Toward in silico trials to estimate the risk of aseptic loosening in new joint replacement designs

Aseptic loosening remains the most common failure mode of total joint arthroplasty (TJA), defined as the failure of the bone-implant fixation in the absence of infection. Studying the effect of implant designs on failure risk is essential to achieving long-term clinical success. Conventional approaches, such as in vivo experiments, animal studies, and clinical trials, are used to assess implant safety and predict aseptic loosening risk. However, these methods are limited by biological irrelevance, species differences, and high costs or long durations. In contrast, in silico trials have emerged as a powerful alternative to investigate the individual joint biomechanics and support the orthopaedic device’s safety and/or efficacy. Particularly, finite element analysis (FEA) model enables detailed analysis of TJA outcome by involving specific patients, implant design and even the surgical procedures. The aim of this thesis is to develop an in silico trial to estimate the risk of aseptic loosening in an anatomical cementless hip stem design. FEA models were used to predict the primary stability (the relative micromotion in each point of the bone-implant contact surface induced by physiological activities such as level walking or stair climbing) of a commercial stem design (Apta, Adler Ortho), for which extensive clinical experience is available. Specific thresholds derived from published experimental mechanobiology studies were used to decide which portion of the stem surface may experience fibrous differentiation, the first step toward aseptic loosening. This basic model was then used to explore the primary stability of the Apta stem as a function of the femoral size, the subject’s weight, the pre-operative bone quality (degree of osteopenia), and the surgical uncertainty on implant sizing. To support large-scale assessments, an automated FEA pipeline for orthopaedic devices was implemented, allowing for efficient batch processing of patient data, physiological loading cases, implant variations, or surgical plans in the future.