Bone health deterioration is a major public health issue increasing the risk of fragility fracture with a substantial associated psychosocioeconomic impact. In the lumbar spine, physical deconditioning associated with ageing and chronic pain is a potential promoter of bone structural degradation. General guidelines for the limitation of bone loss and the management of pain have been issued, prescribing a healthy lifestyle and a minimum level of physical activity. However, there is no specific recommendation regarding targeted activities that can effectively maintain lumbar spine bone health in populations at risk.
The aim of this thesis was to develop a new predictive computational modelling framework for the study of bone structural adaptation to healthy and pathological conditions in the lumbar spine. The approach is based on the combination of a musculoskeletal model of the lumbar spine and lower limbs with structural finite element models of the lumbar vertebrae. These models are built with bone and muscle geometries derived from healthy individuals. Based on daily living activities, musculoskeletal simulations provide physiological loading conditions to the finite element models. Cortical and trabecular bone are modelled with shell and truss elements whose thicknesses and radii are adapted to withstand the physiological mechanical environment using a strain driven optimisation algorithm.
This modelling framework allows to generate healthy bone architecture when a loading envelope representative of a healthy lifestyle is applied to the vertebrae, and identify influential activities. Prediction of bone remodelling under altered loading scenarios characteristic of lumbar pathologies can also be achieved.
The modelling approach developed in this thesis is a powerful tool for the investigation of bone remodelling in the lumbar spine. Preliminary results indicate that locomotion activities are insufficient to maintain lumbar spine bone health. Specific recommendations to limit the effect of physical deconditioning related to muscle weakening back pain are suggested. The approach is also promising for the investigation of other lumbar pathologies such as age related osteoporosis and scoliosis.