|Home | About | Journals | Submit | Contact Us | Français|
To determine the impact of hormones on the biomechanical properties of the vagina and its supportive tissues following surgical menopause in young vs middle aged rats.
Long-Evans rats [4-month virgin (N = 34), 4-month parous (N = 36), and 9-month parous (N = 34)], underwent ovariectomy (OVX) or sham surgery. OVX'd animals received hormones [estrogen (E2) or estrogen plus progesterone (E2 + P4)], placebo, or the Matrix Metalloproteinase inhibitor (CMT-8). Animals were sacrificed after 8 weeks and the biomechanical properties of the vagina and supportive tissues determined. Data was analyzed using a one-way analysis of variance and post-hoc tests.
OVX induced a rapid decline in the biomechanical properties of pelvic tissues in young but not middle aged rats. Supplementation with E2, E2 + P4, or CMT-8 restored tissues of young rats to control levels with no effect on middle aged tissues. Parity did not impact tissue behavior.
OVX has a differential effect on the tissues of young vs middle aged rats.
Pelvic organ prolapse is a common disease in which loss of support to the vagina results in herniation of the bladder, uterus and/or rectum into the vaginal canal and often through the genital hiatus. The primary risk factor for the development of prolapse is vaginal childbirth; however, most women do not develop prolapse until years after childbirth indicating that other factors play a role (1, 2). Age is a defined risk factor as both the incidence and prevalence of prolapse increase with advancing age (2 – 4). The role of menopause in the increase in prolapse in aging women is not clear (5, 6). In part, this is due to lack of a clear definition of menopause and the inherent difficulties in distinguishing age related effects from the effects of gradual hormone deprivation. Most women under the age of 40 and older than 60 are clearly premenopausal and postmenopausal respectively; however, the hormonal status of the majority of women between those ages is highly variable and consequently, difficult to define.
The vagina is supported by the levator ani muscles and the supportive connective tissue attachments between the vagina and the pelvic sidewall (7). A normally supported vagina provides support to the pelvic organs. Failure of vaginal support is thought to result in prolapse (8). At menopause, the steady decline in ovarian function is associated with a multitude of tissue specific changes including atrophy of the labia, vagina and uterus. Estrogen and progesterone receptors have been identified in the vagina and its supportive vaginal tissues (9, 10). Collagen I is decreased in the supportive tissues following menopause relative to collagen III (11, 12) and the expression of several enzymes that degrade collagen (the Matrix Metalloproteinases) is increased in the absence of estrogen and progesterone and suppressed in their presence (10, 13, 14). However, the precise functional impact of estrogen and progesterone deprivation on the supportive capacity of the vagina and its supportive tissues is not known. One possible mechanism to account for the acceleration of prolapse in older women is a rapid progressive deterioration in the biomechanical properties of the vagina and its supportive tissues due to a decline in estrogen and progesterone with menopause.
In this study, we sought to clearly define the impact of loss of the ovarian hormones estrogen and progesterone on the biomechanical properties of the vagina and supportive tissues. To overcome the limitations associated with the use of small tissue biopsies from humans, we employed a rat model previously developed by our lab in which the biomechanical properties of the vagina and the adjacent supportive tissues can be tested as a complex (without disruption of insertion and attachment points) under loading conditions that simulate downward distension (15). To clearly define menopausal status, we used surgically ovariectomized animals. We hypothesized that following ovariectomy (OVX), the biomechanical properties of the vagina and supportive tissues measured as a decreased ultimate load at failure (the maximum force a specimen can withstand) and a decrease in linear stiffness (the ability of the tissue to resist permanent deformation) would decline. We then aimed to determine whether supplementation with estrogen and progesterone following ovariectomy could delay or obviate this deterioration. Since matrix metalloproteinases (MMPs) had previously been shown to play a role in degradation of the pelvic floor following menopause, we further asked whether supplementation with an inhibitor of MMPs, the chemically modified tetracycline −8 (CMT-8) would exert a protective effect similar to hormones (16, 17). To determine whether age or parity impacted the magnitude of response to each treatment, we compared the effect of hormones on younger vs middle aged and virgin vs parous rats.
In these studies, we used young cycling 4-month-old rats as a model for maximal connective tissue support in young women aged 20 to 30. Middle-aged rats (9 months) with prolonged cycles were used as a model for middle aged women approaching ovarian senescence. Animals were divided into aged matched virgin vs parous and parity matched young vs middle aged to determine the impact of age and parity on the vagina and supportive tissues.
A total of 110 cycling 4-month (young) and 9-month (middle aged) old Long-Evans rats (Harlan Laboratories, Indianapolis, IN): 4-month virgin (N = 40), 4-month parous (N = 36), and 9-month old parous (N = 34) were obtained and treated according to IACUC guidelines (IACUC #03-003). Animals arrived 2 to 4 weeks prior to treatment to allow time to acclimate and were housed in standard plastic cages where they were allowed free access to water and chow. Animals were maintained on a 12–12 h light-dark cycle. Within each group, the following treatments were performed: sham surgery (incision made and ovary visualized prior to closure); ovariectomy (OVX); OVX followed immediately by supplementation with hormone or placebo (details of hormonal treatments listed below). To control for the effect of surgery and anesthesia, six 4-month old control animals were sacrificed and tested without any intervention. To test whether Matrix Metalloproteinases (MMPs) were involved in the biomechanical decline of the vagina and supportive tissues, 6 OVX'd animals in each group (4 month virgin, 4 month parous and 9 month parous) were administered an inhibitor of MMPs (CMT-8) as a daily oral gavage (15 mg/kg body weight). Eight weeks following intervention, animals were sacrificed. Our decision to sacrifice at 8 weeks was based on a pilot study in which we determined that this was the minimum amount of time following ovariectomy necessary to achieve a 15–20% decrease in the ultimate load at failure and a 25–30% decrease in stiffness in 4- month old rats.
For surgical intervention and hormone supplementation, animals were treated as previously described (18). Briefly, animals were sedated through the inhalation of isoflurane (2–4% flow; Abbott Laboratories, North Chicago, IL) and underwent ovariectomy. Sham surgeries were performed to control for the effects of surgery and anesthesia. Torbutrol (10% v/v; Fort Dodge Animal Health, Fort Dodge, Iowa) was administered to all rats following surgery to reduce pain. Rats undergoing ovariectomy are known to overeat, therefore, all animals were pair fed until the date of sacrifice to achieve similar weights. For supplemented animals, a ninety-day matrix-driven delivery release pellet in the active form of estrogen (E2, 0.1mg) or estrogen plus progesterone (E2, 0.1mg and P4, 10 mg) was implanted in subcutaneous fat immediately following ovariectomy (Innovative Research of America, Sarasota, FL). Six animals in the 4 - month old parous group underwent OVX followed by insertion of a placebo silastic pellet (Innovative Research of America, Sarasota, FL). A chemically-modified tetracycline (CMT-8) at 15mg/kg body weight was dissolved according to manufacturer's instruction (CollaGenex Pharmaceuticals, Newtown, PA), and administered to animals through a curved rounded-end feeding needle (Roboz Surgical Instrument Company, Gaithersburgh, Maryland) two days prior to ovariectomy and continued for the entire treatment duration. Weights, eating habits and activity of all animals were monitored daily to ensure treatments were not compromising health. Prior to sacrifice each rat was weighed. Mid vaginal width (distance from outermost edge of one side of the vagina to the outermost edge of the opposite side) was measured using a laser micrometer (LS-3060T, Keyence Corporation, Osaka, Japan) as previously described (19) on a subgroup of animals. We obtained this measurement to provide objective data of the impact of OVX and OVX with supplementation on gross vaginal tissue morphology since the vagina is a well defined hormonally responsive structure. Control and sham OVX animals were in similar parts of the menstrual cycle (estrous and metestrous) as determined by vaginal smears at the time of sacrifice.
With increased intra-abdominal pressure, the vagina and its supportive tissues act as a collective unit to resist downward distension. Therefore, for this study, all rats were tested according to a protocol previously developed by us in which the vagina and its supportive tissues are tested as a complex with all of the insertion and attachment points left intact (15). By this method, immediately following sacrifice, animals were dissected to isolate the soft tissues and bones of the pelvis and lower spine. Specimens were continually moistened with 0.9% sodium chloride manipulated over an ice bath to prevent tissue breakdown. The hind limbs were amputated with skin, fur, abdominal muscles, and anterior peritoneum. All excess fat was carefully excised. The uterine horns were trimmed leaving the cervix, vagina, its supportive tissues, the pelvic muscles and bones intact. Each dissection was carried out in the same fashion by one of two individuals (NH, KD).
The freshly dissected specimens were then prepared for biomechanical testing. Prior to mounting the specimen, the mid-vaginal width was measured using a laser micrometer (LS-3060T, Keyence Corporation, Osaka, Japan) interfaced to a computer program. For testing, the lower spine was secured in polymethylmethacrylate (PMMA), and placed inside a custom made cylindrical clamp which was fixed to the base of a materials testing machine (SmartTest EMS, Enduratec, Minnetonka, MN; displacement resolution 0.025 mm). The distal 5 mm of the vagina was placed inside a customized soft-tissue clamp which was fixed to a load-cell, (SM-1000N, Interface, Scottsdale, Arizona; resolution 0.015N), and attached to the crosshead of the machine. Specimens were preloaded to 0.15 N, followed by ten-cycles of preconditioning at an distension rate of 25 mm/min between 0 and 2 mm. A uniaxial load to failure test simulating downward distension of the vagina and its supportive tissues was conducted for each specimen following preconditioning at the same rate. Data points were collected every 0.02 seconds using software provided by Enduratec. The data was imported and analyzed in Excel (Excel, Microsoft Corp, Redmond, WA).
The uniaxial load to failure test simulates downward distension of the vagina and supportive tissues by pulling the distal vagina along its longitudinal axis. Thus, the resulting biomechanical parameters represent the force (load) by which the vagina and supportive tissues collectively resist downward distension as they attempt to maintain their normal anatomic relationships. Each load to failure test generates a load distension curve. In this study three biomechanical parameters are derived from the curve: linear stiffness (N/mm), ultimate load at failure (N), and maximal distension (mm). Linear stiffness is defined as the steepest positive slope measured over a 1 mm interval of distension for each specimen. It is a measure of the specimen's ability to resist distension, i.e. maintain normal anatomical relationships. Ultimate load and maximal distension define the point of tissue failure on the load-distension curve. Thus, ultimate load at failure is a measure of the maximum sustainable force of the tissues or maximum resistance to distension, and is defined as the highest load on the load-distension curve prior to tissue disruption. Maximal distension is the distension that corresponds with the ultimate load and describes the distance the tissues could be pulled before tissue disruption.
Based on preliminary data (virgin versus mid-pregnant data), 5 rats were required in each group to achieve 80% power to detect a 30% decrease in ultimate load and linear stiffness at the 0.05 significance level. From each load to failure test a load-distension curve was generated. From this curve, data points corresponding to the linear stiffness (N/mm), ultimate load at failure (N), and maximal distension (mm) were obtained and exported into Excel (Excel, Microsoft Corp, Redmond, WA). All load-distension curves generated in this study were shaped similarly with defined toe, linear and failure regions.
Statistical analyses were performed using SPSS software (Version release 14.0.1, SPSS Inc., Chicago, IL) and evaluated at the 0.05 significance level. Since the values of the skewness and kurtosis statistics did not indicate a significant departure from a normal distribution, biomechanical properties including ultimate load at failure, linear stiffness, and ultimate distension were analyzed using Student's t-tests or one-way analysis of variance, where appropriate. Post-hoc pairwise comparisons between the sham operated animals and the ovariectomized animals, with and without hormone or CMT-8 supplementation, were made using Dunnett's two-tailed t-tests. Multivariable linear regression was used to determine whether weight, age, or parity altered the association between biomechanical properties and ovariectomy, with and without hormone or CMT-8 supplementation.
As shown in Table I, mid vaginal width, measured by a laser micrometer, changed according to hormonal status. This value was lowest in young animals following OVX and OVX with placebo, and highest in young sham operated animals with intact ovaries (Figure I) providing a corroborative quantitative measurement of the impact of OVX on tissue morphology (ie atrophy). Supplementation with CMT-8 did not change mid-vaginal width relative to OVX while treatment with estrogen and progesterone significantly increased width to sham OVX levels. In spite of pair-feeding, OVX young and middle-aged animals overall were heavier compared to the sham OVX group (Tables I – IV). Analysis of weight as an independent variable, however, using multiple variable regression modeling showed that weight did not affect any of the biomechanical outcomes (P > 0.05).
Comparison of young sham operated controls to untreated control animals revealed no differences in linear stiffness, ultimate load and maximal distension at failure indicating that there was no effect of surgery or anesthesia on the biomechanical properties of the vagina and its supportive tissues (Table I). In both the virgin and parous 4-month old animals, OVX induced a a 40% decrease in linear stiffness and a 30% decrease in ultimate load at failure relative to sham OVX signifying a more distensible and weaker tissue complex (Table II). OVX had no effect on the maximal distension of the tissue at the time of failure in either group of 4 month old rats.
As demonstrated in Tables II and III, supplementation with estrogen or estrogen plus progesterone immediately following the removal of the ovaries in either virgin or parous young animals obviated the decrease in linear stiffness and ultimate load that occurred following OVX without supplementation (OVX and OVX plus placebo). Daily oral gavage with CMT-8, a matrix-metalloproteinase inhibitor, following OVX had a similar effect with no difference in linear stiffness and ultimate load at failure in this group relative to the sham OVX'd group. Comparison of young virgin and parous animals demonstrated that parity did not impact any of the biomechanical outcomes (P > 0.05). In Figure 2, representative examples of the load distension curves of 4-month virgin tissues after OVX with and without supplementation are shown.
Age had a significant effect on the functional tissue response to OVX. As shown in Table IV, vaginal width decreased following OVX and was restored to sham OVX values following supplementation with estrogen and progesterone (E2 + P2) similar to young animals but differed in that it was not restored by E2 alone in the middle aged group. When controlling for multiple comparisons, in contrast to the young rats, OVX in the 9 month old parous animals was not associated with a significant decline in ultimate load at failure or linear stiffness. Similarly, supplementation with hormone(s) or CMT-8 had no effect.
Pelvic organ prolapse is a common disease that affects the live of millions of women worldwide. Although menopause has been shown to be a risk factor in multiple large observational studies, the precise functional impact of loss of ovarian function on the pelvic supportive tissues is not known. In the study outlined here, we used a rat model to determine the effect of a surgically induced menopause on the biomechanical properties of the vagina and its supportive tissues tested as a complex using loading conditions that simulate downward distension (mimicking maternal valsalva, cough etc). The primary findings of this study were that although OVX significantly compromised the biomechanical properties of these tissues in young cycling rats, OVX did not have a similar impact on the tissues of middle aged rats. Parity and weight did not impact biomechanical outcomes. Supplementation with hormones (estrogen alone or estrogen and progesterone) or the matrix metalloproteinase inhibitor (CMT-8) prevented the tissue deterioration that occurred with OVX in young animals. With age there was no apparent effect of hormones or CMT-8.
The differential response to ovariectomy in young and middle aged animals indicates that the age at which ovaries are removed may determine the magnitude of the impact on pelvic tissue supportive capacity. In this study, middle aged rats underwent OVX at age 9 months and then the properties of their tissues were tested 8 weeks later. In rats, the onset of irregular menstrual (estrous) cycles occurs between 9 and 12 months indicating a decline in ovarian function. With continued decline in ovarian function, the rat persists in a state of estrous (low estrogen phase) and the vagina becomes cornified. It is likely that the failure of OVX to impact tissue distensibility (stiffness) and the load at which the tissue failed in middle aged rats reflects an alteration in tissue responsiveness to the hormonal milieu with age. As the animal approaches ovarian senescence, it is likely that the specific component of tissues that determine these biomechanical parameters (most likely the subepithelium and muscularis) is less receptive to hormonal stimuli. Interestingly, the epithelium remains responsive to hormones as reflected in the changes in mid-vaginal width which were confirmed by light microscopy (data not shown). Future, biochemical studies are needed to further study the interaction of age and hormones on the structural proteins in these tissues such as collagen and elastin. The current data, however, strongly implicate ovariectomy at a young age without hormone therapy as a risk factor for the development of inferior tissue biomechanical properties and consequently, pelvic organ prolapse in vulnerable tissues.
The rapid decline in biomechanical parameters following OVX in young animals shows that the end result of OVX is a weaker tissue (decreased ultimate load) that is less resistant to tissue distension (decreased linear stiffness). Supplementation with hormones maintained the biomechanical properties at the levels of those observed in sham operated animals demonstrating that hormones inhibit the physiologic basis for the deterioration of the vagina and supportive tissues or induces a compensatory mechanism that counteracts this decline.
Several studies in the literature provide insight into the biochemical mechanism by which hormones may “rescue“ the vagina and supportive tissues. In a study using a primate model, Clark et al demonstrate that following ovariectomy, estrogen therapy increased mRNA expression for collagen subtypes I and collagen III compared to no hormone (11). In addition, the group showed that estrogen increased the expression of cystatin C - an inhibitor of collagen proteases (20). In a previous study of the arcus tendineous fasciae pelvis, a structure known to be important in support of the anterior vaginal wall, we found that the ratio of [collagen I : (collagens III + V)], a general indicator of tensile strength of a tissue, was decreased in postmenopausal women relative to premenopausal women. The altered collagen ratios were not present in postmenopausal women who had been on hormone therapy for at least one year (10). In addition, estrogen suppresses the expression of matrix metalloproteinsaes −2 and −9, enzymes known to be important in the degradation of collagen, in cells derived from the arcus tendineous (14, 21). This is supported by our finding in this study that an inhibitor of MMPs, CMT-8, also maintained the vagina at the sham OVX'd condition. Together, the data suggest that estrogen with or without progesterone exert a beneficial effect on the vagina and supportive by stimulating collagen synthesis and preventing collagen degradation. Future studies will determine whether a similar mechanism accounts for the differences observed in this study.
In our study, the vagina and its supportive tissues were analyzed as a complex (without disruption of insertion and attachment points) in a reproducible and quantifiable manner. Since the vagina and the soft tissues that support the vagina are highly interdependent and act in concert to collectively support the pelvic organs, we believe that this approach is optimal for studying the behavior of these tissues. One of the primary limitations to studying the biomechanical behavior of the vagina and its supportive tissues in humans is that by necessity study samples are limited to small biopsies or tissue that has been excised as part of a prolapse repair (22, 23). While small biopsies provide local data on a specific tissue, they reveal little insight into the mechanism by which this group of tissues (i.e. vagina, paravaginal attachments to the pelvic sidewall, uterosacral ligaments, and perineal membrane) functions as a unit in vivo. Alternatively, rats provide a well-controlled system in which the behavior of the vagina and supportive tissues can be studied completely intact. In this way, mechanistic studies can be performed in a safe time efficient manner. The major limitation is that rats may have differences in connective tissue metabolism than humans and may respond differently to hormones. Historically, however, rats have provided insight into important mechanisms by which the vasculature, uterus and cervix change in response to hormones in pregnancy (18, 24, 25).
In summary, we have used a controlled system to demonstrate a differential impact of ovariectomy on the biomechanical properties of the vagina and supportive tissues in young vs middle aged rats.
The impact of ovariectomy (OVX) differentially affects the biomechanical properties of the vagina and supportive tissues of young vs middle aged rats.
Typical curves from 4-month old virgin rats following sham ovariectomy (OVX), OVX alone, OVX followed by estrogen (E2) or estrogen and progesterone (E2 + P4) or the matrix metalloproteinase inhibitor CMT-8. The peak of the curve represents the load required to induce tissue failure while the slope of the curve represents the linear stiffness of the specimen or its ability to resist an applied load. Both these parameters are highest in sham operated and lowest in OVX'd animals. Supplementation with hormones or CMT-8 restored load and linear stiffness close to sham operated values.
We would like to thank Dr. Brad Zerler of CollaGenex Pharmaceuticals for generously providing CMT-8.
Supported by R01 HD045590
A portion of this work was presented at the 26th Annual Scientific Meeting of the American Urogynecologic Society