Abstract
Purpose:
To determine the applicability and long-term outcome of endoscopic injection of dextranomer/hyaluronic acid (Dx/HA) to correct incontinence in patients who had previously undergone continent urinary reconstruction.
Patients and Methods:
A retrospective cohort study was performed of all patients who underwent Dx/HA injection at our institution from January 2001 to June 2011. All patients had adequate bladder capacity and compliance on maximized medical therapy before injection. “Success” was defined as either “continence” (daytime dry interval >3 hours) or “improvement” (daytime dry interval >2 hours).
Results:
A total of 22 children (16 females and 6 males) had Dx/HA injections for persistent incontinence from either bladder neck (7), Mitrofanoff (10), or both (5). Median age at injection was 13 years (range 4–21). Children underwent an average of 1.6 injection sessions per patient with an average of 2.6 mL of Dx/HA per session. At a median follow-up of 72 months (range 4–104), 19 (86.4%) patients had successful results (16 continent, 3 improved). For those incontinent from bladder neck, 42% became continent after 1, 75% after 2, and 83% after 3 injections, with a success rate of 91% (10 continent, 1 improved). For those incontinent from Mitrofanoff, 20% became continent after 1, and 73% after 2 injections, with a success rate of 86% (11 continent, 2 improved).
Conclusions:
Endoscopic injection of Dx/HA to correct residual incontinence in selected children after urinary reconstruction appears to be safe and effective, achieving a dry interval in more than two-thirds of patients.
Introduction
I
The treatment options for persistent urinary incontinence secondary to poor outlet resistance include an open reoperative reconstruction such as a revision of the continence mechanism, inserting an artificial urinary sphincter, or as a last resort, closure of the bladder neck. 1,3,4 Those procedures, however, are associated with several risks and inconveniences such as a prolonged inpatient hospitalization, long operative times, and significant morbidity. A minimally invasive therapy with a high success rate and low morbidity would be an ideal approach to manage incontinence in this patient population.
Endoscopic injection of the continence mechanism with a bulking agent is a minimally invasive treatment, associated with a lower morbidity. Injection products that have been descibed in the literature include bovine collagen, Macroplastique®, Teflon®, and dextranomer/hyaluronic acid (Dx/HA) copolymer (Deflux®). 5
We present our experience using endoscopic injection of Dx/HA to correct persistent leakage in patients who had previously undergone continent urinary reconstruction.
Patients and Methods
Patient population
Medical records of all children and young adults who underwent Dx/HA injection surgery at a single pediatric hospital from January 2001 to June 2011 were reviewed retrospectively. Outcome data were collected from hospital records after obtaining Institutional Review Board approval. Data points collected included patient demographics, clinical symptoms, radiographic imaging, indications for treatment, intraoperative details, postoperative continence, and complications. Once the surgical treatment was indicated, the choice of an open procedure or endoscopic injection was offered to the patient and family, and the surgical approach was thus determined based on a shared decision making. All patients had adequate bladder capacity and compliance—based on formal urodynamic studies—after being put on maximized medical therapy (anticholinergics) before injection.
Endoscopic injection procedure
Endoscopy was performed as outpatient surgery. After adequate general anesthesia, the patient was placed in the lithotomy position. The same endoscopic procedure was used irrespective of patient gender, with either a pediatric 9.5F or 10F panendoscope, depending on surgeon preference.
Bladder neck injection (BNI) was performed preferably in an antegrade approach via the Mitrofanoff channel if it existed; otherwise, the retrograde approach via the native urethra was performed. After careful identification of the ureteral orifices, submucosal injection was performed with a 3.7F injection needle at the 3, 6, 9, and 12 o'clock positions circumferentially at the bladder neck and posterior urethra until adequate mucosal coaptation was subjectively believed to have taken place.
Mitrofanoff injection was performed in a retrograde fashion in the submucosa at the level of the valve mechanism, until a good mucosal coaptation was achieved. Intermittent catheterization was resumed immediately after the procedure.
Study design
The primary outcome was to assess the success of endoscopic injections. The secondary outcome was the complications rate.
Evaluation and outcomes measures
Based on a review of clinical notes and questionnaires in this retrospective study, patients were categorized as being continent, improved, or incontinent. “Continence” was defined as at least a 3-hour daytime dry interval, while “improvement” was defined as an increase in the daytime dry interval to at least 2 hours. “Success” was defined as either continence or improvement. Intraoperative and postoperative adverse effects were assessed according to the Clavien classification. 6
Results
During the study period, a total of 22 patients underwent 37 endoscopic injections with Dx/HA for urinary incontinence after failed reconstruction (male, n=6 [27%]; female, n=16 [73%]).
Median patient age at the initial urinary reconstruction was 7 years (range 3–16). Initial CUR included BNR in 16, Mitrofanoff in 21, and bladder augmentation in 8 patients. Etiologies included 9 with myelomeningocele, 4 with anorectal malformations, 3 with cloacal anomalies, 3 with epispadias-bladder extrophy complex, 2 with cerebral palsy, and 1 with Eagle-Barrett syndrome. Postoperatively, incontinence was from the neourethra in 7, from the Mitrofanoff in 10, and from both in 5 patients. After being put on maximized medical therapy and verifying their adequate bladder capacity and compliance on formal urodynamic studies, those patients were selected for endoscopic injection of Dx/HA.
Median patient age at time of endoscopic injection was 13.3 years (range 4–26). At a median follow-up of 72 months (range 4–104), 16 (72%) patients were continent, 3 (14%) patients have improved, with an overall success rate of 86%. No intra- or postoperative complications were encountered. Tables 1 and 2 illustrate the outcomes according to the patient's underlying diagnosis, and the number of injection sessions, respectively. In all patients who needed additional injections, cystoscopy identified that the reason of failure was inadequate mucosal coaptation. The outcomes according to the injection site are presented below.
BNI
For the 12 patients with leaking from the bladder neck: 6 had one, 4 had two, and 2 had three injections. Five (45%) patients were continent when one injection session was performed. Four additional patients (a total of 75%) were continent when two injection sessions were performed. One additional patient (a total of 83%) were continent when three injection sessions were performed. One patient (8%) improved when three injection sessions were performed. The overall success rate for a BNI was 92%. Those who became continent needed an average of 1.4 injections with 3.0 mL Dx/HA per session. Those who improved needed an average of 1.4 injections with 3.0 mL Dx/HA per session. Those with an unchanged outcome received an average of one injection with 4.4 mL Dx/HA per session.
Mitrofanoff injection
For the 15 patients with leaking from the Mitrofanoff: 5 had one, 9 had two, and 1 had three injections. Three patients (20%) were continent when one injection session was performed. Eight additional patients (a total of 73%) were continent when two injection sessions were performed. One (7%) patient improved when one injection session was performed. One additional patient (a total of 14%) improved when three injection sessions were performed. The overall success rate for a Mitrofanoff injection was 87%. Those who became continent needed an average of 1.3 injections with 2.0 mL Dx/HA per session. Those who improved needed an average of two injections with 3.2 mL Dx/HA per session. Those with an unchanged outcome received an average of 1.5 injections with 3.1 mL Dx/HA per session.
Discussion
The use of bulking agents to manage structural incontinence in children had been used many years ago with various types of agents, such as polytetrafluoroethylene paste and bovine collagen, which has fallen out of favor. Dx/HA is nonmigratory, nonallergenic, nonmutagenic, and nonimmunogenic and currently approved by the Food and Drug Administration for the endoscopic treatment of vesicoureteral reflux in children. Dx/HA-based implants have been described in the literature for incontinence in the pediatric population. 7
BNI
Guys and associates 8 demonstrated a low success rate (47%) after injection of the incontinent bladder neck with polydimethylsiloxane, a silicone-based product. With the introduction of Dx/HA, studies showed that endoscopic injection into the bladder neck after urinary reconstruction appears safe and can increase the daytime dry interval in 79%. 9,10 Furthermore, they demonstrated that continence rates after BNI were improved in those with previous bladder augmentation (75% vs 67% without bladder augmentation). 11 More recently, another series studies the long-term follow-up of the effects of BNI after BNR. Despite the reconstructed urethra, the authors did not report any difficulty injecting the bulking agent. 12
The effectiveness of repeated injections has also been reported previously. On one hand, De Vocht and colleagues 12 and Caione and Capozza 13 stated that repeated BNI yields no additional benefits. On the other hand, Block and coworkers 14 suggested that up to three BNI had to be given to achieve maximal effectiveness, which is more consistent with our findings.
Mitrofanoff injection
In the past, formal management of incontinence from a continent catheterizable channel would necessitate surgical recreation of the original continence mechanism. Multiple series, however, proposed that a minimally invasive approach to address channel incontinence could spare these patients an additional open surgery. Kaefer and colleagues, 15 in 1997, was the first to introduce endoscopic injection of the channel with the use of collagen. Earlier series using polymethylsiloxane demonstrated success rates of up to 66%. 16,17 Dx/HA then had a better continence rate. 18,19 With a continence of 86%, Roth and associates 18 concluded that endoscopic treatment of Mitrofanoff incontinence should be attempted before open channel revision is performed. While some patients will need multiple injections, a majority of their patients became continent after a single injection.
In our center, we have been using endoscopic injection of Dx/HA as an alternative option before a formal open procedure for the treatment of patients for urinary incontinence after a failed BNR or Mitrofanoff stoma. While some might be concerned about the potential effect of a previously injected bulking agent on a later open revision, several authors mention that surgery at the bladder neck or Mitrofanoff after endoscopic injection of Dx/HA was not considered more difficult than in noninjected patients. 18,20 With up to 8 years follow-up, our results reveal a success rate of 86%. We also demonstrated that some patients might benefit from a repeated injection, with most of our patient population being completely continent on the first two sessions.
The limitations of this study include its retrospective nature where we present our experience with a small group of heterogeneous patients and no control group. Another limitation is the relatively short follow-up period in some patients to assess for potential late failures. Our study is considered underpowered; therefore, we cannot generate meaningful statistical data, but can only report trends.
Conclusions
Endoscopic injection of Dx/HA to correct residual incontinence after urinary reconstruction appears to be safe and effective, achieving complete continence or significant improvement in more than two-thirds of patients. Repeated injections may be necessary to attain a successful outcome. A trend exists for better results with larger amounts of Dx/HA injected. Endoscopic injection of Dx/HA should be considered as first-line treatment in this patient population.
Disclosure Statement
No competing financial interests exist.