Transanal robotic versus endoscopic submucosal dissection for complex distal colorectal lesions: a comparative study

Abstract Background Endoscopic submucosal dissection (ESD) is technically challenging for complex colorectal lesions, while endorobotic submucosal dissection (ERSD) has emerged as a transanal minimally invasive alternative offering high-definition 3D visualization and improved instrument dexterity. However, evidence-based technique selection criteria remain unclear. This study compares the technical success and operative efficiency between ERSD and ESD using a lesion size-stratified analysis to guide technique selection. Methods Retrospective analysis of prospectively collected data from consecutive patients who underwent ERSD or ESD for distal colorectal lesions at a tertiary referral center (2020–2024). All procedures were performed by a single expert endoluminal surgeon to eliminate learning curve bias. The Da Vinci SP robot was used for ERSD. The primary outcomes included en bloc resection rates, operative time, and perioperative complications. Lesions were stratified by surface area: small (< 10 cm 2 ), medium (10–25 cm 2 ), and large (≥ 25 cm 2 ). Statistical analysis assessed the relationship between lesion size and operative time for both techniques. Results A total of 164 patients were analyzed (ERSD: n = 79, ESD: n = 85). ERSD patients had significantly larger lesions (median surface area 18.4 vs 6.3 cm 2 , p < 0.001) while maintaining similar overall operative times (75 vs 78 min, p = 0.42). En bloc resection rates were comparable (97.5% vs 96.5%, p = 0.99). Intraoperative perforation occurred in 2.5% of ERSD cases compared to 4.8% of ESD cases ( p = 0.75); all perforations were full-thickness defects successfully managed endoscopically or transanally without additional surgical intervention. Postoperative complications occurred in 10.1% vs 8.2% ( p = 0.88). The majority of complications were Clavien–Dindo Grade I–II, with no Grade IV or V events in either group. Operative time correlated with specimen size for both techniques (ERSD: r = 0.59; ESD: r = 0.46; both p < 0.001). Size-stratified analysis revealed equivalent times for small lesions (71 vs 67 min, p = 0.85) but significant ERSD advantages for medium (65 vs 119 min, p = 0.03) and large lesions (87 vs 142 min, p = 0.04), despite robotic setup overhead. Conclusion Endorobotic submucosal dissection demonstrated substantial time savings for lesions ≥ 10 cm 2 without compromising safety or resection quality, suggesting that lesion size may serve as a useful parameter for technique selection between ERSD and ESD.