Efficacy and safety of modified endoscopic mucosal resection for rectal neuroendocrine tumors: a meta-analysis

Purpose Rectal neuroendocrine tumors are rare with good prognosis. Several endoscopic methods such as endoscopic polypectomy, endoscopic submucosal dissection (ESD), endoscopic mucosal resection (EMR), and modified endoscopic mucosal resection (m-EMR) are used in the treatment of rectal neuroendocrine tumors. Although m-EMR is derived from traditional EMR, it has not been widely used in clinical practice. In this study, we compared the efficacy and safety of EMR and m-EMR in the treatment of rectal neuroendocrine tumors by performing a meta-analysis. Materials and Methods We searched PubMed, Web of Science, and EMBASE index up to the end of January 2017 for all published literature about EMR and m-EMR in the treatment of rectal neuroendocrine tumors. Results A total of 11 studies involving 811 patients were included. The pooled data suggested that there was a significantly higher rate of histologic complete resection and endoscopic complete resection among patients treated with m-EMR than those treated with EMR (histologic complete resection: OR = 0.23, 95 % CI = 0.10-0.51, p < 0.01; endoscopic complete resection: OR = 0.13, 95 % CI = 0.02-0.74, p = 0.02). The procedure time of EMR was longer than m-EMR (MD = 2.40, 95 % CI = 0.33-4.46, p = 0.02). There was a significantly higher rate of vertical margin involvement among patients treated with EMR than those treated with m-EMR; whereas, there was no significant difference of lateral margin involvement between the m-EMR and EMR groups (vertical margin involvement: OR = 5.00, 95 % CI = 2.67-9.33, p < 0.01; lateral margin involvement: OR = 1.44, 95 % CI = 0.48-4.37, p = 0.52). There was no significant difference in mean tumor size among patients treated with m-EMR versus those treated with EMR ( MD = -0.30, 95 % CI = -0.75-0.14, p = 0.18); further, there was no significant difference in endoscopic mean sizes of the tumor and pathological mean sizes of the tumor between the m- EMR and EMR groups (endoscopic mean sizes of the tumor: MD = 0.20, 95 % CI = -0.44-0.84, p = 0.43; pathological mean sizes of the tumor: MD = 0.62, 95 % CI = -0.68-1.92, p = 0.05). No significant differences were detected among the treatment groups with regard to complications (bleeding: OR = 0.87, 95 % CI = 0.39-1.95, p = 0.73; complications (bleeding and perforation): OR = 0.87, 95 % CI = 0.40-1.88, p = 0.73). Conclusion The efficacy of m-EMR are better than EMR among patients undergoing endoscopic treatment of rectal neuroendocrine tumors, and the safety of m-EMR is equivalent to EMR treatment.