Ethics
The current study was approved by the institutional review board of Zhejiang university and conducted in accordance with the Declaration of Helsinki. The study design and manuscript preparation fully followed the guidelines from the STROBE statement. Informed written consent was obtained from all patients before endoscopic treatment.
Patients
The study design is a retrospective observational cohort study. We firstly screened the data in our endoscopic system with the Chinese item of “endoscopic submucosal dissection” and the English abbreviation of ESD, for the period between January 2011 and December 2017. The search restriction also includes “patients receiving colonoscopy”. Patients receiving ESD were selected according to the Japanese Gastroenterological Endoscopy Society Guidelines 10. Actually, some patients who receive so called “ESD” is just using submucosal snaring instead of submucosal dissection, after successful circumstantial mucosa incision. We categorized those patients into the group of ESS-CMI. The rest patients were further categorized into the ESD group. The primary goal of endoscopic resection is effective and safe treatment without severe complications such as mass bleeding and perforation need for surgery. Data on patients’ and neoplasm’s characteristics, recurrence, treatment, follow-up and disease status (type, severity and location ) were collected after medical records review and prepared for further analysis. Recurrence was defined as the detection of local recurrence or distant metastasis during follow-up.
All lesions were incidentally identified by screening colonoscopy with various symptoms such as hematochezia, defecation difficulty, change of defecation habit, et al. The biopsy with histological evaluation were obtained and evaluated by independent pathologist prior to the procedure. All patents received enhanced contrast CT scan to evaluate the disease and rule out potential metastasis. Some patients were also further evaluated by endoscopic ultrasonography.
Procedures of ESD and ESS-CMI
As shown in Figure 1, the common base for ESD and ESS-CMI is circumferential mucosa incision using a single-channel endoscope (GIF-Q260, Olympus, Tokyo, Japan) and electrosurgical generator with microprocessor controlled current (VIO 300D Endo Cut Q Effect 3; ERBE Elektromedizin, Tuebingen, Germany), including the sequential steps as: roundly labeling at the site 5mm outside the border line of the neoplasm with Argon plasma coagulation (ERBE, Elektromedizin, Tuebingen, Germany) or Dual knife (Olympus, Tokyo, Japan); injecting saline solution mixed with indigo carmine and noradrenalin into the submucosal layer under the neoplasm; performing circumferential incision at 5 mm outside the marking spot using Dual knife or Hook knife (Olympus, Tokyo, Japan). After thorough circumferential incision and fully lifting of the submucosal layer, the following procedure was furcated. For ESS-CMI, the single submucosal snaring was performed while for ESD, the submucosal layer was directly dissected using the same knife or IT knife (Olympus, Tokyo, Japan). Thereafter, the similar procedures of hemostasis and perforation prevention were routinely carried out for both ESS-CMI and ESD. During the process, precoagulation was performed on visible vessels by using hemostatic forceps (FD-410R, Olympus, Tokyo, Japan).
Sequential steps were carried out for sample management. At first, resected specimen was generally evaluated, including the length of longest side and the square of surface area. Light microscopy was then applied to determine the histological type, depth of invasion, lateral and vertical resection margin involvement and lymphovascular invasion. The degree of hyperplasia was determined by independent pathologist and divided into mild, moderate and severe. Mitotic count and Ki-67 index were assessed for determination of tumor grade under request. For those SMTs that are difficult to diagnosis, appropriate immunohistochemical staining was used.
Cost-effective analysis
Generally, the cost-effective analysis includes two main aspects: treatment effect and complication as well as the cost and efficiency. Among them, en bloc resection was defined as resection of the entire lesion in a single piece. The treatment effect was evaluated by the combination of negativity of vertical cutting edge (NVCE), negativity of side cutting edge (NSCE), no interstitial infiltration (NII) and no vascular infiltration (NVI). The treatment complications usually include mass bleeding in operation that needs emergent surgery (MBO), delayed hemorrhage (DH), intra-operative perforation (IP), delayed operative perforation (DP). The cost and efficiency mainly include operation time (OT), relative operation time (ROT, OT/surface area), overall cost, operation related equipment cost (OREC) and average hospital stay (AHS). OT was defined as the period from identification of the neoplasm to completion of the dissection/snaring and bleeding/perforation control. Perforation was defined as rectal wall penetration observed during the endoscopic procedure or detected after endoscopy by radiology. Hemorrhage was defined as massive bleeding in the operation or hematochezia after operation that required hemostasis or transfusions. En bloc resection was defined as one-piece resection endoscopically. Endoscopic complete resection was defined as en bloc resection endoscopically without tumor involvement to the lateral and the vertical margins of the resected specimens.
Follow up
All patients were periodically followed up by colonoscopy between approximately 6-12 months after endoscopic resection. If recurrent or remnant tumor was suspected, biopsy from the resected scar was performed.
Statistics
Continuous variables were expressed as the means and standard deviations, while the differences in the groups were analyzed using unpaired t test or Mann-Whitney test as appropriate. For categorical variables, X2 test is used. if the total sample is over 40 and the frequency of each cell in the crosstabs is over 5, pearson-chi-square is used; if the total sample is less than 40 or the frequency of any cell in the crosstabs is less than 5, fisher’s exact test is sued. P value <0.05 was considered statistically significant. SPSS version 19.0 was used for the whole analysis .