Retroperitoneoscopic Nephrectomy/Heminephrectomy in Children Planned,Performed,and Managed by Supervised Senior Pediatric Surgical Trainees

Introduction

Laparoscopic surgery has gained acceptance in pediatric urology in the last decade because it is minimally invasive, is associated with decreased postoperative discomfort, involves shorter recovery, and yields improved cosmesis. ¹ Laparoscopic nephrectomy was performed for the first time by Clayman et al. ² in 1991, using a transperitoneal approach, and Figenshau et al. ³ reported the first pediatric laparoscopic nephroureterectomy in 1994. Retroperitoneoscopic nephrectomy (R-neph) in adults was first reported by Gauer et al. ⁴ and Kerbl et al. ⁵ in 1993 and has been introduced at multiple institutions for treating pediatric patients.^6–8 However, R-neph is still not generally favored by pediatric surgeons for various reasons, including lack of experience of retroperitoneal anatomy compared with pediatric urologists, concern about long operative times and related complications, and litigation. ⁹ Thus, there is a dearth of urologic cases that general pediatric surgeons can manage, with little opportunity to master retroperitoneoscopy. ¹⁰

We decided to compare morbidity, perioperative safety, and learning curves between a series of R-neph patients, managed and treated by senior pediatric surgical trainees (SPST), supervised by a single board-certified pediatric surgeon (BCPS) (A.Y.), and a series of conventional open nephrectomy and heminephrectomy (O-neph) patients managed and treated by BCPS.

Materials and Methods

The medical records of all patients who had had nephrectomy or heminephrectomy at our institution between 2004 and 2011 were reviewed retrospectively. From 2004 to 2008, all nephrectomies and heminephrectomies were conventional O-neph procedures performed through a flank incision. R-neph was introduced in 2009 and is currently our treatment of choice.

For O-neph, all surgery was planned and performed by 4 BCPS. However, for R-neph, preoperative planning, surgery, and postoperative management were arranged and performed by 5 SPST, each of whom was between 33 to 37 years old, with 5–8 years of experience as trainee pediatric surgeons, awaiting board certification, under the supervision of one BCPS (A.Y.). As a prerequisite, each SPST had to complete 20 cases of major laparoscopic or thoracoscopic surgery in our animal laboratory as the operating surgeon, involving procedures such as Nissen fundoplication, esophagoesophagostomy, and advanced urologic surgery including pyeloplasty, and participate actively in at least 30 cases of major laparoscopic or thoracoscopic surgery at our institution as an assistant (surgeon or camera assistant), including surgery for esophageal atresia, congenital diaphragmatic hernia, choledochal cyst, biliary atresia, anorectal malformation, or Hirschsprung's disease, to allow each SPST to experience both operating directly on animals and the clinical management of patients.

In all cases (both O-neph and R-neph), general anesthesia was established, and the patient was placed in the lateral decubitus position. In O-neph, simple or subcapsular nephrectomy was performed through a 5–10-cm incision commencing one finger-breadth below the tip of the 12th rib. In R-neph, retroperitoneal access was achieved by placing the first trocar through an incision one finger-breadth below the tip of the 12th rib, using either the open Hasson technique (for the first 5 of 11 cases) or the Optie technique (for the latter 6 of 11 cases). In all R-neph cases, carbon dioxide was insufflated at a pressure of 8–12 mm Hg, and the tip of the retroperitoneoscope (laparoscope) was used to create a working space. A second 5-mm trocar was inserted posteriorly in front of the lumbosacral muscle. A third 5-mm trocar was inserted in the midaxillary line, 10–15 mm from the top of the iliac crest, and a fourth 5-mm trocar was inserted in the anterior axillary line close to the iliac crest. Insufflation pressure was maintained at <12 mm Hg by increasing the flow rate of CO₂ progressively from 1 to 2 L/min. For nephrectomy cases, simple nephrectomy was usually performed. The diseased kidney was freed along the renal parenchyma after Gerota's fascia and fat were incised. Any adhesions were resected, and the renal artery and vein were freed and clipped with titanium clips near the abdominal aorta and inferior vena cava, respectively. The renal pedicle was usually resected with a Harmonic^® scalpel (Ethicon Endo-Surgery, Cincinnati, OH). The distal ureter was dissected as far as possible and ligated. The diseased kidney was excised and placed in a specimen bag, and one of the four trocar sites was enlarged to allow the intact specimen, or smaller crushed pieces, to be removed. During heminephrectomy (partial nephrectomy), the upper (or lower) pole vessels were identified and clipped in R-neph and were tied and ligated in O-neph. The upper pole was identified by color changes after vessel ligature and primarily by the difference in aspect between the normal lower pole and dilated dysplastic upper pole. A drainage tube was placed at the operative site for 2 or 3 days. All patients were treated with prophylactic antibiotics for 5 days.

All medical records were reviewed retrospectively to compare mean operative time (MOT), incidence of complications, requirement for postoperative analgesia, and progress of postoperative recovery. Data were analyzed using standard statistical methods. Demographic data and surgical outcome were compared using Student's t test. The chi-squared test or Fisher's exact test was used for analyzing the incidence of complications. For all statistics, .05 was used to determine significance.

Results

We treated 20 cases with O-neph from 2004 to 2008 and 11 cases with R-neph from 2009 to 2011, using a four-trocar placement technique. In O-neph, there were 14 total nephrectomies and 6 upper pole nephrectomies, indicated for symptomatic multicystic dysplastic kidney (MCDK) in 6, symptomatic duplex kidney in 6, incontinence secondary to an ectopic ureter in 4, and hydronephrosis with no functioning kidney in 4. In R-neph, there were 9 total nephrectomies and 2 upper pole nephrectomies, indicated for MCDK in 3, duplex kidney in 3, ectopic ureter in 3, and hydronephrosis in 2. Mean age at surgery was 41 (range, 4–109) months in O-neph and 52 (23–123) months in R-neph, which was not significantly different. Median weight at the time of surgery was 14 (range, 9–27) kg in O-neph and 19 (13–32) kg in R-neph. One heminephrectomy (partial nephrectomy) case in the R-neph group required conversion to O-neph because of the technical difficulty of dividing the upper and lower pole without injuring the unaffected side; the peritoneum was preserved intact at the time of conversion.

MOT was 137 (range, 85–290) minutes in O-neph and 197 (116–341) minutes in R-neph. Changes in MOT over time in R-neph are shown in Figure 1. Operating time decreased from 281 minutes for the first case to 116 minutes for the last case, and MOT for the first 5 R-neph cases was 249 minutes and for the last 6 cases was 153 minutes (P<.05). Mean blood loss was 17 (range, 1–55) mL in O-neph and 12.5 (2–40) mL in R-neph. There were no transfusions and no intraoperative complications. One heminephrectomy (partial nephrectomy) patient in each group developed postoperative urinoma that was managed conservatively. Mean duration of postoperative bed rest was 1.0 day in O-neph and 0.6 days in R-neph. Mean requirement for fentanyl was 21.5 (range, 10–40) μg/kg in O-neph and 4.4 (0–20) μg/kg in R-neph, which was significantly different (P<.05). Nonsteroidal anti-inflammatory drug suppositories were required for a mean of 2.3 (range, 0–5) days postoperatively in O-neph and 1.1 (0–2) days in R-neph, which was also significantly different (P<.05). Full oral feeding was resumed postoperatively after a mean of 1.6 (range, 1–2) days in O-neph and 1.2 (1–2) days in R-neph (Table 1).

OPEN IN VIEWER

FIG. 1.

Change in operating time in retroperitoneoscopy nephrectomy. Mean operating time for the first 5 retroperitoneoscopy nephrectomy and heminephrectomy cases was 249 minutes, and that for the last 6 cases was 153 minutes (P<.05).

Discussion

Surgeons might feel that minimally invasive surgery is sometimes unnecessarily complicated because they do not have a good surgical field to look at or cannot physically feel what they are doing. When retroperitoneal procedures are performed using minimally invasive surgery, there are additional limitations related to space and maneuverability with which board-certified senior staff themselves may not be confident, so there is almost no logical basis for younger colleagues to be placed in charge. Although there are limitations related to this study—namely, the small number of cases and its retrospective nature—we believe we have indicated that SPST can indeed plan, perform, and manage a “technically advanced” procedure reliably, with safety.

With any surgery, patient safety is of paramount concern, and the main advantage of the retroperitoneoscopic approach is that it avoids opening the peritoneum, which better simulates the open approach, and if any complication arises, it is restricted to this space.^11–13 However, the retroperitoneal space is small, which makes R-neph difficult for beginners ¹¹ ; thus there is a learning curve that has to be overcome. Although, for training using a transperitoneal approach, there are examples in the literature of using a training box or animal models, ¹⁴ there is no wet model available for R-neph, and few animal models have been described, ¹⁵ with the implication that, because of the reduced space, greater skill is required, and a good training model would be useful for both beginners and experienced surgeons alike. ¹⁵ To this end, we developed a specific program to give the five senior trainees involved in this study experience of a specific number of surgical cases involving both clinical and animal laboratory experience, with exposure to at least five retroperitoneoscopic procedures in our animal laboratory before they treated any patient.

In fact, in recent years, R-neph has been adopted and modified by some pediatric laparoscopic surgeons because it minimizes some of the risks associated with transperitoneal laparoscopy.^16–18 In 1999, Borer et al. ¹⁸ were the first to describe pediatric R-neph in the prone position, and in 2002, Urbanowicz et al. ¹⁷ reported their initial experience with retroperitoneoscopic access using three ports in 12 children including complete nephrectomy, nephroureterectomy, and upper pole heminephrectomy. Their MOT was 110 minutes, and length of hospital stay was 3 days. In 2003, El-Ghoneimi et al. ¹⁶ also reported using a three-trocar technique for retroperitoneoscopic nephroureterectomy and partial nephroureterectomy in 15 patients. Their MOT was 150 minutes, with minimal blood loss, and mean hospital stay was 1.4 days. They reported no intraoperative complications except for one conversion to open surgery because of a large peritoneal tear, which can be a complication of the retroperitoneal approach especially in infants, because of difficulty establishing adequate retropneumoperitoneum. This is actually the most common complication of R-neph, and the main cause of conversion to open surgery. ¹⁹

In adult urology, Gill et al. ¹¹ reported an overall complication rate of 4.7% for retroperitoneoscopic approaches to several procedures in an international survey of experience at 36 centers, including pneumothorax and renal vein injury as the most common visceral and vascular complications. Rassweiler et al. ¹³ described intraoperative complications and repeat intervention rates of 9% and 4.6%, respectively, in a series of 65 simple nephrectomies performed at one center. The specific age of patients who underwent nephrectomy is not stated in either series.^11,13

Of note is that the exact incidence of specific complications is not known in the pediatric population because there are so few pediatric series in the literature. Based on adult reports, we can postulate what complications may be likely to arise in children, but in our study, only 1 case of R-neph required conversion to O-neph, even though senior trainees were in charge. We believe our R-neph is safer is because we use four trocars, instead of making do with three. This allows us to establish and develop the retropneumoperitoneum carefully, reducing the risk for compromising the peritoneum. Once enough space is available, the third and fourth trocars can be inserted. Thus, the surgeon's hands are freed to concentrate on dissecting the renal hilum if a grasper is inserted through one of the trocars to provide countertraction.

With regard to technical issues, initial MOT in our series was longer than in other reports but decreased with time, indicating there is a learning curve. Initially, in R-neph, the first trocar was placed using the open Hasson technique, and later it was placed using the Optie technique. We believe this change was possible because of experience gained from working in the retroperitoneal space over time; however, for less experienced general pediatric surgeons, the open Hasson technique would be less traumatic. Thus, the importance of experience cannot be overemphasized because with minimally invasive surgery, “hands on” advice does not impart knowledge as in open surgery; rather, each skill must be mastered with confidence before a procedure can be performed reliably. In other words, during open surgery, any surgeon involved in the operation is usually scrubbed and can only give sound advice by demonstrating directly to one younger surgeon at a time, whereas with minimally invasive surgery, several younger surgeons can observe and learn by using monitors, either in the operating room or elsewhere. Thus, there is more opportunity to master skills and decrease learning curves. Higashihara et al. ²⁰ reported on the learning curve associated with laparoscopic (transperitoneal) nephrectomy. In their study, operating time for benign indications stabilized at almost 200 minutes after 10 cases, which is approximately the same as for open surgery. Rassweiler et al. ¹³ reported a learning curve for general retroperitoneoscopy of 50 human cases, which of course must be related to the surgeon's background level of experience. In our study, there was a significant difference between MOT for the first 5 cases and the last 6 cases, suggesting that the learning curve for R-neph may be relatively short for well-trained staff. As our case numbers increase, we will be able to give a more accurate recommendation about the learning curve for R-neph.

The current study suggests that despite the image of retroperitoneoscopic surgery being an advanced technique, it would appear that SPST are capable of planning, performing, and managing R-neph patients. Unfortunately, general pediatric surgeons have little constant exposure to urologic surgery involving the retroperitoneal space, including nephrectomy and heminephrectomy, so a systematic training program, involving practice of technical skills in an animal laboratory and clinical case experience, would appear to best overcome technical learning curve issues as early as possible.