Explanation of how Device Functions 5

Manufacturing Process 6

Materials used in Device 6

Mechanical Assessment 9

Shore Hardness 9

Density 11

Compression Modulus 13

Energy absorption 15

Dynamic Mechanical Analysis 17

Martindale Abrasion 18

Biocompatibility assessment 20

Perspiration Resistance 20

Cytotoxicity assay 21

References 25

Background

Foot Orthoses (FOs) are used for the treatment of varied foot conditions which disrupt the ‘normal’ form or anatomy of the foot. When foot bones and muscles are out of this normal alignment the loads which are applied to them can cause further disruption within the foot causing damage to surrounding tissues and result in pain (Donatelli, 1987). FOs to correct these problems are currently available with different levels of customisation; prefabricated off the shelf FOs, FOs which are customizable by a clinician or custom-made FOs designed based on the shape of the end users foot (Williams et al., 2010).

Traditional customised FO techniques use vacuum forming of materials onto a cast representation of the patient’s foot or via milling of EVA sheets based on a scan of the foot geometry. These methods require significant post adaptation to incorporate custom features (Telfer et al., 2012). Additive manufacture (AM) provides the capacity to build in customisation with both geometric and material combinations and minimal post processing (Salles and Gyi, 2012). This study will use foot scan data to produce geometrically identical FOs with both EVA milled and AM methods to allow a direct comparison of the effect of production method.

The use of gait analysis for the characterisation of FOs effects is well demonstrated in the literature (Mayich et al., 2014; Mills et al., 2010). Although comparisons of two FOs produced with identical geometries have not previously been conducted the outcome variables outlined in the study will provide a comprehensive dataset to assess equivalence between the current practice (EVA milled) and the novel production method (additive manufacture).

Foot Orthotic Effectiveness

Literature illustrates that the pressure reduction effects of orthotics average 20% +/- 10% (1SD) (e.g. table below).

Concepts and Designs

Generic orthotic design and process

The generic design to be used in the clinical trial is a 6mm thick orthosis which has a top surface fitted to surface which will be in contact with the foot. To generate a custom fit for each patient a 3D digital foot scan will be taken for each foot (left and right) using a Foot 3D system (INESCOP, Spain). This generates 3D models of the feet which are exported in STL format. A CAD package (Rhinoceros 5, McNeel, USA) with insole design software (Custom 3Din, INESCOP, Spain) is then used to digitally mould the top surface of the orthotic insole to the foot scan data, generating a patient specific insole model.

The Custom made orthotic design is based on clinical prescription processes in place at the orthotics department of the East Lancashire Hospitals NHS Trust. For the diabetic foot this involves the use of custom made orthoses which are shaped to fit the patient’s foot. This has been shown to have a positive significant effect on the load distribution across the foot and the reduction of risk for ulceration in the foot within multiple high quality research studies (Zequera et al., 2007; Zequera et al. 2010; Yuk San Tsung et al., 2004).

Prior to manufacture the orthotic insole model is converted into solid parts by closing all open surfaces. Each part is then inspected using a fault checking software designed for additive manufacture (Netfabb basic, Netfabb, Germany). Parts are assigned a material code to allow appropriate material selection prior to printing. A single insole may consist of multiple parts, these are not physically distinct parts but instead represent areas of the insole which can be produced in different materials. Because of this it is necessary that all parts fit perfectly together within the final assembly model of the orthotic insole. Assembly files are loaded into the Objet 500 Connex 3D Printer (Stratasys, USA) and materials are assigned to insole parts based on file name coding.

Explanation of how Device Functions

Total contact insoles give a greater level of fit to the bottom surface of the foot throughout the stance phase of gait, allowing load to be distributed more evenly. Improved support and load distribution permits more effective offloading of regions which are at risk from peak plantar pressures. This combined with adaptations to the orthotic insoles such as increasing arch support and cut-outs halt the degradation of at-risk tissue in the foot and prevent the progression of pathologies such as diabetes. Current orthotic treatment methods adjust the level of support primarily by changing the level of fit to the foot. Although this is effective at reducing pressure it can require increasing the thickness of orthotics to such a degree that bespoke foot wear is required to accommodate their use. Additive manufactured orthotics allow a second method of adjusting the level of support by changing the material properties across the insole. For example where traditionally the arch would be increased in height it can instead be increased in hardness to provide equivalent support without increased orthotic thickness. The potential to reduce orthotic thickness whilst maintaining effective pressure reduction may reduce the requirement for costly orthotic

Manufacturing Process

The devices will be produced by additive manufacture using an Objet 500 Connex 3D Printer (Stratasys, USA). The Objet 500 is a multi-material 3D printer which utilises dual-jetting technology to combine two base polymers, in specific concentrations, as composite digital materials. This permits a spectrum of stable materials, varied in material properties, to be printed within a single product. Replacing the need for multiple and costly post production processes currently used to provide varied support or padding within an orthotic insole. Layers of liquid photopolymer are deposited in 30 micron print layers by the print head, in a similar process to that used by inkjet printers. These layers are then cured using a UV light which chemically bonds the liquid to form solid layers. The layers build up with each pass of the print head to form the final product which requires no further curing. The Default settings on the Objet500 machines will be used for production, following the maintenance/calibration guidelines of the machine manufacturer to ensure consistent quality of production. Following these procedures will ensure that operating temperatures, UV curing and material deposition are all consistent. In order to ensure correct digital material selection, orthotic files will be coded and given an ID marking. This code will be produced by the by the clinician when the orthotic insole model is created. When the material is assigned by the technician, they can refer to the file name. This can then be checked before the parts are run, with each digital build tray file saved to a server for any future reference.

Materials used in Device

FDM-TB

A rubber like material with tear resistance, tensile strength, hardness and elongation at break tailored for applications requiring non-slip or soft surfaces.

FDM-VW

A rigid opaque material with high dimensional stability for applications requiring high detail control.

Material Blends used for Orthotic Insole Production

Values are approximate based on machine material usage (this does not account for material loss)

Mechanical Assessment

Material, Mechanical and Biological tests were conducted at two sites, The University of Salford, UK and The Institute for footwear and related technologies (INESCOP), Spain. Testing at the University of Salford testing was conducted by Dr Daniel Parker Testing at INESCOP was overseen by Dr Cristina Llobell Andrés

Shore Hardness

Durometer type A with; a pressure foot of 18 mm ±0,5 mm and a central hole of diameter 3 mm ±0,1 mm; an indenter formed from a hardened-steel rod of diameter 1,25 mm ±0,15 mm for type A durometers; An indicating device calibrated directly in terms of units ranging from 0 for the maximum protrusion of 2,50 mm ±0,02 mm to 100 for zero protrusion obtained by placing the pressure foot and indenter in firm contact with a suitable flat, hard surface (e.g. glass); A calibrated spring: This is used to apply a force, F, expressed in millinewtons, to the indenter in accordance with the following equation:

All test pieces were conditioned immediately before testing for a minimum period of 24h at standard laboratory temperature (23 °C ± 2 °C, (50 ± 5) % relative humidity) specified in EN 12222. The same temperature conditions were used throughout any single test or series of tests intended to be comparable.

General: The test piece was placed on a flat, hard, rigid surface (e.g. glass). Pressure was applied, to the test piece, with the durometer foot, as rapidly as possible but without shock, keeping the foot parallel to the surface of the test piece and ensuring that the indenter is normal to the materials surface. Test time: A force sufficient only to obtain firm contact between the pressure foot and the test piece was applied. The standard test time is 3 s for vulcanized rubber and 15s for thermoplastic rubber. Due to the likeness of the materials used to thermoplastic rubbers the 15s time was used for this testing.

Five measurements of hardness were taken at different positions on the test piece at least 6 mm apart. The median value was then determined.

The materials measured are within the range of traditionally used materials for the production of orthotic insoles. The capacity to produce this wide range of material hardness in a single production process is expected to be of benefit for the production of orthotic insoles.

Density

Analytical balance: Accurate to ± 1 mg. Balance pan straddle: Convenient size to support the beaker and permit determination of the mass of the test piece in water (for method A). Beaker: 250 cm3 capacity (or smaller if necessitated by the design of the balance) (for method A)

The test piece was weighed in air and then when immersed in freshly boiled and cooled distilled water contained in the beaker by means of a fine filament hooked onto the balance.

Three test pieces were assessed and the mean value is reported. The density, expressed in megagrams per cubic metre, is given by the formula:

ρw is the density of water; m1 is the mass of the rubber, determined by weighing in air; m2 is the mass of the rubber less the mass of an equal volume of water, determined by weighing in water, both at standard laboratory temperature.

This method is accurate to the nearest 0,01 Mg/m3. For most purposes, the density of water at standard laboratory temperature may be taken as 1,00 Mg/m3. However, for precise work, a factor to take account of the density of water at the test temperature shall be used.

Material density increased with increasing percentage of FDM-VW material. Total difference between M1 and M7 was small and is not expected to restrict material selection or use in when orthotic insole is designed.

Compression Modulus

The Instron test machine was evaluated to conform with Class 1 of ISOs 7500-1 and 5893 performing measurement of force to accuracy of ±1 N, and also of measuring the test piece thickness under load with an accuracy of ± 0.1 mm. Compression of the test piece was between a support surface and a compression plate; the compression plate had a controllable uniform relative rate of motion in the vertical direction of 5 to 100mm/min ±20%. The supporting surface was a smooth, flat, horizontal and rigid surface, larger than the test piece. The compression plate was larger than the test piece, it was plane and smooth, but not polished, and it was maintained parallel to the supporting surface

Materials were tested less than 72 h after manufacture Samples and test pieces were protected from light as completely as possible during the interval between production and testing. Prior to the test, the test pieces were be conditioned, undeflected and undistorted, for at least 16 h at 23 °C ± 2 °C, (50 ± 5) % relative humidity.

The dimensions of the test piece were measured prior to testing to calculate the area of the load-bearing face. The test piece was positioned such that the applied force acted along the centre line of the test machine, Compression was controlled at 5 (±1) mm/min by means of the compression plate until the compression strain applied equalled that specified in the material specification. Decompression of the test piece was conducted at the same rate until the separation between the compression plate and the base plate was equal to the initial test piece thickness. This was repeated four times and on the fourth compression cycle the force in Newtons and displacement in mm were recorded.

Three test pieces were assessed, and the mean value is reported. The compression modulus was calculated from the 4th loading cycle at 10% and 20% strain. The compression modulus is given, in megapascals, by the formula

A is the original cross-sectional area, in square millimetres, of the test piece;

Stiffness at 10% and 20% strain increased with increasing Vero White content. Greater

resolution of material stiffness is available at lower end due to exponential curve.

Energy absorption

Three test pieces were assessed and the mean value is reported. The hysteresis was calculated from the 4th loading cycle from the loading and unloading curve data. The hysteresis is given as a ratio between energy used in loading and energy used in unloading.

Energy lost within the materials increased as concentration of FDM-VW material increased. The ability to identify materials with high hysteresis is expected to improve the design of orthotics for the absorption of energy during stance, potentially reducing the work required to dampen high energy impacts within the soft tissues.

Dynamic Mechanical Analysis

Specimens of the seven materials of dimensions 35x15x2 mm were tested using a Q800 DMA tester (TA Instruments, USA) by heating the sample from -80 to 80ºC with a heating rate of 5ºC/min, a frequency of 1 Hz, and amplitude of 5µm.

In this study, the dynamic mechanical properties of the seven materials under test as a function of the temperature were evaluated. Bending deformation mode was used by a single cantilever where the sample was clamped at both ends and flexed at one end.

The evolution of the storage (E’) and loss (E’’) moduli, as well as the stiffness and Tan δ, as a function of the temperature were determined.

The Dynamic Mechanical Analysis (DMA) is considered a suitable technique for the characterisation of the materials M1 to M7. The use of the single cantilever clamp is appropriate to determine some mechanical properties of these materials, such as, stiffness or damping, reported as modulus and tangent delta, respectively. The highest modulus values (E’ and E’’) correspond to sample n7, in accordance with its stiffness, while the lowest values of modulus and stiffness are related to sample n1 (more flexible). Regarding tangent delta results, a maximum value appears (related to molecular structure changes) in the tested materials. These maximum values shift to higher temperatures when the stiffness of the material increases. Furthermore, when the flexibility of the materials increases, a second peak appears, indicating probably a polymer blend.

Martindale Abrasion

An Abrasion machine as specified in the standard which utilised the following features:

The test specimens were fitted to the martindale abrasion device and a constant pressure of 12 kPa ± 0,2 kPa was maintained between the specimen carrier and the abradant table. The movement of the device created a Lissajous figure occupying an area of 60 mm ± 1 mm 60 mm ± 1 mm (see Figure below). Each Lissajous figure requires 16 eliptical motions (revolutions) of the test specimen carrier and the speed of operation of the tester was 5 rad/s □□0,4 rad/s 1).

Specimen inspection was conducted under bright indirect light to identify signs of damage. Specimens were compared to untested blanks of the same material. Any abrasion, pilling and discolouration was recorded with the descriptions; None, very slight, slight, moderate, severe, almost complete, and complete. Any holes or damage to surface layers are also recorded.

The testing bodies’ general recommendations for footwear insoles (end product) are as follows:

(1) The material is excessively soft; it deformed and became worn out around the edge of the clamping ring.

All materials were resistant to wear. The flexibility of materials M1-M4 was high which resulted in deformation during testing, which may affect performance.

Biocompatibility assessment

A biological evaluation of the materials used in the medical device was conducted due to the lack of information which currently exists regarding their suitability for use in a medical device. The testing was conducted by the Institute for footwear and related technologies (INESCOP), Spain. Testing at INESCOP was overseen by Dr Cristina Llobell Andrés

Perspiration Resistance

histidine monohydrochloride monohydrate: 5,00 g sodium chloride: 5,00 g disodium hydrogen orthophosphate dihydrate: 2,50 g Conditioning and environment After preparation, the solution was brought to pH 8 with 0,1 M sodium hydroxide solution.

The test pieces were subjected to successive ageing cycles applying the following treatments:

The table below shows the average results for dimensional and weight changes obtained after five ageing cycles. Testing bodies general recommendations for footwear insoles (end product) are as follows:

The materials were within acceptable levels of resistance to perspiration. The materials all demonstrated high dimensional stability and did not display changes to mechanical integrity.

Cytotoxicity assay

Samples were sterilised in an autoclave for 20 min at 121^oC prior to testing to remove effects of microbiological contamination.

The culture medium and culture conditions used are the standard for this cell type (DMEM medium plus L-glutamine, incubation at 37ºC, 5% CO2).

For the cytotoxicity assay, HaCaT cells were seeded in 60 mm diameter Petri dishes and cultured until 80% confluence. Samples were assayed in triplicates, and a negative and a positive control were added, also in triplicates.

To perform the test, the culture medium was removed, and fresh medium was added. Each sample was carefully placed in the centre of the corresponding Petri dish, taking care not to move the specimens, thus preventing physical trauma to the cells.

According to the standard, the sample has to cover about 1/10 of the useful surface growth area of the culture plates, which is considered 21cm² for the 60mm Petri dishes.

For the qualitative evaluation, samples were removed, and cells were examined microscopically. Changes in general morphology, vacuolisation, detachment, cell lysis and membrane integrity were assessed. Samples were graded according to the following reactivity grades:

Vital staining was used to determine quantitatively the reduction of cell viability, by means of cell counting. The vital staining used was a combination of SYTO9 and propidium iodide. SYTO 9 stains all cells (visible as bright green staining under fluorescence light), allowing the visualisation and evaluation of their morphology. Propidium iodide allows viable cells to be differentiated from dying or dead cells, because while viable cells have intact cellular membranes, able to exclude the molecule, damaged cells cannot exclude the molecule that stains cell nuclei; therefore, cells are observed in bright red under the corresponding fluorescence filter.

After removing the sample, cells were examined under a fluorescence microscope. By using the corresponding fluorescence filter, cells in representative fields of the Petri dishes were counted, and dead or dying cells were referred to the number of total cells.

Testing bodies general recommendations for footwear insoles (end product) are as follows:

Quantitative reduction of cell viability by more than 30% is considered a cytotoxic effect

The qualitative assessment was performed assigning values to the effect of the assayed material over the cultured cell, ranging from “0” (non reactivity) to 4 (severe).

After the contact period almost all the cells appeared vacuolisated, in the cells under the sample as well as in the rest of the culture plate.

A. Healthy cells (control), as seen under transmitted light. B. Cells exposed to FDM-TB for 24h. Arrows show: 1= Dividing cell, 2= Detaching cell, 3= Apoptotic vesicles, 4= Vacuoles, 5= Detached cell.

After the contact period, a reactivity ranging between light and slight was observed, which mainly affected the area under the sample.

A. Healthy cells (control), as seen under transmitted light. B. Cells exposed to FDM-VW for 24h. Arrows show: 1= Dividing cell, 2= Detaching cell, 3= Apoptotic vesicles, 4= Apoptotic cell with cytoplasmic blebbings.

The additive materials are not suitable for direct contact with skin and should always be used in combination with a top cover material or coating. This aligns with the standard practice at East Lancashire Hospitals NHS Trust, where a top cover material is routinely fitted to orthotic insoles used for diabetes treatment.

References

Donatelli, R.A., 1987. Abnormal biomechanics of the foot and ankle. J. Orthop. Sports Phys. Ther. 9, 11–16.

Mayich, D.J., Novak, A., Vena, D., Daniels, T.R., Brodsky, J.W., 2014. Gait Analysis in Orthopedic Foot and Ankle Surgery—Topical Review, Part 1 Principles and Uses of Gait Analysis. Foot Ankle Int. 35, 80–90.

Mills, K., Blanch, P., Chapman, A.R., McPoil, T.G., Vicenzino, B., 2010. Foot orthoses and gait: a systematic review and meta-analysis of literature pertaining to potential mechanisms. Br. J. Sports Med. 44, 1035–1046.

Salles, A.S., Gyi, D.E., 2012. The specification of personalised insoles using additive manufacturing. Work 41, 1771–1774.

Telfer, S., Gibson, K.S., Hennessy, K., Steultjens, M.P., Woodburn, J., 2012. Computer-aided design of customized foot orthoses: Reproducibility and effect of method used to obtain foot shape. Arch. Phys. Med. Rehabil. 93, 863–870.

Tsung B Y, Zhang M, Mak A F, Wong M W. 2004. Effectiveness of insoles on plantar pressure redistribution. J Rehabil Res Dev 41(6A):767-74.

Williams, A.E., Nester, C., Mathieson, I., 2010. Pocket Podiatry: Footwear and Foot Orthoses, 1e, 1 edition. ed. Churchill Livingstone.

Zequera M, Stephan S, Paul J. 2007. Effectiveness of moulded insoles in reducing plantar pressure in diabetic patients. Conf Proc IEEE Eng Med Biol Soc 2007:4671-4

Zequera M, Solomonidis S. 2010. Performance of insole in reducing plantar pressure on diabetic patients in the early stages of the disease. Conf Proc IEEE Eng Med Biol Soc. 2010:2982-5.

Study	Design/location	%Pressure relief achieved
Poon	Metatarsal dome	-16.5
Hodge	Orthosis with dome	-15.7
	Orthosis with bar	-20.5
Holmes	Met 1	-11.1
	Met 2	-27.6
	Met 3 & 4	-22.7
	Met 5	-16.7
Jackson	Orthosis with dome	-11.8
	Orthosis with bar	-21.3
Lee	Orthosis, dome 10mm proximal	-8.6
	Orthosis, dome 5mm distal	-16.8
	Orthosis with bar	-10.2
Kang	Orthosis with dome	-11.9
Lott	Orthosis with dome	-18.8
Lin	Orthosis with cutout at region of interest	-42.9
	Orthosis with cutout plus arch support	-48.3
Duffin	Cushioning	-21.8
(callus study)	Custom orthosis	-17.7
	Custom orthosis with cushioning	-27.67
	MEAN pressure reduction (SD)	-20.4 (10.4%)

Sample	FDM-TB	FDM-VW
M1	100	0
M2	94.3	5.7
M3	90.5	9.5
M4	85.6	14.4
M5	80.8	19.2
M6	74.1	25.9
M7	63.4	36.6

Sample	Shore Hardness (A)
M1	26.5
M2	35.2
M3	42.2
M4	49.1
M5	57.2
M6	67.0
M7	79.5

Sample	Density (g/cm³)
M1	1.130
M2	1.135
M3	1.140
M4	1.144
M5	1.150
M6	1.154
M7	1.161

Sample	Peak Strain	Peek Strain	Stiffness 01	Stiffness 02
M1	0.242	0.320	1.131	1.673
M2	0.243	0.502	1.871	2.681
M3	0.243	0.722	2.770	3.922
M4	0.240	1.142	4.642	6.371
M5	0.233	1.734	7.298	10.432
M6	0.215	2.897	14.440	20.530
M7	0.187	5.984	39.091	60.355

Sample	Load Energy	Unload Energy	Hysteresis (%)
M1	0,034	0,029	15,988
M2	0,051	0,042	18,194
M3	0,072	0,054	24,202
M4	0,110	0,074	32,610
M5	0,161	0,090	43,809
M6	0,249	0,102	58,878
M7	0,447	0,127	71,526

	Direction L	Directi on T	(%)
M1	-0.8	-0.4	-2.5
M2	0	-0.4	-2.3
M3	0	0	-2.0
M4	0	0	-1.5
M5	0	0	-1.4
M6	0	0	-1.2
M7	0	0	-0.5

Table of Contents