Femoral component migration in total knee arthroplasty: randomized study comparing cemented and uncemented fixation of the Miller-Galante I design.

The Miller-Galante I knee replacement was inserted in 25 women and three men (33 knees) with osteoarthrosis. All patients received a TiAlV femoral component with a commercially pure titanium fiber-mesh undersurface. Cemented or cementless fixation was used based on a randomization protocol. Micromotions of the femoral components were recorded with roentgen stereophotogrammetric analysis during the first 2 postoperative years. The magnitude of migration did not differ between cemented and uncemented fixation. The number of nonmigrating prostheses decreased from 21 (12 cemented and nine uncemented) at 3 months to six (three cemented and three uncemented) at 24 months. In both groups, the magnitude of prosthetic tilting about the longitudinal axis (internal-external rotation) was as large as that about the transverse axis (flexion-extension). Rotation into extension was as common as rotation into flexion. The largest translations were recorded at either of the posterior condyles. In 10 uncemented components, radiolucent lines were seen at the distal interface postoperatively. Proximal migration of the femoral component was recorded in these knees, and the width of the lines decreased or the lines disappeared totally at 24 months. After 2 years, lines were noted around four cemented and four uncemented replacements, mainly anteriorly or distally. All of these prostheses migrated. One prosthesis, revised because of malalignment, displayed pronounced migration after an initial period of stability. Bone ingrowth was observed anteriorly and anterodistally despite the presence of motions of 1 mm or more.