Transforming growth factor-beta 1 and 2 in the synovial-like interface membrane between implant and bone in loosening of total hip arthroplasty.

OBJECTIVE

Development of a synovial-like membrane in the implant-bone or cement-bone interface has been linked to aseptic loosening of total hip arthroplasties (THA). This tissue consists of a fibrous stroma containing blood vessels and macrophages, but with relatively few lymphocytes, compared to "autoimmune" rheumatoid synovitis. Our aim was to examine transforming growth factor-beta (TGF-beta) in the synovial-like membrane of the interface and pseudocapsular tissue of loose THA and compare it to control knee synovial membrane.

METHODS

Twenty samples obtained from 10 patients with loose THA at revisions performed for aseptic loosening and 10 samples of knee synovial membrane as controls were analyzed for TGF-beta expression using rabbit antihuman TGF-beta 1 and TGF-beta 2 IgG in immunohistochemical staining. Results were quantitated by a semi-automatic VIDAS image analysis system.

RESULTS

Immunoperoxidase staining disclosed TGF-beta in macrophages and fibroblasts and also in some vascular endothelial cells and in occasional lymphocytes. Image analysis showed an increased number of positive cells/mm2 of both TGF-beta 1 (2327 +/- 212 vs 946 +/- 136; p < 0.01) and TGF-beta 2 (2292 +/- 594 vs 311 +/- 113; p < 0.01) compared to the control tissue. Increased expression of both TGF-beta 1 and TGF-beta 2 was also shown in the pseudocapsule (3210 +/- 585 and 1796 +/- 214). Use of cement or type of alloy did not seem to have any great effect on local expression of TGF-beta.

CONCLUSION

Profibrotic and immunosuppressive TGF-beta are increased in the synovial-like membrane in periprosthetic tissues around loose hip prostheses. They may play a role in the formation, maintenance, and growth of the interface tissue, and thus in the aseptic loosening of THA.