Computational Framework for Determining Patient-Specific Total Knee Arthroplasty Loading
The demand for total knee arthroplasty (TKA) is increasing steadily. In 2007, Kurtz et al. [1] predicted that TKA procedures would increase from 402,100 in 2003 to 3.48 million by 2030. Recent US national inpatient survey data have borne out these trends [2, 3]. Furthermore, demand is growing fastest in people younger than 65 [4] — patients who will need their implants to last the longest. The major factors limiting prosthesis longevity involve wear of the polyethylene bearing surfaces. Wear continues to be a problem at the knee; for example, advances that reduce hip implant wear such as crosslinking of polyethylene are not widely used in TKA due to fears of early material breakdown under knee loading conditions [5]. Preclinical TKA testing is performed with knee wear simulators under generic walking conditions. Efforts are ongoing by us [6] and others [7] to improve the physiological relevance of current testing standards. Nevertheless, a simulator would need to run ∼eight months continuously to simulate 20 years of walking, assuming one-million steps per year and speed of one cycle per second. As a complementary tool, computational models can test multiple conditions efficiently and ensure a faster turnaround time in the design process to eliminate inferior designs earlier. The purpose of this work is to describe a computational framework for predicting TKA loading, and ultimately implant longevity, on a patient-specific basis. The rationale is that, after developing a patient-specific computational framework, TKA designs of any material and under any patient behavior can be modulated to promote contact conditions best for implant longevity.