Background colorectal endoscopic submucosal dissection (ESD) remains a technical challenge, but traction devices show promise in making this procedure easier. However, the efficacy of traction techniques for colorectal ESD is still unknown for inexperienced endoscopists. Methods We selected 400 patients who underwent colorectal ESD performed by four inexperienced endoscopists. Each patient in the traction-assisted ESD (TA-ESD) group was matched to a patient in the conventional ESD (C-ESD) group according to propensity scores. Results One-to-one propensity score-matching analysis created 87 matched pairs. The self-completion rate in the TA-ESD group is significantly higher than that in the C-ESD group (100% [87/87] vs. 92% [80/87], p < 0.001). The median resection speed was significantly faster in the TA-ESD group than that in the C-ESD group (27 mm2/min [IQR, 19.5–47.3] vs.18 mm2/min [IQR, 13.5–33.8], p < 0.001) and the procedure time in the TA-ESD group was significantly shorter than that in the C-ESD group (33 min [IQR, 27–47] vs.53 min [IQR, 38–73], p < 0.001). However, the histologic complete resection rate was not significantly different between the TA-ESD and C-ESD groups (93.1% [6/87]) vs. 96.6% [3/87], p < 0.1888, respectively). The en bloc resection rate (96.6%) and perforation rate (4.6%) were equivalent between the TA-ESD group and the C-ESD group. Conclusion Traction techniques seem to improve resection speed and self-completion rate of colorectal ESD for inexperienced endoscopists.
Read full abstract