Cementless knee replacement. Experimental, clinical and radiostereometric investigations of the tibial component

Abstract

Aseptic loosening is the major failure mode in joint replacement surgery. The aim of the present study was to evaluate different fixation principles of the tibial component in cementless total knee replacement. An experimental study of different surface structures (ceramic and titanium) was performed. Cementless knee replacements with macrointerlock design were followed prospectively. New ceramic coated tibial component concepts were developed and compared with porous coated fixation. Implant stability was assessed in vivo by Radiostereometry (RSA) and correlated to the amount of bone mineral in the proximal tibia.Conical implants made of smooth or rough structured titanium or with hydroxyapatite (HA) coating were inserted in the proximal tibia prior to prosthetic replacement in osteoarthrotic patients. After 3 months, increased bone apposition was found in the rough structured and HA-coated implants on the order of 50% of the measured area. The smooth implants were mostly encapsulated in fibrous tissue.144 uncemented Freeman-Samuelson prostheses were followed prospectively for a minimum of 5 years. The fixation of the tibial components depended on macrointerlocking pegs and did not aim at bone ingrowth. A high failure rate in terms of revision and radiographic loosening was observed.51 knees were stratified to either a Freeman-Samuelson hydroxyapatite-coated (FS HA) or a porous coated Miller-Galante II (MG II) prosthesis. After 5 years, the clinical results were excellent in both groups. RSA revealed that after 5 years the FS HA knees had less migration with regard to MTPM (mean MTPM: FS HA 0.6 mm, MG II 0.8 mm, p= 0.03) and maximum subsidence (mean max subsidence: FS HA -0.2 mm, MG II -0.5 mm, p= 0.01). At 2 months and 1 year, there was a general tendency toward less inducible displacement in the HA-coated prostheses. 40 knees were allocated to a Miller-Galante II prosthesis, with or without hydroxyapatite/ tricalciumphosphate (HA/TCP) coating on the tibial undersurface. After 2 years, the HA/TCP knees displayed less anterior or posterior tilt and subsidence. The MTPM values at 2 years were low in both groups (0.5 mm). There were more radiolucent lines under the tibial tray and around the stem in the uncoated group. HA/TCP coating seems to improve the quality of the interface between the tibial component and the bone.The amount of bone mineral (BMA) was measured by triple-energy absorptiometry in the medial tibial condyle postoperatively, after 1 year and 4-5 years in 38 knees in the FS HA-MG II study. The overall mean BMA decrease was 26% after 4-5 years. The FS HA group had more pronounced reduction of BMA, thus indicating stress shielding of the proximal tibia. Radiodense lines around the stem of the FS HA knees occurred frequently as a sign of bony fixation. There was an inverse correlation between subsidence and change in BMA after 1 year but no correlation was found after 4-5 years. The stability obtained in these modern uncemented knee prostheses is encouraging. Ceramic coating of the tibial component improved the fixation and seems to be a promising alternative to cemented fixation.