比较关节镜下全内缝合对比内-外缝合技术对半月板损伤的修复疗效，提供临床循证依据。

检索5个中英文数据库中关节镜下半月板全内缝合与内-外缝合的对照研究，包括PubMed、Web of Science、荷兰《医学文摘》(EMBASE)、中国知网及万方。检索时间为建库至2020年5月，结果限定于中英文文献。参照设定的标准，筛选纳入患者为关节镜下初次行半月板缝合术、无基础疾病或不合并其它关节疾病的文献。数据提取后汇总愈合优良率、并发症、手术时间、Lyshlom评分、Tegner评分5项指标，文献质量评价采用Cochrane评价手册及纽卡斯尔-渥太华量表，应用RevMan 5.3软件对二分类变量及连续变量行统计分析。

纳入10项研究，共631例患者(全内缝合组340例，内-外缝合组291例)。全内缝合法术后半月板愈合优良率与内-外缝合法相近，差异无统计学意义[随机对照组：风险比(RR)＝1.02，95%可信区间(CI)(0.79，1.31)，P＝0.88；非随机对照组：RR＝1.00，95%CI (0.92，1.09)，P＝0.99]；全内缝合技术并不能降低术后并发症发生率，全内缝合法与内-外缝合法差异无统计学意义[随机对照组：RR＝0.65，95%CI (0.22，1.92)，P＝0.43；非随机对照组：RR＝0.50，95%CI (0.25，1.97)，P＝0.05]；全内缝合组患者平均手术时间较内-外缝合组短，差异有统计学意义，随机对照组缩短16 min，非随机对照组缩短46 min[随机对照组：平均差(MD)＝-16.31, 95%CI (-30.56，-2.07)，P＝0.02；非随机对照组：MD＝-46.60，95%CI (-60.79，-32.41)，P<0.001]；全内缝合与内-外缝合对患者膝关节功能影响相似，两组患者术后Lyshlom评分[随机对照组：MD＝6.24，95%CI (-11.39，23.86)，P＝0.49；非随机对照组：MD＝1.15，95%CI (-0.40，2.69)，P＝0.15])与Tegner评分[MD＝0.00 95%CI (-0.20，0.21)，P＝0.97]差异无统计学意义。

与内-外缝合法相比，关节镜下全内缝合在半月板愈合优良率、并发症发生率、膝关节功能方面并无优势，但能缩短平均手术时间。全内缝合手术花费大，而疗效与内-外缝合相当，临床医师选择全内缝合法时应慎重。

To provide a clinical evidence-based medicine by comparing the efficacy of all-inside versus inside-out suture techniques in arthroscopic meniscal injury repair surgery.

Five English and Chinese databases were searched for controlled studies about arthroscopic meniscus all-inside versus inside-out suture, including PubMed, Web of Science, Excerpt Medica Database(EMBASE), China Knowledge Network Infrastructure, and Wanfang. The search was conducted from the establishment of the database to May 2020, and the results were limited to English and Chinese literature. According to the defined criteria, patients who first underwent arthroscopic meniscus suture and had no underlying disease or other joint disease were screened for inclusion. After data extraction, five indicators were pooled including excellent healed rate, complications, operative time, Lyshlom score, and Tegner score. The quality of the literature was evaluated using the Cochrane System Evaluation Manual and the Newcastle-Ottawa scale (NOS). The analysis of dichotomous and continuous variables was performed by RevMan 5.3 software.

Ten studies with a total of 631 patients (340 in the all-inside suture group and 291 in the inside-out suture group) were included. There was no statistically significant difference in rate of meniscal excellent healed. The result of all-inside suture was similar to inside-out suture [the randomized control trial (RCT) group: the risk ratio (RR)=1.02, 95% confidence interval (CI) (0.79, 1.31), P=0.88; non-RCT group: RR=1.00, 95% CI (0.92, 1.09), P=0.99]. The all-inside suture technique did not reduce the postoperative complication rate, and there was no statistically significant difference between the all-inside suture method and the inside-out suture method [RCT group: RR=0.65, 95% CI (0.22, 1.92), P=0.43; non-RCT group: RR=0.50, 95% CI (0.25, 1.97), P=0.05]. The mean operative time of patients in the all-inside suture group was shorter than the Inside-out suture group, with a statistically significant difference of 16 min in RCT group and 46 min in non-RCT group [RCT group: mean difference (MD)=-16.31, 95% CI (-30.56, -2.07), P=0.02; non-RCT group: MD=-46.60, 95% CI (-60.79, -32.41), P<0.001]. The effect of all-inside and inside-out suture on the knee function of patients was similar, and there were no statistically significant difference in both group for postoperative Lyshlom score [RCT group: MD=6.24, 95% CI (-11.39, 23.86), P=0.49; non-RCT group: MD=1.15, 95% CI (-0.40, 2.69), P=0.15] and Tegner score [MD=0.00 95% CI (-0.20, 0.21), P=0.97].

Compared with the inside-out suture technique, the arthroscopic all-inside suture has no advantage in terms of meniscal excellent healed rate, complication rate or knee function. However, all-inside suture can shorten the mean operation time. The clinician should be thoughtful in choosing the all-inside suture as the procedure is more costly and the outcomes are comparable to inside-out suture.