Characterization of long-term femoral-head-penetration rates. Association with and prediction of osteolysis

Abstract

Background: We examined the relationship between long-term femoral-head-penetration patterns and osteolysis in a ten-year follow-up study of a well controlled patient population. The purposes of this study were to characterize the linearity of long-term head-penetration patterns over time, to describe the relationship between ten-year true wear rates and osteolysis, and to determine whether the occurrence of osteolysis at ten years could be predicted by penetration data obtained prior to five years.

Methods: Temporal femoral-head-penetration patterns were examined at a minimum of ten years after forty-eight primary total hip arthroplasties. The arthroplasties were performed with the use of an Arthropor acetabular cup (Joint Medical Products) and a thirty-two-millimeter-diameter cobalt-chromium femoral head (DePuy). Using a computer-assisted radiographic technique, we evaluated two-dimensional head penetration on serial annual radiographs. Linear regression analysis modeled penetration-versus-time data as a line for each patient. The slope of the regression line indicated the true wear rate for each patient. In a subgroup of thirty-four hips for which three annual radiographs had been made less than five years after the arthroplasty, we compared early head-penetration patterns with the later occurrence of osteolysis.

Results: For all forty-eight hips, the true wear rate averaged 0.18 millimeter per year (range, 0.01 to 0.44 millimeter per year) and temporal head-penetration patterns tended to be linear (mean r2 = 0.91 +/- 0.16). Osteolysis at ten years was strongly associated with increasing true wear rates (p < 0.001). Osteolysis did not develop in any of the nine hips with a true wear rate of less than 0.1 millimeter per year. However, osteolysis developed in nine (43 percent) of twenty-one hips with a rate between 0.1 and less than 0.2 millimeter per year, in eight of ten hips with a rate between 0.2 and 0.3 millimeter per year, and in all eight hips with a rate of greater than 0.3 millimeter per year. Evaluation of early true wear rates as a predictor of late osteolysis showed a similar relationship.

Conclusions: This study demonstrates that true wear rates tend to be constant and that increased true wear is significantly associated with osteolysis at ten years after the operation. A similar relationship was also found at the early follow-up interval, indicating that early true wear rates (determined from serial radiographs) might enable orthopaedists to predict if patients are at risk for the development of osteolysis.

Clinical relevance: On the basis of these findings, we use temporal femoral-head-penetration data in our practice to evaluate polyethylene inserts in asymptomatic patients, to estimate the time to component wear-through, and to adjust the frequency of follow-up evaluations for monitoring the development of osteolytic lesions in at-risk patients.