[Analysis of robotic natural orifice specimen extraction surgery on 162 cases with rectal neoplasms]

Objective: To explore the safety and feasibility of da Vinci robot surgical systems in natural orifice specimen extraction surgery (NOSES) for rectal neoplasms. Methods: A descriptive cohort study was used. Inclusion criteria: (1) age ≥18 years old; (2) diagnosis of rectal cancer by biopsy via colonoscopy or benign neoplasm locating in rectum that could not be resected locally through the anus; (3) R0 resection can be achieved by preoperative evaluation; (4) the CDmax (maximum circumferential diameter) was ≤5 cm or specimens could still be extracted from the anus despite a CDmax exceeding 5 cm but was along the longitudinal axis of the rectum. Exclusion criteria: (1) emergency operation due to gastrointestinal obstruction, perforation, or bleeding; (2) distal metastasis, induding lung, bone, or liver, that could not be resected simultaneously; (3) history of abdominal surgery or any other contraindications for robotic surgery. Clinicopathological data of 162 patients with rectal neoplasms who underwent robotic NOSES at the General Surgery Department of the Second Xiangya Hospital of Central South University from March 2016 to July 2019 were retrospectively collected. Of 162 patients, 94 were male and 68 were female; the average age was (57±13) years; the average BMI was (23.5±3.2) kg/m(2); the average distance from tumor to the anal verge was (8.2±2.9) cm. Five trocars were used to perform total mesorectal excision (TME), and the descending colon artery was preserved. Sterile endoscope sleeve for the specimen extraction was inserted into the pelvic cavity through the anus, and the resected specimen was pulled out through the sleeve. Outcomes of safety (operation time, intraoperative blood loss and postoperative morbidity of complication) and oncological outcomes (number of lymph nodes harvested, rate of lymph node metastasis and rate of positive resection margin) were collected. Results: All the 162 cases completed robotic NOSES successfully with no conversion to laparotomy. The average operation time was (188.7±79.8) minutes; the average blood loss was (47.1±33.2) ml; the average and the maximum CDmax of specimens were (3.4±1.5) cm and 12 cm respectively. A total of 154 patients underwent robotic TME. One underwent robotic TME plus resection of liver metastasis; one underwent robotic TME plus partial transverse colectomy; two patients underwent robotic TME plus ovariectomy; another two underwent robotic TME plus hysterectomy; one patient underwent robotic TME plus left partial nephrectomy due to renal angioleiomyoma; another one underwent robotic TME plus ureteral repair due to intraoperative injury of the left ureter. All the specimens were extracted through the anus. Protective ileostomy was performed in 6.8% (11/162) of the patients. The average number of lymph node harvested was 14.9±5.1. According to pathological reports, 156 neoplasms were adenocarcinoma. Tis stage was 1.3% (2/156), T1 stage was 9.0% (14/156), T2 stage was 26.3% (41/156), T3 stage was 35.9% (56/156), and T4 stage was 27.6% (43/156). Lymph node metastasis accounted for 34.6% (54/156), and simultaneous liver metastasis was observed in one case. Circumferential resection margins (CRMs) and upper and lower resection margins were negative in all the patients. The average postoperative feeding time and postoperative hospital stay were (4.2±4.1) days and (11.4±7.7) days, respectively. Postoperative morbidity of complication was 12.3% (20/162). The incidence of anastomotic leakage was 4.9% (8/162), of which only 4 cases (2.5%) received ileostomy. Within postoperative 90-day, no anal dysfunction or death were found. Conclusion: Robotic NOSES for rectal neoplasms is safe and feasible.