Background: Twenty subjects were analyzed in vivo with use of video fluoroscopy to determine if the femoral head separates from the acetabular component during normal gait and to determine if the amount of separation differs between metal-on-metal and metal-on-polyethylene total hip prostheses.

Methods: Ten subjects had been treated with a metal-on-metal total hip arthroplasty and ten, with a metal-on-polyethylene total hip arthroplasty. All of the prostheses were implanted by the same surgeon utilizing the same surgical technique, and all were judged to be clinically successful (a Harris hip score of >90 points). Each subject walked with a normal gait on a level treadmill while under fluoroscopic surveillance. The two-dimensional fluoroscopic videotapes were then converted into three-dimensional images with use of a computer-automated model-fitting technique. Each implant was analyzed at various flexion angles to assess the amount of femoral head sliding.

Results: No femoral head sliding was observed in the subjects with a metal-on-metal implant, whereas all ten subjects with a metal-on-polyethylene implant had sliding that was greater than our threshold value of 0.75 mm. The average amount of femoral head sliding in these subjects was 2.0 mm, and the sliding was observed during the swing phase of gait. The sliding was typically seen medially while the femoral head remained in contact with the acetabular component superolaterally.

Conclusions: Femoral head sliding commonly occurred following traditional metal-on-polyethylene total hip arthroplasty but not after metal-on-metal arthroplasty. These kinematic data may be of value in future hip-simulation studies to better duplicate wear patterns observed in retrieval analyses, assist in the understanding of the lubrication and wear rates of metal-on-metal designs, and facilitate designing of prosthetic components that minimize wear and optimize hip kinematics.