Longitudinal Impact of Mesh Implantation and Menopause on Vaginal Tissue Structure

Objectives: Urogynecologic meshes have been used to overcome the high failure rates of native tissue repair yet hampered by mesh-related complications. Extracellular collagen architecture and muscularis in the vaginal wall are key to its structural integrity. Here we aimed to define the longitudinal impact of a polypropylene mesh on the vaginal tissue structure in a rat model with or without surgical menopause.

Methods: Female Wistar rats (9-12mos, 285–350g, n=57) were used, with n≥5 in each subgroup. A polypropylene mesh was implanted on the anterior and posterior vagina via sacrocolpopexy following transabdominal supracervical hysterectomy with bilateral ovariectomy (OVX) for 3-, 7- and 42-days or without OVX for 42 days. Sham groups underwent the same procedures without mesh implantation. Vaginal tissues were processed for H&E, Masson’s Trichrome, and picrosirius red staining. Images were taken under a polarized or light microscope. Thickness of subepithelium (subE) and muscularis (Mus) was measured at 100μm intervals, differentiating areas between and beneath mesh fibers in groups with mesh (Figure 1). Thick/aligned (R), thin/disorganized (G), and intermediate (Y) collagen fibers were measured pixel by pixel using a customized MATLAB-based program and normalized by tissue area. Collagen content was represented by % of fiber area/tissue area (Figure 1). Kruskal-Wallis with post-hoc and Wilcoxon matched-pair tests were performed (significance p< .05).

Results: In groups without mesh, the thickness of subE and Mus was both increased by 40% and 80% (p=.024, .022) relative to normal at 3-day post-surgery. The collagen content was simultaneously decreased by 35% at 3 days (p < .001), characterized by a predominant thin and disorganized fibers (58%). These parameters restored approximately to normal at 7- and 42- days (Figure 2). A comparable pattern was observed in groups with mesh. Comparing the groups with vs. without mesh, no differences in layer thickness were found at all time points, while the former had a higher ratio of thick/thin collagen fibers over time, most significantly at 42 days (p=.001), indicating a mesh-induced collagen remodeling and likely fibrotic response. Comparing the mesh groups with vs. without OVX, the thickness of subE and Mus beneath but not between the mesh fibers was both decreased by 31% and 41% (p=.021, .013) in the former at 42 days. Paradoxically, the collagen content tended to be 11% higher with OVX (p=.14), predominantly composed of thick fibers (85%) with 2.4-fold increase in the R/G ratio (p < .001) and a 38% decrease of intermediate fibers (p < .001), consistent with a greater fibrotic response.

Conclusions: Surgery with or without mesh implantation induces transient vaginal tissue edematous changes at the very early postoperative stage. In the long term, menopause-associated estrogen decrease may accentuate the negative effect of mesh on the vaginal tissue structure and intensify mesh-induced tissue fibrosis, increasing the risk of mesh implications.