Use of urinary NGAL in steroid-resistant vs. steroid-sensitive nephrotic syndrome: a systematic review and meta-analysis

Abstract

Background

Nephrotic syndrome is a common glomerular disorder. Treatment typically begins with corticosteroids, but patient response varies. Differentiating between steroid-sensitive nephrotic syndrome (SSNS) and steroid-resistant nephrotic syndrome (SRNS) early in the disease course is important, as SRNS is associated with a higher risk of poor long-term outcomes. Neutrophil gelatinase-associated lipocalin (NGAL), a biomarker released in response to tubular injury, has emerged as a potential non-invasive marker for renal damage.

Methods

We conducted a systematic review and meta-analysis of studies reporting NGAL levels in SSNS and SRNS, based on the PRISMA guidelines. A comprehensive literature search was conducted using PubMed, Scopus, Web of Science, ScienceDirect, and the WHO Virtual Health Library Regional. The statistical analysis was performed using a random-effects model to estimate the standardized mean difference (SMD) with a 95% confidence interval.

Results

A total of 16 studies were included. Meta-analyses revealed significantly higher urinary NGAL levels in both SSNS and SRNS patients compared to healthy controls. Urinary NGAL levels were significantly higher in SSNS and SRNS patients compared to healthy controls, with SMD = 0.78 (95% CI: 0.434–1.128, P < .001) and SMD = 2.56 (95% CI: 1.152–3.971, P < .001), respectively. Patients with SRNS had markedly higher urinary NGAL levels than those with SSNS (SMD = 1.889, 95% CI: 0.819–2.959, P < .001). ROC analyses across several studies demonstrated moderate to strong discriminative ability of urinary NGAL in distinguishing between SRNS and SSNS.

Conclusion

Urinary NGAL demonstrated strong potential as a non-invasive biomarker for distinguishing between SRNS and SSNS, supporting its clinical utility in early diagnosis, risk stratification, and management.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12882-025-04420-9.

Introduction

Nephrotic syndrome is a clinical syndrome defined by massive proteinuria (greater than 40 mg/m^2 per hour) responsible for hypoalbuminemia (less than 30 g/L), with resulting hyperlipidemia, edema, and various complications [1, 2]. Most patients respond to corticosteroid treatment, which can induce remission in approximately 80% of children and is known as steroid-sensitive nephrotic syndrome (SSNS). However, a significant proportion of patients still have steroid-resistant nephrotic syndrome (SRNS), which increases the risk of developing chronic kidney disease and requires treatment supplemented with immunesuppressant agents [1–3].

Neutrophil gelatinase-associated lipocalin (NGAL) is a protein belonging to the lipocalin superfamily, initially found in activated neutrophils. It has been shown that many other types of cells, including the kidney tubules, may produce NGAL in response to various injuries [4, 5]. NGAL expression is rapidly up-regulated in response to renal injury and has proven to be a robust marker for acute kidney injury. The increase in NGAL production and release from tubular cells after harmful stimuli may represent a self-defensive mechanism involving activation of specific iron-dependent pathways, which likely also underlie NGAL’s role in promoting renal growth and differentiation [4–6].

There is a need for finding diagnostic tests that accurately predict steroid responsiveness in nephrotic syndrome or distinguish SRNS from SSNS [6]. SRNS is associated with a higher risk of progression to chronic kidney disease and often necessitates more aggressive or alternative immunosuppressive therapy. Response to steroid therapy is considered a reliable indicator of prognosis, and therefore all children are treated initially with high-dose corticosteroids [6]. Recently, studies have aimed to find markers that could predict a patient’s response to steroids at the time of diagnosis, supporting early therapeutic decision-making. Benefits of early identification of SRNS include avoidance of prolonged exposure to high-dose steroids, thereby reducing the risk of steroid-related toxicity, and enabling earlier consideration of further interventions [7, 8]. In addition, identification of such markers could provide timely risk stratification and prognostic information about the patient’s response to medications and progression to end-stage renal disease [6]. Moreover, NGAL has been shown to be a strong indicator of disease progression in patients with chronic kidney disease [9, 10]. Therefore, this systematic review and meta-analysis aimed to determine if urine NGAL measurements vary between patients with SRNS, SSNS, and healthy controls.

Methods

Search approach and studies inclusion criteria

This systematic review was conducted following the established guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Additional file 1) [11]. The review protocol was earlier registered on the Open Science Framework platform (https://osf.io/u86s5). In order to identify all relevant published research, we undertook an inclusive electronic literature search across PubMed, Web of Science, ScienceDirect, and the World Health Organization Virtual Health Library Regional Portal (WHO VHL). The search strategy did not impose any limitations based on geographical location or publication date, aiming to capture the widest possible range of relevant studies.

The search involved the published studies from inception up to March 2025, using the key words: (“nephrotic syndrome” OR “nephrosis”) AND (“neutrophil gelatinase-associated lipocalin” OR “neutrophil gelatinase associated lipocalin” OR “NGAL” OR “uNGAL” OR “lipocalin” OR “lipocalin-2” OR “LCN2 protein”). The references in the included articles were screened to make sure that no relevant studies have been missed. In addition, we conducted a search in Google Scholar along with our searches in the databases to identify potentially relevant grey literature. All publications were imported into Rayyan software (QCRI, Doha, Qatar; http://rayyan.qcri.org) to for titles/abstracts screening and duplicates deletion [12].

Inclusion and exclusion criteria

Following the literature search, the studies selection method involved a two-step approach. Firstly, three independent reviewers screened titles and abstracts of all identified articles to find potentially relevant studies. Those judged potentially relevant studies then underwent full-text assessment to determine their final eligibility based on the pre-defined inclusion criteria. Specifically, we included observational studies (cross-sectional, case-control, and cohort studies) that provided one or more of the following: (I) comparing NGAL or NGAL/Cr measurements between pediatric patients with SRNS and SSNS, (II) comparing NGAL or NGAL/Cr measurements in pediatric patients with nephrotic syndrome (SSNS or SRNS) vs. their healthy counterparts, (III) reporting diagnostic accuracy measures for NGAL or NGAL/Cr, (IV) reporting associations of NGAL or NGAL/Cr with markers of renal impairment or disease characteristics. On the contrary, studies were excluded if they fell into the following categories: studies lacking data of interest, studies on other kidney diseases or other biomarkers, reviews, editorials, case reports, abstracts from conferences, and non-English studies.

Quality assessment, data extraction, and certainty of evidence

The quality of included studies was assessed for potential bias using Joanna Briggs Institute critical appraisal checklists (https://jbi.global/critical-appraisal-tools). These checklists facilitate assessment of the risk of bias in study design, execution, and data analysis. Data retrieved from every study included study characteristics (author, year, country, sample size) and patient characteristics (age, sex, diagnosis, renal function test, NGAL, and NGAL/Cr levels). For studies reporting median and interquartile range (IQR), we converted these values to mean and standard deviation (SD) using the transformation method described by Wan et al. [13]. All discrepancies were resolved by discussion and consensus.

The GRADE framework (Grading of Recommendations Assessment, Development and Evaluation) was applied for analyzing confidence in the estimates for the outcomes. Level of evidence was categorized into four ranks: high, moderate, low, or very low. The assessment began at low certainty of evidence because of the inherent limitations of observational studies. The evidence certainty of was then downgraded or upgraded based on the GRADE domains.

Statistical analysis

The statistical analyses were performed by using the meta package in R version 4.4.3 to calculate the pooled standardized mean difference (SMD) and its 95% confidence intervals (CI). The random effects model, using the DerSimonian–Laird method, was preferred for calculation of the pooled effect sizes due to the presented studies heterogeneity, which was evaluated using the I² statistic. Publication bias assessment was done using both Begg’s and Egger’s regression tests [14–16]. When there was evidence of a publication bias, the Duval and Tweedie trim-and-fill method was applied to account for potentially missing studies [16]. The leave-one-out sensitivity analysis was conducted to evaluate the robustness of the meta-analysis by assessing the influence of individual studies on the outcomes. The significance level for all analyses was set at 0.05.

Results

Studies characteristics

The schematic flow of the study identification and selection process is presented in Fig. 1. The initial electronic database search yielded a total of 291 records. Following the removal of duplicate entries, 155 studies remained. These 155 studies underwent a thorough title and abstract screening process. During this stage, 137 studies were excluded due to irrelevance. Full texts of the remaining records were screened. Ultimately, a total of 16 studies were selected for the systematic review [17–32] (Fig. 1). A detailed assessment of the risk of bias across studies is presented in (Additional file 2).

Fig. 1.

low chart for studies selection process

Study publications ranged from 2010 to 2023 and included patients from seven countries. The patients with idiopathic nephrotic syndrome were generally divided into groups based on their response to steroid therapy (SSNS and SRNS). Some studies also included healthy individuals. All of the included studies assessed pediatric populations. All of the included studies indicated exclusion criteria, which were mainly fever, gross hematuria, acute kidney injury, CKD, urinary tract infection, and nephrotic syndrome secondary to systemic diseases, diabetes mellitus, viral infections, or autoimmune diseases. Measurements of urinary NGAL were done using various forms of enzyme-linked immunosorbent assay kits. The main features of the selected studies, as well as the risk of bias assessment, are presented in Tables 1 and 2.

Table 1.

Baseline characteristics of the studies included in the review

Study	Year	Country	Sample size	SRNS vs. SSNS	Male n, (%)	Age (years) Mean ± SD / median (range)
Bennet et al. [17]	2012	US	29	15 vs. 14	19 (65.5%)	8.75 ± 5.3
Bennet et al. [18]	2017	US	50	20 vs. 30	34 (68%)	9.4 ± 2.6
Bienias et al. [19]	2015	Poland	59	n/a	49 (83%)	13.7 (4–18)
Chehade et al. [20]	2013	Switzerland	44	n/a	n/a	10 (5–13)
Choudhary et al. [21]	2020	India	84	28 vs. 28	54 (60.7%)	8.06 ± 4.3
El-Gamasy et al. [22]	2017	Egypt	120	45 vs. 45	99 (82.5%)	5.4 ± 2.6
Ezzat et al. [23]	2011	Egypt	45	12 vs. 15	28 (62.2%)	8.7 ± 2.9
Gacka et al. [24]	2016	Poland	31	n/a	n/a	9.2 (3–12)
Gheissari et al. [25]	2013	Iran	104	n/a	69 (66.7%)	10.4 ± 3.73
Jwda et al. [26]	2022	Iraq	57	20 vs. 20	n/a	(range 3–13)
Kumar et al. [27]	2023	India	45	15 vs. 15	28 (62.2%)	(range 1–12)
Murugesan et al. [28]	2023	India	70	3 vs.32	(58.5%)	5.3 ± 3.2
Nickavar et al. [29]	2015	Iran	52	27 vs. 25	34 (65.4%)	(range 1–16)
Ochocinska et al. [30]	2018	Poland	102	14 vs. 29	53 (52%)	9 ± 5
Pradeepchandran et al. [31]	2022	India	79	7 vs. 72	44 (55.7%)	7.18 ± 2.86
Wasilewska et al. [32]	2010	Poland	37	n/a	24 (74%)	9.88 ± 5.73

Table 2.

Summary of main results on the association of urinary NGAL with nephrotic syndrome

Study	NGAL measurement	Main findings
Bennet et al.	ELISA kit (BioPorto Diagnostics, Gentofte Denmark)	Urinary NGAL were higher in SRNS (172.3 ng/ml, IQR 18.8–789) than both SSNS (6.3 ng/ml, IQR 4.9–9.9) and healthy controls (6.5 ng/ml, IQR 4.2–9.1). NGAL was inversely correlated with serum with eGFR. (r = -.5, p < .001).
Bennet et al.	ELISA kit (BioPorto Diagnostics, Gentofte Denmark)	Urinary NGAL were significantly higher in SRNS (30.77, 95CI% 15.01–63.08) than SSNS patients (5.57, 95CI% 3.10–10.00)
Bienias et al.	ELISA kit (USCNK, China)	Serum NGAL (12.36, range 3.7–23.03) and NGAL/Cr ratio (0.30, range 0.03–2.23) were higher in nephrotic syndrome patients than healthy controls [(4.77, range 2.1–10.4) and (0.12, range 0.01–1.07)]. Serum NGAL was not correlated with number of relapses.
Chehade et al.	ELISA kit (Architect i1000SR (Abbott Diagnostics, North Chicago, IL)	Urinary NGAL/Cr were higher in patients with FSGS than in patients with MCD (8.3, IQR 2.1–27.4) vs. (l, IQR 0.5–1.8). The AUC for the NGAL/Cr ratio as a biomarker to distinguish FSGS from MCD was 0.76
Choudhary et al.	ELISA kit (not specified)	Urinary NGAL were significantly higher in SRNS (701.12 ± 371.64 ng/mL) than in patients with SSNS (252.87 ± 66.34 ng/mL). Urinary NGAL had a strong discriminative power at a concentration of 13.1 ng/ml to detect SRNS (AUC: 0.90), based on the ROC curve analysis. Urinary NGAL correlated with eGFR and urine vitamin d binding protein levels.
El-Gamasy et al.	ELISA kit (BioPorto Diagnostics, Gentofte Denmark)	Urinary NGAL were significantly higher in SRNS (47.13 ± 14.06) than SSNS patients(24.13 ± 67.85). Urinary NGAL had a strong discriminative power to detect SRNS (AUC: 0.92). Urinary NGAL correlated with Cr level and disease duration.
Ezzat et al.	ELISA kit (not specified)	Serum (73.4 ng/mL ± 16.4) and urinary NGAL (5.8 ± 5.5) were higher in SSNS patients than healthy controls [(45.6 ng/mL ± 16.7)and (2.9 ± 1.1)].
Gacka et al.	ELISA kit (BioPorto Diagnostics, Gentofte Denmark)	No statistically significant correlation between the cyclosporine concentration and urinary NGAL concentration. However, NGAL was correlated with the duration of treatment
Gheissari et al.	ELISA kit (BioPorto Diagnostics, Gentofte Denmark)	Serum NGAL were higher in nephrotic syndrome patients (25.62 ng/mL ± 13.45) than healthy controls (12.86 ng/mL ± 3.77), with a significant correlation with proteinuria (r = .488), systolic blood pressure (r = .341), and diastolic blood pressure (r = .410).
Jwda et al.	ELISA kits (Abbott i1000)	Serum and urinary NGAL were significantly higher in SRNS (174.3 ± 18.4) than SSNS patients (80.9 ± 9.0), with strong discriminative powers to detect SRNS based on the ROC curve analyses.
Kumar et al.	ELISA kits (Fine test, Wuhan, China)	Urinary NGAL levels were significantly higher in SRNS (50 IQR 40.1–58.9) than SSNS patients (8.68, IQR 4.97–11.65). Urinary NGAL had a strong discriminative power to detect SRNS (AUC: 0.95).
Murugesan et al.	ELISA kit (USCN Life Science Inc., Boston, US)	Urinary NGAL levels were higher in patients with first episode nephrotic syndrome (74.4, IQR 38.4–123.8) compared to healthy individuals (54.6, IQR 15.9–117.7).
Nickavar et al.	ELISA kit (BioPorto Diagnostics, Gentofte Denmark)	Urinary NGAL/Cr ratio was significantly higher in SRNS (1.15, IQR 0.15–11.36) than SSNS patients (0.20, IQR 0.10–0.32).
Ochocinska et al.	ELISA Kit (BioVendor, Czech Republic)	There was no significant difference in the serum and urinary NGAL levels between SRNS and SSNS patients. Serum NGAL concentration were higher in patients on triple therapy (cyclosporine A, mycophenolate mofetil, and prednisone) compared to those on a double regimen without cyclosporine.
Pradeepchandran et al.	ELISA kit (Elabscience, Houston, USA)	Lower urine NGAL (< 10 ng/mL) was associated with steroid responsiveness in the first episode of nephrotic syndrome at 3 months. Higher urinary NGAL (> 10 ng/mL) was associated with hematuria and hypertension.
Wasilewska et al.	ELISA kit (BioPorto Diagnostics, Gentofte Denmark)	Serum and urinary NGAL, as well as NGAL/Cr ratio increased during the course of cyclosporine A treatment. Serum cyclosporine A was positively correlated with both urinary NGAL and NGAL/Cr ratio. NGAL/Cr ratio was positively correlated with serum Cr values and inversely with eGFR. However, the discriminative power for NGAL/Cr ratio to detect cyclosporine nephropathy, based on the ROC curve analysis, was weak.

steroid-sensitive nephrotic syndrome (SSNS), steroid-resistant nephrotic syndrome (SRNS), focal segmental glomerulosclerosis (FSGS), minimal change disease (MCD), Neutrophil gelatinase-associated lipocalin (NGAL), NGAL/creatinine (NGAL/Cr), estimated glomerular filtration rate (eGFR), Receiver Operating Characteristic (ROC), Area Under the Curve (AUC)

Urinary NGAL measurements in patients with nephrotic syndrome

We conducted meta-analyses to assess differences in the mean urinary NGAL levels between patients with SRNS, SSNS patients, and healthy individuals. The results indicated that urinary NGAL levels were significantly higher in both SSNS and SRNS patients compared to healthy controls (Figs. 2 and 3). Furthermore, direct comparison between SRNS and SSNS groups revealed significantly higher urinary NGAL levels in patients with SRNS (Fig. 4) (Table 3).

Fig. 2.

Forest plot for the pooled SMD of urinary NGAL between SRNS and SSNS patients

Fig. 3.

Forest plot for the pooled SMD of urinary NGAL between SSNS and healthy individuals

Fig. 4.

Forest plot for the pooled SMD of urinary NGAL between SRNS and healthy individuals

Table 3.

Summary of diagnostic performance in differentiating pediatric nephrotic syndrome subtypes

Study	Biomarker	Comparison	Cutoff Value	AUC (95% CI)	Sensitivity (%)	Specificity (%)
Bennett et al.	Urinary NGAL/Cr	SRNS vs. SSNS	15 ng/mg	0.91	88.0%	88.0%
Bienias et al.	Urinary NGAL	SRNS vs. SSNS	13.1 ng/mL	0.90	86%	89%
Chehade et al.	Urinary NGAL/Cr	FSGS vs. MCD	17 ng/mg	0.76 (0.57–0.95)	77.0%	78.0%
Choudhary et al.	Urinary NGAL	SRNS vs. SSNS	13.1 ng/mL	0.90 (0.82–0.98)	85.7%	89.3%
El-Gamasy et al.	Urinary NGAL	SRNS vs. SSNS	315 ng/mL	0.90	86.7%	93.3%
Jwda et al.	Urinary NGAL	SRNS vs. SSNS	137 ng/mL	NA	95%	95%
Kumar et al.	Urinary NGAL	SDNS vs. (SRNS + (SSNS)	40.02 ng/mL	0.95 (0.90-1.00)	80.0%	93.3%
Nickavar et al.	Urinary NGAL/Cr	SRNS vs. SSNS	0.46 ng/mg	0.76 (0.62–0.90)	63.0%	92.0%

FSGS: focal segmental glomerulosclerosis, MCD: minimal change disease, SDNS: steroid-dependent nephrotic syndrome

Specifically, patients with SSNS had significantly higher urinary NGAL levels than healthy individuals, with SMD = 0.782 (95% CI: 0.434 to 1.128, P < .001) (Fig. 3). Heterogeneity was low according to the I² test (32.2%). There was no evidence of publication bias based on Begg’s test (P = 1.00) and Egger’s test (P = .639). Likewise, patients with SRNS had significantly higher urinary NGAL levels than healthy individuals, with SMD = 2.56 (95% CI: 1.152 to 3.971, P < .001) (Fig. 4). Heterogeneity was high according to the I² test (92.2%). The publication bias was not significant for Egger’s test (P = .466) and Begg’s test (P = 1.00).

When comparing SRNS with SSNS, patients with SRNS had significantly higher urinary NGAL levels, with SMD = 1.889 (95%CI: 0.819 to 2.959, P < .001) (Fig. 2). Heterogeneity was high according to the I² test (93.7%). There was no evidence of publication bias based on Begg’s test (P = .293) and Egger’s test (P = .160).

Urinary ngal/cr ratio measurements in patients with nephrotic syndrome

Three studies assessed differences in the urinary NGAL/Cr ratio levels between patients with SRNS and SSNS patients. The meta-analysis results showed a non-statistically significant trend toward higher NGAL/Cr levels in SRNS compared to SSNS patients, with pooled SMD = 0.35 (95% CI: − 0.076 to 0.780, P = .107) (Fig. 5). Heterogeneity was low according to the I² test (I² = 10.8%). No publication bias was detected (Begg’s test P = .117; Egger’s test P = .392) (Additional file 3).

Fig. 5.

Forest plot for the pooled SMD of urinary NGAL/Cr between SRNS and SSNS patients

Serum NGAL measurements in patients with nephrotic syndrome

Only two studies (Gheissari et al. and Ezzat et al.) compared serum NGAL levels between patients and healthy controls. both showing significantly higher levels among nephrotic syndrome patients. In contrast, comparisons between SRNS and SSNS patients (Jwda et al. and Ochocinska et al.) yielded inconsistent results (Table 2).

Sensitivity analyses and certainty of evidence

Leave-one-out sensitivity analysis was performed to assess the influence of individual studies on the pooled effect estimates. The sensitivity analysis showed that all re-analyses, where each study was removed one at a time, yielded a statistically significant pooled effect estimate, indicating that no single study excessively influenced the overall results or caused excessive change to the pooled estimates (Additional file 4).

The lowest pooled estimate was obtained by removing the Jwda et al. study, which resulted in a much lower but still significant estimate (SMD = 1.16, P = .003). On the other hand, the highest estimate was obtained by removing the Ochocinskaet et al. study, which resulted in a much higher but still significant estimate (SMD = 1.99, P < 001). Since the studies by Jwda et al. and Kumar et al. demonstrated significant outlier effects in the meta-analyses, we conducted re-analyses excluding both studies. However, the results remained significant and the overall conclusions did not change (Additional file 5).

Subgroup analysis by geographic region showed variability. Studies from India (k = 2) and the Middle East (k = 3) showed larger and statistically significant effects (SMD = 2.50 and 2.78, respectively) compared to studies from Caucasian populations (k = 2), which showed a smaller, non-significant effect (SMD = 0.14). A test for subgroup differences was statistically significant (Q = 8.32, P = .016), indicating that effect sizes differed significantly by region (Additional file 5).

The certainty of evidence, evaluated by GRADE criteria, was rated as low for urinary NGAL outcomes and very low for the NGAL/Cr ratio outcome. The high heterogeneity in urinary NGAL results was counteracted by the large magnitude of effect and absence of other downgrading factors. The evidence for the NGAL/Cr ratio was downgraded to very low due to imprecision, as the confidence interval crossed the line of no effect and the conclusion was based on a limited number of studies (Additional file 6).

Association of NGAL measurements with markers of renal impairment

Most studies have evaluated the association of NGAL measurements with renal function and severity of renal impairment (Table 2). Estimated glomerular filtration rate (eGFR) and serum creatinine levels were the most common markers of kidney function assessed for their correlations with urinary and/or serum NGAL levels in the included studies. The studies were consistently showing a positive correlation between serum creatinine levels and a negative correlation with eGFR, with statistically significant results in most of the studies. The association between NGAL levels and proteinuria magnitude was assessed in a few studies, with inconsistent findings.

Several studies used Receiver Operating Characteristic (ROC) analysis to evaluate the diagnostic and discriminative performance of NGAL, reporting Area Under the Curve (AUC) values. Most showed that urinary NGAL and NGAL/Cr had moderate to strong ability to distinguish SRNS from SSNS (Table 4). A few studies examined NGAL’s association with cyclosporine use or potential toxicity. Gacka et al. and Wasilewska et al. reported correlations between urinary NGAL and duration of cyclosporine treatment. Ochocinskaet et al. observed higher serum NGAL concentration in patients treated with a triple-drug regimen including cyclosporine, compared to those on a double regimen without cyclosporine. However, Wasilewska et al. concluded that the NGAL/Cr ratio had weak discriminative power for detecting specific cyclosporine nephropathy (Table 2).

Table 4.

Summary of the meta-analyses results for the NGAL and ngal/cr measurements

Biomarker	Comparison	No. of studies	Pooled SMD	P value	Ref.
Urinary NGAL	SRNS Vs. SSNS	7	1.889 (0.819 to 2.959)	< 0.001	[18, 21, 22, 26, 27, 29, 30]
Urinary NGAL	SSNS Vs. healthy controls	5	0.78 (0.434 to 1.128)	< 0.001	[18, 21–23, 26]
Urinary NGAL	SRNS Vs. healthy controls	4	2.56 (1.152 to 3.971)	< 0.001	[18, 21, 22, 26]
NGAL/Cr ratio	SRNS Vs. SSNS	3	0.35 (-0.076 to 0.780)	0.107	[28–30]

Discussion

This systematic review and meta-analysis indicates that urinary NGAL levels are significantly elevated in SRNS compared to SSNS and healthy controls, suggesting its clinical utility as a non-invasive biomarker for identifying steroid resistance in idiopathic nephrotic syndrome at an early stage. The general trend across studies conducted in diverse populations, including children from India, Egypt, Europe, and North America, confirms the association between NGAL elevation and steroid resistance in nephrotic syndrome. This reproducibility supports NGAL’s potential for broad clinical application and integration into diagnostic workflows.

The elevated NGAL expression observed in SRNS can be attributed to ongoing tubular injury, which distinguishes it pathologically from SSNS [21]. NGAL is synthesized in renal tubular epithelial cells in response to inflammation, ischemia, and structural damage. Its presence in urine reflects subclinical renal stress, often preceding detectable changes in serum creatinine or eGFR [19, 23, 30]. In studies assessing these parameters, NGAL showed significant inverse relationships with eGFR and positive associations with both serum creatinine and proteinuria, reinforcing its biological role as a surrogate marker of renal insult.

Several investigations demonstrated high diagnostic performance for NGAL, with AUC values frequently exceeding 0.90. These findings highlight the strength of NGAL in accurately stratifying patients by their likelihood of responding to corticosteroids. Although the NGAL/Cr ratio showed slightly lower AUC values, typically ranging between 0.75 and 0.85, and slightly higher levels in SRNS [29, 30], its utility may remain relevant in clinical practice. Adjusting for creatinine concentration provides a pragmatic approach to normalize urinary biomarkers, enhancing consistency across varying hydration states and improving cross-sectional and longitudinal comparability.

In addition to its diagnostic function, urinary NGAL may assist in differentiating histopathological variants of nephrotic syndrome [20]. Furthermore, elevated NGAL concentrations have been linked to calcineurin inhibitor exposure, particularly cyclosporine, suggesting potential application in monitoring nephrotoxicity during immunosuppressive therapy [21].

Notably, the strengths of this review include sensitivity analysis, which further reinforces the validity of the pooled estimates. The convergence of findings from multiple independent investigations in diverse populations strengthens the evidence base for NGAL as a clinically relevant biomarker.

Nonetheless, the review has certain limitations. First, the predominance of pediatric data also limits generalizability to adult patients, who may present with different histological patterns or treatment responses. The lack of longitudinal data restricts conclusions regarding the prognostic value of NGAL over time or its responsiveness to changes in disease activity and treatment. Several studies had limited sample sizes or uneven group distributions, affecting statistical power. Moreover, sampling during various disease phases or in patients receiving concurrent therapies could confound biomarker interpretation. Although nearly half of the studies measured NGAL level using ELISA kits from the same manufacturer, differences in the types of ELISA kits used by other studies may have affected the overall results. Lastly, the studies lack data on renal tissue NGAL immunohistochemistry to correlate their levels with histopathological findings such as tubular injury and interstitial inflammation.

Future research should focus on establishing universally accepted cutoff thresholds for urinary NGAL and assessing urinary NGAL performance across diverse clinical settings. Investigating NGAL alongside other emerging biomarkers such as kidney injury molecule-1 may yield diagnostic panels that improve specificity and capture disease heterogeneity more effectively [20, 28]. Additionally, studies should explore the longitudinal utility of NGAL in predicting relapse, remission, and treatment response. Understanding its kinetics during immunosuppressive therapy and its relationship with histopathological progression may enhance its role as a dynamic biomarker for disease monitoring. Of particular interest is NGAL’s application in predicting steroid toxicity, nephrotoxicity, and resistance profiles, especially in vulnerable pediatric populations where therapeutic precision is critical.

Conclusion

Urinary NGAL demonstrated significant potential as an early, non-invasive marker for identifying steroid resistance in nephrotic syndrome. The diagnostic accuracy, biological plausibility, and relevance across populations support its inclusion in clinical assessment protocols. With further validation, NGAL could play a central role in personalizing treatment strategies, minimizing unnecessary steroid exposure, and improving outcomes in pediatric patients with nephrotic syndrome.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

SSNS

Steroid-sensitive nephrotic syndrome

SRNS

Steroid-resistant nephrotic syndrome

FSGS

Focal segmental glomerulosclerosis

MCD

Minimal change disease

SDNS

Steroid-dependent nephrotic syndrome

NGAL

Neutrophil gelatinase-associated lipocalin

NGAL/Cr

NGAL/creatinine

SMD

Standardized mean difference

CI

Confidence intervals

eGFR

Estimated glomerular filtration rate

ROC

Receiver Operating Characteristic

AUC

Area Under the Curve

GRADE

Grading of Recommendations Assessment, Development and Evaluation

Author contributions

SM conceptualized the research idea. AA, AA, and IA managed and planned the research process. SM, SA, and SO undertook database searches and articles screening. AA, TE, and SM undertook risk of bias assessment. SM, IA, AE, YK, and HK, extracted and summarized data. SM analyzed data. AA, AA, IA, and NA drafted the manuscript. All authors read and approved the final manuscript. SM conceptualized the research idea. AA, AA, and IA managed and planned the research process. SM, SA, and SO undertook database searches and articles screening. AA, TE, and SM undertook risk of bias assessment. SM, IA, AE, YK, and HK, extracted and summarized data. SM analyzed data. AA, AA, IA, and NA drafted the manuscript. All authors read and approved the final manuscript.

Funding

No fund.

Data availability

Data are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.