Patient‐specific in silico models can quantify primary implant stability in elderly human bone

• Published in: Journal of orthopaedic research Vol. 36; no. 3; pp. 954 - 962
• Main Authors: Steiner, Juri A., Hofmann, Urs A.T., Christen, Patrik, Favre, Jean M., Ferguson, Stephen J., van Lenthe, G. Harry
• Format: Journal Article
• Language: English
• Published: United States 01-03-2018
• Subjects: Aged; Aged, 80 and over; Bone Screws; Computer Simulation; Female; Finite Element Analysis; Humans; Humerus - diagnostic imaging; Male; micro‐CT; Patient-Specific Modeling; patient‐specific FEA; primary implant stability; X-Ray Microtomography; micro-CT; patient-specific FEA; primary implant stability
• ISSN: 0736-0266; 1554-527X
• DOI: 10.1002/jor.23721

ABSTRACT
Secure implant fixation is challenging in osteoporotic bone. Due to the high variability in inter‐ and intra‐patient bone quality, ex vivo mechanical testing of implants in bone is very material‐ and time‐consuming. Alternatively, in silico models could substantially reduce costs and speed up the design of novel implants if they had the capability to capture the intricate bone microstructure. Therefore, the aim of this study was to validate a micro‐finite element model of a multi‐screw fracture fixation system. Eight human cadaveric humerii were scanned using micro‐CT and mechanically tested to quantify bone stiffness. Osteotomy and fracture fixation were performed, followed by mechanical testing to quantify displacements at 12 different locations on the instrumented bone. For each experimental case, a micro‐finite element model was created. From the micro‐finite element analyses of the intact model, the patient‐specific bone tissue modulus was determined such that the simulated apparent stiffness matched the measured stiffness of the intact bone. Similarly, the tissue modulus of a small damage region around each screw was determined for the instrumented bone. For validation, all in silico models were rerun using averaged material properties, resulting in an average coefficient of determination of 0.89 ± 0.04 with a slope of 0.93 ± 0.19 and a mean absolute error of 43 ± 10 μm when correlating in silico marker displacements with the ex vivo test. In conclusion, we validated a patient‐specific computer model of an entire organ bone‐implant system at the tissue‐level at high resolution with excellent overall accuracy. © 2017 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:954–962, 2018.