目的:本系统综述和荟萃分析的目的是使用同质亚组分析比较体外吻合(EA)和体内吻合(IA)的腹腔镜右结肠切除术(LRC)与机器人右结肠切除术(RRC)。
方法:MEDLINE,Scopus,截至2020年4月,我们在WebofScience数据库中检索了LRC和RRC对至少一项短期或长期结局的前瞻性或回顾性研究.主要结果是住院时间(LOS)。次要结果包括手术和病理结果,生存,和总成本。使用三个同质亚组比较LRC和RRC:不区分吻合类型,仅限EA,只有IA。使用均差(MD)和随机效应模型进行汇总数据分析。
结果:选择了448项研究中的37项。LRC组纳入的患者为21,397例,RRC组为2796例。不管吻合的类型,RRC显示LOS较短,减少失血,较低的转化率,排气时间较短,与LRC相比,总体并发症发生率较低,但手术时间更长,总成本更高。在EA子组中,RRC显示类似的LOS,更长的手术时间,与LRC相比,成本更高,其他结果相似。在IA亚组中,与LRC相比,RRC显示出更短的LOS和更长的操作时间,其余结果没有差异。
结论:大多数收录的文章都是回顾性的,提供低质量的证据和有限的结论。IA的更频繁使用似乎解释了RRC相对于LRC的优势。
OBJECTIVE: The aim of the present systematic review and meta-analysis is to compare laparoscopic right colectomy (LRC) versus robotic right colectomy (RRC) using homogeneous subgroup analyses for extra-corporeal anastomosis (EA) and intra-corporeal anastomosis (IA).
METHODS: MEDLINE, Scopus, and Web of Science databases were searched up to April 2020 for prospective or retrospective studies comparing LRC versus RRC on at least one short- or long-term outcome. The primary outcome was the length of hospital stay (LOS). The secondary outcomes included operative and pathological results, survival, and total costs. LRC and RRC were compared using three homogeneous subgroups: without distinction by the type of anastomosis, EA only, and IA only. Pooled data analyses were performed using mean difference (MD) and random effects model.
RESULTS: Thirty-seven of 448 studies were selected. The included patients were 21,397 for the LRC group and 2796 for the RRC group. Regardless for the type of anastomosis, RRC showed shorter LOS, lower blood loss, lower conversion rate, shorter time to flatus, and lower overall complication rate compared with LRC, but longer operative time and higher total costs. In the EA subgroup, RRC showed similar LOS, longer operative time, and higher costs compared with LRC, the other outcomes being similar. In the IA subgroup, RRC showed shorter LOS and longer operative time compared with LRC, with no difference for the remaining outcomes.
CONCLUSIONS: Most included articles are retrospective, providing low-quality evidence and limiting conclusions. The more frequent use of the IA seems to explain the advantages of RRC over LRC.