Frequency dependence of viscoelastic properties of medical grade silicones

In vitro investigation into the forces involved during lipofilling

Breast augmentation using implants is the most common aesthetic and reconstructive breast surgical procedure. Complications such as implant rupture maybe related to surgical technique and damage to the implant. Autologous fat transfer (lipofilling) using metallic cannulae has become a standard adjunctive, yet there is little evidence on lipofilling safety in the presence of implants. The aims of this study are to verify the effects of different cannulae and to quantify the forces applied by surgeons during lipofilling. Silicone gel-filled textured implants (200 mL), mounted on a specially constructed mould were ruptured with two different cannulae: type A (hole at tip: sharp) and type B (hole away from tip: blunt), driven at three speeds (10, 100 and 1000 mm/min), and the force at rupture was recorded. In addition, the maximum 10 forces over a 30-s period applied by 11 plastic surgeons against a breast implant in an in vitro environment were recorded using a load cell attached to a type-A cannula. Statistical analysis of comparative results was performed using t-tests, with p < 0.05 considered significant. Results showed that the implant ruptured at forces up to 25% lower when cannula A was used compared to cannula B. This supports current technique in lipofilling in the use of a blunt tipped cannula. There was a significant difference between some displacement rates only, due to the viscoelastic nature of the material. The tactile force that surgeons use during lipofilling was modelled in vitro and showed a range of maximum forces between 0.23 and 16.8 N, with a mean maximum value of 6.9 N. Limitation of this study is that it may not reflect in vivo behaviour of breast implants. More studies are needed to confirm the safety of breast lipofilling in the presence of implants using these data as a starting point.

Viscoelastic properties of mitral valve leaflets: An analysis of regional variation and frequency-dependency

The aim of this study was to determine the regional variation in viscoelastic properties of mitral valve leaflets over a range of physiological and patho-physiological frequencies. This included comparisons to be made between anterior and posterior leaflets, anterior leaflet clear and rough zones, and radial and circumferential leaflet orientation. Dynamic mechanical analysis was used to determine frequency-dependent viscoelastic properties. The valve leaflets were dissected from eight porcine hearts. The leaflets were loaded under a sinusoidal tensile displacement, with a mean dynamic peak to trough strain of 11%, applied to all leaflet samples at nine different frequencies, ranging from 0.5 to 10 Hz. The anterior leaflet has higher storage and loss stiffness than the posterior leaflet. The storage stiffness of circumferential tissue is greater than that of radially oriented valve tissue (2.0 ± 1.6 N/mm cf. 1.7 ± 0.9 N/mm; p < 0.05); however, the loss stiffness is greater for radial tissue (0.15 ± 0.07 cf. 0.14 ± 0.09 N/mm; p < 0.05). Likewise, the storage stiffness of the anterior leaflet clear zone is greater than that of the rough zone (2.4 ± 1.6 cf. 2.1 ± 1.2; p < 0.05), but the loss stiffness is greater for the rough zone (0.17 ± 0.09 N/mm cf. 0.14 ± 0.08 N/mm; p < 0.05). In conclusion, the viscoelastic properties of porcine mitral valve leaflets have regional variations, with dynamic stiffness being dependent on circumferential or radial orientation and on location at a clear or rough zones.

Effect of accelerated aging on the cross-link density of medical grade silicones

Four specimens of Nagor silicone of different hardness (soft, medium and hard) were swollen, until they reached equilibrium (i.e. constant mass) in five liquids at 25°C, before and after accelerated aging. For the specimens swollen before accelerated aging, the greatest swelling was obtained in methyl cyclohexane, while for the specimens swollen after accelerated aging, the greatest swelling was obtained in cyclohexane. The cross-link density, υ, was also calculated from the swelling measurements for all the specimens, before and after accelerated aging, using the Flory-Rehner equation. The softer silicones, which swelled the most, had lower υ values than harder silicones. The amount of swelling (measured in terms of ϕ) and υ varied significantly (p<0.05) in some cases, between the different silicone hardness and between different liquids. Furthermore, the cross-link density, υ, significantly (p<0.05) increased after accelerated aging in most liquids.Note: ϕ is defined as the volume fraction of polymer in its equilibrium swollen state. A probability value of statistical significance of 0.05 or 5% was selected, hence if a p value of less than 0.05 was obtained, the null hypothesis was rejected (i.e. significant if p<0.05).

Effect of accelerated aging on the viscoelastic properties of a medical grade silicone

The viscoelastic properties of cylinders (diameter 5 mm, height 2.2 ± 0.2 mm) of Nagor silicone elastomer of medium hardness, were investigated before and after the specimens had undergone accelerated aging in saline solution at 70°C for 38, 76 and 114 days (to simulate aging at 37°C, for 1, 2 and 3 years, respectively). All sets of specimens were immersed in physiological saline solution at 37°C during testing and the properties were measured using dynamic mechanical analysis (DMA). A sinusoidal cyclic compression of 40 N ± 5 N was applied over a frequency range, f, of 0.02-25 Hz. Values of the storage, E', and loss, E″, moduli were found to depend on f; the dependence of E' or E″ on the logarithm (base 10) of f was represented by a second-order polynomial. After accelerated aging, the E' and E″ values did not increase significantly (p<0.05). Furthermore, scanning electron microscopy (SEM) showed that accelerated aging did not affect the surface morphology of silicone. Attenuated total reflectance Fourier transform infra-red spectroscopy (ATR-FTIR) showed that accelerated aging had a negligible effect on the surface chemical structures of the material. Differential scanning calorimetry (DSC) showed no changes to the bulk properties of silicone, following accelerated aging.

Feasibility of using mixtures of silicone elastomers and silicone oils to model the mechanical behaviour of biological tissues

Mixtures of silicone elastomer and silicone oil were prepared and the values of their Young’s moduli, E, determined in compression. The mixtures had volume fractions, ϕ, of silicone oil in the range of 0–0.73. Measurements were made, under displacement control, for strain rates, ε·, in the range of 0.04–3.85 s⁻¹. The behaviour of E as a function of ϕ and ε· was investigated using a response surface model. The effects of the two variables were independent for the silicones used in this investigation. As a result, the dependence of E values (measured in MPa) on ϕ and ε· (s⁻¹) could be represented by E=0.57−0.75ϕ+0.01loge(ε·). This means that these silicones can be mixed to give materials with E values in the range of about 0.02–0.57 MPa, which includes E values for many biological tissues. Thus, the mixtures can be used for making models for training health-care professionals and may be useful in some research applications as model tissues that do not exhibit biological variability.

Effect of lubricants on friction in laboratory tests of a total disc replacement device

Some designs of total disc replacement devices have articulating bearing surfaces, and these devices are tested in vitro with a lubricant of diluted calf serum. It is believed that the lubricant found in total disc replacement devices in vivo is interstitial fluid that may have properties between that in Ringer's solution and diluted calf serum. To investigate the effect of lubricants, a set of friction tests were performed on a generic model of a metal against metal ball-and-socket total disc replacement device. Two devices were tested: one with a ball radius of 10 mm and other with a ball radius of 16 mm; each device had a radial clearance of 0.015 mm. A spine simulator was used to measure frictional torque for each device in axial rotation, flexion-extension and lateral bending at frequencies of 0.25-2 Hz, under 1200 N axial load. Each device was tested with two different lubricants: a solution of new born calf serum diluted with deionised water and Ringer's solution. The results showed that the frictional torque generated between the bearing surfaces was significantly higher in Ringer's solution than in diluted calf serum. The use of Ringer's solution as a lubricant provides a stringent test condition to detect possible problems. Diluted calf serum is more likely to provide an environment closer to that in vivo. However, the precise properties of the fluid lubricating a total disc replacement device are not known; hence, tests using diluted calf serum may not necessarily give the same results as those obtained in vivo.

Polymer-on-metal or metal-on-polymer total disc arthroplasty: does it make a difference?

STUDY DESIGN

Mechanical testing of total disc arthroplasty (TDA).

OBJECTIVE

To compare the friction between a polymer socket-on-metal ball and metal socket-on-polymer ball TDA.

SUMMARY OF BACKGROUND DATA

A degenerate intervertebral disc can be replaced by TDA. The most common designs have a ball and socket articulation; the contact between the surfaces leads to friction. Friction needs to be minimized to prevent loosening and wear. One of the common material combinations in disc arthroplasty devices is the articulation of a metal socket on polymer ball. However, the combination of a polymer socket on metal ball (which is used in hip arthroplasty) has not been investigated for TDA.

METHODS

TDA models with either a polymer socket/metal ball or a metal socket/polymer ball were manufactured with ball radii of 10 and 14 mm, each with a radial clearance of 0.35 mm. Samples were tested using a spine simulator with a lubricant of diluted newborn calf serum. Each sample was subjected to an axial load of 1200 N; motions of flexion-extension, lateral bending, and axial rotation were then applied at frequencies of 0.25 to 2 Hz. Frictional torque was measured to compare the performance of the TDAs.

RESULTS

The frictional torque was found to be significantly higher for a disc with a metal socket/polymer ball than for a disc with a polymer socket/metal ball for both 10 and 14 mm radii in axial rotation, lateral bend, and extension. The frictional torque in flexion (0°-6°) was not found to be significantly different between the 2 different material combinations. However, when the flexion motion was reduced to 0° to 2°, frictional torque in the metal socket/polymer ball was found to be significantly higher than the polymer socket/metal ball.

CONCLUSION

TDA with a combination of a polymer socket/metal ball has lower friction than the conventional TDA with metal socket/polymer ball. This conclusion has implications in the design of TDA.

Viscoelastic properties of bovine articular cartilage attached to subchondral bone at high frequencies

Background

Articular cartilage is a viscoelastic material, but its exact behaviour under the full range of physiological loading frequencies is unknown. The objective of this study was to measure the viscoelastic properties of bovine articular cartilage at loading frequencies of up to 92 Hz.

Methods

Intact tibial plateau cartilage, attached to subchondral bone, was investigated by dynamic mechanical analysis (DMA). A sinusoidally varying compressive force of between 16 N and 36 N, at frequencies from 1 Hz to 92 Hz, was applied to the cartilage surface by a flat indenter. The storage modulus, loss modulus and phase angle (between the applied force and the deformation induced) were determined.

Results

The storage modulus, E', increased with increasing frequency, but at higher frequencies it tended towards a constant value. Its dependence on frequency, f, could be represented by, E' = Alog_e (f) + B where A = 2.5 ± 0.6 MPa and B = 50.1 ± 12.5 MPa (mean ± standard error). The values of the loss modulus (4.8 ± 1.0 MPa mean ± standard deviation) were much less than the values of storage modulus and showed no dependence on frequency. The phase angle was found to be non-zero for all frequencies tested (4.9 ± 0.6°).

Conclusion

Articular cartilage is viscoelastic throughout the full range of frequencies investigated. The behaviour has implications for mechanical damage to articular cartilage and the onset of osteoarthritis. Storage modulus increases with frequency, until the plateau region is reached, and has a higher value than loss modulus. Furthermore, loss modulus does not increase with loading frequency. This means that more energy is stored by the tissue than is dissipated and that this effect is greater at higher frequencies. The main mechanism for this excess energy to be dissipated is by the formation of cracks.