Abstract:[Objective] To systematically evaluate the clinical consequences of the extramedullary (EM) localizations versus the intramedullary (IM) counterpart for femoral osteotomy in primary total knee arthroplasty (TKA). [Methods] The controlled studies on EM and IM for femoral osteotomy in TKA were searched from data bases, including CNKI, Wanfang, VIP, Chinese Biomedical Literature Database, PubMed, Embase, Cochrane Library and Web of Science databases. A metaanalysis was performed using Stata 15.0 software. [Results] Eight randomized controlled studies were included in this study, with a total of 804 patients, including 404 cases in the EM group and 400 cased in the IM group. As results of the meta-analysis, The EM group was significantly superior to the IM group in terms of coronal angle of femoral prosthesis (RR=0.184, 95%CI 0.07-0.45, P<0.001), and the blood loss (MD=-161.24, 95%CI -233.93~-88.55, P=0.016). However, there were no significant differences in terms of coronal alignment of lower limbs (RR=1.2, 95%CI 0.28~5.21, P=0.809), sagittal angle of femoral prosthesis (RR=0.18, 95%CI 0.07~0.45, P=0.541), operative time (MD=-0.74, 95%CI -4.04~-2.57, P=0.665) between the two groups. [Conclusion] EM localization for femoral osteotomy in TKA get more accurate coronal angle of femur prosthesis with less blood loss over the IM localization, while the two techniques are similar in terms of coronal position alignment of lower limb, sagittal angle of femur prosthesis and operation time.