Objective Bovine collagen has been successfully used for the endoscopic treatment of urinary incontinence (UI) and vesico-ureteric reflux (VUR) in children for more than 8 years, although its long-term efficacy has sometimes: been questioned. Its failure is generally ascribed either to technical difficulties or to instability of collagen at the site of the implant. To assess the efficacy of the procedure we evaluated the children treated at this hospital between 1990 and 1993. Patients and methods A total of 156 children, aged 5 months to 13 years, were treated for VUR. Over the same period, 25 patients aged between 2 and 14 years (nine with neuropathic bladder and 16 with exstrophy-epispadias complex who were still incontinent after bladder neck reconstruction) had periurethral or pericervical glutaraldehyde cross-linked botine collagen injection for UI. Five more children with exstrophy-epispadias complex (aged 1-3 years) underwent periurethral collagen injection to stimulate bladder enhancement and allow subsequent bladder neck reconstruction. Results In children treated for VUR, a single injection proved successful in 72.2% of cases (127 ureters); a second collagen injection raised the success rate to 81%. Continence improved in all nine neuropathic bladders and in 10 of 16 children with exstrophy-epispadias complex treated for UI after bladder neck reconstruction. In four of the five exstrophy-epispadias complex patients who were treated to stimulate bladder enhancement, bladder capacity increased by 25%. Conclusions Endoscopic treatment of VUR seems to be a valid alternative to open surgery, even though concerns remain about the long-term efficacy of collagen implantation. An important distinction should be made between early and late failure of the procedure. Early failure, which we define as persistence of reflux, is usually due to incorrect technique or technical difficulties. Late failure, or recurrence of reflux, which has previously been attributed to the biodegradability of collagen, seems to be due to the displacement of the injected collagen. Micturition itself or high bladder pressure (such as detrusor instability) could be responsible for the displacement of the injected collagen medially and distally, where it can no longer support the submucosal ureteric tunnel. In the treatment of urinary incontinence, both the implant technique and the choice of the site of injection seem to have a considerable effect on the results. In our experience, endoscopic collagen injection is effective in the treatment of both urinary incontinence and VUR in paediatric patients. Accurate selection of patients and technical adjustments and refinements are essential to obtain the best results.
