Abstract
Data on endoscopic submucosal dissection (ESD) for residual neoplasia after incomplete resection (ER) of large non-pedunculated colorectal polyps (LNPCPs) remains scarce. We aimed to evaluate and compare outcomes of ESD in treatment naïve (control) lesions vs. those with prior failed ER.
Multicenter propensity-score (PS) match study of ESDs performed for LNPCPs between January 2021 to September 2024. The following covariates were used for PS calculation: age, sex, and lesion characteristics (size, location, morphology, histology). Endpoints included: rates of en-bloc and R0 resection, adverse events and neoplasia recurrence on surveillance.
A total of 1447 consecutive patients underwent ESD during the study period. PS match resulted in the selection of 361 (control) and 184 (prior failed ER) strictly matched 2:1 pairs. En-bloc and R0 resection rates were similar between the control and prior failed ER arms: 91.7% vs 89.7%; p=.44 and 80.9% vs. 81.0%; p=.98, respectively. There was no difference in the rate of perforation between the control and prior failed ER arms (4.7% vs. 4.4%; p=1.00), whereas there was non-statistically significant trend towards higher delayed bleeding in the prior failed ER group (1.63% vs. 0.83%, relative risk: 1.96; p=0.40). Neoplasia recurrence on surveillance was 3.6% in the control and 5.8% in the prior failed ER group (p=0.32).
ESD can be performed safely and effectively as a salvage therapy after failed attempt at ER of LNPCPs. ESD may be selectively considered as part of our endoscopic armamentarium for the management of these difficult-to-treat lesions. support current guideline endorsed indications for ESD for the treatment of residual neoplasia after incomplete ER.