Association of Renal Scarring With Number of Febrile Urinary Tract Infections in Children

Key Points

Question  Is there an association between the number of febrile urinary tract infections and the risk of renal scarring in children?

Findings  This post hoc analysis of data from 2 studies found that the odds of renal scarring after a second febrile infection were approximately 12 times greater than after a single febrile infection.

Meaning  Identification of biomarkers that could noninvasively identify children at risk for subsequent febrile infections at the time of the first urinary tract infection may help reduce morbidity.

Importance  Although prior studies have suggested that the risk of renal scarring gradually increases with each febrile urinary tract infection (UTI), recent and detailed data are lacking.

Objective  To evaluate how the risk of renal scarring is associated with the number of febrile UTIs.

Design, Setting, and Participants  A post hoc analysis was performed from June 2018 to April 2019 of data collected in the context of 2 multicenter prospective studies (RIVUR [Randomized Intervention for Children With Vesicoureteral Reflux], conducted from June 2007 to June 2013, and CUTIE [Careful Urinary Tract Infection Evaluation], conducted from May 2008 to October 2013), of children with a first UTI without baseline renal abnormalities who were followed up for 2 years for febrile recurrences.

Exposure  Number of known febrile UTIs.

Main Outcomes and Measures  Renal scarring was defined as decreased uptake of tracer associated with the loss of contours or cortical thinning on a technetium 99m dimercaptosuccinic acid renal scan obtained at study exit or approximately 4 months after the last febrile UTI.

Results  A total of 345 children were included (307 girls and 38 boys; mean [SD] age, 19.4 [18.2] months; 221 with vesicoureteral reflux and 124 without vesicoureteral reflux). The incidence of renal scarring was 2.8% (95% CI, 1.2%-5.8%) after 1 febrile UTI, 25.7% (95% CI, 12.5%-43.3%) after 2 febrile UTIs, and 28.6% (95% CI, 8.4%-58.1%) after 3 or more febrile UTIs. The odds of renal scarring after a second febrile infection were 11.8 (95% CI, 4.1-34.4) times greater than after a single febrile infection, and the odds of renal scarring after 3 or more febrile infections were 13.7 (95% CI, 3.4-54.4) times greater than after a single febrile infection.

Conclusions and Relevance  Although relatively few children have 2 febrile UTIs, those who do have a substantially higher risk of renal scarring compared with children with a single febrile UTI. This finding suggests that research should focus on the identification of biomarkers that could noninvasively identify children at risk for subsequent febrile infections. More research is also needed to understand the molecular basis of the increased risk of renal scarring among children with recurrent febrile UTIs.

Renal scarring is an important sequela of febrile urinary tract infections (UTIs) in childhood. Understanding the association between the number of episodes of febrile UTI and the risk of renal scarring can help develop evidence-based management strategies for children with this frequently occurring problem.

We are not aware of recent data on the risk of renal scarring in children according to the number of episodes of febrile UTI; to our knowledge, the last study with such data was conducted before the advent of the technetium 99m dimercaptosuccinic acid (DMSA) renal scan.¹ The findings of that study suggested that the risk of renal scarring increases steadily with each febrile UTI (ie, after accounting for the proportion of children with preexisting lesions, the risk of renal scarring in children with 2 febrile UTIs was approximately twice the risk of those with 1 febrile UTI).¹ The objective of our observational study was to examine the association between the number of febrile UTIs and the risk of acquiring new renal scars.

In this investigation, we performed a post hoc analysis from April 10, 2018, to April 4, 2019, of data from 2 multicenter studies (RIVUR [Randomized Intervention for Children With Vesicoureteral Reflux], conducted from June 2007 to June 2013, and CUTIE [Careful Urinary Tract Infection Evaluation], conducted from May 2008 to October 2013)^2,3 in which children with UTIs without known genitourinary abnormalities were enrolled and prospectively followed up for approximately 2 years. At the end of both studies, a DMSA scan was obtained to determine the presence and severity of renal scarring. Both studies were approved by the institutional review boards of the participating institutions, and written informed consent was obtained prior to any study procedures. The University of Pittsburgh institutional review board reviewed this post hoc analysis and deemed it as exempt because data were deidentified.

The methods of these studies have been previously reported. In brief, in the RIVUR study, children 2 to 71 months of age presenting with grades I to IV vesicoureteral reflux (VUR) diagnosed after a UTI (index UTI) were enrolled at 19 clinical centers throughout North America.³ Children in the RIVUR trial were randomized to receive antimicrobial prophylaxis or placebo. Children without VUR from 3 of the 19 RIVUR sites were enrolled in the parallel CUTIE study.² In both studies, a DMSA scan was performed within 16 weeks of the index UTI to identify children with preexisting renal dysplasia or scarring.

In both studies, episodes of febrile UTI were carefully documented during the 2-year follow-up period.^2,3 We considered the number of febrile UTIs from the time of the index UTI (including the index UTI) until the time of the outcome DMSA scan. A febrile UTI was defined by the presence of a documented temperature of 38°C or more within 24 hours of diagnosis of UTI, pyuria on urinalysis, and 50 000 colony-forming units (CFU)/mL or more of a single organism on culture from a specimen obtained via catheterization or 100 000 CFU/mL or more from a clean voided sample. For this analysis, we excluded children with a history of protocol-defined UTI prior to the index episode (including 1 child with 10 000-50 000 CFU/mL on culture whose early DMSA scan was consistent with moderate pyelonephritis), children with preexisting renal dysplasia (defined by an early DMSA scan with altered renal volume and contours), children without a baseline DMSA scan (in whom dysplasia could therefore not be assessed), and children who were lost to follow-up before a late DMSA scan could be obtained. Thus, all children included were identified at the time of their first UTI, had no evidence of renal dysplasia, and had a follow-up DMSA scan to evaluate renal scarring. Because identification and treatment of VUR no longer represents the standard of care for children with a first UTI, and because the use of prophylaxis substantially reduces the number of subsequent reinfections, we also excluded children in the RIVUR trial who received antimicrobial prophylaxis.

Renal scarring was defined as a decreased uptake of tracer associated with the loss of contours or cortical thinning on a DMSA scan obtained at study exit or approximately 4 months after the last febrile UTI. For children with more than 1 late DMSA scan, we used the last scan obtained. We calculated exact confidence intervals⁴ and used the Cochran-Mantel-Haenszel statistic to assess whether the incidence of renal scarring associated with the number of febrile UTIs differed between children with and children without VUR. P < .05 was considered to be statistically significant and all statistical tests were 2-sided.

A total of 345 children (221 with VUR and 124 without VUR) were included (Figure 1). A total of 50 children had no febrile UTIs (ie, children who had an index UTI that was not accompanied by fever and had no subsequent febrile UTIs), 246 children had 1 febrile UTI, 35 children had 2 febrile UTIs, 13 children had 3 febrile UTIs, and 1 child had 4 febrile UTIs. Because the incidence of renal scarring was similar among children from the RIVUR and CUTIE studies, we combined them for the remaining analyses. The incidence of renal scarring was 0% (0%-7.1%) in those with no febrile UTIs, 2.8% (95% CI, 1.2%-5.8%) after 1 febrile UTI, 25.7% (95% CI, 12.5%-43.3%) after 2 febrile UTIs, and 28.6% (95% CI, 8.4%-58.1%) after 3 or more febrile UTIs (Figure 2). The odds of renal scarring in children with 2 febrile UTIs were 11.8 (95% CI, 4.1-34.4) times greater than in children with 1 febrile UTI, and the odds of renal scarring in children with 3 or more febrile UTIs were 13.7 (95% CI, 3.4-54.4) times greater than in children with 1 febrile UTI.

Our data suggest that the risk of renal scarring increases substantially with a second febrile UTI. In addition, our data suggest that, if prevention of renal scarring is a goal, identification of strategies that could identify children at low or high risk after the first febrile UTI would be important.

Historically, the preferred strategy to accomplish this task has been to obtain a voiding cystourethrogram after a child’s first UTI and to treat those children with VUR with long-term antimicrobial prophylaxis. Despite the results of the RIVUR study (which showed a large decrease in the number of febrile reinfections in those treated with antimicrobial prophylaxis),³ this approach has fallen out of favor, appropriately in our view, because (1) children without VUR can have febrile reinfections, albeit at a lower rate than children with VUR, and, as a result, they may develop scarred kidneys; (2) most children with VUR never have a febrile reinfection and thus do not develop renal scarring; and (3) long-term research on the magnitude of risk of permanent sequelae attributable to renal scarring has been limited and conflicting. Our data confirm that only a minority of children with a febrile UTI have a second febrile UTI, lending support to the recommendations proposed in the most recent Guidelines from the American Academy of Pediatrics against performing invasive screening tests (ie, obtaining a voiding cystourethrogram) after a first febrile UTI.⁵

A second strategy to stratify children into low-risk and high-risk categories for recurrent UTIs and subsequent renal scarring is to use noninvasive means. Some investigators have attempted to use data that are routinely collected during clinical care to determine the risk of recurrent UTIs. A review of the literature suggests that bladder and bowel dysfunction is the strongest independent factor associated with recurrence.^2,6-10 A previous study suggested that screening for bladder and bowel dysfunction should be routine,⁸ and the new data presented here further support such screening. Nevertheless, bladder and bowel dysfunction explains only a small amount of the total variability in reinfection rates and only in the subgroup of children who are toilet trained. Thus, we are enthusiastic about studies examining novel urinary or serum biomarkers. Given the increasing availability of unbiased high-throughput omics approaches, we are optimistic that such marker(s) will be identified in the near future.

The reasons for the dramatic increase in the rate of renal scarring in children with recurrent febrile UTIs are unclear. Recent animal studies suggest that naive mice exposed to uropathogenic Escherichia coli have increased long-term susceptibility to reinfection with uropathogens, possibly mediated by remodeling of their bladder urothelial cells and changes in molecular response to inflammation.¹¹ Perhaps remodeling of the bladder and/or kidney after a first febrile UTI also contributes to poorer outcomes among children. Alternatively, the immune response may differ in those with a history of previous febrile UTIs. Further research in this area is needed to help elucidate the underlying pathophysiologic characteristics and inform new diagnostic and therapeutic measures.

To our knowledge, this is the largest report on the incidence of renal scarring in children with no febrile UTIs, none of whom developed renal scarring. These data support the notion that children with afebrile UTIs are at very low risk of developing renal scarring, and thus their UTIs could be managed less aggressively.

Our study had several limitations. Although the focus of this article is not the absolute risk of renal scarring, the rates that we observed were lower than in previous studies.^12,13 The reasons underlying this difference could include the different clinical characteristics of the children enrolled or the differences in the evaluation of renal scarring (eg, requiring alterations in renal contour and not just photopenia on DMSA scans, requiring adjudication of 2 readers, and timing of the DMSA scan). However, this difference does not detract from the main objective of this analysis (ie, to evaluate the rates of renal scarring associated with the cumulative number of febrile UTIs).

Although the proportion of children with UTIs who have febrile recurrences is relatively small, the risk of renal scarring for those who do is substantially higher. This finding suggests that research should focus on the identification of biomarkers that could noninvasively identify children at risk for subsequent febrile infections. More research is also needed to understand the molecular basis of the increased risk of renal scarring in children with recurrent febrile UTIs.

Accepted for Publication: April 24, 2019.

Corresponding Author: Nader Shaikh, MD, MPH, Division of General Academic Pediatrics, Children’s Hospital of Pittsburgh, University of Pittsburgh Medical Center, 4401 Penn Ave, Pittsburgh, PA 15224 (nader.shaikh@chp.edu).

Published Online: August 5, 2019. doi:10.1001/jamapediatrics.2019.2504

Author Contributions: Dr Shaikh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Shaikh, Hoberman.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Shaikh, Kurs-Lasky, Hoberman.

Critical revision of the manuscript for important intellectual content: Shaikh, Haralam, Hoberman.

Statistical analysis: Shaikh, Kurs-Lasky, Hoberman.

Obtained funding: Hoberman.

Administrative, technical, or material support: Shaikh, Haralam, Hoberman.

Supervision: Haralam, Hoberman.

Conflict of Interest Disclosures: Drs Shaikh and Hoberman reported receiving grants from the National Institute of Diabetes and Digestive and Kidney Diseases during the conduct of the study. No other disclosures were reported.

Funding/Support: The data collected for this analysis were supported by grants U01 DK074059, U01 DK074053, U01 DK074082, U01 DK074064, U01 DK074062, and U01 DK074063 from the National Institute of Diabetes and Digestive and Kidney Diseases.

Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.