Hypothesis / aims of study
Urinary incontinence and pelvic organ prolapse are common debilitating gynecologic conditions in postmenopausal women. It is estimated that the lifetime risk of undergoing a single operation for incontinence or prolapse by age 80 was about 11.1%. Urogynecologic meshes, including mid-urethral slings and prolapse meshes, are important implantable devices that have been used to overcome the high failure rates of native tissue repair. However, the use of mesh is hampered by mesh-related complications, typically mesh exposure into vaginal lumen with the vaginal wall continuity breached. The impact of mesh on the vagina and the mechanism underlying the pathogenesis of complications are not completely understood.
Vaginal muscularis, composed of smooth muscle fascicles is key to the integrity of vaginal wall and plays important roles in vaginal tone maintenance and sexual function, thus having important implications for a women’s quality of life. Our group and others have shown that the vaginal smooth muscle (VSM) is highly sensitive and responsive to mechanical cues, undergoing degenerative changes following the implantation of stiffer and heavier meshes [1, 2]. In addition, menopausal estrogen decrease induces vaginal atrophy including both molecular and functional changes of VSM. Considering that women who receive mesh implantation are mostly post-menopausal, it is important to understand how the VSM is impacted by both mesh implantation and menopause. While these factors have been individually studied, scarce data is available on their combined effect. In addition, diabetes is a pandemic affecting approximately 1/5 of aging women 45 years or older. Women with diabetes experience ~5-fold increase of risk of mesh exposure, the mechanism of which is unknown. While studies have shown that diabetes negatively impact vascular smooth muscle [3], its impact on the VSM in the context of mesh implantation and/or menopause is unclear.
In this study, we aimed to define the impact of mesh implantation on the VSM in the situation of menopause and/or diabetes by comparing the VSM thickness using a rat model. We hypothesized that mesh implantation, menopause-associated estrogen decrease, and hyperglycemia have a synergistic negative impact on the VSM.
Study design, materials and methods
A rat model was used. Three interferences were introduced: (1) bilateral ovariectomy (OVX); (2) diabetes; (3) mesh implantation. Based on a power analysis using our previous data comparing stiffness of vagina between rats with vs. without OVX [3], at least 4 animals in each group were needed to achieve 80% power to detect meaningful differences with p<0.05. All animal studies were approved by the University Institutional Animal Care and Use Committee. In total, 100 middle-aged (9 – 12 mos) female Wistar rats were used, 80 with OVX, 52 with diabetes, and 60 with mesh implantation. In the OVX groups, outcomes were evaluated at 3- (very early, n=23), 7- (early, n=28), and 42-days (late, n=29) post-surgery. In the no-OVX groups, the outcomes were evaluated at 42 days due to the chronic effect of estrogen (n=20).
Specifically, a polypropylene mesh (Restorelle) was implanted on the anterior and posterior vagina via a modified sacrocolpopexy following supracervical hysterectomy with or without bilateral OVX. Sham surgeries were performed following the same procedures without mesh implantation. Diabetes was induced using a single iv of streptozotocin (STZ) at 45mg/kg. OVX and mesh implantation were performed 2 weeks following the development of hyperglycemia. At different time points following surgery, proximal vaginal tissues were collected for gross morphology and histological analysis. Briefly, the cross-sectional tissue sections at 7µm were obtained following embedding and cryo-sectioning. Masson Trichrome procedures were performed and large images at 100x were taken for overall morphological observation and VSM thickness quantification. The VSM thickness was measured at intervals of approximately 100μm. In samples with mesh implantation, the VSM thickness was measured in areas between mesh fibers and beneath the mesh fibers (mesh impacted) (Figure 1).
Mann-Whitney U, Kruskal-Wallis, and related samples Wilcoxon signed rank tests were used for statistical analysis with p<0.05 set as significance.
Results
All three interventions were successfully implemented. Diabetes was effectively induced with the development of hyperglycemia (≥300mg/dL), polydipsia, and polyuria at 3 days following STZ injection.
Without mesh implantation, OVX induced a decrease of VSM thickness over time in the normoglycemic groups, starting from 7 days post-surgery (Figure 2). Specifically, the VSM thickness at 3 days was not different between the OVX and no-OVX groups with median at 183µm. The value was decreased 45% at 7 days and 49% at 42 days following OVX. In contrast, the change of VSM under the impact of OVX was not significant in the diabetic groups (Figure 2).
Mesh implantation generally induced thinning of VSM in areas beneath the mesh fibers as compared to areas at fiber intervals (p<0.001), with the decrease being 16% and 19% at 7- and 42-days post- implantation (p=0.015, 0.002, respectively). This effect was not significant at 3 days (p=0.721). No differences between VSM thickness at mesh fiber intervals and Shams were observed (all p>0.05). With OVX, the VCM thickness in the normoglycemic rats was decreased 50% in areas at fiber intervals (p=0.039) and 51% in areas beneath the fibers (p=0.003) at 42 days post-surgery, as compared to the no-OVX rats (Figure 1). This effect was not observed in the diabetic rats (all p>0.05). Interestingly, under the impact of diabetes, the VCM thickness in the no-OVX rats was decreased 61% in both areas at fibers intervals (p=0.013) and beneath the fibers (p=0.020) at 42 days post-surgery, as compared to the normoglycemic rats (Figure 1). This effect was not observed in the OVX rats (all p>0.05).
Interpretation of results
This is the first study demonstrating the interactive effect of mesh implantation, menopause, and diabetes on the vaginal smooth muscle structure. Consistent with literature, our findings showed that mesh implantation had a negative impact on the VSM thickness, predominantly affecting the areas directly beneath the mesh fibers. This scenario provides additional evidence that stress shielding, incurred by a soft tissue (vaginal VSM) in contact with a stiff material (polypropylene mesh), might be an underlying mechanism for the mesh implantation associated VSM degeneration. With the longitudinal study design, we further showed that the negative impact of mesh occurred early following the mesh placement and lasted for a long term, supporting a high sensitivity of VSM to mechanical stimulation.
In addition, our results showing that OVX induced VSM atrophy is in line with the literature. Importantly, we showed that the negative impact of OVX on the VSM was deteriorated with mesh implantation in the long term (42 days), suggesting that estrogen plays a protective role in maintaining vaginal structural integrity and that an abrupt decrease of estrogen had a synergistic negative impact on the VSM with mesh implantation.
Interestingly, diabetes did not affect the gross morphology of VSM in both no-OVX and OVX conditions without mesh implantation. The negative impact of diabetes was exhibited only when mesh was implanted under no-OVX condition in the long term. Thus, it is likely that the adverse influence of diabetes on the VSM was executed through a mechanism attenuating the protective effect of estrogen. As a result, the benefits of topical estrogen on the vaginal tissue structure might be compromised in women with diabetes.