Adaptations of the Rat Vagina in Pregnancy to Accommodate Delivery

Daucher, James A. MD1; Clark, Katherine A. BA2; Stolz, Donna B. PhD2; Meyn, Leslie A. MS1; Moalli, Pamela A. MD, PhD1

doi: 10.1097/01.AOG.0000246798.78839.62
Original Research

OBJECTIVE: To characterize ultrastructural changes in the rat vagina in pregnancy, delivery, and postpartum, focusing on collagen architecture and smooth muscle cell morphology.

METHODS: The vagina of four virgin, four midpregnant, four late pregnant, four immediate, and four late post–vaginal-delivery rats were examined by transmission electron microscopy. Images were classified into one of four categories based on collagen fibril area fraction, with group 1 containing the highest number of collagen fibers per unit area and group 4 containing the lowest. Smooth muscle cells were characterized into three cell types (“synthetic,” “intermediate,” and “contractile”) based on the volume fraction of cytoplasm occupied by organelles compared with myofibrils.

RESULTS: Quantitative analysis demonstrated that 76% of collagen fibers in virgin rats were categorized as group 1 or 2 compared with 49% in midpregnant, 40% in late pregnant, and 23% in immediate postpartum animals (P=0.006). Late postpartum tissue seemed similar to virgin tissue (77%). Midpregnant (37%), late-pregnant (34%) and immediate postpartum animals (43%) contained a higher proportion of synthetic smooth muscle cells compared with virgins (20%) and late postpartum animals (21%) (P=.02). Contractile smooth muscle cells predominated in virgin (64%) and late postpartum animals (70%) compared with midpregnant (42%), late pregnant (50%) and immediate postpartum (50%, P=.05).

CONCLUSION: In pregnancy, collagen fiber area decreased while smooth muscle cells transformed from a contractile to a synthetic phenotype. The late postpartum period returned to prepregnant levels for both collagen and smooth muscle cell morphologies. It is likely that these changes represent adaptations to minimize trauma to the vagina during passage of the fetus.

The vagina undergoes significant ultrastructural changes during pregnancy and at the time of delivery that facilitate passage of the fetus.

From the 1Magee-Womens Research Institute and the Department of Obstetrics & Gynecology at Magee Womens Hospital; and 2Center for Biological Imaging, Pittsburgh School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania.

See related article on page 136.

This study was supported by the National Institute of Child Health and Human Development—RO1 NIH HD045590.

Corresponding author: Pamela A. Moalli, MD, PhD, Magee-Womens Research Institute, 300 Halket Street, Pittsburgh, PA 15213; e-mail:

Article Outline

Pelvic organ prolapse is a disfiguring and disabling condition that affects millions of American women.1 Women with pelvic organ prolapse often suffer from chronic pelvic pressure, sexual dysfunction and urinary and fecal incontinence.2 In the United States alone, approximately 200,000 women undergo surgery for prolapse each year, with an estimated direct medical cost that exceeds 1 billion dollars annually.3 Unfortunately, very little is known of the pathogenesis of prolapse.

In addition to the levator ani muscles, the vagina plays a critical role in providing support to the pelvic organs.2,4 Vaginal support arises from the connective tissue attachments between the vagina and the pelvic sidewall and the vaginal wall itself. The vagina and supportive connective tissues are composed of a variable amount of collagen, elastin, and smooth muscle.5 The well-supported vagina in turn provides support to the urethra, bladder, uterus, and rectum. Failure of vaginal support results in prolapse of the pelvic organs.6

Histologically, the vaginal wall consists of four distinct layers: 1) a superficial layer of stratified squamous epithelium, 2) a subepithelial layer of dense connective tissue composed primarily of collagen and matrix, 3) a layer of smooth muscle (the muscularis), and 4) an adventitia composed of loose connective tissue. The subepithelium is comprised primarily of collagen type III and confers the biomechanical properties of distensibility and flexibility to the vagina.7–9 The vaginal muscularis is composed of an inner circular and an outer longitudinal layer of smooth muscle.

In the nonpregnant state, vaginal smooth muscle is thought to be under a constant contractile tone,10 allowing the vagina to maintain its shape in response to increases in intra-abdominal pressures and other environmental stimuli.11 Smooth muscle cells can alter their phenotype from a contractile to a synthetic one.12,13 Synthetic smooth muscle cells secrete matrix and have no contractile tone whereas contractile smooth muscle cells contain abundant myofilaments and function as contractile entities.14,15

The number of vaginal deliveries a woman experiences during her lifetime significantly increases her risk of developing pelvic organ prolapse.4 The exact pathophysiology behind this association is not clear. One explanation is for vaginal delivery to occur with minimal maternal injury, specific adaptations must take place within the vagina and its supportive tissues that allow for safe passage of the fetus. If these adaptations are incomplete (inadequate tissue preparation) or exceeded (fetus too large for passageway), then injury ensues.

In this study, we sought to increase our understanding of the adaptations that take place in the vagina that afford passage of a fetus with minimal maternal injury. As a measure of the adaptive process, we focus on changes in collagen microarchitecture and smooth muscle cell morphology in the vaginal subepithelium and muscularis. Specifically, we quantitate changes in the collagen fibril-to-matrix ratios (fibril area fraction) and the conversion of smooth muscle myocytes from a contractile to a synthetic phenotype. We have chosen to study this process in the rat, because it is highly adapted to efficient reproduction over its lifespan, delivering multiple fetuses with little apparent injury (adaptations are rarely exceeded). Ultimately, by increasing our understanding of the adaptations in the rat, we hope to improve our interpretation of the human condition in which we believe injury occurs as a result of maternal adaptations that are incomplete or exceeded.

Back to Top | Article Outline


A total of 20 rats (Long Evans, Charles River Laboratories Inc. Wilmington, MA) aged 3 months were killed in accordance with Institutional Animal Care and Use Committee guidelines. The study was approved by the Institutional Animal Care and Use Committee at Magee-Womens Hospital. Rats were grouped according to gestational age and included 4 virgin, 4 midpregnant (day 12–14), 4 late pregnant (day 20–22), 4 immediate postpartum (0–4 hours) and 4 late postpartum (3 weeks) animals. The midvagina was excised and divided into two halves along the longitudinal axis. One half was imbedded in Optimal Cutting Temperature media (Sakura Finetek Inc., Torrance, CA), cut into 5–7-μm sections and stained with Gomori trichrome for preliminary analysis of gross morphologic features. The second half was double-fixed in 2.5% glutaraldehyde and 1% osmium tetra oxide, dehydrated, and embedded in resin. Thick sections (300 nm), stained with toluidine blue, were analyzed for gross morphologic features (cell density, thickness of epithelium, appearance of collagen and matrix) and orientation for ultrathin sections. Ultrathin sections (50–60 nm) were cut with a glass knife on a Reichert Ultracut ultramicrotome (Reichert, Inc., Depew, NY), collected on copper grids, stained with uranyl acetate and lead citrate and examined by transmission electron microscopy using a JEOL transmission electron microscope 1210 EX (JEOL, Peabody, MA). All specimens were systematically examined from the stratified epithelium through the subepithelium and muscularis to the adventitia.

For collagen analysis, serial measurements were taken in the matrix of the subepithelium. Four categories of collagen fiber formation were established based on fibril area fraction at 80,000x, with group 1 containing the highest number of collagen fibers per unit area (fibril area fraction) and group 4 containing the least. Specifically, group 1 contained more than 80% fibril area relative to matrix; group 2 contained 60–79% fibril area fraction; group 3 contained 40–59% fibril area fraction, and group 4 contained less than 40% fibril area fraction (Fig. 1). Quantification of fibril area fraction was performed using Metamorph threshold-imaging software (Molecular Devices Corp., Chicago, IL). Fifty images per animal (1,000 images total) were viewed under transmission electron microscopy, classified into one of the four groups, and recorded as a median percentage.

Smooth muscle analysis was performed using transmission electron microscopy at 8,000. Three smooth muscle cell types were defined based on the volume fraction of cytoplasm occupied by organelles compared with myofibrils. This classification system of smooth muscle cells is a modification of a previously described method by Chen et al.12 Cells with more than more than 75% of their cytoplasmic volume occupied by myofibrils were termed “contractile,” whereas those with more than 75% of their cytoplasmic volume occupied by organelles were termed “synthetic.” All other cells were termed “intermediate” (Fig. 2). Cytoplasmic volume percentages were quantified using Metamorph threshold-imaging software. Four sequential quadrants per animal were examined within the muscularis for smooth muscle morphology, with cell types being classified into one of the three designated groups. A total of 80 images were analyzed. Totals were recorded as a median percentage.

Two individuals (J.D. and K.C.) each performed a separate analysis of the 1,080 images. The interrater reliability was 90%. Data were analyzed using SPSS 13.0 statistical software (SPSS Inc., Chicago, IL), and statistical tests were evaluated at the .05 significance level. Due to the small sample size, nonparametric statistical methods were employed. Differences in cell type proportions between gestational age groups were evaluated using the Kruskal-Wallis test. Sample size was determined from previously published data demonstrating a 30% decrease in biomechanical properties between virgin and pregnant rats.16 The median percentage of contractile smooth muscle in 4 virgin rats was 64. Based on these data, 4 rats in each group would provide 80% power to detect at least a 30% difference in cell type percentage at the .050 two-tailed significance level.

Back to Top | Article Outline


Baseline characteristics obtained for the animals used in the study included median animal weight (virgin: 245 g; midpregnant: 267 g; late pregnant: 350 g; immediate postpartum: 260 g; late postpartum: 285 g), median number of pups per litter (11), and estrous cycle staging. Three of the virgin rats were in the metestrous phase (low estrogen and progesterone levels) and one was in the proestrous phase (peak estrogen and progesterone levels) of the estrous cycle. As can be seen in the trichome stained cross sections of the vagina (Fig. 3B), pregnancy resulted in progressive changes in the vaginal wall. In both midpregnant and late pregnant animals there was an increase in the number of papillae, resulting in an increase in vaginal surface area. The epithelium became mucified with mucous secreted into the vaginal lumen. There was an increase in matrix deposition in the subepithelium and muscularis.

Transmission electron microscopy analysis confirmed the changes in the vaginal subepithelium and muscularis seen by light microscopy. Transmission electron microscopy studies of collagen formation demonstrated that collagen fibers within the subepithelium of the virgin rats were composed of highly aligned fibrils within compact fiber bundles, with little matrix between fibrils (Fig. 1A). In pregnancy, the fibrils and fibers were less densely packed, with an increase in matrix between fibers and between fiber bundles (Fig. 1B and 1C). Postpartum animals contained fibers that were the least organized, with extensive amounts of extracellular matrix present (Fig. 1D). Late postpartum collagen morphology resembled virgin tissue.

Quantitative analysis (Table 1) showed that in the virgin rat vaginal tissue, 76% of the collagen fibers were categorized as group 1 or 2 compared with 49% in the midpregnant, 40% in late pregnant, and 23% in the immediate postpartum animals (P=.006). Because collagen groups 1 and 2 were not statistically different in any stage of pregnancy or postpartum, we merged them in the final statistical analysis. In contrast, collagen fibers categorized as group 3 predominated in the midpregnant (38%) and late (41%) pregnant animals (P=.05), whereas group 4 predominated in the immediate postpartum animals (41%; P=.004). The late postpartum group showed a profile similar to virgin tissue, with 77% of collagen fibers categorized as group 1 or 2. Thus, the data support a decrease in the ratio collagen relative to matrix in pregnancy reaching its lowest values immediately postpartum, with a complete recovery to baseline in the late postpartum period.

Smooth muscle cell architecture was also noticeably altered as pregnancy progressed. Virgin smooth muscle cells were tightly spaced with extensive gap junction visible between cells (Fig. 2A). Myofibrils within contractile smooth muscle cells were compact and uniform with numerous dense bodies (focal condensations within the cells cytoplasm that serve to hold and organize actin and myosin filaments). During pregnancy, smooth muscle cells appeared hypertrophied and became more diffusely spaced. There was an increase in the proportion of synthetic smooth muscle cells, characterized by prominent cytoplasmic organelles, including mitochondria, Golgi complexes, rough sarcoplasmic reticulum, and cytoplasmic vacuoles (Fig. 1B and 1C). Synthetic smooth muscle cells localized to the periphery of the muscularis layer and were surrounded by increased amount of matrix material. Smooth muscle cells within the immediate postpartum animals appeared hypertrophied, with a return of condensed and compact myofibrils within contractile cells. The prominent features of the synthetic smooth muscle cells phenotype were reduced in the immediate postpartum animals. Smooth muscle cells in the late postpartum period appeared similar to smooth muscle cells in the virgin tissue.

Quantitative analysis of the smooth muscle cell phenotypes, shown in Table 2, lists the median percentage of smooth muscle cell types within each animal group (see methods for definitions of contractile, intermediate, and synthetic). Contractile smooth muscle cells showed a significant decrease during pregnancy (mid and late) and the immediate postpartum period when compared with virgin rats (42% compared with 50% and 50% compared with 64%; P=.05), with a return to prepregnant levels in the late postpartum period (70%). Synthetic smooth muscle cells increased in pregnancy (mid/late) and the immediate postpartum period when compared with virgin (37% compared with 34% and 43% compared with. 20%; P=.02), with a return to prepregnant levels in the late postpartum period (21%). Intermediate smooth muscle cells showed no statistically significance difference among groups.

Back to Top | Article Outline


To accommodate the passage of a fetus with minimal maternal injury, the vagina must undergo significant changes during pregnancy and at the time of delivery. This study represents an in-depth descriptive and quantitative analysis of the histologic and ultrastructural adaptations of the rat vagina in pregnancy to achieve delivery. We demonstrate that vaginal tissue during pregnancy and at the time of delivery, is characterized by profound cellular and extracellular alterations throughout the vaginal wall relative to virgin tissue. The histologic changes during pregnancy include thickening and mucification of the epithelium and increased extracellular matrix material within the subepithelium and muscularis.

Ultrastructurally, collagen formation in virgin tissue contains highly compact collagen fibers with little extracellular matrix between fibers. During pregnancy, collagen becomes less densely packed with increased matrix between fibrils and fiber bundles. During the immediate postpartum period, collagen fibers are the least organized, lack orientation, and have extensive amounts of extracellular matrix present. The morphology of smooth muscle cells in pregnancy and the immediate postpartum period is also altered. Although virgin vaginal tissue is characterized by a predominance of contractile smooth muscle cells composed of filamentous material with few cytoplasmic organelles, in pregnancy there is an increase in the proportion of synthetic smooth muscle cells. In the postpartum period there is a rapid return of contractile smooth muscle cells.

Previous studies have demonstrated that collagen and smooth muscle cells provide structural support to tissues.17 In this study, we hypothesize that collagen within the subepithelium and smooth muscle of the muscularis are most likely the constituents of the vagina that mediate the changes in pregnancy that allow vaginal distension and passage of a fetus. Conversely, damage to these constituents at the time of delivery when adaptations are exceeded may ultimately contribute to the diminished supportive capacity that is present in women with prolapse.

It is well known that within a tissue, collagen is a key determinant of tissue tensile strength. Within a collagen bundle, however, the microarchitectural packing of fibrils also contributes to the biomechanical behavior of the tissue.18 For example, a decrease in the compactness of collagen fibrils (decreased fibril area fraction) allows for greater tissue elasticity and distensibility, whereas compactly packed collagen fibrils increases tissue strength and rigidity.19 Manabe and Yoshida20 performed a descriptive analysis of partial-thickness vaginal tissue from women before, during, and immediately after vaginal delivery using light and electron microscopy. They observed an increase in collagen dispersion in pregnancy as compared with the nonpregnant tissue. Zhao et al,21 using a relaxin knockout mouse model, showed that histologic samples of pregnant wild-type mice have a much looser pattern of collagen when compared with virgin samples but no difference in total collagen content.

Smooth muscle cells are a highly specialized cell whose principal function is contraction and regulation of tone. In the contractile state, smooth muscle cells proliferate at an extremely low rate and exhibit low synthetic activity.22 Smooth muscle cells express a unique repertoire of contractile proteins, ion channels, and signaling molecules that are required for the cell's contractile function.23 Unlike skeletal or cardiac muscle that is terminally differentiated, smooth muscle cells within adult animals retain remarkable plasticity and can undergo profound and reversible changes in phenotype in response to changes in local environmental cues.23 An example of smooth muscle cells plasticity can be seen during vascular development when the smooth muscle cell plays a key role in morphogenesis of the blood vessel and exhibits high rates of proliferation, migration, and production of extracellular matrix components such as collagen, elastin, and proteoglycans.24 Similarly, in response to vascular injury, the smooth muscle cell dramatically increases its rate of cell proliferation, migration, and synthetic capacity and plays a critical role in vascular repair.25 Defects in smooth muscle cell differentiation are seen in many disease states, including asthma,26 obstructive bladder disease,27 gastrointestinal disorders,28 and many forms of cancer.29

The ability of smooth muscle cells to undergo complex changes in the cellular function is referred to as “phenotypic modulation” or “phenotypic switching,” where smooth muscle cells alter their appearance and function between a contractile phenotype and a synthetic phenotype.29 Morphologically, contractile smooth muscle cells contain numerous myofilaments with few cytoplasmic organelles, whereas synthetic smooth muscle cells are characterized by an extensive synthetic apparatus,29 including large amounts of smooth and rough sarcoplasmic reticulum, Golgi complexes, free ribosomes, increased nuclear-to-cytoplasmic ratios and an overall decrease in myofilaments.29 Several factors have been implicated in signaling the switch in smooth muscle cell phenotype from a contractile to a synthetic morphology, including mechanical changes (stretch), hormones, changes in extracellular matrix composition, reactive oxygen species, and thrombin.29 Rabbit aorta smooth muscle cell subcultures, undergoing cyclic stretching on elastin membranes, convert to the synthetic phenotype as measured by their increase in glycosaminoglycan production.30 After stimulation with a single dose of estradiol dipropionate, contractile smooth muscle cells in the rat uterus demonstrate an increase in the number of cellular organelles also consistent with a conversion to the synthetic phenotype.31

The primary limitation of this study was our failure to blind the individuals interpreting the collagen and smooth muscle cell morphologies to the stages of pregnancy. Indeed, when performing the analysis, as we became more familiar with the ultrastructural changes in collagen and smooth muscle typical of each pregnancy phase, blinding was no longer possible. To minimize bias, however, all samples were separately viewed and scored by two investigators (J.D. and K.C.), with a high rate of concordance between them (90%). Additionally, the 4 categories for collagen quantification and 3 morphologic cell types for smooth muscle cell quantification were standardized before analysis, with the primary aim of minimizing interpretive error. In future studies, this methodology (collagen groups based on fibril area fraction and smooth muscle types based on the amount of cytoplasmic myofibrils vs. organelles) can be used as the basis for quantitative measures of changes in tissue structure.

Collagen and smooth muscle cells are two key structural components of vaginal tissue providing dynamic properties critical to facilitating the vagina's central role in maintaining pelvic organ support. The mechanism by which pregnancy and vaginal delivery affect the vagina is an area of ongoing investigation. In the rat, we have demonstrated that the ultrastructural changes in collagen and smooth muscle morphology of the vaginal wall that occur before, during, and after pregnancy explain, in part, the mechanism by which the vagina adapts to pregnancy and facilitates passage of a fetus while minimizing injury to the vagina. In the rat, adaptations in pregnancy and at the time of delivery are complete with minimal maternal injury and rapid recovery. It is likely that in humans, maternal adaptations are often incomplete or exceeded, resulting in immediate injury and eventually prolapse.

Back to Top | Article Outline


1. Weber AM, Richter HE. Pelvic organ prolapse. Obstet Gynecol 2005;106:615–34.
2. DeLancey JO. Pelvic organ prolapse: clinical management and scientific foundations. Clin Obstet Gynecol 1993;36:895–6.
3. Popovic Jr, Kozak LJ. National hospital discharge survey: annual summary, 1998. Vital Health Stat 13 2000;148:1–194.
4. Norton PA. Pelvic floor disorders: the role of fascia and ligaments. Clin Obstet Gynecol 1993;36:926–38.
5. Petros PE, Ulmsten UI. An integral theory of female urinary incontinence. Experimental and clinical considerations. Acta Obstet Gynecol Scand Suppl 1990;153:7–31.
6. Mant J, Painter R, Vessey M. Epidemiology of genital prolapse: observations from the Oxford Family Planning Association Study. Br J Obstet Gynaecol 1997;104:579–85.
7. Anderson DW, Edwards TK, Ricketts MH, Kuivaniemi H, Tromp G, Stolle CA, et al. Multiple defects in type III collagen synthesis are associated with the pathogenesis of abdominal aortic aneurysms. Ann N Y Acad Sci 1996;800:216–28.
8. Ottani V, Martini D, Franchi M, Ruggeri A, Raspanti M. Hierarchical structures in fibrillar collagens. Micron 2002;33:587–96.
9. Moalli PA, Talarico LC, Sung VW, Klingensmith WL, Shand SH, Meyn LA, et al. Impact of menopause on collagen remodeling in the arcus tendineus fasciae pelvis. Am J Obstet Gynecol 2004;190:620–7.
10. Munarriz R, Kim SW, Kim NN, Traish A, Goldstein I. A review of the physiology and pharmacology of peripheral (vaginal and clitoral) female genital arousal in the animal model. J Urol 2003;170:S40-4: discussion S44–5.
11. Kim NN, Min K, Pessina MA, Munarriz R, Goldstein I, Traish AM. Effects of ovariectomy and steroid hormones on vaginal smooth muscle contractility. Int J Impot Res 2004;16:43–50.
12. Chen YH, Chen YL, Lin SJ, Chou CY, Mar GY, Chang MS, et al. Electron microscopic studies of phenotypic modulation of smooth muscle cells in coronary arteries of patients with unstable angina pectoris and postangioplasty restenosis. Circulation 1997;95:1169–75.
13. Kanda K, Matsuda T. Mechanical stress-induced orientation and ultrastructural change of smooth muscle cells cultured in three-dimensional collagen lattices. Cell Transplant 1994;3:481–92.
14. Lindqvist A, Nordstrom I, Malmqvist U, Nordenfelt P, Hellstrand P. Long-term effects of Ca(2+) on structure and contractility of vascular smooth muscle. Am J Physiol 1999;227:C64–73.
15. Chamley JH, Campbell GR. Mitosis of contractile smooth muscle cells in tissue culture. Exp Cell Res 1974;84:105–10.
16. Moalli PA, Howden NS, Lowder JL, Navarro J, Debes KM, Abramowitch SD, et al. A rat model to study the structural properties of the vagina and its supportive tissues. Am J Obstet Gynecol 2005 Jan;192:80–8.
17. Moalli PA, Shand SH, Zyczynski HM, Gordy SC, Meyn LA. Remodeling of vaginal connective tissue in patients with prolapse. Obstet Gynecol 2005;106:953–63.
18. Niyibizi C, Kavalkovich K, Yamaji T, Woo SL. Type V collagen is increased during rabbit medial collateral ligament healing. Knee Surg Sports Traumatol Arthrosc 2000;8:281–5.
19. Jones PA, Werb Z. Degradation of connective tissue matrices by macrophages. II. Influence of matrix composition on proteolysis of glycoproteins, elastin, and collagen by macrophages in culture. J Exp Med 1980;152:1527–36.
20. Manabe Y, Yoshida Y. Collagenolysis in human vaginal tissue during pregnancy and delivery: a light and electron microscopic study. Am J Obstet Gynecol 1986;155:1060–6.
21. Zhao L, Samuel CS, Tregear GW, Beck F, Wintour EM. Collagen studies in late pregnant relaxin null mice. Biol Reprod 2000;63:697–703.
22. Owens GK, Kumar MS, Wamhoff BR. Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev 2004;84:767–801.
23. Owens GK. Regulation of differentiation of vascular smooth muscle cells. Physiol Rev 1995;75:487–515.
24. Gonzalez-Crussi F. Vasculogenesis in the chick embryo. An ultrastructure study. Am J Anat 1971;130:441–60.
25. Campbell GR, Chamley-Campbell JH. Invited review: the cellular pathobiology of atherosclerosis. Pathology 1981;13:423–40.
26. Panettieri RA Jr. Airway smooth muscle: an immunomodulatory cell. J Allergy Clin Immunol 2002;110:S269–274.
27. Levin RM, Haugaard N, O'Connor L, Buttyan R, Das A, Dixon JS, et al. Obstructive response of human bladder to BPH vs. rabbit bladder response to partial outlet obstruction: a direct comparison. Neurourol Urodyn 2000;19:609–29.
28. Golder M, Burleigh De, Belai A, Ghali L, Ashby D, Lunniss PJ, et al. Smooth muscle cholinergic denervation hypersensitivity in diverticular disease. Lancet 2003;361:1945–51.
29. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature 2000;407:249–57.
30. Leung DY, Glagov S, Mathews MB. Cyclic stretching stimulates synthesis of matrix components by arterial smooth muscle cells in vitro. Science 1976;191:475–7.
31. Bo WJ, Odor DL, Rothrock M. The fine structure of uterine smooth muscle of the rat uterus at various time intervals following a single injection of estrogen. Am J Anat 1968;123:369–84.
© 2007 The American College of Obstetricians and Gynecologists