Comparison of short-term outcomes between Chinese surgical robot “Micro Hand S” assisted and laparoscopic total mesorectal excision for rectal cancer

Background As the rst domestically produced Chinese surgical robot for clinical use, the Micro Hand S robot has been applied to total mesorectal excision (TME) for rectal cancer in our center since 2017. The aim of this study was to evaluate the safety and feasibility of the Micro Hand S robot-assisted TME (R-TME) in comparison with laparoscopic TME (L-TME). Methods Between May 2017 and February 2018, patients who underwent R-TME and L-TME in the Third Xiangya Hospital were prospectively included in this study. The data including intraoperative outcomes, postoperative outcomes, pathological outcomes, and functional outcomes were analyzed and compared. Results Totally, 30 patients underwent L-TME and 21 patients underwent R-TME. The patient characteristics of the two groups were similar. Regarding intraoperative and postoperative outcomes, the R-TME group had signicantly less blood loss (95.2 vs. 130.0 ml, p = 0.022), shorter time to rst atus (2.1 vs. 2.9 d, p = 0.017) and rst liquid intake (2.8 vs. 3.7 d, p = 0.027) than the L-TME group. There were no signicant difference in the operative time, conversion, protective ileostomy, length of hospital stay, pain score, complications, and 30-day mortality between the two groups (p > 0.05). In addition, the pathological and functional outcomes also did not differ signicantly. Conclusion The Micro Hand S robot-assisted total mesorectal excision is safe and feasible for rectal cancer. However, prospective and large sample size trials are needed to further conrm this conclusion. protective ileostomy), postoperative outcomes (time to rst atus, time to rst liquid intake, length of hospital stay, pain score, postoperative complications and 30-day mortality), pathological outcomes (lymph nodes, proximal/distal/circumferential margins, histological differentiation, quality of TME) and functional outcomes (the Wexner Incontinence score, the International Prostate Symptom Score-IPSS, and the 5-item version of the International Index of Erectile Function-IIEF-5).


Introduction
Since the laparoscope was rst applied in colorectal diseases in 1991 [1], laparoscopic surgery for rectal cancer has been adopted by more and more medical centers. Laparoscopic surgery improves short-term outcomes such as better cosmesis, faster recovery and shorter hospital stay than open surgery [2].
Meanwhile, the laparoscope imposes technical challenge for surgeons, including poor exibility, unavoidable physiological tremor and inferior ergonomics. Surgical robotic system overcoming the disadvantages of the laparoscope are developed and introduced in clinical use. Nowadays, the da Vinci surgical robotic system, approved by the United States Food and Drug Administration (FDA) in 2000, is the only commercial surgical robot system. It demonstrates excellent performance, especially in a narrow space such as during rectectomy [3]. To deliver more value to Chinese patients, Central South University in cooperation with Tianjin University developed the rst domestically produced Chinese minimally invasive surgical robot system, Micro Hand S, in 2013 ( Fig. 1). It is a master-slave robot that consists of a surgeon console, a slave cart, and a stereo image viewer [4]. Similar to the da Vinci surgical robot, the Micro Hand S surgical robot demonstrates 3D visual imaging, motion scaling, tremor ltering, and wristed instrumentation with seven degrees of freedom. The Micro Hand S was successfully applied in some simple surgeries including acute appendicectomy, gastric perforation repair, cholecystectomy [4].
Recently, our center performed more complicated surgery such as total mesorectal excision (TME) with the Micro Hand S surgical robot (R-TME). In order to evaluate the safety and feasibility of R-TME, we taken laparoscopic TME (L-TME) as a reference treatment in the present study. For the rst time, we compared the short-term outcomes of R-TME and L-TME for rectal cancer in terms of intraoperative outcomes, postoperative outcomes, pathological outcomes and functional outcomes.

Patients And Methods
The prospectively collected data of all patients who underwent R-TME and L-TME were reviewed retrospectively between May 2017 and February 2018. A preoperative colonoscopy with biopsy was performed in all patients and rectal cancer was histologically proven. A pelvic MRI examination was preoperatively performed to evaluate the clinical stage. Neoadjuvant chemoradiation therapy (CRT; 50.4 Gy in 25 fractions for 5 weeks and 5-uorouracil based chemotherapy) was administered if the circumferential resection margin (CRM ≤ 1 mm) was suspicious, threatened, or positive or if lymph nodes that escaped the TME plane were detected by MRI. After the completion of neoadjuvant CRT, a pelvic MRI examination was performed for restaging and surgery was performed 6-8 weeks later. We excluded the patients who underwent palliative resection, concurrent resection, emergency operation and patients with double or recurrent tumor. Patients with distant metastasis, severe dysfunction of vital organs, history of abdominal and/or pelvic surgery were also excluded from this study. The selection of laparoscopic or robotic surgery was made based on patient's decision after an full explanation of each procedure. All patients were followed at least 6 months.
The data included patient demographics, tumor characteristics (tumor size, distance from anal verge, clinical TNM stage), intraoperative outcomes (blood loss, operative time, conversion rate, blood transfusion, protective ileostomy), postoperative outcomes (time to rst atus, time to rst liquid intake, length of hospital stay, pain score, postoperative complications and 30-day mortality), pathological outcomes (lymph nodes, proximal/distal/circumferential margins, histological differentiation, quality of TME) and functional outcomes (the Wexner Incontinence score, the International Prostate Symptom Score-IPSS, and the 5-item version of the International Index of Erectile Function-IIEF-5).
Conversion to open surgery was de ned as transformation from the initial approach (robot or laparoscope) to open surgery except specimen extraction. A postoperative complication was de ned as adverse event within 30 days after the operation. All postoperative complications were recorded and graded using the Clavien-Dindo classi cation, with grade III or above considered to be a severe complication [5]. Then, the comprehensive complication index (CCI) was calculated based on the Clavien-Dindo classi cation [6]. All histopathological examinations were performed by an experienced pathologist. An involved distal resection margin (DRM) was identi ed if malignant cells were found at the cutting edge by microscopy. Accordingly, an involved circumferential resection margin (CRM) was identi ed if malignant cells were found within 1 mm or less from the circumferential resection margin by microscopy [7]. The quality of TME was graded according to the protocol proposed by Quirke and classi ed macroscopically as complete, nearly complete, or incomplete [8]. Functional outcomes were evaluated at pre-and post-operation through questionnaires. The sphincter and urinary function were evaluated using the Wexner Incontinence and IPSS scores (a higher score indicating worse function on a scale of 0-20 or 0-35, respectively). The sexual functions was assessed using IIEF-5 score (a lower score indicating worse function on a scale of 0-25).
All procedures were performed by a single surgeon who had performed more than 100 laparoscopic colorectal resections. Informed consent was obtained from all patients. The study was approved by the Ethical Committee of the Third Xiangya Hospital and was registered on clinicaltrial.gov (NCT02752698).

Robotic procedure
We used a three-armed Micro Hand S surgical robot and a four-port technique. The arrangement of the ports are displayed in Fig. 2. A 12-mm trocar was inserted through supraumbilical port for the 3D camera (Camera port). Two special 10-mm trocars were inserted through two operating ports, located at 5-8cm lateral to the right side of the umbilicus (Port A) and 3-4 cm above the intersection of the umbilicus and the left anterior axillary line (Port B), to connect the right/left robotic instrument, respectively. A 5-mm trocar was inserted through assistant port located at 10 cm above of the right operating port (Assistant port). We adopted a total robotic technique and followed the principle of TME in all cases. After the establishment of the pneumoperitoneum, the patients were placed in lithotomy position with 20-30°T rendelenburg and up to 15-20° right tilt. Then the Micro Hand S surgical robot was docked at the left side of the patient by adjusting the location of the three passive arms of the patient cart. First, we mobilized colon and ligatured of the inferior mesenteric vessels. The sigmoid and descending mesocolon were dissected from medial to lateral. Medial mobilization started with the sigmoid mesocolon. The left ureter was identi ed and preserved. The inferior mesenteric artery/vein (IMA/IMV) were respectively divided at its origin or at the inferior margin of the pancreas. When the splenic exure was completely dissected from posterior tissue, we then turned to lateral mobilization until the completely mobilization of sigmoid and descending colon and splenic exure. Second, we dissected the pelvic cavity, rstly the posterior dissection, then the anterior, and lastly the lateral dissection. Posterior dissection progressed along the space between the mesorectal fascia anteriorly and the presacral fascia (Waldeyer's fascia) posteriorly until to the level of the fourth sacral vertebra. The hypogastric nerves were identi ed and preserved carefully. Anterior dissection started with the incision of the anterior peritoneal brim and continued over the plane of Denonvilliers' fascia. Lateral dissection progressed until to the level of the lateral ligaments of the rectum. Once the dissection was completed, the robot was undocked and the rectum was transected using an Endo-GIA. The specimen was retracted through an abdominal incision. A circular stapler was inserted through the anal and an intracorporeal end-to-end anastomosis was created under laparoscope. An ileostomy was created upon a surgeon's decision during the operation if necessary. Pictures of the operation are shown in Fig. 3.
Laparoscopic procedure L-TME was performed with ve working ports according to general standardized laparoscopic principles. The standard operation method from the robotic procedure was followed, including the ligation of the mesenteric vessels, dissection of sigmoid mesocolon and mesorectum, and creation of an anastomosis.

Statistical analysis
Data analysis was performed using SPSS 20.0 software. Continuous variables with a normal distribution were expressed as the mean ± standard deviation () and compared using Student's t test. Continuous variables with abnormal distribution were expressed as median (interquartile range, IQR) and compared using the Mann-Whitney U test. Categorical variables were expressed as the number of cases (%) for categorical variable and compared using Pearson's chi-square or Fisher's exact test. The differences were considered statistically signi cant at p < 0.05.

Patient characteristics
A total of 30 patients in the L-TME group and 21 patients in the R-TME group were included. All patients were discharged successfully. One patient in the L-TME group and two patients in the R-TME group received neoadjuvant CRT (p = 0.073). The mean distance from the anal verge of the R-TME group was higher than the L-TME group, but there was no statistically different (8.2 vs. 7.1 cm, p = 0.082). The baseline patient characteristics of the two groups are displayed below (p > 0.05) ( Table 1).

Perioperative outcomes
The intra-and postoperative outcomes are listed in Table 2. No deaths or readmissions or reoperations in either group occurred within 30 days after operation. Low anterior resection (LAR) was performed in the majority of the two groups (90.0% vs. 95.2%, p = 0.876). The R-TME group was associated with less blood loss (95.2 vs. 130.0 ml, p = 0.022), a shorter time to rst atus (2.1 vs. 2.9 d, p = 0.017) and a shorter time to rst liquid intake (2.8 vs. 3.7 d, p = 0.027) than the L-TME group. The average operative time of the R-TME group was longer than that of the L-TME group; however, the difference was not signi cant (288 .3 vs. 265.0 min, p = 0.082). One patient who initially underwent R-TME experienced an intraoperative complication of an ureter injury and was converted to open surgery; the conversion rate was similar between the two groups (4.8% vs. 0%, p = 0.412). Protective ileostomy was created in 18 patients of the R-TME group and 8 patients of the L-TME group (p = 0.124). There were no signi cant differences in terms of length of hospital stay.   Table 3. In total, 23 complications in 20 patients were identi ed: 17 patients (33.3%) had a single complication and 3 patients (5.8%) had more than one complication, but only the high-grade complications were listed. The most common complications were wound complication (9.8%), followed by pulmonary complication, urinary retention, ileus and uid collection (5.8%). The overall morbidity were 36.7% in the L-TME group and 42.3% in the R-TME group.
The frequencies of complications were similar between the two groups (all p > 0.05). According to the Clavien-Dindo classi cation, the details of the postoperative complications were as follows. In the L-TME group, 7 patients (23.3%) experienced grade I complications, 3 patients (10.0%) grade II and 1 patients (3.3%) grade III; 4, 4, 1 patients (19.0%, 19.0%, 4.8%) in the R-TME group (p = 0.822). None experienced grade IV or grade V complication. One patient in the L-TME group had a postoperative intra-abdominal abscess and underwent reoperation, which was classi ed as grade IIIb. One patient in the R-TME group had postoperative intraluminal bleeding and underwent endoscopic hemostasis, which was classi ed as grade IIIa. The rate of severe complications (≥ grade III) was 3.3% for the L-TME group and 4.8% for the R-TME group (p = 1.000). The average CCI score was 19.5 ± 9.9 and 19.2 ± 6.2 (p = 0.740).

Functional outcomes
Sixteen male patients in the L-TME group and 12 male patients in the R-TME group were included for evaluating sphincter, urinary and sexual functions at pre-operation and 3, and 6 months post-operation. The patient characteristics and functional outcomes in male patients are listed in Table 5. The Wexner Incontinence scores were higher at 3 months compared with pre-operation for the L-TME group (10.3 ± 3.8 vs. 4.1 ± 3.5, p < 0.001) and the R-TME group (9.5 ± 4.1 vs. 5.1 ± 3.8, p = 0.012). The scores then decreased, and were comparable, to the preoperative level at 6 months. The scores of both groups at the same time points were not signi cantly different (p > 0.05). The IPSS and IIEF-5 scores showed the similar trend as above. However, the IPSS scores for both groups did not decreased to pre-operation at 6 months (p < 0.05). The IIEF score was comparable to pre-operation at 6 months in the R-TME group (p = 0.136), but not in the L-TME group (p = 0.044). The follow-up period was 7-12 months. During the follow-up period, no recurrence or metastases was observed in both groups.

Discussion
Currently, the main therapy for rectal cancer is laparoscopic resection. Rectal cancer is located in a narrow and rigid space of the pelvic cavity that has many important vessels and nerves. This situation poses a great technical challenge to surgeons. Furthermore, TME as a standard operation for rectal cancer, requires complicated laparoscopic manipulations to reach the extremes of the pelvis for the complete resection of the rectum and peripheral lymph nodes [9]. The above situation with the addition of using a straight and rigid laparoscopic instrument makes laparoscopic TME surgery a highly di cult technique. The Micro Hand S robot system, the rst domestically produced Chinese minimally invasive surgical robot system, possess the following features: an 3D vision imaging system contributing to a more accurate spatial orientation, motion scaling allowing for a more precise manipulation, and a wristed instrument increasing the dexterity of the instrument. All of these characteristics facilitate TME surgery in a small space. Therefore, based on the above consideration, we selected rectal cancer surgery to evaluate the safety and feasibility of the Micro Hand S robot in it's initial phase. In the current study, we found that the R-TME group was with a less blood loss, shorter time to rst atus and rst liquid intake than the L-TME group, without compromising on other outcomes.
The patient characteristics of the two groups were comparable. Notably, the number of patients who received neoadjuvant CRT was small because selective neoadjuvant CRT was administered in this study. In fact, for tumors and lymph nodes located in the TME plane without suspicious or positive CRM, standard TME surgery can promise a curative resection, and neoadjuvant CRT is not indispensable [10].
Moreover, neoadjuvant CRT may induce tissue edema, which increases the di culty of pelvic dissection and the risk for intra-and postoperative complications [11]. Therefore, selective neoadjuvant CRT was reasonable in this study. In addition, neoadjuvant CRT decreases the number of positive lymph nodes [12], so the number of positive lymph nodes in this study (median: 2.5 in the R-TME group and 2.0 in the L-TME group) was higher than that in other series [7,13].
The R-TME group had less intraoperative blood loss (95.2 vs. 130.0 ml) than the L-TME group (p < 0.05).
To our experience, with a high resolution view, the surgeon can easily identify the potential space of anatomy with few vessels and the dissection can be performed more smoothly. The wristed instruments unlike the rigid laparoscopic instruments demonstrated exible performance during robotic surgery.
What's more, the Micro Hand S was equipped with motion scaling strategies [14] and the surgeons can downscale the motion of instruments at the ratio of 1:3, 1:6, 1:10 as needed. In the narrow pelvic cavity, the surgeon adopted a ratio of 1:3 to realize a precise dissection. All these features improved accuracy of the manipulations which were helpful to reduce blood loss. Meanwhile, these technical advantages helped to reduce bowel injury and promised a quicker recovery of the bowel.
In this study, the mean operative time was not signi cantly different between the R-TME and L-TME groups (288.3, 265.0 min), although the operative time of the R-TME group was 23.3 min longer than that of the L-TME group. However, several studies have shown that the da Vinci surgical robot-assisted TME had a signi cant longer operative time than L-TME for rectal cancer [15][16][17]. The reasons below may contribute to this fact: (i) The patients characteristics may be responsible for this fact, such as the higher tumor location in the robotic group. The distance from anal verge re ected the depth of dissection in the pelvis and surgical di culty [18]. Therefore, the relative high location of lesions may decrease surgical di culty, thus reducing the operative time; (ii) In this study the surgeon had rich experience using the da Vinci surgical robotic system for rectectomy. Meanwhile, the Micro Hand S and da Vinci robot operated in a similar manner. This situation dramatically facilitated the learning curve of the Micro Hand S robot for rectal resection. The surgeon overcame the learning curve and exhibited a stable operation performance after a few cases; (iii) The adopted trocar position, as seen in Fig. 2, permitted completion of the surgery after the initial docking of the robot and avoided a re-docking, further reducing the operative time.
Encouragingly, our data was similar to the result of the ROLLAR trial for robotic surgery (288.3 vs. 298.5 min) [19]. A low conversion rate is an important advantage of robotic surgery for rectal cancer, and the reported conversion rate ranged from 1.31-5.68% [20][21][22]. In our study, one patient was converted because of an ureteral injury due to adhesion, and the conversion rate (4.76%) was acceptable.
The comprehensive complication index (CCI) that integrated all postoperative complications has shown to be an effective measure to evaluate the differences in treatment effects [23]. The current study demonstrated that both groups had similar CCI. In the previous studies the most common complication was anastomotic leakage, but it did not occur in our study. The low anastomotic leakage rate was related in part to the low rate of neoadjuvant CRT and the inclusion of major upper-middle cases.
The quality of the TME specimen and the circumferential surgical margin (CRM) associated with recurrence rate are important prognostic factors for rectectomy [26,27]. In the current study, 19 patients (90.6%) were classi ed as the grading of "complete" in the R-TME group. Similarly, in the literature, Kim et al. [28] reported 32 "complete" patients (97.0%) and Aselmann et al. [29] reported 43 "complete" patients (97.7%). However, the high-quality evidence from the ROLLAR trail showed that 178 patients (75.4%) were classi ed as "complete" [19], signi cantly lower than our result. The differences might be in uenced by the higher tumor location in our study. A clear CRM with a distance of > 1 mm is the most acceptable standard [30]. In our study, two patients in the R-TME group had positive CRMs (< 1 mm) because of the low tumor location and there were no local recurrence during the follow-up. The pathological outcomes indicated that the Micro Hand S robot-assisted TME can be successfully performed. With the aid of the 3D view, the robotic system provided the possibility of visualizing the membrane anatomy and dissected along the space between the two layers of the pelvic fascia around the rectum, which was particularly bene cial to achieve curative resection.
Functional outcomes were assessed in three aspects: sphincter, bladder and sexual functions. Because of inadequate data about female patients, the analysis only included male patients. The scores worsened after operation and progressively improved at 3/6 months post-operation, with no signi cantly difference between the two groups. As shown by Luca et al. [31], although the features of surgical robot were helpful in identi cation and preservation of pelvic autonomic nerves, no evidence shown a superiority in performing nerve-sparing rectal cancer surgery. The scores at 6 months post-operation did not reach the level of pre-operation and need to take longer time to observe the change in the future study.
The limitations of this study were as follows. First, the study was with a small sample size and a study with large sample size is essential to gain a better understanding of robotic surgery. Second, we included the robotic cases in the learning phase, which may bias our results. Third, the Micro Hand S robot as the rst generation, is still in the clinical stage. Therefore, we did not include complicated cases in this study, such as ultra-low rectal cancers. This selection bias inevitably impaired the objectivity of the results. By performing complicated cases in the future, the advantages of robotic surgery will be fully demonstrated.
In addition, the long-term oncologic outcomes were not included in this study and should be evaluated in future studies.

Conclusion
The present study validates that TME assisted by the Micro Hand S robot is safe and feasible for rectal cancer and the Micro Hand S robot may be an alternative for rectal cancer. A randomized controlled trial needs to be conducted in the future to further con rm our results. The long-term oncological parameters of R-TME will also be assessed in future studies. PRM:Proximal resection margin; R-TME:Robot-assisted TME; TME:Total mesorectal excision; VAS:Visual analogue score Declarations Ethics approval and consent to participate:

Abbreviations
The study was approved by the Ethical Committee of the Third Xiangya Hospital.

Consent for publication:
Written informed consent for publication of clinical images was obtained from the patient. A copy of the consent form is available for review by the Editor of this journal. The "Micro Hand S" surgical robotic system: surgeon console (right), 3D imaging system (middle) and slave surgical cart (left)