Whole-body vibration in the defence maritime environment: analysis and simulation of vertebral cancellous bone

Abstract

Whole-body vibration has been shown to increase the risk of low back pain, especially during extreme exposures such as on marine craft which can reach peak loads of 20g during “slamming”. Wedge fractures and trabecular damage of vertebrae have been noted at these high acceleration events. There is a need of a quantitative link between whole-body vibration and spinal damage, with possible tools for prediction. There is currently little known about the role trabecular damage plays under damage from Whole-body vibration, as well as a lack of robust and repeatable trabecular fatigue FE techniques. A fatigue model for trabecular vertebral bone was developed in four steps: Fatigue testing of porcine vertebral cores; validation of a novel element material method; fatigue simulation of a porcine core to select the failure method; prediction using the validated material mapping model with the best failure method on human vertebral cubes. The fatigue tests were carried out on porcine trabecular cores loaded at 2Hz with varying normalised stress values until fatigue failure. Signal analysis was used to examine the vibrational statistics as per ISO 2631-1. This was done to both compare the statistical approaches used in measuring vibration and quantifying a link with in-vivo damage. Vibration Dose Value exposure was found to be the best predictor of failure within these tests. Its 4th order averaging accounted for minute differences in acceleration that RMS could not, even at the low frequency tested. Fatigue of porcine bone has not been extensively examined in the literature and experimental results indicate that there are significant differences in its fatigue behaviour compared to human and bovine bone. Currently there is a need to calibrate the material models used in finite element simulations to achieve parity with experimental testing. This thesis validates a novel greyscale mapping technique which does not require calibration. This was done on human trabecular cores taken at different orientations, with both experimental and finite element simulations. With these tissue material ii properties the simulations showed good agreement in terms of mechanical response in all three directions. Fatigue was calculated using finite element analysis on a porcine core which was validated against experimental results. Three methods were tested for this: A stress based model which bases the element failure criteria in respect to the cycle number; a model which calculates failure by the specific element stress and a strain model which fails elements based upon total element strain. This was conducted using a direct iterative approach using linear isotropic material properties with failure calculated after each cycle, keeping down computational costs. All methods took roughly the same amount of time for a load step. Failure was predicted much sooner in comparison to the experimental with the specific element stress and strain models. The method which varies failure based on cycle count was selected as it was the most accurate. As porcine fatigue testing has not been examined the results were difficult to compare and differed from previous experiments on human and bovine tissues. Using the validated material model and the best performing fatigue method this was then applied to Human trabecular specimens to estimate the fatigue life. The cubes were then loaded in the main physiological direction from in-vivo loading. This predicted most of the expected mechanical behaviour during fatigue including a linear relationship between damage fraction and modulus reduction. It also highlights the importance of angular orientation in regards to trabecular fatigue life. Although it tended to underestimate the fatigue life of bone, it was in good agreement with the literature over the normalised stress range tested. The differences in simulated fatigue behaviour and the literature, seen previously with porcine tissue, were not apparent here. With further study and validation this model has the potential to improve the understanding of trabecular fatigue failure using vibration exposure as the model stimulus