Effectiveness and safety of endoscopic submucosal dissection for residual or recurrent colorectal neoplasia: Meta-analysis.

Background and study aims: Endoscopic submucosal dissection (ESD) is a potentially surgery-sparing technique for colorectal neoplasia resection. Outcomes of ESD for residual or recurrent colorectal neoplasia are not well described. This meta-analysis aimed to evaluate the effectiveness and safety of ESD in treating residual or recurrent colorectal neoplasia.

Patients and methods: We searched MEDLINE and Embase up to July 24, 2023 for studies on ESD for residual or recurrent colorectal neoplasia at prior surgery or endoscopic resection sites. The primary outcome of the meta-analysis was R0 resection; secondary outcomes included recurrence, adverse events (AEs), procedure time, and hospitalization length. Pooled effect sizes were obtained using inverse variance random effects models. Subgroup analyses were based on study location, lesion size, and endoscopist experience.

Results: From 1,133 abstracts, data from 25 observational studies were included, reporting on 863 residual or recurrent lesions treated with ESD. R0 resection was achieved in 80.7% of patients (95% confidence interval 72.7-86.7%, I ² = 81%) of patients, whereas recurrence occurred in 2.0% (0.7-5.1%, I ² = 0%). Incidence of delayed bleeding and delayed perforation were 1.8% (0.7-4.2%, I ² = 0%) and 1.9% (0.6-6.3%, I ² = 35%), respectively. The former was independent of country of study, recurrent lesion size, or endoscopist experience. Mean procedure duration was 80.4 minutes (66.6-94.2, I ² = 96%) and hospitalization length was 4.2 days (2.0-6.4, I ² = 98%).

Conclusions: This meta-analysis suggests that ESD is effective and safe for treating residual or recurrent colorectal neoplasia after previous resection, with further prospective validation studies needed to compare ESD with other endoscopic resection methods and surgery in this context.