系统评价在动物实验模型中，消化内镜机器人辅助系统相较于传统手动操作对内镜黏膜下剥离术(ESD)的影响。

检索PubMed、Embase、Web of Science、Cochrane Library、中国知网、万方数据、VIP、CBM等数据库自建库至2025年12月发表的相关文献，收集所有在动物模型中比较机器人辅助ESD与常规ESD的研究。根据纳入研究的干预方式，将实验对象分为机器人辅助组和常规组，由2名研究者独立完成文献筛选与数据提取。采用RevMan 5.4、StataMP 18软件进行Meta分析。主要结局指标为整块切除率与黏膜下盲切率，次要结局指标包括总手术时间、黏膜下剥离时间及解剖速度。

共纳入11项动物实验研究(包含离体及活体模型)。分析结果显示，与常规操作相比，机器人辅助能显著提升整块切除率(OR＝8.35，95%CI: 2.63, 26.56，P＝0.0 003)，并将黏膜下盲切率降低(MD＝－19.49，95%CI: －25.03, －13.95，P<0.00001)。在手术效率方面，机器人辅助缩短总手术时间(MD＝－14.23 min，95%CI: －18.99, －9.48，P<0.00 001)和黏膜下剥离时间(MD＝－10.48 min，95%CI: －16.83, －4.13，P<0.00 001)。机器人辅助组的解剖速度亦显著高于常规组(MD＝30.85 mm²/min, 95%CI: 18.54, 43.17，P＝0.0003)。针对总手术时间的Egger′s检验提示该指标无明显发表偏倚(P＝0.147)，其余结局指标因纳入研究数量限制未进行发表偏倚评估。

在现有动物实验证据中，内镜机器人辅助系统可以优化手术视野，在提升ESD整块切除率、降低盲切风险、以及缩短手术耗时方面展现出明确优势。该技术显示出解决ESD技术难点的潜力，但其临床效益仍需通过进一步研究证实。

To systematically evaluate the impact of robotic-assisted digestive endoscopic systems versus conventional manual operation on ESD in animal experimental models.

Searches were performed in databases including PubMed, Embase, Web of Science, Cochrane Library, CNKI, Wanfang Data, VIP, and CBM from inception to December 2025 to identify animal studies comparing robotic-assisted ESD with conventional ESD.According to the intervention methods of the included studies, the experimental subjects were divided into a robotic-assisted group and a conventional group.Two researchers independently screened the literature and extracted data.Meta-analysis was conducted using RevMan 5.4 and StataMP 18.The primary outcome measures were en bloc resection rate and blind dissection rate; secondary outcomes included total procedure time, submucosal dissection time, and dissection speed.

Eleven animal experimental studies (ex vivo and in vivo) were included.Robotic assistance significantly increased the en bloc resection rate (OR=8.35, 95%CI: 2.63, 26.56, P=0.0 003) and reduced the blind dissection rate (MD=-19.49, 95%CI: -25.03, -13.95, P<0.00 001). Total procedure time MD=-14.23 min, 95%CI: -18.99, -9.48, P<0.00 001) and submucosal dissection time (MD=-10.48 min, 95%CI: -16.83, -4.13, P<0.00 001) were significantly shorter in the robotic-assisted group. Dissection speed was also higher (MD=30.85 mm²/min, 95%CI: 18.54, 43.17, P=0.0 003). Egger′s test for the total procedure time indicated no significant publication bias (P=0.147), and publication bias assessment was not performed for the remaining outcome indicators due to the limited number of included studies.

Based on current animal evidence, robotic-assisted endoscopic systems can optimize the surgical field, demonstrating clear advantages in improving the en bloc resection rate of ESD, reducing the risk of blind cutting, and shortening operative time.