Stress Distribution at Hip Joint During Level Walking

Summary

In 1950, Bressler and Frankel [1] calculated dynamic hip joint force for the first time. Since then, researchers have reported the results obtained by various methods [2]. These methods are divided into two categories, based either on a mathematical model or on replacing the femoral head with a prosthesis. Of course the latter method is not universally acceptable. We adopted the model of Gilbert et al. [3], this being the simplest one, consisting of a system of two rigid segments and disregarding the change in center of pressure (COP) which moves from heel to toes. This model is acceptable under ideal special conditions in which only “bone to bone” force is calculated, that is, the pressure, based on the contraction of muscles which pass over joints, is disregarded. In other words, this model is suitable when there is ideally efficient walking with little contraction. Under this supposition, the force of the hip joint during level walking was calculated in the sagittal and frontal planes, using force platform, kinesiologic, and somatotype data for the stance phase. Two normals (a 29-year-old male; subject 1, and a 70-year-old female; 2) were examined in this study. Subject 1 walked rapidly and 2 walked slowly. We compared the components of the resultant force of the femoral head. In subject 1, the resultant force had a two-peak curve, the first peak reaching 2.7 times the body weight (bw); the second peak reached 2.2 times bw and the valley between the two peaks was 0.7 times bw. In subject 2, who had a slow gait, there was no finding of peaks as with subject 1, the shape being trapezoid; the maximal value was found in the latter half of the stance and was 0.88 times bw. In subject 1, the fore-aft components exceeded the vertical one, while in subject 2, the fore-aft component was larger than that of the ground reaction force, but did not exceed the vertical component at the femoral head. We depicted the stress distribution at the joint surface of the hip based on the calculated components, using a rigid body spring model (RBSM). Stress distribution was greatest posteriorly after heel contact; in the middle of the stance the stress to the vertical side was not so large. In subject 2, the stress was moderate in all phases of the stance. Since there is no way of verifying directly whether the three femoral components calculated above are acceptable, we attempted to verify these components indirectly, by comparing the vertical components calculated in the sagittal and frontal planes. In subject 1, the shape of the components was similar from heel contact to mid-stance, but suddenly after mid-stance the components in the sagittal plane decreased. In subject 2, the shape of the components was similar in all phases of the stance. Our findings could be useful for determining stress distribution at the intact hip joint surface of healthy subjects during level walking at a slow speed. However, in the latter half of the stance phase, with rapid walking, a more complete model, particularly in regard to the foot, would be preferable.