Association between contact hip stress and RSA-measured wear rates in total hip arthroplasties of 31 patients

BACKGROUND: The main concern in the long run of total hip replacements is aseptic loosening of the prosthesis. Optimization of the biomechanics of the hip joint is necessary for optimization of long-term success. A widely implementable tool to predict biomechanical consequences of preoperatively planned reconstructions still has to be developed. A potentially useful model to this purpose has been developed previously. The aim of this study is to quantify the association between the estimated hip joint contact force by this biomechanical model and RSA-measured wear rates in a clinical setting. METHODS: Thirty-one patients with a total hip replacement were measured with RSA, the gold standard for clinical wear measurements. The reference examination was done within 1 week of the operation and the follow-up examinations were done at 1, 2 and 5 years. Conventional pelvic X-rays were taken on the same day. The contact stress distribution in the hip joint was determined by the computer program HIPSTRESS. The procedure for the determination of the hip joint contact stress distribution is based on the mathematical model of the resultant hip force in the one-legged stance and the mathematical model of the contact stress distribution. The model for the force requires as input data, several geometrical parameters of the hip and the body weight, while the model for stress requires as input data, the magnitude and direction of the resultant hip force. The stress distribution is presented by the peak stress-the maximal value of stress on the weight-bearing area (p(max)) and also by the peak stress calculated with respect to the body weight (p(max)/W(B)) which gives the effect of hip geometry. Visualization of the relations between predicted values by the model and the wear at different points in the follow-up was done using scatterplots. Correlations were expressed as Pearson r values. RESULTS: The predicted p(max) and wear were clearly correlated in the first year post-operatively (r = 0.58, p = 0.002), while this correlation is weaker after 2 years (r = 0.19, p = 0.337) and 5 years (r = 0.24, p = 0.235). The wear values at 1, 2 and 5 years post-operatively correlate with each other in the way that is expected considering the wear velocity curve of the whole group. The correlation between the predicted p(max) values of two observers who were blinded for each other's results was very good (r = 0.93, p < 0.001). CONCLUSION: We conclude that the biomechanical model used in this paper provides a scientific foundation for the development of a new way of constructing preoperative biomechanical plans for total hip replacements.