Back to 2016 Fall Congress

BACKGROUND:
The VURx, a 6-point weighted scoring system based on gender, timing of VUR on voiding cystourethrogram (VCUG), ureteral abnormalities, and VUR grade, is a validated scoring system designed to predict spontaneous reflux resolution or at least 2 grade improvement based on the initial VCUG in children diagnosed before two years of age. We aimed to assess whether the VURx predicts VUR resolution/improvement in children diagnosed after 2 years of age.
.
METHODS:
Electronic medical records from two institutions were queried for patients with primary VUR diagnosed at >24 months of age, with at least one additional VCUG prior to 18 years of age. Exclusion criteria included secondary VUR and interval between first and last VCUG >2 standard deviations from the mean of the cohort. The VURx scoring is: female - 1 pont; VUR timing: early/mid -3, late- 2; voiding only- 1; ureteral abnormalities(diverticulum, complete duplication)- 1; high grade VUR(IV/V)- 1. Resolution was defined as no VUR and improvement was ≥2 grade decrease in maximum VUR grade. Parametric survival models using the HAZARD procedure in SAS (v 9.3, Cary, NC) were used to identify variables associated with resolution/ improvement in the early (24 months) and later periods (>24 months) following diagnosis.
RESULTS:
271 patients (21M, 250F) met inclusion criteria with mean age at diagnosis of 4.0 ± 2.1 years and median VUR grade of 2. VUR timing was early/mid filling in 92(34.0%), late filling in 145(53.5%), and voiding only in 34(12.6%). Resolution/improvement by VURx score: 1- 1(100%); 2- 25(67.6%); 3- 48(37%); 4-18(21.4%); 5/6- 4(18.2%). Female gender (HR 0.22[95% CI 0.1-0.5], p< 0.001) and timing of VUR (late: HR 0.19[0.07-0.57], p=0.003; early/mid: 0.40[0.18-0.87], p=0.022) were significant predictors of non-resolution. Time to resolution (months) based on the VURx score was: ≤2: 15.6 months (95%CI 11.0-13.8); 3: 34.7 (25.4-44.1); 4; 55.9 (40.1-inf); ≥5: 30.3 (29.5-inf). High grade (IV/V) VUR was not predictive of resolution/improvement in early [HR 0.53(0.07-4.10)] or late [HR 0.94 (0.27 - 3.31)] periods. Ureteral abnormalities were associated with persistent VUR as no patient with them had early resolution/improvement (HR=0), and this was not associated with later resolution [HR 0.91 (0.30-2.75)]. BBD at diagnosis was not a significant predictor of resolution/improvement in both early [HR 0.94 (0.38-2.33)] and later [HR 0.71 (0.35-1.42)] time periods. A VURx score of 3 (HR 0.28[0.13-0.59], p=0.001) and 4 (HR 0.19[0.08-0.49], p< 0.001) predicted lack of resolution/improvement in early resolvers; however, the VURx and its components were not predictive of resolution/improvement in late resolvers (p>0.05). A parametric survival model of resolution/improvement by VURx score is illustrated in the Figure.
.
CONCLUSIONS:
VURx reliably predicts primary VUR resolution or >2 grade improvement in children diagnosed at ≥2 years of age. VUR resolution/improvement is significantly less likely to occur 24 months after diagnosis in these children. Further study is warranted to identify additional predictive components for the VURx in this population.