目的系统评价腹腔镜和开腹手术比较对交界性卵巢肿瘤复发的影响。方法计算机检索Pub Med、The Cochrane Library(2015年11期)、Web of Science、EMbase、CNKI、Wan Fang Data和CBM数据库,搜集腹腔镜和开腹手术比较对交界性卵巢肿瘤复发影响的研究文献,检索年限均从建库至2015年11月。由2位评价员独立筛选文献、提取资料并评价纳入研究的偏倚风险后,采用Rev Man 5.3软件进行Meta分析。结果最终纳入19个队列研究。NOS量表评分结果显示,10个研究总分<7分,9个研究总分≥7分。Meta分析结果显示:虽然腹腔镜组总术后复发率高于开腹组[OR=1.75,95%CI(1.05,2.91),P=0.03],但在保守手术[OR=1.22,95%CI(0.71,2.08),P=0.47]和根治手术[OR=4.38,95%CI(0.85,22.68),P=0.08]两个亚组中,两组术后复发率差异无统计学意义。此外,腹腔镜手术组的肿块直径明显小于开腹手术组[MD=–6.88,95%CI(–8.15,–5.61),P<0.000 01];且术中肿块破裂发生率较开腹手术组高[OR=3.99,95%CI(2.54,6.26),P<0.000 01]。结论腹腔镜手术治疗交界性卵巢癌时,因对肿块径线要求较高,且有增加术中肿瘤破裂率及术后复发的潜在风险,应谨慎考虑应用范围。受纳入研究的质量及纳入病例数限制,腹腔镜在交界性卵巢癌中的运用需更多高质量、大样本研究进一步验证。 Objective To evaluate the effect of laparoscopic and laparotomy on the relapse of borderline ovarian tumors. Methods The databases of Pub Med, The Cochrane Library (2015-11), Web of Science, EMbase, CNKI, Wan Fang Data and CBM were searched by computer to collect the literatures about the relapse of laparoscopic and laparotomy on borderline ovarian tumors. Search years are from the library to November 2015. After two reviewers independently screened the literature, extracted data, and assessed the risk of being included in the study, Meta-analysis was performed using Rev Man 5.3 software. The results eventually included 19 cohort studies. The results of the NOS scale showed that 10 studies had a total score of <7 points and nine studies had a total score of ≥7 points. Meta-analysis showed that although the total postoperative recurrence rate in laparoscopic group was higher than that in open group [OR = 1.75,95% CI (1.05, 2.91), P = 0.03], in conservative surgery [OR = 1.22,95% CI (0.71,2.08), P = 0.47] and radical surgery [OR = 4.38,95% CI (0.85,22.68), P = 0.08]. There was no significant difference in recurrence rate between the two subgroups. In addition, laparoscopic surgery group was significantly smaller than the diameter of the tumor group laparotomy [MD = -6.88,95% CI (-8.15, -5.61, P <0.000 01]; and the incidence of intraoperative rupture of the laparotomy group High [OR = 3.99, 95% CI (2.54, 6.26), P <0.000 01]. Conclusions Laparoscopic surgery in the treatment of borderline ovarian cancer requires careful consideration of the range of applications due to the high requirements for the diameter of the tumor mass and the increased risk of intraoperative rupture of the tumor and postoperative recurrence. Due to the quality of the included studies and the inclusion of case limits, laparoscopy needs more high-quality, larger sample studies to further validate the use of borderline ovarian cancer.