Robotic single‐site versus multiport radical hysterectomy in early stage cervical cancer: An analysis of 62 cases from a single institution

Abstract Background This study aimed to compare the surgical outcomes and cost of robotic single‐site radical hysterectomy (RSSRH) versus robotic multiport radical hysterectomy (RMPRH) with pelvic lymph node dissection in early stage cervical cancer. Methods Sixty‐two patients with early stage cervical cancer were recruited between November 2011 and July 2017 and underwent RSSRH (20 patients) and RMPRH (42 patients) for early stage cervical cancer using the da Vinci Si Surgical System (Intuitive Surgical). Results There were no significant difference between the two groups in most of parameters. However, postoperative hospital discharge and total hospital costs for RSSRH were significantly shorter than RMPRH (both p < 0.001). However, lymph node retrieval of RMPRH was significantly higher than RSSRH in (18.0 vs. 9.5, respectively; p < 0.001). Conclusions RSSRH has comparable surgical outcomes to the RMPRH method. RSSRH could be considered a surgical option in a well‐selected patient group.

(LESS) is one of the newest innovations in MIS and has several potential applications in gynaecologic oncology. Despite the more advanced surgical methods, LESS presents various surgical challenges, including a limited range of motion due to the parallel angle of surgical instruments, difficulty in manipulating a flexible camera and positioning of surgical instruments in a limited space through a small skin incision. 7,8 To overcome these surgical challenges, robotic single-site surgery (RSSS) platforms have been developed recently. RSSS, which is one of the best advanced forms and an FDA-approved alternative to LESS, emerged in 2013 as an attempt to (1) combine easier manipulation; (2) improve a magnified three-dimensional view; and (3) provide wider range of motion with wristed instruments acquired with robotic technology using a single skin incision. 9,10 However, there is still a lack of studies addressing the surgical outcomes of robot-assisted operations, since robot-assisted radical hysterectomy is a relatively novel surgical technique and its surgical outcomes have not been investigated in randomised controlled trials. 11 The aim of the study was to compare the surgical outcomes and total costs of robotic single-site versus multiport radical hysterectomy with PLND in early stage cervical cancer. The stages of cervical cancer were classified according to the International Federation of Gynecology and Obstetrics (FIGO) classification revised in 2018. 12 Patients with a preoperative diagnosis of early cervical cancer (FIGO stage IA1 with lymphovascular space invasion, IA2 and IB1) were selected. There was no specific contraindication to robot-assisted radical hysterectomy operation, except for the following conditions: (1) evidence of metastasis to other organs in preoperative imaging and (2) high-risk pathology (ex. neuroendocrine tumour) on preoperative cervical biopsy. Although there were no restrictions for patients related to body mass index (BMI) or previous abdominal surgeries, robotic multiport radical hysterectomy (RMPRH) was performed in patients with a BMI of 30 or higher or a history of previous surgery with high probability of intrapelvic adhesion.

| Patients and basic characteristics
The total operative time was subdivided as follows: (1) preparation time (the time from the first incision to the end of port replacement); (2) docking time (the time from insertion of the robotic arms through trocars to introduction of the robotic instruments); (3) console time (the real surgical time, measured from the first manipulation by the surgeon to the last manipulation to repair the vaginal cuff); and (4) closure time (the time from release of docking to finishing the skin suture). The total operation time was calculated from the preparation time to closure time. Intraoperative parameters included estimated blood loss, requirement for blood transfusion, conversion to multiport laparoscopy or laparotomy and intraoperative complications. Postoperative parameters included length of hospital stay, total hospital charge, haemoglobin change, lymph node retrieval and postoperative complications.  13 The resected uterus and adnexa were removed through the vagina, and the vagina cuff was repaired with a continuous suture by V-LocTM (Covidien), which is a unidirectional barbed suture with a curved needle in all cases.

| Robotic single-site radical hysterectomy
This single-site instrument is a multiple-channel single port composed of a robotic, 8.5-mm, high definition with a three-dimensional endoscope, two types of curved robotic cannulas and one 5-mm accessory cannula. A single 2.5-cm vertical periumbilical incision was usually made to the left of the umbilicus, performed using an open Hasson approach. The lubricated single-site port was then inserted into the abdominal cavity, and the lower rim of the single-site port was clamped using atraumatic Kelly forceps. After checking the other organs, pneumoperitoneum was made at a pressure of 12 mmHg with carbon dioxide. A trocar for the camera and a three-dimensional, 8.5-mm endoscope (30°) were inserted carefully along the endoscopic cannula and the abdominal cavity was inspected to confirm the feasibility of the RSSRH operation. The position of the da Vinci robotic body was situated between the widened patient's feet. One 5 � 250-mm curved cannula (Arm 2) was inserted through the designated lumen until the end line of the cannula was visible in the field of endoscope vision.
While the other cannula (Arm 1) was inserted using the same method, the already inserted cannula was held by the assistant to prevent displacement. Finally, two curved cannulas were positioned in cross position to avoid collision, and then a monopolar hook (Arm 2) and bipolar fenestrated bipolar grasper (Arm 1) were placed in each arm of the cannulas for the right-handed surgeon.
The assistant's 5-mm accessory cannula was inserted to perform several functions in the procedure 1 : suction and irrigation 2 ; coagulation and cutting simultaneously by the LigaSure 5-mm blunt tip (Covidien) 3 ; and insertion of V-LocTM 2-0 sutures (Covidien), which is a unidirectional barbed suture used exclusively with a straightened needle.

| Robotic multiport radical hysterectomy
In the RMPRH, a 12-mm trocar was placed at 5 cm cranial to the umbilical level after the creation of a pneumoperitoneum to Intuitive endoscope was used during all operations.

| Statistics
The data collected from hospital medical records were reported as median or percentages for continuous and categorical variables, respectively. Differences between the RSSRH and RMPRH groups were tested using the χ 2 test, t test and analysis of covariance (ANCOVA) tests for categorical and continuous variables, respectively. p-values of less than 0.05 were considered statistically significant. Data were analysed using SPSS version 19.0 (SPSS).

| RESULTS
The basic characteristics of the patients are shown in Table 1.
Overall, the descriptive characteristics of the two groups were similar. The median age of patients in the RSSRH group was higher than RMPRH, but the difference was not statistically significant (50.5 vs. 46.0 years, respectively; p = 0.13). In both groups, a history of caesarean section was reported by approximately 30% of patients (30.0% in RSSRH; 35.7% in RMPRH), but there were no cases of conversion to either laparoscopy or laparotomy in both groups. The most common histologic type was squamous cell carcinoma (65% in RSSRH; 66.7% in RMPRH) and the most common FIGO staging was IB1 (80% in RSSRH; 76.2% in RMPRH).
The operative outcomes of the patients are shown in Table 2.
Overall, there was no statistical difference between parameters associated with intraoperative outcomes and there were no major intraoperative complications in either groups. The median total Oncological outcomes between the two surgical methods were also explored and these were not statistically different ( Table 3). The median follow-up time was 38 months. There was no significant difference between the two groups in the risk factor related to recurrence in postoperative biopsy. Adjuvant therapy was confirmed in approximately 30% in both groups, and recurrence was confirmed in one patient in the RSSRH group and in two patients in the RMPRH group. All three patients with recurrence died, and one of them who underwent RMPRH died due to small bowel perforation during chemoradiation.
Tables 4-6 are re-verified by using the ANCOVA test for statistical differences in Table 2 such as postoperative hospital discharge, total hospital charge and lymph node retrieval. Confound variables were age, BMI, number of adjuvant treatment and number of recurrence and these confound variables were corrected to compare the results according to the surgical method. The results showed significant statistical differences between the two groups for all three parameters. Specifically, the postoperative hospital discharge and total hospital charge were lower in the RSSRH group whereas lymph node retrieval was larger in the RMPRH group (p < 0.001). JANG ET AL. However, most of the reported gynaecological cancers were endometrial cancers, and only a few studies have reported results on the use of the RSSS methods in patients with cervical cancer. [15][16][17][18] Hence, we have applied the RSSS method with pelvic node dissection to low-risk cases with gynaecologic malignancies, such as early stage cervical cancer. In our study, RSSRH showed similar or better intraoperative and postoperative results than RMPRH, except for lymph node retrieval. There was no difference in estimated blood loss, blood transfusion, total operation time, haemoglobin change and overall complications, which is an indicator of the safety of the surgical method. In addition, there was an advantage in terms of postoperative hospital discharge and hospital charge. The ANCOVA test was applied to test statistical differences between the two groups.
After correction for potential confounding factors linked to the performance of the two surgical methods in different periods and to the differences in sample size between the two groups. Overall, the results showed that the RSSRH method has better results in other parameters, except for lymph node retrieval.
However, there are several challenges when performing RSSRHs.
First, the RSSRH may present high technical difficulties for the operator. In our centre, we started performing RSSRH in 2015, but our centre has performed since 2013 more than 300 cases of RSSS in benign and other gynaecologic malignant conditions. The notable surgical experience derived from the performance of a high number of RSSS procedures may also be linked to operation time. This factor may explain why we observed a shorter time for the RSSRH method but no difference in operation time, but the same pattern was not seen for the RMPRH method. An important limitation of the RSSRH is that there is less lymph node retrieval compared RMPRH procedures. This result suggests a limitation in accessing the retroperitoneal space for PLND compared to multiport settings, even though surgeons are provided with a wider range of motion with wristed instruments in the single-port setting acquired with robotic technology. More importantly, this result can affect the oncologic

| CONCLUSIONS
This study has showed that the RSSRH procedure was safer, more feasible, cost-effective and had better short-term perioperative outcomes than RMPRH. This technique could also be used to train residents and surgical fellows in well-selected cases, although longterm rates of complications and postoperative radiotherapy or chemotherapy associated with the procedures need to be explored.
Randomized trials are needed to determine whether robotic singlesite techniques may offer clinical advantages over conventional procedures. Abbreviations: ANCOVA, analysis of covariance; BMI, body mass index; RMPRH: robotic multiport radical hysterectomy; RSSRH: robotic singlesite radical hysterectomy.