Utility of Urinary Biomarkers Neutrophil Gelatinase-Associated Lipocalin and Kidney Injury Molecule-1 as a Marker for Diagnosing the Presence of Renal Scar in Children with Vesicoureteral Reflux (VUR): A Cross-Sectional Study.

Aim: To explore the possibility of using urinary biomarkers neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury molecule-1 (KIM-1) to assess the presence of renal scars in children with Vesicoureteric Reflux (VUR). Materials and
Methods: This cross-sectional study was conducted in 94 children aged 0-16 years diagnosed with VUR in the Department of Pediatric Surgery, JIPMER. Urinary biomarkers were measured using the enzyme-linked immunosorbent assay kits, normalized with urinary creatinine (Cr) and compared with severity of VUR (low grade [I and II] and high grade [III, IV, and V]), presence or absence of renal scar in VUR patients and severity of renal scar. Independent Student's t-test, Mann-Whitney U-test, and analysis of variance Kruskal-Wallis test were used for comparison, and receiver operating characteristic (ROC) curve analysis for predicting the accuracy of biomarkers in detecting the presence of renal scars.
Results: The median urinary NGAL (uNGAL) value was higher in children with renal scar (1.49 ng/mL) than those without renal scar (0.58 ng/mL) and was statistically significant (<0.001). Whereas median uNGAL/Cr was higher in children with renal scar (0.07) than those without renal scar (0.03) but was not statistically significant (P = 0.06). Urinary KIM-1 and urinary KIM-1/urinary Cr (uKIM-1/Cr) was not found to be a significant predictor of renal scar. The difference of uNGAL/Cr was comparable between the grades of renal scar but was not statistically significant. On ROC curve analysis, uNGAL had area under the ROC curve (AUC) of 0.769 with 71% of both specificity and sensitivity, whereas uNGAL/Cr was found to be a poor predictor of renal scar with AUC of 0.611, 60% sensitivity, and 61.2% specificity.
Conclusion: uNGAL can serve as a noninvasive marker for diagnosing the presence of renal scar in children with VUR and a multicentric more extensive cohort study may be needed to strengthen or negate its role. Copyright:

INTRODUCTION

Vesicoureteric reflux (VUR) is one of the most common congenital abnormalities of the urinary tract.[1] It is defined as “a pathologic retrograde backflow of urine from the urinary bladder through the vesicoureteral junction into the ureter and kidney,”[2] with a prevalence of around 1.6% in the general population. These VUR patients are predisposed to pyelonephritis which may eventually end up in sequelae such as focal ischemia, interstitial damage, fibrosis, and partially irreversible renal parenchymal scarring (RPS). Long-term complications such as hypertension, proteinuria, and chronic kidney disease (8.5%–18%) can be prevented with early intervention.[3456] Hence, the early detection of renal scarring is essential in patients with VUR for timely intervention.

Dimercaptosuccinic acid (DMSA) renal scintigraphy is considered the gold standard to detect renal scarring and holds a high sensitivity when it comes to the detection of renal scar.[7] However, the exposure to ionizing radiation, invasiveness of the procedure, and expensive nature are the pitfalls which limit its utility on a regular basis. Therefore, the measurement of tubular biomarkers such as neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury molecule-1 (KIM-1) in urine showing the presence and progression of renal scars may prove to be of potential benefit in the evaluation of renal scarring in patients with VUR which will be noninvasive, easy to measure, with no radiation exposure and can be repeated.

NGAL belongs to the lipocalin family and is expressed in neutrophil granules, monocytes, macrophages along with renal epithelial tubular cells. It is secreted into the urine by the thick ascending loop limb of the loop of Henle and the collecting ducts of the kidney. Elevated levels of urinary NGAL (uNGAL) excretion have been demonstrated in children with renal tubular damage depending upon the severity which is due to the overproduction of NGAL and decreased NGAL reabsorption. uNGAL has been demonstrated to be a sensitive biomarker of damage to the proximal kidney and chronic tubulointerstitial injury,[8] and recently, uNGAL has been evaluated in multiple studies as a potential marker for VUR and renal scarring.[34910]

KIM-1, also known as T-cell immunoglobulin mucin domain-1, is a Type 1 transmembrane glycoprotein and serves as a signalling molecule.[11] KIM-1 is undetectable in healthy kidneys and is upregulated after renal ischemia or some form of nephrotoxic stress. Studies have reported that urinary KIM-1 (uKIM-1) levels may also be a sensitive and noninvasive marker of tubulointerstitial injury,[2] as uKIM -1 levels endow with the reliable detection of renal injury in its early evolving phases.[12]

We hypothesized these urinary biomarkers such as NGAL and KIM-1 can be potential marker of renal scar among the children with VUR. Hence, we planned a study with an intent to measure uKIM-1 and NGAL in patients with VUR and evaluate their role in being a useful marker for the detection of renal scar in these patients.

MATERIALS AND METHODS

A cross-sectional observational study was conducted in the department of pediatric surgery in collaboration with the department of biochemistry and department of nuclear medicine in a tertiary center from December 2017 to October 2019 after obtaining ethical approval from the Institutional Ethics Committee (Human Studies). All children aged between age 0 and 16 years diagnosed with vesicoureteral reflux (VUR) on micturating cystourethrogram-primary VUR and VUR secondary to posterior urethral valves were included in the study. Patients with the duplex system, urolithiasis, ureterocele, associated pelvi-ureteric junction obstruction, renal parenchymal diseases, primary hypertension, estimated glomerular filtration rate (eGFR) of <50% of age-appropriate cutoff in children of 2 years and e-GFR of <30 ml/min/m2 in children older than 2 years [Figure 1] were excluded from the study. In addition, patients with acute kidney injury and recent history of urinary tract infection (UTI) were also excluded from the study. Informed and written consent/assent from parents/Legally Authorized Representative was obtained. VUR was graded according to the international reflux study group revised grading system Grade I-V[13] and was again grouped into a low grade (Grade I and II) and high grade (Grade III, IV, and V). RPS was diagnosed by 99 mTc-DMSA scintigraphy. Renal scars were defined as focal decreased uptake associated with contracted and loss of volume in the involved cortex. Renal lesions revealed by DMSA scanning were classified according to the DMSA grading system of Imperiale et al.[14]0 = no scar, 1 = one scar, 2 = two scars, and 3 = multiple scars, more than 2 with renal parenchymal subversion. The grade of RPS was cumulated in case of bilateral involvement.

Figure 1

Study population

15 ml of spot urine was collected from all patients and stored at 80° Celsius until analysis. Urinary biomarkers NGAL and KIM-1 and urinary creatinine (Cr) (normalized to avoid dilution effect) were measured using a commercially available enzyme-linked immunosorbent assay (ELISA) kit and correlated with DMSA scan done on the same day.

Independent variables such as age, gender, height, weight, and eGFR were recorded along with the outcome variables urinary biomarker levels of NGAL and KIM-1 with urinary and serum Cr value.

Statistical analysis

The statistical analysis was performed using the IBM SPSS version 19.0 (IBM Corp, Armonk, New York, USA). The distribution of categorical variables such as gender and renal scarring was expressed in terms of frequency and percentages. The continuous variables such as age, height, weight, and level of urinary biomarker NGAL and KIM-1 were expressed by mean with standard deviation or median with interquartile range (IQR) range depending on the normality of the variables. The comparison of NGAL and KIM-1 with other categorical variables mentioned above was carried out using the independent Student's t-test or Mann–Whitney U test for two groups based on the normalcy and when the levels of urinary biomarkers were compared with severity of renal scar in patients with VUR one-way analysis of variance/Kruskal–Wallis test was used. The prognostic accuracy of NGAL and KIM-1 for predicting the presence of renal scars was expressed using the receiver operating characteristic (ROC) curve. The sample size was estimated using nMaster software version 2.0 based on the anticipated sensitivity of NGAL in diagnosing renal scarring at 92%, at 5% level of significance with 5% precision with 95% confidence, the minimum sample size was estimated to be 114 in total. All the statistical analyses were carried out at 5% level of significance, and P < 0.05 was considered statistically significant.

RESULTS

A total of 94 patients were enrolled in the study, of which 19 (20.2%) were female and 75 (79.8%) were males with a male-to-female ratio of 4:1. Among the 94 patients with VUR, 39 (41.4%) patients had unilateral VUR and 55 (58.5%) patients had bilateral VUR. Of the 39 patients with unilateral VUR, 25 were right-sided VUR and 14 were left-sided VUR.

Demographic profiles and various markers of renal scar in VUR

Of the 94 patients of VUR in our study, 45 (47.9%) had renal scars and 49 (52.1%) had no renal scar in either kidney. The median age of patients of VUR with renal scars was 36 months with an IQR of (12, 60) months. In contrast, the median age of patients without renal scars was 12 months with an IQR of (4.5, 24.5), and this difference was statistically significant (P < 0.001). Similarly, there was a significant difference in weight and height between the groups with mean weight and the height of the child with the renal scar was significantly less than children without renal scar [Table 1].

Table 1

Comparison of demographic profiles and various markers with respect to presence and absence of renal scars in vesicoureteric reflux

Parameters	Renal scars		P
	Present (n=45)	Absent (n=49)
Age (months), median (IQR)	36 (12,60)	12 (4.5,24.5)	<0.001*
Height (cm), mean±SD	92.01±20.09	72.70±12.72	<0.001#
Weight (kg), mean±SD	12.88±5.45	8.26±3.02	<0.001#
Serum creatinine (mg/dL), mean±SD	0.55±0.258	0.436±0.126	0.007#
Urine creatinine (mg/dL), mean±SD	25.20±13.22	14.65±8.61	<0.001#
eGFR, mean±SD	75.51±23.40	73.55±24.27	0.72#
Serum urea (mg/dL), median (IQR)	24 (8)	16 (11)	<0.001*
KIM-1 (ng/mL), mean±SD	4.06±1.86	3.94±1.77	0.73#
NGAL (ng/mL), median (IQR)	1.49 (2.15)	0.58 (0.85)	<0.001*
KIM-1/urinary creatinine, median (IQR)	0.16 (0.15)	0.28 (0.35)	0.007*
NGAL/urinary creatinine, median (IQR)	0.07 (0.2)	0.03 (0.9)	0.06*

*Mann-Whitney U test, #Independent student t-test. SD: Standard deviation, IQR: Interquartile range, EGFR: Estimated glomerular filtration rate, KIM-1: Kidney injury molecule-1, NGAL: Neutrophil gelatinase-associated lipocalin

The mean urinary Cr and serum Cr in patients of VUR with renal scars were 25.20 ± 13.22 mg/dl and 0.55 ± 0.258 mg/dl, respectively, as compared to the patients without renal scars whose mean urine Cr was and serum Cr was 14.65 ± 8.61 mg/dl and 0.436 ± 0.126 mg/dl, respectively. These differences were statistically significant (P < 0.001). These differences are mainly due to the significant difference age, height and weight between the two groups, and further, the serum and urine Cr is influenced by muscle mass and tubular secretion. The mean eGFR in patients of VUR with renal scar was 75.51 ± 23.40 ml/min/1.73 m2 and in patients without renal scar was 73.55 ± 24.27 ml/min/1.73 m2 and is not statistically significant.

The median (IQR) urine NGAL in patients of VUR with renal scars was 1.49 (2.15) ng/mL and in patients without renal scars was 0.58 (0.85) ng/mL, and this difference was statistically significant (P < 0.001). However, the normalized uNGAL/urine Cr between patients of VUR with renal scars and without renal scars was comparable and was higher in children with a renal scar but was not statistically significant (P = 0.06), as shown in Figure 2.

Figure 2

Representation of urinary biomarkers urinary Neutrophil gelatinase-associated lipocalin/creatinine and urinary kidney injury molecule-1/creatinine to renal scar

The mean urine KIM-1 in patients of VUR with renal scars was 4.06 ± 1.86 ng/mL as compared to the patients without renal scars whose mean urine KIM-1 was 3.94 ± 1.77 ng/mL, and this difference was not statistically significant (P = 0.73). However, the normalized uKIM -1/urinary Cr in patients of VUR with renal scars was low, i.e., 0.16 (0.15) as compared to the patients without renal scars whose median (IQR) KIM-1/urinary Cr was 0.28 (0.35). This difference was statistically significant (P < 0.001) with an inverse correlation, as shown in Figure 2.

Urinary biomarkers concerning the severity of renal scars

In the group with one scar, the median (IQR) uNGAL was 1.10 (1.2), in the group with two scars and multiple scars were 0.70 (0.01) and 1.67 (3.46) with statistically significance (P = 0.045). At the same time, normalized median (IQR) NGAL/urinary Cr in the group with one scar, two scars, multiple scars were 0.06 (0.15), 0.02 (0.05), 0.11 (0.23), respectively, with no statistical significance (P = 0.091).The mean KIM-1 and KIM-1/urinary Cr are not comparable with the severity of the renal scar, and the P values were 0.324 and 0.216, respectively, which were not statistically significant in predicting the severity of renal scar, as shown in Table 2.

Table 2

Comparison of urinary biomarkers in vesicoureteric reflux patients with respect to severity of renal scars

Parameters	One scar (n=8)	Two scars (n=5)	Multiple scars (n=32)	P
KIM-1 (ng/mL)	4.62±1.91	4.88±2.68	3.80±1.71	0.324*
NGAL (ng/mL)	1.10 (1.2)	0.70 (0.01)	1.67 (3.46)	0.045#
Urinary KIM-1/creatinine, mean±SD	0.32±0.28	0.24±0.21	0.20±0.13	0.216*
Urinary NGAL/creatinine, median (IQR)	0.06 (0.15)	0.02 (0.05)	0.11 (0.23)	0.091#

#Kruskal-Wallis test, *ANOVA. KIM-1: Kidney injury molecule-1, NGAL: Neutrophil gelatinase-associated lipocalin, SD: Standard deviation, IQR: Interquartile range

Urinary biomarkers and renal scar with unilateral and bilateral VUR patients

The mean KIM-1/urinary Cr in patients with unilateral VUR with the renal scar was 0.23 ± 0.17, and in the group with no renal scars, it was 0.49 ± 0.42. The difference was statistically significant (P = 0.012) but inversely correlating. However, the mean uNGAL/urinary Cr was not statistically significant in unilateral VUR, as shown in Table 3. The normalized mean uNGAL/urinary Cr patients with bilateral VUR with the renal scar were 0.18 ± 0.2, and with no renal scars, it were 0.07 ± 0.11. The difference was statistically significant (P = 0.025) However, the mean KIM-1/urinary Cr levels in bilateral VUR were not statistically significant, as shown in Table 3.

Table 3

Comparison of urinary biomarkers in unilateral and bilateral vesicoureteric reflux with respect to presence or absence of renal scars

Laterality	Parameters	With renal scar (n=15)	Without renal scar (n=24)	P
Unilateral VUR	KIM-1 (ng/mL)	3.78±1.50	4.15±1.83	0.515#
	NGAL (ng/mL)	0.99 (1.16)	0.58 (0.7)	0.038*
	Urinary KIM-1/creatinine, mean±SD	0.23±0.17	0.49±0.42	0.637#
	Urinary NGAL/creatinine, median (IQR)	0.05 (0.09)	0.04 (0.014)	0.921*
Bilateral VUR		(n=30)	(n=25)
	KIM-1 (ng/mL)	4.21±2.02	3.74±1.74	0.36#
	NGAL (ng/mL)	3.42±3.36	1.05±1.19	0.001#
	Urinary KIM-1/creatinine, mean±SD	0.23±0.18	0.34±0.32	0.139#
	Urinary NGAL/creatinine, median (IQR)	0.18±0.2	0.07±0.11	0.025#

*Mann-Whitney U test, #Independent student t-test. KIM-1: Kidney injury molecule-1, NGAL: Neutrophil gelatinase-associated lipocalin, SD: Standard deviation, IQR: Interquartile range, VUR: Vesicoureteric reflux

Urinary biomarkers concerning the severity of reflux

Urinary biomarkers (NGAL, KIM-1, NGAL/Cr, and KIM-1/Cr) were not significant whether the VUR is low grade or a high grade in unilateral VUR. The median (IQR) uNGAL in the group with low-grade bilateral VUR was 0.26 (0.35) ng/mL as compared to 1.47 (2.36) ng/mL in the group with high-grade bilateral VUR which was statistically significant (P = 0.001). The normalized median (IQR) uNGAL/Cr in the group with low-grade bilateral VUR was 0.01 (0.02) as compared to 0.07 (0.18) in the group with high-grade bilateral VUR which was also statistically significant (P = 0.005). However, uKIM-1/Cr was not statistically significant [Table 4].

Table 4

Comparison of urinary biomarkers in unilateral and bilateral vesicoureteric reflux with respect to severity of reflux

Laterality	Parameters	Low grade reflux (n=14)	High grade reflux (n=25)	P
Unilateral VUR	KIM-1 (ng/mL)	4.19±1.69	3.90±1.77	0.62#
	NGAL (ng/mL)	0.03±0.17	0.04±0.02	0.933#
	Urinary KIM-1/creatinine, mean±SD	0.43±0.33	0.37±0.39	0.637#
	Urinary NGAL/creatinine, median (IQR)	1.08±1.63	1.05±0.96	0.68#
Bilateral VUR		(n=5)	(n=50)
	KIM-1 (ng/mL)	3.25±1.08	4.07±1.95	0.362#
	NGAL (ng/mL)	0.26 (0.35)	1.47 (2.36)	0.001*
	Urinary KIM-1/creatinine, mean±SD	0.21±0.16	0.20±0.25	0.854*
	Urinary NGAL/creatinine, median (IQR)	0.01 (0.02)	0.07 (0.18)	0.005*

*Mann-Whitney U test, #Independent student t-test. KIM-1: Kidney injury molecule-1, NGAL: Neutrophil gelatinase-associated lipocalin, SD: Standard deviation, IQR: Interquartile range, VUR: Vesicoureteric reflux

Receiver operating characteristic curve of urinary biomarkers and renal scarring VUR

Overall, in the entire cohort of VUR patients, uNGAL was a predictor with the area under the ROC curve (AUC) of 0.76, having a sensitivity of 71.1% and specificity of 71.4% with a cutoff value of 0.9725 ng/ml to predict scarring. However, the normalized uNGAL/urinary Cr was a poor predictor of the renal scar with AUC of 0.611, having sensitivity and specificity of 60% and 61.2%, respectively. The ROC curve for uKIM-1, KIM-1/urinary Cr was not found to be statistically significant for the renal scar with AUC of 0.523 and 0.337, respectively. The details of the ROC analysis of various urinary biomarkers to renal scars are described inTable 5 and Figure 3.

Table 5

Receiver operating characteristic analysis of urinary biomarkers in vesicoureteric reflux patients with respect to presence or absence of renal scar

Parameters	Cut-off value	Sensitivity (%)	Specificity (%)	AUC (95% CI)	P
KIM-1 (ng/mL)	4.0315	53.3	59.2	0.523 (0.41-0.64)	0.705
NGAL (ng/mL)	0.9725	71.1	71.4	0.769 (0.67-0.87)	<0.001
Urinary KIM-1/creatinine	0.19	48.9	30.6	0.337 (0.23-0.47)	0.007
Urinary NGAL/creatinine	0.045	60	61.2	0.611 (0.49-0.73)	0.06

KIM-1: Kidney injury molecule-1, NGAL: Neutrophil gelatinase-associated lipocalin, AUC: Area under the curve, CI: Confidence interval

Figure 3

Receiver operating characteristic curve of urinary biomarkers and estimated glomerular filtration rate to the presence or absence of renal scar

DISCUSSION

UTI is the most common complication of VUR progressing to pyelonephritis, eventually leading to renal scarring (reflux nephropathy) and chronic kidney disease.[1516] Conventionally serum urea, serum Cr, and Cr clearance are used as a marker of renal injury. In clinical practice, we usually use serum urea or serum Cr to assess the renal function and DMSA scintigraphy to detect renal scars. Serum Cr is a poor marker of early renal damage. It is limited by sensitivity, specificity, and greatly influenced by muscle mass and tubular secretion. Besides, its level starts increasing only after substantial kidney damage has occurred with a time lag.

Similarly, serum urea also lacks specificity and can be high in dehydration and various catabolic conditions. Thus, urea and Cr are not considered a marker of early renal damage. To date, there are no useful biomarkers that can predict the development of renal scars. Noninvasive methods such as urinary biomarkers would greatly facilitate the identification of renal scar early, and thus, and early intervention can prevent renal damage further. In this study, we have evaluated the levels of uNGAL and KIM-1and normalized with urinary Cr in VUR patients as a potential marker for the detection and progression of the renal scar.

In our study, we found that the normalized uNGAL/Cr was though comparably higher in VUR with scarred kidney but not statistically significant between the two groups. However, Ichino et al.[4] and Parmaksiz et al.[3] in their studies found a significantly higher uNGAL/urinary Cr in patients of VUR with scar kidney than patients having no kidney scar. These differences observed in our study may be due to various factors. First, there were statistically significant differences in age, height, and weight between the two groups in our study. Second, the majority of patients in our study group were between 12 and 60 months of age. In contrast, it has been seen that infants excrete high uNGAL as renal cell proliferation is still going on at this stage of life,[3] and this might have skewed our analysis another way. Third, many patients in our study group (VUR with scar group) had small kidneys and not excluded from the study as was done in the study by Ichino et al.[4] It has been seen that small kidneys associated with VUR are usually congenital, and they lack sufficient regenerative renal tubular cells to produce high levels of NGAL.[3]

The role of uKIM-1/Cr in detecting renal scar is not much known. Unlike Toker et al.[2] who had shown that the level of uKIM-1/Cr was significantly increased in VUR with renal scar group and are inclined to increase with the degree of renal scarring, our results shows that uKIM-1/Cr do not have much role in detecting renal scar. In contrast, our result shows that uKIM-1/Cr decrease in the renal scar group. It does not correlate with the degree of the renal scar. Why uKIM-1 level decreased in patients of VUR with renal scar cannot be clearly explained with the present existing knowledge and literature. However, it may be due to the fact that assessment of uKIM-1 was quantified by an ELISA system that is primarily designed for the analysis of KIM-1 in cell culture supernatants and not explicitly designed for urine samples even though the assay has been validated for urine analysis by Chaturvedi et al.[17] Further, the soluble ectodomain of KIM-1 is stable in urine at the room temperature and over a broad range of PH, yet freezing the urine samples at −80°C and thawing them might have caused preanalysis degradation of uKIM-1.[18] Like our study, Parmaksiz et al.[3] also disprove the role of uKIM-1/Cr in detecting renal scar.

The urinary biomarkers were compared between low-grade VUR (Grade I and II) and high grade (Grade III, IV, and V) among unilateral VUR and bilateral VUR. There was no significant difference observed in unilateral VUR between low-grade reflux and high-grade reflux. However, we observed a significant difference in uNGAL/Cr between low grade and high-grade reflux among the bilateral VUR group, and we feel that this difference is mainly due to the difference in sample size between the two groups (low grade = 5, high grade = 50).

Ichino et al.[4] studied uNGAL/Cr as a biomarker for the progression of renal scarring in patients with VUR. They did not find any correlation of uNGAL/Cr with reflux grade. However, uNGAL/Cr was significantly higher in patients with evidence of renal scar on DMSA irrespective of the grade of reflux. They concluded that estimation of uNGAL/Cr might be useful as a noninvasive diagnostic or prognostic biomarker for renal scarring. In our study, we observed that the levels of uNGAL were significantly higher in patients with unilateral VUR and with concomitant renal scar when compared to patients without renal scarring as detected on DMSA scan, and similarly, the level of uNGAL/urinary Cr was significantly higher in patients of bilateral VUR with renal scarring as compared to patients without renal scarring as detected on DMSA scan. Hence, it is safe to assume that uNGAL values are elevated in patients with VUR, and renal scarring and the levels increase with the extent of involvement of damage, and hence, NGAL can serve as a predictive marker for renal scarring.

On the ROC curve analysis of uNGAL/Cr, Ichino et al.[4] reported a cutoff value of 1 with high sensitivity (89.5%) and specificity (100%) with AUC of 0.947 for diagnosis of the renal scar. Parmaksiz et al.[3] reported a cutoff value of the uNGAL/Cr as 0.58 μg/g Cr on ROC analysis with a sensitivity of 72%, a specificity of 60%, and AUC of 0.740 in predicting renal scar. The ROC analysis of our study revealed that only uNGAL was a fair predictor of the renal scar with a cutoff value of 0.9725 ng/mL and AUC 0.769 ng/mL with 71% both sensitivity and specificity. In contrast, the uNGAL/Cr was a poor predictor to detect scarring with AUC of 0.611 with a sensitivity of 60%, specificity of 61% and cutoff of 0.045. The sensitivity and specificity were much higher in the study by Ichino et al. than ours, and it may be because our study cohort has many patients with small kidneys, the comparison is shown in Table 6.

Table 6

Receiver operating characteristic of urinary neutrophil gelatinase-associated lipocalin/creatinine in different study for predicting renal scars in vesicoureteric reflux patients

Study name	Cut-off value	AUC	Sensitivity (%)	Specificity (%)	P
Ichino et al.[4]	1	0.947	89.5	100	0.0036
Parmaksiz et al.[3]	0.58	0.740	72	60	0.0001
Our study	0.045	0.611	60	61	0.06

AUC: Area under the curve

Forster and Devarajan[10] had suggested from their studies that the difference in NGAL values in patients with and without scarring is statistically significant. However, a single NGAL value lacks any utility and hence longitudinal follow-up of NGAL values in children with recurrent episodes of UTIs may serve as a potential indicator to monitor the progression of renal scars. Studies suggest that uNGAL does not revert to normal after the development of scars.

However, in addition to the role of urinary biomarkers (uNGAL and uKIM-1) in renal scar studied above, studies are there that have suggested role of urinary liver-type fatty acid binding protein,[3] urinary transforming growth factor β1, vascular endothelial growth factor, monocyte chemoattractant protein-1[19] and interleukin-8[20] as a reliable tool for early and noninvasive detection of renal injury, inflammation and fibrosis in VUR.

The significant limitations of our study are; it is a single-center study, the sample size is small, the two groups of VUR (renal scar and no renal scar) were not age-matched and nonexclusion of patients of VUR with congenital small scarred kidney which were likely to have low regenerative tubular epithelial cells to secrete NGAL.

CONCLUSION

Although imaging modalities for the diagnosis and follow-up for VUR in pediatric patients remain the first line of investigation, an addition of urinary biomarkers not only predicts the presence of renal scar in the follow-up which can be further confirmed but will also prevent repeated invasive harmful radiation exposure. Although our study has not shown that scar can be diagnosed “early” by uNGAL, it has shown scar may also be diagnosed “significantly” with “noninvasive” test like uNGAL and may be a useful adjunct to invasive, expensive, and less available DMSA test. However, a multicenter longitudinal study in a large cohort is needed to substantiate its role in predicting the early detection of renal scar by the longitudinal assessment of uNGAL level in patients with VUR.

Financial support and sponsorship

This study was financially supported by Intramural financial grant from the institute (JIPMER).

Conflicts of interest

There are no conflicts of interest.