Background: Colorectal laterally spreading tumours (LSTs) are classified into granular (LST-G) and non-granular (LST-NG) type; each type was sub-grouped into LST-G-H (homogenous) and LST-G-M (nodular mixed) type or LST-NG-F (flat elevated) and LST-NG-FD (pseudodepressed) type, respectively. We assessed the clinicopathological factors associated with clinical outcomes of endoscopic submucosal dissection (ESD) for colorectal LSTs, and conducted follow-up after ESD.
Methods: ESD was performed in 196 patients with 204 LSTs that fulfilled the inclusion criteria for colorectal neoplasms. Clinical outcomes including resectability and curability of ESD and perforation were investigated, and factors related to the outcomes were analysed using logistic regression. One hundred thirty-eight patients received endoscopic follow-up for more than 12 months and metastatic surveys in 79 cases of cancerous LSTs.
Results: The incidence of submucosal cancer was lower in LST-G type. There were no significant differences in outcomes regarding LST macroscopic types. Overall en bloc, complete and curative resection, and perforation rates were 86.8%, 77.5%, 82.8% and 9.8%, respectively. Logistic regression analysis showed higher risk of non-curative resection in LST-G-M than in LST-G-H type. No other factors were associated with outcomes. During median follow-up of 35.5 months, no locally recurrent or metastatic tumours were observed, and overall survival was still 100%.
Conclusions: ESD provides acceptable resectability for colorectal LSTs by facilitating en bloc resection, irrespective of macroscopic types. The relatively long-term outcomes may be excellent, but further evaluation is needed for appropriate treatment strategy for each type of LST.