Long-term clinical outcomes of endoscopic submucosal dissection for colorectal neoplasia with or without the hybrid technique.

ABSTRACT

AIM: The main aim of this study was to compare the long-term outcome of a conventional colorectal endoscopic submucosal dissection (ESD) in which submucosal dissection was continued throughout until the completion of resection (ESD-T) to hybrid endoscopic submucosal dissection (ESD-H) in the colorectum. METHOD: Medical records of 836 colorectal neoplasia patients treated by ESD-T or ESD-H were reviewed. ESD-H was defined as colorectal ESD with additional snaring in the final stage of the procedure. Primary outcomes were the overall and metastatic recurrence rates. Secondary outcomes were short-term outcomes such as the en bloc resection rate, procedure time and adverse events. RESULTS: The overall recurrence rate was higher in the ESD-H than in the ESD-T group (5.7% vs 0.7%, P = 0.001). The metastatic recurrence rate showed no significant difference between these groups (1.4% vs 1.4%, P = 1.000). Multivariate analysis revealed that a failed en bloc resection (hazard ratio 24.097; 95% CI 5.446-106.237; P < 0.001) and larger tumour size (hazard ratio 1.042; 95% CI 1.014-1.070; P = 0.003) were independently associated with overall recurrence. The ESD-H group showed a lower en bloc resection rate (56.8% vs 96.5%, P < 0.001), shorter procedure time (45.6 vs 54.3 min, P < 0.001) and higher perforation rate (10.3% vs 6.0%, P = 0.029). CONCLUSION: Although long-term outcomes in terms of overall recurrence are inferior following ESD-H, a failed en bloc resection and large tumour size are the only independent risk factors for recurrence. Further investigations are warranted to improve the long-term outcomes of ESD-H.