Accepted Manuscript Engineering Mechanical Gradients in Next Generation Biomaterials - Lessons Learned from Medical Textile Design Joanna L. Ng, Ciara E. Collins, Melissa L. Knothe Tate PII: DOI: Reference:
S1742-7061(17)30166-6 http://dx.doi.org/10.1016/j.actbio.2017.03.004 ACTBIO 4772
To appear in:
Received Date: Revised Date: Accepted Date:
21 August 2016 2 March 2017 3 March 2017
Please cite this article as: Ng, J.L., Collins, C.E., Knothe Tate, M.L., Engineering Mechanical Gradients in Next Generation Biomaterials - Lessons Learned from Medical Textile Design, Acta Biomaterialia (2017), doi: http:// dx.doi.org/10.1016/j.actbio.2017.03.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Engineering Mechanical Gradients in Next Generation Biomaterials Lessons Learned from Medical Textile Design
Joanna L. Ng1, Ciara E. Collins1, Melissa L. Knothe Tate1 1
Graduate School of Biomedical Engineering, The University of New South Wales, Sydney, Australia
Corresponding Author: Professor Melissa L. Knothe Tate, Paul Trainor Chair of Biomedical Engineering Graduate School of Biomedical Engineering Samuels Building 509 UNSW Sydney NSW Australia T: +61 02 9385 3924 F: +61 02 9663 2108 E: [email protected]
Engineering Mechanical Gradients in Next Generation Biomaterials Lessons Learned from Medical Textile Design Joanna L. Ng, Ciara E. Collins, Melissa L. Knothe Tate Abstract Nonwoven and textile membranes have been applied both externally and internally to prescribe boundary conditions for medical conditions as diverse as oedema and tissue defects. Incorporation of mechanical gradients in next generation medical membrane design offers great potential to enhance function in a dynamic, physiological context. Yet the gradient properties and resulting mechanical performance of current membranes are not well described. To bridge this knowledge gap, we tested and compared the mechanical properties of bounding membranes used in both external (compression sleeves for oedema, exercise bands) and internal (surgical membranes) physiological contexts. We showed that anisotropic compression garment textiles, isotropic exercise bands and surgical membranes exhibit similar ranges of resistance to tension under physiologic strains. However, their mechanical gradients and resulting stress-strain relationships show differences in work capacity and energy expenditure. Exercise bands' moduli of elasticity and respective thicknesses allow for controlled, incremental increases in loading to facilitate healing as injured tissues return to normal structure and function. In contrast, the gradients intrinsic to compression sleeve design exhibit gaps in the middle range of physiological strains and also inconsistencies along the length of the sleeve, resulting in less than optimal performance of these devices. These current shortcomings in compression textile and garment design may be addressed in the future through implementation of novel approaches. For example, patterns, fibre compositions, and fibre anisotropy can be incorporated into biomaterial design to achieve seamless mechanical gradients in structure and resulting dynamic function, which would be particularly useful in physiological contexts. These concepts can be applied further to biomaterial design to deliver pressure gradients during movement of oedematous limbs (compression garments) and facilitate transport of molecules and cells during tissue genesis within tissue defects (surgical membranes).
1. Introduction Nature abounds with gradients, such as the naturally occurring spatial variations in mechanical stiffness and porosity, which are intrinsic to biological tissues and tissue bounding interfaces. Parallels can be drawn between the spatial architecture of bones and the porosity of sea sponges, as well as between interfacing biological morphologies at the tendon enthesis and those of juxtaposed biological systems in nature [1-3]. Gradient engineering provides a novel approach to engineer biomaterials that emulate the smart, emergent properties of their natural biological counterparts . Such smart properties confer the capacity to adapt to the dynamic environment and thereby induce or instruct changes in biological structure and function . Examples of smart biomaterials include flow-directing biomaterials that mimic the stiffness, porosity and counterintuitive flow properties of bone under mechanical loads , and textile weaves that mimic scaled-up three-dimensional patterns of elastin and collagen fibres comprising the fibrous layer of periosteum, a hyperelastic sleeve that exhibits strain stiffening behavior and thereby splints and strengthens bones under impact loads . Recent efforts have emphasised either top-down quantification of gradients in biological materials, including the resulting smart properties intrinsic to such gradients , or bottom-up engineering of gradients in biomaterials using novel technological approaches [5, 6]. Few published reports have used engineering methods to assess mechanical gradients intrinsic to currently implemented biomaterials such as medical textiles and surgical interface membranes. Such studies are expected to guide the design of next generation biomaterials and devices that incorporate gradient engineering approaches . In addition they may speed medical translation of new design approaches, as they are implemented with processes used to manufacture current, regulatory body-approved devices (FDA-, CE Mark-, and/or TGA-approved) . Non-woven and textile membranes have been applied both externally and internally (cf. Graphical Abstract) to prescribe boundary conditions for medical disorders as diverse as oedema (resulting in swelling and pushing out of normal external or internal physiologic boundaries) and tissue defects (resulting in pulling in of normal external and internal physiologic boundaries). Membranes designed for external use show the greatest range of structure, function, and resulting applications, likely due to their limited number of hurdles of obtaining regulatory approval. For example, compression sleeves for oedema treatment are designed to generate external pressure to compress the circumference of and thereby to reduce swelling of oedematous limbs by facilitating lymph flow to the heart. However, the putative mechanical gradients intrinsic to such sleeves, and that underpin the lymph drainage capacity of such sleeves, are not well established. Internal surgical membrane sleeves can also be designed to impart directed pressure gradients to tissues healing within, e.g. to guide differentiation of progenitor cells through transmission of mechanical and mechanically modulated biochemical cues or to guide nutrient and/or growth factor transport to promote tissue genesis within defects [9-12]. However, surgical membrane design and implementation has been faced with a number of setbacks in the past decade, resulting in device recalls of both solid  and textile (mesh) surgical membrane implants [13, 14]. Hence, as a first step towards engineering medical textiles that incorporate mechanical gradients and enable smart fabric properties ,
this study aims to quantify gradients in currently implemented devices used to treat oedema and tissue defects. These insights are then set in context of engineering and design of next generation medical textiles that incorporate mechanical gradients to increase functionality in both internal and external applications.
2. Materials and Methods Our working hypothesis postulates that the function of bounding membranes used for physiological purposes depends on the stiffness gradients intrinsic to both the materials from which they are made as well as their architecture when implemented in external and internal physiological contexts. Our approach was to test mechanical properties of materials from which external and internal sleeves are made, as well as to determine whether sleeve architecture affects mechanical properties and physiological function. First, we measured the mechanical properties of external, medical compression sleeves to determine potential directional dependence, i.e. anisotropic. We then tested compression sleeve architectures for influence on the mechanical properties of the sleeve. In a second step, we investigated mechanical properties of internal, surgical membranes. For comparison, we quantified and compared mechanical properties of external lymphoedema sleeves and internal surgical membranes with those of elastomeric exercise bands. Exercise (physio) bands were used in colour-coded series to incrementally increase loading and thereby facilitate healing as injured tissues return to normal function. 2.1 Materials 2.1.1. Compression sleeve selection Compression sleeve design varies by class of compression needed (I-III) and size of the arm, which in turn depends on size of the patient and degree of oedema. To make the current study as generalisable as possible, we chose Class II 20-30 mmHg, FDA approved, ready-to-wear, over-the-counter compression arm sleeves as representative. The large size was chosen to maximise materials for testing. Compression sleeves are comprised of a 72% polyamide and 28% Lycra® knit. 2.1.2. Preparation of sleeves for gross analysis and mechanical testing Compression sleeves (Da Yu Enterprise, Changhua County, Taiwan) were cut lengthwise and unfolded to resemble a trapezoidal planar surface (n=4). The sleeves were then divided into four sections along their length, between the wrist (Section 1) and upper arm (Section 4) (1-4, Fig. 1A). Compression sleeve weaves were imaged with a Leica M80 stereomicroscope (Leica Microsystems, CHE) at 25x magnification and characterised using image analysis. Distances between the base yarns were measured using the open source image analysis program, Fiji (Fig. 2) . To assess material anisotropy, swatches of three alignments were investigated, including swatches with longitudinal alignment along the length of the sleeve, circumferential alignment around the limb, and oblique alignment at a 45 o angle according to the weave bias (Fig. 1). Within each sleeve section (1-4), six swatches were cut, each measuring circa 50 mm x 20 mm. Swatch thickness was measured using a digital precision caliper (Mitutoyo Corp., JP), resulting in a mean thickness of 0.45 mm (n=3). Mechanical properties of compression textile specimens were compared with those of graded latex exercise bands (TheraBand Resistance Bands, The Hygenic Corporation, 3
Akron, OH) and silicone elastomer surgical membranes (BioPlexus Corporation, Ventura, CA). Sections were cut to the same size as the compression sleeve swatches (50 mm x 20 mm). Surgical grade silicone membranes are manufactured to 0.05 mm thickness. Triplicates of the seven exercise band grades of the surgical membranes, represented by distinct colour, were investigated in this study. 2.2 Mechanical Testing To measure the elastic modulus, the 50 mm x 20 mm compression sleeve swatches were secured to custom designed testing grips, parallel to the short edge. The custom grips have been described previously in the literature for skin  and periosteum  testing. Once secured, the testing gauge length was 30 mm, and the load cell was tared. The gauge length was then pulled to 40 mm (equivalent to circa 0.3 N prestress) to avoid noise associated with the toe region of the stress-strain curve. Following an identical protocol, elastomeric membrane sections were also prestressed to circa 0.3 N prior to loading. The specimens were cyclically loaded at 0.25 Hz (6 mm/s) for 10 cycles using a Bose Electroforce 3230 Series II mechanical testing machine (Bose Corporation, Eden Prairie, MN). The inputs of the mechanical test were programmed using WinTest 7.1 (Bose Corporation, Eden Prairie, MN). A triangular wave input, demarcating constant strain rate, was applied over a range of physiological displacements from 0 to 12 mm, which was governed by the z-position of the displacement transducer. The lengths at initial and maximum displacements measured circa 40 mm and 53 mm, respectively. The cross-sectional grip area was 20 mm x 0.45 mm (i.e. width x thickness). A 45 N load cell (0.34% error in tension) was used to measure the stress-strain relationship of the samples. 2.3 Data Acquisition Time-force data were acquired at 0.1 sec intervals for the duration of the tensile test using WinTest 7.1 software. 2.4 Calculations for Compression Sleeve Material Properties Force-displacement data were exported, and the elastic modulus (E) was calculated in an automated fashion using a custom Matlab algorithm (Mathworks®, Natick, MA) and engineering stress-strain relationship. The elastic modulus was calculated between 0.2ε and 0.3ε of the stress-strain curve with increments of 0.01ε using the following equation: = ∆σ /∆ε = σ − σ /ε − ε
where E = elastic modulus (N/m2 or Pa), σ = engineering stress (N/m2), and ε = engineering strain, and, = / (2) where F = tensile force applied to the swatch (N), and A = the cross sectional area of the sleeve (m2), i.e. width (w) x thickness (t) of the sample held between the grips. To calculate the hoop or circumferential stress at each section, the elastic modulus from equations (1) and (2), and the strain as specified by the manufacturer (Table 1) were substituted into Hooke’s Law using the following equations:
= ∆/ = ( − )/
where εp = strain, l0 = the circumference of an unstressed sleeve, and lmax = the circumference of a stretched sleeve within physiological bounds, and, = ∙
where σH = hoop or circumferential stress. The four sections of the sleeve were modeled as four cylinders of varying radii (r) (Fig. 1B). The radii of the cylinders were found by rearranging the equation for the circumference of a circle (Table 2). Since t 100% greater than that of sections 3 and 4 (1.79 ± 0.61 and 1.71 ± 0.65, respectively). This suggests that the capacity of the sleeve to exert force on the oedematous arm decreases in an exponential trend from the distal to proximal ends of the sleeve (Fig. 4B). No significantly significant differences were observed in force between sections 3 and 4, indicating that the structural stiffness and sleeve function are quite consistent from the upper arm and towards the shoulder. Sample alignment (longitudinal, circumferential, and oblique) exerts a statistically significant effect on each swatch's resistance to tension (i.e. the force transduced at 12 mm displacement), with the distal most section of the sleeve (1) showing different trends than all other sections (3-4). Excluding the distal most section, circumferentially oriented samples exhibited the greatest resistance to tension, followed by obliquely and longitudinally oriented samples (Fig. 4A,C). Forces measured in the longitudinal direction showed divergent effects depending on sleeve sectioned examined. In section 1, the mean force in the longitudinal direction was 4.58 ± 0.23 N, however by section 2 and onwards this dropped by almost three-fold to 1.59 ± 0.08 N. The resistance of compression sleeve swatches to tensile loading is also reflected in its calculated elastic modulus (E), which are 1.07 ± 0.07 MPa and 0.37 ± 0.02 MPa, respectively (Table 2A). Hence, for compression sleeves, the tensile resistance is directly proportional to the material’s stiffness and elasticity. 6
3.3. Calculations for hoop stress and pressure The strain of ε = 0.4 used for testing was designed to fall within strains specified by the manufacturer for normal compression sleeve wear and use (Table 3), albeit closer to the lower boundary (BL, minimum value) than the upper boundary (BU, maximum value). Calculations for theoretical hoop stress at both lower and upper boundaries show that hoop stress is not significantly different between sections 2-4 (0.204 ± 0.05 MPa and 0.394 ± 0.05 MPa, respectively for BL and BU). However, due to fabric geometry, pressures embued by the sleeves range between 14.512 – 28.931 mmHg at sections 2-4. At the upper strain boundary corresponding to maximally oedematous arm use, calculated pressures (25.74 ± 3.20 mmHg, Table 3) fall within the specified Class II range of 20 – 30 mmHg. At the lower strain boundary, calculated pressures fall below the range specified for Class II garments (16.23 ± 1.77 mmHg, Table 3). 3.4 Mechanical characterisation of elastic bands for external and internal use Overall, the maximum tensile force (i.e. axial force at 12mm displacement) obtained for the graded latex exercise bands of both manufacturers were sufficiently different to be arranged into a decreasing trend (Fig. 5A,B), with grey as the most (4.45 ± 0.15 N) and red as the least (1.16 ± 0.10 N) resistant to tension (Fig. 5B). The force magnitudes obtained for the exercise bands were similar to the compression sleeves, confirming their adequacy as an isotropic (i.e. non-directionally dependent) elastomeric standard against which the compression sleeve swatches can be compared. Within their respective groups, the mean force responses between exercise band grades were all statistically different, suggestive of their sequential mechanical application. Comparing exercise bands according to their thickness showed that force capacity increased linearly according to thickness for both exercise band manufacturers (Fig. 5C). Overall, comparing bands from Manufacturer X and Manufacturer Y, X's forces were higher and had a greater slope, indicating that there was greater increase in force as thickness of the material increased (Fig. 5C). But when exercise bands were normalised against thickness for stress-strain data, the grades that resulted in the highest tensile forces no longer correlated to higher elastic moduli (E) (Supplementary Fig. 1). However, the elastic modulus was similar between bands from the same manufacturer (Fig. 5D). An unpaired t-test between bands from each of the two manufacturers (µEx = 1.463 ± 0.023 MPa, µEy = 1.001 ± 0.029 MPa) showed statistically significant differences (p < 0.0001, n=15 and 20, respectively). These findings indicate that latex exercise bands are an appropriate isotropic control for anisotropic compression sleeves, due to their similarities in peak tensile force and elastic moduli, and that material thickness can be used to modulate mechanical properties. 3.5 Mean comparisons between compression sleeves and elastomeric membranes Compared to anisotropic compression sleeve textiles, elastomeric membranes used for exercise bands and surgical membranes are inherently isotropic. Despite this difference, both materials exhibit axial forces at 12 mm physiological displacement (Fig. 4,5) between 0 and 5 N. Hence, these seemingly disparate materials exhibit comparable force capacity in this context.
The graded exercise bands from each of two manufacturers showed incremental changes in force but the range in force capacity of Manufacturer Y was much greater than that of Manufacturer X (1.15 – 4.46 N and 1.84 – 3.20 N, respectively) (Fig. 5A). The elastomeric silicone membrane, designed for internal use, showed the smallest force, i.e. smaller than that of the elastomeric materials tested by a factor of 10 (0.60 ± 0.09 N) (Supplementary Figure 2). In contrast, the elastic modulus of the silicone membrane (1.68 ± 0.07 MPa) was significantly larger than the mean moduli of the exercise bands from both manufacturers X and Y (1.46 ± 0.09 MPa and 1.00 ± 0.13 MPa, respectively) (Fig. 5D). By superimposing the exercise band grades as sections along the length of the compression sleeves, the differences in means of the force data of both materials were compared (Fig. 6A, B). Different grades in exercise bands were consistently different to other grades, regardless of manufacturing brands. Conversely, section 1 of the compression sleeve showed more than twice the difference between means when compared to other portions of the sleeve. The disparity between section 1 and the other sections suggest that forces do not exhibit a gradual transition from the distal to proximal portions of the sleeve, unlike the exercise bands, even though they do exhibit a decrease in mechanical properties along the sleeve's length. Nonetheless, due to their similar range in peak tensile force and energy expenditure, these findings further confirm that latex exercise bands are an appropriate isotropic, baseline control for comparison of anisotropic compression sleeves (Supplementary Figure 3). 3.6 Energy expenditure during the load cycle Over a range of physiological displacements, materials (both textile and elastomeric membranes) returned to their original starting point after the load cycle, as observed in material hysteresis curves. This indicates that there was no loss in mechanical integrity of the material over the loading cycle (Supplementary Fig. 4). During the load cycle, compression sleeve swatches exhibits a decreasing trend in work capacity (energy expenditure) between the distal and proximal portions of the sleeve (Fig. 3,7A), similar to that observed in the peak force data (Fig. 4A). However, no significant differences were observed in energy expenditures between the longitudinal and oblique swatches. In addition, the circumferential alignment of the compression sleeves, which was shown to have the highest overall resistance to tension, had the least energy expenditure and variation (Fig. 7B). This suggests that compression sleeve energy expenditure (a measure of performance) is most conserved and consistent in the circumferentially oriented swatches, and further confirms the anisotropic nature of compression sleeves. With respect to the exercise bands, trends similar to those observed in the peak force data (Fig. 5) were observed though differences between bands were less significant (Fig. 7 C,D).
4. Discussion Based on the results of this study, bounding membrane materials used for physiological purposes, from anisotropic compression garment textiles to isotropic exercise band and surgical membranes, exhibit similar ranges of resistance to tension under physiologic conditions, but their gradients and resulting performance show stark differences. Considering all exercise bands studied from two manufacturers together, their graded resistance is attributable to their moduli of elasticity and respective thicknesses, which allows for controlled, incremental increases in loading to facilitate healing as injured tissues return to normal structure and function. In contrast, the gradients inherent to compression sleeve and textile design are inconsistent and exhibit gaps in the middle range of physiological strains, as well as inconsistencies along the length of the sleeve. This potentially leads to less than optimal performance of these medical textile devices, which are designed to facilitate lymphatic flow towards the upper body. In addition, the textile anisotropy and structure of compression garments significantly affect material performance as measured by work capacity (energy expended in the hysteresis loop). However, work capacity is less affected by gradations in isotropic exercise bands. While the Class II compression sleeve tested did exhibit some capacity to deliver a gradient in pressure to an oedematous limb, there is room for improvement in textile and garment design to optimise function. These concepts can be applied further to design next generation surgical membrane implants, where currently isotropic elastomeric membranes could be designed to deliver pressure gradients and thereby facilitate transport of molecules and cells during tissue genesis. 4.1 Material properties and mechanical integrity Compression sleeves are tight-fitting garments that treat lymphoedema through volume reduction. The mechanism of volume reduction is to constrict the limb by applying forces that increase muscle stiffness, venous return and lymph drainage [18, 19]. Because sleeves have a smaller circumference than limbs, they are stretched over the limb when worn. In this, the ‘stretch’, or elasticity of the material, inherently applies a pressure through the release of strain energy at the interface between sleeve and limb . The pressures and forces acting on this interface are dependent on the geometry of the limb, geometry of the sleeve, and the material of the sleeve [21, 22]. 4.1.1 Knit pattern and anisotropic properties Jersey knits are weft knitted, i.e. the yarn loops that comprise the courses are connected by interlocking in the wale direction (Fig. 2A). These knits do not allow for inherent elasticity in its ‘weave’ unless elastane or spandex yarns are included in its fibre composition. The incorporation of an inlay elastane thread, parallel to the course of the knit, increases the course density and therefore promotes elasticity of the material [23, 24] by reducing the friction between the interlocking points of the loops . In this study course density was investigated by measuring the distance between the base yarns in the wale direction. The gradual increases in distance between these the wales suggest that course density decreased between sections 1 to 4, and thus the knit is tighter at the wrist portion and looser towards the shoulder (Fig. 2B). In addition, the elastane inlay thread captured using stereomicroscopy is mechanically important in resisting strain and increasing recovery of the material after strain. In the current study, mechanical significance of the inlay thread was evident in both ‘peak
force vs. alignment’ (Fig. 4) and ‘energy expenditure’ datasets (Fig. 7). Of the former, the circumferential direction exhibited the highest resistance to tensile loading. One possible explanation is that the direction of the force was parallel to the larger diameter elastane fibres, which if perceived as a composite material, would directionally resist greater tension. Regarding energy expenditure, energy was most conserved in the circumferential direction, running parallel to the elastane inlay thread. This is because elastane is a material that has the capacity to efficiently store and release kinetic energy . Hence, energy expenditure in the longitudinal alignment, which ran perpendicular to the circumferential direction, was greater because of the reduced capacity to store and release energy in the base yarn. Furthermore, the curvilinear shape, depicting the energy expenditure in the longitudinal direction of section 1, is much greater than that in all other sections and alignments. This is likely due to the fact that this particular swatch was strained at ε= 0.4, which is more than twice the specified upper boundary elongation of ε= 0.171 (Table 2). Hence, the material was strained beyond its mechanical capacity and physiological necessity. Thus, material anisotropy can be attributed to the alignment of the fibres and the material properties of the fibres running in these directions. 4.1.2 Material geometry and elastic modulus Not only is the knit and composition of the fabric able to influence the mechanical properties of the compression sleeve, but the shape and overall geometry play an important role in determining the functional capacity of the sleeve. As mentioned above, sleeves alleviate lymphoedema and fluid congestion through their smallerthan-limb diameter and elasticity. Regarding the former, the correlation between the size of the sleeve and the limb is the reduction factor, in which the sleeve is recommended to be between 10-20% less than the circumference of the limb . In a previous study, reduction factors in this range have exerted a pressure of 20-25 mmHg onto the limb-sleeve interface . The current study confirmed that the theoretical pressures exerted from the sleeve to the arm were typical of a Class II device, though these measurements were not calculated using Laplace’s Law, as in previous studies [25, 27]. This is because the use of Laplace’s Law to predict pressures are only applicable for large diameter cylinders, and the lower boundary of the specified strain of the sleeves was less than the minimum requirement of 25 cm circumference . Material thickness is also a parameter that can be used to moderate strength and resistance to loading. In both exercise band brands, there was a direct correlation between thickness and peak tensile force (Fig. 5C), despite the brands having different elastic moduli. Interestingly, exercise band peak force data (Fig. 5A,B) showed a stepwise decrease in maximum force at 12 mm displacement across different grades. However, this stepwise order was not maintained upon calculating the elastic modulus. For example, forest bands displayed lower force capacity to resist tension than black bands (Fig. 5B), yet appeared to have a greater E value (Supplementary Fig. 5). Due to the alternating pattern in the data (Supplementary Fig. 5), this suggests that thickness is only one of the parameters used to control the strength of the exercise bands; it is unclear what other material properties are also influencing its strength. Hence, material geometry and thickness are parameters that can be used to modulate the reduction factor and strength of fabrics, respectively.
4.2 Marked gradations in compression sleeves Gradients comprising a gradual reduction along the length of the sleeve are a significant factor in the decongestive capacity of compression sleeves. In the current study, this capacity was investigated, through analysis of mechanical properties of the sleeve in four sections along its length. As expected, it was found that the distal portion of the sleeve exhibited higher resistance to tension, hoop stress, greater elastic modulus and interfacing pressure than the proximal portions of the sleeve. This is significant because it aids in the flow of fluid from high pressure to low pressure regions. Thus, this finding confirms that the sleeve geometry correlates to its function. However, in the analysis of the consistency of gradations between sections, it was discovered that these gradations were not equally distributed along the length of the sleeve. The mechanical properties of section 1 appeared to be substantially different than the other 3 sections, while sections 3 and 4 showed no difference in reductive capacity (Fig. 6A). When compared to the exercise bands, which have been proven to exhibit progressive and linear increases in force capacity between grades [28, 29], it becomes clear that compression sleeves lack a gradual stepwise gradation between the sections. This is clinically relevant, because consistent gradual decreases in pressure between the distal and proximal sections of the sleeve are required for lymph drainage, decongestation and overall volume reduction. It is possible that the gradation of compression sleeves may be attributed to the course density of the fabric, which also decreases along the length of the sleeve. 4.3 Limitations Compression sleeves are classified medical devices, of which many are comprised of a jersey base stitch with an inlay spandex thread. A previous study has shown a similar mechanism of action for different compression sleeves . One limitation of the current study was that only one brand of sleeve was tested, although it was selected (Class II, size large, n=4 replicates) to be generalisable to a range of sleeves from different classes and manufacturers. For performance and value comparison of currently commercialised compression sleeves, a Consumer Reports type study would be of benefit, i.e. comparing head-to-head the different sizes, grades and manufacturers of compression armsleeves. However, the ultimate aim of the current study was to test and understand gradients in currently used biomaterials for external and internal boundary functions, and thereby to instruct design of next generation medical textiles for external and internal use. Another limitation of the current study was the inability to obtain data at the upper boundary of strain specified by the compression garment manufacturer. This is due to the limitation of the mechanical testing set up, which was designed for physiological conditions. While exceeding physiological limits, upper boundary data would be useful in identifying the material response at maximum strain and the occurrence, if any, of plastic deformation or excessive energy expenditure in the system. Lastly, the idealisation of the limb as a perfect cylinder with specified circumferences for material strain allowed for calculation of theoretical values for strain pressure and hoop stress. However, limbs are neither perfectly cylindrical nor have identical stiffness around their perimeters. In fact, there is greater strain at joints, particularly during flexion , which would affect the stretch of the knit at the joints and hence
also influence the pressures over these regions. Currently, it is not yet possible to measure with precise accuracy how joint regions or differences in limb geometry or stiffness affect the mechanics of compression sleeves. However, finite element analysis has been used to simulate garment designs on the wearer . Ongoing studies using digital image correlation and magnetic resonance imaging, will allow for more dynamic and physiological measurements in the future. 5. Future directions and conclusion This study has shown that there are many parameters involved in garment design that contribute to its mechanical properties and function, including knit pattern, inlay thread properties, course density, material thickness and sleeve geometry. There are many forms of knitting techniques that are composed through combination of knitting and purling threads. Jersey knits are commonly used in comfortable, loose-fitting wearable fabrics. Since jersey knitted fabrics are inherently not elastic, the elasticity of the compression sleeve arises from the elastane fibres in the yarns. However, elasticity can also be created through knitting techniques and structures. Ribbed knits, which are a combination of knitting and purling, intrinsically promote cross-wise stretch even without the inclusion of elastane. They are known to be tight fitting, to have good dynamic work recovery and are stiffer than jersey knits. It is plausible that both these properties and knitting techniques would be desirable in the design of nextgeneration compression sleeves. Other means to alter material properties, and in particularly the elastic moduli of fabrics, is through weaving. Woven materials consist of a weft thread that is woven through the warp threads, creating a mesh. Weave patterns can be created in high resolution and customised such that any pattern is possible along the length of the sleeve, with the inclusion of material anisotropy. A previous study has shown that woven textiles for compression fabrics have high elastic moduli compared to knitted fabrics . Given the values for elastic moduli found in the present study, weaving may be able to generate textiles that have elastic moduli that are more applicable to human tissue. This may be beneficial for development of biomaterials that are used for stiffer and less pliable tissue than skin. In addition, the results of this study show that course density is another means that can be used to customise the mechanical properties of compression sleeves. It is possible for course density to be modified through the thickness and material properties of the elastic inlay thread, and also through the tension of the wales. Furthermore, since the width of fabrics can be altered through the diameters of yarn, the development of a sleeve that gradually increased in yarn thickness in the wale direction would be beneficial in customising force gradation along its length. The development of such yarns that change in diameter or stiffness along their length have been patented . Taken together, both course density, as well yarn diameter and stiffness, are parameters that can be used to increase the efficacy of pressure gradation of compression sleeves in the future. In conclusion, external compression garments comprise anisotropic textiles, whose performance falls short of achieving spatially graded pressure along the oedematous limb. Based on the data from this study, compression sleeves are not consistently graded along their length, which may hinder lymph drainage. In contrast, the graded resistance of isotropic exercise bands allows for incremental increase in loading and facilitation of healing as injured tissues return to normal function. The anisotropic
properties of compression sleeves depend on the structure of the wale, course and inlay elastic threads in the knit. This shortcoming could potentially be overcome by modifying parameters such as the material composition, stiffness and diameter of the yarns making up the textile, fabric course density, and structure of the overall knit. These insights are expected to instruct future design of medical textiles for both external and internal use.
Acknowledgements This study was supported by the Paul Trainor Foundation and the Australian Postgraduate Award Scholarship scheme of the Australian Department of Education and Training (JLN). Thank you Daniel J. Hageman and Karen Virdi for your collaboration on this ongoing project. Disclosures The current study is fundamental in nature. For full disclosure, MKT has protected the intellectual property underpinning several technologies that can be implemented for design and manufacture of medical textiles and devices. She is also Co-Founder of a start-up company that aims to translate and commercialise these technologies. To date (15 November 2016), none of the authors has received reimbursements, fees, funding or salary from the current or previous (wound down) start-up companies.
Figure Legends Figure 1. Compression sleeve samples and their mechanical testing. A Schematic of compression sleeve swatch preparation. B Cylindrical model of compression sleeve. C State of stress schematic of pressurised thin-walled cylindrical vessel representing a lymphoedema compression sleeve in situ. Figure 2. Quantitative analysis of compression sleeve textile knit characteristics, based on stereoscopic microscope image analysis (25x magnification, with scale bar representing 0.5 mm, on four representative areas imaged in sections 1-4 of each sleeve). Inset shows measurements used to determine weave density of the single needle weft knit, where the weft, or course, of the fabric is staggered vertically, and the wale is defined as the column of loops running along the length of the knit, and is shown by the diagonal fibres that connect the courses. A The distance between the lengthwise running elastic inlay threads (distance between dotted lines, a/2) and the area enclosed by the wales (black square), define the characteristic length ‘a’ of the weave. B The mean characteristic length ‘a’ of the weave along the sleeve with error bar representing ± standard deviation. Figure 3. Schematic of energy expenditure in hysteresis. Energy expenditure is represented by the area bounded by loading (black) and unloading (red) curves (shaded in blue). Figure 4. Peak tensile forces of compression sleeves at physiological displacement. A Individual value plot of tensile force at 12 mm displacement versus fixed factors (alignment and section). B Mean tensile force depicted for each sleeve section. C Mean tensile force depicted as a function of alignment, with data from each section depicted by colour (1,2,3,4). Error bars represent ± standard deviation. Figure 5. Peak tensile forces of latex exercise bands and silicone elastomer surgical membrane under test conditions identical to those of the medical textile swatches (Fig. 4). A Mean tensile force at 12 mm displacement for Manufacturer X exercise bands. B Mean tensile force at 12 mm displacement for Manufacturer Y exercise bands. C Scatterplot depicting the relationship between force (N) and material thickness. D Elastic modulus of exercise bands and silicone membrane. A-D exhibit significant differences between means unless otherwise stated. Figure 6. Statistical analysis comparing material gradients in compression versus exercise band grades. A Differences between means of tensile force at 12 mm displacement between compression sleeve sections, B Differences between means of tensile force at 12 mm displacement between exercise band grades (no significance if interval crosses dotted line). Figure 7. Analysis of hysteresis curves as a measure of material performance. A Compression sleeve swatches show an exponential decline in energy expenditure between the distal and proximal sections of the sleeve. B Circumferentially oriented swatches has significantly less work capacity (energy expenditure) than other alignments. C Energy expenditure of exercise bands follows a similar trajectory to those of peak resistance tension measured in the bands, albeit with less significant difference between grades (colours) of bands. Error bars represent ± standard deviation.
Supplementary Figure Legends Supplementary Figure 1. Hysteresis curves of exercise bands. A Force displacement data and B Stress - strain data. Supplementary Figure 2. Individual value plot at 12 mm displacement of exercise bands from Manufacturer X, Manufacturer Y, and silicone. Supplementary Figure 3. T-test between superimposed compression sleeve textile sections and exercise band grades. B Peak force data. C Energy expended data. Supplementary Figure 4. Force - Displacement data, hysteresis curves of compression sleeve swatches by section and aligment. A Section 1, B Section 2, C Section 3, D Section 4, E Longitudinal alignment (numbers in legend inset denote section), F Circumferential alignment (numbers in legend inset denote section), G Oblique alignment (numbers in legend inset denote section). Supplementary Figure 5. Elastic moduli of exercise bands, A Manufacturer X, and B Manufacturer Y.
Tables Table 1. Mean compression sleeve measurements before and after physiological strain according to manufacturing specifications Section 1 2 3 4
Radius initial (cm) 2.5 2.9 3.2 3.6
Circumference initial (cm) 15.8 18.2 19.8 22.8
Radius final (cm) 2.7 3.7 4.1 4.9
Circumference final (cm) 17.0 23.5 26.0 30.5
Table 2A. Elastic modulus of compression sleeve by section and alignment Section
E (Mean) (MPa)
E (SD) (MPa)
Table 2B. Elastic modulus of exercise band grades Grade
E (Mean) (MPa)
E (SD) (MPa)
Table 3. Calculations for circumferential stress and pressure inside a thin-walled pressure vessel at final circumference Strain (B L, BU)
FTensile (N) (B L, BU)
Hoop Stress (MPa) (BL, BU)
Radius (cm) (BL, BU*) 2.706 , 2.944
0.076 , 0.171
0.509 , 1.144
0.057 , 0.127
7.049 , 14.575
3.740 , 4.536
0.291 , 0.566
1.801 , 3.499
0.200 , 0.389
18.054 , 28.931
4.138 , 5.093
0.313 , 0.616
1.778 , 3.499
0.198 , 0.389
4.854 , 5.968 0.338 , 0.645 0.618 1.879 , 3.586 0.209 , 0.398 Bolded- Specified circumference from manufacturer corresponding to Class II 20-30 mmHg
P (mmHg) (BL, B U)
16.117 , 25.767 14.512 , 22.534
compression sleeve, size L. *BL: Lower boundary; BU: Upper boundary.
References  M.L. Knothe Tate, M. Detamore, J.R. Capadona, A. Woolley, U. Knothe, Engineering and commercialization of human-device interfaces, from bone to brain, Biomaterials 95 (2016) 35-46.  M.L. Knothe Tate, R. Steck, E.J. Anderson, Bone as an inspiration for a novel class of mechanoactive materials, Biomaterials 30(2) (2009) 133-140.  M. Benjamin, H. Toumi, J.R. Ralphs, G. Bydder, T.M. Best, S. Milz, Where tendons and ligaments meet bone: attachment sites (‘entheses’) in relation to exercise and/or mechanical load, Journal of anatomy 208(4) (2006) 471-490.  B.M. Holzapfel, J.C. Reichert, J.-T. Schantz, U. Gbureck, L. Rackwitz, U. Noeth, F. Jakob, M. Rudert, J. Groll, D.W. Hutmacher, How smart do biomaterials need to be? A translational science and clinical point of view, Advanced drug delivery reviews 65(4) (2013) 581-603.  J. Ng, L. Knothe, R.M. Whan, U. Knothe, M.L. Knothe Tate, Scale-up of Nature's Tissue Weaving Algorithms to Engineer Advanced Functional Materials, Scientific Reports 7 (2017) 40396.  M.L. Knothe Tate, Smart body armor inspired by flow in bone, Smart Structures and Systems 7(3) (2011) 223-228.  S.F. Evans, D. Docheva, A. Bernecker, C. Colnot, R.P. Richter, M.L. Knothe Tate, Solid-supported lipid bilayers to drive stem cell fate and tissue architecture using periosteum derived progenitor cells, Biomaterials 34(8) (2013) 1878-1887.  B.P. Partlow, C.W. Hanna, J. Rnjak‐Kovacina, J.E. Moreau, M.B. Applegate, K.A. Burke, B. Marelli, A.N. Mitropoulos, F.G. Omenetto, D.L. Kaplan, Highly tunable elastomeric silk biomaterials, Advanced functional materials 24(29) (2014) 4615-4624.  S.R. Moore, G.M. Saidel, U. Knothe, M.L. Knothe Tate, Mechanistic, Mathematical Model to Predict the Dynamics of Tissue Genesis in Bone Defects via Mechanical Feedback and Mediation of Biochemical Factors, PLoS computational biology 10(6) (2014) e1003604.  M.L. Knothe Tate, H. Chang, S.R. Moore, U.R. Knothe, Surgical membranes as directional delivery devices to generate tissue: testing in an ovine critical sized defect model, PloS one 6(12) (2011) e28702.  S.F. Evans, H. Chang, M.L. Knothe Tate, Elucidating multiscale periosteal mechanobiology: a key to unlocking the smart properties and regenerative capacity of the periosteum?, Tissue Engineering Part B: Reviews 19(2) (2013) 147-159.  S.F. Evans, J.B. Parent, C.E. Lasko, X. Zhen, U.R. Knothe, T. Lemaire, M.L. Knothe Tate, Periosteum, bone's “smart” bounding membrane, exhibits direction‐ dependent permeability, Journal of Bone and Mineral Research 28(3) (2013) 608-617.  Therapeutic Goods Administration, Urogynaecological surgical mesh implants statement provided to the 7.30 Report (ABC) by the TGA. , 2012 (accessed August 15.2016).  U.S. Food and Drug Administration, UPDATE on Serious Complications Associated with Transvaginal Placement of Surgical Mesh for Pelvic Organ Prolapse: FDA Safety Communication. , 2011 (accessed August 15.2016).
 J. Schindelin, I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, Fiji: an open-source platform for biological-image analysis, Nature methods 9(7) (2012) 676-682.  J.M. Mansour, B.R. Davis, M. Srour, R. Theberge, A method for obtaining repeatable measurements of the tensile properties of skin at low strain, Journal of biomechanics 26(2) (1993) 211-216.  S.H. McBride, S.F. Evans, M.L. Knothe Tate, Anisotropic mechanical properties of ovine femoral periosteum and the effects of cryopreservation, Journal of biomechanics 44(10) (2011) 1954-1959.  M. King, A. Deveaux, H. White, D. Rayson, Compression garments versus compression bandaging in decongestive lymphatic therapy for breast cancer-related lymphedema: a randomized controlled trial, Supportive Care in Cancer 20(5) (2012) 1031-1036.  K. Suehiro, S. Honda, H. Kakutani, N. Morikage, M. Murakami, O. Yamashita, K. Ueda, M. Samura, K. Hamano, A Novel Arm Sleeve for Upper Extremity Lymphedema: A Pilot Study, Annals of vascular diseases 7(2) (2014) 134.  J. Umar, T. Hussain, M. Maqsood, Modeling the mechanical and compression properties of polyamide/elastane knitted fabrics used in compression sportswear, The Journal of The Textile Institute (2015) 1-13.  R. Liu, Y.-L. Kwok, Y. Li, T.-T. Lao, X. Zhang, Effects of material properties and fabric structure characteristics of graduated compression stockings (GCS) on the skin pressure distributions, Fibers and Polymers 6(4) (2005) 322-331.  R. Liu, Y.-L. Kwok, Y. Li, T.-T. Lao, Fabric mechanical-surface properties of compression hosiery and their effects on skin pressure magnitudes when worn, Fibres & Textiles in Eastern Europe 18(2) (2010) 79.  O. Troynikov, E. Ashayeri, M. Burton, A. Subic, F. Alam, S. Marteau, Factors influencing the effectiveness of compression garments used in sports, Procedia Engineering 2(2) (2010) 2823-2829.  M. Senthilkumar, Effect of spandex input tension, spandex linear density and cotton yarn loop length on dynamic elastic behavior of cotton/spandex knitted fabrics, Journal of Textile and Apparel, Technology and Management 7(4) (2012).  L. Macintyre, Designing pressure garments capable of exerting specific pressures on limbs, Burns 33(5) (2007) 579-586.  E. Maklewska, A. Nawrocki, J. Ledwoń, K. Kowalski, Modelling and designing of knitted products used in compressive therapy, Fibres & Textiles in Eastern Europe (5 (59)) (2006) 111-113.  W.Y. Leung, D.W. Yuen, S.P. Ng, S.Q. Shi, Pressure prediction model for compression garment design, Journal of Burn Care & Research 31(5) (2010) 716-727.  G.M. Santos, G. Tavares, G.d. Gasperi, G.R. Bau, Mechanical evaluation of the resistance of elastic bands, Brazilian Journal of Physical Therapy 13(6) (2009) 521526.  P. Page, A. Labbe, R.V. Topp, Clinical force production of thera-band elastic bands, J Orthop Sports Phys Ther 30(1) (2000) 47-8.  J. Tarrier, A. Harland, R. Jones, T. Lucas, D. Price, Applying finite element analysis to compression garment development, Procedia Engineering 2(2) (2010) 3349-3354.
 R. Zouari, S.B. Amar, A. Dogui, Experimental Analysis and Orthotropic Hyperelastic Modelling Of Textile Woven Fabric, Journal of Engineered Fabrics and Fibers (JEFF) 9 (2014) 91-98.  M.L. Knothe Tate, A Substrate. PCT/AU2014/000813. WIPO WO2015/021503. Filed 15 August 2014.
Statement of significance External and internal biomaterial membranes prescribe boundary conditions for treatment of medical disorders, from oedema to tissue defects. Studies are needed to guide the design of next generation biomaterials and devices that incorporate gradient engineering approaches, which offer great potential to enhance function in a dynamic and physiological context. Mechanical gradients intrinsic to currently implemented biomaterials such as medical textiles and surgical interface membranes are poorly understood. Here we characterise quantitatively the mechanics of textile and nonwoven biomaterial membranes for external and internal use. The lack of seamless gradients in compression medical textiles contrasts with the graded mechanical effects achieved by elastomeric exercise bands, which are designed to deliver controlled, incremental increases in loading to facilitate healing as injured tissues return to normal structure and function. Engineering textiles with a prescient choice of fibre composition/size, type of knit/weave and inlay fibres, and weave density/anisotropy will enable creation of fabrics that can deliver spatially and temporally controlled mechanical gradients to maintain force balances at tissue boundaries, e.g. to treat oedema or tissue defects.