DISCREPANCY IN VESICO-URETERIC REFLUX DIAGNOSIS BETWEEN VOIDING CYSTOURETHROGRAPHY AND VIDEO-URODYNAMICS
Matthieu Peycelon, MD, Konrad M. Szymanski, MD, MPH, M. Francesca Monn, MD, MPH, Mark P. Cain, MD, Richard C. Rink, MD, Rosalia Misseri, MD.
Riley Children Hospital; Indiana University, School of Medicine, Indianapolis, IN, USA.
Background. Voiding cystourethrography (VCUG) is the gold standard for the diagnosis of vesicoureteral reflux (VUR). However, this test does not account for bladder capacity and detrusor pressures. Videourodynamics (VUDS) has been advocated for the detection of physiological VUR. The aims were to compare the results of VCUG and VUDS in the detection and grading of VUR in a group of children having both exams in a one-year window and to assess if discrepancy between VCUG and VUDS affects management.
Methods. Patients who underwent a VCUG and VUDS within a 365-day period without medical or surgical intervention were retrospectively analyzed (2011-2018). We extracted data on demographics, indications, and VUR characteristics (high grade >=4, side). Analyses were based on patients and renal units (RU). Sensitivity and specificity were calculated with VCUG as the standard. Statistical analysis was performed with Fisher's, McNemar, and Stuart-Maxwell tests.
Results. Fifty-four children (28 girls) with 108 renal units (RU) were studied sequentially using VCUG and UDS. VCUG was performed at median age 2.1 years (IQR 2 weeks-6.1 years) and then VUDS after a median of 4.3 months (IQR 2.9-8.9).
Indications for both tests were voiding dysfunction in 38.9% children, neonatal protocol for spina bifida (SB) in 33.3% infants, and urinary tract infections in 27.8% children with SB.
VUR was detected in 51.8% of patients on VCUG, 35.2% on VUDS, and 31.5% on both. On VCUG, 18.5% had high grade VUR and 25.9% had bilateral. On VUDS, 14.8% had high grade VUR and 16.7% had bilateral. Six children (11.1%) had VUR on VCUG but not on VUDS. Two children(3.7%) had no VUR on VCUG but did on VUDS. Compared to VUDS, VCUG detected significantly more VUR overall (p=0.03) and individually captured more VUR that was high grade, low grade, bilateral, and unilateral (p<=0.04). Sensitivity of VCUG to detect any VUR was 60.7%(17/28) with a specificity of 92.3%(24/26).
Looking at RUs, VUR was detected in 38.8% on VCUG, 25.9% on VUDS, and 21.3% on both. On VCUG, 11.1% had high grade VUR and 25.9% had bilateral. On VUDS, 7.4% had high grade VUR and 18.5% had bilateral. Compared to VUDS, VCUG detected significantly more VUR overall (p=0.01), and specifically VUR that was high grade, low grade, bilateral, and unilateral (p<=0.04). Sensitivity of VCUG to detect any VUR in a RU was 54.8%(23/42) with a specificity of 92.4%(61/66).
VUDS and VCUG detect VUR differently from both a laterality and grading status (perfect correlation for bilateral, unilateral, high grade, low grade and no VUR was 43%, 36%, 66.7%, 46.7%, and 92% respectively).
Twenty patients (37.0%) did not have the same findings on VCUG and VUDS which implied a different clinical management.
Conclusions. VUDS is a poor screening tool to detect VUR in comparison to VCUG. However, VUDS confirms that there is no VUR seen on VCUG. The results of these methods in a group of selected patients showed a significant difference in the detection of any VUR, and presence of high grade and bilateral VUR which may impact final decision making.
Back to 2019 Abstracts