Materials Science and Engineering C 35 (2014) 220–230

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

Microstructure and mechanical properties of synthetic brow-suspension materials Kyung-Ah Kwon a, Rebecca J. Shipley a, Mohan Edirisinghe a,⁎, Daniel George Ezra b, Geoffrey E. Rose b, Andrew W. Rayment c, Serena M. Best c, Ruth E. Cameron c a b c

Department of Mechanical Engineering, University College London, Torrington Place, London WC1E 7JE, UK Moorfields Eye Hospital, UCL Institute of Ophthalmology NIHR Biomedical Research Centre for Ophthalmology, 162 City Road, London EC1V 2PD, UK Department of Materials Science and Metallurgy, Cambridge Centre for Medical Materials, University of Cambridge, Pembroke Street, Cambridge CB2 3QZ, UK

a r t i c l e

i n f o

Article history: Received 2 August 2013 Received in revised form 2 October 2013 Accepted 29 October 2013 Available online 11 November 2013 Keywords: Ptosis Brow-suspension Materials Microstructure Mechanical properties

a b s t r a c t Levator palpebrae superioris (LPS) is a muscle responsible for lifting the upper eyelid and its malfunction leads to a condition called “ptosis”, resulting in disfigurement and visual impairment. Severe ptosis is generally treated with “brow-suspension” surgery, whereby the eyelid is cross-connected to the mobile tissues above the eyebrow using a cord-like material, either natural (e.g. fascia lata harvested from the patient) or a synthetic cord. Synthetic brow-suspension materials are widely used, due to not requiring the harvesting of fascia lata that can be associated with pain and donor-site complications. The mechanical properties of some commonly-used synthetic brow-suspension materials were investigated — namely, monofilament polypropylene (Prolene®), sheathed braided polyamide (Supramid Extra® II), silicone frontalis suspension rod (Visitec® Seiff frontalis suspension set), woven polyester (Mersilene® mesh), and expanded polytetrafluoroethylene (Ptose-Up). Each material underwent a single tensile loading to the failure of the material, at three different displacement rates (1, 750 and 1500 mm/min). All the materials exhibited elastic–plastic tensile stress–strain behaviour with considerable differences in elastic modulus, ultimate tensile strength, elastic limit and work of fracture. The results suggest that, as compared to other materials, the silicone brow-suspension rod (Visitec® SFSS) might be the most suitable, providing relatively long-lasting stability and desirable performance. These findings, together with other factors such as commercial availability, cost and clinical outcomes, will provide clinicians with a more rational basis for selection of browsuspension materials. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Ptosis is a condition characterised by drooping of one or both upper eyelids, resulting in visual impairment and disfigurement; as ptosis can be congenital or acquired, it impairs the quality of life of many people from infancy to the aged. Ptosis arises from many causes, such as eyelid tumours, injury to the oculomotor nerve, or maldevelopment of the levator palpebrae superioris muscle (LPS) that is responsible for lifting the upper eyelid. Amongst the various methods for treatment of ptosis, “brow-suspension” surgery is usually performed where there is poor or absent LPS function: during the surgery, the upper eyelid is internally attached to the frontalis muscle using brow-suspension materials to aid the poorly-functioning LPS [1–3]. Fascia lata is commonly used as a brow-suspension material due to its high efficacy and low rate of complications (such as granuloma formation, infection or long-term extrusion) as compared with other

⁎ Corresponding author. E-mail address: [email protected] (M. Edirisinghe). 0928-4931/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2013.10.031

suspensory materials [4–7]. However, harvesting the fascia lata requires an extra incision on the patient's leg, that may result in early postoperative leg pain or impaired gait, and carries a risk of persistent scarring [8]; the extra surgery also increases the operative time and the risk of infection, and harvesting sufficient fascia lata may be problematic in small children [4,6]. These limitations with the biological fascia lata have led to the use of alternative, synthetic filamentous materials — such as silicone rod [9–12], polyester mesh [13–16], expanded polytetrafluoroethylene (ePTFE) strip [17–19], monofilament polypropylene [20,21] and braided polyamide [22–24]. These synthetic materials, being readily available and easy to handle, are used widely in a brow-suspension surgery — either as a permanent solution, or as a temporary suspensory material in very small children. When implanted, any brow suspension material will be subjected to a rapid tensile stretch during blinking and therefore the mechanical characteristics of such materials are important, as described in the examples below: • Stress–strain relationship: the stress or strain induced by a blinking action must lie within the elastic region of stress–strain curves for

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

the material so that the material can return to its original configuration after the stretch during blinking. • Tensile strength and work of fracture, the latter is related to the “toughness” of the material: this property provides an indication of how much strength/load the material can endure before its fracture — which might be practically useful during knot-tying. • Elastic modulus or rigidity: to comprehend the degree of pliability of the material for its handling and control during the surgical procedure and knot security. Many papers have reported the mechanical properties of surgical suture materials such as Gore-Tex®, Prolene®, Mersilene® and monofilament nylon (Ethicon®) [25–32], as they are widely used in several branches of surgery — such as tendon repairs and cardiovascular, gynaecological or orthopaedic surgeries. To date, however, there have been no reported comparisons of the mechanical properties of commonly-used brow-suspension materials. Furthermore, most reported tensile tests for suture materials have been performed at a strain rate ranging from 3 to 50 mm/min; considering that the peak speed during blinking can reach 15 000 mm/min [33], the reported values of tensile strength and elastic modulus might not represent the actual ones in this particular usage. The surgeon's choice of brow-suspension material is based upon various considerations, such as the patient's age and overall health condition, the intended duration for the implant, the professional experience during previous ophthalmic procedures, and some knowledge of the biological and physical properties of the material. The material selection is a key factor in achieving a stable and long-lasting lift of the dysfunctional upper eyelid and, in order to assist surgeons in their choice of material, the mechanical properties of suspensory materials are investigated in this study. This study is focused on five commonly-used synthetic suspensory materials and tensile tests were performed to estimate the stress–strain relationship, ultimate tensile strength, work of fracture and elastic modulus. Three different strain rates, ranging within the capability of the tensile-testing apparatus, were employed to assess the influences of the strain rate on the mentioned mechanical characteristics. In addition, the microstructure of the suspensory materials was examined before and after the mechanical tests. 2. Materials and methods 2.1. Materials The following commonly-used synthetic brow-suspension materials were purchased and used as received: 4-0 monofilament polypropylene (Prolene®; Ethicon Ltd., UK), 3-0 sheathed braided polyamide (Supramid Extra® II; S. Jackson Inc., USA), Silicone frontalis suspension rod (Visitec® Seiff frontalis suspension set (Visitec® SFSS); Beaver-Visitec International Ltd., UK), Woven polyester mesh (Mersilene® mesh; Ethicon Ltd., UK), Expanded polytetrafluoroethylene (Ptose-Up; FCI Ophthalmics Inc., USA). 2.2. Tensile testing Uniaxial tensile tests of the unknotted brow-suspension materials were performed using a testing machine (Hounsfield, Redhill, England) controlled by a computer software (QMat 5.44, SDL Atlas, Rock Hill, SC, USA). Different grips were used to securely hold the materials, depending on their shape. For suture-type materials (Prolene®, Supramid Extra® II and Visitec® SFSS), two polished aluminium cylinders of 10 mm diameter were used in series. Each suture material was passed once around the aluminium cylinders and fixed to metal pegs with a

221

slip knot at each end in order to ensure the secure anchorage of the material during the tensile tests; the slip knot being applied in order to minimise the local stress concentration around the knot during the tests. This mounting set-up could not be used for mesh-type materials (Mersilene® mesh and Ptose-Up), while retaining their original crosssections, since applying slip knots would have caused appreciable deformation of the materials, most likely generating high local stress concentration. Hence, apposing rectangular clamps with rubber contact surfaces were used instead to securely grip the mesh-type specimens, each end of a specimen being mounted by manually tightening the screws. A double-sided adhesive tape was also needed to prevent slippage of the Ptose-Up samples on the rubber surfaces during tests. Once a specimen was mounted onto the testing apparatus with a nominal gauge length of 50 mm for the suture-type materials and 15 mm for the mesh-type materials, the overhead grip was programmed to move upward at a pre-determined displacement rate until the specimen broke. For each material three different displacement rates were investigated: “slow” (1 mm/min), “intermediate” (750 mm/min) and “fast” (1500 mm/min). Due to its substantial capacity for elongation, the Visitec® SFSS silicone rod was tested using a “slow” displacement rate of 5 mm/min and, because of the limited availability of test material, Ptose-Up was only tested at two rates (“slow” and “fast”). Three samples of each suspensory material were tested at the different displacement rates mentioned above, and the tests were conducted at 18.2 ± 0.2 °C with 34.2 ± 0.5% humidity. For each specimen the applied load (N) and the resulting strain (%) were continuously recorded using a 250 N load cell with 0.1 N accuracy and an external laser extensometer (Hounsfield S500), respectively. Small strips of red reflective tape were applied at the top and the bottom of the grips to allow direct strain measurements by the laser extensometer during tensile testing. The load recorded was converted to engineering stress by dividing the load by the initial specimen dimensions. The elastic limit was determined to be the force at which the linear elastic region of force–strain curve finished and the yield tensile strength (σy) was identified with the standard 0.2% offset yield strength; that is, the stress at which the stress–strain curve deviates by a strain of 0.2% from the linear elastic region of the curve, or as the stress at which the stress–strain curve levelled off. The corresponding strain (εy) was taken as the yield tensile strain. The failure load (Fmax) and the ultimate tensile stress (UTS) were determined as the maximum force and the maximum engineering stress reached on the force–strain curve and the stress–strain curve obtained, respectively, and the corresponding strain (%) was taken as the ultimate tensile strain. In addition, the elastic modulus and work of fracture were calculated as the gradient of the stress–strain curve in the initial linear region and the area under the curve up to the fracture point (summing the area of the trapezoids defined by pairs of points), respectively. When the suture-type materials did not break between the aluminium cylinders the data obtained were ignored. Similarly, when the mesh-type materials broke at the edge of the rectangular grip the data obtained were ignored so that the accurate mechanical properties of the materials could be determined. 2.3. Structural characterisation The size (such as the diameter of the cross-section for suture-type materials and the thickness for mesh-type materials) of each suspensory material was measured with an optical microscope (Nikon Eclipse Me600) and a digital calliper (Mitutoyo). The measurements were taken at 3 random places along each specimen and three specimens for each material were analysed. The surface morphology and cross-section of each suspensory material were examined using a scanning electron microscope (SEM, JEOL JSM-6301F). Secondary electron imaging mode was used at an accelerating voltage of 3 kV with the associated software (SemAfore) to collect images. Prior to imaging, all the specimens were sputter coated with

222

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

gold for 120 s at 20 mA. In order to study the cross-section of the materials, the specimens were sectioned with scissors. The SEM was also used to analyse any morphological changes on the fractured surfaces after tensile tests. The specimens of each material tested in slow (1 mm/min) and fast (1500 mm/min) strain rates were collected and examined. 2.4. Statistical analysis The significant difference amongst the characteristic values for the measured properties of suspensory materials was examined using one-way analysis of variance (ANOVA) in the software package Origin 9.0 (OriginLab Corporation). The Tukey, post-hoc test was employed and p b 0.05 was considered to be statistically significant. 3. Results and discussion The surface morphology and cross-section of each brow-suspension material are shown in SEM micrographs (Fig. 1). The morphology and microstructure varied considerably between the materials. Table 1 summarises the microstructure of the commonly-used synthetic browsuspension materials.

Prolene® was shown to be a monofilament with a smooth surface as reported by Karaca and Hockenberger [34] although a few scratch marks and some impurities were observed (Fig. 1a). Its diameter was measured to be 201 ± 2 μm. Fig. 1b shows the cross-section and surface of the Supramid Extra® II suture. It was a multifilament as observed by Kook et al. [23]. Although the cross-sectioning procedure seemed to have caused slight deformation on the filaments, approximately 50 filaments, each having fairly regular diameter of 23 ± 2 μm and circular shape, were observed to be enclosed in a smooth outer sheath. Each filament appeared to stay in place, being evenly distributed within the outer sheath of diameter 235 ± 3 μm since each filament could be distinguished. Some scratch marks and debris were also seen on the sheath surface. The cross-section and surface of Visitec® SFSS, are shown in Fig. 1c, the cross-section appearing to be smooth, but white small regions of size ranging from 50 to 150 nm with average 95 ± 33 nm being observed on the surface at a high magnification. These might be an artefact produced during the processing, possibly owing to a different phase containing silicon or different molecular weight of the silicone polymer chain. The diameter of the rod was measured to be 799 ± 4 μm and a high degree of debris was observed on the surface although no scratch mark were noted. Mersilene® mesh was measured to have a varying width ranging from 6 to 10 mm with thickness of 0.26 mm. Under the

Fig. 1. Scanning electron micrographs of the commonly-used brow-suspension materials (a) Prolene®; (b) Supramid Extra® II; (c) Visitec® Seiff frontalis suspension set; (d) Mersilene® mesh; (e) Ptose-Up.

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

223

Table 1 A summary of microstructure of the synthetic brow-suspension materials (dia. =diameter; ave. = average). Sample

Raw Material

Microstructure

Prolene® Supramid Extra® II Visitec® Seiff frontalis suspension set Mersilene® mesh

Polypropylene Polyamide Silicone

Smooth monofilament with dia. of 201 ± 2 μm ~50 filaments (each dia. of 23 ± 2 μm) enclosed in a smooth outer sheath (dia. of 235 ± 3 μm) Solid rod with dia. of 799 ± 4 μm

Polyethylene terephthalate (PET) polyester Expanded tetrafluoroethylene (ePTFE)

Macroporous structure constructed by ~15 filaments (each dia. of 15 μm) woven into a mesh; in the oval-shaped pores, long axis = 1.40 ± 0.05 mm and short-axis = 0.75 ± 0.02 mm; overall thickness = 0.26 mm Microporous structure having 2 mm width and 112 ± 8 μm thickness; ePTFE fibrils (each dia. of 150 ± 17 nm) adhered to form a fibre (~1.29 ± 0.28 μm) and several fibres are connected into nodes; inter-nodal spacing = 43.4 ± 7.6 μm

Ptose-Up

SEM (Fig. 1d), it was observed that a bundle of approximately 15 filaments, each having a regular circular cross-section (15 μm in diameter), were woven into a mesh, forming oval-shaped pores. This observation is supported by the work of Iglesia et al. [26] and Brun et al. [35]. The longaxis and short-axis of the oval mesh pores were measured to be 1.40 ± 0.05 mm and 0.75 ± 0.02 mm, respectively. Some debris was seen on the filament surfaces as well as in between the filaments. Ptose-Up was measured to be 2 mm wide and 112 ± 8 μm thick although the typical thickness has been reported to be 0.35 mm by the manufacturer [36]. This appreciable discrepancy might be caused during the measurement by the digital calliper since the material was very fragile and porous, and hence the material might have been slightly compressed in spite of the great care during the measurement. The SEM micrographs of Ptose-Up (Fig. 1e) revealed that it has a microporous structure constructed by interconnected nodes and thin fibrils at a high density. ePTFE fibrils of 150 ± 17 nm in diameter appeared to be adhered to each other, forming a fibre of 1.29 ± 0.28 μm in diameter. Several fibres seemed to be connected to nodes, which were on average 43.4 ± 7.6 μm apart longitudinally. Moreover, these interconnected nodes with fibres seemed to form a laminar structure, which appeared to be stacked on top of each other. Unlike Mersilene® mesh, no weaving was observed in Ptose-Up. Fig. 2 represents typical force–strain curves as well as stress–strain curves of the five synthetic brow-suspension materials tested at 1500 mm/min displacement rate. Four of the materials – Prolene®, Supramid Extra® II, Mersilene® mesh and Ptose-Up – exhibited similar force–strain curves, with a steep linear increase in force up to about 50–100% in strain. The Visitec® SFSS, however, exhibited a substantially different tensile behaviour as compared to the other materials, reaching the failure load (Fmax ) at an order of magnitude higher strain (~ 1100% in strain). While Supramid Extra® II had the greatest Fmax , followed by Prolene® and Mersilene® mesh, then Ptose-Up and Visitec® SFSS, both the Supramid Extra® II and the Prolene®

showed the greatest UTS — followed by the others on stress–strain curves with more than an order of magnitude difference in maximum stress between the two groups. The effect of testing (displacement) rate is shown in Fig. 3. The values of the key mechanical properties of these materials determined from the stress–strain curves are summarised in Table 2. It is noteworthy that the coefficient of determination (R2) of the linear regression line to estimate the elastic modulus was 0.99 ± 0.01 for all the brow-suspension materials regardless of the displacement rate. Moreover, apart from Ptose-Up tested at the “slow” rate, all suspensory materials actually broke at the maximum stress value; that is UTS, although the stress–strain curves appeared to continue even after this point, especially for the curves recorded at 750 and 1500 mm/min displacement rates, showing a near-linear line from the UTS to zero (Fig. 3). This near-linear line is the “residual” of the tensile testing process performed at high displacement rates such as 750 and 1500 mm/min. The testing machine had to be manually stopped once the specimen broke hence there was always a time lag between breakage of the specimen and cessation of the stress–strain recording — this resulting in the “residual” points. However, there were no such “residual” points on the stress–strain curve of Ptose-Up tested at 1 mm/min displacement rate, since it actually broke at near-zero stress level: For example, the strain at UTS for Ptose-Up tested at 1 mm/min displacement rate was determined to be 26.8 ± 9.1% (Table 2), but the strain at rupture was determined to be 69.5 ± 18.3%. The stress–strain curves of Prolene®, Mersilene® mesh and Ptose-Up looked similar to those reported by other researchers in the literature [25,27,31,35,36]. Regardless of displacement rates, all the mesh-type materials (Mersilene® mesh and Ptose-Up) exhibited a toe-region up to about 5–20% in strain, followed by a linear-region, then a plasticregion. No such toe-region could be observed on the stress–strain curves of suture-type materials (Prolene®, Supramid Extra® II and Visitec® SFSS). This brief toe-region at the beginning might be due to alignment

Fig. 2. Typical data of brow-suspension materials from tensile tests conducted at 1500 mm/min displacement rate (a) force–strain curves, (b) stress–strain curves; the inset in (b) depicting the stress–strain curves at a smaller scale; Curves: blue — Prolene®, red — Supramid Extra® II, green — Visitec® Seiff frontalis suspension set, purple — Mersilene® mesh, and light blue — Ptose-Up.

224

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

Fig. 3. Stress–strain curves obtained from tensile tests at different displacement rates: (a) Prolene®, (b) Supramid Extra® II, (c) Visitec® Seiff frontalis suspension set, (d) Mersilene® mesh, (e) Ptose-Up; Curves: blue—at 1500 mm/min rate, red—at 750 mm/min rate, and green—at 1 mm/min rate.

of the polyester or ePTFE fibres in the direction of the stretch during the tensile tests because, unlike the suture-type materials, mesh-type materials have a macro- or micro-porous structure with free space for the fibres to move around. In the present work, increasing the displacement rate from 1 to 750, or to 1500 mm/min, significantly increased the elastic limit, UTS and elastic modulus (p b 0.05) for Prolene®, although no such statistically significant difference was found between the results performed at 750 and 1500 mm/min displacement rates. This result agrees well with the findings by Nilsson [37] who investigated the effect of displacement rate on mechanical properties of Prolene® thread by conducting tensile tests at varying rates ranging from 2 to 600 mm/min. Supramid Extra® II also showed the same effect of increasing displacement rate, although no statistically significant difference was found in elastic modulus. Nevertheless, the elastic modulus of Supramid Extra® II tested at 750 or 1500 mm/min displacement rate exhibited a greater mean value than the one tested at 1 mm/min rate. No significantly different values of strain at elastic limit, strain at fracture and work of fracture were

found for both materials. Like Nilsson [37], Catanese III et al. [27] showed that increasing strain rate from 0.05 to 2.4% s−1 increased the yield stress as well as UTS for medical grade tubular ePTFE material. This study also shows a significant difference of these mechanical properties for Ptose-Up, but with the opposite trend; that is, increasing the displacement rate decreased the elastic limit, Fmax, UTS and elastic modulus. The different sample shape might be one of possible reasons for this opposite effect: in this study, Ptose-Up was a mesh-type material with thickness of 0.35 mm and porosity of 50 μm [36] and tightly gripping it could have reduced the thickness, thereby introducing local concentrated stress near the grip edges. Hence, pulling the material at a fast displacement rate might have caused a tear near the grip edges, resulting in the premature failure of the sample. It is noteworthy that about 75% of the Ptose-Up specimens tested at 1500 mm/min displacement rate failed at a site near the grips. Compared with those tested at 5 mm/min, Visitec® SFSS showed significantly greater UTS, strain at rupture and work of fracture for the samples tested at 750 mm/min, but no trend was observed as significantly different results were not

Table 2 A summary of the key mechanical properties of the synthetic brow-suspension materials investigated in this work (the result values are mean ± s.d. and the ultimate tensile strain was the same as the strain at rupture for all the suspensory materials except Ptose-Up tested at 1 mm/min displacement rate). Sample

Strain rate (mm/min)

Elastic limit (N)

Yield strength, σy (MPa)

Yield strain, εy (%)

Failure load, Fmax (N)

Ultimate tensile strength, UTS (MPa)

Ultimate tensile strain (%)

Elastic modulus (MPa)

Work of fracture (MJ/m3)

Prolene®

1 750 1500 1 750 1500 5 750 1500 1 750 1500 1 1500

9.35 15.1 15.2 21.6 23.4 23.7 4.73 4.74 5.39 17.6 24.3 21.3 9.40 6.69

296 478 481 498 538 547 9.43 7.44 11.0 8.57 11.7 11.1 20.4 14.5

96.7 80.8 73.9 82.9 80.6 84.3 814 798 917 48.0 54.7 52.1 20.0 40.9

13.0 16.6 16.2 21.6 23.4 23.7 5.09 6.99 6.47 22.2 25.7 21.3 9.97 6.87

411 524 510 498 538 547 10.2 13.9 12.9 11.1 12.2 11.1 21.6 14.9

160 93.2 79.8 82.9 80.6 84.3 962 1550 1260 67.7 61.1 52.1 26.8 42.9

312 612 694 598 702 694 1.15 1.17 1.19 21.6 25.0 27.7 165 59.7

388 271 226 186 201 214 58.3 132 96.6 4.18 3.82 2.37 8.10 2.45

Supramid Extra® II

Visitec® Seiff frontalis suspension set

Mersilene® mesh

Ptose-Up

± ± ± ± ± ± ± ± ± ± ± ± ± ±

1.43 1.7 1.4 0.3 0.7 0.6 1.23 0.84 0.49 8.4 3.6 5.5 0.89 1.00

± ± ± ± ± ± ± ± ± ± ± ± ± ±

45 54 44 7 15 15 2.46 1.67 1.4 2.69 0.5 0.7 1.9 2.2

± ± ± ± ± ± ± ± ± ± ± ± ± ±

36.5 15.3 18.8 9.2 7.0 9.5 253 230 160 19.5 8.6 3.3 4.4 6.9

± ± ± ± ± ± ± ± ± ± ± ± ± ±

1.2 0.7 0.9 0.3 0.7 0.6 0.97 0.23 0.15 4.2 5.0 5.5 0.35 1.17

± ± ± ± ± ± ± ± ± ± ± ± ± ±

39 21 29 7 15 15 1.9 0.5 0.3 0.4 0.9 0.7 0.8 2.5

± ± ± ± ± ± ± ± ± ± ± ± ± ±

61 6.4 8.5 9.2 7.0 9.5 217 170 271 2.7 4.0 3.3 9.1 6.5

± ± ± ± ± ± ± ± ± ± ± ± ± ±

62 99 152 62 42 90 0.16 0.15 0.13 0.4 4.3 3.6 17 12.1

± ± ± ± ± ± ± ± ± ± ± ± ± ±

197 7 24 7 24 22 24.1 18 29.1 0.51 0.48 0.15 1.60 0.65

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

Fig. 4. Comparison of elastic limit, elastic modulus and work of fracture of commonly-used synthetic brow-suspension materials.

obtained between the samples tested at 1500 and 5 mm/min displacement rates. The difference shown in this result might be due to batchto-batch variation of the material. In addition, Mersilene® mesh was found to have significantly lower strain at rupture and the work of

225

fracture for the samples tested at 1500 mm/min rate as compared to the others tested at 1 and 750 mm/min rates; there was, however, no significant difference in any other mechanical properties estimated in this study (that is, the elastic limit, yield tensile strain, UTS or elastic modulus). Polymers are well known for their viscoelastic properties [25,29,38]. Applying high strain rates, therefore, was expected to affect the elastic modulus and UTS, since such action would most likely favour the elastic nature of the materials. However, no such trend could be observed in this study and only Prolene® and Supramid Extra® II showed such a response. A wider range of strain rate might be required to be able to draw more information on their viscoelastic nature. Considering values of the brow-suspension materials tested at the fastest displacement rate, 1500 mm/min (Table 2), the failure load (Fmax) was the greatest for Supramid Extra® II and Mersilene® mesh, ~ 21–24 N, and the lowest for Visitec® SFSS and Ptose-Up, ~ 7 N. Prolene® exhibited an intermediate value of ~ 16 N as depicted in Fig. 2a. The elastic limit demonstrated the same pattern. However, this changed when the force values were converted to stress values as shown in Fig. 2b: Supramid Extra® II and Prolene® had a substantially greater, namely an order of magnitude greater, ultimate tensile stress (UTS) than the others (510–550 MPa), while the other materials had a UTS of 10–15 MPa. This difference may be primarily due to the fact that the apparent cross-sectional area of each brow-suspension material was used in the estimation of the UTS, not taking into account of any voids or space. Therefore, Mersilene® mesh ended up having a

Fig. 5. Scanning electron micrographs of Prolene® tensile test fracture surfaces: (a) 1 mm/min and (b) 1500 mm/min displacement rate.

226

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

comparable UTS value to Visitec® SFSS and Ptose-Up, which showed the lowest Fmax. Likewise, Prolene® showed a comparable UTS value to Supramid Extra® II, which exhibited the greatest Fmax since Prolene® has a relatively smaller cross-sectional area than Supramid Extra® II. The elastic modulus followed the same pattern as UTS — that is, Supramid Extra® II and Prolene® exhibited the greatest elastic modulus of ~700 MPa while the rest of the materials exhibited a one or two orders of magnitude lower elastic modulus, 30–60 MPa for Mersilene® mesh and Ptose-Up, and 1 MPa for Visitec® SFSS. Regarding the yield strain and strain at UTS, only Visitec® SFSS showed a significantly different value of ~1000% compared with the others, which had strains ranging from 40 to 85%. Moreover, Supramid Extra® II and Prolene® demonstrated the greatest work of fracture of ~ 200 MJ/m3 whereas Mersilene® mesh and Ptose-Up gave the lowest work of fracture (~ 2 MJ/m3). Visitec® SFSS gave an intermediate work of fracture, ~100 MJ/m3. These results are summarised in Fig. 4. The fracture surface morphology of each of the materials after tensile testing at 1 and 1500 mm/min displacement rates was examined by microscopy (Figs. 5–9). Prolene® tested at 1500 mm/min displacement rate showed a globular appearance on the fracture surface with a V-notch sticking out and this result has been documented to be associated with high-speed tensile break [39]. Stretching the material at a high-speed would cause heating, which could not have been dissipated to the surroundings. Consequently, the polymer, namely polypropylene (PP), softened and resulted in the localised flow of the material. In addition, the V-notch

indicates where the fracture was initiated. On the other hand, Prolene® tested at 1 mm/min rate failed by axial splitting, as reported by Karaca and Hockenberger [34] who performed tensile testing on varying sizes of polypropylene suture materials at 200 mm/min. They attributed this phenomenon to the chemical structure of the material since PP is anisotropic due to strong C\C covalent bonds along the molecular axis, but weak van der Waals forces existing between the C\C chains. This highly anisotropic nature of PP also resulted in the peeled-off fracture point. For Supramid Extra® II tested at the 1500 mm/min rate, the polyamide filaments seemed to be in place as in the untested material (Fig. 1b) with similar filament lengths. The outer sheath enclosing these filaments appeared to be intact although a small crack could be seen on its surface. However, when the material was tested at 1 mm/min the outer sheath did rupture, scattering the polyamide filaments. The filaments had varying lengths, indicating some of them fractured before others. This might be why the stress–strain curve of this material tested at 1 mm/min showed fluctuations (Fig. 3b). Some of the broken filament ends of Supramid Extra® II tested at 1 mm/min showed a pointed edge while most of the broken filament ends of the material tested at 1500 mm/min showed mushroom-like ends, which are normally associated with high-speed tensile fracture [39]. No discernible differences on the fracture surface of Visitec® SFSS silicone rod could be observed between tests at 1 or 1500 mm/min displacement rate. Furthermore, the fracture surfaces looked the same to the cross-sectioned surface of the material before the tensile test

Fig. 6. Scanning electron micrographs of Supramid Extra® II tensile test fracture surfaces: (a) 1 mm/min and (b) 1500 mm/min displacement rate.

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

227

Fig. 7. Scanning electron micrographs of the tensile test fracture surfaces of Visitec® Seiff frontalis suspensions set: (a) 1 mm/min and (b) 1500 mm/min displacement rate (the crack was caused by the electron beam damage during SEM).

Fig. 8. Scanning electron micrographs of Mersilene® mesh tensile test fracture surfaces: (a) 1 mm/min and (b) 1500 mm/min displacement rate.

228

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

(Fig. 1c), namely a smooth surface with small white nano-sized regions clearly seen at a high magnification; this result suggesting that Visitec® SFSS ultimately suffered a typical brittle fracture and agrees well with the literature [39], since highly extensible elastomeric fibres have been documented to finally fail with the brittle mode of fracture. Mersilene® mesh failed by the breakage of inter-woven polyester fibres (Fig. 8a and b). Although the width of the mesh was seen to be reduced near the failed region for both cases, (a) and (b), the longaxis of the oval pores seemed to be much larger for the material tested at 1 mm/min than for the one tested at 1500 mm/min. In addition, while the polyester fibres showed a smooth surface for the material tested at 1 mm/min rate some surface damage was observed in the fibres tested at 1500 mm/min. These were most likely caused by the fast pulling action of the fibres, thereby abrading the surface. Ptose-Up stretched at 1 mm/min resulted in a thin and long ePTFE fibre bundle at the fracture region while no such extensive deformation could be observed for the samples stretched at 1500 mm/min. On the plan-view (Fig. 9a), nodes were still clearly visible for the material tested at 1 mm/min and most of the ePTFE fibres connected to the nodes seemed to be elongated in the direction of stretching, coalescing into a flat sheet-like surface in some parts. However, no such organised feature could be observed for the material tested at 1500 mm/min. After fracture, there still existed an ordered array of fibres and nodes as seen on the untested surface, but in some parts of the material surface

the fibres were no longer parallel to each other in the direction of loading. The fibres and nodes were fused together and they could not be distinguished. This random orientation of the fused fibres and nodes could be the reason for the lower tensile strength and elastic modulus (Fig. 3e) for this material tested at 1500 mm/min compared with the one tested at 1 mm/min. Other factors should also influence the ophthalmic surgeon's choice of synthetic brow-suspension material — for example, the cost of the materials (Table 3). Comparing the price for ~150 mm long sample of each suspensory material, Ptose-Up is the most expensive material N£100. Visitec® SFSS costs about £30 while the rest of the materials costs less than £5. However, the mechanical properties of the materials are also crucial as they might predict possible outcomes. For example, at a given strain in the initial linear region of the stress–strain curve, the higher the elastic limit suggests that greater stress and consequently force are used; that is if a blink results in ~ 40% strain of the suspensory material implanted, Visitec® SFSS and Ptose-Up requires much less force to achieve such strain compared with Supramid Extra® II and Mersilene® mesh as illustrated in Fig. 4. The elastic modulus suggests not only the degree of pliability but also the possibility of a “cheese-wire” effect — namely, the higher elastic modulus meaning a greater stiffness and, consequently, a greater chance of “cheese-wiring” of the material cutting through the soft tissues underneath. Work of fracture, which is a function of how

Fig. 9. Scanning electron micrographs of Ptose-Up tensile test fracture surfaces: (a) 1 mm/min and (b) 1500 mm/min displacement rate.

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

229

Table 3 Size and cost of the commonly-used synthetic brow-suspension materials (Price in July 2013; Dia. = diameter). Material

Size

No. of product/pack

Cost (£)

Prolene® Supramid Extra® II Visitec® Seiff frontalis suspension set Mersilene® mesh Ptose-Up

Length = 900 mm, Dia. = ~0.2 mm Length = ~450 mm, Dia. = ~0.24 mm Length = 400 cm, Dia. = 0.8 mm Approximately 7 mm × 150 mm strips cut from each sheet of material 2 mm × 150 mm with 0.35 mm thickness

24 12 3 Variable 1

~100 ~85 ~180 ~1–2 ~115

much energy can be absorbed before the material failure, also gives some indication of the material behaviour once it is implanted in the body. Jacobs [40] reported the strength of the orbicularis oculi muscle, which is responsible for eyelid closure, for healthy humans to be about 60–70 g, equivalent to 0.6–0.7 N. Table 4 shows how much each brow-suspension material investigated in this study could be extended at the given load, ~0.7 N. Table 4 indicates that Visitec® SFSS alone will allow significantly greater — possibly even complete eyelid closure during normal blinking, since the other tested brow-suspension materials are estimated to result in only 2–17% change in strain during closing action of eye. The most-commonly used brow-suspension material, autogenous fascia lata has been reported to have ultimate tensile strength of about 50 MPa with N90% ultimate tensile strain [41]. This finding also suggests that Visitec® SFSS resembles more closely to human fascia lata compared with the other synthetic materials.

penetrating into the soft tissue underneath, thereby increasing the stability of the suspension. Strain rate influenced the mechanical behaviour of Prolene®, Supramid Extra® II and Ptose-Up. Increasing the strain rate was shown to increase the elastic limit, UTS and elastic modulus for Prolene® and Supramid Extra® II, while the opposite trend was observed for Ptose-Up. No such dependence was observed for Visitec® SFSS or the Mersilene® mesh. The effect of the strain rate on the fracture surface morphology was clear since the fracture ends from tests at 1 and 1500 mm/min displacement rates were significantly different for most of the materials. Visitec® SFSS alone showed no discernible difference and the effect for Mersilene® mesh was not as significant as for Prolene®, Supramid Extra® II and Ptose-Up. A further study is on-going with more testing including cyclic loading so that tests even-more closely reflect what the suspensory materials will experience once they are implanted in human body. Acknowledgements

4. Conclusions This study demonstrated that mechanical properties of commonlyused synthetic brow-suspension materials vary considerably, although most exhibited near-linear mechanical behaviour with an initial linear elastic region almost up to the fracture point (apart from Ptose-Up tested at 1 mm/min displacement rate). Ignoring different material geometries or sizes, when direct comparisons of the material properties at 1500 mm/min displacement rate were made, it was found that Prolene® and Supramid Extra® II exhibited substantially greater elastic modulus, UTS and work of fracture as compared with the others. In addition, the yield strain was lowest for Ptose-Up and greatest for Visitec® SFSS, with an order of magnitude difference. Definitive conclusions about the materials might require more sophisticated testing, such as after long-term exposure to body-fluids. However, purely in relation to the yield strain, elastic limit, elastic modulus and work of fracture deduced from these conventional mechanical testing results, the silicone rod (Visitec® SFSS) appears to have particularly suitable mechanical properties for the long-lasting treatment of ptosis, since it would require a relatively low force to stretch, with fairly reasonable work of fracture. At the reported load of orbicularis oculi, it was estimated that Visitec® SFSS would stretch ~ 80% in strain, most likely resulting in complete eyelid closure during blinking. Its low elastic modulus would also substantially reduce the risk of the material

Table 4 A comparison of extensibility of brow-suspension materials at the resultant load of closing action of an eye. Material

Strain (%) at ~0.7 N

Prolene® Supramid Extra® II Visitec® Seiff frontalis suspension set Mersilene® mesh Ptose-Up

2.43 5.75 80.3 8.67 17.2

± ± ± ± ±

0.70 3.28 17.2 1.97 4.4

This study was supported by an EPSRC-UCL Post-doctoral Mobility Award made to the authors and subsequent funding by the Dean of Engineering Sciences at University College London (UCL) to Edirisinghe. Authors Ezra and Rose acknowledge funding by the Department of Health through the award made by the National Institute for Health Research to Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology for a Specialist Biomedical Research Centre for Ophthalmology. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health. References [1] J.J. Dutton, A Color Atlas of Ptosis: A Practical Guide to Evaluation and Management, P G Publishing Pre Ltd., 1989. [2] B.C. Edmonson, A.E. Wulc, Otolaryngol. Clin. N. Am. 38 (2005) 921–946. [3] S.A. Fox, Surgery of Ptosis, The Williams & Wilkins Company, Baltimore, USA, 1980. [4] J.S. Crawford, Ophthalmic Surg. 8 (4) (1977) 31–40. [5] J.C. Flanagan, C.B. Campbell, Trans. Am. Ophthalmol. Soc. 79 (1981) 227–242. [6] Y. Takahashi, I. Leibovitch, H. Kakizaki, Open Ophthalmol. J. 4 (2010) 91–97. [7] B.N. Wasserman, D.T. Sprunger, E.M. Helveston, Arch. Ophthalmol. 119 (2001) 687–691. [8] S.M. Wheatcroft, S.J. Vardy, A.G. Tyers, Br. J. Ophthalmol. 81 (1997) 581–583. [9] S.R. Carter, W.J. Meecham, S.R. Seiff, Ophthalmology 103 (1996) 623–630. [10] M.J. Lee, J.Y. Oh, H.-K. Choung, N.J. Kim, M.S. Sung, S.I. Khwarg, Ophthalmology 116 (2009) 123–129. [11] C.R.J. Leone, J.W. Shore, J.V. Van Gemert, Ophthalmic Surg. 12 (1981) 881–887. [12] P.J. Rowan, G.S. Hayes, South. Med. J. 70 (1977) 68–69. [13] R.N. Downes, J.R.O. Collin, Br. J. Ophthalmol. 73 (1989) 498–501. [14] C.R. Hintschich, M. Zürcher, J.R.O. Collin, Br. J. Ophthalmol. 79 (1995) 358–361. [15] P. Mehta, P. Patel, J.M. Olver, Br. J. Ophthalmol. 88 (2004) 361–364. [16] T.K. Sharma, H. Willshaw, Eye 17 (2003) 759–761. [17] J.M. Gonzalez, Ocul. Surg. News 21 (4) (2003). [18] J.-M. Ruban, C. Burillon, E. Tabone, M. Mallem, C. Donne, D. Milea, Orbit 15 (1996) 67–76. [19] F.J. Steinkogler, A. Kuchar, E. Huber, E. Arocker-Mettinger, Plast. Reconstr. Surg. 92 (1993) 1057–1060. [20] K. Chow, N. Deva, S.G.J. Ng, Eye 25 (2011) 735–739. [21] R.M. Manners, A.G. Tyers, R.J. Morris, Eye 8 (1994) 346–348. [22] J.A. Katowitz, Arch. Ophthalmol. 97 (1979) 1659–1663. [23] K.H. Kook, H. Lew, J.H. Chang, H.Y. Kim, J. Ye, S.Y. Lee, Am. J. Ophthalmol. 138 (2004) 756–763. [24] R.A. Saunders, C.M. Grice, J. Pediatr. Ophthalmol. Strabismus 28 (1991) 271–273.

230

K.-A. Kwon et al. / Materials Science and Engineering C 35 (2014) 220–230

[25] D.E.W. Holmlund, Ann. Surg. 184 (1976) 189–193. [26] C.B. Iglesia, D.E. Fenner, L. Brubaker, Int. Urogynecol. J. 8 (1997) 105–115. [27] J. Catanese III, D. Cooke, C. Maas, L. Pruitt, J. Biomed. Mater. Res. A 48 (1999) 187–192. [28] T.M. Lawrence, T.R.C. Davis, J. Hand Surg. 30A (2005) 836–841. [29] L. Rennie, W. Fleming, D. Clark, C. Ellerton, R. Bosanquet, Eye 8 (1994) 343–345. [30] T. Sardenberg, S.S. Müller, P.B. De Almeida Silvares, A.B. Mendonça, R.R. De Lima Moraes, Acta Orthop. Bras. 11 (2003) 88–94. [31] J.A. von Fraunhofer, R.J. Storey, B.J. Masterson, Biomaterials 9 (1988) 324–327. [32] J.A. von Fraunhofer, R.S. Storey, I.K. Stone, B.J. Masterson, J. Biomed. Mater. Res. 19 (1985) 595–600. [33] K.-A. Kwon, R.J. Shipley, M. Edirisinghe, D.G. Ezra, G. Rose, S.M. Best, R.E. Cameron, J. R. Soc. Interface 10 (85) (2013) 20130227. [34] E. Karaca, A.S. Hockenberger, J. Biomed. Mater. Res. B Appl. Biomater. 87B (2008) 580–589. [35] J.L. Brun, L. Bordenave, F. Lefebvre, R. Bareille, C. Barbié, F. Rouais, C. Baquey, Bio-Med. Mater. Eng. 2 (1992) 203–225. [36] FCI Ophthalmics, http://www.fci-ophthalmics.com/lid-repair(accessed on 1st October 2013). [37] T. Nilsson, Scand. J. Plast. Reconstr. Surg. 16 (1982) 97–100. [38] F. Vizesi, C. Jones, N. Lotz, M. Gianoutsos, W.R. Walsh, J. Hand Surg. 33A (2008) 241–246. [39] J.W.S. Hearle, B. Lomas, W.D. Cooke, I.J. Duerden, Fibre Failure and Wear of Materials: An Atlas of Fracture, Fatigue and Durability, Ellis Horwood Limited, Chichester, England, 1989. [40] H.B. Jacobs, Br. J. Ophthalmol. 38 (1954) 560–566. [41] C.M. Gratz, J. Bone Joint Surg. Am. 13 (1931) 334–340.

Daniel Ezra is a consultant ophthalmic and oculoplastic surgeon at Moorfields Eye Hospital and Honorary Lecturer at the UCL Institute of Ophthalmology. He is a faculty member of the NIHR Biomedical Research Centre for Ophthalmology and is also the research lead for the Oculoplastics department.

Kyung-Ah Kwon completed her PhD in Department of Materials Science and Metallurgy at University of Cambridge in July 2012. She is currently a research associate at University College London (UCL), working in blinking and browsuspension analysis, brow-suspension material characterisation and nano-structured composite scaffold fabrication and characterisation.

Andrew Rayment MA is the senior technical officer for Mechanical Testing at the Department of Materials sciences and metallurgy, University of Cambridge. A Chartered Mathematician and Molecular Biologist he is working on several projects involving; bio-materials, nano-indentation and test method development.

Rebecca J Shipley DPhil is a Lecturer in Biomedical Engineering in the Department of Mechanical Engineering at University College London. Her research lies in mathematical modelling of biological systems with a focus on vascular tissues and tissue engineering systems. She completed her DPhil in Applied Mathematics from Oxford in 2009 and has published over 25 journal papers. In 2011 she was awarded Young Researcher of the Year for the UK Tissue and Cell Engineering Society.

Serena Best PhD, FREng, FIMMM, FBSE is co-director of the Cambridge Centre for Medical Materials, Department of Materials Science and Metallurgy of the University of Cambridge. She has published over 200 journal papers and holds 9 patents in the fields of biomaterials and skeletal repair. In 2011 she was awarded both the Chapman Medal and the Kroll Medal by the Institute of Materials, Minerals and Mining.

Mohan Edirisinghe DSc holds the Bonfield Chair of Biomaterials in the Department of Mechanical Engineering at University College London. He has published over 350 journal papers and his most recent research is on advanced biomaterials structure formation and analysis using new methods. He has been awarded many prizes for his research including the Royal Society Brian Mercer Innovation Feasibility Award (thrice — 2005, 2009 and 2013) and the 2012 UK Biomaterials Society President's prize.

Ruth Cameron is a Professor Materials Science and a Director of the Cambridge Centre for Medical Materials at the Department of Materials Science at the University of Cambridge and a fellow of Lucy Cavendish College, Cambridge. She is a fellow of both the Institute of Physics and of the Institute of Materials Minerals and Mining.

Geoffrey Rose graduated BSc Pharmacology, MBBS and MRCP. Postgraduate ophthalmic training culminated in award of FRCS in 1985 and FRCOphth at its foundation in 1988. In 1990 the University of London granted an MS doctorate for corneal research and, in 2004, a Doctor of Science in Ophthalmology and Ophthalmic Surgery. Professor Rose was appointed consultant to Moorfields Eye Hospital in 1990 and is also a Senior Research Fellow of the NIHR Biomedical Research Centre of the Institute of Ophthalmology. He is a Past-President of the British Oculo-Plastic Surgical Society.