Oyama T. published the first study about counter traction in ESD in 2002, using the clip with line method. After that, endoscopists around the world working on different traction strategies to facilitate ESD procedure. Xia. et al reported in a meta-analysis that traction ESD was superior to standard ESD in terms of perforation rate and resection speed, confirming the importance of traction method.
In our study, the procedure time of the RAC-ESD group was longer than that of the conventional ESD group. We believe that this can be explained by the fact that the resected area of RAC-ESD group was slightly larger than that of the conventional ESD group, although the difference was not significant. Overall, the resection speed of the RAC-ESD group was significantly faster than that of the conventional ESD group. The en bloc resection rate, R0 resection rate and the curative resection rate of both groups are quite satisfied, similar to the other studies in Asia[10, 11]. There was no significant difference in oncological and safety data between the two group may be owing to the operators involved in this study were experienced in ESD technique. The learning curve in previous study confirmed that ESD with traction could also be used in the training progress of inexperienced ESD operators[12, 13], as it continuously optimizes exposure of the submucosal space, making the process of submucosal dissection much easier.
Several studies have demonstrated the effectiveness of inner traction method in colorectal ESDs using clips with either rubber band or looped thread[12, 14, 15]. In their study, the rubber band was already a commercial product but has not been approved for medical use in many other countries. The production of looped thread was comparatively complicated for the need of measuring the length of thread and the knot-tying process. In our study, we use a kind of rubber band made of latex examination gloves, which are extremely easy to get among medical centers, suitable to be cut into circles and above all, inexpensive. Unlike several other previously reported traction systems, the rubber band and clip can pass through the working channel of the scope without scope removal. The elasticity of the rubber band makes it more adaptable in the ESD practice. Operators can increase or decrease the countertraction by control the air volume in the colorectal space. Moreover, when dealing with the giant lesions which cannot be handled by a single traction, a third or a fourth clip can be applied for traction, attaching the rubber band to the opposite site of the remnant lesion to obtain continuous countertraction as we have previously reported.
Although many previous studies have confirmed the superiority of traction ESD compared with conventional ESD, few studies discussed that in which kinds of lesions traction method could show the greatest effect. The subgroup analysis of our study showed that RAC-ESD improved the resection speed in lesions located in transverse colon to ascending colon, lesions macroscopically presenting as lateral spreading tumor (LST) and lesions with large diameter equal to or greater than 2cm. ESD is not easy to perform in transverse colon to ascending colon owing to the poor scope maneuverability and limited use of gravitation. Iacopini et al reported that positional changes did not harness the power of gravity in 22% of colonic ESD cases, compared with only 3% of rectal ESD cases. Moreover, lesions presenting as LST and larger lesions are also thought to be more challenged for endoscopist, especially inexperienced endoscopist to perform ESD, as the technical difficulty of procedure increases with the size of the lesion. Thus, the advantage and efficiency of RAC-ESD for these lesions were shown more obviously, counting for the differences displayed in the subgroup analysis, even for skillful endoscopists, as the operators included in our study are both highly experienced in ESD (performing more than 500 colorectal ESD).
The propensity score-matched method was used to reduced bias in this study. Propensity scoring was calculated by using the above-mentioned confounders including localization, macroscopic type and the large diameter equal to or larger than 2cm or not, as these confounders should possibly influence the difficulty of ESD process and thus interfere with the clinical outcomes between groups. These basic characteristics were balanced between groups after propensity score-matched, making the technical data between the RAC-ESD group and the conventional ESD group more comparable.
There are a few limitations to this study. First, it was a single center, retrospective study susceptible to selection bias. Although the use of propensity score matching allowed us to balance the two groups, some confounders that were not incorporated into the propensity score, may still influence the results. However, we consider it difficult to conduct randomized trials between RAC-ESD and traditional ESD as the former has shown to be much superior in challenging cases. Second, the two operators included in our study are both experienced endoscopists in ESD. We did not discuss the effectiveness of RAC-ESD strategy in the learning process for trainees in ESD. Finally, as the rubber band we used in this study was made out of latex gloves, although we did not encounter any allergy complications during the traction process, we are unaware that whether patients being allergic to latex would develop an allergic reaction when the latex come into contact with the colonic wall. Therefore, for patients with definite medical history of severe latex allergy, RAC-ESD might shall be avoided in the process of endoscopic treatment.
In conclusion, RAC-ESD is a safe, effective and cheap method to treat colorectal neoplasms, especially in lesions presenting particular difficulty: larger lesions, lesions macroscopically presenting as lateral spreading tumor and lesions located in right-sided colon. A multicenter study including inexperienced endoscopist is needed to confirm the reproducibility and its effect in the training process of ESD.