Underwater versus standard endoscopic submucosal dissection for superficial colorectal neoplasms: a propensity score matched analysis

Underwater ESD (UESD) has recently emerged as a less complex technical alternative to standard endoscopic submucosal dissection (SESD), but only small retrospective data are currently available. Our aim was to evaluate the effectiveness and safety of UESD, as compared to the current standard of care. We performed a 1-to-4 nearest-neighbor retrospective propensity-score-matched-analysis between all UESD cases and all SESD cases performed and collected prospectively in a Western and an Eastern interventional endoscopy referral center, respectively. The primary outcomes were the rate of resections with histologically negative lateral and deep margins and the en-bloc resection rate. The secondary outcomes were procedure speed (mm2/min) and procedure time. The secondary outcomes were procedure speed (mm2/min) and procedure time. Overall, 100 patients for UESD and 400 patients for SESD were included, matched for all baseline parameters. According to the Kudo and Paris classifications, in both groups most of lesions were laterally spreading tumours, granular type (LST-G) (60.7% vs. 55%). The mean size of lesions in the SESD group was 40.9 ± 14.7 mm. In the UESD group, the mean size of lesions was 57.4 ± 27.9 mm. UESD yielded higher rate of en-bloc resection (100% vs. 86.5%, p < 0.001) and lower rate of perforation (1% vs. 9.5%, p = 0.009). No significant difference between the two techniques was observed in terms of vertical R0 resection rate, speed and time of dissection, and delayed bleeding rate. Our findings suggest that UESD allows to achieve resection speed and procedure times comparable to those of expert Japanese endoscopists, with a significant reduction in the risk of perforation.