Efficacy and safety of endoscopic submucosal dissection for colorectal neoplasia: a systematic review.

BACKGROUND AND STUDY AIMS: Endoscopic submucosal dissection (ESD) has been proposed for large colorectal lesions, due to the high risk of recurrence following endoscopic mucosal resection. However, data on the efficacy and safety of colorectal ESD are still controversial. The aim of the current systematic review was to assess the efficacy and safety of colorectal ESD. METHODS: A detailed Medline search of papers published during the period 1999-2010 was performed, using the search terms "Endoscopic submucosal dissection," "Colorectal neoplasia," "Colon," or "Rectum." Published studies that evaluated ESD for colorectal lesions were assessed using well-defined inclusion/exclusion criteria, including histological confirmation and surgery for complications. The process was independently performed by two authors. Forest plots on primary (i.e. histologically verified R0 resection and surgery for ESD complications) and secondary end-points were produced based on random-effect models. Heterogeneity was assessed using the I2 statistic. Risk for within-study bias was also ascertained. RESULTS: A total of 22 studies (20 Asian, two European) provided data on 2841 ESD-treated lesions. The per-lesion summary estimate of R0 resection rate was 88% (95%CI 82%-92%; I2=91%). At meta-regression, carcinoid vs. non-carcinoid series (R0 93% vs. 87%; P=0.04) and Asian vs. European series (R0 88% vs. 65%; P=0.03) appeared to explain the detected heterogeneity. The per-lesion summary estimate of surgery for ESD complications was 1% (95%CI 0%-1%) with a moderate degree of heterogeneity (I2=49%). However, subgrouping of these results according to histological tumor types was not available in the reviewed studies. CONCLUSIONS: ESD appeared to be an extremely effective technique to achieve R0 resection of large colorectal lesions. The very low rate of surgery for complications also shows the potential safety of this approach.