Abstract Articular cartilage, despite its resilience, is vulnerable to low-energy impacts that initiate microcracks within its collagen network, potentially leading to osteoarthritis (OA). This study integrates experimental evidence and a multiscale computational framework to predict the initiation and propagation of such microdamage. Building on prior experimental characterizations of type II collagen fibril orientation, stress?stretch behavior, and failure properties, we employed the Finite Elements of Multiscale Mixtures (FE2M) framework within FEBio (University of Utah) to couple macroscale cartilage deformation with microscale fibril mechanics. We generated Statistically Equivalent Representative Volume Elements (SERVEs) to replicate the anisotropic collagen architecture of the superficial zone. The combined approach enabled simulations of impact and cyclic compression that replicate in vitro loading conditions. Validation against data from second harmonic generation microscopy demonstrated strong agreement between experimentally measured microcrack fractions and FE2M-predicted fiber failure fractions. Sensitivity analyses confirmed that model responses were robust to moderate variations in fibril stiffness parameters (c1, c3). High-impact simulations revealed broader stress distributions and greater fibril-level heterogeneity, underscoring the role of stress, rather than stretch, as a more reliable predictor of failure. Although computational and biological variability introduced uncertainty at relatively high loading conditions, the FE2M framework accurately captured multiscale mechanical behavior and microdamage trends. This validated multiscale approach provides a predictive and mechanistically grounded tool for investigating cartilage degeneration, offering potential applications in assessing injury risk, optimizing surgical interventions, and informing preventative strategies for OA progression.
Safari et al. (2026) studied this question.