(L-455) A Computational Study of Wrinkling and Porosity for Novel Auxetic Membrane Designs to Aid in the Repair of Pelvic Organ Prolapse in Women

Introduction:: Pelvic organ prolapse is a condition that occurs primarily in women when support to the pelvic organs becomes compromised. It results in roughly 1 in 10 women seeking surgical repair during their lifetime. Unfortunately, between 10-30% of repairs using surgical mesh can result in complications such as pain and mesh exposures. Our group has shown that these complications are associated with mesh wrinkling and pore collapse, likely through a mechanical signaling pathway. To address this, we have established a new repair device that we call an auxetic membrane, whose pores increase in size when deformed. This study aims to characterize wrinkling and porosity across membranes with varying pore size and thickness. Specifically, we hypothesize that the auxetic pore geometry would dominate the membrane behavior, leading to flattening for all designs (i.e. minimal wrinkling). We also hypothesized that changes in porosity with tensioning would be similar. This computational study is a step toward identifying the design parameters that optimize membrane performance and can also provide insights into the design of surgical mesh for improved outcomes in pelvic organ prolapse repairs.


Materials and Methods:: Ten membranes, with varied strut (fiber) width and thickness, were designed using CAD software (Solidworks). Half of the membranes had sharp edges and half had rounded edges. The membranes were discretized using fTetWild (https://github.com/wildmeshing/fTetWild) into between 77k and 128k tetrahedral elements. The material was assumed to be neo-Hookean with a modulus of 64 MPa and a Poisson’s ratio of .425. Two longitudinally oriented point loads located at 25% and 75% along the width of the top edge were used to stretch each membrane to 200% of its original length. These points were free to move in all other degrees of freedom. Along the bottom edge, boundary conditions were prescribed to corresponding points along the width. One point was fixed in all directions, while the other was fixed in only the longitudinal direction. These boundary conditions simulate the membrane being pulled with two sutures tied to either end and allows the membrane to wrinkle freely in response to the applied tension. Simulations were performed in PolyFEM, a novel open-source solver that implements incremental potential contact, which guarantees no element inversions or penetration with contact. To assess wrinkling, a curvature index was measured using the geometries’ cross-sections at 50% of the longitudinal length of the membranes. It was defined as the ratio of a membrane’s arc length divided by its width minus 1. Porosity was determined by calculating the volume of empty space relative to the total volume (volume of empty space + volume of membrane material) and reported as a percentage.




Results, Conclusions, and Discussions:: The membrane designs that demonstrated the greatest amount of wrinkling were 7, 8, 9, and 10 (Figure 1), which were the thinnest. Those with a higher in-plane thickness had curvature indexes that were 1-2 orders of magnitude less than the thinner membranes and appeared flatter. Membranes 3 and 4 curved the least (curvature indexes of 0.009 and 0.003, respectively). Porosity increased by 20.4% on average between the undeformed and deformed states for all designs (Table 1). Membrane designs 1-6 had a higher rest average rest porosity of 55.4% while membranes 7-10 had a lower average rest porosity of 36.5%. Rounded edges had little effect on wrinkling behavior and resulted in a small decrease in porosity (between 1.2% for pair 5 & 6 and 2.5% for pair 9 & 10).
In conclusion, our study used a computational approach to investigate the impact of geometric design parameters on wrinkling and porosity of auxetic membrane being designed as a replacement to surgical mesh for the treatment of pelvic organ prolapse repairs. Despite the auxetic nature of these membranes (i.e. consistent with a material that has a negative Poisson’s ratio), our simulations revealed that membrane thickness is a major contributor to wrinkling, which is consistent with wrinkling theory for non-auxetic thin elastic sheets. On the other hand, we found that geometric differences mostly affected the rest porosity, but not the change in porosity with tensioning. These findings will aid in the design of devices that minimize wrinkling and maximize porosity in order to promote a host response to the device that lessens the risk for complications. Future studies will manufacture and implant these membranes in-vivo to determine how these behaviors affect the biological response of the host.



Acknowledgements (Optional): :

References (Optional): :