Background
Pelvic organ prolapse (POP) is a multifactorial disease that manifests as the herniation of the pelvic organs into the vagina. Surgical methods for prolapse repair involve the use of a synthetic polypropylene mesh. The use of this mesh has led to significantly higher anatomical success rates compared with native tissue repairs, and therefore, despite recent warnings by the Food and Drug Administration regarding the use of vaginal mesh, the number of POP mesh surgeries has increased over the last few years. However, mesh implantation is associated with higher postsurgery complications, including pain and erosion, with higher consecutive rates of reoperation when placed vaginally. Little is known on how the mechanical properties of the implanted mesh itself change in vivo. It is assumed that the mechanical properties of these meshes remain unchanged, with any differences in mechanical properties of the formed mesh-tissue complex attributed to the attached tissue alone. It is likely that any changes in mesh mechanical properties that do occur in vivo will have an impact on the biomechanical properties of the formed mesh-tissue complex.
Objective
The objective of the study was to assess changes in the multiaxial mechanical properties of synthetic clinical prolapse meshes implanted abdominally for up to 90 days, using a rat model. Another objective of the study was to assess the biomechanical properties of the formed mesh-tissue complex following implantation.
Study Design
Three nondegradable polypropylene clinical synthetic mesh types for prolapse repair (Gynemesh PS, Polyform Lite, and Restorelle) and a partially degradable polypropylene/polyglecaprone mesh (UltraPro) were mechanically assessed before and after implantation (n = 5/ mesh type) in Sprague Dawley rats for 30 (Gynemesh PS, Polyform Lite, and Restorelle) and 90 (UltraPro and Polyform Lite) days. Stiffness and permanent extension following cyclic loading, and breaking load, of the preimplanted mesh types, explanted mesh-tissue complexes, and explanted meshes were assessed using a multi-axial (ball-burst) method.
Results
The 4 clinical meshes varied from each other in weight, thickness, porosity, and pore size and showed significant differences in stiffness and breaking load before implantation. Following 30 days of implantation, the mechanical properties of some mesh types altered, with significant decreases in mesh stiffness and breaking load, and increased permanent extension. After 90 days these changes were more obvious, with significant decreases in stiffness and breaking load and increased permanent extension. Similar biomechanical properties of formed mesh-tissue complexes were observed for mesh types of different preimplant stiffness and structure after 90 days implantation.
Conclusion
This is the first study to report on intrinsic changes in the mechanical properties of implanted meshes and how these changes have an impact on the estimated tissue contribution of the formed mesh-tissue complex. Decreased mesh stiffness, strength, and increased permanent extension following 90 days of implantation increase the biomechanical contribution of the attached tissue of the formed mesh-tissue complex more than previously thought. This needs to be considered when using meshes for prolapse repair.
Pelvic organ prolapse (POP) is a multifactorial disease that manifests as the herniation of the pelvic organs into the vagina. When conservative approaches do not relieve symptoms, surgery is a widely used treatment option. Since the introduction of synthetic meshes in hernia repair, surgical methods for prolapse repair have also involved the use of a similar type of synthetic mesh. The use of this mesh has led to significantly higher anatomical success rates compared with native tissue repairs, and therefore, despite recent warnings by the Food and Drug Administration regarding the use of vaginal mesh, the number of POP mesh surgeries has increased over the last few years. However, mesh implantation is associated with higher postsurgery complications, including pain and erosion, with higher consecutive rates of reoperation when placed vaginally.
Ideally, a POP mesh should function to support the damaged tissue, augmenting new tissue growth. It is also important that the mesh matches the biomechanical properties of the surrounding vaginal tissue so that it is mechanically compliant. Complications involving the use of mesh have decreased in alignment with the increasing use of light-weight meshes (20-30 g/m 2 ) containing large pores.
These lightweight, highly porous meshes may be favoured in POP repair because they enhance tissue integration and reduce encapsulation, potentially improving the biomechanical properties of the formed mesh-tissue complex (MTC). Poor mechanical compliance between the implanted mesh and surrounding tissue may lead to stress shielding and complications such as erosion of the surrounding tissue and dyspareunia.
Little is known on how the mechanical properties of the implanted mesh itself change in vivo. It is assumed that the structure of nondegradable polypropylene (PP) meshes remains intact and that the mechanical properties of these meshes remain unchanged, with any differences in mechanical properties between implanted mesh and formed MTC attributed to the attached tissue. However, studies have determined some in vivo degradation of PP hernia and POP meshes. It is likely that any changes in mesh mechanical properties that occur will have an impact on the biomechanical properties of the formed MTC. We hypothesize that if polymer degradation does occur, then it will change the mechanical properties of the implanted mesh itself and thus contribute to the observed changes in reported biomechanical properties of the formed MTC.
The aim of this study was to investigate changes in the mechanical properties of nondegradable and partially degradable PP-based clinical POP meshes, following implantation up to 90 days, using a rat abdominal hernia model. An additional aim was to determine the influence of these changes on the biomechanical properties of the formed MTC for a range of mesh types.
Materials and Methods
Clinical meshes
Three clinical nondegradable PP meshes (Restorelle, Coloplast Pty Ltd, Notting Hill, Australia; Polyform Lite, Boston Scientific, Marlborough, MA; and Gynemesh PS, Ethicon, Johnson and Johnson, North Ryde, Australia) and a partially degradable PP/polyglecaprone mesh (UltraPro [Ethicon, Johnson and Johnson]), were assessed for their structural properties, pore size, thickness, mass, and porosity.
Optical micrographs of the clinical meshes ( Figure 1 , A–D) were captured using a light microscope (M40; Wild Heerbrugg, Heerbrugg, Switzerland), with an attached digital camera (DFC290; Leica, North Ryde, Australia). Image analysis software (Image-Pro Plus; Media Cybernetics, Silver Spring, MD) was used to measure small pore size (less than 1000 μm in length [ranging 78–188 pores/mesh type]) and large pore size (greater than 1000 μm in length [ranging 16–36 pores/mesh type). It was also used to calculate mesh porosity (void fraction of mesh) based on a solid mesh area of 127 mm 2 for each mesh type (1 mesh per mesh type). Mesh types were measured for thickness (4–6 meshes per mesh type) using electronic digital callipers and mass/area (4–6 meshes per mesh type).
Animals
Sprague Dawley rats were housed in the Monash Medical Centre Animal Facility in Clayton, Australia, in compliance with the National Health and Medical Research Council guidelines for the care and use of laboratory animals. Ethics approval was obtained from the Monash Medical Centre Animal Ethics Committee A (2009/50) on Nov. 24, 2009.
The rats were kept in closed cages with food and water supply ad libitum under controlled environmental conditions at 20°C and a 12 hour day/night cycle. More commonly used nondegradable meshes were chosen for study up to 30 days of implantation to assess the influence of mesh weight on mechanical degradation in vivo and resultant mesh-tissue complex properties. Of these meshes, the Polyform Lite, which was a typical midrange weight, nondegradable mesh, was extended to 90 days of implantation and compared with a partially degradable mesh, the UltraPro Lite mesh (Ethicon, Johnson and Johnson).
Thirty rats were randomly divided into 6 experimental groups (5 rats/group), with the groups consisting of Restorelle (Coloplast Pty Ltd), Polyform Lite (Boston Scientific), and Gynemesh PS (Ethicon, Johnson, and Johnson) implanted for 30 days and UltraPro (Ethicon, Johnson, and Johnson) and Polyform Lite (Boston Scientific) implanted for 90 days. Five rats served as a control (tissue from the abdominal wall was collected from rats that did not undergo any surgical procedure).
Meshes were cut into 3.0 × 2.5 cm pieces and implanted subcutaneously in the abdomen of each rat, as previously described. The abdominal hernia model is an acceptable initial animal model to evaluate the mechanical behaviour of mesh types, including vaginal mesh. The animals were sacrificed at day 30 (Polyform Lite, Gynemesh, Restorelle) and day 90 (Ultrapro, Polyform Lite) (n = 5/group per time point). Explanted meshes were dissected with a 0.5 cm border of adjacent tissue including the skin and underlying attached muscle. Mesh-tissue complexes for biomechanical analyses were immediately frozen at –20°C until testing.
Mechanical testing
The same test method was used to mechanically assess all samples types. Preimplanted meshes were soaked in phosphate-buffered saline for up to 2 hours prior to testing, and all other sample types were wet with phosphate-buffered saline during testing.
In testing the MTC and the explanted mesh, the MTC was tested first, and following this, the MTC was separated into mesh and tissue counterparts, carefully removing the nonadherent tissue from the mesh, with minimal strain on the mesh to avoid permanent deformation of the mesh. The explanted mesh counterpart was then tested in a new position. This method of separation and testing was validated by comparing the mechanical properties of pretested and not pretested explanted meshes, with pretesting having no effect on the mechanical properties of the mesh when the mesh was tested in a new position.
Samples were tested using an Instron Tensile Tester (5557; Instron Corp, Canton, MA) with a load cell of 1 kN. Samples were secured between 2 embossed metal plates, both with an aperture of 20 mm (for penetration of the steel ball). For the MTC, the synthetic mesh was placed uppermost, nearest the steel ball. Rubber sheeting was used to avoid sample slippage during testing. During testing a rounded steel rod, of a diameter of 10 mm, was pushed through the sample at a crosshead speed of 10 mm/min to a preload of 0.5 N. Samples were then cyclically loaded from 0 to 5 N for 10 cycles and then from 5 to 10 N for 10 cycles, at 10 mm/min, and then extended until the rupture of the sample.
Load-elongation curves were plotted from the generated data, and from these curves the stiffness (newtons per millimeter) in the linear region, permanent extension (millimeters), and the breaking load (newtons) were calculated. Mesh stiffness was taken to be the slope of the load-elongation curve after all cyclic activity and prior to the first sign of sample rupture, measured using best fit linear regression between 2 reference points (with R-squared value greater than 0.95). Permanent extension was noted as the increase in length at 0 N following the first set of cyclic loading, from 0 to 5 N.
Statistics
R statistical software package (free software) was used for statistical analysis. The Shapiro-Wilk test showed normally distributed residuals for the stiffness and breaking load results but not the permanent extension results. Quantitative data are reported as mean ± SEM, with breaking load and stiffness results statistically analyzed using a 2-way analysis of variance, with Bonferroni correction to determine significant differences. Permanent extension results were statistically analysed using multiple t tests. Values of P < .05 were considered to be statistically significant.
Results
Demographic data
All rats were 12 weeks old at the time of implantation and did not show significant differences in their pre- or postoperative weight. No rats died during the operation or the study period.
Preimplant mesh properties
The 4 clinical meshes showed different structural properties in terms of their mass per area, ranging from 19 ± 0 g/m 2 for Restorelle to 56 ± 1 g/m 2 for UltraPro, and thickness, ranging from 0.2 ± 0.0 mm for Polyform Lite to 0.5 ± 0.0 mm for UltraPro ( Table ). Small differences in mesh porosity were determined between mesh types, with Gynemesh the least porous (67%) and Polyform Lite the most porous (73%) ( Table ).
| Mass/area, g/m 2 | Thickness, mm | Porosity, % | Small pore size, μm | Large pore size, μm | |
|---|---|---|---|---|---|
| UltraPro | 56 ± 1 | 0.5 ± 0.0 | 69 | 335 ± 12 | 2704 ± 65 |
| Gynemesh PS | 40 ± 0 | 0.4 ± 0.0 | 67 | 488 ± 16 | 1913 ± 78 |
| Polyform Lite | 26 ± 0 | 0.2 ± 0.0 | 73 | 613 ± 11 | 2027 ± 36 |
| Restorelle | 19 ± 0 | 0.3 ± 0.0 | 72 | 230 ± 6 | 2092 ± 42 |
Pore size was grouped into small (less than 1000 μm) and large (greater than 1000 μm). Small pore size differed between mesh types, with Restorelle possessing the smallest, at 230 ± 6 μm, and Polyform Lite possessing the largest, at 613 ± 11 μm. Less variability was determined for large pore size between mesh types, ranging from 1913 ± 78 μm to 2092 ± 42 μm for all meshes apart from the UltraPro mesh, which had a larger pore size at 2704 ± 65 μm ( Table ).
Preimplanted clinical meshes were assessed for multiaxial mechanical properties and results compared with each other and with native rat tissue: mesh stiffness, permanent extension, and breaking load results are provided in Figure 2 , A–C. Comparing mesh types, UltraPro and Gynemesh were of similar stiffness and breaking load ( Figure 2 , A and C), with these meshes significantly stiffer than Polyform Lite ( P < .001) and Restorelle ( P < .0001) ( Figure 2 A), and with significantly higher breaking loads than Polyform Lite ( P < .01 and P < .001, respectively) and Restorelle ( P < .001 and P < .0001, respectively) ( Figure 2 C). Slight nonsignificant differences in permanent extension were determined between mesh types ( Figure 2 B).
Compared with the rat tissue, UltraPro, Gynemesh, and Polyform Lite were all significantly ( P < .0001, P < .0001, P < .01, respectively) stiffer and of higher breaking load than the rat tissue ( Figure 2 , A and C). The mesh types, taken together, had significantly ( P < .05) lower permanent extension than the rat tissue ( Figure 2 B). The preimplanted Restorelle mesh was closest in mechanical properties to the native rat tissue ( Figure 2 , A–C).
Changes in mesh mechanical properties following implantation
To identify whether the properties of the MTC were due to tissue ingrowth or changes in the mesh mechanical properties following implantation, explanted mesh mechanical properties were compared with those of the preimplanted mesh. Mesh stiffness, permanent extension, and breaking load results after 30 and 90 days of implantation are provided in Figure 3 , A–C, and D–F, respectively.
We found a decrease in mesh stiffness and breaking load and an increase in permanent extension following 30 days ( Figure 3 , A–C) and 90 days ( Figure 3 , D–F) of implantation for some mesh types. Significant decreases in mesh stiffness were determined for Gynemesh ( P <0.001) and Restorelle ( P <0.05), and in breaking load for Gynemesh ( P <0.0001) ( Figure 3 , A and C) at 30 days. At 90 days UltraPro and Polyform Lite significantly decreased in stiffness ( P < .0001 and P < .001, respectively) ( Figure 3 D) and breaking load ( P < .01 and P < .05, respectively) ( Figure 3 F), with nonsignificant increases in permanent extension for both mesh types ( Figure 3 E).
With increased implantation time (between days 30 and 90), there were continued decreases in stiffness and breaking load for the Polyform Lite mesh, with these differences significant ( P < .05) for breaking load. In terms of loss in mesh strength (calculated as the percentage change in the average breaking load before and after implantation), at 30 days Gynemesh had the greatest loss in strength (51%), and this was followed by Restorelle (22%) and Polyform Lite (12%) meshes. At 90 days, Polyform Lite had a greater loss in strength (57%) than the UltraPro mesh (39%).
Mesh-tissue complex biomechanical properties
Explanted MTCs, comprising mesh plus attached abdominal wall muscle, were tested for multiaxial mechanical properties and compared with those of the explanted mesh (separated from the MTC). Stiffness, permanent extension, and breaking load results following 30 ( Figure 4 , A–C) and 90 ( Figure 4 , D–F) days of implantation were determined.
