Fixed- versus mobile-bearing UKA: a systematic review and meta-analysis

Two design concepts are currently used for unicondylar knee arthroplasty (UKA) prostheses: fixed bearing (FB) and mobile bearing (MB). While MB prostheses have theoretical advantages over their FB counterparts, it is not clear whether they are associated with better outcomes. A systematic review was conducted to examine survivorship differences and differences in failure modes of between FB and MB designs. PubMed, Scirus and Cochrane library databases were searched for medial UKA outcome studies. A total of 44 papers, involving 9,463 knees, were eligible. Outcomes examined included knee function, survivorship and the reasons for, and incidence of, revision for FB and MB prostheses. Random effects meta-analysis was employed to obtain pooled revision rate estimates. Where available, cause-specific time to revision was extracted. Mean follow-up was 8.7 years for FB and 5.9 years for MB prostheses. There were no other relevant baseline differences. The overall crude revision rate for FB and for MB prostheses was 0.90 (95 % confidence interval (CI) 0.65-1.21) and 1.51 (95 % CI 1.11-1.93) per 100 component years, respectively. After stratification on follow-up time and age, the revision rates were not substantially different, aside for younger patients in short term from studies with short-term follow-up. No essential differences between the two designs were observed. MB and FB UKA designs have comparable revision rates. As our study is based on predominantly observational data, with large variations in reporting standards, inferences should be drawn with caution. IV.