Abstract
Abstract
Background
Robotic-assisted surgery (RAS) is increasingly used for colorectal cancer (CRC), but its clinical and oncologic advantages over conventional laparoscopy (LS) remain uncertain. Prior meta-analyses (e.g., Huang et al., Front Oncol 13;1273378, 2023; Zou et al., BMC Surg 25;86, 2025) have included overlapping RCTs but vary in methodology, scope, and analytical transparency. This review aims to provide an updated, independently re-analyzed synthesis of RCTs published from 2015 to 2025, with full PRISMA compliance, explicit analytic reproducibility, and expanded evaluation of bias and evidence certainty.
Methods
A systematic review and meta-analysis was conducted according to PRISMA guidelines. The protocol was retrospectively registered in PROSPERO (Registration ID:
CRD420251237158
). PubMed, Embase, and Cochrane CENTRAL were searched (January 1, 2015–January 31, 2025). Full reproducible search strings, PICOS criteria, and inclusion/exclusion rules were predefined. Only RCTs comparing RAS vs. LS for malignant colorectal disease were included. Data extraction was performed independently by two reviewers. Meta-analyses used DerSimonian–Laird random-effects models; standardized procedures were applied for converting medians/IQRs into means/SDs and for continuity corrections in zero-event trials. Risk of bias was assessed using Cochrane RoB 2.0, and evidence certainty was graded using GRADE.
Results
A total of 12 RCTs encompassing 3,107 patients met the inclusion criteria. RAS resulted in significantly lower conversion-to-open rates (OR 0.42; 95% CI 0.28–0.63; I²=18%) compared with LS. Operative time was consistently longer with RAS (MD + 23.8 min; 95% CI 14.2–33.4; I²=67%). Overall postoperative complications (Clavien–Dindo ≥ II) were comparable (OR 0.91; 95% CI 0.76–1.13; I²=22%). Length of stay showed a small but significant reduction with RAS (MD − 0.8 days; 95% CI − 1.3 to − 0.2; I²=49%).
Pathologic outcomes showed lower circumferential resection margin (CRM) positivity with RAS (OR 0.59; 95% CI 0.41–0.85). Lymph node retrieval was slightly higher with RAS (MD + 0.71 nodes; 95% CI 0.25–1.18). Distal margins and TME completeness were equivalent. No RCT reported mature long-term oncologic outcomes; evidence remains limited to short-term surrogates.
Conclusions
In contemporary RCTs, RAS provides fewer conversions and slightly better pathologic surrogates, while maintaining similar morbidity compared to LS. The main trade-off remains longer operative time and higher resource use. True oncologic equivalence cannot be confirmed until long-term RCT data mature. Advanced imaging (e.g., SOMATOM Force CT), age-specific MIS evidence, and the emergence of endoluminal robotic systems are likely to shape future refinements in technique and patient selection.