Robot‐assisted laparoscopic augmentation ileocystoplasty and Mitrofanoff appendicovesicostomy in children: Step‐by‐step and modifications to UChicago technique

Abstract Objective To describe the step‐by‐step techniques and modifications for robot‐assisted augmentation ileocystoplasty and Mitrofanoff appendicovesicostomy in a pediatric population with updated institutional results. Introduction Robot‐assisted laparoscopic augmentation ileocystoplasty with Mitrofanoff appendicovesicostomy (RALIMA) protects the upper urinary tract and reestablishes continence in patients with refractory neurogenic bladder. Robotic assistance could provide the benefits of minimally invasive surgery without the challenges of pure laparoscopy. Here, we focus on the outcomes of RALIMA with salient tips and modifications of the technique. Methods We performed a retrospective review of our robotic database and identified 24 patients who underwent attempted robot‐assisted laparoscopic augmentation ileocystoplasty (RALI) between 2008 and 2017 by a single surgeon at an academic center. Outcomes of interest included operative time, hospitalization time, postoperative complications, and change in bladder capacity. RALI and all concomitant procedures were performed using the da Vinci® surgical system (Intuitive Surgical, Sunnyvale, CA, USA). Results Of 24 patients, 20 successfully underwent RALI. Eighty percent underwent concomitant appendicovesicostomy (APV), 40% underwent antegrade continence enema channel formation (ACE), and 30% underwent a bladder neck procedure. Mean operative time was 573 minutes and the most recent RALIMA was 360 minutes. The average return to regular diet was 3.9 days and length of stay was 6.9 days. Mean change in bladder capacity was 244% postoperatively. Thirty‐day complications were noted in 35% of patients; one Clavian grade I (5%) complication, five grade II (25%) complications, and one grade IIIb (5%) complication. With a median follow‐up of 83.1 months we note a 25% incidence of bladder stones, 15% upper tract stones, 5% incidence of bladder rupture, and 5% small bowel obstruction. No patients required re‐augmentation in the follow‐up period. Conclusions RALI has similar functional outcomes and complications when compared with the open augmentation ileocystoplasty literature. RALI is desirable due to favorable pain control with decreased length of stay. Long‐term outcomes after RALI are similar to the open approach. As the operative time is currently the largest point of criticism with the robotic approach, we discuss modifications to decrease the operative time.


| SURG I C AL TECHNI QUE
*Indicates modified technique.

| Preoperative preparation
We do not use a bowel preparation preoperatively. [5][6][7] Patients are encouraged to continue a regular diet until midnight the night before surgery. Patients are admitted to the hospital postoperatively on the day of surgery. A preoperative antibiotic protocol of weight-based cefazolin, metronidazole, and gentamicin is administered within the 30 minutes before skin incision and are continued for 24-48 hours.
Patients with ventriculoperitoneal shunts also receive one dose of prophylactic vancomycin preoperatively. A neurosurgery consult is obtained prior to surgery for these patients.

| Patient positioning, port placement, and robot docking
The patient is positioned in a supine semilithotomy with 10 degrees Trendelenburg. A Foley catheter is placed sterilely on the field. A nasogastric tube is inserted for the duration of the surgical procedure. A 12-mm camera port is placed in a supraumbilical position using Hasson's technique*. We have previously described umbilical camera port placement; however, we find the supraumbilical port placement allows for easy identification and dissection of the appendix and bowel. If utilizing the DaVinci X or Xi® robot, an 8-mm camera port is placed in a similar position. After establishing pneumoperitoneum, the 8-mm robotic working arm ports are placed laterally at the level of the umbilicus in the midclavicular line. A 5-mm assistant port is placed in the left upper quadrant, inferior to the costal margin and in the midclavicular line (this can be substituted for a large assistant port if staples are utilized for bowel anastomosis). A fourth robotic working arm port can be placed at the site of stoma creation in the right iliac fossa for patients who are greater than 12 years of age or 5 feet tall due to space restriction.

| Diagnostic peritoneoscopy
We recommend beginning the case with diagnostic peritoneoscopy and lysis of adhesions if necessary. This step facilitates the ease of appendix isolation, especially the suprahepatic locations in patients with VP shunts. If present, a VP shunt can be placed into an endocatch bag. The appendix is identified, ensuring adequate length and vascularity to allow for successful appendicovesicostomy. The evaluation of the appendix and intra-abdominal anat-

| Appendiceal isolation and harvest
A traction suture can be placed at the tip of the appendix to aid in dissection and manipulation. A 4-0 Vicryl suture (polyglactin) is placed as a stay suture and a mesenteric window with adequate blood supply is developed. The appendix is then excised from the cecum. If a short appendix is noted, a cecal flap can be created to ensure adequate length and to avoid stomal stenosis.* In those who require antegrade continence enema channel creation, the length of the appendix will determine the need for a split technique vs a cecal flap. The defect in the cecum is closed in two layers.

| Ileal loop isolation and anastomosis
A 20-cm ileal segment is isolated 20 cm proximal to the ileocecal junction for the cystoplasty patch. Percutaneous stay stitches placed in the proximal and distal ends of the bowel are performed with Keith needles. This maneuver provides traction of the bowel and allows for easier isolation and anastomosis. A premeasured umbilical tape is used to ensure accurate measurement of the bowel segments. After ensuring mesenteric length and that the ileal segment will reach the bladder, the ileal loop is transected. We demonstrate the division of the mesentery with Harmonic scalpel®* and bipolar forceps to reduce bleeding and facilitate the dissection.
Bowel continuity is re-established by hand sewn single-layer seromuscular anastomosis using 5-0 PDS in children, or 4-0 PDS in adults. We start the anastomosis on the antimesenteric border using a running stitch toward the mesentery on the posterior wall. On the anterior wall of the bowel, a separate stitch is run from the mesenteric border toward the antimesenteric border. The mesenteric defect is closed to prevent the possibility of closed loop bowel obstruction.

| Detrusorotomy and extravesical appendicovesicostomy
In the case of a short appendix (as such in the video), an oblique extravesical appendicovesicostomy with stoma formation in the right iliac fossa can be created. Otherwise, we suggest the intravesical approach with posterior wall implantation to reduce the operative time.* The detrusorotomy can be made in the coronal plane to reduce bleeding.* When performing the intravesical approach, the appendix is brought to the posterior wall and oriented according to planned stoma site creation (umbilical location proceed with midline anastomosis, while right iliac fossa requires an oblique anastomosis). The bladder is distended with sterile saline and a detrusorotomy is made along the posterior bladder wall in the coronal plane. The previously placed stay suture at the tip of the appendix allows for easy manipulation while minimizing direct handling of the appendix. The appendix is spatulated and anastomosed to the bladder mucosa with interrupted 5′0 PDS II® Medline Industries (polydioxanone) stitches. An 8 French feeding tube is placed within the appendix. After the anastomosis is performed, the detrusor muscle is closed over the appendix with 3-0 or 4-0 Vicryl in continuous fashion without tightening. We do not fenestrate the mesentery or tack the appendix to the bladder wall.

| Ileal detubularization
The previously isolated ileal segment is now detubularized along the antimesenteric border with a harmonic scalpel*. This allows for a reduction in operative time by reducing the bleeding. Stay sutures are placed at the proximal and distal ends of the ileal patch to prevent torsion of the mesentery.

| Cystotomy and patch ileoystoplasty
The cystotomy is performed in the coronal plane. A thick-walled bladder is often encountered, and we find the harmonic scalpel aids in hemostasis and decreases operative time compared to our previous use of monopolar scissors.* We then turn attention to the augmentation with ileal patch. The detubularized bowel is sutured to the apices of the cystotomy. Utilization of the 4th arm can aid in retraction and exposure. We now use a barbed quill suture to perform the posterior bowel bladder anastomosis in a continuous fashion.* We utilize either a 2-0 Quill TM suture (Surgical specialties corporation), Vicryl, or polydioxanone (PDS). In our experience, placement of only one suprapubic catheter often leads to dislodgement and clogging; therefore, two suprapubic catheters are placed percutaneously to provide maximal drainage.* The anterior bladder bowel anastomosis is then performed, working from the apices toward the midline. The augmented bladder is filled with sterile water to identify leakage.

| Maturation of appendix stoma to right iliac fossa
The appendix can be brought to the predetermined stoma site with the assistance of a stay suture.

| ME THODS
After obtaining approval from our institutional IRB, we identified patients who underwent robot-assisted laparoscopic ileocystoplasty

| RE SULTS
A total of 24 patients were scheduled to undergo RALI; however, four were converted to open surgery and therefore excluded from the final analysis. Reasons for open conversion include two patients with kyphosis limiting intra-abdominal space on insufflation and two patients with dense adhesions and unfavorable appendiceal anatomy. Table 1 displays patient characteristics. The median age was 11.7 years old and 60% of patients were male. The median weight in kilograms was 44.3 and median BMI was 19.9. Eight patients (40%) had VP shunts.
Only two patients (10%) had a history of posterior urethral valves (PUV). Our most recent patient underwent RALIMA with an operative time of 360 minutes. Mean time to regular diet was 3.9 days and mean length of stay was 6.9 days. Mean change in bladder capacity was increased by 244% postoperatively which is displayed in Table 3.  Length of stay, days (95% CI) 6.9 (5.8-8) Complications are described in Table 4

| D ISCUSS I ON
The robotic approach has been favored over open for decreased pain, cosmesis, and improved tissue handling without compromising outcomes across many pediatric operations. [8][9][10] The laparoscopic approach to augmentation cystoplasty has not been widely accepted due to steep learning curve and increased operative time. 3,11 Robotassisted surgery offers the ability to shorten the learning curve and allow minimally invasive surgery to become accessible even in more complex reconstructive procedures. RALI have been shown to decrease postoperative opioid use when compared to open equivalent as well as a trend toward decreased length of stay. 2,12 Although RALIMA has its advantages, appropriate patient selection is key. Anatomic considerations, such as kyphosis leading to poor intra-abdominal space, prior surgery leading to dense adhesions, and appendiceal anatomy can necessitate open conversion. 12 The patient population requiring this surgery often may be wheelchair bound or have contractures, further leading to difficulty positioning.
In our series we describe one patient who had a brief neuropraxia attributed to positioning. Additionally, unforeseen compliance is-  15 Although operative time remains the major criticism of the robotic approach, with increasing experience, this gap is closing.
The following modifications have enabled a shorter operative time as displayed in Figure 3. The supraumbilical camera port placement allows for easier handling of the bowel and appendix. The use of the harmonic scalpel has improved operative times when performing the cystotomy, detubularizing the ileum and dividing the mesentery. We find the use of this device has superior hemostasis when compared to the monopolar/bipolar robotic energy sources.
Making the detrusorotomy in the coronal plane is another way to decrease bleeding and therefore operative time. Utilizing a port site for the APV can also aid in decreasing operative time.