Obstetrics & Gynecology:
Robotic Hysterectomy and Pelvic–Aortic Lymphadenectomy for Endometrial Cancer
Seamon, Leigh G. DO; Cohn, David E. MD; Richardson, Debra L. MD; Valmadre, Sue MD; Carlson, Matthew J. MD; Phillips, Gary S. MAS; Fowler, Jeffrey M. MD
From the Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, and the Center for Biostatistics, The Ohio State University College of Medicine, Columbus, Ohio.
See related editorial on page 1198.
Corresponding author: Jeffrey M. Fowler, MD, The Ohio State University College of Medicine, M-210 Starling-Loving, 320 West Tenth Avenue, Columbus, Ohio 43210-1228 Jeffrey.Fowler@osumc.edu.
Financial Disclosure The authors have no potential conflicts of interest to disclose.
OBJECTIVE: To report the learning curve and outcomes after our first 105 patients underwent robotic hysterectomy and pelvic–aortic lymphadenectomy for the comprehensive staging of endometrial cancer.
METHODS: We prospectively collected patient demographics, operative times, complications, pathologic results, and length of stay on all patients who underwent robotic hysterectomy pelvic–aortic lymphadenectomy for clinical stage I or occult stage II endometrial carcinoma.
RESULTS: One hundred five patients at The Ohio State University between March 2006 and April 2008 underwent exploration with the intent of robotic hysterectomy pelvic–aortic lymphadenectomy. Ninety-two (87.6%) were completed robotically and 13 (12.4%) were converted. The probability of conversion was 15% (95% confidence interval [CI] 8.4–25.7), 24% (95% CI 12.4–39.9), 35% (95% CI 15.9–59.6), and 48% (95% CI 19.1–77.8) for a body mass index of 40, 45, 50, and 55 kg/m2, respectively. The median body mass index was 34 kg/m2 (range 19–58). In patients who underwent a robotic hysterectomy pelvic–aortic lymphadenectomy (n=79, 75%) or a robotic hysterectomy–pelvic lymphadenectomy (n=6, 5.7%), the average operating time from skin opening to closure was 242 minutes (±50 minutes). The median estimated blood loss was 99 mL (±83 mL). The median number of lymph nodes recovered was 29 (range 9–56), 21 (range 5–40) pelvic nodes and 9 (range 2–21) aortic nodes. The median length of stay was 1 night. After analysis of the data, we determined approximately 20 cases are needed to gain proficiency.
CONCLUSION: Early experience demonstrates that robotic hysterectomy pelvic–aortic lymphadenectomy for endometrial cancer is feasible, with approximately 20 procedures needed to gain proficiency.
LEVEL OF EVIDENCE: III
Endometrial carcinoma is the most common gynecologic malignancy in the United States, with 40,100 cases per year and 7,470 deaths.1 The majority of patients with endometrial carcinoma present with apparent early stage disease, and the initial management is surgical. Comprehensive surgical staging best defines the biologic nature of the disease and allows the gynecologic oncologist to make accurate postoperative treatment decisions. Nevertheless, endometrial cancer remains the most heterogeneously managed gynecologic malignancy, with disparity in the extent of surgical staging and postoperative treatment recommendations.2–3
The cornerstone of staging is pelvic and paraaortic lymphadenectomy. Traditionally, this is accomplished by laparotomy through a midline incision. Improvements in technology and training have resulted in the ability to perform complex surgical procedures using laparoscopy. Avoiding the morbidity of an abdominal incision has many advantages for the patient, including improved cosmesis, less blood loss, decreased need for postoperative analgesia, and a quicker recovery. Since the first description of minimally invasive surgery for endometrial cancer staging by Childers et al4 in 1993, the laparoscopic management of endometrial carcinoma has proven technically feasible, but is limited by a difficult learning curve, patient factors, surgeon experience, and great variability in accomplishing complete surgical staging.5–11 Other disadvantages of the laparoscopic approach include two-dimensional vision, dependence on assistance skill, decreased range of motion and degrees of freedom of instruments, and ergonomic disadvantages to the surgeon.
The da Vinci surgical system (Intuitive Surgical, Sunnyvale, CA) was approved by the U.S. Food and Drug Administration for gynecology in April 2005. Although many disadvantages of laparoscopy are minimized by the robotics platform, scant literature exists regarding the feasibility and learning curve for comprehensive surgical staging using the da Vinci surgical system (Intuitive Surgical) in patients with endometrial carcinoma. Our objective is to describe the initial 105 consecutive cases and experience with da Vinci hysterectomy pelvic and aortic lymphadenectomy and discuss the learning curve.
MATERIALS AND METHODS
At The Ohio State University, two gynecologic oncologists (J.M.F. and D.E.C.) began performing gynecologic robotic surgery in January 2006 and initiated the robotics platform for the comprehensive staging of endometrial carcinoma in March 2006. Institutional review board approval at The Ohio State University was obtained. We prospectively collected patient demographics, operative times, complications, all conversions to laparotomy, pathologic results, and length of stay on all patients at one institution who underwent robotic hysterectomy pelvic–aortic lymphadenectomy for clinical stage I or occult stage II endometrial carcinoma from March 2006 to April 2008. The definitions for operative times are noted in Table 1. Complications were defined as the following surgical or postoperative problems: any major vessel, nerve, or gastrointestinal injury, venous-thromboembolic, cardiac, pulmonary, gastrointestinal, renal/urinary tract, or urologic events.
The surgical team consisted of the primary surgeon, bedside assistant (fellow or resident, who occasionally served as the vaginal assistant), vaginal assistant, and a robotics-dedicated scrub technician and circulating nurse. A preliminary report on our initial experience as well as a detailed description and video of our robotic set-up and port placement have previously been published.12 Our management goal in endometrial cancer patients, regardless of preoperative grade, includes comprehensive robotic staging and includes pelvic washings, total hysterectomy, bilateral salpingo-oophorectomy (BSO), and pelvic–aortic lymphadenectomy when feasible.
We determined significant relationships using the two-sample t test, analysis of variance, or Kruskal-Wallis test, depending on the distribution of the continuous data. Pearson’s χ2 was used to determine the association between categorical outcomes. Linear regression was used to model the relationship between console time and body mass index (BMI), whereas logistic regression was used to model the relationship between whether a patient was converted to laparotomy and BMI. Locally weighted regression was used to generate smoothed lines that represent operation time over the sequence of 105 operations. The smoothed line is fit using weighted least squares, giving more weight to the point whose response is being estimated and less weight to points further away.13 Before analysis, the authors arbitrarily determined the definition for “proficiency of the procedure” to be the point at which the slopes for the locally weighted regression curves begins to flatten (room, skin, and console times). Efficiency was defined as the point at which the curve becomes linear. All analyses were run using Stata 10.0 (StataCorp LP, College Station, TX).
One hundred five robotic cases for apparent clinical stage I or occult stage II endometrial carcinoma were attempted, and 92 (87.6%) were completed robotically. The patient demographics are described in Table 2. The spectrum of procedures is listed in Table 3. Sixty-one percent of the patients were obese, with 30.5% Class III or more obesity,14 and 25% percent had at least two other comorbidities in addition to obesity. Ninety-two (87.6%) were successfully completed, and 13 (12.4%) were converted to laparotomy for the following indications: 11 poor exposure (n=7 difficulty packing the small bowel, n=4 dense adhesions), one technical difficulty, and one for a lost 4×4-in X-ray detectable sponge. Despite a lower median BMI for the first 50 cases (median BMI 33 kg/m2; range 19–52 kg/m2) compared with the next 55 (median BMI 36 kg/m2; range 20–58), 62% of the conversions (8 of 13) occurred within the first 50 cases. A single case was converted beyond the 65th procedure.
In 13 patients (12.4%), a pelvic–aortic lymphadenectomy was not performed due to morbid obesity (BMI range 47–58 kg/m2), a preoperative histology of grade 1 endometrioid adenocarcinoma, and no obvious myometrial invasion noted intraoperatively combined with the inability to safely expose the anatomy required to complete the dissection. In six (46.1%) of these patients, robotic pelvic lymphadenectomy in addition to robotic hysterectomy with or without BSO was feasible with a median pelvic node count of 21 (range 17–26). Five (38.6%) underwent robotic hysterectomy with or without BSO only. Two of these 13 patients (15%) were converted for morbid obesity (BMI 52 and 45 kg/m2) and inability to obtain and maintain exposure. Even after laparotomy, pelvic lymphadenectomy was not feasible in one of these patients. Although comprehensive surgical staging (robotic hysterectomy pelvic–aortic lymphadenectomy) including aortic lymphadenectomy in these 13 patients was not feasible, the final stage and grade of the patients undergoing robotic hysterectomy with or without BSO only (n=5) compared with robotic hysterectomy pelvic lymphadenectomy (n=6) only were as follows: three patients with TIAg1 Nx Mx, two patients with TIBg1 Nx Mx compared with three patients TIAg1 N0 Mx, one patient with TIAg2 N0 Mx (preoperative grade 1), and two patients with TIBg1 N0 Mx, respectively. Thus, no high- or intermediate-risk patients were identified who did not undergo comprehensive staging.
For the completed robotic hysterectomy pelvic–aortic lymphadenectomy (n=79) or robotic hysterectomy pelvic lymphadenectomy (n=6) cases, surgical–pathologic factors are listed in Table 4, and operative times are summarized in Table 5. The median age was 60 years (range 35–82 years), with a BMI of 34 kg/m2 (±8.9), and median weight of 178 pounds. The preoperative grade was noted as grade 1, 2, 3 for 71%, 15%, and 10% (n=60, 15, 10) of patients, respectively. The median tumor size was 3.4 cm (range: no residual to 8 cm). On final histology, the tumors were classified as endometrioid, mucinous, clear cell, or serous in 87%, 2.3%, 7%, and 3.4% (n=74, 2, 6, 3) of patients, respectively. Positive washings were found in three patients (3.5%).
Room, skin, and console times significantly improved as the sequence of robotic operations increased with time (Fig. 1). Console time increased 8.5 minutes for every 1 unit increase in BMI (P=.005) after adjusting for sequence of robotic operation and whether a patient was converted to laparotomy, because these were both significant confounders. The learning curve (defined in the Materials and Methods section) seems to be approximately 20 cases (Figure 1). Based on visual clues from Figure 1, the sequence of robotic operations was dichotomized at 20, and the two groups were compared with the following results for mean room, skin, and console times, respectively (all P<.001): 368 compared with 288, 296 compared with 228, 236 compared with 167. Although a significant operating room time difference was noted, the estimated blood loss (98 mL compared with 120 mL, P=.392) and number of lymph nodes (29 compared with 26, P=.134) were not significant when comparing the first 20 to those procedures after 20.
There was no statistical difference in the operative time, estimated blood loss, or length of stay when comparing patients with a normal BMI (less than 25, n=14) with those patients overweight (BMI 25–29, n=24), obese (BMI 30–39, n=35), or morbidly obese (BMI 40 or more, n=20). The mean BMI in those patients converted to laparotomy was significantly higher than that of patients completed robotically (40±6.6 compared with 34±9.1, P=.008), and conversion rates increase with BMI. The probability of conversion was 15% (95% CI 8.4–25.7), 24% (95% CI 12.4–39.9), 35% (95% CI 15.9–59.6), and 48% (95% CI 19.1–77.8) for a BMI 40, 45, 50, and 55 kg/m2, respectively. The odds of being converted increase 11% for every 1 unit increase in BMI after adjusting for sequence of operation. In addition, the feasibility of completing a robotic aortic lymphadenectomy is 67% and approximately 35% with a BMI 45 and 50 kg/m2 or more, respectively.
Three patients were transfused 10 units of blood due to intraoperative or postoperative anemia as a result of the surgical procedure. Thirteen percent of patients (11 of 85) experienced complications. Perioperative injuries include an inferior vena cava injury that was endoscopically controlled and two patients with gastrointestinal events. One of these patients (BMI 50 kg/m2) developed shock after an unrecognized small bowel injury that required exploration and small bowel resection and resulted in a 47-day hospital stay. Another patient presented postoperatively with a bowel obstruction due to robotic trocar evisceration that resolved after exploration and repair of the port site.
A delay in discharge was seen for one patient with pneumonia and in another patient with postoperative hemorrhage who was hemostatic after the dissection; however, a 9.1 mg/dL drop in hemoglobin was noted postoperatively (computed tomography scan revealed a 6-cm hematoma). Her hemoglobin stabilized quickly and she did not require transfusion. The following four readmissions were noted: severe Clostridium difficile colitis, chest pain, vaginal leaking of peritoneal fluid, and pelvic abscess. In another patient, a seroma in the midline port resulted in an unscheduled outpatient evaluation. An additional patient presented with a fascial dehiscence after the midline camera port site required extension to retrieve an enlarged uterus that failed vaginal extraction.
In this report, we describe our series of clinical stage I or occult stage II endometrial carcinoma patients undergoing robotic exploration with the intent of comprehensive staging. Limited published data exist regarding the robotics platform in endometrial cancer, which includes three studies totaling 12 cases and only five patients that clearly underwent comprehensive staging.15–17
Many gynecologic oncologists advocate minimally invasive surgery as the procedure of choice for endometrial carcinoma.11,18–20 Most of the disadvantages of the laparoscopic approach are related to physician or patient factors. A steep learning curve exists in laparoscopy that is due to counterintuitive motion, surgeon training, and experience. Mastering laparoscopic lymphadenectomy is difficult and requires patience, time, and acceptance of a longer initial learning curve for surgical trainees and mentors who train them. Patient factors include obesity, previous surgeries, or a large uterus.
Morbid obesity is not only a major risk factor for developing endometrial cancer, but is also a leading limitation to laparoscopic staging. The average BMI in studies reporting successful laparoscopic staging ranges from 25–40 kg/m2.7,18–22 In fact, laparoscopy may be the preferred approach to these patients who would potentially benefit the most from minimal-access surgery.19,20 Thus, appropriate obese candidates should be considered for minimally invasive surgery. Although each patient should be evaluated independently and comprehensive staging is less likely to be completed in this patient population regardless of surgical approach, our data suggests that a BMI between 50–55 kg/m2 may be a threshold for comprehensive robotic endometrial cancer staging feasibility. In our experience, patients in whom a pelvic–aortic lymphadenectomy is not feasible robotically are generally unable to be adequately completed at laparotomy.
Although the published data are limited, our study is comparable to other feasibility studies for robotic gynecologic oncology procedures in regard to most patient characteristics, estimated blood loss, lymph node count, length of stay, operative times, and complication rates.15–17 Although “complications” are variably defined and difficult to compare between publications, complication rates of 8.2–16.1% after laparoscopy for endometrial cancer seems acceptable, comparable to our robotic data, and perhaps improved over laparotomy.23–24
Our data support the surgical capability of performing a complete robotic staging for endometrial cancer. The average BMI is higher than most previously reported laparoscopic studies,6,8,18 and the majority of our patients underwent comprehensive staging. The conversion rate in the present series compares favorably to laparoscopy (23%) (Walker J, Mannel R, Piedmonte M, Schlaerth J, Spirtos N, Spiegel G, et al. Society of Gynecologic Oncologists 37th Annual Meeting 2006, abstract 22, Gynecol Oncol 2006;101:S11–12). As expected, the rate of conversion of robotic to open is higher in obese women compared with those of normal BMI.
As with any new procedure, there is a learning curve when applying the robotics platform to endometrial cancer staging, and for advanced laparoscopic procedures, this is difficult to define.23 Generally, this is dependent on the experience of the surgeon, the involvement of trainees, the thoroughness of the lymph node dissection, and patient factors. Robotics may offer a shorter learning curve for minimally invasive surgery. When compared with laparoscopy, the robotics platform enables the surgeon to more readily transfer open techniques to a minimal-access setting. This is demonstrated in the urologic literature by only 20 cases needed for proficiency in robotic prostatectomies.25 Similarly, for robotic hysterectomy pelvic–aortic lymphadenectomy for endometrial cancer, proficiency is approximately 20 cases; however, efficiency continues to improve.
We acknowledge that most surgeons learning robotic surgery have experience with laparoscopic endometrial cancer staging (Kim K, Seamon LG, Fowler JM, Cohn DE. Initial experience with abdominal, laparoscopic, and robotic endometrial cancer staging surgery: a single surgeon’s experience at the beginning of the learning curve. Proc Min Invas Robotic Assoc 2008. Third International Congress of the Minimally Invasive Robotic Association, January 24–26, 2008, Rome, Italy). Unlike the learning curve for laparoscopy,22,25,26 in which there is a decreased number of lymph nodes as well as an increased estimated blood loss noted during the initial adoption of the procedure, we did not find a statistically significant difference in these robotic outcomes. As experience increased, we noted significant improvement in times without compromise in comprehensive staging.
We speculate that although the advantages of the robotics platform, including a shorter learning curve, will eventually allow a wider application of minimally invasive surgery in a more heterogeneous population, the potential disadvantages are numerous. Sahabudin et al27 outlined several key elements to developing a successful robotics program, including adequate funding, extensive training of the surgical team, and renovation of the operating suite for robotics. This requires an adequate volume of procedures, financial commitment from the institution, well-oiled teams, surgeon expertise, and minimizing disposable instrumentation.
Minimally invasive surgical approaches are expanding in gynecology. We believe that robotic surgery represents an improvement over laparoscopy, especially for complex procedures. Nevertheless, robotics is not a substitute for meticulous surgical technique, expertise in pelvic, abdominal, and retroperitoneal anatomy, knowledge of the natural history of disease, and applying appropriate indications for surgery.
1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer Statistics, 2008. CA Cancer J Clin 2008;58:71–96.
2. Fanning J. Long-term survival of intermediate risk endometrial cancer (IG3, IC, II) treated with full lymphadenectomy and brachytherapy without teletherapy. Gynecol Oncol 2001;82:371–4.
3. Keys HM, Roberts JA, Brunetto VL, Zaino RJ, Spirtos NM, Bloss JD, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: A Gynecologic Oncology Group study [published erratum appears in Gynecol Oncol 2004;94:241–2]. Gynecol Oncol 2004;92:744–51.
4. Childers JM, Brzechffa PR, Hatch KD, Surwit EA. Laparoscopically assisted surgical staging (LASS) of endometrial cancer. Gynecol Oncol 1993;51:33–8.
5. Wong CK, Wong YH, Lo LS, Tai CM, Ng TK. Laparoscopy compared with laparotomy for the surgical staging of endometrial carcinoma. J Obstet Gynaecol Res 2005;31:286–90.
6. Fram KM. Laparoscopically assisted vaginal hysterectomy versus abdominal hysterectomy in stage I endometrial cancer. Int J Gynecol Cancer 2002;12:57–61.
7. Eltabbakh GH, Shamonki MI, Moody JM, Garafano LL. Laparoscopy as the primary modality for the treatment of women with endometrial carcinoma. Cancer 2001;91:378–87.
8. Malur S, Possover M, Michels W, Schneider A. Laparoscopic-assisted vaginal versus abdominal surgery in patients with endometrial cancer—a prospective randomized trial. Gynecol Oncol 2001;80:239–44.
9. Holub Z, Jabor A, Bartos P, Hendl J, Urbánek S. Laparoscopic surgery in women with endometrial cancer: the learning curve. Eur J Obstet Gynecol Reprod Biol 2003;107:195–200.
10. Magrina JF, Mutone NF, Weaver AL, Magtibay PM, Fowler RS, Cornella JL. Laparoscopic lymphadenectomy and vaginal or laparoscopic hysterectomy with bilateral salpingo-oophorectomy for endometrial cancer: morbidity and survival. Am J Obstet Gynecol 1999;181:376–81.
11. Fowler JM. The role of laparoscopic staging in the management of patients with early endometrial cancer. Gynecol Oncol 1999;73:1–3.
12. Seamon LG, Cohn DE, Valmadre S, Richardson DL, Jayjohn LA, Jenson C, et al. Robotic hysterectomy and lymphadenectomy for endometrial cancer: technical aspects and details of success—the Ohio State University method. J Robotic Surg 2008;2: June 10 [Epub ahead of print].
13. Cleveland WS. Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc 1979;74:829–36.
14. National Institutes of Health. National Heart, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. Available at: http://www.nhlbi.nih.gov/guidelines/obesity/ob_gdlns.htm
. Retrieved June 23, 2008.
15. Reynolds RK, Burke W, Advincula AP. Preliminary experience with robot-assisted laparoscopic staging of gynecologic malignancies. JSLS 2005;9:149–58.
16. Marchal F, Rauch P, Vandromme J, Laurent I, Lobontiu A, Ahcel B, et al. Telerobotic-assisted laparoscopic hysterectomy for benign and oncologic pathologies: initial clinical experience with 30 patients. Surg Endosc 2005;19:826–31.
17. Field JB, Benoit MF, Dinh TA, Diaz-Arrastia C. Computer-enhanced robotic surgery in gynecologic oncology. Surg Endosc 2007;21:244–6.
18. Barakat RR, Lev G, Hummer AJ, Sonoda Y, Chi DS, Alektiar KM, et al. Twelve-year experience in the management of endometrial cancer: a change in surgical and postoperative radiation approaches. Gynecol Oncol 2007;105:150–6.
19. Tozzi R, Malur S, Koehler C, Schneider A. Analysis of morbidity in patients with endometrial cancer: is there a commitment to offer laparoscopy? Gynecol Oncol 2005;97:4–9.
20. Abu-Rustum NR. CO(2) pneumoperitoneum or the Bookwalter: choose your access and exposure. Gynecol Oncol 2005;97:1–3.
21. Kueck AS, Gossner G, Burke WM, Reynolds RK. Laparoscopic technology for the treatment of endometrial cancer. Int J Gynaecol Obstet 2006;93:176–81.
22. Scribner DR Jr, Walker JL, Johnson GA, McMeekin DS, Gold MA, Mannel RS. Laparoscopic pelvic and paraaortic lymph node dissection in the obese. Gynecol Oncol 2002;84:426–30.
23. Fleisch MC, Newton J, Steinmetz I, Whitehair J, Hallum A, Hatch KD. Learning and teaching advanced laparoscopic procedures: do alternating trainees impair a laparoscopic surgeon’s learning curve? J Minim Invasive Gynecol 2007;14:293–9.
24. Gil-Moreno A, Díaz-Feijoo B, Morchón S, Xercavins J. Analysis of survival after laparoscopic-assisted vaginal hysterectomy compared with conventional abdominal approach for early-stage endometrial carcinoma: a review of the literature. J Minim Invasive Gynecol 2006;13:26–35.
25. Köhler C, Klemm P, Schau A, Possover M, Krause N, Tozzi R, et al. Introduction of transperitoneal lymphadenectomy in a gynecologic center: analysis of 650 laparoscopic pelvic and/or paraaortic transperitoneal lymphadenectomies. Gynecol Oncol 2004;95:52–61.
26. Holub Z, Jabor A, Bartos P, Hendl J, Urbanek S. Laparoscopic surgery in women with endometrial cancer: the learning curve. Eur J Obstet Gynecol Reprod Biol 2003;107:195–200.
27. Sahabudin RM, Arni T, Ashani N, Arumuga K, Rajenthran S, Murali S, et al. Development of robotic program: an Asian experience. World J Urol 2006;24:161–4.
This article has been cited 1 time(s).
© 2008 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Looking for ABOG articles? Visit our ABOG MOC II collection. The selected Green Journal articles are free through the end of the calendar year.
ACOG MEMBER SUBSCRIPTION ACCESS
If you are an ACOG Fellow and have not logged in or registered to Obstetrics & Gynecology, please follow these step-by-step instructions to access journal content with your member subscription.
Data is temporarily unavailable. Please try again soon.