Femoral Remodeling Around Well-Fixed Cemented Charnley Total Hip Arthroplasty: A Minimum Twenty Year Follow-Up Study

Introduction: In 1968, Charnley observed two unusual patterns of remodeling of the proximal femur in response to the femoral component of well-fixed total hip arthroplasties: cortical thinning leading to endosteal widening around the femoral component and hypertrophy of the distal femoral cortex. No previous study has specifically investigated the quantitative remodeling characteristics of the proximal femur to well-fixed cemented femoral components or correlated the patterns of change with patient demographic variables. The aim of this study was to provide a quantitative analysis of the long-term femoral remodeling at minimum 20-year radiographic follow-up as well as analyze those patient demographic variables which may contribute to the observed patterns of remodeling. Methods: The study group consisted of 130 primary cemented Charnley femoral components performed by a single surgeon, which were all well-fixed at minimum 20-year radiographic follow-up. All radiographs were classified into one of three groups based on the gross appearance: distal hypertrophy, endosteal loss, or normal. The gross remodeling type was statistically analyzed with respect to four patient demographic variables: sex, age at surgery, body mass index, and postoperative activity level. All radiographs were digitized and analyzed using a digital edge detection program on Matlab®. The total femoral diameter, medial cortical width, and lateral cortical width were measured at four points: the inferior aspect of the lesser trochanter, the tip of the femoral component, the midpoint between the inferior aspect of the lesser trochanter, and the tip of the femoral component, and 2.5 cm distal to the tip of the of the femoral component. All measurements were corrected for magnification by scaling to the known diameter of the femoral head. The total femoral diameter, endosteal diameter, medial cortical thickness, and lateral cortical thickness were recorded at each level. Average values at the four points were compared for each of the three gross remodeling types for statistical significance. IRB approval was obtained. Results: According to the radiographic appearance of the proximal femur, 20 hips (15%) showed endosteal loss, 18 hips (14%) showed distal cortical hypertrophy, and 92 hips (71%) showed normal remodeling characteristics. Statistical analysis showed that the distal cortical hypertrophy group had a significantly younger age at surgery than both the endosteal loss group (p=0.001) and the normal remodeling group (0.007). There was no significant difference between the three groups in regards to gender, activity level, or BMI. The average percent change in the proximal femoral dimensions are shown in the tables below: Normal Remodeling Total width Endosteal width Medial Cortex Lateral Cortex Proximal 7.1% 5.4% -14.2% 0.4% Midstem 7.0% 7.6% 11.1% 8.4% Tip 11.1% 14.9% 23.2% 3.6% Distal 3.9% 7.5% 5.3% 3.6% Distal Cortical Hypertrophy Total width Endosteal width Medial Cortex Lateral Cortex Proximal 2.6% 2.3% -9.2% -6.1% Midstem 15.5% 20.8% 54.1% 22.9% Tip 29.6% 43.4% 95.9% 26.1% Distal 24.8% 7.0% 26.5% -1.6% Endosteal Loss Total width Endosteal width Medial Cortex Lateral Cortex Proximal 4.0% 3.7% -27.1% -27.8% Midstem 4.1% 6.6% -3.5% -15.5% Tip 6.8% 9.9% -1.2% -12.9% 2.5cm 3.5% 12.2% 3.2% -13.8% The distal cortical hypertrophy group showed a significantly greater increase in medial cortical thickness and endosteal width at the midstem, implant tip, and 2.5 cm distal to the tip (p<0.05) when compared with the endosteal loss and normal remodeling groups. The endosteal loss group showed a significantly decreased lateral cortical thickness at the lesser trochanter, midstem, and distal points when compared to the normal remodeling group (p<0.05). Discussion: This study provides information about the long-term effect of a cemented femoral component on bone remodeling of the proximal femur. This study showed that the proximal femoral dimensions differ even in femurs with well-fixed components over prolonged time intervals. Remodeling characteristics are difficult to predict preoperatively as the only demographic variable that correlated with remodeling type was younger patient age with distal cortical hypertrophy. This study should provide comparison for long term studies of cementless designs to determine whether bone remodeling is different with these cementless designs.