Medial knee cartilage is unlikely to withstand a lifetime of running without positive adaptation: a theoretical biomechanical model of failure phenomena

Abstract

Runners on average do not have a high risk of developing knee osteoarthritis, even though running places very high loads on the knee joint. Here we used gait analysis, musculoskeletal modeling, and a discrete-element model of knee contact mechanics to estimate strains of the medial knee cartilage in walking and running in 22 young adults (age 23 ± 3 years). A phenomenological model of cartilage damage, repair, and adaptation in response to these strains then estimated the failure probability of the medial knee cartilage over an adult lifespan (age 23–83 years) for 6 km/day of walking vs. walking and running 3 km/day each. With no running, by age 55 the cumulative probability of medial knee cartilage failure averaged 36% without repair and 13% with repair, similar to reports on incidence of knee osteoarthritis in non-obese adults with no knee injuries, but the probability for running was very high without repair or adaptation (98%) and remained high after including repair (95%). Adaptation of the cartilage compressive modulus, cartilage thickness, and the tibiofemoral bone congruence in response to running (+1.15 standard deviations of their baseline values) was necessary for the failure probability of walking and running 3 km/day each to equal the failure probability of walking 6 km/day. The model results suggest two conclusions for further testing: (i) unlike previous findings on the load per unit distance, damage per unit distance on the medial knee cartilage is greater in running vs. walking, refuting the “cumulative load” hypothesis for long-term joint health; (ii) medial knee cartilage is unlikely to withstand a lifetime of mechanical loading from running without a natural adaptation process, supporting the “cartilage conditioning” hypothesis for long-term joint health.

Biomechanics of the medial meniscus in the osteoarthritic knee joint

“…For this reason, such models were not useful for assessing the influence of degenerative changes on the pathomechanism of degenerative meniscal injury. The process of cartilage degeneration and damage was simulated in recent works ( Liukkonen et al, 2017 ; Miller & Krupenevich, 2020 ). Peters et al (2018) suggested that computational models of the knee joint should be constructed using material parameters and geometry from cohorts with consistent disease state.…”

Effects of Obesity on Medial Tibiofemoral Cartilage Mechanics in Females—An Exploration Using Musculoskeletal Simulation and Probabilistic Cartilage Failure Modelling

“…The peak net (N) and normalized medial tibiofemoral forces (BW) were calculated using data derived from the joint reaction analysis. The cumulative medial tibiofemoral load was computed as the quotient of the mean stance phase medial tibiofemoral force and the stride length [ 35 ]. Pilot walking data shows a minimal detectable difference (MDC) of 0.27 BW and a high level of reliability (ICC = 0.951) for the peak medial tibiofemoral force ( Table S1 ).…”

“…The medial tibiofemoral contact forces obtained from OpenSim were input into a model of medial knee contact mechanics which was utilized to calculate tibiofemoral cartilage stress and strain via adapted MATLAB source code [ 35 ]. Pilot walking data shown MDCs of 0.01 and 0.16 MPa and high reliability (ICC ≥ 0.945) for indices of medial tibiofemoral stress and strain ( Table S1 ).…”

“…The power law coefficient (1.0), Weibull coefficient (1.03) and power law exponent (12.9) were incorporated as constants in the quantification of time until failure. These parameters were extracted from Miller & Krupenevich, [ 35 ], who fit a power law function to the loading cycles to failure data of Riemenschneider et al [ 63 ].…”

“…However, although quantification of tibiofemoral joint kinetics is now feasible, there remain difficulties in exploring the influence of distinct mechanical and physiological conditions on the instigation, progression and temporal profile of knee OA. Therefore, probabilistic modelling of cartilage of stress and strain induced accumulative damage may be valuable for determining the effects of obesity on probability of osteoarthritic degeneration over a lifetime of cyclic loading [ 35 ]. However, probabilistic cartilage modelling has not been adopted to explore differences in medial tibiofemoral cartilage failure probability between obese and healthy weight individuals.…”