Management of bone loss in revision total knee arthroplasty with tantalum cones: Primary aseptic revision versus multiple revisions.

Abstract

The increasing number of revisions following total knee arthroplasty (TKA) has led to the appearance of new challenges, including the management of bone defects. The use of porous tantalum cones is one of the options to address this bone loss. We carried out a retrospective study to compare the medium-term outcomes of single and multiple TKA revisions in the context of massive bone defects using tantalum cones. We analyzed: (1) the survival of the revision implants, (2) osseointegration of the cone, (3) the clinical and functional outcomes, (4) complications. Between July 2011 and May 2019, 79 metaphyseal reconstructions (MRs) were performed in 58 patients (mean age 68 ± 11 (41-89)) who received tantalum cones during the procedure. Revisions for infection were excluded. The massive bone defects were classified as Anderson Orthopaedic Research Institute types 2B/3. The cone had to be removed in 12 patients (16 MRs). Implant survival free of all-cause revision and cone survival free of aseptic loosening were calculated. Osseointegration of the cones was determined on radiographs. A subset of patients (n = 46) underwent a clinical assessment including the OKS, KSS, KOOS, satisfaction level and knee flexion. All of the patients (n = 58) underwent a radiological assessment. Complications were documented during the follow up period. The mean follow up was 73.1 months ± 29.8 (12-143). In all, 91 cones (48 at the femur and 43 in the tibia) had been implanted during 79 MR procedures: 40 in the primary revision group (R-primary) and 39 in the multiple revision group (R-multiple). Survivorship of the implant free of all-cause revision at 6 years was better in the R-primary group, 97.5 % (95 %CI: 93 % to 100 %) than in the R-multiple group, 70.5 % (95 %CI: 57 % to 87 %) (P < 0.01). Survivorship of the cone free of aseptic loosening at 6 years was comparable between the two groups: 100 % (95 %CI 100 % to 100 %) for R-primary and 96.8 % (95 %CI 91 % to 100 %) for R-multiple (P = 0.26). At the final assessment, 10 of 91 cones had no bone ingrowth; all of them were in the R-multiple group (P = 0.012). The functional outcomes were better in the R-primary group.The mean OKS was 31.5 ± 8.7 (16-45) vs. 25.7 ± 9.1 (9-41) in the R-multiple group (P = 0.014). The mean KSS global was 146.2 ± 32.3 (69-197) vs. 136.8 ± 20.9 (98-182) in the R-multiple group (P = 0.082). The KOOS global was 61.3 ± 17 (26-84) vs. 48.3 ± 15.7 (18-79) in the R-multiple group (P < 0.01). In the R-primary group, 78 % of patients (22/28) were satisfied or very satisfied with the outcome vs. 61 % (11/18) in the R-multiple group (P = 0.25). The mean knee flexion at the final assessment was 102.7 ± 24.2 (20-125) in the R-primary group vs. 98.3 ± 16.5 (70-120) in the R-multiple group (P = 0.36). A re-revision was needed in 16 of the 79 MRs (20 %). The most common cause of failure was a periprosthetic infection in almost half of these cases (eight of 16). Only one re-revision was needed in the R-primary group, while 15 were needed in the R-multiple group (P < 0.001). MR using porous tantalum biomaterials is a viable technique for managing massive bone defects during TKA revision. Multiple revisions were associated with worse implant survivorship and worse functional outcomes than primary (or first) revisions, along with having a higher likelihood of infection-related complications. The survivorship of the cones free of aseptic loosening was comparable between primary and repeat revisions. Loosening was more frequent on the femoral side than the tibial side.