What are the predictive factors leading to ureteral obstruction following endoscopic correction of VUR in the pediatric population?

Background It is extremely important to not only address the short-term success following endoscopic correction of vesicoureteral reflux (VUR) but also the long-term efficacy and safety of the tissue augmenting substance utilized for endoscopic correction. Objective This study retrospectively evaluated all cases of ureterovesical junction (UVJ) obstruction following endoscopic treatment of VUR over the last 5 years utilizing two tissue augmenting substances, with special emphasis on the safety of Vantris (R), and performed clinical and histological review of these patients. Methods The study population comprised 2495 patients who underwent endoscopic correction of VUR utilizing Deflux (R) (1790) and Vantris (R) (705). Tissue sections were stained with hematoxylin & eosin and trichrome, and examined under a light microscope. Nine primary obstructive megaureters after ureteral reimplantation served as controls. Results Nine (0.5%) children (three female and six male) in the Deflux group and nine (1.3%) (five females and four males) in the Vantris group developed UVJ obstruction and required ureteral re-implantation. Obstruction developed during the period ranging 2-49 months (average 16 months) following endoscopic correction. The primary reflux grade was III in seven, IV in six, and V in six children. The mean volume of the injected material in all obstructed patients was 1.2 +/- 0.6 cc (mean +/- SD). Histopathological analysis revealed a pseudocapsule composed of fibrous tissue and foreign-body giant cells surrounding the Vantris implant in all patients. The distal part of the ureters demonstrated significant ureteral dilatation without ureteral fibrosis. In all patients, additional biopsies from the muscularis propria adjacent to the injection site were examined and showed no significant abnormalities. There was an increased collagen deposition in the juxtavesical segment of the obstructive ureters following Deflux and Vantris injections, and of primary obstructive megaureter. No significant difference was found in the tissue response between Deflux and Vantris patients and controls. Statistical analysis of the nonhomogeneous population demonstrated higher obstruction rates in patients from the Vantris group. However, no statistical difference was demonstrated regarding the obstruction rate in the homogenous group with relation to gender, age and reflux grade group of patients. Moreover, univariate analysis revealed that Grade V reflux, the presence of beak sign on the reviewed pretreatment, and inflamed bladder mucosa upon injection were significant independent risk factors leading to obstruction. Discussion This study suggested that the underlining ureteral pathology lead to UVJ obstruction following Vantris injection. There was increased collagen deposition in the juxtavesical segment of the obstructive ureters following Vantris injection. Furthermore, these findings were similar to those discovered in patients who underwent endoscopic correction with Deflux, and in patients who required ureteral reimplantation due to primary obstructive megaureter. Additional biopsies from the muscularis propria adjacent to the injection site showed no significant abnormalities, ironing out the fact that Vantris did not led to adverse tissue reaction following injection. Univariate analysis further ironed out the hypothesis that underlying ureteral pathology was responsible for the increased incidence of UVJ obstruction and demonstrated that Grade V reflux, the presence of beak sign on the reviewed pretreatment VCUG, and inflamed bladder mucosa upon injection were significant independent risk factors leading to obstruction. Conclusion Data showed that Vantris injection did not lead to any different ureteral fibrosis or inflammatory changes to the tissue augmenting substances utilized in past and present clinical practice, and therefore did not seem to increase the incidence of UVJ obstruction. High reflux grade, presence of obstructive/refluxing megaureter and inflamed bladder mucosa were the only statistically significant and independent predictive factors for UVJ obstruction following endoscopic correction of VUR.