Assessing 3D printable density-graded lattice structures to minimize risk of tissue damage from compression-release stabilized sockets

Abstract

Background:

Pressure, shear stress, and friction can contribute to soft tissue damage experienced by a residual limb. Current compression/release stabilized (CRS) socket designs may pose a risk to soft tissue from abrupt compression differences within the socket.

Objectives:

Density-graded lattice structures are investigated for their potential to mitigate risk of tissue damage by assessing their ability to produce more gradual transitions between high-compression and low-compression areas.

Study Design:

A full factorial experimental design was used to reveal the effects of changes among three variables: lattice geometry, density alteration, and displacement magnitude. A total of 144 experimental conditions were examined.

Methods:

Lattice samples representing areas of compression and release based on a novel cushioned transhumeral level CRS style socket design were 3D printed. Compression testing was performed on 2 types of lattice structures which incorporated 1 of 8 design elements to alter density and axial stiffness. The effect on stiffness of the sample as a function of lattice type and density alteration was recorded under 3 loading conditions.

Results:

The offset diamond lattice type with blend radius density alterations produced the only samples meeting criteria set for compression areas of the socket. No samples satisfied criteria for release areas. Transitional density lattices that gradually tapered between the best performing compression and release values were successfully produced.

Conclusions:

Transitional density lattices offer promise for mitigation of soft tissue damage through minimization of compression differentials throughout the socket. Wider implications for this research include use in sockets for other levels of amputation and in orthotics. Future work will focus on lattice optimization to improve release behavior within a modified CRS socket.

Introduction

Pressure, shear stress, and friction can contribute to soft tissue damage experienced by a residual limb.^1–3 While compression/release stabilized (CRS) sockets show evidence of increasing stabilization and user control of prostheses when compared with full-coverage sockets and have the additional benefit of distributing concentrated pressures across the length of the residual bone,^4,5 the abrupt pressure difference and associated shear stresses between the compression and release areas of current CRS designs may pose an unnecessary risk for soft tissue damage to the residuum.^6,7 Density-graded lattice structures for use in a modified transhumeral CRS socket design are investigated as a strategy to mitigate the risk of soft tissue damage by testing their ability to produce more gradual transitions between high-compression and low-compression areas in the socket.

Comparing CRS sockets with full-coverage sockets

Full-coverage sockets have an unbroken surface area that interacts with a person's residual limb. CRS sockets alternate areas of direct longitudinal compression of the soft tissue along the length of the residual bone with open “release” areas that receive the tissue displaced by this compression⁵ (Figure 1).

Figure 1.

The full-coverage socket pictured on the left has an unbroken surface area, while the CRS socket on the right alternates areas of direct compression on the soft tissue of the residual limb with open “release” areas that allow the compressed tissue a place to expand. Here, a transhumeral CRS socket compresses soft tissue intermittently along the shaft of the humerus bone.

Each style of socket has benefits and drawbacks. In a comparison study by Resnik et al, individuals with transhumeral amputations were fitted with both a full-coverage socket and a CRS style socket. Results of dynamic kinematic assessment in this study provided evidence that the use of CRS sockets resulted in better control of the limb within the socket than traditional full-coverage sockets. CRS sockets were also associated with less sweating and slippage; however, users found CRS sockets less comfortable and more difficult to don than their traditional counterparts. It should be noted that this was a small study containing only 2 participants.⁴

Alley et al developed the CRS socket design to increase range of motion, distribute pressure more evenly across the bone, and improve user control over movement of the prosthesis.⁵ The authors demonstrated that range of motion is diminished when the residual bone must first move through layers of soft tissue before its force can be transmitted onto a prosthetic device, causing it to move.⁵ Precompressing the soft tissue down the length of the bone in selected areas allows motion input from the user to pass more directly from the bone to the device to perform the intended movement in the prosthesis. Areas of compression along the length of the shaft of the residual bone were designed to stabilize the socket in its position and more evenly distribute the load of a lifted object over the length of the bone, rather than concentrating much of the load at a point on its distal end.^5,8 Alley and Albuquerque had fitted over 20 upper limb patients with 1 or more sockets during development of the CRS socket design. The authors reported that, in all cases in which the patient had a preexisting socket, the CRS socket allowed for a greater range of motion and improved stability and that it increased the load capacity users could handle with their terminal devices.⁵

Schofield et al utilized pressure sensor data obtained from the socket/residual limb interface of a full-coverage socket to reveal areas on a transhumeral residual limb that experience the most pressure while holding the prosthesis in 2 flexed positions (at elbow only to 90° or at shoulder only to 90°). They tested each position using the weight of the prosthesis alone and then with the weight of the prosthesis plus an additional external 1 kg weight at the terminal device. Pressure was most often concentrated near the tip of the residual humerus and in the tissue near the axilla.⁹

Because the humerus bone exists within the length of the residual limb surrounded by muscle, fat, nerves, blood vessels, and other soft tissue, when the weight of a prosthesis or a lifted object is added, the system begins to behave like a first-class lever in which the distal end of the humerus acts as the fulcrum. The weight causes a moment, or rotation of the prosthesis, around this pivot point. As the prosthesis rotates counterclockwise (Figure 2), the bone, to support the additional weight, spins clockwise resulting in pinching of the soft tissue between the tip of the humerus and the prosthetic device. At the same time, the rotating prosthesis also pinches the tissue near the axilla between the prosthesis and the bone, thus aligning with the findings of the previously cited pressure sensor research. The CRS socket approach, as described by Alley et al, serves as a method to avoid this uneven distribution of load and pressure across the residual limb by precompressing the soft tissue down the length of the bone, allowing the bone and prosthesis to move as one rather than in opposing rotations.⁵

Figure 2.

(a) The humerus is positioned roughly centered within the residual limb surrounded by muscle, fat, and other soft tissue. (b) As the weight of an object is added to a terminal device, the prosthesis begins to rotate counterclockwise, while the humerus rotates clockwise. The soft tissue at the distal end of the humerus and near the axilla becomes pinched.

Despite some of the advantages of CRS sockets reported in the aforementioned studies, an unmentioned drawback is the increased risk for soft tissue damage to the residual limb from the intentional placement of high-compression areas directly next to areas intended to supply no compression, the open release areas from which soft tissue bulges. As evidenced by Leon Bennett in 1971, large changes in compression levels over short distances can contribute to soft tissue damage due to shear stress.⁷ In that study, cysts developed in the soft tissue of patients' residua just above the brim of their sockets, proximal to where their prostheses touched the skin. This led Bennett to create 2 low-shear socket designs still in use today that gradually taper the compression experienced within the tissue at the brim of the socket over a given distance.

It is now understood that much of the socket-related tissue damage found in residual limbs traces back to localized disruptions of blood flow and impaired lymphatic drainage.^9,10 The combination of pressure, its associated shear stress in the tissue, the friction between a device and the skin, and the interaction of these factors with patient-specific tissue attributes contribute to this damage.^1,2,6,9,10 While current literature highlights the role of such risk factors as they relate to prosthetic socket use in general, to the best of the authors' knowledge, no current studies expressly link incidences of tissue damage specifically to CRS socket use.¹¹ It should thus be noted that risk of tissue damage in relation to CRS socket use as discussed in this article is based on the presence within CRS designs of factors for which evidence of a correlation with tissue damage is known.

Density-graded lattice structures

In this research, flexible density-graded lattice samples are (1) tested to determine their capacity to serve as adequate compression and/or release areas and (2) investigated as a strategy to gradually taper compression experienced in soft tissue between areas of compression and release over a controlled distance for use in a modified CRS socket design. Similar lattice structures are being explored in research and industry in applications such as shoe soles and foot orthoses.^12–16 Levels of firmness and support within different areas of the application can be individualized based on a specific person's needs. Firmness may be dictated by a local variation of beam or wall dimensions within a single unit cell, the geometry that is repeated to form a lattice structure, or by increasing or decreasing the number of unit cells present in a given space. Variations of either factor can impact the stiffness of the resultant lattice structure. The process of achieving control of lattice stiffness is the density-graded lattice strategy which is evaluated in this study in the context of a novel transhumeral prosthetic socket design. Ideally, the process could be modified for use in other levels of upper or lower limb sockets.

Methods

Design strategy

To retain the reported benefits of current CRS socket designs, which include stability, range of motion, and increased control over the prosthesis, placement of compression and release areas along the humerus remained similar in the model proposed in this study. However, using density-graded lattices, the stark compression differences between areas of compression and release can be made more gradual, thus decreasing the risk for tissue damage. The ability to modify lattice densities to control firmness levels in cushioning can allow clinicians to pinpoint precise areas within the socket where soft tissue should or should not be compressed or be in contact with it at all.

Figure 3(a) shows an example of a single unit cell building block. When unit cells are repeated and stacked, a 3-dimensional lattice structure is created. Numerous building block geometries exist, each with its own cushioning behavior that can be fine-tuned. Low, medium, and high-density values can be achieved using the same building block throughout a device by varying its beam or wall thickness (Figure 3(b)).

Figure 3.

(a) An example of a single unit cell. (b) Unit cell repeated to form a lattice, and lattice beam thickness altered. This change in density can vary cushioning behavior in the sample. (c) Initial computer-aided design concept showing how various lattice structures of high-density, intermediate-density, and low-density lattice structures could be organized within a CRS style socket.

Autodesk Fusion 360 (Autodesk, San Rafael, CA) and nTop (nTop Inc, New York, NY) software packages were used to create a novel transhumeral CRS style socket insert design with both compression (high pressure) and release (low pressure) lattice cushioning regions. In Figure 3(c), the pink areas correspond to higher density lattice regions, thus serving the function of compression areas of the socket, while blue areas correspond to lower density lattice regions serving the function of release areas of the socket. The gray areas serve as intermediate density lattices that bridge the high-density and low-density regions. Ideally, the final printed socket will require only 1 type of lattice structure with variable density that when altered can meet requirements for either compression or release areas based on the density assigned to the structure in each area. The socket is designed to be printed in biocompatible flexible thermoplastic polyurethane and to have a soft, breathable perforated “skin” layer which contacts the residual limb on one side and the variable density lattice cushion on the other. This single piece should lock into the rigid outer prosthetic upper arm encasement during use and be removable for cleaning to prevent infection or skin irritation.¹⁷ In most transhumeral prostheses, harness systems provide much of the suspension. CRS style sockets with distributed compression areas can help facilitate this suspension and enhance rotational stability.¹⁸ The increase in air flow and subsequent reduction of sweating enabled by a socket with perforated “skin” and open lattice structures may additionally be beneficial for user comfort.^19,20 With a lower limb lattice-modified CRS socket, suspension often relies more on the connection between a user's skin and the device. In this case, use of a closed “skin” lattice socket interface with a liner would be more appropriate.

Lattice samples

Lattice sample cubes with side dimensions based on test rig manufacturer recommendation of 5.08 cm (2 in) and bounding box volumes of 131.1 cm³ (8 in 3) were designed using nTop software and manufactured on Ultimaker 2 Extended+ (Ultimaker B.V., Netherlands) fused filament 3D printers. Samples were printed without support materials, as printing materials were nondissolvable and thus could not be removed from the complex inner cavities of the lattice structures. Leaving them inside would have compromised the intended target for stiffness testing within the sample. Thresholds for lattices deemed adequate to serve as compression or release areas within a modified CRS socket were set to a minimum of 65 N at 1 cm (0.39 in) displacement (>25,187.55 Pa compressive stress at 19.69% strain) and a maximum of 10 N at 3 cm (1.18 in) displacement (<3875.01 Pa compressive stress at 59.06% strain), respectively. It should be noted that compression area thresholds were established based on an intentionally challenging functional transhumeral use case, as was detailed by Myers et al, rather than on thresholds that unequivocally prevent tissue damage.²¹ This is because no such universal thresholds exist that will apply to every person. The risk for tissue damage is based on the interaction of several factors such as applied loads (their magnitude, duration, and direction) and a number of patient-specific conditions such as the age of an individual, the amount of tissue between the bone and the surface of skin, whether that tissue is muscle or fat, skin elasticity, and skin moisture levels.^1,2,22,23 In practice, determination of levels of compression to place on tissue within specific areas of the socket will be based on a clinician's expertise and the patient-specific conditions presented.

Lattice cushioning samples were printed using black Essentium 80A-Z flexible thermoplastic polyurethane. The printing profile for these samples is detailed in Appendix A (http://links.lww.com/POI/A278). Compression testing was performed on samples using 2 types of unit cells: (1) offset diamond and (2) re-entrant auxetic. The offset diamond unit cell geometry was designed to be bending-dominant according to the Maxwell Stability Criterion for 3D structures, and the re-entrant auxetic unit cell was created to exhibit a negative Poisson ratio when compressed as described by Myers et al.²¹ The intent was to ensure containment of materials within the original sample volume and thus, in practice, prevent compression in 1 area of the socket from affecting the stiffness in neighboring areas of the socket. Each lattice sample incorporated 1 of 8 design alterations that regulated the effective density and level of firmness in each sample: (1) blend radius 0.0 mm; (2) blend radius 0.5 mm; (3) blend radius 1.0 mm; (4) thickness 0.4 mm; (5) thickness 0.5 mm; (6) number of unit cells in xyz: 3; (7) number of unit cells in xyz: 4; and (8) number of unit cells in xyz: 5 (Figure 4). Changes in blend radius were designed to alter only the connection points between unit cells, to stiffen only the lattice areas where most movement would be regulated during compression. A change in thickness refers to the thickness of all beams or walls of the lattice structure, and the number of unit cells in xyz describes the number of unit cells used to span the length, depth, and height of the sample cube. Each sample was printed with a smooth, thin, flexible, perforated top surface to mimic the surface of the socket where it would be in contact with the skin.

Figure 4.

Image of the 2 lattice unit cell types tested in the study. Examples of the design alterations to lattice density performed on the offset diamond lattice type and an image of a printed sample being compression tested are displayed.

Samples were compression tested using an Instron 5542 machine paired with BlueHill 2 software (Instron, Norwood, MA). The effect on stiffness as a function of lattice type and density alteration under 3 loading conditions was recorded: the force required to displace a sample by (1) 1 cm, (2) 2 cm, or (3) 3 cm in compression.

Each of the tested samples contained a uniform lattice structure throughout. Three replicates of 16 sample conditions were tested at each loading displacement setting for a total of 144 compression conditions being tested. After compression testing was complete, intermediate lattice structures were developed by gradually tapering lattice density from the conditions leading to the firmest compression area value (highest stiffness samples) to the best performing release condition settings linearly within a single sample.

Results

Results indicated the widest range of change in stiffness occurred using an offset diamond lattice type paired with the blend radius method of density alterations, producing the only samples meeting high stiffness criteria set for compression areas of the socket. The highest stiffness sample recorded required 94.83 N force to compress the sample to 1 cm displacement (36,745.15 Pa compressive stress at 19.69% strain). This exceeds the force required to meet the compression area threshold by 29.83 N. None of the tested samples satisfied the low stiffness criteria set for release areas (Figure 5(a)). The lowest stiffness sample recorded required 22.61 N to compress to 3 cm displacement (8762.28 Pa compressive stress at 59.06% strain), which is 12.61 N more than the maximum allowable force to meet the requirement for release areas. In other words, samples meant to be firm were firm enough to serve as compression areas in the socket, but samples meant to be very soft were not flexible enough to provide regions for soft tissue release. Transitional density lattices which taper linearly from conditions that produced lattice samples with the highest stiffness on one end of the sample to conditions which produced lattice samples with the lowest stiffness on the other end were produced (Figure 5(b)).

Figure 5.

(a) Blend radius design alterations produce samples with the widest range of firmness/stiffness levels and meet thresholds set for compression areas of the modified CRS socket design. No samples meet requirements set for release areas. (b) Samples of blend radius settings creating low-density (A), high-density (B), and transitional-density lattice (C).

Conclusions

In this study, samples were produced that met criteria for socket areas meant for tissue compression. While some samples far exceeded criteria set for compression areas, the minimum level that meets the criteria for firmness of compression areas would be considered the most optimal toward user comfort. None of the samples tested met the criteria set for release. Given these findings, if the 2 lattice types used in this study were the only types of lattice structures to be tested, the design for future modified CRS socket models would be altered to reintroduce open areas for release of compressed soft tissue. In this case, lattice density would decrease significantly toward the opening and would thus serve to gradually taper the shear stress experienced throughout the tissue as it began to emerge. It is possible the greater sensitivity of the offset diamond/blend radius conditions over all other tested conditions can be attributed in part to its intentionally bend-dominant design, as compared with the hybrid beamed and walled structure of the re-entrant auxetic lattice. This likely contributed to the former's capacity to achieve lower stiffness outcomes in comparison with the latter. Thicker blend radii may impede the bend-dominant quality the geometry displays at a uniform thickness, thus allowing for greater stiffnesses to be achieved.

The testing procedure described in this study could be applied to evaluate additional types of lattice structures beyond those assessed here for their adequacy to serve as compression and release areas in future iterations of density-graded lattice modified CRS style sockets. Beyond this, assessment of which characteristics of each unit cell structure have the greatest impact on stiffness would be useful.

Lattice structures transitioning from firmest density conditions to best-performing low-density lattice conditions within a single sample were produced, offering promise for mitigation of soft tissue damage through minimization of pressure gradients that can exacerbate shear stress. As a point of discussion, there is an assumption in this study that the potential clinical benefits of a CRS socket would be preserved, even after a graded variation of pressure was applied between different areas of the socket. The ability to create controlled transitions in stiffness implies that clinicians could use a density-graded lattice padding strategy to create highly targeted areas of firmness and softness throughout any type of socket. Pressure-tuned padding could be used to accommodate areas of bony prominences, neuromas, areas intended to offload pressures or redistribute weight, or any number of other anatomical or physiological features, while these gradual shifts in stiffness enhanced tissue safety overall. Implications for use of targeted variable firmness padding could extend to several types of orthotics, as well, including spinal orthotics. Beyond the functional targeted stiffness needs required in such devices, comfort and sweating are often issues. A breathable, softer interface may enhance the user experience while continuing to offer rehabilitative support where needed.

One of the limitations of this study was that the loading used normal forces only. The intent was to quickly screen out lattice structures that would have no potential to serve as either compression or release areas of a modified CRS socket. In future testing, were structures identified that could successfully meet such criteria, testing that included the quantification of shear loading would be a useful way to simulate various use scenarios. Another limitation of the study is that structures tested were created using only a low-cost planer fused filament fabrication printing process. Mechanical properties of such parts can vary based on print process parameters and print orientation. While compression tests conducted in this study were nondestructive to the test samples, it is likely that since the lattice beams created with this process were composed of several printed layers, if longer term cyclical loading were applied to similar samples over time, partial breakage may be possible. Were a nonplaner process used instead to create lattice beams composed of a single layer, as in the research of Poddar, such lattice beams may be more resistant to damage if exposed to extended cyclic or multidirectional fatigue loading conditions.²⁴

An additional area for potential exploration is examining whether a flexible lattice modified CRS style socket might prove beneficial in helping to mechanically compensate for issues related to minor decreases in residual limb volume or allow for more comfortable accommodation of minor increases. The compressive nature of such a socket may make it possible for minor daily volume deviations to occur without sacrificing stability of the prostheses on the residual limb.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported, in part, through funds from the Orthotic and Prosthetic Education and Research Foundation, Inc. (OPERF).

Declaration of conflicting interest

The author(s) disclosed no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this article is available in this article. Direct URL citation appears in the text and is provided in the HTML and PDF versions of this article on the journal's Web site (www.POIjournal.org).

Supplementary Material

poi-49-76-s001.docx ^{(11.9KB, docx)}

poi-49-76-s002.pdf ^{(329.1KB, pdf)}