GYNECOLOGIC
ONCOLOGY
44,
271-276 (1992)
Laparoscopic Para-Aortic Lymph Node Sampling: Development of a Technique JAY HERD, M.D., Division
of Gynecologic
JEFFREY M. FOWLER, M.D., Oncology,
Department
DAVID SHENSON, SILAS LACY, AND F. J. MONTZ,
of Obstetrics & Gynecology, Los Angeles, California
University 90024
of California
Los Angeles-CHS
M.D.‘,* Room
24-127,
Received August 15, 1991
Clinical staging is woefully inadequatein predicting lymph nodemetastasisand dictating prudent radiation ports in women undergoingtherapy for cervical malignancies.Surgical staging with evaluationof the precavallymph nodes(via a transperitoneal or extraperitonealapproach)hasassociatedratesof complication that are felt by many clinicians to be excessive.A laparoscopic approach could avoid many of these complicationswhile still obtainingvaluableinformation r?gardingspreadof disease.After experimentationwith different forms of laparoscopicapproaches in the swinemodel, we have found an easyand efficient way to visualize and samplethe i&arena1 precaval lymph nodes.The laparoscopeis placed suprapubically, and accessoryoperating trocars are placed suprapubicallyand periumbilically. Laparostopically placed suturesare placedon the incisedposterior peritoneum. The sutures exit through the periumbilical trocars. When drawn taut, the suturespull the peritoneumin a cephalad and ventral fashion so as to act as a tent, retracting the intraperitoneal contentsand exposingthe precaval area. Using this techniquein the swine,we were ableto remove88% of all rightsidecommoniliac and paraaorticnodeslaparoscopicallywithout injuring surroundingstructures. Evaluation of this techniquein humansshouldbe considered. o 1~ Academic PRSS, IX.
INTRODUCTION
Cervical malignancies are officially staged clinically, though many authors believe that surgical staging is beneficial in tailoring therapy, therefore increasing rates of cure [1,2]. Traditionally, this surgical staging has included evaluation of the precaval lymph nodes via a transperitoneal or extraperitoneal approach. Both approaches have associated complications that are felt by many clinicians to be excessive. A laparoscopic approach could ’ Sponsored by American Award 90-0143. * To whom
reprint
requests
Cancer should
Society
Career
Development
be addressed.
avoid many of these complications while still obtaining valuable information regarding the spread of disease. Laparoscopic pelvic lymph node sampling/dissection has been described and at present is widely utilized by urologic surgeons [3]. Laparoscopic paraaortic lymph node sampling has not yet been described or used in this setting. This disuse is predominantly due to technical difficulties in the visualization of precaval and retroperitoneal structures. At laparotomy, retracting of the bowel and access to the precaval space is accomplished with ease. This is not so readily accomplished using an endoscopic approach. We herein describe a method that we have developed for laparoscopically sampling the paraaortic lymph nodes, visualizing intraperitoneal contents, obtaining peritoneal washings and intraperitoneal biopsies, and thus avoiding a large abdominal wall incision. MATERIALS
AND METHODS
Female swine were utilized for these investigations. All animals were housed at the UCLA vivarium. This facility is fully accredited by the American Association for Animal Laboratory Care. Animal care and surgical procedures were performed in accordance with the standards described in the National Institutes of Health’s Guide for Care and Use of Laboratory Animals, in compliance with the Federal Animal Welfare Act. Mean animal weight was 21 kg. After a lo-day quarantine during which the animals were allowed access to water and chow ad libitum, animals were fasted for 24 hr. The swine were preanesthesized with ketamine (10 mg/kg). Following endotracheal intubation and induction of systemic halothane anesthesia, animals were placed in the supine position. Their anterior abdominal walls were shaved and prepped with an iodine solution. Following a
271 lWO-82W92 $1.50 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
272
HERD
ET AL.
VIDEO hmt4m-m
Grasper
FIGURE
cut-down procedure, an arterial cannula was inserted into the swine’s carotid artery to obtain continuous blood pressure and pulse monitoring throughout the procedure. A Verres needle was inserted into the peritoneal cavity periumbilically. After verification of needle placement, a pneumoperitoneum was achieved with a CO2 insufflator at a flow rate of 1 liter/min and a pressure of 10 mm Hg. After induction of pneumoperitoneum, the Verres needle was removed. A small stab incision was made above the pubic symphysis and a lo-mm trocar was inserted into the peritoneal cavity through the stab incision. The operating laparoscope was inserted into the lo-mm Surgiport (U.S. Surgical), and continuous, pressure-controlled CO2 flow was restored. The light source and video camera were attached and the peritoneal cavity and structures were carefully examined utilizing a video monitor. To facilitate small and large bowel mobilization and retraction, the animals were placed in the Trendelenburg position (30”-45”). Under laparoscopic visualization, three additional 5mm trocars were placed at the level of the umbilicus (Fig. 1). Using a probe, the small bowel was mobilized away
1
from the bifurcation of the right common iliac vessels. The posterior peritoneum was incised using hooked scissors, lateral to the right common iliac vessels. These scissors were introduced into the peritoneal cavity through the left lateral 5-mm trocar. The incision was extended along the right common iliac toward the bifurcation of the aorta. This incision was made at the level of the sacrum. An O-Chromic Endoloop (Ethicon) was passed through the right-sided 5-mm Surgiport, and atraumatic forceps were inserted through the loop and used to grasp the cephalad margin of the incised peritoneum, lateral to the bifurcation of the right common iliac vessels. The loop was tied and the forceps were removed. The peritoneum was grasped a second time, medial to the bifurcation with an Endoloop through the left-sided accessory trocar. The procedure was repeated through the umbilical trocar to grasp the pelvic peritoneum a third time at the apex of the incision. The three Endoloops were used to lift the peritoneum off the dorsal aspect of the peritoneal cavity. This drape of peritoneum effectively acted as a retractor, elevating the small bowel, distal ileum, cecum, and ascending colon (Fig, 2). The Endoloops exited the
LAPAROSCOPIC
273
LYMPH NODE SAMPLING
FIGURE
anterior abdominal wall through the trocars and were stabilized with hemostats at the desired tension. Though the sutures exited through the trocars, instruments could easily be placed and manipulated through the trocars. Tissue dissection was performed using aquadissection, scissors, and the blunt dissecting probe. Dissection started at the bifurcation of the common iliac vessels and extended to the renal vein. A second lo-mm Surgiport trocar was inserted suprapubically, through which Endoclips (specially designed titanium clips applied with a multiuse, preloaded, disposable applicator) were placed to clip lymphatics and vessels prior to transection and through which lymph nodes were retrieved. Additional Endoloops were attached as needed to assist with the tenting of the peritoneum as dissection continued cephalad along the common iliac vessels. Upon completion of the procedure, the Surgiports and instruments were removed. The swine then underwent a celiotomy through a vertical midline incision. The peritoneal cavity and its contents were examined for damage and complications due to the laparoscopic procedure. Using a transperitoneal approach, any remaining para-aortic and pelvic lymph
2
nodes were removed and kept separate from these removed laparoscopically. The animals were euthanized and closed en masse with running 0 PGA suture. RESULTS We performed a total of six surgeries in an attempt to develop a technique of performing laparoscopic pelvic and paraaortic lymphadenectomy . The first four surgeries involved the actual development of the technique. The final two surgeries involved the application and testing of this technique. In case 1, an attempt was made to perform the paraaortic lymph node sampling with the laparoscope and accessory trocars placed in “traditional” locales (subumbilically and suprapubically, respectively). At the same setting an attempt was made to operate directly through the peritoneum without employing a specific means of elevating the extensive porcine bowel out of the operating field. With the laparoscope placed in the subumbilical position there was very little mobility allowed and visualization was suboptimal. Similarly, placing the operating instruments through suprapubic trocars mean that they
274
HERD
had to travel a greater distance to reach the field of operation. Often the instruments would seem to “disappear” in the loops of small bowel, which led to a high level of surgeon frustration. Superimposed on these tactical problems was the voluminous bowel, which even the steepest amounts of Trendelenburg position and aggressive attempts at manipulation with blunt probes were unable to mobilize out of the surgical field. The procedure was aborted after over 210 min of frustration. In case 2 an attempt was made to perform the entire procedure via a retroperitoneal approach. The Verres needle was placed 5 cm below the costal margin through the right flank. An attempt to keep the Verres needle in the retroperitoneal space and out of the peritoneal cavity was, to the surgeons amazement, successful. After a pneumoretroperitoneum was obtained, the lo-mm trocar was placed at the site of the Verres needle, with two 5-mm trocars being placed, under direct visualization, 10 cm caudally. Initial dissection went smoothly, but at the psoas muscle the hog peritoneum was adherent to the tendon. Attempts to free the peritoneum, so as to continue the retroperitoneal dissection, were unsuccessful. A series of large rents in the peritoneum occurred with a total escape of the carbon dioxide into the peritoneal cavity and loss of the pneumoretroperitoneum. After complete decompression of the pneumoperitoneum, a similar attempt was made to establish a pneumoretroperitoneum on the contralateral (left) side. This was unsuccessful, and after over 180 min the procedure was aborted with, again, very frustrated surgeons. In case 3, the decision was made to place the Verres needle and trocars as described under Materials and Methods. It was noted that placing them in this “inverted” fashion allowed excellent visualization of the middle and upper abdomen and close proximity to the para-aortic region. The problem now was mobilization of the bowel. With the animal in steep Trendelenburg position as described above, an attempt was made to keep the bowel packed away from the right common iliac vessels and overlying peritoneum. A sling of Gore surgical membrane was fashioned and an attempt made to hold it in place with pickups placed through the lateral 5-mm trocars. Though this allowed initial adequate visualization so that a peritoneal incision could be made and dissection undertaken, over time, with the ventilation of the animal, the small bowel would slip under the sling and obscure vision. After the Gore surgical membrane sling was removed, an attempt to pack the bowel into the upper abdomen using plastic field drapes, which were easily placed into the peritoneal cavity through the lo-mm trocars, was made. Unfortunately, these were similarly ineffective as they would be displaced by the ventilatory effort of the animal and obscure the operative field, which was partially responsible for an inadvertent arteriotomy
ET AL.
and massive intraperitoneal hemorrhage that could not be controlled laparoscopically. The third case was aborted after over 240 min. In our fourth case, we had our first experience using what became, with slight modifications, our final technique. Trocars and instruments were placed as described under Materials and Methods. Unfortunately, the incision in the posterior peritoneum was placed too far lateral to the right common iliac vessels and, though an adequate drape of peritoneum was obtained, the surgeons were disoriented and an attempt at a laperoscopic nephrectomy transpired. The procedure was terminated 100 min after initiation because of uncontrollable hemorrhage from the partially resected kidney. We did learn, however, the limitations of the microtip cautery as regards any visible vessels. This experience led us to select prophylactic clipping of visualized vessel over attempts to control bleeding/oozing with cautery once it had occurred. However, we continued to use cautery to perform dissection through tissue devoid of appreciated vessels. In case 5, a total of six lymph nodes were removed (two high common and four para-aortic) from the precaval space without vascular, intestinal, or urinary tract injury. Subsequent laparotomy revealed an additional two lymph nodes (one low common and one lateral to the vena cava). Hence, we retrieved 75% of the infrarenal precaval lymph nodes. The total operative time for this procedure was 120 min. In case 6, a total of eight lymph nodes were removed from the right side (two high common iliac and six paraaortic). Subsequent laparotomy failed to demonstrate any residual high common or precaval lymph nodes. In addition, no vascular, intestinal, or urinary tract injuries were noted. The total operative time for this procedure was 95 min. DISCUSSION Definitive staging of cervical cancer has prognostic significance as well as therapeutic implications. Cervical cancer is still clinically staged by FIGO, although clinical staging underestimates the extent of disease in 29-48% of cases [4-61. “Extended clinical staging” using noninvasion tools has been used to improve diagnostic accuracy [7]. These studies (computed tomography, magnetic resonance imaging, ultrasonography, lymphangiography, etc.) are useful if positive findings are noted. However, they are severely limited in their ability to detect microscopic metastasis [B]. Fine-needle aspiration (FNA) directed by these noninvasive studies can predict disease if positive (low false-positive rate), but has a significant false-negative rate of 14%. A further confounding problem is that 10% of specimens are unsatisfactory, giving FNA a diagnostic accuracv. of 74% 191. La
LAPAROSCOPIC
LYMPH
Due to the inaccuracies of traditional and extended clinical staging, microscopic metastases to the paraaortic lymph nodes are commonly unappreciated. Obviously, if treatment is directed solely at pelvic disease and the paraaortic region is pathologically involved but not included in the radiation port, failure is inevitable [8]. Multiple studies have evaluated the prognostic and therapeutic benefits of preradiation staging laparotomy to assess peritoneal cytology and paraaortic lymph node status in patients with cervical cancer [4-6,9-161. The underlying theme in these studies is that the status of the paraaortic lymph nodes is a statistically significant prognostic factor of patient survival [lo]. Sampling of the pelvic and paraaortic nodes has been performed at celiotomy (transperitoneal) and via an extraperitoneal approach [ 15-171. The benefits of the transperitoneal approach are the abilities to assess the pelvic and intraabdominal findings, obtain washings, and take directed biopsies. The complications associated with the transperitoneal approach followed by radiation therapy are those of bowel adhesions and radiation damage that can affect up to 30% of patients [ 151. The benefits of the retroperitoneal approach are less bowel manipulation and adhesion formation and, allegedly, fewer postradiation bowel complications (2.5%) [15]. Th e mean blood loss in one study of extraperitoneal bilateral pelvic and paraaortic lymphadenectomy was 300 cc, and the mean postoperative hospital stay was 8 days [15]. A method for sampling the pelvic and paraaortic lymph nodes that would minimize adhesion formation, incision length, and hospital stay, without compromising the “gold standard” of histologic confirmation, would be ideal. Operative laparoscopy has allowed the gynecologist to perform, in an outpatient setting, operations that previously required laparotomy and hospitalization. Although no prospective randomized trials of laparoscopy versus laparotomy with respect to adhesion formation have been done in the human, there is evidence in the animal model that laparoscopy causes significantly less postoperative adhesions than does laparotomy [ 181. Laparoscopic pelvic lymphadenectomy has been studied in both the swine and the human [2,19,20]. These papers fail to address the status of the paraaortic nodes, data which are essential in the making of rational treatment decisions. We have developed a method for laparoscopically sampling the precaval lymph nodes. Prior to the development of the Endoloop retraction technique we attempted other methods of visualizing the paraaortic area. A retroperitoneal (flank) approach with insufflation and laparoscopic insertion retroperitoneally afforded poor visibility and inability to identify vital structures. Different forms of retractors were used through the laparoscope including plastic drapes and Gore surgical membrane sheet. These were ineffectual in keeping the bowel out of the operating field.
275
NODE SAMPLING
The swine model presented unique problems that are not met in the human: the small bowel in the swine is abundant, complicating any attempt at retraction of it, and the posterior parietal peritoneum is weak, such that the Endoloops would occasionally pull through the peritoneum, requiring replacement. With video monitoring and knowledgeable assistants, the dissection of both the common iliac and the paraaortic nodes can be done under the magnification of the laparoscope with minimal bleeding due to the pneumoperitoneum and use of aquadissection, cautery, and Endoclips when necessary. The obturator and hypogastric lymph nodes are probably best sampled with the laparoscope in the traditional periumbilical position. Modification of trocar insertions and operating positioning is possible, and further retraction of the “peritoneal tent” can also be accomplished by tilting the distal tip of the 5-mm sleeve (Endoloops already attached) cephalad. The data so far on the sensitivity and specificity of laparoscopic pelvic lymphadenectomy with its low associated morbidity have made it a very attractive form of surgical staging [19]. Hospitalization is usually only overnight and the time interval from sampling to institution of radiation therapy is minimal. It appears that laparoscopic pelvic and paraaortic lymph node sampling offers the surgeon a unique port of entry for performing the same surgical procedure accomplished utilizing an extraperitoneal or transperitoneal approach, while lessening the risk of negative side effects. Thorough human studies will need to be performed before either traditional means of surgical staging are abandoned or universal laparoscopic node sampling is recommended for all patients with cervical cancer. REFERENCES 1. Jones, W. B. Surgical approaches for advanced or recurrent cancer of the cervix, Cancer 60, 2094-2103 (1987). 2. Averette, H. E., Donato, D. M., Lovecchio, J. L., and Seven, B. Surgical staging of gynecologic malignancies, Cancer 60, 2010-2020 (1987). 3. Shuessler, W. W., Vancaillie, T. G., Reich, H., and Griffith, D. P. Transperitoneal endosurgical lymphadenectomy in patients with localized prostate cancer, J. Ural. 145, 988-991 (1991). 4. Lagasse, L. D., Creasman, W. T., Shingleton, H. M., Ford, J. H., and Blessing, J. A. Results and complications of operative staging in cervical cancer: Experience of the Gynecologic Oncology Group, Gynecol. Oncol. 9, 90-98 (1980). 5. Averette, H. E., Dudan, R. C., and Ford, J. H. Exploratory celiotomy for surgical staging of cervical cancer, Am. 1. Obstet. Gynecol. W, 1090-1096 (1972). 6. LaPolla, J. P., Schlaerth, J. B., Gaddis, O., and Morrow, C. P. The influence of surgical staging on the evaluation and treatment of patients with cervical carcinoma, Gynecol. Oncol. 24, 194-206 (1986). 7. Berek, J. S., and Hacker, N. F. Practical gynecologic oncology. Williams & Wilkins, Baltimore, pp. 241-283 (1989).
276
HERD ET AL.
8. Heaps, J. M., and Berek, J. S. Surgical staging of cervical cancer. Clin. Obsret. Gynecol. 33(4), 852-862 (1990). 9. McDonald, T. W., Morley, G. W., Choo, Y. C., Shields, J. J., Cordoba, R. B., and Naylor, B. Fine needle aspiration of paraaortic and pelvic lymph nodes showing lymphangiographic abnormalities, Obstet. Gynecol. 33, 71-75 (1983). 10. Podczaski, E. S., Palombo, C., Manetta, A., Andrews, C., Larson, J., De Geest, K., and Mortel, R. Assessment of pretreatment laparotomy in patients with cervical carcinoma prior to radiotherapy, Gynecol. Oncol. 33, 71-75 (1989). 11. Potish, R. A., Twiggs, L. B., Okagaki, T., Prem, K. A., and Adcock, L. L. Therapeutic implications of the natural history of advanced cervical cancer as defined by pretreatment surgical staging, Cancer
56, 956-960
15.
16.
17.
(1985).
12. Rubin, S. C., Brookland, R., Mikuta, J. J., Mangan, C., Sutton, G., and Danoff, B. Para-aortic nodal metastasis in early cervical carcinoma: Long term survival following extended-field radiotherapy, Gynecol. Oncol. 18, 213-217 (1984). 13. Tewfik, H. H., Buchsbaum, H. J., Latourette, H. B., Lifshitz, S. G., and Tewfik, F. A. Para-aortic lymph node irradiation in carcinoma of the cervix after exploratory laparotomy and biopsyproven positive arotic nodes, Int. J. Radial. Oncoi. Biof. Phys. 8, 13-18 (1982). 14. Lovecchio, J. L., Averette, H. E., Donato, D., and Bell, J. 5-year
18.
19.
20.
survival of patients with periaortic nodal metastasis in clinical Stage IB and IIA cervical carcinoma, Gynecol. Oncol. 34, 43-45 (1989). Berman, M. L., Lagasse, L. D., Watring, W. G., Ballon, S. C., Schlesinger, R. E., Moore, J. G., and Donaldson, R. C. The operative evaluation of patients with cervical carcinoma by an extraperitoneal approach, Obstet. Gynecol. SO, 658-664 (1977). Ballon, S. C., Berman, K. L., Lagasse, L. D., Petrilli, E. S., and Castaldo, T. W. Survival after extraperitoneal pelvic and paraaortic lymphadenectomy and radiation therapy in cervical carcinoma, Obstet. Gynecol. 57, 90-95 (1981). Berman, M. L., Lagasse, L. D., Ballon, S. C., Watring, W. G., and Tesler, A. Modification of radiation therapy following operative evaluation of patients with cervical carcinoma, Gynecol. Oncof. 6, 328-332 (1978). Luciano, A. A., Maier, D. B., Koch, E. I., Nulsen, J. C., and Whitman, G. F. A comparative study of postoperative adhesions following laser surgery by laparoscopy versus laparotomy in the rabbit model, Obstet. Gynecol. 74, 220-224 (1989). Querleu, D., Leblanc, E., and Castelain, B. Laparoscopic pelvic lymphadenectomy in the staging of early carcinoma of the cervix, Am. J. Obstet. Gynecol. 164, 579-581 (1991). Gersham, A., Daykhovsky, L., Chandra, M., Danoff, D., and Grundfes, W. S. Laparoscopic pelvic lymphadenectomy, 1. Laparoendoscopic Surg. 1, 63-68 (1990).
ONCOLOGY
44,
271-276 (1992)
Laparoscopic Para-Aortic Lymph Node Sampling: Development of a Technique JAY HERD, M.D., Division
of Gynecologic
JEFFREY M. FOWLER, M.D., Oncology,
Department
DAVID SHENSON, SILAS LACY, AND F. J. MONTZ,
of Obstetrics & Gynecology, Los Angeles, California
University 90024
of California
Los Angeles-CHS
M.D.‘,* Room
24-127,
Received August 15, 1991
Clinical staging is woefully inadequatein predicting lymph nodemetastasisand dictating prudent radiation ports in women undergoingtherapy for cervical malignancies.Surgical staging with evaluationof the precavallymph nodes(via a transperitoneal or extraperitonealapproach)hasassociatedratesof complication that are felt by many clinicians to be excessive.A laparoscopic approach could avoid many of these complicationswhile still obtainingvaluableinformation r?gardingspreadof disease.After experimentationwith different forms of laparoscopicapproaches in the swinemodel, we have found an easyand efficient way to visualize and samplethe i&arena1 precaval lymph nodes.The laparoscopeis placed suprapubically, and accessoryoperating trocars are placed suprapubicallyand periumbilically. Laparostopically placed suturesare placedon the incisedposterior peritoneum. The sutures exit through the periumbilical trocars. When drawn taut, the suturespull the peritoneumin a cephalad and ventral fashion so as to act as a tent, retracting the intraperitoneal contentsand exposingthe precaval area. Using this techniquein the swine,we were ableto remove88% of all rightsidecommoniliac and paraaorticnodeslaparoscopicallywithout injuring surroundingstructures. Evaluation of this techniquein humansshouldbe considered. o 1~ Academic PRSS, IX.
INTRODUCTION
Cervical malignancies are officially staged clinically, though many authors believe that surgical staging is beneficial in tailoring therapy, therefore increasing rates of cure [1,2]. Traditionally, this surgical staging has included evaluation of the precaval lymph nodes via a transperitoneal or extraperitoneal approach. Both approaches have associated complications that are felt by many clinicians to be excessive. A laparoscopic approach could ’ Sponsored by American Award 90-0143. * To whom
reprint
requests
Cancer should
Society
Career
Development
be addressed.
avoid many of these complications while still obtaining valuable information regarding the spread of disease. Laparoscopic pelvic lymph node sampling/dissection has been described and at present is widely utilized by urologic surgeons [3]. Laparoscopic paraaortic lymph node sampling has not yet been described or used in this setting. This disuse is predominantly due to technical difficulties in the visualization of precaval and retroperitoneal structures. At laparotomy, retracting of the bowel and access to the precaval space is accomplished with ease. This is not so readily accomplished using an endoscopic approach. We herein describe a method that we have developed for laparoscopically sampling the paraaortic lymph nodes, visualizing intraperitoneal contents, obtaining peritoneal washings and intraperitoneal biopsies, and thus avoiding a large abdominal wall incision. MATERIALS
AND METHODS
Female swine were utilized for these investigations. All animals were housed at the UCLA vivarium. This facility is fully accredited by the American Association for Animal Laboratory Care. Animal care and surgical procedures were performed in accordance with the standards described in the National Institutes of Health’s Guide for Care and Use of Laboratory Animals, in compliance with the Federal Animal Welfare Act. Mean animal weight was 21 kg. After a lo-day quarantine during which the animals were allowed access to water and chow ad libitum, animals were fasted for 24 hr. The swine were preanesthesized with ketamine (10 mg/kg). Following endotracheal intubation and induction of systemic halothane anesthesia, animals were placed in the supine position. Their anterior abdominal walls were shaved and prepped with an iodine solution. Following a
271 lWO-82W92 $1.50 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
272
HERD
ET AL.
VIDEO hmt4m-m
Grasper
FIGURE
cut-down procedure, an arterial cannula was inserted into the swine’s carotid artery to obtain continuous blood pressure and pulse monitoring throughout the procedure. A Verres needle was inserted into the peritoneal cavity periumbilically. After verification of needle placement, a pneumoperitoneum was achieved with a CO2 insufflator at a flow rate of 1 liter/min and a pressure of 10 mm Hg. After induction of pneumoperitoneum, the Verres needle was removed. A small stab incision was made above the pubic symphysis and a lo-mm trocar was inserted into the peritoneal cavity through the stab incision. The operating laparoscope was inserted into the lo-mm Surgiport (U.S. Surgical), and continuous, pressure-controlled CO2 flow was restored. The light source and video camera were attached and the peritoneal cavity and structures were carefully examined utilizing a video monitor. To facilitate small and large bowel mobilization and retraction, the animals were placed in the Trendelenburg position (30”-45”). Under laparoscopic visualization, three additional 5mm trocars were placed at the level of the umbilicus (Fig. 1). Using a probe, the small bowel was mobilized away
1
from the bifurcation of the right common iliac vessels. The posterior peritoneum was incised using hooked scissors, lateral to the right common iliac vessels. These scissors were introduced into the peritoneal cavity through the left lateral 5-mm trocar. The incision was extended along the right common iliac toward the bifurcation of the aorta. This incision was made at the level of the sacrum. An O-Chromic Endoloop (Ethicon) was passed through the right-sided 5-mm Surgiport, and atraumatic forceps were inserted through the loop and used to grasp the cephalad margin of the incised peritoneum, lateral to the bifurcation of the right common iliac vessels. The loop was tied and the forceps were removed. The peritoneum was grasped a second time, medial to the bifurcation with an Endoloop through the left-sided accessory trocar. The procedure was repeated through the umbilical trocar to grasp the pelvic peritoneum a third time at the apex of the incision. The three Endoloops were used to lift the peritoneum off the dorsal aspect of the peritoneal cavity. This drape of peritoneum effectively acted as a retractor, elevating the small bowel, distal ileum, cecum, and ascending colon (Fig, 2). The Endoloops exited the
LAPAROSCOPIC
273
LYMPH NODE SAMPLING
FIGURE
anterior abdominal wall through the trocars and were stabilized with hemostats at the desired tension. Though the sutures exited through the trocars, instruments could easily be placed and manipulated through the trocars. Tissue dissection was performed using aquadissection, scissors, and the blunt dissecting probe. Dissection started at the bifurcation of the common iliac vessels and extended to the renal vein. A second lo-mm Surgiport trocar was inserted suprapubically, through which Endoclips (specially designed titanium clips applied with a multiuse, preloaded, disposable applicator) were placed to clip lymphatics and vessels prior to transection and through which lymph nodes were retrieved. Additional Endoloops were attached as needed to assist with the tenting of the peritoneum as dissection continued cephalad along the common iliac vessels. Upon completion of the procedure, the Surgiports and instruments were removed. The swine then underwent a celiotomy through a vertical midline incision. The peritoneal cavity and its contents were examined for damage and complications due to the laparoscopic procedure. Using a transperitoneal approach, any remaining para-aortic and pelvic lymph
2
nodes were removed and kept separate from these removed laparoscopically. The animals were euthanized and closed en masse with running 0 PGA suture. RESULTS We performed a total of six surgeries in an attempt to develop a technique of performing laparoscopic pelvic and paraaortic lymphadenectomy . The first four surgeries involved the actual development of the technique. The final two surgeries involved the application and testing of this technique. In case 1, an attempt was made to perform the paraaortic lymph node sampling with the laparoscope and accessory trocars placed in “traditional” locales (subumbilically and suprapubically, respectively). At the same setting an attempt was made to operate directly through the peritoneum without employing a specific means of elevating the extensive porcine bowel out of the operating field. With the laparoscope placed in the subumbilical position there was very little mobility allowed and visualization was suboptimal. Similarly, placing the operating instruments through suprapubic trocars mean that they
274
HERD
had to travel a greater distance to reach the field of operation. Often the instruments would seem to “disappear” in the loops of small bowel, which led to a high level of surgeon frustration. Superimposed on these tactical problems was the voluminous bowel, which even the steepest amounts of Trendelenburg position and aggressive attempts at manipulation with blunt probes were unable to mobilize out of the surgical field. The procedure was aborted after over 210 min of frustration. In case 2 an attempt was made to perform the entire procedure via a retroperitoneal approach. The Verres needle was placed 5 cm below the costal margin through the right flank. An attempt to keep the Verres needle in the retroperitoneal space and out of the peritoneal cavity was, to the surgeons amazement, successful. After a pneumoretroperitoneum was obtained, the lo-mm trocar was placed at the site of the Verres needle, with two 5-mm trocars being placed, under direct visualization, 10 cm caudally. Initial dissection went smoothly, but at the psoas muscle the hog peritoneum was adherent to the tendon. Attempts to free the peritoneum, so as to continue the retroperitoneal dissection, were unsuccessful. A series of large rents in the peritoneum occurred with a total escape of the carbon dioxide into the peritoneal cavity and loss of the pneumoretroperitoneum. After complete decompression of the pneumoperitoneum, a similar attempt was made to establish a pneumoretroperitoneum on the contralateral (left) side. This was unsuccessful, and after over 180 min the procedure was aborted with, again, very frustrated surgeons. In case 3, the decision was made to place the Verres needle and trocars as described under Materials and Methods. It was noted that placing them in this “inverted” fashion allowed excellent visualization of the middle and upper abdomen and close proximity to the para-aortic region. The problem now was mobilization of the bowel. With the animal in steep Trendelenburg position as described above, an attempt was made to keep the bowel packed away from the right common iliac vessels and overlying peritoneum. A sling of Gore surgical membrane was fashioned and an attempt made to hold it in place with pickups placed through the lateral 5-mm trocars. Though this allowed initial adequate visualization so that a peritoneal incision could be made and dissection undertaken, over time, with the ventilation of the animal, the small bowel would slip under the sling and obscure vision. After the Gore surgical membrane sling was removed, an attempt to pack the bowel into the upper abdomen using plastic field drapes, which were easily placed into the peritoneal cavity through the lo-mm trocars, was made. Unfortunately, these were similarly ineffective as they would be displaced by the ventilatory effort of the animal and obscure the operative field, which was partially responsible for an inadvertent arteriotomy
ET AL.
and massive intraperitoneal hemorrhage that could not be controlled laparoscopically. The third case was aborted after over 240 min. In our fourth case, we had our first experience using what became, with slight modifications, our final technique. Trocars and instruments were placed as described under Materials and Methods. Unfortunately, the incision in the posterior peritoneum was placed too far lateral to the right common iliac vessels and, though an adequate drape of peritoneum was obtained, the surgeons were disoriented and an attempt at a laperoscopic nephrectomy transpired. The procedure was terminated 100 min after initiation because of uncontrollable hemorrhage from the partially resected kidney. We did learn, however, the limitations of the microtip cautery as regards any visible vessels. This experience led us to select prophylactic clipping of visualized vessel over attempts to control bleeding/oozing with cautery once it had occurred. However, we continued to use cautery to perform dissection through tissue devoid of appreciated vessels. In case 5, a total of six lymph nodes were removed (two high common and four para-aortic) from the precaval space without vascular, intestinal, or urinary tract injury. Subsequent laparotomy revealed an additional two lymph nodes (one low common and one lateral to the vena cava). Hence, we retrieved 75% of the infrarenal precaval lymph nodes. The total operative time for this procedure was 120 min. In case 6, a total of eight lymph nodes were removed from the right side (two high common iliac and six paraaortic). Subsequent laparotomy failed to demonstrate any residual high common or precaval lymph nodes. In addition, no vascular, intestinal, or urinary tract injuries were noted. The total operative time for this procedure was 95 min. DISCUSSION Definitive staging of cervical cancer has prognostic significance as well as therapeutic implications. Cervical cancer is still clinically staged by FIGO, although clinical staging underestimates the extent of disease in 29-48% of cases [4-61. “Extended clinical staging” using noninvasion tools has been used to improve diagnostic accuracy [7]. These studies (computed tomography, magnetic resonance imaging, ultrasonography, lymphangiography, etc.) are useful if positive findings are noted. However, they are severely limited in their ability to detect microscopic metastasis [B]. Fine-needle aspiration (FNA) directed by these noninvasive studies can predict disease if positive (low false-positive rate), but has a significant false-negative rate of 14%. A further confounding problem is that 10% of specimens are unsatisfactory, giving FNA a diagnostic accuracv. of 74% 191. La
LAPAROSCOPIC
LYMPH
Due to the inaccuracies of traditional and extended clinical staging, microscopic metastases to the paraaortic lymph nodes are commonly unappreciated. Obviously, if treatment is directed solely at pelvic disease and the paraaortic region is pathologically involved but not included in the radiation port, failure is inevitable [8]. Multiple studies have evaluated the prognostic and therapeutic benefits of preradiation staging laparotomy to assess peritoneal cytology and paraaortic lymph node status in patients with cervical cancer [4-6,9-161. The underlying theme in these studies is that the status of the paraaortic lymph nodes is a statistically significant prognostic factor of patient survival [lo]. Sampling of the pelvic and paraaortic nodes has been performed at celiotomy (transperitoneal) and via an extraperitoneal approach [ 15-171. The benefits of the transperitoneal approach are the abilities to assess the pelvic and intraabdominal findings, obtain washings, and take directed biopsies. The complications associated with the transperitoneal approach followed by radiation therapy are those of bowel adhesions and radiation damage that can affect up to 30% of patients [ 151. The benefits of the retroperitoneal approach are less bowel manipulation and adhesion formation and, allegedly, fewer postradiation bowel complications (2.5%) [15]. Th e mean blood loss in one study of extraperitoneal bilateral pelvic and paraaortic lymphadenectomy was 300 cc, and the mean postoperative hospital stay was 8 days [15]. A method for sampling the pelvic and paraaortic lymph nodes that would minimize adhesion formation, incision length, and hospital stay, without compromising the “gold standard” of histologic confirmation, would be ideal. Operative laparoscopy has allowed the gynecologist to perform, in an outpatient setting, operations that previously required laparotomy and hospitalization. Although no prospective randomized trials of laparoscopy versus laparotomy with respect to adhesion formation have been done in the human, there is evidence in the animal model that laparoscopy causes significantly less postoperative adhesions than does laparotomy [ 181. Laparoscopic pelvic lymphadenectomy has been studied in both the swine and the human [2,19,20]. These papers fail to address the status of the paraaortic nodes, data which are essential in the making of rational treatment decisions. We have developed a method for laparoscopically sampling the precaval lymph nodes. Prior to the development of the Endoloop retraction technique we attempted other methods of visualizing the paraaortic area. A retroperitoneal (flank) approach with insufflation and laparoscopic insertion retroperitoneally afforded poor visibility and inability to identify vital structures. Different forms of retractors were used through the laparoscope including plastic drapes and Gore surgical membrane sheet. These were ineffectual in keeping the bowel out of the operating field.
275
NODE SAMPLING
The swine model presented unique problems that are not met in the human: the small bowel in the swine is abundant, complicating any attempt at retraction of it, and the posterior parietal peritoneum is weak, such that the Endoloops would occasionally pull through the peritoneum, requiring replacement. With video monitoring and knowledgeable assistants, the dissection of both the common iliac and the paraaortic nodes can be done under the magnification of the laparoscope with minimal bleeding due to the pneumoperitoneum and use of aquadissection, cautery, and Endoclips when necessary. The obturator and hypogastric lymph nodes are probably best sampled with the laparoscope in the traditional periumbilical position. Modification of trocar insertions and operating positioning is possible, and further retraction of the “peritoneal tent” can also be accomplished by tilting the distal tip of the 5-mm sleeve (Endoloops already attached) cephalad. The data so far on the sensitivity and specificity of laparoscopic pelvic lymphadenectomy with its low associated morbidity have made it a very attractive form of surgical staging [19]. Hospitalization is usually only overnight and the time interval from sampling to institution of radiation therapy is minimal. It appears that laparoscopic pelvic and paraaortic lymph node sampling offers the surgeon a unique port of entry for performing the same surgical procedure accomplished utilizing an extraperitoneal or transperitoneal approach, while lessening the risk of negative side effects. Thorough human studies will need to be performed before either traditional means of surgical staging are abandoned or universal laparoscopic node sampling is recommended for all patients with cervical cancer. REFERENCES 1. Jones, W. B. Surgical approaches for advanced or recurrent cancer of the cervix, Cancer 60, 2094-2103 (1987). 2. Averette, H. E., Donato, D. M., Lovecchio, J. L., and Seven, B. Surgical staging of gynecologic malignancies, Cancer 60, 2010-2020 (1987). 3. Shuessler, W. W., Vancaillie, T. G., Reich, H., and Griffith, D. P. Transperitoneal endosurgical lymphadenectomy in patients with localized prostate cancer, J. Ural. 145, 988-991 (1991). 4. Lagasse, L. D., Creasman, W. T., Shingleton, H. M., Ford, J. H., and Blessing, J. A. Results and complications of operative staging in cervical cancer: Experience of the Gynecologic Oncology Group, Gynecol. Oncol. 9, 90-98 (1980). 5. Averette, H. E., Dudan, R. C., and Ford, J. H. Exploratory celiotomy for surgical staging of cervical cancer, Am. 1. Obstet. Gynecol. W, 1090-1096 (1972). 6. LaPolla, J. P., Schlaerth, J. B., Gaddis, O., and Morrow, C. P. The influence of surgical staging on the evaluation and treatment of patients with cervical carcinoma, Gynecol. Oncol. 24, 194-206 (1986). 7. Berek, J. S., and Hacker, N. F. Practical gynecologic oncology. Williams & Wilkins, Baltimore, pp. 241-283 (1989).
276
HERD ET AL.
8. Heaps, J. M., and Berek, J. S. Surgical staging of cervical cancer. Clin. Obsret. Gynecol. 33(4), 852-862 (1990). 9. McDonald, T. W., Morley, G. W., Choo, Y. C., Shields, J. J., Cordoba, R. B., and Naylor, B. Fine needle aspiration of paraaortic and pelvic lymph nodes showing lymphangiographic abnormalities, Obstet. Gynecol. 33, 71-75 (1983). 10. Podczaski, E. S., Palombo, C., Manetta, A., Andrews, C., Larson, J., De Geest, K., and Mortel, R. Assessment of pretreatment laparotomy in patients with cervical carcinoma prior to radiotherapy, Gynecol. Oncol. 33, 71-75 (1989). 11. Potish, R. A., Twiggs, L. B., Okagaki, T., Prem, K. A., and Adcock, L. L. Therapeutic implications of the natural history of advanced cervical cancer as defined by pretreatment surgical staging, Cancer
56, 956-960
15.
16.
17.
(1985).
12. Rubin, S. C., Brookland, R., Mikuta, J. J., Mangan, C., Sutton, G., and Danoff, B. Para-aortic nodal metastasis in early cervical carcinoma: Long term survival following extended-field radiotherapy, Gynecol. Oncol. 18, 213-217 (1984). 13. Tewfik, H. H., Buchsbaum, H. J., Latourette, H. B., Lifshitz, S. G., and Tewfik, F. A. Para-aortic lymph node irradiation in carcinoma of the cervix after exploratory laparotomy and biopsyproven positive arotic nodes, Int. J. Radial. Oncoi. Biof. Phys. 8, 13-18 (1982). 14. Lovecchio, J. L., Averette, H. E., Donato, D., and Bell, J. 5-year
18.
19.
20.
survival of patients with periaortic nodal metastasis in clinical Stage IB and IIA cervical carcinoma, Gynecol. Oncol. 34, 43-45 (1989). Berman, M. L., Lagasse, L. D., Watring, W. G., Ballon, S. C., Schlesinger, R. E., Moore, J. G., and Donaldson, R. C. The operative evaluation of patients with cervical carcinoma by an extraperitoneal approach, Obstet. Gynecol. SO, 658-664 (1977). Ballon, S. C., Berman, K. L., Lagasse, L. D., Petrilli, E. S., and Castaldo, T. W. Survival after extraperitoneal pelvic and paraaortic lymphadenectomy and radiation therapy in cervical carcinoma, Obstet. Gynecol. 57, 90-95 (1981). Berman, M. L., Lagasse, L. D., Ballon, S. C., Watring, W. G., and Tesler, A. Modification of radiation therapy following operative evaluation of patients with cervical carcinoma, Gynecol. Oncof. 6, 328-332 (1978). Luciano, A. A., Maier, D. B., Koch, E. I., Nulsen, J. C., and Whitman, G. F. A comparative study of postoperative adhesions following laser surgery by laparoscopy versus laparotomy in the rabbit model, Obstet. Gynecol. 74, 220-224 (1989). Querleu, D., Leblanc, E., and Castelain, B. Laparoscopic pelvic lymphadenectomy in the staging of early carcinoma of the cervix, Am. J. Obstet. Gynecol. 164, 579-581 (1991). Gersham, A., Daykhovsky, L., Chandra, M., Danoff, D., and Grundfes, W. S. Laparoscopic pelvic lymphadenectomy, 1. Laparoendoscopic Surg. 1, 63-68 (1990).
