Identification of capillary blood pressure levels at which capillary collapse is likely in a tissue subjected to large compressive and shear deformations

Abstract

Pressure ulcers (PU) are localised damage to skin and underlying tissues, caused by sustained tissue deformations and ischaemia. PU typically appear in insensitive or immobile patients, e.g. those with spinal cord injury (SCI) or geriatric patients. As these patients often experience fluctuations in blood pressure, and are also exposed to high-shear loads in their weight-bearing soft tissues during wheelchair sitting or bed rest, we used an inverse finite element method to determine the effects of capillary blood pressure (CBP) and shear deformations on occurrence of mechanical collapse in capillaries. We studied collapse in straight, U-shaped and bifurcated capillaries. All model configurations were consistent in demonstrating that the level of CBP has a considerable influence on the likelihood of capillary collapse in the physiological CBP range, particularly if shear is present. Our modelling therefore suggests that low CBP is a ‘suspect’ risk factor for PU in SCI and geriatric patients.

Keywords:

• pressure ulcer
• deep tissue injury
• finite element model
• spinal cord injury
• elderly
• risk factor

Notes

^1. Submitted to a Special Issue of Computer Methods in Biomechanics and Biomedical Engineering on Identification of Material Parameters using Inverse Methods (Guest Editors: Drs. Stéphane Avril and Sam Evans)

^2. In preliminary analyses, we tested the effects of using denser meshes for the capillary wall. Though denser meshes provided better resolution of the strain/stress distributions across the deformed wall, they did not affect the collapse behaviour of capillaries, in terms of the eccentricity parameter (Equation (4)), and hence, for cost-effectiveness, it was decided to use a single layer of elements for meshing the capillary walls (Figure 2(c)). Likewise, using elements of higher order for the capillary wall (the element types that were used are specified in Table 1) did not affect the behaviour of the eccentricity parameter.