Association of Urinary Tract Infection Guidelines With Vesicoureteral Reflux Screening

In 2011, the American Academy of Pediatrics (AAP) issued guidelines recommending against using routine voiding cystourethrograms in children aged 2 to 24 months who present with febrile urinary tract infection and instead recommending obtaining a renal ultrasonographic image and reserving voiding cystourethrographic testing for specific circumstances.¹ We sought to examine the outcomes associated with these guidelines on (1) diagnosis patterns among children with underlying vesicoureteral reflux and (2) detection rates of vesicoureteral reflux per screening voiding cystourethrogram performed for all children. We hypothesized an improvement in the detection rates of vesicoureteral reflux at the expense of delayed diagnoses among children with vesicoureteral reflux.

Using Cliniformatics Data Mart (Optuminsight), a deidentified national claims database of 75 million individuals, we identified children younger than 2 years who were diagnosed with vesicoureteral reflux after initial voiding cystourethrogram from January 2007 to December 2016. All outpatient, emergency department (ED), and inpatient encounters with principal diagnosis codes of urinary tract infection or pyelonephritis (International Statistical Classification of Diseases, Ninth Revision, Clinical Modification codes 599.0, 590.11, 590.80, and 590.81 and Tenth Revision, Clinical Modification codes N39.0, N10, and N11.0) were recorded. Children with diagnosis codes for which voiding cystourethrographic testing is routinely performed despite the absence of urinary tract infection (eg, myelomeningocele, ureteropelvic junction obstruction, posterior urethral valves) were excluded. This study was exempted from review by the Michigan Medicine institutional research board, which granted a consent waiver because of the use of a deidentified administrative database.

The following outcomes of interest were included: (1) undiagnosed days, defined as the period between the initial infection-associated encounter and vesicoureteral reflux diagnosis date, (2) multiple ED encounters within undiagnosed days (selected since outpatient encounters frequently occur without active infection, and ED encounters typically precede inpatient encounters), and (3) detection rate of vesicoureteral reflux per voiding cystourethrogram ordered per quarter of the year. Single-group, interrupted time-series analyses assessed for differences in trends of undiagnosed days and detection rate of vesicoureteral reflux before and after the 2011 AAP guidelines were published in September 2011.² Pearson χ² tests were used to assess the association between multiple infection-associated ED encounters before and after the 2011 AAP guidelines were released. Statistical analyses were performed from May 2018 to August 2018 using statistical software (Stata 15/SE [StataCorp]), with 2-sided P values less than .05 considered statistically significant.

Among 2997 children with new vesicoureteral reflux diagnoses, 2077 children and 920 children were diagnosed during the pre-AAP and post-AAP periods, respectively. There was a significant increase in the trend of undiagnosed days between the pre-AAP and post-AAP periods (change in slope, 1.4% [95% CI, 0.33%-2.54%]; P = .01). The interrupted time-series analysis is shown in Figure 1. The mean (SD) number of undiagnosed days increased from 49 (8) days just before the release of the 2011 AAP guidelines to 186 (30) days at the end of 2016 (P < .001). Children diagnosed with vesicoureteral reflux after the 2011 AAP guidelines were more likely to have had multiple infection-associated ED encounters prior to vesicoureteral reflux diagnosis (post-AAP: 63 of 920 patients [6.8%]; pre-AAP: 62 of 2077 patients [3.0%]odds ratio, 2.35 [95% CI, 1.08-5.18]; P = .02).

There was no change in the trend of detection rates of vesicoureteral reflux per screening voiding cystourethrogram between the pre-AAP and post-AAP periods. The interrupted time-series analysis is shown in Figure 2. The overall detection rate of vesicoureteral reflux was 24%.

Voiding cystourethrogram use has significantly decreased since the 2011 AAP guidelines were published.³ This helped avoid a substantial number of screening studies in healthy children, reducing unnecessary morbidity and costs associated with routine vesicoureteral reflux screening. Despite these positive outcomes, we identified a subset of children with underlying vesicoureteral reflux who experienced a delay in diagnosis and increased risk of repeated infections prior to diagnosis. We also demonstrate that detection rates of vesicoureteral reflux diagnosis per screening voiding cystourethrogram have been stable since the 2011 AAP guidelines were published. This suggests that, rather than the implementation of a more selective approach, screening studies are simply being performed less frequently.

This study has limitations. As with other population-based studies using administrative claims data, the diagnosis of an active infection is inferred using diagnosis codes rather than confirmatory diagnostic studies. In addition, there will be a subset of children with active infection who are diagnosed and treated on an outpatient basis.

Collectively, these findings highlight the need to develop methods that will improve identification of children presenting with febrile urinary tract infection who are at high risk of clinically significant vesicoureteral reflux.

Accepted for Publication: May 3, 2019.

Corresponding Author: Ted Lee, MD, Department of Urology, Michigan Medicine, 1500 E Medical Center Dr, Ann Arbor, MI 48109 (lee.ted.md@gmail.com).

Published Online: September 30, 2019. doi:10.1001/jamapediatrics.2019.3168

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

Concept and design: Basu, Park, Wan.

Acquisition, analysis, or interpretation of data: Triner, Basu, Park.

Drafting of the manuscript: Triner, Wan.

Critical revision of the manuscript for important intellectual content: Basu, Park, Wan.

Statistical analysis: Triner, Basu.

Obtained funding: Wan.

Administrative, technical, or material support: Basu, Park, Wan.

Supervision: Park, Wan.

Conflict of Interest Disclosures: Dr Wan is the pediatric urology editor for the Journal of Urology, for which he receives an honorarium annually; he also served on the safety panel of a trial run by Allergan Inc. No other disclosures were reported.