Pelvic organ prolapse is a common disease that prevents women from enjoying a full and active life. In the United States, approximately 200,000 women1 have surgery to repair prolapse annually, with a cost to society that exceeds one billion dollars per year.2 Anterior vaginal wall prolapse, clinically known as cystocele, is the most common form of pelvic organ prolapse.3
At present, our understanding of anterior vaginal wall support mechanisms is focused primarily on paravaginal attachments.4,5 Recently two additional observations have been made. First, levator ani muscle damage is seen in women with prolapse.6–9 Second, anterior compartment prolapse is highly correlated (r=0.73) with loss of apical support.10 How these observations help explain anterior compartment failure has been unclear and is a goal of the present study.
A common approach to analyzing how a complex biologic system works is to represent the behavior of interest with a mathematic model. Such models can yield useful mechanistic insights and even predict system responses with reasonable accuracy.11 One class of such models is the “lumped parameter” model in which the relevant features of the biologic system are divided into relatively few subunits. Through simplifying assumptions, the overall properties of each subunit can be approximated by an equation. Hence, the relationship between the tensile force and length of the pubovisceral muscle can most simply be represented by a lumped parameter representing the development of active contractile force and another representing the passive resistance of the muscle to stretch. Overall system behavior is then predicted by assembling and solving the system of equations that represent the various subunits and how they physically interact with one another.
The aim of this study, therefore, was to develop and test a first-generation model to investigate how changes in pelvic floor muscle support and apical connective tissue support interact to affect anterior vaginal wall prolapse under conditions of raised intra-abdominal pressure.
MATERIALS AND METHODS
A sagittally-symmetric, lumped parameter, biomechanical model of the anterior vaginal wall and its support system was developed based on our previous anatomic work.5,12–14 As a first step, the dimensions and orientation of the anterior vaginal wall and its support system were measured from midsagittal plane magnetic resonance scans of 10 healthy volunteers. This sample size was judged sufficient, based on the authors' previous experience, to yield representative geometry. These women were recruited from 2001 to 2003 as controls in an case–control study, approved by University of Michigan IRBMED, comparing findings in women with normal support to women with pelvic organ prolapse. They denied having incontinence or prolapse symptoms and had demonstrated normal support on pelvic organ prolapse quantification system examination15 (no vaginal wall point lower than 1 cm above the hymen). These women had a mean age of 58.8 years (±standard deviation of 12.2 years) and parity of 2.45 (±1.03). Axial, sagittal, and coronal magnetic resonance images of the pelvic floor region were taken at 5-mm intervals as previously described.16 The dimensions and orientation of the anterior vaginal wall and levator plate were measured in the midsagittal plane at rest. The mean length of the vaginal wall was 7.0 (±1.0) cm, with a mean vaginal inclination angle of 62.4° (±7.8°). The mean length of the levator plate was 10.4±0.9 cm, with a mean levator inclination angle of 47.9° (±4.8°).
A three-dimensional model of the levator ani muscles, vagina, rectum, cardinal and uterosacral ligaments (“apical supports”), pubic bone, and sacrum was created from a representative subject's magnetic resonance (MR) images using 3D Slicer 2.1b1 (open-source software available at http://www.slicer.org/), as previously described.17,18 The attachments of the levator ani muscle to the pubic bone and the lines of action of the apical supports were located based on three-dimensional reconstructions. A two-dimensional, sagittally-symmetric, lumped parameter, biomechanical model was then created by projecting the three-dimensional geometry of anterior vagina wall and its support system onto the midsagittal plane (Fig. 1). The anterior vaginal wall was modeled as a deformable and stretchable membrane lying on the rectum, which was supported by the levator plate. The anterior vaginal wall was fixed at the perineal membrane, which was assumed to be stationary during Valsalva. For simplicity, the levator plate was modeled as a rigid “trap door” hinged on the sacrum. To isolate the anterior vaginal wall effects, no relative motion was assumed to occur between the rectum and the levator plate. The pubovisceral portion of the levator ani muscle was considered to control levator plate inclination with respect to the pelvis and was modeled using both an active contractile element and a passive elastic spring element (Fpvm=Factive+Fpassive).
Apical support, provided to the vagina in Level I by the cardinal ligament and the uterosacral ligament, was represented by two passive elastic springs (Fig. 1) attached to the top of the vaginal wall. An exponential relationship between force and elongation for anterior vaginal wall membrane and all passive springs (including passive material property of pubovisceral muscle and the apical supports) was assumed to present the nonlinear (hyperelastic) material property of biologic tissue, that is,
where C1 and C2 are the material property parameters, l is the current length, and l0 is the initial length. The material property parameters of the vaginal wall were held constant through these simulations to focus on the interaction of apical supports and levator activity. The vaginal wall, pubovisceral muscle and apical supports material properties used were based on values from existing literature.19,20 Table 1 shows the dimension and material property of each model element.
The model was implemented in Matlab (The MathWorks Inc., Natick, MA). To model a Valsalva maneuver, the model abdomen was pressurized with 70 cm H2O of intra-abdominal pressure to represent a maximal Valsalva maneuver.21 In a woman with normal support, clinical observations show that a maximal Valsalva maneuver causes little displacement of the pelvic floor if the pelvic floor muscles are actively contracted so as to equilibrate the downward intra-abdominal pressure acting on the levator plate.22 Therefore, in our model we assumed that the intact pubovisceral muscle can increase its contractile force sufficiently above the baseline resting tone so as to maintain hiatus closure under increased intra-abdominal pressure, as observed in women with normal support. Therefore, the model anterior vaginal wall was considered to be completely supported by the rectum and levator plate, without tension on the cardinal and uterosacral ligaments. The mean maximal vaginal closure force was assumed to be 2.8 N, which is the average value reported in the literature23,24 (Fig. 2A). Because the force exerted by the pubovisceral muscles holds the trap door or urogenital hiatus closed, intra-abdominal pressure cannot induce inferior–superior tension in the anterior vaginal wall because there is no pressure differential being generated.
To use the model to simulate the two mechanisms underlying prolapse, we modeled the pubovisceral impairment (indicated by Defect) by simulating loss of both the pubovisceral muscle's active contractile force (Factive) and also its passive spring constant (derived its passive resistance to stretch in the uncontracted state), that is,
Impairment of the connective tissue apical support (indicated by Defect) was modeled as a decrease in their elastic properties, that is
Model simulations were run with impairments of 0%, 20%, 40%, 60%, 80 % decrements in the maximal force that the pubovisceral muscle could generate and 0%, 50%, 90% increases in the elasticity of the apical support. Because we focused on the interaction between pubovisceral muscle and apical supports, the material property of the vaginal wall was held constant in all simulations. The primary outcome measure was the change in anterior vaginal wall curvature showing anterior vaginal wall deformation, and secondary outcome measures were the downward angular displacement of the levator plate and the downward translation of the top of the anterior vaginal wall under the action of 70 cmH2O intra-abdominal pressure. We used the descent of the most dependent point of the vaginal wall as a measure of prolapse size, as shown in Figure 2B.
General model behavior under load is shown in Figure 2B. When the system with impaired muscle and connective tissue was loaded with maximal intra-abdominal pressure (Valsalva), downward rotation of the levator plate occurred when the defective pubovisceral muscle could no longer generate sufficient force to counterbalance the abdominal pressure increase. In this situation, the model trap-door begins to “open,” such that the lower part of the anterior vaginal wall is no longer supported by the levator ani plate. The unsupported distal region of vaginal wall becomes the tensile structure that separates a higher intra-abdominal pressure zone and a lower atmospheric pressure zone. This pressure differential causes the downward “bulging” (deformation) of the distal anterior vaginal wall in the areas not supported by the levator ani muscle. This, in turn, induces a superior–inferior tension force in the upper anterior vaginal wall. The apical supports are now placed under tension as they resist the downward movement of the upper vaginal wall until a new mechanical equilibrium is achieved in the system.
The specific simulation results with different combinations of pubovisceral and apical support impairment are shown in Figure 3. Here, the degree of anterior vaginal wall descent is shown as the distance below the perineal membrane, which lies at the level of the hymenal ring. For the results in the top row (Fig. 3A, 3B, and 3C), the apical supports had the normal material parameter constants. The bottom row shows the results with 90% apical support impairment. In the first column (Fig. 3A and 3D), the pubovisceral muscle has normal active contractile function and normal passive material properties, whereas the second and third columns show the effects of increasing pubovisceral muscle impairment by 60% and 90%, respectively.
When the pubovisceral muscle has normal properties (Fig. 3A), no anterior wall deformation is seen. With pubovisceral impairment (Fig. 3B and 3C), some degree of anterior vaginal prolapse occurs. With normal muscles, but impaired apical support (Fig. 3D), no displacement occurs, because the vaginal wall is not subjected to a pressure differential.
The interaction between pubovisceral muscle defect and apical support impairment is shown in the bottom row (Fig. 3E and 3F). When the apical support impairment was superimposed on the pubovisceral muscle impairment, there was a loss of resistance to the descent of apical anterior vaginal wall, resulting in a much larger prolapse. The prolapse size, D, defined as the largest descent of the most independent point, is also shown in Figure 3. A 90% impairment of apical support led to an increase in anterior wall prolapse from 0.3 cm to 1.9 cm (a 530% increase) at 60% pubovisceral muscle impairment (Fig. 3C compared with 3F), and from 0.7 cm to 2.4 cm (a 240% increase) at 80% pubovisceral muscle impairment (Fig. 3B compared with 3E).
The prolapse size is shown as a function of apical support impairment for different degrees of pubovisceral impairment in Figure 4. The size of the prolapse increased exponentially with increasing apical support impairment.
To verify that the simulation results were reasonable approximations of the interactions between connective tissue and muscular support of vaginal wall, these findings were compared with a typical magnetic resonance image of a patient with cystocele in both the resting state as well as under a maximal Valsalva (Fig. 5). In the resting state, the patient has a relatively normal configuration. But with increasing intra-abdominal pressure associated with a Valsalva maneuver, the levator plate is pushed posteriorly and downwards to a more vertical orientation, the anterior vaginal wall is pushed inferiorly by way of the hiatus, whereas the uterus is pulled down by anterior vaginal wall. This scenario matches well with our model simulation, thereby helping to validate the behavior of our model.
This study presents the development of a simple two-dimensional model that was used to investigate how anterior vaginal wall muscular and connective tissue supports affect the size of anterior vaginal wall prolapse. The model is suitable for examining first-order effects of the interaction between these supportive elements in the presence of impairments in the pubovisceral muscle and apical support tissues. The model is focused on anterior vaginal wall prolapse because it is the most common form of pelvic organ prolapse3 and a form that can be used to initiate meaningful investigation into muscle–connective tissue interaction. We chose to focus on the apical connective tissue supports as a result of recent data10 showing a strong correlation between apical descent and anterior compartment prolapse, recognizing that further work will be needed to add other elements of the support system. This model shows important qualitative results on how a combination of muscular and connective tissue impairments affects prolapse magnitude. The resulting anterior wall prolapses also mimic those seen clinically, thereby providing validity for our model.
The clinical significance of this article lies in its demonstration of the interaction between apical connective tissue support and the levator ani muscle impairments that result in anterior vaginal wall prolapse. This links research showing that the levator muscles are abnormal in women with pelvic organ prolapse6–9 with observations of connective tissue abnormalities in pelvic organ prolapse.4,14,25 Our simulations also suggest that to some degree normal muscle and normal connective tissue function can compensate for injury to the other system. It is when both systems have impairments that the larger anterior vaginal wall prolapse is predicted. This can partly explain why women with levator ani muscle defect are more likely to have prolapse,9 but not all women with levator ani muscle defects have developed anterior vaginal wall prolapse.
When the intact pubovisceral muscle maintains urogenital hiatus closure, there is no tension in the vaginal wall, because it is well supported inside the abdominal cavity by the levator plate, and therefore no pressure differential acts across it. This assumption is based on clinical observations in nulliparous women that reveal minimal descent of the pelvic floor during Valsalva when the muscles are contracted. When the pubovisceral muscle is defective and its contractile force is reduced, then the model predicted that less hiatal closure force is generated, which agrees with clinical measurements showing a 30% reduction in force among women with stress incontinence.24 At a certain level of pubovisceral defect, the muscles are no longer able to maintain hiatal closure, and the distal vagina is now exposed to a pressure differential as it is interposed between the high abdominal pressure zone and lower atmospheric pressure. This pressure differential deforms the vaginal wall, driving it inferiorly toward the low pressure region. The behavior of our model extends previous theoretical work of the concept of the “pelvic valve.”26
It is important to recognize that there are inherent limitations in this type of research. In this first simulation we chose to focus the analysis on the anterior vaginal wall and have made some simplifying assumptions. For example, it was assumed there is no deformation of levator plate or rectal movement relative to the levator plate. Second, this model is focused on the interaction between muscular support and apical suspension on anterior vaginal wall support. In this two-dimensional model, the paravaginal supports of the anterior vaginal wall that attach laterally to the pelvic sidewall were not specifically modeled because this requires a more complex, 3-dimensional, approach. From a structural mechanics point of view, paravaginal support adds a reinforcing element to the mid portion of the vaginal wall, which will reduce its descent through the hiatus. In our model, we found that we needed to adjust the property of the vaginal wall tissue to be 10 times stiffer than published experimental results to obtain reasonable model results. This need may reflect the lack of paravaginal support. Also, stretch is not necessarily uniform along each element of the model as assumed; it can vary locally along and across the elements, especially if the thickness varies. The effect of nonuniform stretch in these structures requires further investigation, particularly if systematic regional variations in properties are found. Finally, the elements in the model were assumed to have hyperelastic material properties, with increasing resistance to stretch as is indeed found in many collagenous structures, but it is a limitation that no time-dependent behavior was represented. Despite these simplifications, we believe this simple model captures the qualitative behavior of the system and yields useful predictions. In the future, distributed parameter models, such as finite element models, may prove to be more accurate and better for the three-dimensional analyses that are needed. This is a first attempt to use biomechanical modeling to study the mechanics of pelvic organ support. The model offers an opportunity to analyze fundamental concepts related to how injury of levator ani muscle and impairment of connective tissue can lead to anterior vaginal wall prolapse.
1. Boyles SH, Weber AM, Meyn L. Procedures for pelvic organ prolapse in the United States, 1979–1997. Am J Obstet Gynecol 2003;188:108–15.
2. Subak LL, Waetjen LE, van den Eeden S, Thom DH, Vittinghoff E, Brown JS. Cost of pelvic organ prolapse surgery in the United States. Obstet Gynecol 2001;98:646–51.
3. Hendrix SL, Clark A, Nygaard I, Aragaki A, Barnabei V, McTiernan A. Pelvic organ prolapse in the Women's Health Initiative: gravity and gravidity. Am J Obstet Gynecol 2002;186:1160–6.
4. Richardson AC, Lyon JB, Williams NL. A new look at pelvic relaxation. Am J Obstet Gynecol 1976;126:568–73.
5. DeLancey JO. Anatomic aspects of vaginal eversion after hysterectomy. Am J Obstet Gynecol 1992;166:1717–24.
6. Tunn R, Paris S, Fischer W, Hamm B, Kuchinke J. Static magnetic resonance imaging of the pelvic floor muscle morphology in women with stress urinary incontinence and pelvic prolapse. Neurourol Urodyn 1998;17:579–89.
7. Singh K, Jakab M, Reid WM, Berger LA, Hoyte L. Three-dimensional magnetic resonance imaging assessment of levator ani morphologic features in different grades of prolapse. Am J Obstet Gynecol 2003;188:910–5.
8. Hoyte L, Schierlitz L, Zou K, Flesh G, Fielding JR. Two- and 3-dimensional MRI comparison of levator ani structure, volume, and integrity in women with stress incontinence and prolapse. Am J Obstet Gynecol 2001;185:11–9.
9. Hoyte L, Jakab M, Warfield SK, Shott S, Flesh G, Fielding JR. Levator ani thickness variations in symptomatic and asymptomatic women using magnetic resonance-based 3-dimensional color mapping. Am J Obstet Gynecol 2004;191:856–61.
10. Summers A, Winkel LA, Hussain HK, DeLancey JO. The Relationship between Anterior and Apical Compartment Support. Am J Obstet Gynecol 2006;194:1438–43.
11. Alexander RM. Modelling approaches in biomechanics. Philos Trans R Soc Lond B Biol Sci 2003;358:1429–35.
12. DeLancey JO. Structural support of the urethra as it relates to stress urinary incontinence: the hammock hypothesis. Am J Obstet Gynecol 1994;170:1713–20.
13. DeLancey JO. Structural anatomy of the posterior pelvic compartment as it relates to rectocele. Am J Obstet Gynecol 1999;180:815–23.
14. DeLancey JO. Fascial and muscular abnormalities in women with urethral hypermobility and anterior vaginal wall prolapse. Am J Obstet Gynecol 2002;187:93–8.
15. Bump RC, Mattiasson A, Bo K, Brubaker LP, DeLancey JO, Klarskov P, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–7.
16. Chou Q, DeLancey JO. A structured system to evaluate urethral support anatomy in magnetic resonance images. Am J Obstet Gynecol 2001;185:44–50.
17. Chen L, Hsu Y, Ashton-Miller JA, Delancey JO. Measurement of the pubic portion of the levator ani muscle in women with unilateral defects in 3-D models from MR images. Int J Gynaecol Obstet 2006;92:234–41.
18. Hsu Y, Chen L, Delancey JO, Ashton-Miller JA. Vaginal thickness, cross-sectional area, and perimeter in women with and those without prolapse. Obstet Gynecol 2005;105:1012–7.
19. Yamada H. Strength of biological materials. Baltimore (MD): Williams & Wilkins; 1970. p. 205–70.
20. Bartscht KD, Delancey JO. A technique to study the passive supports of the uterus. Obstet Gynecol 1988;72:940–3.
21. Howard D, Miller JM, DeLancey JO, Ashton-Miller JA. Differential effects of cough, Valsalva, and continence status on vesical neck movement. Obstet Gynecol 2000;95:535–40.
22. Shafik A, Doss S, Asaad S. Etiology of the resting myoelectric activity of the levator ani muscle: physioanatomic study with a new theory. World J Surg 2003;27:309–14.
23. Morgan DM, Kaur G, Hsu Y, Fenner DE, Guire K, Miller J, et al. Does vaginal closure force differ in the supine and standing positions?. Am J Obstet Gynecol 2005;192:1722–8.
24. Morin M, Bourbonnais D, Gravel D, Dumoulin C, Lemieux MC. Pelvic floor muscle function in continent and stress urinary incontinent women using dynamometric measurements. Neurourol Urodyn 2004;23:668–74.
25. Jackson SR, Avery NC, Tarlton JF, Eckford SD, Abrams P, Bailey AJ. Changes in metabolism of collagen in genitourinary prolapse. Lancet 1996;347:1658–61.
26. Porges RF, Porges JC, Blinick G. Mechanisms of uterine support and the pathogenesis of uterine prolapse. Obstet Gynecol 1960;15:711–26.