Risk factors associated with recurrent urinary tract infections in children with bladder dysfunction: a retrospective cohort study

Introduction

Urinary tract infection (UTI) is a commonly encountered condition in children, particularly in individuals with predisposing factors. In children, UTIs can be severe and potentially lead to renal damage, especially in cases of recurrent UTIs (rUTIs) (1). Bladder dysfunction represents a significant risk factor for the occurrence of UTIs (2,3). Such dysfunction may stem from structural or nervous system abnormalities, which can be congenital or acquired later in life. Bladder dysfunction can be attributed to neurologic deficits such as spinal dysraphism, sacral agenesis, and cerebral palsy. Alternatively, it may result from structural anomalies, such as posterior urethral valve (PUV), or anomalies of the urinary bladder, as observed in conditions like prune belly syndrome and cloacal anomalies (4). These bladder dysfunctions often manifest as urinary retention, increased or decreased frequency of urination, and hesitancy in initiating urination. In many cases, complete voiding remains unattainable, leading to urine retention in the bladder and subsequently increasing susceptibility to UTIs (5).

However, bladder dysfunction rarely exists in isolation; it is frequently compounded by co-morbidities that further exacerbate the risk of rUTIs. A critical co-morbidity is bowel dysfunction, specifically constipation (1). Rectal distension from constipation can mechanically compress the bladder, causing bladder instability and obstructing proper voiding. Additionally, vesicoureteral reflux (VUR) is a common co-morbidity in this population (6). The combination of high intravesical pressure caused by bladder dysfunction and the retrograde flow of urine in VUR creates a “perfect storm” for bacterial ascent and upper tract infection.

Over the long term, bladder dysfunction may result in damage to the upper urinary tract and kidneys, leading to conditions like hydronephrosis, hydroureter, VUR, and nephropathy (4,7). Notably, in the spina bifida patient population, approximately 25–50% exhibit detrusor sphincter dyssynergia, a condition that generates elevated pressure within the bladder, causing damage to the urinary tract and subsequent renal tissue. This condition significantly raises the risk of UTIs (8). Moreover, rUTIs, if left unmanaged, can ultimately lead to renal scarring and kidney failure (9). Despite the severity of these outcomes, clinicians frequently struggle to determine which patients will progress to recurrent infections and renal damage, leading to variation in prophylaxis and surveillance strategies.

Currently, there is a limited research data concerning the risk factors for UTIs in pediatric patients with bladder dysfunction. Existing research has primarily concentrated on specific subgroups, such as children with spina bifida and neurogenic bladder (10-12). Our study aimed to identify risk factors contributing to rUTIs in a broader cohort of pediatric patients with bladder dysfunction. We explored these factors in children with dysfunction arising from either neuropathic causes or structural anomalies of the lower urinary tract, excluding those with purely functional voiding disorders, to determine which clinical parameters are most predictive of recurrence. We present this article in accordance with the STROBE reporting checklist (available at https://pm.amegroups.com/article/view/10.21037/pm-25-127/rc).

Methods

This study was conducted as a retrospective cohort analysis. We examined the medical records of pediatric patients under the age of 18 years who visited either the pediatric urology or nephrology clinics at Siriraj Hospital between January 2010 and April 2020. We included patients diagnosed with bladder dysfunction secondary to two primary etiologies:

• Neuropathic bladder: International Classification of Diseases, 10th Revision (ICD-10): G95.8, N31.0-31.9.
• Structural anomalies of the lower urinary tract: ICD-10: Q64.1, Q64.2, Q64.3, Q64.5. Specific conditions in this group included PUV, prune belly syndrome, and bladder exstrophy.

Patients with bladder dysfunction attributed to functional causes (e.g., dysfunctional voiding in a structurally and neurologically normal urinary tract) were excluded. Additionally, patients with a follow-up period of less than 6 months were excluded to ensure a sufficient observation window for the detection of recurrent infections.

The comprehensive dataset we collected encompassed the following variables: age, gender, underlying diseases or anomalies, micturition or the method for bladder drainage, the history of UTIs, urine culture results, antibiotic prophylaxis, serum chemistry data, as well as findings from kidney-ureter-bladder (KUB) ultrasounds and voiding cystourethrography (VCUG). Renal function assessment relies on the estimated glomerular filtration rate, calculated using Bedside Schwartz Formula (2009 update): estimated glomerular filtration rate (eGFR) = (0.413 × height in cm)/serum creatinine (mg/dL). The constant (k) of 0.413 was selected as serum creatinine was quantified using the enzymatic method. We defined pyuria as the presence of 10 or more leukocytes per high-power field in a centrifuged urine specimen. Microscopic examination was performed at 400× magnification by trained laboratory technicians averaging counts over 10 fields, consistent with institutional protocols during the study period. The criteria for diagnosing a UTI included a positive urine culture combined with pyuria and the presence of one or more symptoms such as fever, loin pain, hematuria, or evolving urination symptoms (13). The presence of bacteria greater than 10³ and 10⁵ colony-forming unit (CFU)/mL for samples obtained by catheter and midstream clean catch, respectively, indicated positive urine culture. The occurrence of two or more UTIs within a 6-month period constituted the definition of rUTI (10,14).

Statistical analysis

The sample size was determined based on the requirements for multivariable logistic regression analysis. To ensure model stability and minimize overfitting, we applied the “events per variable” principle, aiming for a minimum of 10 outcome events (rUTIs) for each independent variable included in the final model. Anticipating the inclusion of approximately 5 to 7 potential risk factors in the multivariable analysis, and estimating an rUTI prevalence of roughly 40% in this high-risk population (10,12), a minimum total sample size of 120 patients was deemed necessary to achieve adequate statistical power. The statistical analysis was conducted using IBM SPSS Statistics 26. Categorical data are presented as frequencies and percentages. For data with symmetric distributions, we presented means and standard deviations and applied the sample t-test. For data with asymmetric distributions, we reported medians and interquartile ranges and performed the Mann-Whitney U test. When analyzing categorical data, we opted for either the Chi-square test or Fisher’s exact test as appropriate. The multivariable analysis to identify risk factors associated with rUTIs was carried out using a forward stepwise multiple logistic regression approach, incorporating likelihood ratio statistics. Variables with a P value <0.1 in the univariate analysis were considered for inclusion in the multivariable model. This method was selected to explore potential predictors in this heterogeneous population. Sensitivity analysis: given the relatively small number of rUTI events and the potential for sparse data bias, we performed a confirmatory sensitivity analysis using Firth’s penalized logistic regression. For final models and all other comparisons, statistical significance was defined as a P value <0.05 (two-tailed). Missing data for co-variates were minimal (<5%) and were handled using a complete-case analysis approach.

Ethics

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional review board of Siriraj Hospital (IRB approval No. Si359/2021) and individual consent for this retrospective analysis was waived.

Results

Of 173 cases identified (Figure 1), 53 cases were excluded due to no bladder dysfunction (39 cases), other non-neurogenic causes (2 cases), a follow-up period of less than 6 months (8 cases) and missing records (4 cases). A total of 120 patients with an average age of 7.7 years were included in the study. Among the participants, 66 were male (55%) and 54 were female (45%), corresponding to a male-to-female ratio of 1.2:1. The majority of comorbidities (65.8%) were attributed to spinal cord diseases, encompassing conditions such as spinal dysraphism, sacral agenesis, tethered cord, cord compression, and injury. Brain-related diseases accounted for 27.5% of cases, including conditions such as hydrocephalus, Chiari malformation, epilepsy, Down syndrome, microcephaly, and cerebral palsy. Moreover, 15% of the patients presented with global developmental delay (GDD), while 10% had anorectal malformations.

Figure 1 Study population selection process flowchart. UTI, urinary tract infection.

The predominant cause of bladder dysfunction was neurogenic bladder (78.3%), while lower urinary tract obstruction accounted for 21.7%. Lower urinary tract obstruction encompassed conditions such as PUV, anterior urethral valve, and urethral stricture. Regarding bladder management, approximately 41.7% (n=50) of the patients utilized clean intermittent catheterization (CIC). The majority of patients requiring CIC had underlying neurogenic bladder dysfunction (94.8%), while a smaller subset had severe myogenic failure secondary to valve obstruction. One-third (34.2%) could void spontaneously without catheterization. Indwelling catheters were required for 11.7% of patients, and 5.8% underwent cutaneous vesicostomy.

Moreover, 38.3% of the patients received antibiotic prophylaxis, with co-trimoxazole being the most frequently prescribed antibiotic. Nearly half of the patients (49.2%) received anticholinergic therapy, specifically oxybutynin, to manage intravesical pressure (Table 1). Urodynamic studies were performed in a limited subset of patients (n=18). Findings in this subgroup included detrusor overactivity (n=8), underactive detrusor contractility (n=7), and atonic bladder (n=5). Due to the small sample size and incomplete data for the majority of the cohort, statistical analysis associating specific urodynamic patterns with UTI recurrence was not performed.

From a total of 120 patients, 118 underwent KUB ultrasound examinations, and 102 patients underwent VCUG. Among them, 56.7% presented with hydronephrosis, with the majority being bilateral (37.2%). Furthermore, 60.8% exhibited abnormalities in the urinary bladder, including distortion, diverticula, and trabeculation. Among the 102 patients who underwent VCUG, 45% were diagnosed with vesicoureteral reflux, occurring both bilaterally and unilaterally in nearly equal proportions. Most cases demonstrated a high-grade severity level.

Regarding the incidence of UTIs, out of all patients observed over a 6-month period, 15% experienced a single UTI, while 12.5% (n=15) had rUTIs. Among those with rUTIs, 5% had two episodes, 3.3% had three episodes, and 4.2% had four episodes. Out of a total of 62 urine specimens collected, nearly half of them were infected with Escherichia coli (48.4%), followed by Klebsiella pneumoniae at 24.2%, and Pseudomonas aeruginosa at 14.5%. Other infectious agents included Candida albicans, Candida tropicalis, Acinetobacter baumannii, Proteus mirabilis, Stenotrophomonas maltophilia, and Serratia marcescens. Furthermore, it was observed that both Escherichia coli and Klebsiella pneumoniae were extended-spectrum beta-lactamase (ESBL) producers, accounting for 30.6% of all identified pathogens.

In the comparative analysis between the groups with and without rUTIs, statistically significant differences were observed (Table 2). The rUTI group was significantly younger, with a median age of 1.0 years (IQR, 0.5–2.2 years) compared to 8.4 years (IQR, 3.8–12.1 years) in the non-recurrent group (P<0.001). A higher incidence of GDD was observed in the recurrent group (33.3% vs. 12.4%, P=0.049). Additionally, patients with rUTIs were more likely to be on antibiotic prophylaxis (86.7% vs. 31.4%, P<0.001), reflecting the higher clinical concern in this subgroup. Furthermore, ultrasound KUB findings (Table 3) indicated a higher prevalence of hydronephrosis and hydroureter in the rUTI group. Specifically, bilateral hydronephrosis was present in 60% of the recurrent group compared to 35% in the non-recurrent group.

In the initial standard multivariable logistic regression analysis (Table 4), after adjusting for potential confounders, younger age and GDD appeared as independent predictors. However, to address the potential for sparse data bias given the limited number of outcome events (n=15), a rigorous sensitivity analysis was performed using Firth’s penalized logistic regression. In this final, penalized model, younger age remained a highly significant, robust independent predictor; the risk of rUTI decreased by approximately 29% for every one-year increase in patient age (adjusted OR 0.7, 95% CI: 0.6–0.9, P<0.001). Conversely, the association for GDD was attenuated. While GDD maintained a strong effect size, indicating a trend toward increased risk, it no longer reached statistical significance after penalization (adjusted OR 3.6, 95% CI: 0.9–14.3, P=0.06).

Discussion

Key findings

This study investigated risk factors for rUTIs in pediatric patients with bladder dysfunction arising from both neurogenic and structural etiologies. We found that 27.5% of patients developed symptomatic UTIs, and 12.5% had recurrent episodes within 6 months. Following rigorous statistical penalization to account for the small number of outcome events, younger age emerged as the only significant independent predictor of recurrence. While GDD demonstrated a strong clinical trend toward increased risk, it did not reach statistical significance after this adjustment. Crucially, our analysis revealed a distinct and concerning pattern of antimicrobial resistance: Nearly one-third of isolated pathogens were ESBL producers, particularly Escherichia coli and Klebsiella pneumoniae, with higher prevalence among those with rUTIs. Children with recurrent infections also had higher rates of antibiotic prophylaxis, which was associated with increased resistance.

Strengths and limitations

A major strength of this study is the inclusion of a diverse cohort combining both neurogenic bladder and structural lower urinary tract anomalies. By analyzing these groups together, we highlight that the downstream consequences of “bladder dysfunction”, specifically urinary stasis and infection risk, share a common clinical pathway regardless of the primary diagnosis. Detailed chart review minimized misclassification of outcomes and ensured accurate assessment of risk factors. However, the retrospective design is subject to information bias. The frequency of CIC and subtle urinary symptoms may have been underreported. Some patients may have received treatment at outside facilities, leading to incomplete capture of UTI episodes. Variability in ultrasound reporting limited assessment of hydronephrosis severity. Additionally, the relatively small sample size (n=120) and low number of recurrence events necessitated the use of penalized regression. This rigorous methodological approach appropriately attenuated the initial significance of GDD, highlighting the importance of adjusting for sparse data bias to avoid overstating risk factors in pediatric urology studies. Finally, urodynamic data was available for only a subset of patients, limiting our ability to correlate specific pressure-flow patterns with infection risk.

Comparison with similar research

Our findings are consistent with previous reports showing high UTI burden in children with bladder dysfunction. Chaudhry et al. found 20% of neurogenic bladder patients developed rUTIs, while Jiang et al. reported a rate of 24.2% (11,12). Similar to Chaudhry et al., our results demonstrated that younger age increased recurrence risk, with a 30% increase per year decrease in age. In contrast, Jiang et al. observed a slight increase in risk with advancing age, although their rUTI group was only marginally older (11). Studies in VUR patients have also shown particularly high recurrence rates in infancy, especially under 6 months (15).

While our univariate analysis suggested that GDD might also play a significant role—which is consistent with prior studies linking severe cognitive impairment to lower urinary tract dysfunction and UTI susceptibility (16-19)—our penalized regression revealed that this association requires larger cohorts to confirm as an independent predictor. In contrast, spinal cord disease—while a common cause of bladder dysfunction—was not an independent risk factor in our cohort, although prior research in spina bifida populations has demonstrated associations with VUR, trabeculated bladder, and high detrusor pressures (10,11,20,21).

Most significantly, our microbiological findings diverge from older cohorts. While Escherichia coli and Klebsiella pneumoniae remain dominant, the high prevalence of ESBL producers (32.8%) in our study exceeds rates reported in previous community-based pediatric series. This mirrors the rising trend of multidrug resistance in this pediatric population (22-24).

Explanations of findings

The elevated risk of rUTIs in younger children likely reflects immature host defenses, greater frequency of urinary stasis, and higher susceptibility to VUR during infancy (15,25-27). GDD may contribute through impaired bladder and bowel function, and abnormal voiding patterns (16-19). The observed clinical trend suggesting an association between GDD and rUTIs likely stems from a combination of impaired mobility, inability to communicate voiding urges, and reliance on caregivers for hygiene and catheterization, all of which predispose to bacterial introduction and stasis.

The lack of association between hydronephrosis and rUTIs in multivariable analysis aligns with prior studies in neurogenic bladder patients, suggesting that hydronephrosis is more a marker of bladder dysfunction severity than a direct predictor of recurrence (10-12,21).

The high prevalence of ESBL-producing organisms and increased resistance in patients on antibiotic prophylaxis highlights the unintended consequences of long-term antimicrobial use, consistent with prior studies demonstrating a two- to three-fold increased risk of resistant infections in spina bifida patients receiving prophylaxis (13,28,29). The alarming rate of ESBL production and its association with antibiotic prophylaxis is the most clinically pressing finding. This confirms that in the modern era, continuous antibiotic prophylaxis may effectively suppress sensitive flora while facilitating the emergence of difficult-to-treat resistant organisms.

Implications and actions needed

These findings carry immediate clinical implications. First, young patient age should be treated as a major “red flag” for recurrence. While GDD did not reach independent statistical significance in this model, clinicians should still maintain a high index of suspicion for these patients, who may benefit from aggressive bowel management and caregiver education rather than just medication. Second, the high rate of ESBL production necessitates a critical shift in antibiotic stewardship. The routine use of long-term prophylaxis in children with bladder dysfunction should be critically re-evaluated. Clinicians should consider moving toward “active surveillance” and treating symptomatic episodes with culture-specific therapy rather than relying on suppression that breeds resistance.

Future prospective, multicenter studies with larger sample sizes are warranted to definitively evaluate the independent clinical impact of co-morbidities like GDD and to trial non-antibiotic prevention strategies.

Conclusions

This study highlights that pediatric bladder dysfunction, whether neurogenic or structural, carries a shared risk profile for recurrent infection that is primarily driven by young patient age. However, the most critical insight is the substantial burden of antimicrobial resistance. The strong association between prophylaxis and resistant organisms serves as a clear warning against the indiscriminate use of suppressive antibiotics. Management strategies must evolve from “suppression” to “stewardship”, focusing on vigilant observation of young, high-risk patients and reserving antimicrobial therapy for active, symptomatic infections.

Acknowledgments

The authors wish to thank the dedicated medical record personnel at Siriraj Hospital for their assistance in retrieving the patient data used in this study. The authors also thank Ms. Saowalak Hunnangkul and Mr. Suthipol Udompunthurak for their support with data management and analysis. The authors acknowledge Ms. Apichaya Yoomeesuk for her assistance with manuscript formatting and support with the submission process.

The abstract has been accepted for presentation at the 15th Asian Congress of Pediatric Nephrology (ACPN2023) on November 23–25, 2023 (Dubai, United Arab Emirates).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://pm.amegroups.com/article/view/10.21037/pm-25-127/rc

Data Sharing Statement: Available at https://pm.amegroups.com/article/view/10.21037/pm-25-127/dss

Peer Review File: Available at https://pm.amegroups.com/article/view/10.21037/pm-25-127/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://pm.amegroups.com/article/view/10.21037/pm-25-127/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional review board of Siriraj Hospital (IRB approval No. Si359/2021) and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Brandström P, Hansson S. Urinary Tract Infection in Children. Pediatr Clin North Am 2022;69:1099-114. [Crossref] [PubMed]
2. Hum S, Liu H, Shaikh N. Risk Factors for the Development of Febrile Recurrences in Children with a History of Urinary Tract Infection. J Pediatr 2022;243:152-7. [Crossref] [PubMed]
3. Keren R, Shaikh N, Pohl H, et al. Risk Factors for Recurrent Urinary Tract Infection and Renal Scarring. Pediatrics 2015;136:e13-21. [Crossref] [PubMed]
4. Wiener JS, Huck N, Blais AS, et al. Challenges in pediatric urologic practice: a lifelong view. World J Urol 2021;39:981-91. [Crossref] [PubMed]
5. Vasudeva P, Madersbacher H. Factors implicated in pathogenesis of urinary tract infections in neurogenic bladders: some revered, few forgotten, others ignored. Neurourol Urodyn 2014;33:95-100. [Crossref] [PubMed]
6. Van Batavia JP, Ahn JJ, Fast AM, et al. Prevalence of urinary tract infection and vesicoureteral reflux in children with lower urinary tract dysfunction. J Urol 2013;190:1495-9. [Crossref] [PubMed]
7. Goetz LL, Klausner AP. Strategies for prevention of urinary tract infections in neurogenic bladder dysfunction. Phys Med Rehabil Clin N Am 2014;25:605-18. viii. [Crossref] [PubMed]
8. de Jong TP, Chrzan R, Klijn AJ, et al. Treatment of the neurogenic bladder in spina bifida. Pediatr Nephrol 2008;23:889-96. [Crossref] [PubMed]
9. Penna FJ, Elder JS. CKD and bladder problems in children. Adv Chronic Kidney Dis 2011;18:362-9. [Crossref] [PubMed]
10. Zhou G, Jiang M, Liu X, et al. Risk factors associated with recurrent febrile urinary tract infection in children with neurogenic bladder who perform clean intermittent catheterization. Neurourol Urodyn 2023;42:1485-90. [Crossref] [PubMed]
11. Jiang M, Deng J, Zhou G, et al. Risk Factors for Recurrent Urinary Tract Infection in Children With Neurogenic Bladder Following Clean Intermittent Catheterization. Urology 2022;164:224-9. [Crossref] [PubMed]
12. Chaudhry R, Balsara ZR, Madden-Fuentes RJ, et al. Risk Factors Associated With Recurrent Urinary Tract Infection in Neurogenic Bladders Managed by Clean Intermittent Catheterization. Urology 2017;102:213-8. [Crossref] [PubMed]
13. Pannek J, Wöllner J. Management of urinary tract infections in patients with neurogenic bladder: challenges and solutions. Res Rep Urol 2017;9:121-7. [Crossref] [PubMed]
14. Dray EV, Clemens JQ. Recurrent urinary tract infections in patients with incomplete bladder emptying: is there a role for intravesical therapy? Transl Androl Urol 2017;6:S163-70. [Crossref] [PubMed]
15. Dias CS, Silva JM, Diniz JS, et al. Risk factors for recurrent urinary tract infections in a cohort of patients with primary vesicoureteral reflux. Pediatr Infect Dis J 2010;29:139-44. [Crossref] [PubMed]
16. de Carvalho Mrad FC, de Bessa J Jr, de Rezende Duarte AM, et al. Prevalence of lower urinary tract symptoms in individuals with Down syndrome. J Pediatr Urol 2014;10:844-9. [Crossref] [PubMed]
17. Bhatt NR, Murchison L, Yardy G, et al. Bladder bowel dysfunction in children with Down's syndrome. Pediatr Surg Int 2020;36:763-72. [Crossref] [PubMed]
18. Silva JA, Alvares RA, Barboza AL, et al. Lower urinary tract dysfunction in children with cerebral palsy. Neurourol Urodyn 2009;28:959-63. [Crossref] [PubMed]
19. Gündoğdu G, Kömür M, Avlan D, et al. Relationship of bladder dysfunction with upper urinary tract deterioration in cerebral palsy. J Pediatr Urol 2013;9:659-64. [Crossref] [PubMed]
20. Timberlake MD, Jacobs MA, Kern AJ, et al. Streamlining risk stratification in infants and young children with spinal dysraphism: Vesicoureteral reflux and/or bladder trabeculations outperforms other urodynamic findings for predicting adverse outcomes. J Pediatr Urol 2018;14:319.e1-7.
21. Seki N, Masuda K, Kinukawa N, et al. Risk factors for febrile urinary tract infection in children with myelodysplasia treated by clean intermittent catheterization. Int J Urol 2004;11:973-7. [Crossref] [PubMed]
22. Ortiz TK, Velazquez N, Ding L, et al. Predominant bacteria and patterns of antibiotic susceptibility in urinary tract infection in children with spina bifida. J Pediatr Urol 2018;14:444.e1-8.
23. Forster CS, Shaikh N, Hoberman A, et al. Uropathogens and Pyuria in Children With Neurogenic Bladders. Pediatrics 2018;141:e20173006. [Crossref] [PubMed]
24. Shaikh N, Hoberman A, Keren R, et al. Predictors of Antimicrobial Resistance among Pathogens Causing Urinary Tract Infection in Children. J Pediatr 2016;171:116-21. [Crossref] [PubMed]
25. Tanaka Y, Oishi T, Ono S, et al. Epidemiology of urinary tract infections in children: Causative bacteria and antimicrobial therapy. Pediatr Int 2021;63:1198-204. [Crossref] [PubMed]
26. Boon HA, Struyf T, Crèvecoeur J, et al. Incidence rates and trends of childhood urinary tract infections and antibiotic prescribing: registry-based study in general practices (2000 to 2020). BMC Prim Care 2022;23:177. [Crossref] [PubMed]
27. Vachvanichsanong P, Dissaneewate P, McNeil E. Childhood recurrent urinary tract infection in southern Thailand. Ren Fail 2013;35:66-71. [Crossref] [PubMed]
28. McKibben MJ, Seed P, Ross SS, et al. Urinary Tract Infection and Neurogenic Bladder. Urol Clin North Am 2015;42:527-36. [Crossref] [PubMed]
29. Zegers SH, Dieleman J, van der Bruggen T, et al. The influence of antibiotic prophylaxis on bacterial resistance in urinary tract infections in children with spina bifida. BMC Infect Dis 2017;17:63. [Crossref] [PubMed]