Abstract Book - 26th ESPU Congress, Joint Meeting with ICCS + SPU + AAP/SOU + AAPU + SFU

103 / 492

S5-8 (P)
RELIABILITY OF VESICOURETERAL REFLUX GRADING AND OTHER FINDINGS ON
VOIDING CYSTOURETHROGRAPHY
Anthony SCHAEFFER1, Saul GREENFIELD2, Alejandro HOBERMAN3, Ron KEREN 4, Ranjiv MATHEWS5, Tej MATTOO6, Gang
CUI7, Michael ZERIN8, Jeanne CHOW 9, Myra CARPENTER7, Anastasia IVANOVA7, Russell CHESNEY 10, Marva MOXEY-
MIMS11 and Caleb NELSON1
1) Boston Children's Hospital, Urology, Boston, USA - 2) Women & Children's Hospital of Buffalo, Urology, Buffalo, USA -
3) University of Pittsburgh, Pediatrics, Pittsburgh, USA - 4) Children's Hospital of Philadelphia, Pediatrics, Philadelphia,
USA - 5) Johns Hopkins Hospital, Urology, Baltimore, USA - 6) Children's Hospital of Michigan, Pediatrics, Detroit, USA -
7) University of North Carolina, Collaborative Studies Coordinating Center, Dept of Biostatistics, Chapel Hill, USA - 8)
Children's Hospital of Michigan, Radiology, Detroit, USA - 9) Boston Children's Hospital, Radiology, Boston, USA - 10)
University of Tennessee Health Sciences Center, Pediatrics, Memphis, USA - 11) National Institutes of Health, National
Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, USA
PURPOSE
Although vesicoureteral reflux (VUR) grading is essential for prognosis and clinical decision-making, the inter-rater
reliability of VUR grading is poorly understood. We sought to determine the inter-observer consistency of VCUG findings
in a large cohort of children with VUR.
MATERIAL AND METHODS
The Randomized Intervention for Children with Vesico-Ureteral Reflux (RIVUR) trial is a randomized controlled trial of
antimicrobial prophylaxis among children with VUR and urinary tract infection. Each enrollment VCUG was read by a
local clinical radiologist, and independently by two blinded RIVUR reference radiologists (reference radiologist
disagreements were adjudicated for RIVUR trial purposes). We compared the three radiology interpretations for VUR
grade and other VCUG findings.
RESULTS
Among the 607 subjects (558 females; median age at VCUG: 11 months) included in the RIVUR trial, 602 non-reference
radiology reports from 91 sites were reviewed and yielded VUR grade information for 524 right and 560 left ureters. All
three radiologist interpretations agreed on VUR grade in only 59% of ureters; two out of three agreed on 39% of
ureters; and all three disagreed on 2% of ureters. Agreement was better (≥92%) for other VCUG findings (e.g. bladder
shape "normal"). The clinical radiologist's VUR grade differed from the adjudicated RIVUR grade by exactly one grade
level in 19% of ureters, and by 2 or more grade levels in 2.6% of ureters. All 3 radiologists agreed on the VUR grade of
both ureters in just 43% of patients.
CONCLUSIONS
There is significant inter-rater variability for grading VUR. This has both research and clinical implications, as study
design, risk stratification, and clinical decision-making are heavily reliant on VUR grades.