Background: Proximal tibial strain in medial unicompartmental knee arthroplasty (UKA) may alter bone mineral density and cause pain. The aims of this retrospective cohort study were to quantify and compare changes in proximal tibial bone mineral density in metal-backed and all-polyethylene medial UKAs, correlating these with outcome, particularly ongoing pain.
Methods: Radiographs of 173 metal-backed and 82 all-polyethylene UKAs were analyzed using digital radiograph densitometry at 0, 1, 2, and 5 years. The mean grayscale of 4 proximal tibial regions was measured and converted to a ratio: the GSRb (grayscale ratio b), where GSRb>1 represents relative medial sclerosis.
Results: In both implants, GSRb reduced significantly to 1 year and stabilized with no differences between implants. Subgroup analysis showed less improvement in Oxford Knee Score in patients whose GSRb increased by more than 10% at 1 year (40/255) compared with patients whose GSRb reduced by more than 10% at both 1 years (8.2 vs 15.8, P=.002) and 5 years (9.6 vs 15.8, P=.022). Patients with persistently painful UKAs (17/255) showed no reduction in GSRb at 1 year compared with a 20% reduction in those without pain (P=.05).
Conclusions: Bone mineral density changes under medial UKAs are independent of metal backing. Medial sclerosis appears to be associated with ongoing pain.
Keywords: bone mineral density; bone strain; digital radiodensitometry; unexplained pain; unicompartmental knee arthroplasty.
Copyright © 2016 Elsevier Inc. All rights reserved.