Prediction of long-term polyethylene wear in total hip arthroplasty, based on early wear measurements made using digital image analysis.

The capability to reliably predict long-term in vivo wear of polyethylene would be of great value for the early identification of problematic total hip designs. Formal quantitative estimates of long-term polyethylene wear were made from a series of 197 patients who had a total hip arthroplasty and who were followed for a minimum of 10 years; the estimates were based on the wear that was apparent radiographically at nominally 2 years after the operation. A newly developed digital image-analysis edge-detection procedure was applied to 1,237 archived follow-up radiographs. The edge-detection measurements were analyzed with a robust regression random-coefficients statistical formulation developed especially to address the distributions of wear rate observed across this population over time. Formal regression equations were reported, which can be used to estimate late-wear depth for a patient radiographed at a 2-year follow-up visit. Series wide, the correlation between predicted and observed wear depths was 0.73 at 4 years, with a correlation decline of approximately 0.03 per additional year.