Systematic review and meta-analysis of endoscopic submucosal dissection versus transanal endoscopic microsurgery for large noninvasive rectal lesions

Background For almost 30 years, transanal endoscopic microsurgery (TEM) has been the mainstay treatment for large rectal lesions. With the advent of endoscopic submucosal dissection (ESD), flexible endoscopy has aimed at en bloc R0 resection of superficial lesions of the digestive tract. This systematic review and meta-analysis compared the safety and effectiveness of ESD and full-thickness rectal wall excision by TEM in the treatment of large nonpedunculated rectal lesions preoperatively assessed as noninvasive. Methods A systematic review of the literature published between 1984 and 2010 was conducted (Registration no. CRD42012001882). Data were integrated with those from the original databases requested from the study authors when needed. Pooled estimates of the proportions of patients with en bloc R0 resection, complications, recurrence, and need for further treatment in the ESD and TEM series were compared using random-effects single-arm meta-analysis. Results This review included 11 ESD and 10 TEM series (2,077 patients). The en bloc resection rate was 87.8 % (95 % confidence interval [CI] 84.3–90.6) for the ESD patients versus 98.7 % (95 % CI 97.4–99.3 %) for the TEM patients ( P < 0.001). The R0 resection rate was 74.6 % (95 % CI 70.4–78.4 %) for the ESD patients versus 88.5 % (95 % CI 85.9–90.6 %) for the TEM patients ( P < 0.001). The postoperative complications rate was 8.0 % (95 %, CI 5.4–11.8 %) for the ESD patients versus 8.4 % (95 % CI 5.2–13.4 %) for the TEM patients ( P = 0.874). The recurrence rate was 2.6 % (95 % CI 1.3–5.2 %) for the ESD patients versus 5.2 % (95 % CI 4.0–6.9 %) for the TEM patients ( P < 0.001). Nevertheless, the rate for the overall need of further abdominal treatment, defined as any type of surgery performed through an abdominal access, including both complications and pathology indications, was 8.4 % (95 % CI 4.9–13.9 %) for the ESD patients versus 1.8 % (95 % CI 0.8–3.7 %) for the TEM patients ( P < 0.001). Conclusions The ESD procedure appears to be a safe technique, but TEM achieves a higher R0 resection rate when performed in full-thickness fashion, significantly reducing the need for further abdominal treatment.