This study was designed to develop a model for predicting bone mineral density (BMD) loss of the femur after total hip arthroplasty (THA) using artificial intelligence (AI), and to identify factors that influence the prediction. Additionally, we virtually examined the efficacy of administration of bisphosphonate for cases with severe BMD loss based on the predictive model. The study included 538 joints that underwent primary THA. The patients were divided into groups using unsupervised time series clustering for five-year BMD loss of Gruen zone 7 postoperatively, and a machine-learning model to predict the BMD loss was developed. Additionally, the predictor for BMD loss was extracted using SHapley Additive exPlanations (SHAP). The patient-specific efficacy of bisphosphonate, which is the most important categorical predictor for BMD loss, was examined by calculating the change in predictive probability when hypothetically switching between the inclusion and exclusion of bisphosphonate.Aims
Methods
We measured the orientation of the acetabular and femoral components in 45 patients (33 men, 12 women) with a mean age of 53.4 years (30 to 74) who had undergone revision of metal-on-metal hip resurfacings. Three-dimensional CT was used to measure the inclination and version of the acetabular component, femoral version and the horizontal femoral offset, and the linear wear of the removed acetabular components was measured using a roundness machine. We found that acetabular version and combined version of the acetabular and femoral components were weakly positively correlated with the rate of wear. The acetabular inclination angle was strongly positively correlated with the rate of wear. Femoral version was weakly negatively correlated with the rate of wear. Application of a threshold of >
5 μm/year for the rate of wear in order to separate the revisions into low or high wearing groups showed that more high wearing components were implanted outside Lewinnek’s safe zone, but that this was mainly due to the inclination of the acetabular component, which was the only parameter that significantly differed between the groups. We were unable to show that excess version of the acetabular component alone or combined with femoral version was associated with an increase in the rate of wear based on our assessment of version using CT.
