Retroperitoneoscopic Nephroureterectomy via Three Trocars in Pediatric Patients with End-Stage Reflux Nephropathy

Introduction

L aparoscopic surgery has gained great acceptance in pediatric urology in the last decade with major advantages such as minimally invasive surgery, decreased postoperative discomfort, shortened recovery period, and improved cosmesis. ¹ Figenshau and colleagues ² performed the initial pediatric laparoscopic nephroureterectomy in 1994. Although some authors recommend nephrectomy and only partial ureterectomy because of the low complication and morbidity rates of the ureteral stumps, ^3,4 Casale and coworkers ⁵ suggested nephroureterectomy or heminephroureterectomy to the level of the bladder hiatus in patients with reflux into a poorly functioning kidney, because 19% of refluxing distal ureteral remnant stumps became symptomatic during follow-up. Therefore, complete nephroureterectomy via an open or a laparoscopic procedure seems to be a reasonable approach in children with end-stage reflux nephropathy.

Laparoscopic nephroureterectomy can be performed either by a transperitoneal or a retroperitoneal approach. ⁶ There is still some controversy regarding the choice of either a transperitoneal or a retroperitoneal approach. Performing nephroureterectomy via a transperitoneal approach seems to be easier than a retroperitoneoscopic approach, because the abdominal cavity has more space for manipulation of laparoscopic instruments. Postoperative bowel adhesions, direct injury to abdominal organs, omental or visceral herniation through port sites or incisions for specimen extraction, and ileus, however, are the complications that might be seen in the transperitoneal approach. Supporters of the retroperitoneal approach believe that retroperitoneoscopic access avoids entry into the peritoneal cavity, thus preventing the complications. ⁷ In addition, it is not common to use a three-trocar retroperitoneal approach for nephroureterectomy with complete ureteral excision in the pediatric age group.

In this study, we present the results of pediatric retroperitoneoscopic nephroureterectomy with complete ureteral excision performed by using three trocars in patients with end-stage reflux nephropathy.

Patients and Methods

Between August 2003 and November 2009, 13 children, ages 15 months to 14 years, who had end-stage reflux nephropathy with primary dysplasia underwent retroperitoneoscopic nephroureterectomy under general anesthesia by the same three urologists. All patients were followed by members of the pediatric nephrology department before and after the operation. The indication for nephroureterectomy was recurrent urinary tract infection (UTI) that cannot be controlled despite prophylactic antibiotic therapy. Split renal functions of the removed kidneys of all patients were below 10% on ^99mtechnetium dimercaptosuccinic acid scintigraphy. The contralateral kidneys of the patients were normal.

The patients were placed in the lateral decubitus position with a kidney break. Three trocars were used in all of the operations. In nine of the patients—one of them was a 15-month-old boy—three 5-mm trocars were used; one trocar was placed in the midaxillary line (MAL) below the tip of the 12th rib, one in the anterior axillary line (AAL) at the level of the umbilicus, and the other in the posterior axillary line (PAL) at the subcostal level. In three patients, one 10-mm trocar was placed in the MAL and the other 10-mm trocar in the PAL and a 5-mm trocar in the AAL. A 10-mm trocar was placed in the PAL and two 5-mm trocars were placed in the MAL and the AAL in one patient.

For insertion of the first trocar, a 5- or 10-mm transverse muscle-splitting incision was made to the skin on the MAL (in bigger children) or the PAL (in smaller children), followed by blunt entry to the retroperitoneum with a hemostat clamp. Then, a 5- or 10-mm trocar, with an excised glove finger securely attached to its distal end, was inserted. The finger (balloon) was then filled with 200 to 250 mL of saline through the tap of the trocar, and it remained in place for 5 minutes to aid in creating the retroperitoneal space. After the inspection of the retroperitoneum with a laparoscope through the glove finger that became thin and transparent, this trocar was replaced with another trocar in the same size that was then used for insertion of a 5- or 10-mm laparoscope.

Pneumoretroperitoneum was created by insufflation of CO₂, and pressure was maintained at 12 mm Hg. After initial exploration, secondary trocars were placed under direct vision. The first visual landmark was the psoas muscle, which was followed medially to the ureter and superiorly to the kidney. Using laparoscopic scissors, aspirator, and dissector, the renal hilum was dissected initially. Each renal artery and vein were ligated with laparoscopic clips and divided, leaving three clips on the patient side. The ureter was identified and dissected downward to the ureterovesical junction (UVJ) until visualization of detrusor fibers. Then, the ureter was ligated and cut with two laparoscopic clips or LigaSure™ device (Covidien, Mansfield, MA) at the UVJ to avoid a residual ureteral stump (Fig. 1). The kidney was dissected free from the retroperitoneal fatty tissues. The nephroureterectomy specimen was extracted through the MAL or PAL incision after a small extension if needed (Figs. 2A, 2B), and a drain was placed into the nephrectomy lodge. If a 10 mm trocar was used, the fascia incision was sutured by 3/0 polyglycolic acid suture, and the other skin incisions were closed with intracuticular running 4/0 Caprosyn™ (Covidien, Mansfield, MA) suture.

OPEN IN VIEWER

FIG. 1.

The ureteral ligation at the ureterovesical junction.

OPEN IN VIEWER

FIG. 2A, 2B.

The nephroureterectomy specimen extraction.

Results

The mean age of the patients was 6.5 ± 4.4 years (15 months–14 years). Seven of the patients were boys and six of them were girls. Eight of the operations were performed on the left and five on the right side. Mean operative time was 137 ± 47 minutes (75–230 min). Estimated blood losses were minimal. Oral intake was started 4 to 6 hours after the operation. Mean hospitalization time was 2.2 ± 0.9 days (2–5 d). After the removal of the drains within the first 24 hours, nine patients were discharged on the second postoperative day. In three patients, the drains were removed 48 hours after the operation. No major complication or conversion to open surgery was seen.

An 8-mm ileal serosal injury occurred during the dissection of the kidney with very dense peripheral adhesions in one patient, however. This serosal injury was immediately repaired retroperitoneoscopically with 4/0 polyglycolic acid Lembert sutures and then the surgery was continued and completed without conversion to transperitoneal or open surgery. This patient was discharged 5 days after the operation without any problem. Also, in four patients, small unintentional bladder openings during the dissection of the UVJ occurred. These bladder openings were immediately sutured retroperitoneoscopically with 4/0 polyglycolic acid sutures, and the operations were continued without conversion to transperitoneal or open surgery. Urethral catheters were placed in these patients at the end of the operations. The urethral catheters were removed within the first week. Millimetric peritoneal tearing occurred during the dissection of the tortuous ureter in three patients. These tearings were too small, and we were able to continue the procedure without any interference. In one patient, subcutaneous emphysema was seen and resolved spontaneously within a few hours.

Overall, postoperative analgesic requirement was minimal. No postoperative analgesic was necessary in two patients. In four patients, a single, in five patients two intramuscular injections of 0.3 mg/kg metamizol sodium were administered. In two patients, ibuprofen 20 mg/kg/d was used for 2 days because of fever without any sign of infection. Mean follow-up was 38.5 ± 28 months (3–78 mos). Cosmetic results were excellent at the initial postoperative visit. No recurrent UTI was seen in any of the patients in the follow-up.

Discussion

Laparoscopic surgery has gained popularity in the pediatric age group with some advantages over standard open techniques, including minimally invasive surgery, excellent visualization of the anatomic structures by magnification, less postoperative pain and discomfort, less analgesic requirement, decreased hospital stay, shortened recovery period, and improved cosmesis. The often reported disadvantages are technical difficulty and increased operative time. As laparoscopists continue to gain experience, surgery becomes less difficult, and operative time and rate of conversion to open surgery continue to decrease.

In 1994, Figenshau and associates ² performed the initial pediatric laparoscopic nephroureterectomy in a 6-year-old girl who had grade V vesicoureteral reflux and reflux nephropathy on the left kidney and a 30-month history of recurrent Pseudomonas urinary infections despite suppressive oral antibiotic therapy. The operation was performed transperitoneally by using five trocars. Total operative time was 5 hours and 35 minutes, estimated blood loss was 50 mL, and the patient was discharged on the second postoperative day. ² This case was followed by other pediatric transperitoneal nephrectomy and nephroureterectomy series. ^{8 –10}

The retroperitoneoscopic approach for nephrectomy would minimize some of the risks of transperitoneal laparoscopy and has been adopted by some pediatric laparoscopic surgeons in recent years. ^7,11,12 In 1999, Borer and colleagues ⁷ described pediatric retroperitoneoscopic nephrectomy in the prone position. In 2002, Urbanowicz and coworkers ¹¹ reported their initial experience with retroperitoneoscopic access via three ports in 12 children including complete nephrectomy, nephroureterectomy, or upper pole heminephrectomy. Their average operating time was 110 minutes, and length of hospital stay was 3 days (2–5 d). In 2003, El-Ghoneimi and colleagues ¹² reported the three-trocar technique for retroperitoneoscopic nephroureterectomy procedures in pediatric patients. They used a three-trocar retroperitoneal approach for partial nephroureterectomy in 15 children. The mean operative time was 152 minutes (75–240 min), and blood loss was minimal (<20 mL). The mean hospital stay was 1.4 days (1–3 d). There was no intraoperative complication except conversion to open surgery in one patient because of large peritoneal tearing.

All of the operations in our series were performed retroperitoneoscopically by using only three trocars. In our series, the mean operative time was 137 ± 47 minutes (75–230 min), similar to the studies of Urbanowicz and associates ¹¹ and El-Ghoneimi and coworkers. ¹² The estimated blood loss, mean hospitalization time, and time to begin oral intake were also comparable with these series. In all of the series, overall analgesic requirements were minimal, no serious perioperative and long-term complications were seen, and cosmetic results were excellent.

The main inconvenience of the retroperitoneal access is the higher incidence of peritoneal tears in infants, which prevent the creation of an adequate retropneumoperitoneum. This is the most common complication and also the main cause of conversion to open surgery. ¹⁰ In our series, small peritoneal tearing occurred in three patients, but we were able to complete the operation retroperitoneoscopically without any need for conversion to the transperitoneal aproach or open surgery. We believe that, even in the case of a large peritoneal tear, it is still possible to complete the operation without conversion to open surgery, although conversion to transperitoneal laparoscopy may usually be necessary. Lam and coworkers ¹³ reported the results of their study that compared retroperitoneoscopic (n = 10) and transperitoneal (n = 12) laparoscopic nephrectomy in childen. Skin to skin operative times were significantly shorter in the retroperitoneoscopic group compared with the transperitoneal group. Also, the postoperative morphine requirement in the retroperitoneoscopic group was about one-third that of the transperitoneal group. They interpreted this result to the lack of peritoneal irritation from any spillage of blood with the retroperitoneoscopic approach and the absence of potential adverse effects of creating a pneumoperitoneum. Patients were discharged home within 2 days of surgery in both groups.

The results of the retroperitoneoscopic group in this study are similar to those in our series. They also affirm our belief by stating that peritoneal tears do not always necessitate conversion to open surgery. Finally, they concluded that retroperitoneal nephrectomy appears to be safe and a quick technique in children, with reduced intraoperative physiologic effects and less postoperative pain. Kim and associates ¹⁴ systematically reviewed the transperitoneal (n = 288) and retroperitoneal (n = 401) approaches for laparoscopic nephrectomy in children from several etiologies including vesicoureteral reflux (60 reteroperitoneoscopic, 72 transperitoneal). They concluded that both approaches are suitable and safe for children with comparable outcomes; therefore, the choice of approach should be individualized according to surgeon preference and patient anatomy.

In this study, the weighted average hospital stay was slightly longer for the retroperitoneoscopic series than for the transperitoneal series (2.5 vs 2.3 d). The weighted average operative time for the retroperitoneoscopic approach was shorter than for the transperitoneal approach (129 vs 154 min). Also, they found no statistically significant differences between the transperitoneal and retroperitoneal approaches in terms of vascular and bowel injuries. This shows that the retroperitoneal approach does not provide additional safety for intraperitoneal organs. Extensive attention should be given even in the retroperitoneal approach, especially in secondary cases with retroperitoneal adhesions. The bowel injuries might also be repaired, however, by using the same three trocars as we performed in a patient who had an ileal serosal injury during the dissection of the kidney from adherent peritoneum in our retroperitoneoscopic series.

Nevertheless, there is no conclusive medical evidence that favors either the retroperitoneal or transperitoneal approach. We preferred the retroperitoneal approach because of easy and rapid visualization of the renal pedicle, absence of interference secondary to intraperitoneal structures and adhesions, avoidance of postoperative intraperitoneal adhesion formation, and more rapid recovery. We used only three trocars, although it is more difficult to perform the procedure through only two instrumentation trocars with one of them used mainly for retraction. Another disadvantage of using only three trocars was partial obstruction of the view by the caudal portion of the psoas muscle during the dissection of the distal ureter and UVJ. It was not a major challenge in our experience, however. On the other hand, working with four or more trocars in the narrow retroperitoneum of a pediatric patient might cause instrument conflicts and surgeon discomfort; therefore, the three-port approach is favorable. In our series, the ureter was dissected down to the UVJ without any major difficulty. We think that performing the operation in the lateral decubitus position provides better access for complete resection of the ureter than the prone position, such as Borzi ¹⁵ reported.

Recently, laparoendoscopic single-site surgery (LESS) has been represented as the latest innovation in the armamentarium of laparoscopic surgery. Since the initial report of LESS nephrectomy by Raman and colleagues ¹⁶ in 2007, several investigators have demonstrated the technical feasibility of a variety of pediatric urologic procedures such as nephroureterectomy, simple nephrectomy, partial nephrectomy, pyeloplasty, and adolescent varicocelectomy. ^17,18 Bayazit and coworkers ¹⁷ reported the initial pediatric LESS nephroureterectomy in 10-year-old girl with end-stage reflux nephropathy. The operative time was 140 minutes, and the blood loss was minimal. The hospital stay was 24 hours. Desai and coworkers ¹⁹ reported their initial 100 patients in LESS including one pediatric nephroureterectomy from severe reflux. The LESS nephroureterectomy was performed in a 13-year-old boy for severe reflux with recurrent pyelonephritis. The operative time was 90 minutes, estimated blood loss was 75 mL, and the hospital stay was 5 days. These results were comparable with our results.

The first comparative study between the LESS nephrectomy technique and the respective laparoscopic approach has been reported by Raman and associates. ²⁰ Eleven LESS nephrectomies were compared with 22 conventional laparoscopic nephrectomies. All operative parameters were similar between the two groups. The only advantage of the LESS procedure was the improved cosmetic result. Thus, the advantages of LESS surgery still exist at the theoretical level, because no clear benefit on the postoperative course and the convalescence of the patients has been proven. The possible advantage of reduced risk for trocar placement-related morbidities does not clearly reflect practice, because the accumulated laparoscopic experience has almost eliminated these complications. Moreover, the major problem of intraoperative ergonomics of instruments should be addressed. Although LESS urologic surgery is feasible for selected indications, it seems far from being a standard of care at present.

Conclusion

This small series shows that retroperitoneoscopic nephroureterectomy using three trocars is a feasible and safe procedure in pediatric patients. Performing the nephroureterectomy through only three 5- to 10-mm incisions instead of two open surgical incisions moves the advantages of laparoscopy one step forward.