De novo adhesions with extraperitoneal endosurgical para-aortic lymphadenectomy versus transperitoneal laparoscopic para-aortic lymphadenectomy: A randomized experimental study

GENERAL OBSTETRICS AND GYNECOLOGY Gynecology

De novo adhesions with extraperitoneal endosurgical para-aortic lymphadenectomy versus transperitoneal laparoscopic paraaortic lymphadenectomy: A randomized experimental study Bruno Occelli, MD,a Fabrice Narducci, MD,a Dominique Lanvin, MD,a Denis Querleu, MD,a Eric Coste, MD,b Bernard Castelain, MD,c David Gibon, PhD,c and Eric LeBlanc, MDc Lille, France OBJECTIVE: The objective of this study was to compare the risk of adhesion formation and the site of postoperative adhesions after transperitoneal versus extraperitoneal laparoscopic para-aortic lymphadenectomy. STUDY DESIGN: A prospective study was conducted on 66 pigs, randomly allocated to 2 groups of 33 animals that underwent transperitoneal versus extraperitoneal laparoscopic para-aortic lymphadenectomy. RESULTS: No statistically significant difference was observed between the 2 techniques in terms of operating time, number of lymph nodes removed, and intraoperative and postoperative morbidity. More lymphoceles were observed in the extraperitoneal laparoscopy group (P = .0002). The overall adhesion rates were 76% in the transperitoneal group and 43.33% in the extraperitoneal group (P = .04). The adhesion score specific to the operative site was lower in the extraperitoneal group (P = .0005). A stereolocalization study showed that almost all adhesions after transperitoneal laparoscopy were situated in the para-aortic external irradiation field, whereas adhesions after extraperitoneal laparoscopy were predominantly situated outside this field. CONCLUSION: Extraperitoneal laparoscopy generates significantly fewer adhesions in the para-aortic irradiation field. (Am J Obstet Gynecol 2000;183:529-33.)

Key words: Laparoscopic surgery, para-aortic lymphadenectomy, randomized controlled study

The risk-benefit trade-off of surgical staging in cases of advanced cervical carcinoma is a matter of controversy.1 Because of the complication rate and the cost of laparotomy for para-aortic lymphadenectomy, some investigators have recently investigated the feasibility of laparoscopic para-aortic lymphadenectomy for this disease.2, 3 It is generally accepted that the risk of radiation enteritis is reduced after extraperitoneal para-aortic lymphadenectomy.4, 5 Randomized experimental studies have also provided evidence of a major reduction in ad-

From the Division of Gynecologic Oncology, Hôpital Jeanne de Flandre,a the Institut de Technologie Médicale, University Hospital,b and the Oscar Lambret Cancer Center.c Supported by grants from the Délégation à la Recherche du CHRU de Lille, and the Ligue Nationale Contre le Cancer. Received for publication April 15, 1999; revised December 22, 1999; accepted January 13, 2000. Reprint requests: Denis Querleu, MD, Hôpital Jeanne de Flandre, CHRU, 2 Avenue Oscar Lambret, 59037 Lille Cedex, France. Copyright © 2000 by Mosby, Inc. 0002-9378/2000 $12.00 + 0 6/1/105736 doi:10.1067/mob.2000.105736

hesion formation rate after laparoscopic lymphadenectomy with respect to both transperitoneal laparotomy and extraperitoneal laparotomy.6, 7 It is therefore logical to investigate the combination of laparoscopy with an extraperitoneal approach. Extraperitoneal laparoscopy was recently developed.8-10 The porcine model has been shown to be feasible and similar to human surgical practice.9-11 Before application of this technique to the clinical setting, we decided to complete our learning curve in a preliminary study12 and then compare transperitoneal laparoscopy with the new extraperitoneal technique in an animal model. The results of this study, which is the first controlled study to compare efficacy and outcome between the 2 laparoscopic approaches, are reported in this article. Material and methods We used crossbred female pigs (Large White–Landrace cross) aged 2 to 3 months and weighing between 18 and 24 kg. All rules concerning animal experimentation were respected (French decree 87-848 of October 19, 1987). 529

530 Occelli et al

All operations were performed by 2 senior residents (Bruno Occelli, MD, and Fabrice Narducci, MD [operators A and B]) experienced in laparoscopic surgery with the animals under general anesthesia. These 2 surgeons had each previously performed 4 para-aortic lymphadenectomies (2 by transperitoneal laparoscopy and 2 by extraperitoneal laparoscopy) in pigs under the supervision of a senior surgeon experienced in these techniques. The limits of lymphadenectomy were the lower edge of the renal veins for the cephalad limit, the aortic trifurcation and the inferior mesenteric artery arising from the ventral surface of the aorta 10 to 15 mm above the aortic trifurcation for the caudad limit, the psoas muscles for the dorsal limits, and the ureters for the lateral limits. The technique used for transumbilical transperitoneal aortic laparoscopy was described in a previous article.6 The technique used for extraperitoneal laparoscopic lymphadenectomy is summarized as follows. The animal was placed in the right lateral decubitus position. A retropneumoperitoneum was created by an open technique, starting with a short vertical incision 3 cm above the left nipple line. The peritoneum was visualized and left intact. The parietal peritoneum was detached in the direction of the spine by gentle lateral movements of the finger. A Blunt Tip (Origin, Menlo Park, Calif) 10-mm trocar was then introduced into this space and connected with the carbon dioxide insufflator. Two 5-mm trocars were then inserted in the anterior axillary line under visual control. The extraperitoneal pressure was maintained at 10 mm Hg. The operating time from incision until skin closure was recorded for each animal. All lymph nodes removed were fixed in 10% neutral formalin and were sent for histologic examination. Intraoperative complications and blood loss were recorded. No intraoperative or postoperative treatments, such as anti-inflammatory drugs or antibiotics, were prescribed. The animals underwent evaluation by laparotomy under general anesthesia 30 to 34 days after the operation to detect and localize any adhesions, lymphoceles (volume after aspiration), or abscesses; to specify the number of lymph nodes remaining in the operative site; to evaluate retroperitoneal fibrosis; and to determine an adhesion score for each pig.6 The animals were killed after each evaluation under general anesthesia. The choice of operator and assistant between operators A and B was randomly determined at the beginning of the study. Operator A killed the pigs operated on by operator B and vice versa. To predict the possible morbidity of each approach during adjuvant radiotherapy in human oncology, an objective stereolocalization study of postoperative adhesions was performed during the killing of the last 20 pigs (n = 10 animals that underwent transperitoneal lymphadenectomy and n = 10 animals that underwent retroperitoneal lymphadenectomy). A 3-dimensional localiza-

September 2000 Am J Obstet Gynecol

tion device was used to determine the relative positions of postoperative adhesions and anatomic structures delimiting the operative site and especially to define their positions with respect to the irradiation field. This device was composed of a video signal acquisition and processing system, 2 standard video cameras, and an infrared cursor. On the day before examination the pigs of this complementary study underwent simulation of paraaortic radiotherapy. The parameters of the irradiation field were defined according to standard human paraaortic radiotherapy criteria, between the superior edge of the first lumbar vertebra and the superior edge of the first sacral vertebra with a field width of 2 cm on either side of the lumbar spine. The para-aortic irradiation field was marked on the abdomen of the animals, which were placed in a dorsal decubitus position on the radiotherapy table, with the light beam produced by the radiotherapy simulator. At the time that the animals were killed the system recorded 4 points of the operating table (operator references) and the skin marking points. During laparotomy the operator used the cursor to record the sites of any adhesions and the anatomic structures delimiting the operative site. Computer processing (number of adhesions detected and number of adhesions in the irradiation field) and a 3-dimensional graphic representation with color coding were performed subsequently. A preliminary study12 on 20 pigs (n = 10 operated on with transperitoneal laparoscopy and n = 10 operated on with extraperitoneal laparoscopy) calculated the number of animals required to demonstrate a significant difference of the adhesion score between the 2 approaches. For a power of the study of 90% (ie, β [risk] = 10%) and α [limit of significance] = 5%) the number of pigs required in each group was 27, corresponding to a minimum of 54 animals for this study. After a 20% risk of post–random assignment exclusion was taken into account, 66 animals were required. Random assignment resulted in 15 transperitoneal laparoscopies and 18 extraperitoneal laparoscopies performed by operator A and 18 transperitoneal laparoscopies and 15 extraperitoneal laparoscopies performed by operator B. Statistical analysis consisted of parametric tests when the sample size of each group was >20—unpaired Student t test for comparison of 2 means and χ2 test and Fisher exact test for expected values