Laparoscopic Radical Hysterectomy with Pelvic Lymphadenectomy for Early, lnvasive Cervical Carcinoma

Transcription

1 November 1998, Vol. 5, No. 4 The Journal of the American Association of Gynecologic Laparoscopists Laparoscopic Radical Hysterectomy with Pelvic Lymphadenectomy for Early, lnvasive Cervical Carcinoma Dong Ho Kim, M.D., and Jong Soo Moon, M.D. Abstract Study Objective. To demonstrate the feasibility and evaluate the efficacy of laparoscopic radical hysterectomy with pelvic lymphadenectomy for early, invasive cervical cancer. Design. Prospective study (Canadian Task Force classification 11-2). Setting. University-affiliated hospital. Patients. Eighteen women (age range yrs) with early, invasive cervical cancer. Intervention. Laparoscopic radical hysterectomy with pelvic lymphadenectomy. Measurements and Main Results. Diagnoses were squamous cell carcinoma in 15 patients and adenocarcinoma of the cervix in 3; these were graded microcarcinoma in 6 and stage Ib (<4 cm) in 12. Mean operating time was approximately 363 +_ 65 minutes (range rain). Blood loss averaged 619 +_ 297 ml (range ml). The average number of pelvic lymph nodes obtained was (range 14-40). Specimen weight averaged g (range g). Surgical margins were clear in all patients. No procedure was converted to laparotomy, There were no major intraoperative complications involving injury to main blood vessels, nerves, bowel, bladder, or ureters. Conclusion. In our experience, laparoscopic radical hysterectomy with pelvic lymphadenectomy is acceptable in accordance with the standards of gynecologic oncology. Despite the longer operating time than traditional abdominal radical hysterectomy, all patients recovered as quickly as they would after laparoscopic-assisted vaginal hysterectomy. We believe that this procedure could be an alternative to abdominal radical hysterectomy for selected women, especially those who have stage Ib 1 cervical cancer. (J Am Assoc Gynecol Laparosc 5(4): , 1998) From the Department of Obstetrics and Gynecology, College of Medicine, Chung Ang University, Seoul, Korea (both authors). Supported by Chung-Ang University's special research grants in Address reprint requests to Dong Ho Kim, M.D., Department of Obstetrics and Gynecology, Chung-Ang University, Pil-Dong Hospital, 82-1, 2Ga, Pil- Dong, Chung-Gu, Seoul, Korea; fax Presented at the 24th annual meeting of the American Association of Gynecologic Laparoscopists, Orlando, Florida, November 8-12, Accepted for publication August 3,

2 Laparascopic Surgery for Early, Invasive Cervical Carcinoma Kim and Moon With advances in technology, laparoscopic surgery is increasingly performed in the field of oncology, such as laparoscopic staging of gynecologic malignancies, 1-3 urologic malignancies, 4-s and pelvic and paraaortic lymphadenectomy Some authors ~3-18 suggested that radical hysterectomy for invasive cervical cancer could be performed successfully by laparoscopy with benefits of the vaginal route. Laparoscopic surgery is adapted to lymph node dissection and gives a direct view of the ureters and origin of the uterine arteries, and vaginal surgery allows precise incision of the vaginal cuff. The combination of vaginal and laparoscopic surgery avoids the pain and discomfort of both laparotomy and perineotomy. Materials and Methods Between February 1994 and February 1996, 18 women (age range yrs, average 48.0 yrs) underwent laparoscopic radical hysterectomy with pelvic lymphadenectomy. They all provided informed consents, and were aware that laparotomy would be performed if necessary. Preoperative Assessment The women were admitted several days before surgery for baseline studies consisting of history and physical examination, routine preoperative laboratory investigations, chest radiograph, and computed tomography or magnetic resonance imaging of the abdomen and pelvis. Other standard examinations were excretory urogram, barium enema, cystoscopy, and proctosigmoidoscopy. Based on these results, patients were assigned a FIGO stage: 12 had stage Ibl and 6 stage Ia2 cervical cancer. All women had mechanical bowel preparation consisting of 2 days of saline-soap enemas and 200 ml magnesium citrate orally, followed by a kanamycin 0.1% retention enema. They received antibiotic prophylaxis with cefotaxime 1 g intravenously every 8 hours/day for 2 days before surgery, and one dose started 1 hour preoperatively. Surgical Procedure Procedures were performed under general endotracheal anesthesia. The patient was placed in dorsolithotomy position with 15-degree Trendelenburg and arms tucked by her sides. The vaginal cavity was cleansed and sterilized with povidone-iodine solution. A uterine manipulator was inserted. A Foley catheter was placed in the bladder. Laparoscopic equip- ment used for all cases included a camera attached to a 10-ram laparoscope with light source, two videomonitors, and a high-flow insufflator using CO2. Two large (10- or ll-mm) cannulas were placed in the umbilicus and at midline near the symphysis pubis. Two 5-mm cannulas were placed laterally midway between the umbilicus and anterior superior iliac crest. Suction, irrigation, bipolar coagulation, and suture materials were used. The surgical technique was nearly identical in all cases and was the same as that for laparotomy. Before the operative procedure, all pelvic structures were inspected and the abdomen explored through the laparoscope in a clockwise fashion. If adhesions were seen around the uterus and ovaries, adhesiolysis was performed. Hemostasis was achieved and ovarian (or utero-ovarian) vessels were divided. The procedures were performed in the same fashion as in classic intrafascial supracervical hysterectomy. ~9 The first endosuture was placed on the proximal part of the adnexa, which included the fallopian tube and utero-ovarian and round ligaments, an extracorporeal knot was tied, and a security knot was made. The round ligament was coagulated close to the pelvic wall with bipolar forceps and divided with scissors. A Roeder loop was placed over the pedicle for security. A second endosuture was placed over the infundibulopelvic ligament, including corresponding vessels, in the same manner. The infundibulopelvic ligament was separated with scissors, and two additional Roeder loops were applied to the stump. The same procedures were performed on the opposite side. The retroperitoneum was opened upward to the level of the aortic bifurcation and down to the level of the uterosacral ligament posteriorly. The ureter was identified at the pelvic brim, traced into the pelvis, and freed from the posterior leaf of the broad ligament. The anterior leaf of the broad ligament was dissected with scissors. The vesicouterine fold was grasped and incised. External iliac vessels were separated from the psoas muscle, and the external iliac artery was freed from its fascial sheath and separated from the underlying vein. External and common iliac lymph nodes were removed from vein surfaces by blunt or sharp dissection. We prefer early pelvic lymphadenectomy because it allows excellent skeletonization of hypogastric and uterine vessels at their origins. The obturator fossa was entered laterally and fatty nodal tissue mobilized to identify obturator nerve and vessels. Obturator lymph nodes were removed as 412

3 November ] 998, Vol. 5, No. 4 Thejournal of the American Association of Gynecologic Laparoscopists corresponding vessels and nerves were skeletonized. This bundle of tissue was freed by first dividing its distal attachments and continuing dissection cephalad along the internal iliac artery. The paravesical space was developed by grasping obliterated hypogastric arteries where they are crossed by the round ligament, approximately 1 to 2 cm inferomedial to the internal inguinal ring, and pulling them in a medial direction. Dissection was aided by introducing bipolar forceps and separating the jaws of a grasping forceps. The pararectal space was developed by peeling apart the leaf of the broad ligament, making a small opening in the dense areolar tissue at the base of the broad ligament with scissors, and insinuating a grasping forceps into the pararectal space and widening the space by separating the jaws of the instrument. Rectovaginal and vesicovaginal spaces were developed with blunt dissection while the uterus was pushed cephalad. Uterine vessels were identified, skeletonized, and ligated with loops at their origin, and cut and coagulated with bipolar forceps (Figure 1). The ureter was dissected and unroofed with scissors and bipolar forceps to the point at which it entered the bladder. The uterosacral ligament was endosutured, ligated, and cut with scissors (Figure 2). The lateral parametrium was coagulated and transected approximately 2 or 3 cm lateral to the cervix. Dissection was carried out down to 2 cm below the cervix. The same procedures were performed on the opposite side. As the FIGURE 2. View of dissection of the left ureter. The ureter is freed down to the level of the point where it enters the bladder. An endosuture is placed on the deep uterosacral ligament. uterus was pushed cephalad by an assistant, anterior or posterior colpotomy was performed. At this point, the laparoscopic procedure was stopped and radical hysterectomy was completed vaginally by incising the vagina 2 cm distal to the cervix and removing the uterus vaginally. The vaginal vault was closed with continuous locking sutures of 1-0 chromic catgut placed transvaginally in most cases to shorten operating time, but in a few cases it was closed laparoscopically. Pneumoperitoneum was reestablished with CO2 and peritonealization was performed with endosutures. We approximated anterior and posterior pelvic peritoneum loosely to prevent postoperative intestinal adhesions, and placed a Robinson drain right and left of lateral pelvic spaces. The pelvis was thoroughly irrigated with saline containing antibiotics. Results FIGURE 1. After transection of the right uterine artery, the uterine artery was grasped and ligated with loop at its origin. The ureter is visible on the left side of the hypogastric artery, and obturator nerve on the right side of the picture. We successfully completed 18 cases of laparoscopic radical hysterectomy with pelvic lymphadenectomy. Patients' characteristics are shown in Table 1. Fifteen women (83%) had squamous cell carcinoma and 3 (17%) had adenocarcinoma of the cervix; 6 (33%) had microcarcinoma and 12 (67%) had stage Ibl disease. Mean operating time was approximately minutes (range min). Blood loss averaged ml (range ml; Table 2). Five patients 413

4 Laparoscopic Surgery for Early, Invasive Cervical Carcinoma Kim and Moon TABLE 1. Patient Characteristics Mean SD Age Disease Pathologic (yrs) Parity Stage Diagnosis 47 4 Ibl Squamous large cell 70 6 Ibl Squamous large cell 65 3 Ibl Squamous large cell 62 6 Ibl Squamous large cell 32 1 la2 Squamous large cell 52 4 Ibl Squamous large cell 44 3 Ibl Squamous large cell 44 2 Ibl Adenocarcinoma 46 4 la2 Squamous large cell 59 6 la2 Squamous large cell 29 3 Ibl Adenocarcinoma 40 3 Ibl Squamous large cell 65 4 la2 Squamous large cell 43 2 Ibl Adenocarcinoma 39 2 Ibl Squamous large cell 48 3 Ibl Squamous large cell 38 3 la2 Squamous large cell 37 2 la2 Squamous large cell (28%) required transfusions for blood loss over 1000 ml. The average number of pelvic lymph nodes obtained was (range 14-40). Lymph nodes were positive in two patients (11%), 2/23 in one and 4/14 in the other. Average specimen weight was g (range g). Surgical margins were clear in all patients. The average vaginal margin was 2.03 cm (range cm). Average postoperative stay was days. There were no major intraoperative complications involving injury to main blood vessels and nerves, bowel, bladder, or ureters. None of the procedures was converted to laparotomy. Postoperatively, two women (11%) had vault infections. Five patients (28%) had transient bladder dysfunction, which was defined as voiding difficulty, with more than 100 ml residual urine after removal of the indwelling Foley catheter on the fifth postoperative day; they all returned to normal within 2 weeks. Lymphocele, pelvic hematoma, and bowel obstructions were not observed. One woman (5.6%) experienced ureterovaginal fistula on postoperative day 14, and one developed postirradiation ureteral stricture. Two women (11%) had febrile morbidity above 38 ~ C, and one had transient right leg edema. The six women who required radiotherapy began treatment within 15 days postoperatively. Discussion Many authors assert that laparoscopic-assisted vaginal hysterectomy (LAVH) is associated with lower postoperative complication rates, more rapid recovery, greater cost effectiveness, and many other advantages compared with abdominal hysterectomy. In fact, before laparoscopic lymph node dissection, the major disadvantage of radical vaginal hysterectomy was the need to make an abdominal incision for pelvic lymphadenectomy and the Schuchardt perineal incision. Some authors reported a series of select cases of radical hysterectomy done solely by laparoscopic technique. 17 The main difficulty with vaginal radical hysterectomy is identifying and unroofing the ureter. Laparoscopy affords the surgeon an adequate, minimally invasive alternative to standard open pelvic lymphadenectomy. Vaginal incision, transection of uterosacral ligaments, and transection of cardinal ligaments at the proper level are the only vaginal tasks. Laparoscopic assistance may reduce the risk of injury to ureters and may avoid perineal incision. 13 The laparoscopic procedure is more limited than the open abdominal procedure. Laparoscopic vision is downward from the umbilicus, so to obtain good exposure, it is preferable first to divide infundibulopelvic and round ligaments, which are the nearest obstacles to vision. In gaining access to retroperitoneum, however, we believe there is no difference between dividing the round ligament first or the infundibulopelvic ligament first. Early in our series we performed some pelvic lymph node dissections before radical hysterectomy, and in others vice versa. In general, we preferred to do the pelvic lymph node dissection first, unless the uterus was so large that it obscured the visual field, as it provides an excellent view of pelvic anatomy. When the uterus is pushed cephalad by an assistant for developing paravesical and pararectal spaces, the ligaments and the space between them can be easily distinguished due to differences in texture of the tissues. Those spaces are composed of areolar tissue, and after they are divided they are developed without much difficulty. It is easier to identify ureters by laparoscopy than by unassisted vision, so the special anatomic landmark 414

5 November 1998, Vd. 5, No. 4 The Journal of the American Association of Gynecologic Laparoscopists TABLE 2. Results of Laparoscopic Radical Hysterectomy Mean SD Operating EBL Uterine No. Lymph Length of Time (min) (ml) weight (gm) Nodes Postoperative Stay RT a a EBL = estimated blood loss; RT = radiation therapy. alymph node involvement. is unnecessary. Even in obese patients, the ureter's course in the lower part of ovarian vessels (on the pelvic brim) is easily recognized. We cut retropelvic peritoneum, recognize the ureter, and separate it in accordance with its course. During ureteral unroofing, it is advantageous to use bipolar forceps with relatively thin tips. To prevent ureteral damage, the ureteral tunnel was widened by separating the jaws of grasping forceps, and the upper portion of the tunnel was lifted up and the irrigator was placed on the ureter, preventing bipolar current from flowing into the ureter directly. To prevent heat damage to the ureter, saline irrigation was performed during bipolar coagulation. The uterosacral ligament can be divided easily with bipolar coagulation without bleeding. However, if it is too thick or too deep, endosuture ligation saves time. We have never used laser or monopolar current in any laparoscopic surgery, but only bipolar current selectively. As we are familiar with endosuture, we prefer it when cutting and coagulating with bipolar forceps are expected to take a long time. We transected the cardinal ligament laparoscopically in most cases. Paravesical and pararectal spaces were developed laparoscopically, and both cardinal ligaments were endosutured 2 or 3 cm lateral to the cervix and tra_nsected laparoscopically with scissors and bipolar forceps. In some cases the cardinal ligaments were divided with EndoGIA staplers. We do not agree with the claim that there is no difference in development of pelvic adhesions whether pelvic peritoneum is closed or open. We always close the pelvic peritoneum after hysterectomy. During laparoscopic radical hysterectomy, we approximate anterior and posterior pelvic peritoneum loosely, and place Robinson drains in order to recognize postoperative bleeding from the pelvic cavity early and prevent formation of lymphocele. It is not clear, based on our experience, that comparison of laparoscopic radical hysterectomy versus laparoscopic-assisted radical vaginal hysterectomy is important. We believe the former approach has more advantages. It provides better vision, and thus may enhance precise dissection of cardinal and uterosacral ligaments, be less likely to expose the patient to infection, and result in less blood loss. Some, however, maintain that the laparoscopic approach is more difficult and takes longer to perform. ~3 415

6 Laparoscopic Surgery for Early, Invasive Cervical Carcinoma Kim and Moon We believe laparoscopic radical hysterectomy should be developed, rather than laparoscopic-assisted radical vaginal hysterectomy, just as LAVH has given way to laparoscopic hysterectomy and total laparoscopic hysterectomy. The more vaginal the procedures, the more possible ascending infection may be. The success of the operations depends on technique, however, so surgeons must choose the approach with which they are familiar, whether it be vaginal or laparoscopic. Mean estimated blood loss in our series, ml, compares favorably with that reported with abdominal hysterectomy. 2~ We did not perform paraaortic lymph node dissection, but dissected up to the level of the common iliac lymph node, because the prevalence of paraaortic nodal metastasis in patients with early invasive cervical carcinoma is less than 1%,r0 a risk that does not match the risk of paraaortic node sampling. However, paraaortic node dissections were performed in a woman with positive common iliac lymph nodes and in two with adenocarcinoma. The average length of combined procedures was approximately 6 hours, which is longer than in other reports. ~3-~8 This may be because the time associated with adhesiolysis and frozen section analysis of pelvic lymph nodes was included, and we performed more laparoscopic procedures. However, operating time is also lengthened by equipment failure and lack of experience. Therefore, it should be shorter with surgeons and operating teams experienced in the surgery. Complications associated with laparoscopic radical hysterectomy with pelvic lymphadenectomy seem acceptable. The only significant complication in this series was ureterovaginal fistula, which required readmission. We believe this complication is not completely avoidable. Our patients had relatively long hospital stays because of regulations concerning the medical insurance system in our country and because they were cared for under our department's guidelines. According to these guidelines, women who undergo abdominal radical hysterectomy remain in the hospital for more than 10 days and the indwelling Foley catheter is in place for 5 days. Even more important than hospital stay is length of recovery, which in these women was as rapid as that after LAVH, despite the longer operating time. Conclusion Successful laparoscopic radical hysterectomy depends on several factors. It is important that the surgeon be a well-trained laparoscopist and oncologist, and it is mandatory that proper equipment be available. It is advantageous to be able to perform the procedure at a single institution with operating room staff who are familiar with the technique. Patient selection and preparation also are extremely important. Obesity, inadequate bowel preparation, and intraperitoneal adhesions make the operation more difficult, although in our experience they are not total contraindications. References 1. Querleu D, Leblanc E, Castelain B: Laparoscopic pelvic lymphadenectomy in the staging of early cervical carcinomas. Am J Obstet Gynecol 164: , Childers JM, Brzechaffa PR, Hatch KD, et al: Laparoscopically assisted surgical staging (LASS) of endometrial cancer. Gynecol Oncol 51:33-38, Querleu D, Leblanc E: Laparoscopic infrarenal paraaortic lymph node dissection for restaging of carcinoma of the ovary or fallopian tube. Cancer 73: , Guazzoni G, Montorsi F, Bergamaschi F: Open surgical revision of laparoscopic pelvic lymphadenectomy for staging of prostate cancer: The impact of laparoscopic learning curve. J Urol 151: , Lemer SE, Chamberlin JW, Fleischmann J: Combined laparoscopic pelvic lymph node dissection and modified belt radical perineal prostatectomy for localized prostatic adenocarcinoma. Urology 43: , Childers JM, Hatch K, Surwit EA: The role of laparoscopic lymphadenectomy in the management of cervical carcinoma. Gynecol Oncol 47:38-43, Kadar N: Laparoscopic pelvic lymphadenectomy in obese women with gynecologic malignancies. J Am Assoc Gynecol Laparosc 2: , Gershman A, Daykhovsky L, Chandra M: Laparoscopic pelvic lymphadenectomy. J Laparoendosc Surg 1:63-458, Fowler JM, Carter JR, Carlson JW: Lymph node yield from laparoscopic lymphadenectomy in cervical cancer: A comparative study. Gynecol Oncol 51: ,

7 November 1998, Vol. 5, No. 4 The Journal of the American Association of Gynecologic Laparoscopists 10. Pastner B, Sedlacek TV, Lovecchio JL: Paraaortic node sampling in small (3 cm or less) stage Ib invasive cervical cancer. Gynecol Oncol 44:53-54, Childers JM, Hatch K, Surwit EA: Laparoscopic paraaortic lymphadenectomy in gynecologic malignancies. Obstet Gynecol, in press 12. Herd J, Fowler JM, Shenson D: Laparoscopic paraaortic lymph node sampling: Development of a technique. Gynecol Oncol 44: , Querleu D: Laparoscopically assisted radical vaginal hysterectomy, Gynecol Oncol 51: , Nezhat CR, Burrell MO, Nezhat FR: Laparoscopic radical hysterectomy with paraaortic and pelvic node dissection. Am J Obstet Gynecol 166: , Canis M, Mage G, Wattiez A: Vaginally assisted laparoscopic radical hysterectomy. J Gynecol Surg 8: , Kadar N, Reich H: Laparoscopically assisted radical Schauta hysterectomy and bilateral laparoscopic pelvic lymphadenectomy for the treatment of bulky stage Ib carcinoma of the cervix, Gynaecol Endosc 2: , Spirtos NM, Schaerth JB, Kimball RE, et al: Laparoscopic radical hysterectomy (type Ill) with aortic and pelvic lymphadenectomy. Am J Obstet Gynecol 174: , Schneider A, Possover M, Kamprath S, et al: Laparos~ copy-assisted radical vaginal hysterectomy modified according to Schauta-Stoeckel. Obstet Gynecol 88: , Kim DH, Lee JC, Bae DH: Clinical analysis of pelviscopic classic intrafascial Serum hysterectomy. J Am Assoc Gynecol Laparosc 2: , Mikuta JJ, Giuntoli RL, Rubin E, et al: The "problem" radical hysterectomy. Am J Obstet Gyneco1128:11%127, Orr JW, Orr PJ, Holimon JL: Radical hysterectomy: Does the type of incision matter? Am J Obstet Gynecol 173: , Artman LE, Hoskins WJ, Bibro MC, et al: Radical hysterectomy and pelvic lymphadenectomy for stage Ib carcinoma of the cervix: 21 year experience. Gynecol Oncol 28:8-13, Massi G, Savino L, Susini T: Schauta-Amreich vaginal hysterectomy and Wertheim-Meigs abdominal hysterectomy in the treatment of the cervical cancer: A retrospective analysis. Am J Obstet Gynecol 168: , Barton DP, Cavanagh D, Roberts WS, et al: Radical hysterectomy for treatment of cervical cancer: A prospective study of two methods of closed suction drainage. Am J Obstet Gynecol 166: ,