Revision total knee arthroplasty: should tibial stems be cemented or uncemented?

Thirty-two consecutive coordinate revision total knee replacements (17 cemented, 15 cementless) with longer than 2-year follow-up were radiographically reviewed for assessment of radiosclerotic and radiolucent lines, fit and fill of the stem, constraint of the prosthesis (posterior stabilized vs. varus–valgus constraint) and whether the stems were cemented or uncemented. Radiosclerosis was noted around uncemented stems and radiolucency was identified at the cement prosthesis interface on anteroposterior and lateral X-rays. Radiolucent or sclerotic lines were rated according to the number of zones involved on most recent follow-up radiographs and were recorded if progressive from initial post-operative radiographs. There were 17 cemented and 15 uncemented tibial stems. Twenty-two posterior stabilized and 10 varus–valgus constrained implants, 17 cemented and 15 uncemented stems and 16–32 stems rated as having poor fit and fill. The cemented stems had fewer zones involving radiolucent–sclerotic lines than the uncemented tibial stems (P=0.006). Prosthesis constraint and stem fit and fill did not statistically influence radiographic signs of loosening. This short-term follow-up study suggests that cement may afford better early stability to the tibial component. The difference in constraint between a posterior stabilized implant and a varus–valgus constrained implant may not be great enough to influence radiographic appearance in the short-term. Similarly, stem fit and fill differences may not have been large enough to demonstrate a difference in radiographic findings.