The incidence of clinically significant osteolysis was based on the number of unrevised THAs (total hip arthroplasties) with at least 1.5 square centimeters of pelvic and/or femoral osteolysis. Osteolysis was defined as an area of localized loss of trabecular bone or cortical erosion that was not apparent on the pre-operative or immediate postoperative radiograph. To obtain lesion sizes, the defects were outlined on the anteroposterior pelvic radiograph and the area of the lesion was measured using Martell's Hip Analysis Suite software. Lesions were considered clinically important if the total area of osteolysis around a hip replacement was at least 1.5 square centimeters.

Minimum 9-year radiographs used to assess osteolysis were available for 79 unrevised hips with Marathon and 68 unrevised hips with Enduron polyethylene.

A single reviewer, who was blinded to the type of polyethylene liner, evaluated femoral head penetration among all unrevised hips with minimum 9-year radiographic follow-up using serial anteroposterior pelvic radiographs. Two-dimensional head penetration was determined for each follow-up radiograph relative to the immediate post-operative (nominal 6-week follow-up) reference view using Hip Suite Analysis version 8.0 with elliptical correction, a validated, computer-assisted technique. A linear wear rate was evaluated for each hip that had a minimum of three follow-up radiographs by using a least-squares linear regression to calculate the slope of the best-fit line for the wear vector magnitude versus time in situ data. The slope from this regression represented the steady-state linear wear rate. The steady-state linear wear rate data from all hips in a group was used to compute a mean polyethylene wear value.

At least 3 head penetration measurements evaluated with Martell's Hip Suite Analysis software were available for 76 unrevised hips with Marathon and 66 unrevised hips with Enduron polyethylene.

The rate of reoperation was based on the number of reoperations in each group. Any additional surgery after a participant's initial hip replacement was considered a reoperation.

All 116 hips randomized to Marathon polyethylene and all 114 hips randomized to Enduron polyethylene were included in the analysis population to determine the rate of reoperation.

The Harris Hip Score measures outcome after hip replacement and is based on a scale from 0 (worst) to 100 (best).

Harris Hip Scores were evaluated for 74 unrevised hips with Marathon and 65 unrevised hips with Enduron polyethylene that had minimum 9-year follow-up and complete data to compute a score.

Patient satisfaction was quantified by asking participants to respond "yes" or "no" to the question, "Are you satisfied with the results of your hip operation?"

Satisfaction rates were evaluated among 84 participants with unrevised hips who had Marathon and 70 participants with unrevised hips who had Enduron polyethylene. These participants had minimum 9-year follow-up and responded to the question about satisfaction.