The optimal technique for component fixation in revision total knee arthroplasty (rTKA) remains controversial: full cementation (FC) versus hybrid fixation (HF), which involves press-fit stem with cement fixation in the metaphyseal and epiphyseal zones. Previous series have either demonstrated the superiority of one or the other of these techniques or their equivalence. However, few studies have compared these 2 methods for rTKA using the Legacy® Constrained Condylar Knee (LCCK) prosthesis (Zimmer, Warsaw, Indiana, USA).

Our hypothesis was that HF of LCCK components is associated with a higher rate of aseptic loosening (AL) than FC.

This was a single-center, multi-surgeon, retrospective study. Primary revisions between January 2010 and December 2014 were included for all indications. The only exclusion criterion was death without revision before the 5-year follow-up. The primary objective of this study was to compare the survivorship of 2 groups of LCCK components (femoral or tibial), depending on whether their stems had been cemented (HF versus FC), taking AL, revised or not, as the endpoint. The secondary objective was to look for other predictive factors of AL.

A total of 75 rTKAs (150 components) were included. The FC group (51 components) had more Anderson Orthopedic Research Institute (AORI) type 2B and type 3 bone defects (BDs) (p<0.001), more reconstructions using trabecular metal (TM) cones (19 FCs and 5 HFs; p<0.001), and bone allografts (p<0.001). At more than 5 years, none of the FC components were loose compared to 10 HF components (9.4%), with 4 of these stems revised. The only significant difference was survivorship without radiographic AL at 9 years (FC=100%; HF=78.6%; p=0.04). The only predictive factor of AL in the HF group was the filling of the diaphyseal canal (p<0.01). The detrimental effect of BD severity (p=0.78) and the protective effect of TM cones were not demonstrated (p=0.21).

Other series studying revisions using the same type of prosthesis also concluded the superiority of FC, not found for other revision prostheses. Despite this study's limitations (retrospective, multi-surgeon, limited sample size, and limited follow-up), all patient outcomes were known, and the difference in survivorship between the groups was very significant.

HF has not been proven effective for the LCCK prosthesis. Better diaphyseal filling, wider metaphyseal bone tunnels enabling better cement injections, and stem designs more appropriate for press-fit fixation could improve these results. TM cones are an interesting avenue for further research.

III; retrospective comparative study.