Incidence, temporal trends and potential risk factors for aseptic loosening following primary unicompartmental knee arthroplasty: A meta-analysis of 96,294 knees

Background Aseptic loosening (AL) is among the major reasons for revision of failed primary unicompartmental knee arthroplasty (UKA). There is an ongoing temporal increase in the use of UKA with a resultant increase in the revision burden. We aimed to evaluate the incidence of, temporal trends and risk factors for AL. Methods Longitudinal studies reporting the incidence of AL following primary UKA were sought from MEDLINE, Embase, Web of Science and Cochrane Library up to 6th April 2020. Incidence and relative risks (RR) (with 95% confidence intervals) were calculated. Results We identified 62 studies for inclusion. Overall, 96,294 primary UKA procedures accounting for 1752 AL cases were included. AL incidence ranged from 0.00% to 22.70% over a 7.7 year weighted mean follow-up. The pooled random effects incidence (95% CI) was 1.77% (1.34–2.25) in the same follow-up period. The annual rate of AL was 0.10% (0.02–0.22). AL incidence increased with length of follow-up, but there was a temporal decrease from the 1970s onwards. Tibial loosening was more common than femoral component loosening: incidence (95% CI) of 1.63% (0.96–2.44) and 0.58% (0.20–1.09) respectively over a weighted follow-up of 6.6 years. Fixed bearing implant design and cemented fixation were both associated with increased AL risk, whereas robotic-assisted surgery was associated with decreased risk. Conclusion The overall incidence of AL following primary UKA is primarily driven by tibial component loosening and there is a temporal decline in rates. The use of mobile bearing, uncemented implants inserted with robotic assisted surgery may reduce the risk of AL.