Abstract
Objective:
Determining when a positive culture represents a urinary tract infection (UTI) versus colonization in children with spina bifida is challenging. Combining markers of host response with urobiome parameters may improve diagnostic accuracy.
Study design:
Cross-sectional pilot study of children with spina bifida who were evaluated for UTI in the Emergency Department. We included children under the age of 22 years with spina bifida who used clean intermittent catheterization and had a urinalysis and urine culture sent. We measured urine neutrophil gelatinase-associated lipocalin (NGAL) and completed 16SrRNA sequencing. We generated receiver operating characteristic curves to assess the predictive accuracy of urinary NGAL and Shannon Diversity Index.
Results:
There were 51 children, 13 with UTI, in this study. Median Shannon Diversity Index was lower in children with UTI compared to those without, while median NGAL was significantly higher in the UTI versus the no UTI group. The AUC for the combination of NGAL and Shannon Diversity was 0.88 (0.78-0.99), which is not significantly different from either the AUC for NGAL or Shannon Diversity Index alone. NGAL was more specific, while Shannon Diversity Index was more sensitive for the diagnosis of UTI.
Conclusion:
NGAL is specific, and Shannon Diversity Index is sensitive, for UTI. However, the combination of the two markers does not have improved predictive accuracy over either alone for the diagnosis of UTI in children with spina bifida who use clean intermittent catheterization.
Introduction:
Children with spina bifida frequently have bacteriuria, with one study reporting a clinically-relevant amount of bacteria present in approximately 70% of asymptomatic children.1 However, determining when a positive urine culture represents a urinary tract infection (UTI) as opposed to asymptomatic bacteriuria is difficult. There is currently no widely-accepted consensus on how to define a UTI in this unique patient population.2 Given the risks of untreated infection, including renal scarring, which one study reported was present in 43% of assessed renal units,3 and the lack of widely-accepted diagnostic guidelines, children with spina bifida are often prescribed antibiotics, leading to higher rates of antibiotic resistant organisms cultured from urine compared to children without spina bifida.4,5
The currently used biomarkers of UTI, which include both leukocyte esterase (LE) and pyuria, have suboptimal predictive accuracy for UTI in children with spina bifida.6 Children with spina bifida often have chronic inflammation of the urothelium due to both their underlying neurologic injury and irritation related urothelial catheterization, resulting in pyuria in the absence of infection.7 Urinary neutrophil-gelatinase associated lipocalin (NGAL) is a novel biomarker that has good predictive accuracy for UTI in both the general pediatrics population as well as children with neurogenic bladders.8,9 Despite the improved diagnostic accuracy of uNGAL for UTI, it remains imperfect: studies to date have found that the Shannon Diversity Index, a measure of alpha diversity that assesses both the number of different species and how balanced those species are within the community, is lower in otherwise healthy children with UTI compared to those without.10 However, the urobiome in people with spina bifida is often different from those without spina bifida,11 and may not exhibit the same pattern during UTI. Further, the urobiome only reflects the microbial ecosystem within the bladder and not the host response to those bacteria. Thus, combining urobiome indices with NGAL may have improved diagnostic accuracy for UTI in children with spina bifida compared to either marker alone as this combination reflects both the changes in the urobiome, as well as the host response to those changes.
The objective of this pilot study was to assess the predictive accuracy of NGAL and the Shannon Diversity Index, both individually and combined, in diagnosing UTI in children with spina bifida who use clean intermittent catheterization. We hypothesized that the predictive accuracy of the combination of NGAL and Shannon Diversity Index for the diagnosis of UTI would be better than that of either marker alone.
Methods:
This was a pilot, prospective cross-sectional study of children with spina bifida who presented to the Emergency Department of one of two urban quaternary care centers (e.g., University of Pittsburgh Medical Center Children’s Hospital of Pittsburgh and Cincinnati Children’s Hospital Medical Center) and were evaluated for UTI between 11/1/2022 and 3/1/2025. Families of eligible children were approached for consent after their residual urine was obtained for use in this study. We obtained informed consent from the parents or legal guardians for participation in this study and obtained assent from children as appropriate by age and / or developmental stage in accordance with ethical guidelines for pediatric research. Those who consented completed several instruments to provide data on presenting symptoms and routine aspects of bowel and bladder care. Any collected urine samples from a child who did not consent or was not able to be reached for consent was discarded. A single IRB at the University of Pittsburgh approved this study (IRB 21020133).
Patients:
Children were eligible for this study if they were less than 22 years of age, had a diagnosis of spina bifida, used clean intermittent catheterization for bladder management, and had a urine sample sent in the Emergency Department to evaluate for UTI as documented in the provider’s note. We excluded children for the following reasons: current immunosuppression, diagnosis of diabetes mellitus, history of taking treatment antibiotics within two weeks of presentation (children on prophylactic antibiotics were eligible), end stage renal disease, or a known kidney or bladder stone. No patient enrolled in this study required surgical intervention by Urology either at the time of enrollment or during their subsequent admission, if they were admitted.
Sample Size:
An a priori sample size of 50 participants was calculated for this pilot study based upon prior data in the literature on Shannon Diversity Index in children with spina bifida and UTI.12 We specifically powered this study to assess a difference in Shannon Diversity Index as there are multiple reports in the literature examining the utility of NGAL in diagnosing UTI in people with neurogenic bladders,9,13,14 but very few reports examining the diagnostic utility of the alpha diversity of the urobiome.
Urine samples:
The urine samples analyzed in this study were residual clinical samples, all of which had been obtained by catheterization as part of routine clinical care. Urine samples were obtained from the clinical laboratory within 24 hours of being collected from the patient, during which time samples were stored at 4° Celsius. After collecting samples from the clinical laboratory, urine was processed in several different ways. We saved a minimum of 2mL of urine for urobiome analysis and added AssayAssure (Sierra Molecular, Princeton, NJ) in a 1:9 ratio prior to freezing at −80° Celsius.15 We then centrifuged the remaining urine and aliquoted the supernatant for NGAL measurements. These samples were also frozen at −80° Celsius. There were no freeze-thaw cycles between freezing at −80° Celsius and when the samples were thawed for analysis.
Definitions:
We defined UTI using a modified definition that is included as part of the Urological Management the Preserve Initial Renal function (UMPIRE) protocol.16 The UMPIRE protocol is a multicenter effort support by the Centers for Disease Control to standardize the approach urologic care for infants and young children with spina bifida. One metric that is followed within the UMPIRE protocol is UTI, which is defined using a consensus-based definition developed and agreed upon by experts in the urological care of children with spina bifida. The definition of UTI that we used in this study was: growth of ≥50,000 colony-forming units per milliliter (cfu/mL) of one or two uropathogens and either >10 urinary white blood cells/high powered field or >2+ LE in a child with two or more of the following symptoms: fever; gross hematuria; abdominal, suprapubic of flank pain; new or worse incontinence; new or worse urinary frequency or urgency; pain with catheterization; and malodorous or cloudy urine. We chose to decrease the bacterial burden cut-off from 100,000 cfu/mL (which was included in the initial UMPIRE definition) to 50,000 cfu/mL in accordance with the American Academy of Pediatrics’ definition of UTI, which uses a cut-off of 50,000 cfu/mL,17 and the lack of empiric data to support that using a higher cut-off is more indicative of infection.
Urobiome analysis:
DNA isolation and 16S sequencing was done by the SeqCenter (Pittsburgh, PA). All DNA extractions for this study were done using the ZymoBIOMICS™ DNA Miniprep Kit (Zymo, Irving, CA) with final DNA concentrations determined by Qubit. DNA samples were then prepared using Zymo Research’s Quick-16S kit (Zymo, Irving, CA) with phased primers targeting the V3/V4 regions of the 16S gene. The specific primer sequences used in this work are found in the supplemental materials (supplemental table 1). Following clean up and normalization, samples were sequenced on a P1 or P2 600cyc NextSeq2000 Flowcell (Illumina, San Diego, CA) to generate 2x301bp paired end reads. Quality control and adapter trimming was performed with bcl-convert (v4.2.4). Primer-dimer sequences identified as PCR artifacts were filtered from the generated FASTQ files based on the following criteria: read length > 150 basepairs, PolyN strings < 10 sequential Ns; PolyG strings < 150 sequential Gs. The team that performed the sequencing was blinded to all patient information.
Bioinformatics:
We used Qiime2 for analysis.18 We used the cutadapt2 plugin to remove primer sequences, and DADA2 to denoise the sequences.19 Denoised sequences were placed into a feature table detailing which amplicon sequence variants (ASVs) were observed in which samples, and how many times each ASV was observed in each sample. Identified ASVs were taxonomically assessed using the Silva 138 99% full-length sequence database20 and the VSEARCH4 utility within Qiime2’s feature-classifier plugin. ASVs were then collapsed to their lowest taxonomic units (strain, species, genus, etc.), and their counts were converted to reflect their relative frequency within a sample. We then combined ASVs into genus-level identification as not all ASV were identifiable down the strain level, but the majority (80% of the initial ASVs) were identified to the taxa level. To decontaminate the data, we first removed as ASVs that were only in the negative controls and not in samples (n=3). We then removed any ASV where there was a higher prevalence of the ASV in the negative control compared to the sum of the entire cohort of samples (n=11). We then used decontam on the remaining samples and individually examined the data for each ASV to identify if they were contaminants.21 In the end, we did not remove any of the remaining ASVs.
NGAL measurement:
We used ELISA to measure urine NGAL according to the manufacturer’s instruction. (Bioporto, Grusbakken, Denmark) We did not normalize urine NGAL results by urine creatinine given the different biologies of urine NGAL (produced after the glomerulus)22 and urine creatinine (filtered by the glomerulus), and the fact that children with spina bifida often have altered muscle mass, thus affecting their urine creatinine.23 Finally, reports in the literature support that normalization by urine creatinine does not alter trends in urine NGAL levels when assessing for UTI.8 Members of the research team who performed the ELISA and calculated the subsequent values were blinded to all patient information.
Statistics:
The main analysis compared children with and without UTI. We compared medians with Wilcoxon Rank Sum test and means with t-test. We compared categorical variables with the chi-square (or Fisher’s Exact, as relevant) test. We used PERMANOVA (permutational multivariate analysis of variance), a non-parametric statistical test used to determine whether groups differ in their overall multivariate composition, to assess differences in the composition of the urobiome between children with and without UTI (i.e., beta-diversity). We used the package pROC24 in R25 to generate receiver operating characteristic curves for each biomarker by plotting sensitivity versus 1-specificity across all possible threshold values with UTI diagnosis as the binary outcome. We generated the receiver operating characteristic curve of the combined biomarker model using logistic regression–derived predicted probabilities. We calculated the area under the curve (AUC) using the trapezoidal method. We determined optimal cut-off values for each marker using the 'best' function in pROC, which maximizes Youden's Index (sensitivity + specificity - 1), and used DeLong’s test within the pROC package to compare ROC curves. We did not include other variables in the model given the limited number of children in the UTI group. Similarly, we also did not account for clustering by site. There was no missing data within the data presented for our limited cohort.
Results:
We initially identified 255 children that were eligible for enrollment, of which 51 were ultimately enrolled in this study. (Supplemental figure 1) Children in this study had a median age of 7.0 (3.0-12.5) years with a slight male predominance. (Table 1) There was no difference in age, sex, or use of antibiotic prophylaxis between those with and without UTI. There was no difference in mean age or sex between children who did and did not consent to participate. There was also no difference in the proportion of children admitted to the hospital who did and did not have UTI. The most common diagnosis among children who did not meet our definition of UTI was viral infection (n= 12, 31.6%), followed by UTI (n=10, 26.3%), headache (n=5, 13.2%), other bacterial infection (n=4, 10.5%), non-infectious gastrointestinal issues (n=4, 10.5%), and other (n=3, 2.6%). Of the ten children diagnosed with UTI who did not meet our criteria, six had urine culture with unidentified mix growth of organisms, one had a negative urine culture, two did not have any of the included symptoms, and one only had cloudy urine.
Table 1.
Patient demographics and clinical characteristics
| No UTI (n=38) | UTI (n=13) | |
|---|---|---|
| Mean age | 7.6 (5.7) | 9.9 (6.0) |
| Age (categorical) | ||
| < 3yrs | 8 (21.1) | 1 (7.7) |
| 3-<12 years | 20 (52.6) | 7 (53.8) |
| >12 years | 10 (26.3) | 5 (38.5) |
| Female | 16 (42.1) | 7 (53.8) |
| Antibiotic prophylaxis | 6 (15.8) | 4 (30.8) |
| Routinely use enemas / flushes | 20 (52.6) | 9 (69.2) |
| Hospital admission | 18 (47.4) | 8 (61.5) |
Laboratory parameters are shown in table 2. There were differences in both LE and degree of urine white blood cells between groups. Most children in the UTI group had 3+ LE , whereas most children in the no UTI group had negative LE. Similarly, most children in the UTI group had >50 urinary white blood cells whereas most children in the no UTI group had less than 10 urinary white blood cells. Of note, 22.2% of children in the no UTI group had >50 urinary white blood cells. There was no difference in the proportion of children with positive nitrites on their UA between groups. (Table 2) The most common uropathogen isolated on urine culture in the UTI group was E. coli (53.8%) followed by Proteus (30.8%) and Klebsiella (15.4%). Most cultures did not have any growth in the no UTI group (55.2%). The most common organisms that were cultured from children in the non-UTI group were E. coli (15.8%), Klebsiella (10.5%) and Enterococcus (5.3%). (supplemental table 2). In the no UTI group, 10 (58.8%) cultures with any growth had >100,000 CFU/mL of bacterial growth, 1 (5.9%) was between 50,000-100,000 CFU/mL, and 6 (35.3%) were between 10,000-50,000 CFU/mL. There were 14 children without UTI who had positive nitrites on their UA. The most common reason that these children did not meet UTI criteria was lack of a positive urine culture (three children had negative urine cultures, five children had unidentified mixed growth), followed by not meeting symptom criteria (n=5), and not having pyuria in their urine (n=1). Thirteen out of the 14 of children with positive nitrites but without UTI had a urobiome predominated by Gram negative bacteria.
Table 2:
Urinalysis Results
| No UTI (n=38) | UTI (n=13) | p-value | |
|---|---|---|---|
| Mean pH | 6.6 (1.0) | 6.5 (0.9) | 0.65 |
| LE | <0.01 | ||
| None | 16 (42.1) | 0 (0) | |
| trace | 3 (7.9) | 0 (0) | |
| 1+ | 7 (18.4) | 3 (23.1) | |
| 2+ | 4 (10.5) | 2 (15.4) | |
| 3+ | 8 (21.1) | 8 (61.5) | |
| Positive nitrites | 14 (36.8) | 8 (61.5) | 0.22 |
| Urinary white blood cells* (per high-powered field) | <0.01 | ||
| <10 | 19 (52.7) | 0 (0) | |
| 10-20 | 7 (19.4) | 1 (7.7) | |
| 21-50 | 2 (5.6) | 4 (30.8) | |
| >50 | 8 (22.2) | 8 (61.5) | |
2 children without UTI did not have a microscopic urinalysis sent
The community structure of urobiomes from children with UTI versus those without UTI is shown in relative abundance plots in Figure 1. There is no difference in beta diversity between the urobiomes of those with and those without UTI (PERMANOVA=0.42). Median (interquartile range) Shannon Diversity Index were lower in children with UTI compared to those without (0.0082 (0.13) vs 0.76 (2.13), p<0.01). Median NGAL was significantly higher in the UTI versus the no UTI group (208.0 (1493.0) ng/mL vs 20.7 (60.4) ng/mL, p<0.01). When we examined whether there was a change in urobiome structure by NGAL value, we found that children with less diverse urobiomes that had higher proportions of Escherichia tended to have higher NGAL, whereas children whose urobiomes were more diverse tended to have lower NGAL.(Supplemental figure 2).
Figure 1:
Relative abundance of taxa (genus level) in the urobiome of children with spina bifida who do and do not have UTI. Each color represents a different genus, with the most abundant taxa identified in the legend.
The AUC for NGAL for the diagnosis of UTI is 0.81 (0.63-0.98) at a cut-off of 177.1 ng/mL. Comparatively, the AUC for Shannon Diversity Index for UTI is 0.77 (0.62-0.91) at a cut-off of 0.32. NGAL is more specific (0.88 (0.42-0.98)) while Shannon Diversity Index is more sensitive (0.81 (0.45-1.0)) for the diagnosis of UTI. The combination of both NGAL and Shannon Diversity Index had an AUC of 0.88 (0.78-0.99), which is not significantly different from the AUC of either Shannon Diversity (p=0.10) or NGAL (p=0.15) alone. (Table 3)
Table 3:
Predictive Accuracies
| AUC | Cut-off | Sensitivity | Specificity | |
|---|---|---|---|---|
| NGAL | 0.81 (0.63-0.98) | 177.1 | 0.73 (0.27-0.91) | 0.88 (0.42-0.98) |
| Shannon Diversity Index | 0.77 (0.62-0.91) | 0.32 | 0.81 (0.45-1.0) | 0.63 (0.33-0.90) |
| Combination | 0.88 (0.78-0.99) | -- | 0.82 (0.10-1.0) | 0.90 (0.5-1.0) |
Discussion:
In this pilot study, we report that both urine NGAL and Shannon Diversity Index are significantly different in children with spina bifida who have a UTI versus those who do not. We found that NGAL is a specific marker, while Shannon Diversity Index is a sensitive marker, of UTI. However, the AUC of the combination of NGAL and Shannon Diversity index is not different than the AUC of either alone.
Our results are consistent with prior reports documenting the utility of NGAL as a marker of UTI in both children and adults with neurogenic bladders.9,13,14 Our cut-off values are somewhat similar to one study of children with neurogenic bladders, which was 229ng/mL.9 The other study that reports urine NGAL in children with spina bifida reported NGAL levels that were normalized by urine creatinine, and thus are not directly comparable.14 However, our methods are distinct from the other papers in the pediatric literature in three critical ways: First, our study included prospective evaluation of symptoms instead of relying on retrospective chart review in one study.9 Second, we also included assessments of the urobiome along with NGAL measurements, which was not done in prior work. Finally, our population were all children who presented to the ED and had a UA to assess for UTI, which is distinct from other studies, both which also included children undergoing urodynamics in addition to those who presented to the ED.
There are studies that compare the urobiome in otherwise healthy children with and without UTI that report a decrease alpha diversity in UTI, similar to what we present here.10 While other studies in the literature have described the urobiome in children with spina bifida and compared it between those with and without UTI, our study is distinct for several reasons. First, other work did not use the same urine processing measures,12 most notably adding AssayAssure prior to freezing urine samples.15 Our samples, which had AssayAssure included prior to freezing, are a more accurate representation of the urobiome than those preserved without a DNA preservative.15 Second, our clinical cohort is less heterogenous than that published previously, thus allowing for greater confidence that our results accurately represent the pediatric spina bifida population.12 A more recent study compared the baseline urobiome of children with spina bifida to healthy controls, but did not include children who were either evaluated for, or diagnosed with, UTI in their analysis.26
One interesting finding of our work is the lack of difference in beta diversity between children with spina bifida who do and do not have a UTI. This is not the case in children with normal-functioning bladders, where several reports have found differences in beta diversity in the setting of UTI.10,27 The lack of difference in the urobiome beta diversity between children with spina bifida who do and do not have UTI reflects that many children with spina bifida have perturbed urobiomes compared to children without spina bifida. Indeed, children with spina bifida often have a baseline urobiome that is predominated by bacteria with uropathogenic potential, whereas the urobiome of children without spina bifida tends to have higher proportions of bacterial without uropathogenic potential.12 Therefore, it is likely that in a UTI in a child with spina bifida, there is a change in the proportion of the taxa in the urobiome rather than a complete shift in composition. Indeed, this is reflected in the subgroup of 14 children who did not meet UTI criteria but had positive nitrites on their UA, the majority of which had urobiomes predominated by Gram negative bacteria. One potential explanation for this finding is that nitrites are produced by some Gram negative bacteria, and therefore only indicate the presence of a Gram negative bacteria rather than its interaction with the urothelium. However, the possibility that these children could either have an indolent or early infection cannot be excluded using our data.
We did not find that the combination of NGAL and Shannon Diversity Index significantly improved diagnostic accuracy over either alone. One potential reason for this may be small sample size: we powered our study to find a difference in Shannon Diversity Index but may have under powered the study to examine the combination. It also may be that there is more clinical heterogeneity than anticipated in our cohort, which may have affected our results. However, the real value in the combination of both NGAL and Shannon Diversity is that it allows for optimization of both sensitivity (driven by Shannon Diversity Index) and specificity (driven by NGAL). Although higher sensitivity is helpful, it is the improved specificity, driven by NGAL, will have the real impact on antibiotic prescription. Using previous data on the utility of UA parameters for the diagnosis of UTI in children with spina bifida,6 assessing NGAL as compared to nitrites to determine when to start antibiotics would save 119 courses of antibiotics per 1000 patients. When compared to pyuria, the improved specificity of NGAL would prevent and estimated 425 antibiotic courses per 1000 patients. Thus, the next best step in clinical translation would be to implement NGAL testing in children with spina bifida being evaluated for UTI and clinically monitor those children with low NGAL while holding antibiotics.
Our study is not without limitations. Although we identified 255 eligible children, we were only able to enroll 50 children, of which only a small number met our definition of UTI. Further, while we chose a consensus definition of UTI, it is possible that we have misclassified patients due to difficulty in determining true UTI in this population. This definition also includes the symptoms of cloudy and foul-smelling urine, which we acknowledge can also be present in asymptomatic bacteriuria. It is also possible that due to the heterogeneity in sensation experienced by children with spina bifida and the inclusion of preverbal children in this work, we were not able to capture all symptoms. We also did not include history of UTI as a covariate given our inability to validate whether reported prior episodes met our definition of UTI. Finally, there is no current way to rapidly assess urobiome alpha diversity, thus limiting the ability to translate these results into clinical care. However, given the rapid rate at which biomedical technology is makes advances, and it is very possible that there will soon be a method to rapidly assess the urobiome’s alpha diversity. Despite these limitations, our study also has several strengths, which include the prospective collection of symptoms, the combination of urobiome analysis with markers of host response, and that we enrolled children from two separate sites. Finally, we only included children who were assessed for UTI.
Conclusion:
Here, we show that the combination of urine NGAL and Shannon Diversity Index of the urobiome have good predictive accuracy for the diagnosis of UTI in children with spina bifida. Next steps will include validation of these findings in a larger cohort, inclusion of a broader population of children with spina bifida such as those who do not catheterize and those who have undergone a bladder augment, and understanding the community dynamics of the urobiome. The combination of these assays may help identify children with spina bifida who are unlikely to have a UTI and can be safely monitored off antibiotics.
Supplementary Material
Supplemental Figure 1: CONSORT diagram indicating the number of eligible children, those who met exclusion criteria, and the number of children ultimately enrolled. Of note, exclusion categories are not mutually exclusive.
Supplemental Figure 2:Relative abundance plot of urobiomes of all included children arranged in order by urine NGAL values, with the highest NGAL on the left and the lowest NGAL on the right.
Funding:
This work was supported by the NIH (1K23DK129783), and in part by G. Nicholas III and Dorothy B. Beckwith.
Footnotes
Conflict of Interest: None of the authors have any conflicts of interest to disclose
Credit Author Statement: CF was involved in conceptualization, data curation, performed the formal analysis, acquired the funding, developed the methodology, administered the project, and wrote the first draft. BS, AG, CC, and KT were involved in investigation, methodology, data curation, and critical review of the manuscript. SDR and NS were involved in methodology, data curation, data interpretation, supervision, and critically reviewed the manuscript. All authors approved submission of the final version.
References:
- 1.Schlager TA, Dilks S, Trudell J, Whittam TS, Hendley JO. Bacteriuria in children with neurogenic bladder treated with intermittent catheterization: Natural history. The Journal of Pediatrics. 1995;126(3):490–496. doi: 10.1016/S0022-3476(95)70477-9 [DOI] [PubMed] [Google Scholar]
- 2.Forster CS, Kowalewski NN, Atienza M, Reines K, Ross S. Defining Urinary Tract Infections in Children With Spina Bifida: A Systematic Review. Hosp Pediatr. 2021;11(11):1280–1287. doi: 10.1542/hpeds.2021-005934 [DOI] [PubMed] [Google Scholar]
- 3.Omay S, Yavuz S, Akkas BE, Omak Ö, Kaya M, Kaya EM. Correlation between the clinical features of patients with spina bifida and the findings of DMSA scintigraphy. Eur J Pediatr. 2025;184(12):792. doi: 10.1007/s00431-025-06601-y [DOI] [PubMed] [Google Scholar]
- 4.Ortiz TK, Velazquez N, Ding L, et al. Predominant bacteria and patterns of antibiotic susceptibility in urinary tract infection in children with spina bifida. Journal of Pediatric Urology. 2018;14(5):444.e1–444.e8. doi: 10.1016/j.jpurol.2018.03.017 [DOI] [PubMed] [Google Scholar]
- 5.Forster CS, Courter J, Jackson EEC, Mortensen J, Haslam DBD. Frequency of Multi-Drug Resistant Organisms Cultured from Urine in Children on Clean Intermittent Catheterization. Journal of the Pediatric Infectious Disease Society. 2016;6(4):1–7. doi: 10.1093/jpids/piw056 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Forster CS, Miller RG, Gibeau A, et al. Accuracy of Urinalysis for UTI in Spina Bifida. Pediatrics. 2024;154(1):e2023065192. doi: 10.1542/peds.2023-065192 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Schlager T a, Grady R, Mills SE, Hendley JO. Bladder epithelium is abnormal in patients with neurogenic bladder due to myelomeningocele. Spinal cord : the official journal of the International Medical Society of Paraplegia. 2004;42(3):163–168. doi: 10.1038/sj.sc.3101565 [DOI] [PubMed] [Google Scholar]
- 8.Forster CS, Lubell TR, Dayan PS, Shaikh N. Accuracy of NGAL as a Biomarker for Urinary Tract Infection in Young Febrile Children: An Individual Patient Data Meta-Analysis. The Journal of Pediatrics. 2023;258:113394. doi: 10.1016/j.jpeds.2023.113394 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Forster CS, Jackson E, Ma Q, Bennett M, Shah SS, Goldstein SL. Predictive ability of NGAL in identifying urinary tract infection in children with neurogenic bladders. Pediatric nephrology (Berlin, Germany). 2018;33(8):1365–1374. doi: 10.1007/s00467-018-3936-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Marshall CW, Kurs-Lasky M, McElheny CL, Bridwell S, Liu H, Shaikh N. Performance of Conventional Urine Culture Compared to 16S rRNA Gene Amplicon Sequencing in Children with Suspected Urinary Tract Infection. Microbiology Spectrum. 2021;9(3):e01861–21. doi: 10.1128/spectrum.01861-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Groah SL, Pérez-Losada M, Caldovic L, et al. Redefining Healthy Urine: A Cross-Sectional Exploratory Metagenomic Study of People With and Without Bladder Dysfunction. The Journal of Urology. 2016;196(2):579–587. doi: 10.1016/j.juro.2016.01.088 [DOI] [PubMed] [Google Scholar]
- 12.Forster CS, Panchapakesan K, Stroud C, Banerjee P, Gordish-Dressman H, Hsieh MH. A cross-sectional analysis of the urine microbiome of children with neuropathic bladders. Journal of Pediatric Urology. 2020;16(5):593.e1–593.e8. doi: 10.1016/j.jpurol.2020.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Forster CS, Lamanna OK, Rounds A, Sprague BM, Ljungberg I and GS. The association between urine neutrophil gelatinase-associated lipocalin and UTI in people with neurogenic lower urinary tract dysfunction. Spinal Cord. Published online 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gupta S, Preece J, Haynes A, Becknell B, Ching C. Differentiating Asymptomatic Bacteriuria From Urinary Tract Infection in the Pediatric Neurogenic Bladder Population: NGAL As a Promising Biomarker. Topics in Spinal Cord Injury Rehabilitation. 2019;25(3):214–221. doi: 10.1310/sci2503-214 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jung CE, Chopyk J, Shin JH, et al. Benchmarking urine storage and collection conditions for evaluating the female urinary microbiome. Scientific Reports. 2019;9(1):13409. doi: 10.1038/s41598-019-49823-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Routh JC, Cheng EY, Austin JC, et al. Design and Methodological Considerations of the Centers for Disease Control and Prevention Urologic and Renal Protocol for the Newborn and Young Child with Spina Bifida. The Journal of urology. 2016;196(6):1728–1734. doi: 10.1016/j.juro.2016.07.081 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Roberts KB, Downs SM, Finell SME, et al. Urinary tract infection: clinical practice guideline for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months. Pediatrics. 2011;128(3):595–610. doi: 10.1542/peds.2011-1330 [DOI] [PubMed] [Google Scholar]
- 18.Hall M, Beiko RG. 16S rRNA Gene Analysis with QIIME2. In: Beiko RG, Hsiao W, Parkinson J, eds. Microbiome Analysis: Methods and Protocols. Springer; 2018:113–129. doi: 10.1007/978-1-4939-8728-3_8 [DOI] [PubMed] [Google Scholar]
- 19.Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nature methods. 2016;13(7):581–583. doi: 10.1038/nmeth.3869 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Quast C, Pruesse E, Yilmaz P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic acids research. 2013;41(Database issue):D590–6. doi: 10.1093/nar/gks1219 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Davis NM, Proctor DM, Holmes SP, Relman DA, Callahan BJ. Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome. 2018;6(1):226. doi: 10.1186/s40168-018-0605-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Paragas N, Kulkarni R, Werth M, et al. α-Intercalated cells defend the urinary system from bacterial infection. Journal of Clinical Investigation. 2014;124(7):2963–2976. doi: 10.1172/JCI71630 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Quan A, Adams R, Ekmark E, Baum M. Serum creatinine is a poor marker of glomerular filtration rate in patients with spina bifida. Developmental Medicine & Child Neurology. 1997;39(12):808–810. doi: 10.1111/j.1469-8749.1997.tb07547.x [DOI] [PubMed] [Google Scholar]
- 24.Robin X, Turck N, Hainard A, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics. 2011;12(1):77. doi: 10.1186/1471-2105-12-77 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Computing RF for SR: A language and environment for statistical computing. 2013. [Google Scholar]
- 26.De Maio F, Grotti G, Mariani F, et al. Profiling the Urobiota in a Pediatric Population with Neurogenic Bladder Secondary to Spinal Dysraphism. International Journal of Molecular Sciences. 2023;24(9):8261. doi: 10.3390/ijms24098261 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kinneman L, Zhu W, Wong WSW, et al. Assessment of the Urinary Microbiome in Children Younger Than 48 Months. The Pediatric infectious disease journal. 2020;39(7):565–570. doi: 10.1097/INF.0000000000002622 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1: CONSORT diagram indicating the number of eligible children, those who met exclusion criteria, and the number of children ultimately enrolled. Of note, exclusion categories are not mutually exclusive.
Supplemental Figure 2:Relative abundance plot of urobiomes of all included children arranged in order by urine NGAL values, with the highest NGAL on the left and the lowest NGAL on the right.