Negative pressure wound therapy: multi-scale computational study

Since its commercial introduction, Negative Pressure Wound Therapy (NPWT) has become a widely accepted option for managing and treating non-healing wounds. Despite the success of NPWT, its mechanisms of action remain unclear. This has several explanations such as, NPWT is a generic multi-modal technology that can deliver a broad range of treatment goals depending on the patient being treated and these goals can be achieved by altering a range of variables which all add to the complexity of studying the therapy. In addition, variations in treatment regimens, depending on wound type, size and location, make it very difficult to standardise across health care settings and to optimise wound healing outcomes. Moreover, recent findings collectively imply that understanding of the biomechanical environment, such as mechanical stress and oxygen tension on wound bed, is fundamental for further mechanistic investigations of wound healing when NPWT applied. Although current scientific findings from clinical studies can show a facet of its aetiology, there is lack of enabling tools bridging these outcomes with the clinical management strategies of wounds. Hence, our understanding on the biomechanical aspects of the wound tissue during NPWT is rather limited.

Computational modelling has demonstrated its advantages for investigating biomechanical effectiveness of medical devices where clinically relevant complexities can be introduced in each increment and the range of variables can be alternated simultaneously. Hence, the main aim of this thesis is the application of advanced computational modelling as enabling tools to gain insights of the mechanical effects of NPWT on tissues with particular focus on deformation and tissue oxygenation in wound tissue. A finite element (FE) modelling procedure was introduced to predict the mechanical behaviour of different types of wound and its reliability was validated against experimental measurements. The experimental model showed a good comparison with FE model predictions. Then, multi-cohort of parameters, including wound characteristics, its location and surrounding tissue structures, were evaluated.

The wound characteristics study showed the effectiveness of NPWT on wound deformation. FE models were further used in this capacity to evaluate the effects of surrounding tissue structure based on wound location and its characteristics. A full-scale anatomical model was necessary to show full effect of NPWT response phenomena exhibited. The main reason is that in simplified model the magnitudes of displacements and stresses were underestimated when the total surrounding tissue volume relative to the wound volume was not fully implemented. Following that, a multi-scale model to predict the tissue oxygenation changes around the wound cavity subject to NPWT application was developed with varying the wound surface area. Finally, the effect of complex anatomical geometry on tissue oxygenation changes was modelled and the clinical applicability of the established framework for tissue oxygenation changes was demonstrated.

FE models based on multi-parametric studies were compared with currently available literature data to demonstrate its predictive capabilities. The results of this study made valuable contributions to our existing understanding of the mechanics of wound tissue during the NPWT process. Outcomes from the current study allowing stratified NPWT in terms of quantify the level of tissue deformations at specific location in the wounded area with varied pressure levels. With further verification using clinical data as inputs, the developed computational models would aid the stratification of personalised wound management using NPWT for complex wounds.