Laparoscopic Radical Hysterectomy (Type III) with Aortic and Pelvic Lymphadenectomy: Surgical Morbidity and Intermediate Follow-up.
Nick M. Spirtos, MD, Scott M. Eisenkop, MD2, John B. Schlaerth, MD3, and Samuel C. Ballon, MD1
LAPAROSCOPIC RADICAL HYSTERECTOMY (TYPE III) WITH AORTIC AND PELVIC LYMPHADENECTOMY: SURGICAL MORBIDITY AND INTERMEDIATE FOLLOW-UP
Spirtos, N.M., Eisenkop,S.M., Schlaerth, J.B., and Ballon, S.C.,
Objective: To determine the risk of recurrence and quantify morbidity and moratlity in patients consented to undergo laparoscopic radical hysterectomy (Type III) and retroperitoneal lymphadenectomy.
Materials and Methods: 78 consecutive patients with at least 3 years of followup were consented to undergo this surgical procedure using argon beam coagulation (ABC) and endoscopic staplers. All had a Quetelet index of <35. The average age was 41.5 years ( range 26-62 years). 68 patients had squamous cell carcinomas;8 adenocarcinomas and 2 adenosquamous carcinomas of the cervix.
Results: All but five surgical procedures were completed laparascopically. The average operative time was 205 minutes (range 150 to 430 minutes). The average blood loss was 225 ml. (range 50 - 700ml). One patient (1.3%) was transfused. Operative cystotomies occurred in three patients: two were repaired laparoscopically and one required laparotomy. One patient underwent laparotomy due to equipment failure and another to pass a ureteral stent. Two other patients underwent laparotomy to control bleeding sites. The average lymph node count was 34 ( range 19 - 68). Nine patients (11.5%) had positive lymph nodes. All surgical margins were macroscopically negative but three had microscopically positive and/or close surgical margins. One patient developed a ureterovaginal fistula post-operatively requiring re-operation. Follow-up has been provided on a q 3 month basis. There have been four (5.1%) documented recurrences with a minimum of three years follow-up.
Conclusions: Laparoscopic radical hysterectomy (Type III) can be successfully completed in patients with early-stage cervical cancer with acceptable morbidity. Intermediate-term-follow-up validates the adequacy of this procedure.
Laparoscopic Radical Hysterectomy (Type III) with Aortic and Pelvic Lymphadenectomy: Surgical Morbidity and Intermediate Follow-up.
Nick M. Spirtos, MD, Scott M. Eisenkop, MD, John B. Schlaerth, MD, and Samuel C. Ballon, MD
Acceptance of a new surgical technique in oncology requires that technical feasibility be demonstrated and morbidity and mortality associated with it are not prohibitively high. Additionally, short and long-term survival should be comparable to that of the accepted standard therapy. Techniques used to perform radical hysterectomy with aortic and pelvic lymph node dissection laparoscopically are well-described.(1,2) and therefore the feasibility issues have, for the most part been resolved. Much less is known about the morbidity and mortality associated with this procedure and even less has been published regarding short or long-term survival following laparoscopic radical hysterectomy (type III), aortic and pelvic lymph node dissection. We report the surgical morbidity and intermediate-term follow-up on 78 patients with stage Ia2 and IB carcinoma of the cervix with at least three-years of follow-up.
Materials and Methods
From July 1, 1994 through December 1996 eighty-four patients with Stage Ia2 or Ib cervical cancer underwent laparoscopy with the intent to perform a radical hysterectomy (type III) with pelvic and aortic lymph node dissection. Only patients found to have negative paraaortic lymph nodes, clinically normal pelvic lymph nodes at surgery, and no evidence of extracervical disease were to be included in this study population. All patients undergoing this procedure underwent complete history and physical examination. Clinical data collected included patient age, GOG performance status, Quetelet index, and FIGO stage.(3) Tumors were classified according to cell type and grade. Tumor size and depth of invasion as well as vascular space involvement (VSI) and lymph node status and the number resected recorded. Follow-up care was provided on a quarterly basis.
Using previously described techniques with some modification all patients underwent laparoscopic radical hysterectomy (type III) with or without bilateral salpingooophorectomy and pelvic and aortic lymph node dissection.(1,2) Combined general and epidural anesthesia was used in every case, as were prophylactic antibiotics and sequential compression devices and support hose. After positioning the patient with the thighs on the same plane as the abdominal wall and the knees bent at 90 degrees and placed in Allen stirrups. The cervix is grasped with a tenaculum and a sound is passed into the uterine cavity after which it is steristripped to the tenaculum, in order to facilitate its use as a uterine manipulator. Usually five 12- millimeter trocars are placed in the configuration shown in Figure 1. The umbilical trocar is placed first, using a direct-puncture technique. After placing the patient in Trendelenberg position (approximately 30 degrees), the other four trocars are placed under direct visualization, the abdominal cavity inspected, and the retroperitoneal spaces opened. If macroscopically positive lymph nodes or gross parametrial extension identified, the laparoscopic procedure is terminated. Using argon-beam coagulation (ABC) (Conmed, Utica, N.Y.) and counter traction, the aortic, then the pelvic lymph nodes are resected with camera position changed from the suprapubic to the umbilical port site respectively. Adequacy of the lymph node dissection was determined visually. The superior boundary of the lymph node dissection was the inferior mesenteric artery (IMA); the inferior boundary was the external iliac artery at the point it is crossed by the circumflex iliac vein; the lateral boundary was the muscular sidewall of the pelvis; and the posterior boundary was the obturator nerve. The midpoint of the common iliac artery served to divide pelvic from aortic lymph nodes. Lymph nodes were separated into four groups: right and left pelvic and right and left aortic. Lymph node counts were recorded as reported by the Departments of Pathology.
After dissecting the urinary bladder from the cervix and upper vagina and the rectum from the posterior aspect of the vagina, the adnexae are either resected or left in situ using an endoscopic stapler to secure the blood supply at the appropriate level to achieve our purpose. Then the uterine arteries are transected at the level of the hypogastric artery. In the first ten cases endoscopic clips and shears were used and thereafter these vessels were transected with an endoscopic stapler using vascular cartridges. The transection and resection of the paracervical soft tissues including the parametria, paracolpos, and uterosacral ligaments is then undertaken using Endo-Path endoscopic staplers (Ethicon Endosurgery, Cincinnati, Ohio). To achieve an adequate lateral margin down to the level of intended vaginotomy, usually two additional firings of the endoscopic stapling device are required after transecting the uterine vessels. Similarly resection of the uterosacral ligaments require two firings of the endoscopic staplers. The ureter is then dissected free from the overlying uterine artery and vesicouterine ligament using argon beam coagulation and counter-traction. The necessary lateral and posterior mobilization of the ureters and bladder is then undertaken to allow for resection of the most medial aspect of the paracolpos. This step is performed by using a flexible endoscopic stapler placed through the lateral 12 mm trocar sites. With the surgeon placing his non-dominant hand in the vagina to assure an adequate distal margin the vaginal incision is made circumferencially with the ABC. The specimen is removed and the vagina closed endoscopically using an Endo-stitch device in an interrupted or continuous fashion.(US Surgical, Norwalk, Ct) If drains were placed they were introduced through the suprapubic trocar, exiting through the lateral ports sites. In these cases no fascial closure was used. The drains were sutured in place with 3-0 nylon.
Operative time was measured from the time of placement of the first trocar. Estimated blood loss was approximated to the nearest 50cc increment. Hospital stay was counted from the first postoperative day with any fractional day being counted as a full day. Short-term complications occurring within 30 days of surgery were considered perioperative. Patients were discharged home with either a suprapubic or foley catheter in place and not checked for residual urine volumes until at least 2 weeks post-operatively. The catheter were discontinued if the residual measured less than 75 cc.
Disease free interval (DFI) was defined from the date of surgery to radiologic or physical evidence of recurrence. Survival was defined as the time from the date of surgery to death or the date of last follow-up. The Kaplan-Meier method was used to determine the actuarial survival and disease-free interval of the cohort.
Six of the eighty-four patients were found to have, either microscopically or macroscopically positive aortic lymph nodes; macroscopically positive pelvic lymph nodes; or extracervical disease and were excluded from this analysis. The remaining 78 patients serve as the basis of this report. All patients had a GOG performance status of 0 or 1 and a Quetelet index <35 (5,6). The average age was 41.5 years 9 (range 26-62 years). Sixty-eight patients had squamous cell carcinoma; 8 adenocarcinoma; and 2 adenosquamous carcinoma.
All but five had the surgical procedure completed laparoscopically. Two patients required laparotomy to control bleeding sites. One patient had the inferior mesenteric artery lacerated with an endoscopic clip and another had a perforator arising from the anterior aspect of the vena cava transected. One other patient required laparotomy as we were unable to maintain an adequate pneumoperitoneum after performing a retroperitoneal laparoscopic lymphadenectomy. One patient required laparotomy to repair a cystotomy and another to place a ureteral stent.
The average operative time was 205 minutes (range 150 to 430 minutes). The first 26 patients averaged 255 minutes and the last 52 patients averaged 186 minutes. The average estimated blood loss was 250 ml. (range 50 ml. to 700ml.) Only one patient (1.3%) required transfusion. Three patients, all with previous cesarean sections suffered intra-operative cystotomies of which two were repaired laparoscopically while one required laparotomy. This was the extent of the intraoperative complications. The average length of hospital stay was 2.9 days (range 1-7).
Post-operative complications occurred in 7 patients (9.1%). One patient each suffered from a ureterovaginal fistula, deep venous thrombosis, urosepsis, vaginal cuff abscess, or abdominal wall hematoma. Two patients developed pelvic lymphocysts. All bladder catheters were discontinued within 12 weeks of surgery with all but 2 discontinued by 6 weeks. Two patients required a valsalva maneuver to empty their bladders.
Sixty- eight patients had squamous cell carcinoma; 8 adenocarcinoma and 2 adenosquamous carcinoma of the cervix. Surgical margins were negative except in three patients, which were either microscopically positive or described as "close". Twenty-six patients (33%) presented with occult lesions; 38(49%) had lesions from 1- 3 cm; and 14(18%) had lesions measured from 4-6 cm. In each group, recurrence occurred in 1(3.6%); 4(9.5%);and 3(21.4%) patients respectively. Fifty-two patients (66%) were found to have less than 10 mm of invasion and only one patient in the subgroup recurred (2%). Twenty-one patients (27%) were found to have 11-20 mm of invasion and 4 of them, recurred (20%). Five patients (6%) had greater than 20 mm of invasion and 3 recurred (60%). Vascular space involvement was identified in 33 patients (42%) of which 5 recurred, while there were 3 recurrences in the 45 patients without this finding. Rates recurrence and survival in subgroups of patients based on pathologic characteristics are found in Table 1.
The average number of lymph nodes was 34.1 (range 19-68) The average number of aortic and pelvic lymph nodes was 10.3 and 23.8 respectively. Nine patients (11.5%) had positive lymph nodes. Seven patients had unilateral involvement of one or two lymph nodes and two had bilateral involvement of multiple lymph nodes.
Four patients with positive lymph nodes were treated with radiation therapy and one was treated with chemotherapy in combination with radiation therapy. Four patients with a single unilateral lymph node did not receive additional treatment. Two patients with "close" margins both with vascular space involvement (vsi) and negative lymph nodes received radiation. The other patient refused further treatment initially.
To date, eight patients (10.3%) have recurred. The estimated 5-year disease free interval after treatment was 89.7%. (Figure 1) Mean follow-up was 66.8 months; a standard error of 1.78 and 95% confidence interval of (63.29, 70.28) Three patients recurred on the pelvic sidewall with two of these three below the level of obturator nerve. One patient had an isolated nodal recurrence on the external iliac artery just distal to the circumflex iliac vein. There was one recurrence in the vaginal apex and base of the bladder. One patient each had liver and lung metastases and another, involvement of the suprarenal lymph nodes.
Of the 8 patients with recurrence, 4 had been radiated and 3 of those 4 recurred within the treatment field and are dead of disease. It should be noted that 2 of these patients were treated for "close margins". The remaining 4 patients were not treated following initial surgery. All four have subsequently received radiation therapy and ¾ remain alive, although 2 of the survivors have had a short period of follow-up since the recurrence was noted.
There have been five deaths recorded to date. The estimated 5 year survival for the entire cohort was 93.6% (Figure 2) with a mean follow-up of 68.3 months; a standard error of 1.61 and 95% confidence interval of (65.14, 71.47). Three died due solely to recurrent disease. One due to unsuspected heart disease and one due to sepsis following a bowel obstruction suffered greater than 1 month after surgery. The depth of invasion, lesion size, histology, lymph node yield and status, patient age, Quetelet index, or performance status could not be demonstrated to influence survival.
Issues regarding the use of minimally invasive surgical techniques in the patient cervical cancer, center on 1) applicability 2) operative and postoperative morbidity and mortality 3) risk of recurrence and overall survival and 4) adequacy.
Seventy-three of 84 patients (89%) with stage Ia2 and Ib cervix cancer successfully underwent laparoscopic radical hysterectomy with or without adnexectomy and aortic and pelvic lymph node dissection in patients with Quetelet indices < 35. In order to discuss applicability of any procedure it is important to account for every patient in which there was intent to complete that procedure and identify limiting factors. . In six patients the procedure was abandoned due to the presence of macroscopically pelvic lymph nodes, positive aortic lymph nodes, or intraperitoneal disease. In this subgroup we are most concerned about the potential risk associated with the fragmentation and removal of grossly positive lymph nodes. Case reports regarding port site metastases in most gynecologic malignancies are, for the most part, in patients with widespread intraperitoneal disease. (4) However, trocar site metastases have been reported in patients with stage Ib squamous cell carcinoma of the cervix without extracervical disease.(5) If the mechanism associated with these occurrences is related to the laparoscopy this problem can be avoided by simply converting the procedure to a traditional laparotomy. As the incidence of grossly positive lymph nodes or extracervical disease in this series and others is reported to occur in less than 10% of radical hysterectomy candidates it is unlikely that this riddle will be solved and therefore it is probably prudent to either abandon the procedure or make every effort to not fragment positive lymph nodes and remove them either via an endoscopic pouch or colpotomy. (6) This potential limitation is inherent in all minimally invasive surgical procedures that integrate laparoscopic lymphadenectomy.
Other minimally invasive procedures combining laparoscopic lymphadenectomy and vaginal radical hysterectomy are not as universally applicable as the complete laparoscopic radical hysterectomy and aortic and pelvic lymphadenectomy described herein. Parity and the bony configuration of the pelvis are factors that can be limiting when attempting radical vaginal hysterectomy that are of little consequence when performing the procedure laparoscopically.
In this series, patients all had a Quetelet index of less than 35. In our initial 10 patients the Quetelet index was limited to 30. As our experience increased we extended our limits. We have noted that obesity rarely affects our ability to perform the pelvic lymphadenectomy in patients with endometrial cancer but rather it is the dissection of the aortic lymph nodes that is limited with a rising Quetelet index. Currently there is no weight limitation in place at the Women's Cancer Center for patients with cervical cancer who are otherwise candidates for laparoscopic radical hysterectomy and lymphadenectomy. Other contraindications or limits in applicability of this procedure, except for multiple previous surgical procedures, are not unique to laparoscopy per se but rather general medical risks present regardless of the choice of incision size or location.
Intraoperative complications occurred infrequently. Cystotomy occurred in 3/78 patients (3.9%), all of whom had undergone at least 1 cesarean section. This compares favorably to other reports in the literature that even designate this result of dissection as a complication. Two of the cystotomies were repaired laporoscopically using an Endostitch device and 0- polysorb suture. One patient required a small transverse incision made suprapubically to facilitate the bladder repair. All cystotomies in this series occurred in the first 50 patients. Since that time our technique has been modified so that during the dissection of the bladder from the vagina the surgeon places his non-dominant hand in the vagina and elevates the anterior wall, while the operating room technician pushes the uterus superiorly. We have found that this maneuver has resulted in improved tactile sensation and local counter traction at the point of dissection.
No patient currently requires intermittent bladder catheterization although 2 patients noted reduced sensation and required a Valsalva maneuver to empty their bladders to their satisfaction. These two patients had large primary lesions, measuring 6 and 7cm and had the upper third of the vagina resected along with the parametrial tissues. The remaining 76 patients were found to have residual urine volumes of less than 75cc within 6 weeks of surgery and return of bladder sensation. Beyond checking residual urine volume no specific urological evaluation has been performed on this group of patients. This aspect of radical pelvic surgery is not often reported as the incidence of this complication could be affected significantly by the extent of pre- and post-operative evaluation undertaken. As a result it is difficult to establish a basis of comparison of our findings. Performing the laparoscopic radical hysterectomy, using an approach similar to the abdominal procedure, allows the surgeon to spare the most posterior aspect of the uterosacral ligaments, which is always transected when performing radical vaginal hysterectomy. Hypothetically, this should result in a comparative reduction in bladder dysfunction, although this has not been proven on a randomized, prospective basis.
One of two patients requiring laparotomy to control bleeding, needed to be transfused. This was the only patient (1.3%), requiring a transfusion. This is in contrast to the reports by Benjamin et al and Covens et al wherein 44% and 34% of patients undergoing radical hysterectomy at their respective institutions were transfused perioperatively (7,8) The obvious advantage of such a reduction in the rate of transfusion is a parallel reduction in the incidence of infectious disease associated with homologous transfusion. The impact of this benefit could arguably be dampened by the use of autologous blood products, as was reported by Benjamin et al. during the period of 1991-3.(7) However, autologous donation is expensive and the amount of blood available to the patient is limited.
The second patient requiring laparotomy to control a bleeding site did not require transfusion. The patient only had an estimated blood loss of 250cc’s but nevertheless required a laparotomy to minimize the blood loss as we were unable to endoscopically control the bleeding site on a purely mechanical basis. As instrumentation improves and experience gained this group of patients will likely decrease in size. Early in our experience approximately 5% of patients with gynecologic cancer undergoing laparoscopic lymphadenectomy required laparotomy to control bleeding sites. (9) The number of laparotomies required to control bleeding have decreased due to the introduction of smaller endoscopic vascular clips which subjectively are more hemostatic, a vaginal sponge placed intraperitoneally at the beginning of the case used to tamponade bleeding sites, and the patience that comes with experience to maintain adequate pressure long enough to achieve hemostasis.
One of the difficulties surrounding this group of patients undergoing unplanned laparotomies is a common practice to describe this event as a complication, a practice with which we disagree. The complication or error in surgical judgement occurs when efforts to avoid laparotomy result in unnecessary blood loss and transfusion. Unplanned laparotomies will always occur in any series of endoscopic procedures and are no more a complication than finding at laparotomy that the initial incision needs to be modified either by extending it or making a second one in order to respond to an unexpected intraoperative finding or event. Lastly, in our experience laparotomy to control bleeding is almost always associated with the aortic lymph node dissection and as a result we undertake this portion of the surgical procedure first, so if laparotomy is required it is performed early in the case. The additional advantages of this approach are that the surgeon is performing the most difficult portion of the procedure when freshest and information gained from the evaluation of the aortic or high common iliac lymph nodes may significantly affect the extent of the surgery required in the pelvis.
One other unplanned laparotomy was undertaken to pass a ureteral stent that could not be passed cystoscopically. Again a small suprapubic incision was made and the hospital stay for this patient was 3 days.
Overall, the average hospital stay was 2.9 days (range 1-7). Compared to other published North American series there is a marked reduction in hospitalization associated with this procedure. Hospital stays of 7-10 days are associated with radical abdominal hysterectomy and lymph node dissection. (7,8, 10) Hospital stays vary substantially, from an average of 4 days to 11 days, in reports of laparoscopically-assisted radical vaginal hysterectomy. Comment should be reserved as differences in culture and health care delivery and reimbursement, rather than medical need may account for this wide variation. is also associated with shorter hospital stays when compared to abdominal procedures and is much more similar to our experience.(11, 12)
Postoperative complications occurred in 7 patients (9.1%). One patient each suffered from a ureterovaginal fistula, deep venous thrombosis, urosepsis, vaginal cuff abscess, or abdominal wall hematoma. Two patients developed pelvic lymphocysts. The incidence of post-operative complications compares favorably to other recent reports in the literature and there were no unusual or unique complications attributable to the laparoscopic aspects of this surgical procedure. (7,8,10,11,12,13) One complication noticeable absent was serious wound infections. In addition to reduced complications associated with blood loss at the time of surgery and subsequent blood product transfusion, late complications related to transfusion should also be reduced. Early in our experience when still using interrupted sutures to close the vaginal cuff some patients presented with a profuse watery vaginal raising concerns regarding urinary tract fistulas. As it developed all but one, the work-up was negative for fistulas, and the profuse discharge was determined to be lymphatic fluid leaking from the cuff. In all cases it resolved spontaneously. Now that the vaginal cuff is closed with a running suture this problem has vanished. Regardless, 2 patients developed symptomatic lymphocysts requiring percutaneous drainage.
Assuming morbidity and mortality associated with a surgical procedure is acceptable then determination of adequacy should be based on an objective basis. The obvious and immediate basis available to assess the adequacy of any radical hysterectomy would be simple measurements of the parametrial and paracolpos as well as the length of vagina resected. In 1996 our first ten laparoscopic radical hysterectomies were (Type III) were reported with the average parametrial margin measuring 3.3cm (range 1.0 – 5.0) and the average vaginal margin 2.15cm. (range 1.0 – 3.5) (9) Bennedetti-Panici et al performed similar measurements to assess the extent of parametrial resection acheivable using vascular clips instead of clamps to secure the margins of the lateral dissection. They reported average parametrial margins of 3.3 cm.(range 1.5 – 4.5) when using Z- clamps and 5.2 cm (range 2.8-7.2) when using vascular clips. No vaginal margins were noted. (13) Our laparoscopic technique appears to provide parametrial margins equal to those using traditional clamps as well as equivalent to those reported by Schneider et al when performing laparoscopically-assisted radical vaginal hysterectomy. (12) On this basis adequate margins have shown to be obtainable. Fanning et al also reported using endoscopic staplers to perform abdominal radical hysterectomy with a notable decrease in operative time and blood loss associated with this technique. (10) Margins were not quantified objectively in that report. It is incomprehensible to think that by passing the same instrument through a 12mm trocar instead of a skin incision of unknown length or orientation would affect the function of the stapler or the adequacy of the resection.
In general, the introduction of laparoscopic procedures in gynecologic oncology has been highly scrutinized on a prospective basis in studies undertaken by the Gynecologic Oncology Group (GOG) and until these feasibility studies are completed, patients undergoing endoscopic surgical procedures cannot be entered into a GOG treatment study. (14) For example patients receiving neoadjuvant chemotherapy followed by radical hysterectomy cannot have that procedure performed endoscopically despite there being no control for size, orientation, or length of surgical incision or for surgeon to surgeon variation in abdominal radical hysterectomies being performed in that study.
GOG protocol #9207, was unique in that the lymph node dissection was performed endoscopically then the patient underwent laparotomy to assess the completeness of the nodal dissection and complete the radical hysterectomy. In that study, Schlaerth et al found all aortic lymph node dissections to be adequate based on surgeons’ evaluations, independent video review, and surgeons’ evaluation of the operative site at laparotomy. In six patients (15%) the only area found to contain lymph node bearing tissue following endoscopic lymphadenectomy was lateral to the common iliac artery. (15) This was more problematic on the left side, probably due to issues of mobilizing the sigmoid colon. This study identified an area in need of special attention if complete pelvic lymphadenectomy is to be performed. At the time of laparotomy no positive lymph nodes were found left in situ and no more than 1 lymph node was found in any one patient.(15) Chu et al undertook a similar study and found no patient to harbor residual lymph nodes at the time of laparotomy. (16) Even with the use of intraoperative lymphography of only 29% of patients reported by Kolbenstvedt and Kolstad were determined to have had a complete lymphadenectomy at the time of abdominal radical hysterectomy.(17) A later study by these authors identified, as did Schlearth et al that the intermediate common iliac nodes were difficult to identify and resect. (18) In this study our average lymph node count was 34.4 (range 19 –68) with the average number increasing with experience. In the last 20 patients the average lymph node count was approximately 44. In the GOG study where 85% of patients were considered to have had a complete lymphadenectomy the lymph node counts averaged 43.2 and Kolbenstvedt and Kolstad, in their last one hundred patients found that an average of 34.8 lymph nodes were removed. (15,17) Undoubtedly, with care, lymphadenectomy can be completed successfully using endoscopic techniques and has the same limitations that have been reported in patients undergoing laparotomy.
The issue of adequacy of the radical hysterectomy performed endoscopically is even more interesting in light of the total lack of knowledge of what constitutes an adequate surgical margin. For a surgical procedure being performed for now over 100 years little objective data exists regarding the adequacy of the parametrial resection. The majority of patients will have negative parametria and negative lymph nodes (58 %) and it would be hard to argue that resecting more of the uninvolved tissues in this setting is beneficial.(19) Approximately 36% of patients will have negative or positive parametria with lymph node metastases present again logic prevents arguing for the importance of the extent of the local tissues when disease is known to be metastatic distant to this site. Less than 10% of patients will have involved parametria and negative lymph nodes and that might truly be affected by the extent of the surgical resection undertaken. So much is written about the radical lateral resection yet parametrial extension can involve tissue anterior to the cervix. (20,21) Simply put, the amount of tissue present to secure an anterior surgical margin is limited due to the presence of the bladder and as a result, even when the parametrial tissues are negative the anterior margin can be significantly less than the lateral margin. Despite these basic anatomic considerations, there is an assumption of adequacy if the skin incision is long enough.
Regardless of the reasons, or lack thereof, there has been a slow integration of endoscopic procedures into the clinical practice of gynecologic oncology. Only 11/52 GOG institutions (21%) have participated in LAP-2, the ongoing randomized trial, in patients with early stage endometrial cancer (laparoscopic lymph node sampling and assisted vaginal hysterectomy and bilateral salpingooophorectomy vs. total abdominal hysterectomy and bilateral salpingooophorectomy and lymph node sampling). In an earlier surgical staging protocol, GOG #33, 35/35 full-member institutions (100%) participated. (14) As a result of this reluctance to continue to be surgical leaders and innovators in the gynecologic community, gynecologic oncologists will be relegated to a lesser role, certainly as it pertains to endoscopic procedures. It is disappointing, given our familiarity with the instrumentation, that gynecologic oncologists have dismissed endoscopic procedures in the most part, while the general surgical world has embraced the technology. In the general surgical literature few if any endoscopic procedures were reported until 1989. Today splenectomy, adrenalectomy, colectomy for malignant and non-malignant conditions are commonplace. The only limitation to the endoscopic resection necessary to perform a radical hysterectomy lies not in the instrumentation but solely with the surgeon.
The ultimate test of the adequacy of this, or any oncologic surgical procedure, is survival. This report, although comprises the outcomes of only 78 patients, is the first providing recurrence rates and survival data with a minimum of three years follow-up in patients undergoing laparoscopic radical hysterectomy and aortic and pelvic lymphadenectomy. Figures 1 and 2 indicate that the total number of recurrences in three years was 8 patients (11%) and overall survival was 92%. These figures compare to those published by Delgado et al in a similar group of patients undergoing radical abdominal hysterectomy and aortic and pelvic lymphadenectomy as well as those of Dargent et al reporting his experience combining laparoscopic lymphadenectomy and radical vaginal hysterectomy.(6,22) The small number of subjects and events (recurrence and death) survival were not associated
These data support the position that radical hysterectomy and retroperitoneal lymph node dissection can be accomplished using any number of approaches and the most important is not incision size or instrumentation but rather surgical expertise and judgement.
Table 1. Patients (%) with Recurrent Disease and Survival Categorized by Surgical and Pathologic Findings