To systematically evaluate the clinical efficacy of single condylar replacement (unicompartmental kenn arthroplasty, UKA) and total knee replacement (total knee arthroplasty, TKA) in the treatment of knee osteoarthritis.

Cochrane Library, Pubmed, Web of science, China national knowledge infrastructure (CNKI), Wanfang database, Chinese biomedical literature service system, published clinical studies on the treatment of knee osteoarthritis by single condyle replacement and total knee replacement were searched by computer. Two researchers screened the literature and extracted the relevant data according to the naught criteria. The inclusion criteria were randomized controlled trials (RCT) or non-randomized controlled trials (no-RCT). The patients underwent total knee arthroplasty or unicondylar replacement for the first time. The KSS score of knee joint, the time required for flexion of knee joint to reach 90 degrees, joint mobility, pain and other outcome indicators were observed and compared.Duplicate publications, original relevant data could not be obtained, full texts could not be obtained, and low-quality literature were excluded. RevMan 5.3.3 software was used for data processing: measurement data using mean differences (MD) for combined statistics, counting information using relative risk (RR) for combined statistics, all the effects data were described by 95% confidence interval (CI). Two researchers extracted data and determined heterogeneity between studies: if P ≥0.05, I²≤50%, using fixed effect model (fixed effects model); if no clinical heterogeneity, the random effects model would be adopted. If data could not be combined, descriptive analysis was performed.

A total of seven RCTS and 13 no-RCTS were included. Meta analysis results showed that the joint activity was measared in two, three, six, 12, 48 months. According to the results, between the two groups after two months [MD=7.31, 95% CI (0.99, 15.61), P=0.08] there was no heterogeneity, and in the mid-and long term, [MD=5.90, 95% CI(1.07, 10.74), P=0.02] there was heterogeneity. The monocondylar group [RCT: MD=-3.36, 95% CI(-6.22, -0.50), P=0.02, no-RCT: MD=-2.94, 95% CI)(-3.51, -2.37), P=0.02] was shorter than the total knee replacement group in the time for 90° flexion.Compared with the total knee replacement group, monocondylar replacement had the same efficacy in the early and middle stage of the postoperative KSS score of the knee joint, but had obvious advantages in the long-term efficacy.Intraoperative blood loss [RCT: MD=-176.99, 95% CI(-205.36, -148.62), P <0.01, no-RCT: MD =-139.08, 95% CI(-150.94, -127.22), P <0.01]and intraoperative drainage[MD=-77.33, 95% CI(-88.18, -66.48), P<0.01]had significant advantages over the total knee replacement group. In terms of operation time [no-RCT: MD=-0.43, 95% CI(-7.27, -6.40), P=0.90], no significant heterogeneity was found between the two groups.

Single condyle replacement shows no abvious advantage in the time needed for postoperative knee joint mobility, buckling 90 degrees, knee KSS score after operation, intraoperative blood loss, intraoperative led traffic obvious advantages, heterogeneity in terms of operation time, but more high quality, further clinical trials of large sample verification is still needed.