EARLY FAILURES AMONG 14,009 CEMENTED AND 1,326 UNCEMENTED PROSTHESES FOR PRIMARY COXARTHROSIS - THE NORWEGIAN ARTHROPLASTY REGISTER, 1987-1992

In the Norwegian Arthroplasty Register, 15,335 primary total hip repla cements (THR) in patients with primary arthrosis were followed for 0-5 .4 years. The Kaplan-Meier estimate of cumulative failure (revision) a fter 4.5 years was 2.7 percent for cemented THR, compared to 6.5 perce nt for uncemented. In patients under 65 years the cumulative revisions for cemented and uncemented THR were 3.3 and 7.9 percent. For the ace tabular components, the cumulative failures were 0.6 percent for cemen ted and 1.7 percent for uncemented, and for femoral components 1.7 and 3.9 percent after 4.5 years. Adjusting for age and sex using a Cox re gression model, 2 times higher rates of failure were found comparing u ncemented to cemented THR. The results for uncemented Prostheses were more unfavorable in young patients. In men and women under 60, the rev ision rates were increased 6 and 3 times, respectively, for patients w ith uncemented THR compared to those with cemented THR. Restriction of the end-point to revision for aseptic loosening gave results similar to the over-all results. No difference between cemented and uncemented THR was seen for revisions due to infection, whereas the most unfavor able results for uncemented THR were seen when revisions due to causes other than infection and aseptic loosening were considered.