Robotic Surgery In The Management Of Endometrial Cancer: A Single-Centre Experience

Introduction

Endometrial carcinoma is the most common gynecological cancer in developed countries with a global age-standardized incidence rate of 8.4 per 100,000 women.¹ In India, there has been an increase in the incidence of endometrial cancer with a total number of 13,328 estimated new cases of endometrial cancer with an estimated 5010 deaths as reported in 2018.^1,2 The changing lifestyle patterns in urban India leading to an increase in obesity, metabolic syndrome, and delayed childbearing are the causes of an increase in endometrial cancer. Surgical management has been the mainstay, traditionally with open surgery.¹

Laparoscopy has become the standard approach for the surgical treatment of endometrial cancer as a large, prospective, randomized trial LAP2 which compared laparotomy with laparoscopy in the staging of endometrial cancer showed that laparoscopy was associated with fewer postoperative complications, a shorter hospital stay with similar overall survival in both groups (laparoscopy and open laparotomy)at 89.8%.³ Increasing implementation of advanced minimally invasive surgeries resulted in the advancement of laparoscopic surgery and the use of robots in surgical practices in the early 90s. However, only after the Food and Drug Administration (FDA) approval for the use of the da Vinci system in gynecology in April 2005, the advantages of robotic surgery for gynecological oncology procedures have emerged.⁴

The several advantages of robotic surgery include increased surgeon comfort with console ergonomics, three-dimensional vision, endo wrist instruments with seven degrees of freedom with the elimination of the fulcrum effect and tremors, increased surgical precision and dexterity, improved patient safety, and shorter learning curve. These have led to increased use of this approach by surgeons.^4,5

The robotic platform (da Vinci S, Intuitive Surgical) was made available to us in July 2015. This retrospective study aimed to assess the perioperative outcomes and role of robotic surgery in the management of endometrial cancer.

Material and Methods

We retrospectively identified 60 women who had undergone robotic-assisted ecological gynecological procedures from June 2015–September 2019 from the operation theatre (OT) register. Of these 32 women had a histological diagnosis of endometrial carcinoma confirmed by the hospital's electronic medical records (EMR). Data was accrued on preoperative parameters like age, body mass index (BMI), previous history of abdominal surgery, and comorbidities. Further, the perioperative parameters including blood loss, console time, surgical time, blood transfusion rate, total lymph node yield, and conversion rate were taken into consideration. The postoperative pain score (the VAS score), length of hospital stay, and postoperative complications were also recorded. The console time and total surgery duration which started from the time of incision to the completion of surgery were noted from the OT register. Histology reports were reviewed for the grade, stage, nodal counts, and completeness of the specimen (International Federation of Gynaecology and Obstetrics [FIGO]2009). The data was collected from the OT register, EMR, and daVinci console records and collated into MS Excel 2016 spreadsheet, and analyzed.

The diagnosis was done by pre-operative imaging which included either contrast abdominopelvic Magnetic resonance imaging (MRI) or computed tomography (CT) scan, and a hematological and biochemical workup was also performed. All the cases were discussed at the Multidisciplinary Tumor Board Meetings (MDTB) and a plan for surgery was finalized. The women were counseled adequately and written informed consent was obtained. Preoperative anesthetic fitness was obtained. They were admitted a night before the surgery to aid mechanical bowel preparation with Exelyte. All the surgeries were performed by Consultant Gynaecological Oncologists who are also certified robotic surgeons. We used the da Vinci S-HD platform with standard three robotic arm ports including the camera port. CO2 pneumoperitoneum was created with Verres intraumbilically and through palmers point for women with previous abdominal surgeries. Intraoperatively intraabdominal pressure of 14 mm Hg was used. A 12 mm port for the camera was placed 2–3 cm superior to the umbilicus. Two 8 mm robotic ports were placed 9 cm from the camera port bilaterally in line with the umbilicus. In addition, two 12 mm ports were placed on the right side, one in the upper right quadrant and the other in the outer third of the right spino-umbilical line for assistance. A pair of monopolar scissors (Intuitive Surgical Inc) was used in the right robotic arm while fenestrated bipolar forceps (Intuitive Surgical Inc) were used on the left robotic arm. The standard 10 mm zero-degree telescope was used. Steep trendelenberg position usually up to 45 degrees was preferred. In cases requiring lymph node dissection, the procedure started with bilateral pelvic lymphadenectomy (including common iliac, external iliac, internal iliac, lateral, intermediate, and obturator nodes on either side) and then proceeded to hysterectomy. The uterine artery was identified early in the operation, isolated, and diathermised/clipped at the origin. A vaginal intrauterine manipulator was used in our series. Colpotomy was done with monopolar scissors over the vaginal tube. The hysterectomy specimen was retrieved through the vagina and nodes were retrieved through the right lateral 12 mm port. After specimen retrieval, the vagina was obliterated with a wet mop to prevent loss of pneumoperitoneum. Vault closure was performed robotically with a 2-0 barbed suture V lock. In cases where the cost was a constraint, we used only 2 robotic instruments (i.e. monopolar scissors and fenestrated bipolar forceps), and vault closure was done vaginally with 2-0 vicryl after retrieval of the surgical specimen. All procedures were completed with single docking. Only in cases where more than one quadrant surgery was performed i.e. nephrectomy, double docking was done with the change of the port positions as per requirement. The estimated blood loss (EBL) was determined by the blood collected in the suction apparatus and the gauze pieces soaked. All women were followed up a week after surgery with histology and were advised standard 5 years follow-up at the center.

Results

The patient characteristics have been described in Table 1.

Of the 32 procedures, 2 were completion surgery with robot-assisted bilateral pelvic lymph node dissection (BPLND) (1 with bilateral salpingo-oophorectomy [BSO]) on women who had undergone Laparoscopic Hysterectomy with Bilateral Salpingo-Oophorectomy (TLH BSO) and Vaginal Hysterectomy for Dysfunctional uterine bleeding (DUB) elsewhere and had postoperative histology of endometrial cancer. Two women who had dual/synchronous malignancy (endometrial carcinoma and renal cell carcinoma [RCC]; 1 clear cell, 1 RCC) underwent nephrectomy (1 partial, 1 radical) as an additional procedure i.e. multi-quadrant surgery. Robot-assisted hysterectomy with bilateral pelvic lymphadenopathy was the most frequently performed procedure (24 women) followed by robotic hysterectomy with bilateral salpingo-oophorectomy only (6 women), and bilateral pelvic lymph node dissection (2 women; 1 with BSO). (Table 2)

The mean console time was 154.125 min (range 80–288 min) while the mean surgical time recorded was 185 min (range 100–340 min). The mean blood loss recorded in our series was 115ml (range 50–600 ml) comparable with other studies (Table 3). Only one complication (vascular) was observed. The right external iliac vein was injured during pelvic lymphadenectomy and it was repaired robotically with 4-0 sutures. The complication was not related to the learning curve. The transfusion rates and conversion rates were zero. The mean hospital stay was 2.8 days (range 2 – 4 days) (Table 3).

The histopathology reports were reviewed for the cell type, grade, stage, myometrial invasion, lymph node count, and other parameters. The most common surgical stage (FIGO 2009) was I A (23 patients) followed by I B (8 patients), II A (1 patient); the most common cell type was endometrioid (30 patients) followed by papillary serous (2 patients). The most common grade was grade 2 (16 patients), followed by grade 1 (13 patients), and grade 3 (1 patient). The mean lymph node yield was 14 (range 4–30).

The average duration of follow-up was 47.875 months (SD 32.52). Four women (12.5%) underwent adjuvant radiotherapy. Of all the women who were followed up, 3 (9.37%) developed recurrence, 2 (6.25%) women had vault recurrence after 2 years of follow-up while 1 (3.12%) woman developed early intraperitoneal recurrence– unfortunately, she succumbed to it (mortality 3.12%).

Discussion

The GOG LAP² trial firmly established the laparoscopic approach as the standard of care in the management of endometrial cancer.³ However, the steep learning curve of operative laparoscopy has been a deterrent in its wider applications. The robotic platform first came into use in 2005, nearly 15 years after, the laparoscopic approach was gaining ground among gynecological oncologists.^4,5 Boggess et al. reported the supremacy of robotic platforms in the management of gynecological malignancy as early as 2008.⁶

Most surgical oncologists in our country have used the open approach preferentially to manage gynecological oncology cases. The shorter learning curve and the technology mimicking open surgery have made its applications easier with many surgeons being able to adopt the technique. Over the past 15 years, as many as 70 centers are offering robotic surgery with 80 robots in India. However, the high investment cost and high cost of disposables have led to comparisons with the laparoscopic approach; questioning the very advantages of robotic surgery over it. Some authors feel that the laparoscopic approach with its nuances such as 3D vision technology can give comparable results compared with robotic,surgery at a lower cost.^7–9 The recent trial on the Laparoscopic Approach to Cervical Cancer (LACC trial; Clinicaltrials.gov NCT00614211) in October 2018, significantly challenged the safety of a minimally invasive approach to radical hysterectomy for early cervical cancer.¹⁰ Owing to the emerging evidence and understanding, the most optimal use of robotic surgery in gynecological oncology is identified in the endometrial cancer patient cohort with morbid obesity, many comorbidities, coupled with complex anatomy as seen from our patient cohort.

The 3D vision improves the visualization of difficult anatomy of the pelvis thus helping in the early identification and ligation of uterine arteries at their origin with endo wrist technology significantly reducing blood loss as is evident from our series. The mean blood loss in our study was 115 ml which was comparable with other studies.^10–13 Also, lymph node dissection can be successfully performed with ease due to 3D technology and 7-degree freedom at the wrist joint.^13–15 In our study the mean lymph node yield was 14 which appears to be adequate and satisfactory and comparable with other studies.^13–15

We have performed two quadrant surgery in a single sitting by a single surgeon (i.e. robotic hysterectomy and nephrectomy) in 5 women with gynecological cancer so far; however, only two women with the diagnosis of endometrial cancer have been included here. The operating time of nearly 4 hr 40 min though prolonged, led to significantly reduced surgeon fatigue thus emphasizing the advantage of robots in multi-quadrant surgery.

The mean console and operative time in our case series were 154.125 mins and 185 mins, respectively which was comparable with the studies conducted by Boggess et al. ⁹ and Bell et al. ¹²

A low intraoperative complication rate (3.12%) was observed in our series. This rate was comparable to other studies.^13–15 Nearly half of the patient cohort (15/32) in our study had previous abdominal surgeries with intraabdominal adhesions. Despite this, there were no conversions in our series due to the improved precision of robotic surgery. Selection bias in terms of disease volume (in most cases early endometrial cancers stage I A ) could have impacted the complication rate. The 24-hour post-operative pain score was low (2-3/10 on VAS). Besides these outcomes, the hospital stay was relatively shortened and the patient could be discharged after 48 hours which corresponded with studies by Seamon et al. ¹⁵ All the histopathology reports reviewed showed the completeness of specimens which justified that no intrauterine manipulator-related complications or port site metastases were observed in our series.The initial follow-up results in our series have reconfirmed its safety, with only 3 recurrences one of which proved fatal. The remaining 31 women remain alive and well at the time of this report.

The limitations of our study are its retrospective nature, the small size of the patient cohort, the lack of comparative laparoscopic/laparotomy data, and cost analysis data from our institution with the strength being the long follow-up.

Conclusion

The results from our short case series are encouraging for robotic surgery in endometrial cancer. However, results from larger centers should be compiled to have robust and meaningful data in the Indian context. Financial constraints pose a major challenge in our setting and are a major limiting factor in the wider applications of robotics. Moreover, insurance coverage does not provide for the entire cost of robotic surgery yet. Newer robotic platforms and hopefully an indigenous platform in the future could help bring down the cost of surgery. Until then robotic surgery in endometrial cancer could be used in selected cases with morbid obesity, previous multiple abdominal surgeries, and multi-quadrant surgery. Technological advancements in robotics and their applications such as single-port robotic surgery and telesurgery are exciting possibilities for the future.

Conflict of interest

The authors have no conflicts of interest to declare.