Natural orifice specimen extraction surgery (NOSES) for colon cancer treatment: a double-center case-matched study of surgical and short-term postoperative outcomes

To the Editor: Natural orifice specimen extraction surgery (NOSES) is an actively developing approach for extracting specimens and performing non-incisional colorectal cancer surgery. Previous studies have shown that its effectiveness and safety in the surgical treatment of rectal cancer are similar to those of conventional laparoscopic surgery. NOSES was demonstrated to be a possible alternative approach to conventional laparoscopy, for which a mini-laparotomy is used to extract the specimen, as there were no differences in surgical characteristics and short-term outcomes.[1] Recently, new promising approaches have been demonstrated for rectal tumors extraction with bowel eversion and extra-abdominal resection.[2] While segmental resections for colon cancer have been widely adopted, more opportunities for NOSES implementation for different colon cancer locations become possible. According to Japanese guidelines, localized forms of colon cancer can be resected within 10-cm resection margins, depending on the feeding branch inflow pattern. Thus, the results of NOSES should be analyzed in the context of segmental colon resection. This study evaluated the surgical and short-term outcomes of NOSES in the segmental resections of colon cancer by comparing with those of conventional laparoscopic resection (CLR) in a multicenter retrospective study based on data from Chinese and Russian surgical centers. This retrospective matched case-control study included patients who had undergone surgical treatment for colon cancer using either NOSES or CLR between 2014 and 2020 in two surgical centers in Moscow, Russia, and Beijing, China. The selection criteria for inclusion in the study were: (1) 18–80 years of age, (2) adenocarcinoma of the colon, (3) tumors diagnosed at the preoperative stage as cTis-3, cN0–N2, (4) tumors ≤ 5 cm in diameter, and (5) a body mass index (BMI) ≤ 35 kg/m2. Patients were excluded if multiple primary cancer, or any signs of laparotomy were present. Firstly, patients were matched based on their clinical diagnosis. Secondly, patients were filtered based on sex, then age was considered (a 5-year difference was acceptable), and lastly, height and weight were matched. If the patients were similar according to these criteria, a matching pair "NOSES-CLR" was formed for further analysis. After matching the inclusion criteria, 92 pairs were formed for comparative analysis. Three pairs were excluded because of the need for conversion from CLR to open surgery: adhesions, found intraoperatively (n = 1), locally advanced tumors, hardly managed laparoscopically (n = 2). Patients diagnosed with pT4b intraoperatively were not excluded from the study if radical R0 surgery was performed laparoscopically. The final study group included 89 patients who had undergone natural orifice specimen extraction. The final control group included 89 patients who underwent conventional laparoscopy. We assessed the surgical outcomes, intraoperative and postoperative complications, and recovery rates in both groups. The patients were adjusted for sex, age, BMI, and tumor size and location. The surgeries were performed by five chief surgeons who had more than 10 years of experience in laparoscopic colorectal surgery. Preoperative staging was based on colonoscopy with biopsy and histological assessment, chest computed tomography (CT), abdominal CT and/or ultrasonography, and pelvic CT and/or magnetic resonance imaging (MRI). Patients who met the abovementioned criteria for inclusion were offered the NOSES procedure preoperatively. This study was approved by the local ethics committee of each hospital and followed the principles of the Declaration of Helsinki. Informed consent for participation in the study was obtained from each of the patients. The preoperative and postoperative assessment of the patients was the same in both groups. Although resection was performed in the same way in both groups, different specimen extraction approaches were used. The size of the removed colon was defined by margins of ≤ 10 cm on the proximal and distal sides of the colon from the tumor. The mesocolon segment associated with the resected colon was excised within the fascial envelope. The D3 lymph node area was removed along the main feeding arteries with superior mesenteric or inferior mesenteric artery dissection, where appropriate. The surgeons were free to choose the number of laparoscopic ports and the site of placement. The right-sided colectomy procedure included mobilization of the right colon and its mesentery, ligation of the feeding vessels confirmed by CT at the root of the colic vessels with D2 or D3 lymph node dissection, where appropriate with complete lymph node dissection. The colic branches that did not feed the resected part of the colon were preserved. NOSES was planned preoperatively, with the transvaginal approach indicated for most right-sided colon resections. In male patients, the specimen after right hemicolectomy was extracted via colotomy in the upper rectum. The procedures for a left-sided colectomy or sigmoid colon resection were performed with isolation and transection of the vascular pedicle with preservation of the left colic artery and/or superior rectal artery, D2 or D3 lymph node dissection, and colonic or colorectal anastomosis. In the NOSES group, the specimens were extracted through the rectal stump or through a separate vaginal incision. The transanal extraction procedure included introduction of a sterile plastic sleeve through the anus and rectum, followed by specimen elimination through the sleeve. Transvaginal extraction involved a 3 to 4 cm transverse posterior colpotomy, the introduction of a sterile plastic sleeve through the vaginal incision, and extraction of the specimen. In the CLR approach, specimens were extracted via midline mini-laparotomy or Pfannenstiel incision. The primary endpoints of the study were the surgical outcomes, recovery rates and the rate of short-term intrahospital complications. The secondary endpoints were the clinicopathological data, the procedure characteristics, length of surgery, blood loss volume, time to first flatus and defecation, length of hospital stay, the resection margin, and the lymph node harvest. Statistical analysis was performed using Pearson's chi-squared test or Fisher's exact test, and the Mann–Whitney U test was performed to determine whether the differences were statistically significant. The differences in categorical data were analyzed using Student's t-test. Quantitative data are reported as the mean and standard error mean (95% Confidence interval [CI]). The statistical analysis was performed using SPSS 26 (SPSS Inc., Chicago, Illinois, USA). No statistically significant differences were observed among the clinical and morphological characteristics of the patients [Supplementary Table 1, https://links.lww.com/CM9/B559]. NOSES was associated with a tendency for a greater number of harvested lymph nodes (25.70 ± 1.41 vs. 27.75 ± 1.54, P = 0.37). It decreased the length of surgery, but the difference was not statistically significant (165.93 ± 6.35 min vs. 181.25 ± 7.02 min, P = 0.15). Blood loss was significantly lower in the NOSES group (40.17 ± 2.81 mL vs. 52.25 ± 4.12 mL, P = 0.03). In patients treated with NOSES, the first defecation was observed earlier (3.85 ± 0.18 days vs. 4.29 ± 0.17 days, P = 0.02). The patients from NOSES group had shorter hospital stays (8.37 ± 0.44 days vs. 9.93 ± 0.46 days, P <0.01). No intraoperative complications were observed in either of the groups, and there were no significant differences in the postoperative complication rates. In the NOSES group, reoperations were performed to prevent stoma formation due of anastomotic leakage. In the CLR group, one patient underwent surgery due to anastomotic leakage. Three cases of anastomotic leakage were diagnosed in the NOSES group, and one in the CLR group. There were two cases of wound infection at the sites of the mini-laparotomy incisions in the CLR group. The incidence of short-term complications was not significantly higher in neither group. No 30-day mortality was observed in either group [Supplementary Table 1, https://links.lww.com/CM9/B559]. NOSES was introduced as a promising alternative approach to specimen extraction, allowing non-incisional radical surgery for colorectal cancer, mostly for distal sigmoid and rectal cancer surgery.[1] However, there have been few studies of NOSES for colon cancer. The implementation of NOSES in left-colon cancer treatment in a prospective randomized control study of 40 patients showed that the postoperative anorectal function, complications, and hospital stay were similar in the CLR and NOSES groups.[3] A propensity score-matched analysis of 45 patients who underwent transcolonic natural orifice specimen extraction after right hemicolectomy for ascending colon cancer showed less postoperative pain, less need for analgesia, and a lower inflammatory response, while the 3-year overall survival was the same.[4] Therefore, the use of this method in segmental colonic resection with complete mesocolon excision for colon cancer, and its oncological safety, is promising, but requires further study. The learning curve associated with NOSES, is one of the factors, limiting the wide implementation of NOSES in colorectal cancer. However, we showed that NOSES decreases the length of operations. Moreover, NOSES resulted in less blood loss in our study. Another factor limiting NOSES adoption is the need for extended resections in colon cancer surgery. In our study, the lymph node number was observed to be slightly higher in the NOSES group, despite the economic segmental resections, performed in both institutions—an average of 27 lymph nodes were found in specimens extracted through the natural orifice in the NOSES group. In addition, no positive resection margins were observed in this group. Therefore, segmental resections with natural orifice specimen extraction can be adopted. Importantly to mention, the time to first flatus and blood loss decreased, resulting in better recovery rates. However, the influence of the patients' physical status and tumor characteristics should be analyzed to prove that extraction type is a significant factor in the studied outcomes. The results of recent meta-analysis and a systematic review support the equivalence of NOSES and CLR in terms of the postoperative outcomes and survival rates, leading to better quality of life. In addition, NOSES was significantly associated with lower intraoperative bleeding (P <0.001). In our study, the primary and secondary outcomes were better in the NOSES group, which is consistent with the meta-analysis by Brincat et al.[5] Our study has several limitations. It was a retrospective comparative study, and randomized prospective controlled studies are necessary to confirm our results. In some cases, tumors extracted through a natural orifice were found to be pT4, but still included in the study as they met the inclusion criteria and were misdiagnosed preoperatively and were found to be feasible for NOSES. Thus, it is important to determine whether this approach is possible in T4 cancer to improve short-term outcomes and survival rates. In summary, this study showed that the NOSES technique is better than CLR in terms of surgical and short-term postoperative outcomes. Both approaches are associated with a similar number of complications; however, NOSES improves the surgical results and recovery rates. NOSES should be considered for wide implementation in segmental colectomies for colon cancer. Conflicts of interest None.