or until PICU discharge, whichever occurred first. The study protocol and consent forms were approved by the Baylor College of Medicine Human Subjects Institutional Review Board prior to study initiation.

Clinical data collection
The clinical variables collected for this study were age, gender, height/weight, admission and discharge diagnoses, renal replacement therapy (RRT) provision and 28-day mortality. Patients were classified as having sepsis if they fulfilled consensus criteria for systemic inflammatory response syndrome, infection, sepsis, severe sepsis or septic shock [16], as determined from PICU admission/discharge summaries and laboratory values. The determination of sepsis status was performed blind to AKI status and uIL-18 levels and was only assigned to subjects who had the diagnosis within the 72 h surrounding the days of urine specimen collection. The paediatric risk of mortality score (PRISM II, a severity of illness/mortality risk measure) was calculated at the day of ICU admission [17].

Laboratory data collection
SCr values were recorded daily as part of routine patient care and retrospectively from the PICU admission day to the day of study enrollment. Estimated creatinine clearance (eCCl) was calculated using the Schwartz formula [18]. Patients were classified daily by pRIFLE criteria for AKI, using the changes in eCCl from baseline eCCl [2]. The pRIFLE criteria for AKI classified patients’ grade of AKI based on changes in eCCl: pRIFLE R (‘Risk’) denotes a ≥25% decrease in eCCl; pRIFLE I (‘Injury’) denotes a ≥50% decrease in eCCl and pRIFLE F (‘Failure’) denotes a 75% decrease in eCCl from baseline renal function. Each subject's first occurrence of any AKI using pRIFLE criteria and the worst pRIFLE stratum (pRIFLEmax) attained in the first 14 days after study enrollment were recorded.

Urine specimen collection
Urine specimens were collected at 2 PM each day, for up to 4 consecutive days, beginning on the day of enrollment or the following day if consent was obtained after 2 PM (Figure 1a). Reasons for not collecting urine samples on all 4 days were bladder catheterization discontinuation, hospital discharge, death or anuria. Urine bags were emptied at 1 PM to collect urine from the previous hour. Anuria was defined as less than 5 ml in the urine collection bag from the hour prior to collection because this was the minimum amount required for processing and storage.


Figure 1
View larger version (35K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Description of urine collection procedures and use of urine specimens with reference to analytic time points: (a) overall urine collection procedure, depicting that study enrollment began shortly after initiation of ventilation and that urine was collected once per day for up to 4 days if urine was available; (b) AKI urine specimens collected prior to AKI development were used to evaluate uIL-18 for early detection of AKI; (c) AKI urine specimens collected within 24 h of AKI by pRIFLE criteria were used to evaluate uIL-18 as a marker of renal injury severity. PICU = paediatric intensive care unit; SCr = serum creatinine; AKI = acute kidney injury; Day 0 = the first day the patient attained AKI; pRIFLEmax = the worst pRIFLE stratum attained; uIL-18 = urine interleukin-18.

 
Specimens were kept on ice until they were centrifuged at 3000 RPM at 4°C for 5 min. The supernatant was aliquotted equally into cryovials and then stored at –80°C until shipment on dry ice to Dr. Edelstein's lab for uIL-18 measurement. IL-18 was measured in human urine using a human IL-18 enzyme-linked immunosorbent assay kit (Medical and Biologic Laboratories, Nagoya, Japan) that specifically detects the mature form of IL-18, as previously described [10,11]. The cross-reactivity of the kit for pro–IL-18 is extremely low. The coefficient of variation of inter- and intra-assay reproducibility for IL-18 concentration ranges from 5 to 10%, corresponding to that reported by the kit manufacturer. The measurements were made in duplicate and in a blinded fashion. Final uIL-18 values were expressed in pg/ml and pg/mg creatinine.

Data management, interpretation and analysis
Using all urine specimens collected, the highest uIL-18 concentration for each patient was denoted as peak uIL-18. Peak uIL-18 concentrations were compared between controls and those with varying degrees of maximal AKI (controls, pRIFLEmax R, I and F).

We also examined for an association between peak uIL-18 and mortality. Using the first urine specimen collected from all subjects, we evaluated whether uIL-18 was an independent predictor of mortality, controlling for the PRISM II score.

For subsequent analyses, only data from urine samples for which SCr was known in the 48 h following urine collection were used (Figure 1b). The data were arranged to define ‘Day 0’ as the first day that a subject attained AKI (an ≥25% decrease in eCCl from baseline). Urine samples collected between 72 h prior to Day 0 (Days –3, –2 and –1) to 72 h after Day 0 (Days 0, +1, +2 and +3) were compared to control uIL-18 concentrations, in order to evaluate whether uIL-18 rose prior to SCr increase. Up to four control urine samples per AKI patient were randomly selected from patients who did not develop AKI, drawn on the same PICU admission day as for the AKI subjects, as representative control uIL-18 concentrations.

We examined whether uIL-18 was predictive of AKI occurrence. Using the first urine specimen collected from AKI subjects who had urine collected prior to AKI development and the first urine specimen collected from controls (Figure 1b), we used logistic regression to determine if uIL-18 concentration predicted AKI development in the following 48 h, independent of the PRISM II score. We also used these urine specimens to calculate the diagnostic characteristics of uIL-18 to predict AKI development.

We evaluated the diagnostic characteristics of uIL-18 collected from Day 0 or Day +1 (within 24 h of AKI attainment) to predict persistent AKI, defined as the lack of complete AKI resolution within 48 h, for 48 h), as an estimate of AKI severity (Figure 1c).

Subgroup analysis of non-septic patients
We repeated all analyses described above, regarding the relationship of uIL-18 and AKI and the evaluation of uIL-18 as a diagnostic marker of AKI, excluding patients with a diagnosis of sepsis. There were too few septic patients to perform any meaningful statistical analyses in this subgroup.

Statistical analysis
Urine IL-18 was non-normally distributed; therefore, non-parametric testing was used to compare uIL-18 concentrations between groups (Mann–Whitney test for two groups and Kruskal–Wallis test for multiple groups). Categorical variables were analyzed using the Fisher's exact test. For the evaluation of diagnostic characteristics, sensitivity and specificity, positive predictive value and negative predictive value were calculated using standard 2 x 2 tables and receiver-operating characteristics (ROC) curves were constructed. Logistic regression was used to determine whether uIL-18 was a predictor of mortality, independent of the PRISM II score and also to determine if uIL-18 concentration was a predictor of AKI development within 48 h. Natural logarithmic transformation was performed on uIL-18 values for inclusion in the regression analyses. All analyses were performed using the Intercooled STATA® statistical software package (College Station, TX, USA).



   Results
Top
 Abstract
 Introduction
 Subjects and methods
 Results
Discussion
 Acknowledgements
 References
 
Patient characteristics
Of the original 150 patients, 137 enrolled in the prospective pRIFLE validation study had urine specimens available for analysis. Five patients were excluded because they had less than two SCr levels drawn and 8 patients had no urine available. Patient characteristics are shown in Table 1 and are described in greater detail in the aforementioned pRIFLE publication [2].


View this table:
[in this window]
[in a new window]

  		Table 1. Patient characteristics

 
Most patients developed AKI in the first week of admission and urine collection generally began very close to the day of attaining AKI (Day 0). Thirty-four patients (24.8%) did not develop AKI during the study period and served as controls; 50 patients (36.5%) developed pRIFLEmax R AKI; 28 (20.4%) pRIFLEmax I and 25 (18.3%) pRIFLEmax F AKI. A total of 420 urine specimens were available from all patients: 4 urine specimens from 69 patients, 3 from 29 patients; 2 from 18 patients and 1 urine specimen from 21 patients. One hundred and four urine specimens were from controls (3.1/control) and 316 urine specimens were from AKI patients (3.1/AKI patient).

Association between peak uIL-18 and AKI
Table 2 demonstrates that peak uIL-18 concentration increased progressively with worsening pRIFLEmax stratum, whether uIL-18 was expressed in pg/ml (P = 0.004) or in pg/mg creatinine (P = 0.0001, Kruskal–Wallis test). All future uIL-18 results are expressed in pg/ml since results expressed in pg/mg creatinine show a similar trend.


View this table:
[in this window]
[in a new window]

  		Table 2. Comparison of peak uIL-18 concentrations between pRIFLEmax strata

 
Association of uIL-18 with mortality
Non-survivors had higher peak uIL-18 concentrations (median [IQR] = 148 [222.9] pg/ml) compared to survivors (56.9 [167.6] pg/ml, P < 0.05). Urinary IL-18 concentrations from the first urine specimen were higher in non-survivors versus survivors (median [IQR] = 114.3 [147.9] versus 24.3 [100.4] pg/ml, P = 0.01). The association of the first urine uIL-18 levels and mortality was independent of the severity of illness by the PRISM II score (adjusted odds ratio: 1.29, 95% CI = 1.01–1.64, P = 0.04).

Urinary interleukin-18 as an early marker of AKI
Figure 2 displays AKI uIL-18 concentrations from 3 days prior to 3 days after AKI development (Day 0), compared to control uIL-18 concentrations. In AKI patients, uIL-18 began to rise at Day –2, peaked at Day 0 and then steadily declined to baseline at Day 3, whereas control uIL-18 concentrations remained unchanged. Urine IL-18 concentrations were significantly higher in AKI versus control patients’ urines from Days 0 to 2 (all P < 0.05).


Figure 2
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Plot of the median values of uIL-18 in patients with AKI (Black) from 3 days before AKI attainment (Day –3) to 3 days after AKI attainment (Day +3). Representative control median uIL-18 levels are shown in gray. The table below the graph displays the number of urine specimens available for analysis on each day examined. *P < 0.05 for the difference between AKI and control uIL-18 concentrations.

 
Urinary interleukin-18 as a diagnostic marker of AKI
The first urine specimen from patients with AKI who had urine collected prior to AKI onset and the first urine specimen from controls were used for the following analyses. Higher uIL-18 was a predictor of developing AKI in the next 24 to 48 h (adjusted odds ratio = 3.7, 95% CI = 1.4 to 9.5, P = 0.008), independent of the PRISM II score (adjusted odds ratio = 1.3, 95% CI = 1.0 to 5.4, P = 0.03) and the interaction term of PRISM II and uIL-18 (adjusted odds ratio = 0.91, 95% CI = 0.85 to 0.97, P = 0.006). Table 3 displays the diagnostic characteristics of different uIL-18 cutoffs to predict AKI development within 24 h. Results for predicting AKI within 48 h were similar, but less strong (not shown). Specificity ranged from 67% for uIL-18 ≥25 pg/ml to 94% for uIL-18 ≥250 pg/ml. Negative predictive value ranged from 82 to 85% at all cutoffs. For all cutoffs, the sensitivity was ≤38% and the positive predictive value was ≤33%.


View this table:
[in this window]
[in a new window]

  		Table 3. Diagnostic characteristics of uIL-18 to predict (1) AKI development within 24 h and (2) predict persistent AKI (duration >48 h)

 
Table 3 also displays the diagnostic characteristics of uIL-18 in AKI patients from Day 0 or 1 (within 24 h of first AKI attainment) to predict severe AKI, defined as AKI duration >48 h. For lower uIL-18 cutoffs (25 and 50 pg/ml) the sensitivity was 79 and 68%, respectively, with specificity of 29 and 50%, respectively; at higher uIL-18 cutoffs (≥75 and 200 pg/ml) specificity was 71 and 93%, respectively. The positive predictive value was ≥80% for cutoffs ≥75 pg/ml, but the negative predictive value was ≤39% at all cutoffs.

Subgroup analyses excluding septic patients
Patients with and without AKI had similar proportions of septic patients (21 versus 22%). Septic patients had higher uIL-18 concentrations than non-septic patients on all 4 days of urine collection (septic versus non-septic median uIL-18, pg/ml: 78.0 versus 25.8, P < 0.02; 42.2 versus 14.0, P < 0.01; 39.9 versus 10.4, P < 0.2 and 51.6 versus 0, P < 0.03).

The time-dependent relationship between uIL-18 and the day of AKI attainment was strengthened when septic patients were removed from the analyses. Figure 3 displays that AKI uIL-18 was higher in ‘non-septic’ patients than in controls between Days –2 and 2 (all P < 0.05).


Figure 3
View larger version (26K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Plot of the median values of uIL-18 in patients with AKI (black) from 3 days before AKI attainment (Day –3) to 3 days after AKI attainment (Day +3), in non-septic patients only. Representative non-septic control median uIL-18 levels are shown in gray. The table below the graph displays the number of urine specimens available for analysis on each day examined. *P < 0.05 for the difference between AKI and control median uIL-18 concentrations. **P < 0.07.

 
When only ‘non-septic patients’ were included in multivariate analysis (with the PRISM II score), the independent relationship between uIL-18 and AKI development in the next 48 h became stronger (adjusted odds ratio = 5.23, 95% CI = 1.61 to 16.84, P = 0.007). All diagnostic characteristics evaluated above were similar in non-septic patients (not shown).



   Discussion
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
Acknowledgements
 References
 
AKI is defined by acute changes in SCr, which is a known late and non-specific marker of AKI [19,20]. AKI is emerging as a public health problem since epidemiologic studies reveal that mild increases in SCr of hospitalized patients are associated with increased morbidity and mortality [4]. Several urinary protein biomarkers have been identified as potential candidates for AKI detection prior to SCr rise.

Animal studies show that intrarenal IL-18 production is involved in the pathophysiology of ischemic AKI independent of neutrophils [14]. Human studies revealed that uIL-18 concentrations rise 24–48 h prior to AKI defined by RIFLE criteria, in a heterogeneous population of critically ill adults [15] and in children undergoing cardiopulmonary bypass (CPB) surgery [11]. In this select paediatric population, the exact timing of kidney injury was known, unlike the PICU setting, where timing of kidney injury is unknown and AKI causes are often multifactorial. The use of uIL-18 in the PICU setting may be further complicated by its role as a mediator of sepsis and its known association with mortality [12,15].

The current study is the first to (1) systematically examine the relationship between uIL-18 and AKI in a heterogeneous group of critically ill children and (2) examine the impact of illness severity on uIL-18 utility as an AKI biomarker. Peak uIL-18 concentrations increased with worsening pRIFLEmax strata. Urinary IL-18 was also associated with patient mortality, consistent with previous studies [10,15], independent of the PRISM II score.

The most important goal for studying AKI biomarkers is AKI detection prior to SCr rise. Urinary IL-18 concentrations began to rise as early as 2 days prior to AKI development by pRIFLE criteria; however AKI uIL-18 was only significantly different from control uIL-18 beginning on Day 0, suggesting that uIL-18 is not an effective early marker of AKI. When septic patients were removed from the analyses, uIL-18 was significantly higher than controls beginning 2 days prior to AKI by pRIFLE criteria. Urinary IL-18 concentrations were associated with development of AKI within 48 h, independent of the PRISM II score. These results mirror those found in critically ill adult patients, where uIL-18 and APACHE II scores (an adult risk of mortality score analogous to the PRISM II score in children) were independent predictors of AKI development in the next 24 h [15]. The relationship between uIL-18 and AKI development in the following 24 to 48 h was strengthened when septic patients were removed from the analysis (the odds ratio increasing from 1.29 to 5.23).

Unfortunately, our sample size was not large enough to evaluate the role of uIL-18 in the subgroup of septic patients. Thus, we suggest that uIL-18 is an early marker of AKI in non-septic critically ill patients and that further study is required in septic patients, who may have variably high uIL-18 concentrations independent of AKI. In addition, because the diagnosis of sepsis mostly occurred very early in PICU admission, very close in time to urine collection, it was not possible to discriminate whether uIL-18 was a better marker of sepsis versus AKI. Future study will need to determine the relative contribution of uIL-18 concentrations from sepsis and AKI.

Though uIL-18 predicted the AKI development in multivariate logistic regression, it is necessary to evaluate the diagnostic characteristics (sensitivity and specificity) to assess the clinical applicability of uIL-18 to predict AKI. The area under the curve (AUC) for uIL-18 to predict AKI development within 24 to 48 h was 0.54; this differs from the AUC found in children undergoing CPB [11]. It may not be surprising that uIL-18 performed less well as a diagnostic test in our heterogeneous paediatric patient population. Moreover, since the timing of the renal ischemic event (aortic clamping) was known in children undergoing CPB surgery, the investigators were able to assess uIL-18 concentrations at specific time points from the ischemic event to the SCr rise. Urine IL-18 had particularly low sensitivity and positive predictive value to predict AKI development; however, the specificity and negative predictive values were similar to those found in previous studies at a cutoff value of approximately 75 pg/ml [11]. This suggests that uIL-18 concentration >75 pg/ml may play a role in the early detection of AKI as a future member of a multiple biomarker panel.

It is often difficult to differentiate fluid-responsive elevations in SCr (‘pre-renal AKI’) from true renal injury as seen with acute tubular necrosis (ATN). Urinary AKI biomarkers may be helpful to differentiate these two conditions as well as be markers of renal injury severity. Urinary IL-18 has previously been shown to be a marker of number of days with AKI [11]. We attempted to evaluate uIL-18 to predict the AKI duration >48 h, as a surrogate marker of fluid responsive AKI. While the AUC was 0.61, the specificity at higher cutoffs of uIL-18 (≥200 pg/ml) was good (93%), suggesting another potential role for uIL-18 in the diagnosis and management of AKI. Unfortunately, we could not rely on other indices of renal hypoperfusion, such as the fractional excretion of sodium, because many of our patients were receiving diuretics. Future research should further explore the role of uIL-18 as a marker that can distinguish pre-renal AKI from ATN as well as being a marker of injury severity.

There were limitations to our study. The primary aim of the study was to describe AKI epidemiology in high-risk critically ill patients [2] and was not powered to specifically detect differences in uIL-18 between patients with versus without AKI. Ideally, we would have collected urine samples from all patients from the day of initiation of ventilation. Although urine specimens were collected within 48 h of initiation of mechanical ventilation, several patients developed AKI prior to urine collection, reducing our sample size for an early AKI detection. The relatively small sample size limited our ability to perform several analyses in the subgroup of septic patients. The illness severity, and thus AKI incidence of our sample, was very high compared to other populations studied [21]. It is possible that the characteristics of uIL-18 as an AKI biomarker in patients who are less severely ill or in general hospitalized patients who may have lower systemic, non-renal IL-18 concentrations, may be different. Future study should elucidate whether our findings are applicable to less severely ill patients. Another limitation to our study is that our AKI definition is based on SCr. SCr is an unreliable marker of glomerular filtration rate (GFR) in the acute setting. Unfortunately, there are no other reference standards upon which to test urinary AKI biomarkers. Therefore, future research should investigate other markers of GFR, such as Cystatin C, to determine whether these are better reference standards upon which urinary biomarkers can be evaluated.

Other urine and serum biomarkers of AKI, such as urine neutrophil gelatinase-associated lipocalin (uNGAL), urine kidney injury molecule-1 (KIM-1) and serum Cystatin C, have been evaluated by other investigators [8,9,11,22]. Urine NGAL has been evaluated only in children and only in those undergoing cardiopulmonary bypass, where the exact timing of kidney injury is known. Therefore, it is not surprising that uNGAL had much higher sensitivity and specificity to predict AKI within 24 to 48 h (>95%) than found in our study [9,11]. Our findings highlight the importance of evaluating AKI biomarkers in different patient populations prior to their widespread use. Future research should eventually evaluate and compare each of these biomarkers in heterogeneous groups of patients, in order to direct their clinical application.

In summary, higher uIL-18 concentrations are associated with worsening AKI severity and uIL-18 measured early after initiation of ventilation is an independent predictor of AKI and of mortality in a heterogeneous group of critically ill children. However, these findings could only be confirmed in non-septic patients. Urine IL-18 rises 2 days prior to SCr in non-septic patients. These findings should stimulate further study in critically ill populations powered to evaluate the effect of sepsis and severity of illness on uIL-18 levels in the setting of AKI.



   Acknowledgements
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
References
 
Dr. Zappitelli is the recipient of a Kidney Research Scientist Core Education and National Training Program post-doctoral research fellowship award. We thank William S. May and Patricia C. Mapua for their assistance with urine collections.

Conflict of interest statement. The authors declare that there are no conflicts of interest associated with this work.



   Notes
 
5KKW and MZ contributed equally to this manuscript and are co-first authors. Back



   References
Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Abosaif NY, Tolba YA, Heap M, et al. The outcome of acute renal failure in the intensive care unit according to RIFLE: model application, sensitivity, and predictability. Am J Kidney Dis (2005) 46:1038–1048.[CrossRef][Web of Science][Medline]
  2. Akcan-Arikan A, Zappitelli M, Loftis LL, et al. Modified RIFLE criteria in critically ill children with acute kidney injury. Kidney Int (2007) 71:1028–1035.[CrossRef][Web of Science][Medline]
  3. Hoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care (2006) 10:R73.[CrossRef][Medline]
  4. Chertow GM BE, Honour M, Bonventre JV, et al. Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. JASN (2005) 16:3365–3370.[Abstract/Free Full Text]
  5. Price J, Mott A, Dickerson H, et al. Worsening renal function in children hospitalized with acute decompensated heart failure: evidence for a paediatric cardiorenal syndrome? Ped Crit Care Med (2007) (accepted).
  6. Schrier RW, Wang W. Acute renal failure and sepsis. N Engl J Med (2004) 351:159–169.[Free Full Text]
  7. Han WK, Bailly V, Abichandani R, et al. Kidney Injury Molecule-1 (KIM-1): a novel biomarker for human renal proximal tubule injury. Kidney Int (2002) 62:237–244.[CrossRef][Web of Science][Medline]
  8. Herget-Rosenthal S, Marggraf G, Husing J, et al. Early detection of acute renal failure by serum cystatin C. Kidney Int (2004) 66:1115–1122.[CrossRef][Web of Science][Medline]
  9. Mishra J, Ma Q, Prada A, et al. Identification of neutrophil gelatinase-associated lipocalin as a novel early urinary biomarker for ischemic renal injury. J Am Soc Nephrol (2003) 14:2534–2543.[Abstract/Free Full Text]
  10. Parikh CR, Jani A, Mishra J, et al. Urine NGAL and IL-18 are predictive biomarkers for delayed graft function following kidney transplantation. Am J Transplant (2006) 6:1639–1645.[CrossRef][Web of Science][Medline]
  11. Parikh CR, Mishra J, Thiessen-Philbrook H, et al. Urinary IL-18 is an early predictive biomarker of acute kidney injury after cardiac surgery. Kidney Int (2006) 70:199–203.[CrossRef][Web of Science][Medline]
  12. Tschoeke SK, Oberholzer A, Moldawer LL. Interleukin-18: a novel prognostic cytokine in bacteria-induced sepsis. Crit Care Med (2006) 34:1225–1233.[CrossRef][Web of Science][Medline]
  13. Melnikov VY, Faubel S, Siegmund B, et al. Neutrophil-independent mechanisms of caspase-1- and IL-18-mediated ischemic acute tubular necrosis in mice. J Clin Invest (2002) 110:1083–1091.[CrossRef][Web of Science][Medline]
  14. Melnikov VY, Ecder T, Fantuzzi G, et al. Impaired IL-18 processing protects caspase-1-deficient mice from ischemic acute renal failure. J Clin Invest (2001) 107:1145–1152.[Web of Science][Medline]
  15. Parikh CR, Abraham E, Ancukiewicz M, et al. Urine IL-18 is an early diagnostic marker for acute kidney injury and predicts mortality in the intensive care unit. J Am Soc Nephrol (2005) 16:3046–3052.[Abstract/Free Full Text]
  16. Goldstein B, Giroir B, Randolph A. International paediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in paediatrics. Pediatr Crit Care Med (2005) 6:2–8.[CrossRef][Medline]
  17. Pollack MM, Ruttimann UE, Getson PR. Paediatric risk of mortality (PRISM) score. Crit Care Med (1988) 16:1110–1116.[Web of Science][Medline]
  18. Schwartz GJ, Haycock GB, Edelmann CMJr, et al. A simple estimate of glomerular filtration rate in children derived from body length and plasma creatinine. Paediatrics (1976) 58:259–263.[Abstract/Free Full Text]
  19. Bellomo R. Defining, quantifying, and classifying acute renal failure. Crit Care Clin (2005) 21:223–237.[CrossRef][Web of Science][Medline]
  20. Devarajan P. Cellular and molecular derangements in acute tubular necrosis. Curr Opin Pediatr (2005) 17:193–199.[CrossRef][Web of Science][Medline]
  21. Bailey D, Phan V, Litalien C, et al. Risk factors of acute renal failure in critically ill children: a prospective descriptive epidemiological study. Pediatr Crit Care Med (2007) 8:29–35.[CrossRef][Web of Science][Medline]
  22. Liangos O, Perianayagam MC, Vaidya VS, et al. Urinary N-acetyl-beta-(d)-glucosaminidase activity and kidney injury molecule-1 level are associated with adverse outcomes in acute renal failure. J Am Soc Nephrol (2007) 18:904–912.[Abstract/Free Full Text]
Received for publication: 23. 5.07
Accepted in revised form: 21. 8.07


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?

Related articles in NDT:

In this issue ...

NDT 2008 23: i. [Extract] [FREE Full Text]  



This article has been cited by other articles:


Home page
 Nephrol Dial TransplantHome page
Z. H. Endre and J. W. Pickering
Outcome definitions in non-dialysis intervention and prevention trials in acute kidney injury (AKI)
Nephrol. Dial. Transplant., October 7, 2009; (2009) gfp501v1.
[Abstract] [Full Text] [PDF]

Home page
 Nephrol Dial TransplantHome page
P. Devarajan
Neutrophil gelatinase-associated lipocalin--an emerging troponin for kidney injury
Nephrol. Dial. Transplant., December 1, 2008; 23(12): 3737 - 3743.
[Full Text] [PDF]

Home page
 CJASNHome page
S. S. Waikar, K. D. Liu, and G. M. Chertow
Diagnosis, Epidemiology and Outcomes of Acute Kidney Injury
Clin. J. Am. Soc. Nephrol., May 1, 2008; 3(3): 844 - 861.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
23/2/566    most recent
gfm638v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in NDT
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (8)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Washburn, K. K.
Right arrow Articles by Goldstein, S. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Washburn, K. K.
Right arrow Articles by Goldstein, S. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?