The biomechanical protective effects of a treatment dressing on the soft tissues surrounding a non‐offloaded sacral pressure ulcer

Abstract

Patients who are immobile endure prolonged bodyweight‐related compressive, tensional and shear loads at their body‐support contact areas that over time may lead to the onset of pressure ulcers (PUs). Approximately, one‐third of the common sacral PUs are severe and classified as category 3 or 4. If a PU has occurred, off‐loading is the basic, commonly accepted clinical intervention; however, in many situations, complete off‐loading of sacral PUs is not possible. Minimising the exposure of wounds and their surroundings to elevated mechanical loads is crucial for healing. Accordingly, in the present study, we aimed to investigate the biomechanical effects of the structural and mechanical properties of different treatment dressings on stresses in soft tissues surrounding a non‐offloaded sacral PU in a supine patient. Using a novel three‐dimensional anatomically realistic finite element modelling framework, we have compared performances of three dressing designs: (a) The Mepilex Border Sacrum (MBS) multilayer anisotropic silicone foam dressing (Mölnlycke Health Care), (b) an isotropic stiff dressing, and (c) an isotropic flexible dressing. Using our newly developed protective efficacy index (PEI) and aggravation index (AI) for assessing prophylactic and treatment dressings, we identified the anisotropic stiffness feature of the MBS dressing as a key design element.

1. INTRODUCTION

A pressure ulcer (PU), also called a pressure injury, is localised damage to skin and underlying soft tissues, usually over bony prominences, as a result of sustained mechanical loads applied to the tissues.1 Patients who are stationary, such as those who are paralysed, unconscious or under anaesthesia during a surgical procedure, endure prolonged bodyweight‐related compressive, tensional and shear loads at the body‐support contact areas, which over time, may lead to the onset of PUs.2, 3, 4

Clinical surveys identify the sacral area as the most common site for PUs associated with prolonged supine position (bedrest).5, 6 Compressive loads applied by the heavy pelvis and shear loads caused by static or dynamic frictional forces (such as when a patient slides downwards in the bed because of gravity or during repositioning) subject the soft tissues around the sacrum to sustained deformations, which may lead to a PU. The high curvature of the nearly rigid sacral bone and the stiffness gradients of the layered soft tissue structures (eg, muscle vs. adipose) intensifies the internal shear stresses and may spread the tissue distortions and associated injury to the depth of the soft tissues. Despite the vast government, academic, industry, and health care service efforts and resources are being now invested in PU prevention, and although clinical studies testing different interventions are now able to report on decline in PUs incidence,7 the prevalence of these wounds is still high, and sometimes they can be unavoidable.8 Approximately, one‐third of the reported PUs are severe and classified as category 3 or 4.6, 9

The fundamental intervention wherever a PU has been diagnosed is off‐loading and pressure redistribution through frequent repositioning.1 These are physical measures taken to remove the mechanical loads, which distort the tissues at the existing or evolving PU, and (because bodyweight forces will need to transfer regardless) relocate these loads so that they act somewhere else in the body. Off‐loading and pressure redistribution are fundamental interventions that help restrain additional tissue damage and promote healing. Accordingly, they are widely used by clinicians who commonly use the phrase—“put about anything on the PU but the patient.” However, there are some patient conditions in which complete off‐loading is not feasible, or there may be lack of patient compliance to positioning. For example, patients requiring cardiopulmonary support through extracorporeal membrane oxygenation (ECMO) are difficult to reposition as the postural changes may dislodge the device. Likewise, patients with spasticity, patients with a severe gastroesophageal reflux disease, patients with dementia or delirium and those with psychiatric disorders are challenging to reposition. In such clinical scenarios, if a sacral PU exists, patients may still spend some or most of their time in a supine position. Accordingly, in these cases, the wound bed and non‐injured tissues surrounding the wound will be subjected to sustained bodyweight forces and associated tissue deformations.2, 3, 4

Wound‐product dressings are used in the management of PUs, for maintenance of a moist wound environment, reduction of the risk of infections, and promotion of healing.10 When a sacral dressing is used on a non‐offloaded sacral PU, as in the aforementioned examples, it interferes with the loading state in the soft tissues at and around the wound. An adequate treatment dressing with a protective biomechanical effect should be able to absorb compressive and shear deformation energy from the wound and surrounding tissues, and hence allow the loaded wound bed to heal despite the transferred loads. At a minimum, a good treatment dressing applied to a non‐offloaded PU should minimise the risk for exacerbation of the existing PU.

The purpose of this study was to investigate the biomechanical effects of the structural and mechanical properties of different treatment dressings on the mechanical stresses developed in soft tissues surrounding a non‐offloaded sacral PU in a supine patient. We specifically focused on the biomechanical performances of the anisotropic multilayered Mepilex Border Sacrum (MBS) dressing as a treatment dressing (ie, when a PU already exists, and is subjected to bodyweight forces), after previously finding the MBS dressing effective for pure prophylaxis of sacral PUs.11, 12, 13 In our published work, we reported that anisotropy, that is, greater stiffness of the multilayer dressing in the direction of the spine vs. a more compliant behaviour laterally (when stretched along the buttocks cheeks) is a critically important design feature in a prophylactic dressing. Nevertheless, the potential contribution of anisotropy in a treatment dressing has not been addressed in the literature thus far.

We compared the biomechanical efficacy of the anisotropic MBS dressing—used as a treatment dressing—against two hypothetical isotropic treatment dressings, one that is substantially stiffer and another that is considerably more flexible than the MBS dressing. Using a novel three‐dimensional (3D) anatomically realistic finite element (FE) modelling framework that incorporates a category‐4 sacral PU, we have analysed the biomechanical efficacy of these three dressing designs in protecting non‐injured superficial as well as deep tissues around a sacral wound bed and away from it.

2. METHODS

To investigate the biomechanical protective effects of a sacral dressing on the loading state in soft tissues of the buttocks after a PU has already formed, we developed three model variants of the buttocks with an existing PU, covered by one of the following dressings: (a) A multilayer anisotropic silicone foam dressing (Mepilex Border Sacrum, MBS, by Mölnlycke Health Care, Gothenburg, Sweden), (b) a hypothetical isotropic stiff dressing (HIS), and (c) a hypothetical isotropic flexible dressing (HIF). An additional forth model variant where the buttocks are bare and not covered by any dressing was also developed, as a reference case for systematic comparisons. In all these four model variants, the buttocks have been laid on a standard, flat medical foam mattress. We specifically aimed to compare effective tissue stress magnitudes and distributions between the four model variants when the buttocks are subjected to combined compressive (body weight) and shear loads (because of gravity's pull causing a tendency to slide in bed). The above comparisons specifically enable to determine whether a dressing with a directional stiffness preference, namely, the MBS dressing, has an enhanced biomechanical protective effect at the wound perimeter or at adjacent non‐wounded tissues, when a sacral PU already exists. We also compared the biomechanical effects of the three dressing designs, MBS, HIS, and HIF, on the soft tissues near the perimeter of the dressing to identify any potential skin irritation because of elevated mechanical stresses or aggravation of the mechanical loading state in tissues caused by the dressing there.

2.1. Geometry

The 3D anatomical model of the buttocks that was used in this work is based on a model that was recently developed by our group for methodological, comparative sacral dressing studies. Briefly, 76 T₁‐weighted axial magnetic resonance imaging (MRI) slices of the weight‐bearing buttocks of a 28‐year‐old healthy female subject were imported to the ScanIP module of the Simpleware software package (Synopsis Co., Mountain View, California) for segmentation of the pelvic bones and soft tissues.14 Details regarding the MRI machine, scan protocol, and medical ethical approval are available elsewhere.11 An open, category 4 PU was incorporated in the 3D buttocks model as a penetrating, truncated conical hole in the soft tissues, which spans from the skin level to a depth of 1.75 cm, reaching and exposing the sacral bone as per the clinical definition of a category‐4 PU (Figure 1A). The diameters of this simulated PU at the skin‐side and bone‐side were 5 and 2.5 cm, respectively (Figure 1B,C).

Figure 1.

Finite element computational modelling of the buttocks with the Mepilex Border Sacrum (MBS) dressing (Mölnlycke Health Care, Gothenburg, Sweden): A, during weight‐bearing (left); geometric model of the main volume of interest (VOI) under the sacrum, including the virtually added category‐4 pressure ulcer (middle); the main VOI covered by the MBS dressing (right); B, the applied loading and boundary conditions in a sagittal cross‐section through the sacrum; and C, the tetrahedral mesh of the sacral region, the MBS dressing and the mattress, in an axial cross‐section through the sacrum

The MBS dressing covering the above PU was modelled with three physical material layers: the polyurethane foam (PUR), the non‐woven (NW) and the airlaid (AL) layers, consistent with our previous published work.11, 12, 13 We further considered the innermost Safetac layer as a tied interface between the soft tissue component and the PUR foam layer; the outermost “backing film” layer was represented as frictional sliding between the AL layer and the foam mattress, again as in our previous work.11, 12, 13 The geometries of the HIS dressing and HIF dressing were identical to the MBS dressing geometry to facilitate consistent comparisons across simulation cases.

2.2. Numerical methods

Meshing of the tissues, dressing, and mattress components was performed using the ScanIP module of Simpleware.14 Four‐node linear tetrahedral elements were used in all model components. Mesh refinements were applied locally in both healthy skin‐dressing interfaces and the wound‐dressing interfaces, as well as in the mattress‐dressing interface to reach convergence of the numerical solutions, and to obtain optimal accuracy of solutions. There were a total of more than 150 k elements in the model variant with the dressing: 17 611 bone elements, 30 041 soft tissue elements, 69 779 dressing elements, and 35 852 mattress elements (Figure 1C). Converging time steps were chosen for numerical data collection, so that the resulting reaction force deviated from the target reaction force by no more than 0.15%. The FE simulations were all set up using the PreView module of FEBio (Version 2.0; University of Utah, Salt Lake City, Utah), analysed using the Pardiso linear solver of FEBio (http://mrl.sci.utah.edu/software/febio) (Version 2.7) and post‐processed using PostView of FEBio (Version 2.2).15 The average time for a solution of the simulations was 12 hours using a 64‐bit Windows 7‐based workstation with Intel Core i7‐5820 K 3.30 GHz CPU and 32 GB of RAM.

2.3. Mechanical properties of the dressing and tissues

The layers of the MBS dressing were all considered anisotropic elastic materials, with a greater material stiffness in the longitudinal direction (of the spine) than in the lateral direction (of the buttocks cheeks). The compressive elastic moduli assigned to the dressing components (Table 1) were based on experimental measurements previously conducted in our laboratory.11 Tensile elastic moduli of the MBS dressing (ie, elastic moduli in the X‐ and Y‐directions, associated with stretching of the dressing as bodyweight loads apply) as well as the Poisson's ratio were extracted from measurements carried out in the testing laboratories of the manufacturer, overviewing the experimental protocol and data (Table 1). Both the HIS and HIF dressings were considered as made of isotropic elastic materials. The elastic modulus of the HIS dressing was taken as equal to the elastic modulus along the stiffer direction of the (anisotropic) MBS dressing, that is, the direction of the spine (the Y‐direction in the currently used orthogonal system). Consistently, the elastic modulus of the HIF dressing was taken as equal to the elastic modulus along the flexible direction of the MBS dressing, that is, the (X‐) direction of the buttocks cheeks (Table 1). The Poisson's ratio was adopted from the literature as 0.258 for both the HIS and HIF dressings16 (Table 1).

Table 1.

Mechanical properties of the model components, including tissues, dressing and the mattress

Modal component		Shear modulus (kPa)	Elastic modulus (kPa)			Poisson's ratio
		G	E x	E y	E z	ν
Soft tissues		2	—	—	—	0.49
Sacrum		—	7 × 106	7 × 106	7 × 106	0.3
Mattress		—	50	50	50	0.3
HIS dressing (by layer)	Airlaid	—	1175	1175	1175	0.258
	Non‐woven		5760	5760	5760
	Polyurethane foam		921	921	921
HIF dressing (by layer)	Airlaid	—	177	177	177	0.258
	Non‐woven		868	868	868
	polyurethane foam		138	138	138
						ν12	ν23
MBS anisotropic dressing (by layer)	Airlaid	—	1175	177	30.6	0.58	0.58
	Non‐woven	—	5760	868	150	1.26	0.258
	polyurethane foam	—	921	138	24	0.4	0.258

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 an elastic modulus of 7 GPa and Poisson's ratio of 0.3.17, 18, 19 The soft tissues were assumed to be nearly incompressible (Poisson's ratio of 0.49), nonlinear isotropic, with their large deformation behaviour described by an uncoupled Neo‐Hookean model with the following strain energy density (SED) function W:

$$W=\frac{G_{\mathrm{ins}}}{2}\left({{\lambda }_1}^2+{{\lambda }_2}^2+{{\lambda }_3}^2-3\right)+\frac{1}{2}K{\left(\text{In}j\right)}^2,$$ (1)

where G _ins (the instantaneous shear modulus) is 2 kPa,19 λ _i (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. We used the material constants reported by Oomens et al20 for Equation (1) and calculated an effective soft tissue G _ins, constituting of 60% skin and 40% fat, as in our previous modelling work of the buttocks.11, 12, 13

Finally, the mattress was considered isotropic linear‐elastic, with an elastic modulus of 50 kPa and Poisson's ratio of 0.3, based on the literature.21, 22, 23

2.4. Boundary conditions

We sought to simulate the combination of the descent of the weight‐bearing sacrum during a supine bedrest and the sliding motion typically caused as gravity pulls the patient's body downwards when placed in a Fowler's position or when the head of the bed is elevated. For this purpose, equal downward and forward displacements of 5.25 mm were applied on the top surface of the model (Figure 1B). The bottom of the mattress was fixed for all motions so that only the buttocks were able to move. For the three model variants with the dressing, frictional sliding contact was defined between the dressing and the mattress, with a coefficient of friction of 0.35.16 During the descent of the buttocks towards the mattress because of bodyweight, the inner surface of the PU and the dressing were pressed together, and as they contacted, the interface between these two surfaces was defined as tied. Tied interfaces were also defined at the bone‐soft tissue boundaries, as well as between the non‐wounded soft tissues and the dressing. For the model variant without the dressing, a frictional sliding interface was defined between the soft tissues and the mattress, with a coefficient of friction of 0.4.16, 21 A total reaction force of 35 N has been obtained in all model variants, which represents approximately 6% of the total bodyweight of the subject that is transmitted focally at the sacral region.

2.5. Biomechanical outcome measures

The outcome measures described below were obtained for the aforementioned displacement boundary conditions, which were identical for all four model variants. We focused on a specific, cubical volume of interest (VOI) within the buttocks, with a size of 27.8 × 17.4 × 5.6 cm³, which contained the sacral bone and surrounding soft tissues, to optimise the numerical analyses. Three sub‐VOIs were defined within the aforementioned main VOI to systematically and regionally compare the effective and maximal shear stresses developed in the soft tissues, in each of the cases where the buttocks and PU were covered by either one of the dressings (MBS, HIS or HIF) as well as in the case that the buttocks and PU were bare. The first sub‐VOI included the superficial elements of the soft tissues surrounding the perimeter of the wound, representing skin around the wound bed, within a distance of 1 cm from the wound edge. The second sub‐VOI included the superficial elements of the soft tissues near the perimeter of the dressing, representing skin under the border of the dressing, within a rectangular area in the size of 2 × 4.8 cm². The third sub‐VOI was a cubical volume that contained the subdermal and deep tissues surrounding the tip of the coccyx bone, with a size of 6.5 × 5.7 × 2.2 cm³. The volumetric exposures of the soft tissues to elevated effective stresses were analysed and then compared between the four model variant cases, for the three VOIs defined above, using stress exposure histogram (SEH) charts, as per the methodology developed in our previous published work.13 As a final step after evaluating the volumetric exposure of the soft tissues to effective stresses and plotting the SEHs for each of the three sub‐VOIs, we determined the protective efficacy index (PEI) for the first and third sub‐VOIs by calculating the relative %‐differences in the area under the SEH for the MBS, HIS, and HIF cases relative to the case where no dressing was used:

$$\mathrm{PEI}\left[\%\right]=100\times \frac{A_{\mathrm{No}\text{dressing}}-A_{\text{dressing}}}{A_{\mathrm{No}\text{dressing}}}$$ (2)

Another index, named the aggravation index (AI), was defined for the second sub‐VOI (near the border of the dressing). For this purpose, we first calculated the absolute value of the relative difference (AVRD) between the area under the SEH of each of the three dressings (MBS, HIS, and HIF) and the no‐dressing curve (in percent):

$$\text{AVRD}\left[\%\right]=100\times \frac{A_{\mathrm{No}\text{dressing}}-A_{\text{dressing}}}{A_{\mathrm{No}\text{dressing}}}$$ (3)

We then determined the AI, which denotes the added biomechanical aggravation to skin on top of the no‐dressing case. Thus, to calculate the AI, we simply added 100% to the AVRD, to represent the baseline (no‐dressing case) onto which the skin aggravation is added.

3. RESULTS

For all four model variants, stress concentrations appeared in skin surrounding the perimeter of the wound (1st sub‐VOI) and in the soft tissues in the vicinity of the tip of the coccyx bone (3rd sub‐VOI). In the three model variants where the buttocks were covered by dressings, elevated stresses also appeared in skin under the border of the dressing.

Effective stress distributions in soft tissues in the three sub‐VOIs, when the body is subjected to combined compression and shear loads, with the MBS dressing, HIS dressing, and HIF dressing, and bare buttocks without a dressing are shown in Figure 2 (surface stresses) and Figure 3 (deep tissue stresses). In the first sub‐VOI, which included the skin surrounding the wound bed, a considerable reduction in magnitudes of effective stresses was demonstrated when dressings were in use, with respect to the bare buttocks case (Figure 2). The greater stress reductions were demonstrated for the MBS and HIS dressings. Specifically, the average effective stress in the first VOI was reduced by 45% for both the MBS and HIS dressings but only by 33% for the HIF dressing (relative to the no dressing case). In the second sub‐VOI, which included skin under the border of the dressing, the average effective stress in the second sub‐VOI increased by 18% for the HIS dressing and by 1% for the MBS dressing, but was 1% less for the HIF dressing, with respect to the no‐dressing case (Figure 2). In deep tissue within the third sub‐VOI, near the coccyx bone, the greatest stress reduction was demonstrated for the MBS dressing that reduced the average effective stress by 18% relative to the bare buttocks case, compared with reductions of 8.5% and 9% for the HIS and HIF dressings, respectively. Similar trends of stress reductions were demonstrated for the maximal shear stresses for all the sub‐VOIs (skin shear stresses and deep tissue shear stresses are shown in Figures 4 and 5, respectively).

Figure 2.

Effective stresses in the soft tissues of the buttocks on the skin surface: A, without a dressing; B, with the Mepilex Border Sacrum (MBS) dressing (Mölnlycke Health Care, Gothenburg, Sweden); C, with a hypothetical isotropic stiff (HIS) dressing; and D, with a hypothetical isotropic flexible (HIF) dressing

Figure 3.

Effective stresses in the soft tissues of the buttocks in cross‐sections through the sacrum, and zoom‐in to the area near the tip of the coccyx: A, without a dressing; B, with the Mepilex Border Sacrum (MBS) dressing (Mölnlycke Health Care, Gothenburg, Sweden); C, with a hypothetical isotropic stiff (HIS) dressing; and D, with a hypothetical isotropic flexible (HIF) dressing

Figure 4.

Maximal shear stresses in the soft tissues of the buttocks on the skin surface: A, without a dressing; B, with the Mepilex Border Sacrum (MBS) dressing (Mölnlycke Health Care, Gothenburg, Sweden); C, with a hypothetical isotropic stiff (HIS) dressing; and (D) with a hypothetical isotropic flexible (HIF) dressing

Figure 5.

Maximal shear stresses in the soft tissues of the buttocks in cross‐sections through the sacrum, and zoom‐in to the area near the tip of the coccyx: A, without a dressing; B, with the Mepilex Border Sacrum (MBS) dressing (Mölnlycke Health Care, Gothenburg, Sweden); C with a hypothetical isotropic stiff (HIS) dressing; and D with a hypothetical isotropic flexible (HIF) dressing

Comparisons of the cumulative volumetric exposures of soft tissues to effective stress in the three sub‐VOIs under combined compression and shear loading of the buttocks, with each of the three types of dressings, as well as without a dressing, are presented in Figure 6. In the first sub‐VOI, all three types of dressings were able to reduce the exposure of skin surrounding the wound bed to high effective stresses with respect to the no‐dressing case. However, the MBS and HIS dressing were more effective in reducing these skin stresses (Figure 6A). The PEI values for the first sub‐VOI consistently demonstrated the greatest biomechanical protective effect for the MBS and HIS dressings (Figure 7A). Specifically, the PEIs of MBS and HIS dressings were 64% and 65%, respectively, whereas the PEI of the HIF dressing was 47% (Figure 7A). In the third sub‐VOI, representing exposure of deep tissues to stress, the MBS was approximately two‐fold better than the HIS and HIF dressings in alleviating effective tissue stresses near the tip of the coccyx bone (Figure 6C). Specifically, the PEI of the MBS dressing was 34%, compared with just 14% and 15% for the HIS and HIF dressings, respectively (Figure 7A). The superior biomechanical protective effect of the MBS dressing vs. the two isotropic dressings HIS and HIF was even more prominent for the high‐effective stress domain as demonstrated in Figure 6C. For the second sub‐VOI, which represents potential irritation to skin at the border of the dressing because of exposure to elevated stresses, the data showed a remarkable difference between the tissue exposure curves. Specifically, the HIS dressing had an AI value of 220% compared with only 102% and 113% for the HIF and MBS dressings, respectively.

Figure 6.

Cumulative volumetric exposures of the soft tissues to effective stresses under combined compression and shear loading: A, On the skin surface in the first sub‐volume of interest (1st sub‐VOI) near the perimeter of the pressure ulcer; B, in the second sub‐VOI, on the skin surface near the border of the dressing; and (C) in the third sub‐VOI near the tip of the coccyx, with zoom‐in to high effective stress values

Figure 7.

A, Protective efficacy index of each of the studied dressings in the first and third sub‐volumes of interest (sub‐VOIs); B, the aggravation index of each of the dressings in the second sub‐VOI. Both the protective efficacy and aggravation index are calculated for each dressing with respect to the bare buttocks case

4. DISCUSSION

In the present study, we investigated the biomechanical protective effect of a sacral dressing on the loading state in the soft tissues of the buttocks after a PU has already formed, near and away from the wound. We specifically focused on determining the biomechanical protection provided to tissues adjacent to the PU by the anisotropy feature of the MBS dressing, which has been extensively studied by our group but in a purely prophylactic context so far.12, 13 Using a new FE modelling framework that incorporates a simulated existing sacral PU, we compared the performances of the MBS dressing that is characterised by strong anisotropy (ie, substantial directional preference of stiffness) against two hypothetical dressings: (a) a HIS and (b) an isotropic flexible dressing. We specifically compered the mechanical stresses developing in the soft tissues near the sacrum because of bodyweight loads when each of the three tested dressing designs has been used, as well as the skin stresses at the borders of the three dressings.

The simulation data have demonstrated three main regions of potential stress concentrations that are manifested when the buttocks are subjected to combine compressive and shear loads associated with bodyweight. These regions were the skin around the wound bed (first sub‐VOI), the borders of the dressings (second sub‐VOI) and deep tissues around the tip of the coccyx bone (third sub‐VOI) (Figures 6 and 7). We found that in the skin surrounding the wound bed (first sub‐VOI), the anisotropic MBS dressing and the HIS dressing performed better in reducing effective stress levels with respect to the HIF dressing. The perimeter of a non‐off‐loaded wound is exposed to substantial bodyweight‐associated compressive loads, and also to simultaneous shear as a result of static or dynamic frictional forces (especially if the head of the bed is elevated). As a result, tissues at the perimeter of the wound sustain large compressive, tensional and shear deformations, which may in turn result in exacerbation of the wound. The MBS dressing that is stiffer in the longitudinal direction (the direction of the spine along which most of the frictional forces apply) and the HIS dressing were able to reduce these deformations by preventing large movements of the wound edges, and as a result, the effective skin stress levels near the wound were reduced (Figure 7A).

The deep tissues surrounding the tip of the coccyx bone (third sub‐VOI) are also at a relatively high risk for developing elevated mechanical stress levels, which imperil tissue viability. This risk has biomechanical reasoning, namely, it is because of the curved bone shape and the stiffness gradients between the bone and soft tissues, and between non‐injured, structurally intact soft tissues vs. the actual wound, which is partially necrotic. The MBS dressing was approximately two‐fold better than both the HIS and HIF dressings in alleviating these deep tissue stresses (Figure 7A), and was especially superior in reducing peak stress levels with respect to the HIS and HIF cases, where peak stresses exceeded even those in the bare buttocks case (Figure 6C). Given that the MBS dressing is compared here to considerably softer (HIF) and considerably stiffer (HIS) dressings that are both isotropic and shaped identically, the supremacy of the MBS in alleviating tissue stresses can be associated with confidence with its anisotropic feature and adequate directional preference of stiffness. This finding is consistent with our previous work regarding the biomechanical efficacy of the MBS in pure prophylaxis of sacral PUs.12, 13

Despite that the present and previously published work showed that a multilayer dressing provides topical biomechanical protection to soft tissues by cushioning, shear absorbance and low‐friction in the outer surface of the dressing, trade‐off effects are likely.11, 12, 13, 21, 22 Specifically, as dressings typically have a different structural stiffness with respect to skin and underlying soft tissues, some extent of a stress concentration, associated with the tissue‐dressing stiffness gradient, is expected at the borders of the dressing (second sub‐VOI). When comparing the three dressing designs that were tested here, we found that the stiffer the dressing is, the greater are the stress concentrations at the borders, which could have been foreseen as a stiffer dressing implies a greater tissue‐dressing stiffness gradient at the borders. Hence, as expected, the HIS dressing was the worst for skin integrity and viability at its borders, and produced stress concentrations that were considerably above those for the MBS and HIF cases (Figure 7B). Interestingly, however, the MBS dressing that is anisotropic and stretchable at the directions of the buttocks cheeks caused only slightly more stress at the border than the HIF dressing (Figure 7B).

The concept of “stress shielding” is well known and widely used in biomechanics, mostly in the context of orthopaedic implants.24, 25 However, in its broader interpretation, stress shielding is the stress removed from a loaded tissue when an artificial structure that is substantially stiffer than the tissue comes in contact with that tissue. In such a case, the artificial structure and tissue share the applied mechanical load, and the portion of the load transferred to the tissue (out of the total load) depends on the difference in stiffnesses between the artificial structure and tissue. Now focusing on dressings (as the specific artificial structure), for the sake of example, consider a dressing made of steel (which is practically rigid compared with skin). In such a case, the difference between the dressing stiffness vs. the tissue stiffness is so large that skin protected by the centre of such a steel dressing would be negligibly deformed and insignificantly stressed. However, at the borders of such a steel dressing, where skin (and underlying tissues) is not fully confined and is able to partially distort, a different phenomenon will occur: Indentation of the dressing borders into the skin will cause severe tissue compression, tension, and shear, which will—if applied in reality—cause cell and tissue death within minutes. Hence, stress shielding in PU prevention is generally undesirable. Namely, shielding a limited soft tissue region at the centre of a too‐stiff dressing comes at the price of over‐distorting tissues away from the dressing centre, particularly around the borders of the dressing, which is why the HIS dressing examined here is theoretically inadequate. This leads to the understanding that to fully characterise the effects of a dressing, one needs to examine the state of tissue stresses both under the centre of the dressing and away from the centre, at or near its borders.

An opposite of a metallic dressing would be a very soft dressing, perhaps even as soft as skin. The use of such a flexible dressing will not involve stress shielding but rather, allow the skin and subdermal tissues to distort, expand and deform under the bodyweight forces in a nearly natural manner for a supine position (both under the centre of the dressing and at its borders). In such a case, the only protective quality of the flexible dressing would be its cushioning effect (as it adds some “compliant” thickness to that of the natural tissues). However, this is clearly insufficient for PU prevention or preventing an existing (loaded) wound from deteriorating further, given that the additional cushioning will be negligible with respect to the one already provided by the mattress. Therefore, like in the stiff dressing case, the HIF dressing studied here is also theoretically inadequate, despite that it is soft and would, therefore, not inflict stress concentrations at its boundaries as was caused by the HIS dressing. These hypothetical examples (HIS, HIF) establish that a balance is required between stress shielding, vs. flexibility and softness that minimise border‐indentation effects but do not effectively protect any tissues. Such balance, which is useful for either prophylaxis or prevention of further damage around an existing wound, is evidently facilitated by the anisotropy of the MBS, as indicated by our combined PEI and AI indexing method (Figure 7).

The anisotropy of a treatment (and also, of a prophylactic) dressing is a critical feature that needs to be carefully considered in the process of engineering design of dressings, and in product evaluations. Based on the current work as well as our previously published articles, it is already clear that the stiffer direction of the dressing must be the direction of the spine, along which most of the tissue distortion occurs, particularly in shear. In addition, however, the ratio of dressing stiffness along the spinal direction over its stiffness along the lateral (buttocks cheeks) direction, which constitutes the extent of anisotropy of the dressing, should be adequate. This ratio should not be too low, to achieve the biomechanical protective effect demonstrated here, but it should also not be too high, so not to cause stress concentrations along the spinal direction of the dressing. The optimal anisotropy value for a treatment (or a prophylactic) dressing is not known at this time (and could be different across patient populations), however, the MBS dressing appears to approach such an optimum given its remarkable biomechanical protective effect, as per the findings of this and our previous work. From a clinical perspective, the MBS dressing not only delivers excellent prophylactic outcomes26, 27, 28 but also leads to better resolution of existing PUs and deep tissue injuries (DTIs), as demonstrated by Sullivan.29 In this latter study, which included 77 adults hospitalised at the Mayo Clinic in the United States, it has been demonstrated that the normal trajectory of tissue destruction associated with DTIs was changed, and DTIs healed, through early intervention with the MBS dressing. This provides strong clinical support for the present computational modelling results, demonstrating that the MBS either prevents DTIs26, 27, 28 or helps DTIs not to deteriorate.29

In the present article, we introduced, for the first time in the wound care literature, two important indices, PEI and AI, which can be used jointly to objectively and quantitatively assess the biomechanical efficacy of any prophylactic or treatment dressing. The PEI reflects the quality of a tested dressing in alleviating superficial (first sub‐VOI) and deep (third sub‐VOI) tissue stresses near a bony prominence (and also, potentially, near an existing wound). This is weighed against the AI that captures the extent of the stress shielding caused by the said preventive/treatment dressing (second sub‐VOI). By considering these two indices together, it becomes straight‐forward to perform a standardised evaluation of the aforementioned trade‐off between stress alleviation and stress shielding, that is, of the stress redistribution in tissues protected by the centre of the dressing as opposed to the potentially resulting increase in tissue stresses at and near the borders of the dressing. Here, we found that this trade‐off was excellent for the MBS dressing (Figure 7). Through its anisotropy, the MBS achieved PEI that is more than twice the ones provided by the HIS and HIS dressings in deep tissues (first sub‐VOI) and as good as that of HIS (but substantially better than HIF) superficially (third sub‐VOI). When examining the stress shielding (trade‐off) aspect, it is shown that the AI for the HIS is considerably greater than that of the MBS (Figure 7B) despite that the MBS has a better PEI in deep tissues and an equivalent PEI superficially (Figure 7A).

Assumptions and limitations are inevitable in any computational modelling work and should be discussed here for completeness. In this first work on the biomechanical efficacy of dressings at the continuum of care between prophylaxis and treatment, the PU (Figure 1) was incorporated virtually in the 3D buttocks model (which has been created from MRI scans of a healthy adult). Hence, the real‐world diversity in specific locations, sizes, depths and shapes of PUs was not considered in our study, nor did we take into account other possible anatomical variations (which are unrelated to the wound). Furthermore, the potential variability in mechanical properties of injured as well as non‐injured tissues was not considered. Nevertheless, our use of the same reference anatomy and biomechanical properties in the modelling framework facilitated the consistent, methodologically systematic quantitative comparisons that are reported here.

To conclude, in this study, we found that the anisotropic MBS dressing is remarkably effective in alleviating tissue stresses superficially and deeply in the wound bed for supine patients for whom a complete off‐loading of the sacral wound is not possible. Minimising the exposure of the wound and its surroundings to elevated sustained mechanical loads is crucial for healing, or at the least for restraining additional tissue damage. We again identified the anisotropic feature, which makes the MBS dressing less stretchable in the longitudinal (spine) direction and more flexible in the lateral (buttocks cheeks) direction as the key design element providing effective redistribution of bodyweight and frictional loads, now demonstrating this for tissues at the environment of an existing wound. Future research may focus on further adjusting the stiffness preference in sacral dressings, which can be potentially different between purely prophylactic dressings vs. treatment dressings (if no off‐loading can be provided).

Schwartz D, Gefen A. The biomechanical protective effects of a treatment dressing on the soft tissues surrounding a non‐offloaded sacral pressure ulcer. Int Wound J. 2019;16:684–695. 10.1111/iwj.13082