To evaluate the clinical outcomes between the computer navigation and conventional methods for total knee arthroplasty(TKA) through meta-analysis.

Pubmed, CNKI, Springer Link, and other databases were retrieved for literatures concerning randomized controlled trial of clinical outcomes of computer-assisted navigation and conventional TKA published from January 1^st 2007 to May 1^st 2017.The related data were selected of the literatures in accordance with clinical randomized controlled trials for primary TKA which set computer navigation as the experimental group; those literatures with suspected data were excluded. The data were analyzed by Manager Review 5.0.

Fifteen randomized controlled trials were eligible for meta-analysis. No significant difference was found in coronal femoral component alignment[MD=-0.03, 95%CI(-0.60, 0.53), P >0.05], coronal tibial component alignment[MD=-0.32, 95%CI(-0.93, 0.29), P>0.05], or in the incidences of mal-alignment of coronal femoral component>3°[RR=0.79, 95%CI(0.60, 1.05), P >0.05], mal-alignment of axial femoral component>3°[RR=0.78, 95%CI(0.60, 1.01), P>0.05], sagittal femoral component alignment >3°[RR =0.82, 95%CI(0.49, 1.39), P>0.05], nor in the postoperative knee society functional scores[MD=5.26, 95%CI(-1.52, 12.04), P>0.05], hemorrhage volume[MD=-17.23, 95%CI(-46.43, 11.97), P >0.05] and the rates of complications[RR =1.29, 95%CI(0.53, 3.15), P >0.05] between the two methods. There was significant difference in operating time [MD=13.4, 95%CI(9.40, 16.67), P<0.01]. Compared with the conventional groups, the patients in the computer navigation group had significantly better lower limbs mechanical axis[MD =-0.67, 95%CI(-1.08, -0.25), P<0.01] and sagittal tibial component alignment[MD =-1.26, 95%CI(-1.80, -0.72), P<0.01], the knee society scores [MD =5.26, 95%CI(-1.52, 12.04), P=0.13] and the range of motion(ROM)[MD =2.36, 95%CI(0.79, 3.93), P<0.01] as well; and showed significantly lower risks of the mechanical axis mal-alignment >3°[RR =0.60, 95%CI(0.50, 0.71), P<0.01], lower incidences of tibial component mal-alignment >3° both in coronal[RR =0.50, 95%CI(0.28, 0.90), P<0.05] and sagittal positions[RR =0.61, 95%CI(0.44, 0.83), P<0.01].

The computer navigation could improve the alignment of the component position and the range of motion; it’s propitious to the postoperative knee function evaluation. Although the computer navigation may prolong the operating time, there is no obvious difference in hemorrhage volume and the rates of complications compared with conventional surgery.