A recent meta-analysis for total knee replacement (TKR) undertaken with navigation demonstrated improved accuracy of implant positioning but did not have sufficient evidence on functional outcomes. This meta-analysis evaluates the functional outcomes for TKR with and without navigation.

We present a randomized prospective and comparative studies on functional outcomes of TKR with and without navigation were identified. The selected articles were tested for publication bias and heterogeneity. Studies presenting the functional outcomes in terms of knee society score, oxford knee score and HSS scores were included in the study. The data was then aggregated by random-effects modelling after which estimated weighted mean differences for individual functional scores were calculated.

Sixty two studies were identified and reviewed independently by two researchers; ten studies fulfilled the inclusion criteria, resulting in 976 cases for the meta-analysis (490 with navigation and 486 without). The mean age of both groups was similar (68 and 69). Results of a meta-analysis are best demonstrated by funnel graphs, forest plot, P values, and confidence intervals. In summary, the weighted mean of difference of KSS score is −0.288, with a p value of 0.867. Moreover, the standard means of difference of Oxford knee score was 0.133, with a P value of 0.257. The HSS scores resulted in a SMD of −0.099 and a ‘p’ value of 0.686. Finally, the range of motion weighted means of difference was 1.428, and a ‘p’ value of 0.228.

This meta-analysis demonstrated that there is no statistical advantage to the use of navigation for TKR in terms of functional outcome; the increased positional accuracy does not impart an improved outcome as has been proposed. This is probably due to the tolerances available in the implant positioning. This meta-analysis is providing the early outcomes & highlighting the necessity of long term studies.