Abstract
Background
Congenital obstructive uropathy (OU) is a leading cause of pediatric kidney failure, representing a unique mechanism of injury, in part from renal tubular stretch and ischemia. Tubular injury biomarkers have potential to improve OU-specific risk stratification.
Methods
Patients with OU were identified in the Chronic Kidney Disease in Children (CKiD) study. “Cases” were defined as individuals receiving any kidney replacement therapy (KRT), while “controls” were age- and time-on-study matched and KRT free at last study visit. Urine and plasma neutrophil gelatinase-associated lipocalin (NGAL), interleukin 18 (IL-18), and liver-type fatty acid-binding protein (L-FABP) levels were measured at enrollment and annually and compared between cases and controls. Urine values were normalized to urine creatinine.
Results
In total, 22 cases and 22 controls were identified, with median (interquartile range) ages of 10.5 (9.0–13.0) and 15.9(13.9–16.9) years at baseline and outcome, respectively. At enrollment there were no differences noted between cases and controls for any urine (u) or plasma (p) biomarker measured. However, the mean pNGAL and uL-FABP/creatinine increased throughout the study period in cases (15.38 ng/ml per year and 0.20 ng/ml per mg/dl per year, respectively, p = 0.01 for both) but remained stable in controls. This remained constant after controlling for baseline glomerular filtration rate (GFR).
Conclusions
In children with OU, pNGAL and uL-FABP levels increased over the 5 years preceding KRT; independent of baseline GFR. Future studies are necessary to identify optimal cutoff values and to determine if these markers outperform current clinical predictors.
Keywords: CKiD, Obstructive uropathy, Biomarkers, NGAL, L-FABP, Kidney, Children
Introduction
Congenital obstructive uropathy (OU) represents a group of rare diseases affecting the fetal urinary tract. Although there are numerous fetal anomalies described that can lead to OU, the vast majority of severe cases results from posterior urethral valves (PUV). The hallmark of PUV is outflow obstruction at the level of the posterior urethra by leaf-like projections resulting in varying levels of injury to the bladder, ureters, and kidneys. This obstruction during a critical time of fetal kidney development often leads to kidney dysplasia. The clinical course however can be quite varied, ranging from fetal demise to subclinical kidney disease. Although early morbidity and mortality is often the result of pulmonary hypoplasia, those who survive the newborn period continue to be at increased risk of progressive kidney disease with OU representing a leading cause of kidney failure and transplantation during childhood [1–3].
Not all children with OU, however, ultimately progress to kidney failure. For those who do, development of kidney failure can occur at any point during childhood or even adulthood with great variability between patients. This spectrum of kidney outcomes lends to the importance of risk stratification and individualized care. In children with PUV, measurements of serum nadir creatinine in the first year of life predict progression based on cutoff values for those patients at clinical extremes. Children with intermediate values, however, show a less predictable course and are much more difficult to risk stratify [4]. Compounding the difficulty in risk stratification for children with a moderate level of baseline kidney injury is that serial insults throughout childhood from recurrent obstruction, nephrotoxins, dehydration, and acute kidney injury (AKI) likely contribute to the variable nature of progression. Thus, markers that can be measured at both baseline and over time are necessary to improve risk stratification of this subset of patients.
There is emerging evidence to suggest that commonly measured clinical biomarkers predict risk of progression in children with chronic kidney disease (CKD) of both glomerular and non-glomerular etiologies, including OU. These potential markers include glomerular filtration rate (GFR), serum phosphate, bicarbonate, albumin and hemoglobin; and urine protein/creatinine (Cr) and microalbumin/Cr ratios [5, 6]. There continues to be, however, the need for more disease-specific biomarkers to improve the accuracy of predicting clinical progression. For instance, in children with OU, markers of tubular damage that are released due to uroepithelial stretch injury may provide valuable information on both initial obstructive injury and ongoing or recurrent obstruction throughout childhood. By focusing on markers that reflect the mechanism of injury in CKD, there is opportunity to develop more personalized care plans and in turn improve patient-specific risk stratification. A review of contemporary literature identified neutrophil gelatinase-associated lipocalin (NGAL), interleukin 18 (IL-18), and liver-type fatty acid-binding protein (L-FABP) as candidate biomarkers of tubular injury.
NGAL is secreted by α-intercalated cells within the kidney collecting ducts in response to ischemic kidney injury. Blood and urine NGAL (uNGAL) levels accumulate following acute and chronic kidney injury, and serum NGAL concentrations highly correlate with serum Cr [7]. In children with CKD, serum NGAL closely correlates with GFR and outperforms estimations of GFR at more advanced stages of disease [8]. uNGAL also predicts obstruction on renal scan in children with hydronephrosis [9]. IL-18 is a pro-inflammatory cytokine and member of the interleukin-1 family with its inactive precursor synthesized in proximal tubule epithelial cells. It is activated by caspase-1 and thought to be a mediator in kidney ischemia-reperfusion injury by acting to induce tubular necrosis [10, 11]. In mice, both inhibition of caspase-1 and IL-18 neutralizing antiserum offer protection during kidney ischemia. In animal models and human patients, serum levels of IL-18 are elevated with kidney injury, including tubular necrosis and delayed graft function [12–14]. L-FABP is expressed in the proximal tubules, where it participates in fatty acid metabolism. Urinary levels of L-FABP (uL-FABP) are increased in states of elevated stress including tubular stretch and ischemia-reperfusion [15]. In both mouse models and human kidney disease patients, uL-FABP levels correlate with the severity of tubulointerstitial injury, making it a promising clinical marker of CKD progression in OU [16].
In the current study, we aimed to determine if baseline measurements and changes over time of urine and plasma NGAL, IL-18, and L-FABP are higher in children with OU who develop kidney failure.
Materials and methods
CKiD is a prospective, observational study of children with mild-to-moderate CKD recruited from 48 North American pediatric nephrology centers. Enrollment for CKiD was initiated in 2005 (CKiD cohort 1) and was initially designed to enroll children ages 1–16 with a GFR of 30–90 ml/min per1.73 m2. In 2011 (CKiD cohort 2), these criteria were further restricted to a GFR of 45–90 ml/min per 1.73 m2 to allow longer follow-up prior to CKD progression. At the time of this analysis, over 1000 children have been enrolled in CKiD. All subjects undergo baseline evaluation and yearly follow-up visits of which details for the study design and methods have been previously published [17]. For this study, we nested a case-control design within the CKiD cohort by matching children with OU who received kidney replacement therapy (KRT), defined as kidney transplant or dialysis, to children with OU who did not receive KRT during follow-up. An analysis of traditional clinical markers to predict kidney failure in this cohort was recently published by our group with this current study focused on specific biomarkers of tubular injury [5].
Cohort development
CKiD was queried to identify children with OU as their primary diagnosis code. Cases were defined as children who progressed to KRT during study follow-up, while controls received no KRT over the same time period. In total, 33 children were identified who met case definition, out of a total 149 patients with OU. Inclusion was further restricted to only those children with at least 3 annual study visits where urine and plasma samples were collected. Matching of cases and controls was performed based on age at baseline and time on study using a random sequence generated SAS Enterprise Guide software. Twenty-two cases and 22 controls remained after matching. Study visit samples included enrollment, most central study visit, and last study visit available (for controls, the visit best matching the last paired case visit was used). Patients are removed from the main CKiD study once they reach the KRT outcome. Thus, no patient received KRT prior to sample collection.
Measurements and data collection
CKiD data on included patients was reviewed to estimate patient GFR, measured by either disappearance of iohexol (iGFR) or estimated as a function of sex, height, serum creatinine, cystatin C, and BUN using the CKiD derived formula which has been previously described [18].
After local IRB exemption, banked samples of urine and plasma from each included CKiD patient were requested for analysis of select markers of tubular injury including NGAL, IL-18, and L-FABP. All samples are initially collected and processed by each CKiD institution under strict protocolization. For urine, protease inhibitors are added to the fresh sample, which is then centrifuged, decanted, and frozen to ≤ − 70 °C in separate cryovials. Subject blood is immediately placed in a plasma separator tube (PST), inverted 8–10 times, and centrifuged under strict protocolization. Serum and plasma are then transferred to separate cryovials for freezing at ≤ − 70 °C. Batched urine, plasma, and serum samples are mailed on dry ice to the NIDDK biorepository quarterly. Select samples identified for this study were shipped on dry ice from the NIDDK biorepository to Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, for analysis. Most samples did have one prior thaw; only 3 (all urine) were previously thawed twice. NGAL analysis was performed via nephelometry on the Siemens BNII nephelometer using the Bioporto reagent set for chemistry analyzer. Urine creatinine was measured using an RXL Xpand chemistry analyzer with standardized enzymatic reagent set, while IL-18 and L-FABP were performed using commercially available ELISA kits (Medical & Biological Laboratories Co., LTD., Nagoya, Japan, and CMIC Co., Tokyo, Japan, respectively). All urine results for NGAL, IL-18, and L-FABP were normalized to urine Cr prior to analysis.
Statistical analysis
Patient characteristics (age, sex, race), follow-up duration, and GFR were described for cases and controls at baseline using frequencies for categorical variables and summary measures (median, interquartile range (IQR)) for continuous variables. The baseline Spearman rank correlations between patient age, urine and plasma biomarkers, and GFR measurements were examined. Baseline and longitudinal differences in urine, plasma, and GFR measurements were compared between cases and controls using the Wilcoxon signed-rank test (for baseline differences) and also modeled in linear mixed-effects models (for baseline and longitudinal differences). Each of the markers was modeled as the dependent variable, with time to KRT and case vs. control status as the independent variables. For all matched case-control pairs, the slope for time to KRT was anchored at the time of KRT for the case. All models included a patient random intercept and random slopes for time to KRT, to allow for patient-specific progression to kidney failure, and an interaction term between time to KRT and case status in order to test for slope differences between cases and controls. iGFR measurement was preferentially used when available, and estimated GFR was included for improved linear modeling for visits where only estimations were calculated. As a sensitivity analysis, each of the longitudinal models for urine and plasma biomarkers was repeated after including baseline GFR as covariate. All analyses were conducted in SAS Enterprise Guide.
Results
The study sample consisted of 44 children, 22 cases and 22 controls. There were no differences noted between cases and controls for age at baseline, sex, and race. Follow-up duration was shorter for cases (median, 4.9 years; IQR, 3.6–5.9) than controls (median, 7.2 years; IQR, 5.5–7.8) due to CKiD study censoring once the patient develops kidney failure. For both groups, median (IQR) age at baseline and outcome measurement were 10.5 (9.0–13.0) vs. 15.9 (13.9–16.9) years, respectively (Table 1). At baseline there were no statistically significant differences noted between cases and controls for any urine or plasma tubular marker measured; however, the mean plasma NGAL (pNGAL) did approach a significant difference (193.02 ng/ml vs. 125.56 ng/ml, respectively, p = 0.07). There was an expected difference in the mean baseline GFR between cases and controls (37.8 ml/min per 1.73 m2 vs. 50.1 ml/min per 1.73 m2, respectively, p = 0.01) (Table 2).
Table 1.
Baseline patient characteristics and follow-up duration
| Cases (n = 22) | Controls (n = 22) | |
|---|---|---|
| Age, y | 10.5 (9.0–13.0) | 10.5 (9.0–13.0) |
| Male sex (%) | 18 (81.8) | 17 (77.3) |
| Non-white race (%) | 6 (27.3) | 6 (27.3) |
| GFR (ml/min per 1.73m2) | 35.6 (27.1–13.4) | 46.9 (34.4–58.7) |
| Follow-up duration, y | 4.9 (3.6–5.9) | 7.2 (5.5–7.8) |
| Age at outcome, y | 15.9 (13.9–16.9) | 15.9 (13.9–16.9) |
Data reported as median (interquartile range, IQR) for continuous variables and n (%) for categorical variables
Controls matched to cases by age at study entry and time on study
Years (y), glomerular filtration rate (GFR), number (n)
Table 2.
Baseline biomarker measurements and differences between cases and controls
| Median (IQR) | Mean (95% CI)b | Mean difference (95% CI)b | |||||
|---|---|---|---|---|---|---|---|
| Cases | Controls | pa | Cases | Controls | Case-Control | pb | |
| pNGAL (ng/ml) | 177.92 (93.15, 279.46) | 118.58 (82.54, 165.69) | 0.06 | 193.02 (140.93, 245.11) | 125.56 (73.47, 177.65) | 67.46 (−6.21, 141.13) | 0.07 |
| pIL-18 (pg/ml) | 317.37 (263.41, 443.47) | 357.48 (259.19, 442.40) | 0.86 | 377.27 (296.08, 458.46) | 345.66 (264.47, 426.85) | 31.61 (−83.22, 146.43) | 0.57 |
| pL-FABP (ng/ml) | 130.14 (95.24, 150.18) | 130.74 (91.04, 153.24) | 0.44 | 131.33 (111.69, 150.96) | 129.82 (110.18, 149.45) | 1.51 (−26.26, 29.28) | 0.91 |
| uNGAL (ng/ml) | 98.75 (10.10, 347.00) | 11.55 (9.38, 185.00) | 0.21 | 553.65 (18.93, 1088.38) | 197.31 (−337.41, 732.03) | 356.34 (−399.87, 1112.56) | 0.34 |
| uIL-18 (pg/ml) | 8.66 (7.36, 12.92) | 8.95 (5.14, 13.08) | 0.84 | 26.15 (3.88, 48.41) | 11.28 (−10.98, 33.55) | 14.86 (−16.63, 46.35) | 0.34 |
| uL-FABP (ng/ml) | 12.49 (5.22, 41.23) | 8.05 (2.53, 29.09) | 0.26 | 33.59 (10.58, 56.61) | 29.74 (5.85, 53.62) | 3.85 (−24.20, 31.90) | 0.78 |
| uNGAL/Cr (ng/ml per mg/dl) | 2.67 (0.35, 9.73) | 0.50 (0.22, 4.15) | 0.14 | 14.57 (3.10, 26.04) | 4.83 (−6.64, 16.30) | 9.74 (−6.48, 25.96) | 0.23 |
| uIL-18/Cr (pg/ml per mg/dl) | 0.25 (0.19, 0.36) | 0.27 (0.12, 0.36) | 0.89 | 1.24 (−0.11, 2.58) | 0.29 (−1.06, 1.63) | 0.95 (−0.95, 2.85) | 0.31 |
| uL-FABP/Cr (ng/ml per mg/dl) | 0.36 (0.14, 1.17) | 0.15 (0.08, 0.75) | 0.32 | 1.19 (0.37, 2.00) | 0.90 (0.06, 1.73) | 0.29 (−0.52, 1.10) | 0.46 |
| GFR (ml/min per 1.73 m2) | 35.59 (27.06, 43.39) | 46.85 (34.38, 58.66) | 0.03* | 37.77 (30.94, 44.61) | 50.14 (43.30, 56.98) | −12.37 (−21.80, −2.93) | 0.01* |
signed rank test,
linear mixed effects models
Interquartile range (IQR), confidence interval (CI), plasma (p), urine (u), neutrophil gelatinase-associated lipocalin (NGAL), interleukin 18 (IL-18), liver-type fatty acid binding protein (L-FABP), creatinine (Cr), glomerular filtration rate (GFR).
Clinical significance (p < 0.05)
Evaluating changes in GFR over time, cases declined on average 3.42 ml/min per 1.73 m2/year faster (95% confidence interval (CI) (− 4.73, − 2.10), p < .0001) compared with controls. Expanding analysis to select markers of tubular injury, over 3 annual measurements, pNGAL increased significantly throughout the study period in cases, at an average rate of15.38 ng/ml per year (95% CI (3.30, 27.46), p = 0.01), while controls showed a slight, non-significant decline (− 2.44 ng/ml per year, 95% CI (− 14.23, 9.35), p = 0.68). Over 3 annual measurements, uL-FABP/Cr also showed an increase over time in cases (0.20 ng/ml per mg/dl per year, 95% CI (0.06,0.34), p = 0.01), while controls were relatively stable (−0.03 ng/ml per mg/dl per year, 95% CI (− 0.17, 0.11), p =0.71). There were no other significant differences over time noted on analysis of remaining plasma (IL-18, L-FABP) or urine biomarkers (NGAL, IL-18) for cases or controls (Table 3). After adjusting for baseline GFR, results for biomarker change over time for cases and controls remained relatively stable (Table 4).
Table 3.
Changes in biomarker measurements over time for cases and controls
| Cases | Controls | Case-controls | ||||
|---|---|---|---|---|---|---|
| Slope (95% CI) | p | Slope (95% CI) | p | Difference (95% CI) | p | |
| pNGAL (ng/ml) | 15.38 (3.30, 27.46) | 0.01* | −2.44 (−14.23, 9.35) | 0.68 | 17.82 (0.91, 34.73) | 0.04* |
| pIL-18 (pg/ml) | −14.25 (−31.69, 3.19) | 0.11 | −6.83 (−24.01, 10.35) | 0.43 | −7.42 (−31.94, 17.10) | 0.54 |
| pL-FABP (ng/ml) | 3.16 (−3.49, 9.82) | 0.34 | −0.96 (−7.26, 5.34) | 0.76 | 4.12 (−5.06, 13.31) | 0.37 |
| uNGAL (ng/ml) | −60.35 (−199.92, 79.22) | 0.39 | 32.14 (−98.88, 163.16) | 0.62 | −92.49 (−284.35, 99.37) | 0.33 |
| uIL-18 (pg/ml) | −2.28 (−8.54, 3.99) | 0.47 | 2.40 (−3.51, 8.30) | 0.42 | −4.67 (−13.30, 3.96) | 0.28 |
| uL-FABP (ng/ml) | 6.70 (2.31, 11.10) | <0.01* | 0.68 (−3.60, 4.95) | 0.75 | 6.03 (−0.13, 12.18) | 0.05 |
| uNGAL/Cr (ng/ml per mg/dl) | 0.86 (−3.15, 4.87) | 0.67 | 0.73 (−3.10, 4.56) | 0.70 | 0.13 (−5.43, 5.69) | 0.96 |
| uIL-18/Cr (pg/ml per mg/dl) | −0.24 (−0.57, 0.09) | 0.14 | 0.07 (−0.24, 0.37) | 0.66 | −0.31 (−0.75, 0.14) | 0.17 |
| uL-FABP/Cr (ng/ml per mg/dl) | 0.20 (0.06, 0.34) | 0.01* | −0.03 (−0.17, 0.11) | 0.71 | 0.23 (0.02, 0.43) | 0.03* |
| GFR (ml/min per 1.73m2) | −4.02 (−4.93, −3.11) | < 0.0001* | −0.61 (−1.57, 0.36) | 0.21 | −3.42 (−4.73, −2.10) | < 0.0001* |
All statistics estimated from linear mixed effects models
Confidence interval (CI), plasma (p), urine (u), neutrophil gelatinase-associated lipocalin (NGAL) interleukin 18 (IL-18), liver-type fatty acid binding protein (L-FABP), creatinine (Cr), glomerular filtration rate (GFR)
Clinical significance (p < 0.05)
Table 4.
Changes in biomarker measurements over time for cases and controls after adjusting for baseline GFR
| Cases | Controls | Case-controls | ||||
|---|---|---|---|---|---|---|
| Slope (95% CI) | p | Slope (95% CI) | p | Difference (95% CI) | p | |
| pNGAL (ng/ml) | 15.03 (2.96, 27.10) | 0.02* | −1.70 (−13.47, 10.06) | 0.77 | 16.73 (−0.17, 33.64) | 0.05 |
| pIL-18 (pg/ml) | −14.55 (−32.04, 2.95) | 0.10 | −6.74 (−23.97, 10.48) | 0.43 | −7.80 (−32.40, 16.79) | 0.52 |
| pL-FABP (ng/ml) | 3.35 (−3.32, 10.03) | 0.31 | −1.10 (−7.41, 5.21) | 0.73 | 4.45 (−4.77, 13.67) | 0.33 |
| uNGAL (ng/ml) | −65.75 (−201.85, 70.35) | 0.33 | 30.41 (−97.31, 158.13) | 0.63 | −96.16 (−283.17, 90.85) | 0.30 |
| uIL-18 (pg/ml) | −2.45 (−8.80, 0.00) | 0.44 | 2.41 (−3.56, 8.38) | 0.42 | −4.86 (−13.60, 3.88) | 0.27 |
| uL-FABP (ng/ml) | 6.53 (2.16, 10.90) | < 0.01* | 0.91 (−3.35, 5.17) | 0.67 | 5.62 (−0.51, 11.75) | 0.07 |
| uNGAL/Cr (ng/ml per mg/dl) | 0.70 (−3.31, 4.72) | 0.72 | 0.94 (−2.90, 4.78) | 0.62 | −0.23 (−5.82, 5.35) | 0.93 |
| uIL-18/Cr (pg/ml per mg/dl) | −0.25 (−0.57, 0.08) | 0.14 | 0.06 (−0.24, 0.37) | 0.68 | −0.31 (−0.76, 0.14) | 0.17 |
| uL-FABP/Cr (ng/ml per mg/dl) | 0.20 (0.05, 0.34) | 0.01* | −0.02 (−0.16, 0.12) | 0.76 | 0.22 (0.02, 0.42) | 0.04* |
All statistics estimated from linear mixed effects models
Confidence interval (CI), plasma (p), urine (u), neutrophil gelatinase-associated lipocalin (NGAL), interleukin 18 (IL-18), liver-type fatty acid binding protein (L-FABP), creatinine (Cr), glomerular filtration rate (GFR)
Clinical significance (p < 0.05)
Discussion
Markers specific to kidney tubular injury have the potential to improve risk stratification in children with OU. A significant barrier to studying this has been the delayed nature of kidney disease progression for many children with OU. By leveraging opportunities in the CKiD study population and utilizing banked biospecimens, this provided a unique opportunity to better define the utility of tubular markers in CKD progression for these children. Through this analysis, we identified pNGAL and uL-FABP/Cr as potential markers of tubular injury that can be followed over time to potentially improve risk stratification.
NGAL measurements in both urine and plasma/serum have become one of the most frequently studied and most promising markers of AKI in both adults and children [19]. There is also an emerging evidence to support NGAL’s role in the prediction of CKD. In a large study of adults with CKD, uNGAL negatively correlated with estimated GFR (eGFR) (r = 0.13, p < 0.0001), and higher levels of uNGAL were associated with an increased 2-year and 10-year risk of kidney failure (HR = 1.62 and HR = 1.25 per doubling, respectively, both p < 0.0001). These associations were lost however once adjusted for eGFR [20]. In a study of children with CKD stages 2–4, sNGAL had excellent agreement with measured GFR and outperformed cystatin C and eGFR in patients with more advanced CKD (measured GFR < 30 ml/min per1.73 m2) [8]. Although these results are encouraging, they represent a heterogeneous population of CKD where the full predictive potential may be masked by varying mechanisms of kidney injury and progression. Given that NGAL is known to originate from the kidney collecting duct and is elevated in patients with hydronephrosis, NGAL may provide more accurate prediction in children with CKD resulting from OU [21].
There is preliminary evidence to suggest NGAL’s potential role in pediatric urinary obstruction. In a study of children with obstructive nephropathy, defined as unilateral hydronephrosis with decreased relative function on renal scintigraphy, sNGAL and uNGAL/Cr were significantly greater compared with hydronephrosis without obstructive nephropathy and controls [22]. Similar results have been reported in children undergoing pyeloplasty for ureteropelvic junction obstruction, where bladder uNGAL levels at time of surgery were significantly elevated compared with controls [23]. Interestingly, on a follow-up evaluation of urine in this population of children after pyeloplasty, despite correction of obstruction, uNGAL levels continued to be elevated. This was in contrast to other antimicrobial peptides evaluated that improve after surgical correction [24]. These results suggest that uNGAL levels not only elevate in response to obstruction but continue to be elevated as a result of ongoing tubular injury despite relief of urinary obstruction. In the current study, uNGAL/Cr was higher at baseline and increased more over time in cases compared with controls but did not reach statistical significance for either.
On further analysis, baseline pNGAL was elevated in cases compared with controls, which approached but did not reach statistical significance. Statistical significance was reached however as we followed case subjects longitudinally, where pNGAL levels increased as they approached the KRT outcome. There was a notably different trend in control subjects who showed no change in pNGAL levels over time. These findings persisted after adjusting for baseline GFR. This result confirms pNGAL’s potential as an independent marker of CKD progression and warrants further investigation.
On analysis of L-FABP, baseline mean plasma concentration was virtually identical between cases and controls (131 ng/ml vs. 130 ng/ml, respectively). This is consistent with prior research in adults with diabetic nephropathy where sL-FABP was associated with eGFR but not predictive of eGFR decline [25]. In our study, uL-FABP in the case and control groups (mean 33.59 ng/ml vs.29.74 ng/ml, respectively) were elevated at baseline compared with previously published values in healthy children (mean, 3.4 ng/ml; IQR, 1.6–6.0) [26]. We also identified a significant yearly increase in uL-FABP and uL-FABP/Cr for the case group as subjects approached the KRT outcome. This was true both without and with adjustment for baseline GFR. Prior research on adults with diabetic nephropathy found similar findings where uL-FABP increased as eGFR declined, correlating with proteinuria and systolic blood pressure [27]. Our results add to a small but growing body of literature suggesting the potential role for uL-FABP in the prediction of CKD progression, specifically those with an obstructive mechanism of injury. Future research is necessary to determine if uL-FABP outperforms current clinical markers of CKD progression.
IL-18 is a pro-inflammatory cytokine, shown to be involved in pathologic changes to the kidney due to multiple disease processes, including urinary obstruction [10, 28–31]. Urinary and plasma IL-18 levels are elevated in patients with CKD and predict negative outcomes including death in patients with AKI [32, 33]. In our analysis of children with mild-to-moderate CKD and OU, we did not identify any difference between urine or plasma IL-18 in cases or controls, at enrollment or over time. At baseline, the mean uIL-18 for cases and controls (26.15 pg/ml vs. 11.28 pg/ml, respectively) was similar compared with expected values published for healthy children (21.6 pg/ml) [26]. For pIL-18, the mean baseline values for cases and controls (377.27 pg/ml vs. 345.66 pg/ml, respectively) were higher than those reported in the literature for healthy children (39.16 pg/ml, 169.7 pg/ml, 17 pg/ml). Variability in analytic platforms utilized for measurement, however, impedes a direct comparison across studies [34–36]. Thus, further research is necessary to better elucidate the role of IL-18 in OU and whether urine or plasma measurements provides any prognostic value.
Notwithstanding the benefits of prospective observational data and biobanking through CKiD, there are several limitations to this study. OU is a clinical outcome resulting from a diverse group of disease processes. Unfortunately, due to disease coding confines, we were not able to characterize specific disease processes which ultimately led to the OU. This likely created a cohort of patients with significant diagnosis heterogeneity. This is especially salient in OU, as patients with PUV and chronic bladder dysfunction experience different risk factors for CKD progression compared with those with a history of upper urinary tract obstruction. Secondly, the inability to assess urinary tract obstruction status, bladder dynamics, or imaging at time of sample collection proves challenging for data interpretation, as markers of tubular injury likely change in response to both chronic disease progression and acute insults. Despite these limitations, however, we were able to identify statistical differences between cases and controls for pNGAL and uL-FABP, where in a more homogeneous population, one would expect these differences to further increase. Lastly, urine collection method varied, as some patients provided voided samples, while others required catheterization. Although this variation could have affected biomarker levels, the kidney tubule origin of the markers selected decreases the likelihood of a significant impact.
In conclusion, pNGAL and uL-FABP levels rose significantly over the 5 years preceding KRT in children with mild-to-moderate CKD from OU. This contrasted with children who did not progress to KRT on follow-up, who had relatively stable pNGAL and uL-FABP levels. This remained stable after adjusting for baseline GFR, suggesting an independent effect. Results of this study suggest that serial measures of pNGAL and uL-FABP may provide predictive value for this intermediate-risk population of children. Future studies are necessary to identify optimal cutoff values and to determine if these markers of tubular damage outperform current clinical indicators of CKD progression.
Acknowledgment
Data in this manuscript were collected by the Chronic Kidney Disease in Children (CKiD) prospective cohort study with clinical coordinating centers (principal investigators) at Children’s Mercy Hospital and the University of Missouri, Kansas City (Bradley Warady, MD), and Children’s Hospital of Philadelphia (Susan Furth, MD, PhD), Central Biochemistry Laboratory (George Schwartz, MD) at the University of Rochester Medical Center, and data coordinating center (Alvaro Muñoz, PhD and Derek Ng, PhD) at the Johns Hopkins Bloomberg School of Public Health.
Funding information The CKiD study is supported by grants from the National Institute of Diabetes and Digestive and Kidney Diseases, with additional funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, and the National Heart, Lung, and Blood Institute (U01DK66143, U01DK66174, U24DK082194, U24DK066116). The CKiD website is located at https://statepi.jhsph.edu/ckid/.
Footnotes
Conflict of interest The authors declare that they have no conflict of interest.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Roth KS, Koo HP, Spottswood SE, Chan JC (2002) Obstructive uropathy: an important cause of chronic renal failure in children. Clin Pediatr (Phila) 41:309–314. 10.1177/000992280204100503 [DOI] [PubMed] [Google Scholar]
- 2.Weaver DJ Jr, Somers MJG, Martz K, Mitsnefes MM (2017) Clinical outcomes and survival in pediatric patients initiating chronic dialysis: a report of the NAPRTCS registry. Pediatr Nephrol 32: 2319–2330. 10.1007/s00467-017-3759-4 [DOI] [PubMed] [Google Scholar]
- 3.Benfield MR, McDonald RA, Bartosh S, Ho PL, Harmon W (2003) Changing trends in pediatric transplantation: 2001 annual report of the North American pediatric renal transplant cooperative study. Pediatr Transplant 7:321–335 [DOI] [PubMed] [Google Scholar]
- 4.McLeod DJ, Szymanski KM, Gong E, Granberg C, Reddy P, Sebastiao Y, Fuchs M, Gargollo P, Whittam B, VanderBrink BA, Pediatric Urology Midwest Alliance (PUMA) (2019) Renal replacement therapy and intermittent catheterization risk in posterior urethral valves. Pediatrics 143:e20182656 10.1542/peds.2018-2656 [DOI] [PubMed] [Google Scholar]
- 5.McLeod DJ, Ching CB, Sebastiao YV, Greenberg JH, Furth SL, McHugh KM, Becknell B (2019) Common clinical markers predict end-stage renal disease in children with obstructive uropathy. Pediatr Nephrol 34:443–448. 10.1007/s00467-018-4107-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Winnicki E, McCulloch CE, Mitsnefes MM, Furth SL, Warady BA, Ku E (2018) Use of the kidney failure risk equation to determine the risk of progression to end-stage renal disease in children with chronic kidney disease. JAMA Pediatr 172:174–180. 10.1001/jamapediatrics.2017.4083 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Devarajan P (2008) Neutrophil gelatinase-associated lipocalin (NGAL): a new marker of kidney disease. Scand J Clin Lab Investig Suppl 241:89–94. 10.1080/00365510802150158 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mitsnefes MM, Kathman TS, Mishra J, Kartal J, Khoury PR, Nickolas TL, Barasch J, Devarajan P (2007) Serum neutrophil gelatinase-associated lipocalin as a marker of renal function in children with chronic kidney disease. Pediatr Nephrol 22:101–108. 10.1007/s00467-006-0244-x [DOI] [PubMed] [Google Scholar]
- 9.Noyan A, Parmaksiz G, Dursun H, Ezer SS, Anarat R, Cengiz N (2015) Urinary NGAL, KIM-1 and L-FABP concentrations in antenatal hydronephrosis. J Pediatr Urol 11(249):e241–e246. 10.1016/j.jpurol.2015.02.021 [DOI] [PubMed] [Google Scholar]
- 10.Dinarello CA (2007) Interleukin-18 and the pathogenesis of inflammatory diseases. Semin Nephrol 27:98–114. 10.1016/j.semnephrol.2006.09.013 [DOI] [PubMed] [Google Scholar]
- 11.Awad AS, El-Sharif AA (2011) Curcumin immune-mediated and anti-apoptotic mechanisms protect against renal ischemia/reperfusion and distant organ induced injuries. Int Immunopharmacol 11:992–996. 10.1016/j.intimp.2011.02.015 [DOI] [PubMed] [Google Scholar]
- 12.Wang S, Chen F, Yang S, Shi J (2018) Interleukin-18. Int Heart J 59:786–790. 10.1536/ihj.17-154 [DOI] [PubMed] [Google Scholar]
- 13.Mohamed Ali OS, Elshaer SS, Anwar HM, Zohni MSE (2017) Relevance of cystatin-C, N-acetylglucosaminidase, and Interleukin-18 with the diagnosis of acute kidney injury induced by cadmium in rats. J Biochem Mol Toxicol 31(11). 10.1002/jbt.21968 [DOI] [PubMed] [Google Scholar]
- 14.Parikh CR, Jani A, Melnikov VY, Faubel S, Edelstein CL (2004) Urinary interleukin-18 is a marker of human acute tubular necrosis. Am J Kidney Dis 43:405–414 [DOI] [PubMed] [Google Scholar]
- 15.Kamijo-Ikemori A, Sugaya T, Matsui K, Yokoyama T, Kimura K (2011) Roles of human liver type fatty acid binding protein in kidney disease clarified using hL-FABP chromosomal transgenic mice. Nephrology 16:539–544. 10.1111/j.1440-1797.2011.01469.x [DOI] [PubMed] [Google Scholar]
- 16.Kamijo A, Sugaya T, Hikawa A, Okada M, Okumura F, Yamanouchi M, Honda A, Okabe M, Fujino T, Hirata Y, Omata M, Kaneko R, Fujii H, Fukamizu A, Kimura K (2004) Urinary excretion of fatty acid-binding protein reflects stress overload on the proximal tubules. Am J Pathol 165:1243–1255. 10.1016/S0002-9440(10)63384-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Furth SL, Cole SR, Moxey-Mims M, Kaskel F, Mak R, Schwartz G, Wong C, Munoz A, Warady BA (2006) Design and methods of the chronic kidney disease in children (CKiD) prospective cohort study. Clin J Am Soc Nephrol 1:1006–1015. 10.2215/CJN.01941205 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schwartz GJ, Munoz A, Schneider MF, Mak RH, Kaskel F, Warady BA, Furth SL (2009) New equations to estimate GFR in children with CKD. J Am Soc Nephrol 20:629–637. 10.1681/ASN.2008030287 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Haase M, Bellomo R, Devarajan P, Schlattmann P, Haase-Fielitz A, NGAL Meta-analysis Investigator Group (2009) 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 54:1012–1024. 10.1053/j.ajkd.2009.07.020 [DOI] [PubMed] [Google Scholar]
- 20.Dubin RF, Judd S, Scherzer R, Shlipak M, Warnock DG, Cushman M, Sarnak M, Parikh C, Bennett M, Powe N, Peralta CA (2018) Urinary tubular injury biomarkers are associated with ESRD and death in the REGARDS study. Kidney Int Rep 3:1183–1192. 10.1016/j.ekir.2018.05.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Forster CS, Devarajan P (2017) Neutrophil gelatinase-associated lipocalin: utility in urologic conditions. Pediatr Nephrol 32:377–381. 10.1007/s00467-016-3540-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bienias B, Sikora P (2018) Potential novel biomarkers of obstructive nephropathy in children with Hydronephrosis. Dis Markers 2018:1015726 10.1155/2018/1015726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gupta S, Jackson AR, DaJusta DG, McLeod DJ, Alpert SA, Jayanthi VR, McHugh K, Schwaderer AR, Becknell B, Ching CB (2018) Urinary antimicrobial peptides: potential novel biomarkers of obstructive uropathy. J Pediatr Urol 14:238 e231–238 e236. 10.1016/j.jpurol.2018.03.006 [DOI] [PubMed] [Google Scholar]
- 24.Gupta S, Nicassio L, Yepes Junquera G, Jackson A, Fuchs M, McLeod D, Alpert S, Jayanthi R, DaJusta D, McHugh K, Becknell B, Ching C (2019) Urinary HIP/PAP and BD-1 indicate surgical success after pediatric ureteropelvic junction obstruction surgery Pediatric Urology Fall Congress Scottsdale, Arizona: https://fallcongressspuonlineorg/multimedia/files/2019/presentations/Thursday/0112_Chingpdf Acccessed May 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Chou KM, Lee CC, Chen CH, Sun CY (2013) Clinical value of NGAL, L-FABP and albuminuria in predicting GFR decline in type 2 diabetes mellitus patients. PLoS One 8:e54863 10.1371/journal.pone.0054863 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bennett MR, Nehus E, Haffner C, Ma Q, Devarajan P (2015) Pediatric reference ranges for acute kidney injury biomarkers. Pediatr Nephrol 30:677–685. 10.1007/s00467-014-2989-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Viswanathan V, Sivakumar S, Sekar V, Umapathy D, Kumpatla S (2015) Clinical significance of urinary liver-type fatty acid binding protein at various stages of nephropathy. Indian J Nephrol 25:269–273. 10.4103/0971-4065.145097 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bani-Hani AH, Leslie JA, Asanuma H, Dinarello CA, Campbell MT, Meldrum DR, Zhang H, Hile K, Meldrum KK (2009) IL-18 neutralization ameliorates obstruction-induced epithelial-mesenchymal transition and renal fibrosis. Kidney Int 76:500–511. 10.1038/ki.2009.216 [DOI] [PubMed] [Google Scholar]
- 29.Daemen MA, van’t Veer C, Wolfs TG, Buurman WA (1999) Ischemia/reperfusion-induced IFN-gamma up-regulation: involvement of IL-12 and IL-18. J Immunol 162:5506–5510 [PubMed] [Google Scholar]
- 30.Faust J, Menke J, Kriegsmann J, Kelley VR, Mayet WJ, Galle PR, Schwarting A (2002) Correlation of renal tubular epithelial cell-derived interleukin-18 up-regulation with disease activity in MRL-Faslpr mice with autoimmune lupus nephritis. Arthritis Rheum 46:3083–3095. 10.1002/art.10563 [DOI] [PubMed] [Google Scholar]
- 31.Striz I, Krasna E, Honsova E, Lacha J, Petrickova K, Jaresova M, Lodererova A, Bohmova R, Valhova S, Slavcev A, Vitko S (2005) Interleukin 18 (IL-18) upregulation in acute rejection of kidney allograft. Immunol Lett 99:30–35. 10.1016/j.imlet.2005.01.010 [DOI] [PubMed] [Google Scholar]
- 32.Parikh CR, Abraham E, Ancukiewicz M, Edelstein CL (2005) Urine IL-18 is an early diagnostic marker for acute kidney injury and predicts mortality in the intensive care unit. J Am Soc Nephrol 16(10):3046–3052. 10.1681/ASN.2005030236 [DOI] [PubMed] [Google Scholar]
- 33.Chiang CK, Hsu SP, Pai MF, Peng YS, Ho TI, Liu SH, Hung KY, Tsai TJ, Hsieh BS (2005) Plasma interleukin-18 levels in chronic renal failure and continuous ambulatory peritoneal dialysis. Blood Purif 23:144–148. 10.1159/000083620 [DOI] [PubMed] [Google Scholar]
- 34.Zhou J, Shi F, Xun W (2018) Leptin, hs-CRP, IL-18 and urinary protein before and after treatment of children with nephrotic syndrome. Exp Ther Med 15:4426–4430. 10.3892/etm.2018.5923 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Wang YB, Shan NN, Chen O, Gao Y, Zou X, Wei DE, Wang CX, Zhang Y (2011) Imbalance of interleukin-18 and interleukin-18 binding protein in children with Henoch-Schonlein purpura. J Int Med Res 39:2201–2208. 10.1177/147323001103900616 [DOI] [PubMed] [Google Scholar]
- 36.Chung HL, Shin JY, Ju M, Kim WT, Kim SG (2011) Decreased interleukin-18 response in asthmatic children with severe myco-plasma pneumoniae pneumonia. Cytokine 54:218–221. 10.1016/j.cyto.2011.02.008 [DOI] [PMC free article] [PubMed] [Google Scholar]