In order to investigate the biomechanical efficacy of sacral prophylactic dressings in protecting soft tissues after absorbing fluids, the authors developed 16 FE model variants representing the buttocks with either the anisotropic multilayer dressing or with a hypothetical dressing (see Table 1). The concept of the hypothetical dressing is based on commercially available dressings that were tested experimentally. However, to create comparable models of the MBS dressing and a hypothetical one, which needed to be identical in shape to the MBS but have different mechanical behavior (and hence, was not a “real” product), the hypothetical elastic moduli obtained in tests8,11 of some commercial products on the same geometric model that had been devised to represent the MBS were used. Specifically, the performance of the anisotropic multilayer dressing (which, according to manufacturer data, preserves its anisotropy feature when wet11) was compared to a hypothetical dressing that loses its anisotropy as fluid contents build up. The amounts of fluids are based on transepidermal water loss (TEWL) values taken from Kottner et al.12 These 3 fluid values include TEWL for lower back and buttocks of approximately 10 (g/[m2/h]), which corresponds to 0.025 (mL/[cm2/day]) in the dressing, and the higher levels (0.075, 0.15 [mL/(cm2/day)]) were taken in order to investigate a more substantial accumulation of fluids — for example, when excess perspiration or incontinence is present. Both types of dressings (anisotropic multilayer versus hypothetical) were tested for wetness levels when pure compressive bodyweight loading was applied and also for a combined compression and shear loading mode, consistent with the current authors’ previously published work.10,11 Effective and maximal shear stresses developed in the soft tissues at the sacral region in supine weight-bearing were systematically compared using a standard hospital mattress for each examined case (see Table 1).
Geometry. A 3-dimensional (3D) anatomical model of the buttocks recently developed by the authors’ group for methodological, comparative sacral dressing studies10,11 was used in this work. Briefly, 76 T1-weighted axial magnetic resonance imaging (MRI) slices of the weight-bearing buttocks of a 28-year-old healthy woman were imported to the ScanIP module of the Simpleware software package (Synopsis Inc, Mountain View, CA) for segmentation of the pelvic bones and soft tissues.13 Details regarding the MRI machine, scan protocol, and medical ethical approval are available elsewhere.10 The authors focused on a volume of interest (VOI) of 27.8 cm x 17.4 cm x 5.6 cm, incorporating the sacral bone and surrounding soft tissues. This allowed researchers to optimize computer power and make the numerical calculations effective where the tissue distortion phenomena relevant to sacral pressure ulcers occurred (see Figure 1a).
As in the authors’ previously published work,8-11 the anisotropic multilayer dressing included 3 physical material layers in the modeling: polyurethane foam (PUR), a nonwoven (NW) layer, and the airlaid (AL) layer. Consistent with previous studies,8-11 the authors considered the innermost Safetac layer as a tied interface between the soft tissue component and the PUR foam layer, and the outermost “backing film” layer was represented as frictional sliding between the AL layer and the mattress (see Figure 1b).8-11 However, the shape of the anisotropic multilayer dressing was not adopted from the authors’ previous work; rather, it was recreated using the ScanIP module of Simpleware to comply with the newest anisotropic multilayer dressing design launched in 2017. To complete the generation of the model geometry, a flat foam mattress was added under the buttocks in the ScanIP module of Simpleware.
Numerical methods. Meshing of the tissues, multilayer dressings, and mattress model components was performed using the ScanIP module of Simpleware.13 Four (4)-node linear tetrahedral elements were used in all model components. In order to obtain optimal accuracy but minimize complexity of the numerical solution and the associated computational power, mesh refinements were applied locally at the skin-dressing and mattress-dressing interfaces.
The FE simulations were set up using the PreView module of FEBio (Ver.1.19; University of Utah, Salt Lake City, UT), analyzed using the Pardiso linear solver of FEBio (http://mrl.sci.utah.edu/software/febio) (Ver. 2.5.0), and post-processed using PostView of FEBio (Ver. 1.10.2).14 Converging time steps were chosen for numerical data collection so that the resulting reaction force was within a 2% difference from the target reaction force (description to follow). The time for solving each simulation case, using a 64-bit Windows 8-based workstation with 2×Intel Xeon E5-2620 2.00 GHz CPU and 64 GB of RAM, ranged between 7 and 12 hours.
Mechanical properties of the dressing and tissues. Constitutive laws and mechanical properties of the tissue components and the mattress were adopted from the literature. Specifically, the sacrum was assumed to be a linear-elastic isotropic material with elastic modulus of 7 GPa and a Poisson’s ratio of 0.3.15-17 The soft tissues were assumed to be nearly incompressible (Poisson’s ratio of 0.49) and nonlinear isotropic, with their large deformation behavior described by an uncoupled Neo-Hookean model with the following SED function W:
where Gins (the instantaneous shear modulus) is 2 kPa,17 λi (I = 1,2,3) are the principal stretch ratios, K (the bulk modulus) is 1 MPa , and J = det(F) where F is the deformation gradient tensor. Specifically, material constants reported by Oomens et al18 were used to calculate an effective soft tissue Gins comprised of 60% skin and 40% fat, as in the authors’ previous modeling work of the buttocks.10
The anisotropic multilayer dressing has significantly different stiffness properties in the vertical versus the horizontal directions (anisotropy), while the hypothetical dressing has less distinct directional stiffness properties. The elastic moduli of the multilayer dressing at the X and Y directions (ie, the spinal and lateral directions, respectively) were measured in the authors’ laboratory and in the anisotropic dressing’s testing facilities with the authors maintaining oversight of the experimental protocol and data (see Table 2). The ratio between the elastic moduli at the Y direction over the X direction was found to be 6.6 for the anisotropic multilayer and approximately 1.8 for other commercially available dressings in the dry condition. The elastic moduli and the ratio of moduli at the Y direction over the X direction also were measured for the 3 levels of wetness in moist dressings: 0.025, 0.075, and 0.15 (mL/[cm2]) (see Table 2). For the hypothetical dressing in its wet conditions, a modulus ratio of unity was assigned based on measurements of commercial prophylactic dressings loaded with the above wetness levels and then mechanically tested in tensile loading at the X and Y directions. In other words, the hypothetical dressing was considered to become linear-elastic isotropic when wet (at any of the above 3 wetness levels), which is a potential material softening response known to exist in some wet porous materials (such as wet paper). A Poisson’s ratio of 0.258 was chosen for all dressings based on published experimental data.19 The mattress was considered isotropic linear-elastic, with an elastic modulus of 50 kPa and a Poisson’s ratio of 0.3, based on literature.8,9,20
Body loads applied to the buttocks and boundary conditions. Downward displacements of 5.5 mm to 6.48 mm were applied on the top surface of the model in order to simulate the descent of the weight-bearing sacrum during supine bedrest or a 45˚ Fowler’s position, with the anisotropic multilayer or hypothetical dressings in the dry and 3 wet dressing conditions. A total reaction force of 40 Newtons was obtained in all simulations that represented approximately 7% of the total bodyweight of the subject; this was transmitted focally at the sacral region. Therefore, the comparison between all simulation cases was conducted under the same (7% bodyweight) conditions for consistency of outcome measures across the different model variants. In the combined compression and shear loading scenario (representing sliding in bed due to gravity), a horizontal displacement of the same magnitude was added in the Y direction. The bottom surface of the mattress was fixed for all motions, and tied interfaces were defined at the bone-soft tissue boundaries as well as between the soft tissues and the dressing. Frictional sliding was defined between the dressing and mattress, with the coefficient of friction set to 0.35.8,19
Biomechanical outcome measures. Effective and maximal shear stresses within the VOI were compared across all simulation cases. Volumetric exposures of soft tissues below the sacrum (in the VOI) to elevated effective stresses also were compared and examined using stress exposure histogram (SEH) charts. As a final step after evaluating the volumetric exposure of soft tissues to stresses and plotting the SEHs, the protective endurance (PE) of the anisotropic multilayer dressing versus the hypothetical dressing (in percents) was calculated as follows: 1) the relative difference in the area under the SEH (A) for dry (d) and wet (w) cases of each dressing was calculated relative to the case in which no dressing was used:
2) the protective endurance (PE) of a given dressing was defined as:
Hence, the PE is an objective, standardized, and quantitative indicator of the preservation of biomechanical protection that a certain dressing provides to the soft tissues while being wet with respect to its (ideal) protective efficacy when it is dry. Because the present modeling is deterministic (and not probabilistic), each combination of dressing conditions (type of dressing, level of moisture, and loading mode as specified in Table 1) was simulated once. A detailed description of the chosen FE computer modeling and simulation approach is provided in the authors’ previous work.9
Data management. The FE simulation data were directly imported to and post-processed using PostView (US National Institutes of Health, Washington, DC), a post-processor software designed to visualize and analyze results from a FEBio analysis.14 The displacement applied on the top surface of the model was increased incrementally for numerical convergence purposes, so the resulting reaction forces between the buttocks and the support surface were within <1.8% difference of the aforementioned target reaction force. The effective and maximal shear stresses data that developed in the soft tissues within a cubical VOI in a size of 9 cm x 9 cm x 2.5 cm were pooled for each dressing type in the dry and 3 wet conditions, under pure compression due to bodyweight and separately under combined compression and shear loading. Further, the volumetric exposure of soft tissues below the sacrum in the aforementioned VOI was compared to elevated effective stresses and a SEH was plotted per each dressing type and dryness/wetness condition. A PE index then was calculated for each dressing type according to the algorithm described. Because all simulation data are deterministic, no variability exists in the modeling outcomes per each specific case of input dataset; hence, statistical analyses were not applicable in this study (please see a detailed explanation in the authors’ previous work9).