Endoscopic submucosal dissection of early cancers, flat adenomas, and submucosal tumors in the gastrointestinal tract.

Background & Aims: Endoscopic submucosal dissection (ESD) is a promising technique in the treatment of large premalignant and early malignant gastrointestinal lesions. In contrast to Japan and Asian countries, few data are available from Western countries. The objective of this study was to assess the feasibility of ESD in a European center, with special regard for the success rate and learning curve.
Methods: Over a 4-year-period, 82 epithelial or submucosal lesions were referred for ESD. Seventy-one ESDs were performed (51 gastric, 17 rectal, 2 esophageal, and 1 duodenal). Resection rates, procedure times, specimen sizes, complications, and recurrences were noted. The mean follow-up period was 15 months.
Results: Specimen size increased significantly (P < .05) and procedural duration decreased significantly (P < .005) over time. En bloc resection rates and R0 en bloc resection rates were 77.1% and 65.7%, respectively, in the first half of the study and increased to 86.1% and 72.2%, respectively, in the second half (P = NS). No recurrence was observed after R0 en bloc resection whereas the recurrence rate was 38.5% after piecemeal resections (P < .001). Two perforations in the first series were treated by surgery; 2 other perforations, 8 minor bleedings, and 2 pyloric stenoses were treated endoscopically.
Conclusions: ESD is technically feasible and shows promising results in this German single-center-study. ESD is time consuming and difficult but shows a learning curve resulting in a decrease of the procedural duration over time. R0 en bloc resection is mostly possible and can avoid the risk of local recurrence.