Changes in endometrial blood vessels in the endometrium of women with hormone replacement therapy-related irregular bleeding

M. Hickey1,5, G. Pillai2, J.M. Higham3, M. Sullivan4, D. Horncastle4, D. Doherty1 and G. Stamp4

1 University of Western Australia, Women and Infants Research Foundation, Department of Obstetrics and Gynaecology, Subiaco, WA 6008, Australia, 2 Nuffield Department of Clinical Laboratory Sciences, University of Oxford, Oxford, 3 Division of Paediatrics, Obstetrics and Gynaecology, Faculty of Medicine Imperial College, and 4 Department of Pathology, Hammersmith Hospital, Imperial College, London, UK

5 To whom correspondence should be addressed. e-mail: mhickey{at}obsgyn.uwa.edu.au


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Irregular bleeding affects up to 60% of hormone replacement therapy (HRT) users. The mechanism of this bleeding is not understood. Reduced endometrial microvascular integrity appears to underlie breakthrough bleeding in pre-menopausal women and the aim of this study was to establish whether similar changes are seen in HRT users and hence to elucidate a possible mechanism of irregular bleeding. METHODS: Endometrium from 34 HRT users with amenorrhoea, irregular bleeding or regular bleeding was assessed for endometrial endothelial cell density (anti-CD34), number of blood vessels per mm2, vascular basal lamina components (laminin, collagen IV and heparan sulphate proteoglycan) and in 32 subjects and 23 controls for perivascular smooth muscle {alpha} (SMA). Findings were compared with a control population of 29 post-menopausal women not using HRT, other sex steroids or tamoxifen and with no vaginal bleeding. Staining intensity was assessed in a blinded fashion in all immunohistochemical studies. RESULTS: Four significant differences in endometrial blood vessels were observed between HRT users and controls: (i) a significantly lower density of endometrial endothelial cells (EC staining for CD34) per mm2 was present in HRT users compared with controls (P < 0.001); (ii) endothelial cells (EC) were predominantly organized within blood vessels (83%) in controls but in HRT users EC were dispersed in the tissues with only 29% in organized vessels (P <0.001); (iii) supportive perivascular cell SMA was significantly reduced in 23 post-menopausal HRT users compared with 23 post-menopausal controls (n = 29, P = 0.013) and (iv) an atrophic or inactive histological pattern of endometrium was more frequently seen in the controls (P < 0.001). CONCLUSIONS: These findings support the hypothesis that exposure to HRT profoundly alters endometrial blood vessels, reducing structural integrity thereby predisposing to irregular bleeding in HRT users.

Key words: bleeding/endometrium/HRT/smooth muscle {alpha}-actin


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Long-term hormone replacement therapy (HRT) is used by increasing numbers of peri- and post-menopausal women in the developed world. HRT with estrogen alone or with added progestogen relieves menopausal symptoms and physical changes associated with depleted endogenous estrogen levels. Combined HRT has also demonstrated a clear benefit in the prevention of osteoporosis and hip fracture and may reduce the risk of colorectal cancer (Women’s Health Initiative, 2002) and Alzheimer’s disease (Fillit, 2002Go).

Many women tolerate bleeding with HRT or have amenorrhoea, but a substantial minority suffer from unpredictable, unscheduled vaginal bleeding and spotting. Irregular bleeding occurs in up to 60% of HRT users (Al-Azzawi and Habiba, 1994Go) and leads to discontinuation of therapy in up to one in three users (Limouzin-Lamothe, 1996Go). This irregular bleeding is often worse, but is not confined to the initial months of HRT use, and occurs in users of both sequential and continuous combined preparations (Ettinger et al., 1998Go). In addition, the expectation of unscheduled vaginal bleeding is thought to deter many women from starting HRT (Newton et al., 1997Go).

Irregular bleeding occurs with all HRT preparations and regimens and no established treatments are available. Investigations to exclude malignancy are often indicated though in most cases no significant pathology is demonstrated. Breakthrough bleeding in progestogen users arises from fragile superficial vessels (Hickey et al., 1999aGo; 2000; Rogers et al., 2000Go). The aim of this study was to determine whether HRT affects endometrial vasculature integrity and structure in a similar fashion and hence to improve understanding of the mechanisms of HRT-related irregular bleeding.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Peri- and post-menopausal women in good general health were recruited from the menopause, reproductive endocrine and general gynaecology outpatient clinics at St Mary’s Hospital, London, UK. Peri-menopausal status was defined as 3–11 months of amenorrhoea in the last 12 months in the presence of menopausal symptoms (Dudley et al., 1998Go) and post-menopausal status as ≥1 year since last menstrual period. Subjects were excluded if they had a history of breast or genital tract malignancy, endocrinopathy or cervical surgery (making outpatient endometrial biopsy difficult). All subjects were required to have had a normal cervical smear in the previous 12 months. Informed consent was obtained from all subjects. The Local Research Ethics Committee of Imperial College School of Medicine at St Mary’s Hospital, London gave approval for this study.

Endometrial biopsies were taken from subjects who had been using sequential or continuous combined HRT for ≥3 months. Sequential HRT was defined as that containing an estrogen preparation taken continuously with a progestogen taken in addition for up to 14 days in a 28 day period. Continuous combined HRT was defined as that contained estrogen and progestogen taken on a daily basis. Biopsies were taken using an endometrial suction curette (Z Sampler; BEI Medical Systems International, USA). Control endometrial tissue was obtained from post-menopausal women undergoing hysterectomy for benign gynaecological disease. These subjects had not been exposed to sex steroids and had no history of post-menopausal bleeding.

Menstrual bleeding patterns in the previous 3 months were classed as: no bleeding; regular bleeding (in cyclic HRT users); or irregular bleeding (in continuous combined or cyclic HRT users). All subjects with irregular bleeding were investigated with transvaginal ultrasound as well as endometrial biopsy. If the endometrial thickness was >4 mm hysteroscopy was performed (Gull et al., 2001Go).

Laboratory studies
Biopsies were formalin-fixed and wax-embedded, and examined blind for histological appearance (haematoxylin and eosin) by one senior gynaecological pathologist from Imperial College School of Medicine and classified according to the criteria of Noyes et al. (1950Go).

Following dewaxing and rehydration, the tissue sections were quenched for endogenous peroxidase activity (0.6% hydrogen peroxide), for 15 min at room temperature. Antigen retrieval was performed to expose antigen sites masked by fixation. CD34 required microwave heating for 10 min in 0.01 mol/l citrate buffer. Laminin and collagen IV required 0.05% protease digestion for 10 min (Sigma–Aldrich, UK). Non-specific sites were blocked with 5% normal goat serum for 10 min.

Sections were incubated with primary antibody overnight at 40°C. The primary antibodies used were mouse anti-human CD34 antigen at 1/100 dilution (Q Bend 10; Dako Ltd, UK), mouse anti-human collagen IV at 1:100 dilution (Dako Ltd), mouse anti-human laminin at 1/1000 dilution, (Sigma–Aldrich), mouse anti-human smooth muscle {alpha} actin (SMA), 1/16 000 (Sigma–Aldrich) and mouse anti-human heparan sulphate proteoglycan (HSPG) (Clone MAB 458; Chemicon International Inc., USA).

Sections were incubated with biotinylated goat anti-mouse immunoglobulin (Dako Ltd), at a dilution of 1/500 for 30 min at room temperature. This was followed by incubation with streptavidin peroxidase (Roche Diagnostics Ltd, UK) at a dilution of 1/500 for 30 min at room temperature. Diaminobenzidine (Sigma–Aldrich) was used as a chromagen for 10 min at room temperature. The sections were counterstained with Cole’s haematoxylin (Pioneer Research Chemical Ltd, UK), dehydrated and mounted in DPX pertex mounting media. Negative and positive controls were included in each staining run. Positive controls were skin for laminin and collagen IV and colon for CD34, SMA and HSPG.

Strong and continuous staining around the vessel periphery identified basal lamina. The presence or absence of each basal lamina component was noted. Assessment was performed in a blinded fashion by a single observer.

Vessel counting
Samples were considered suitable for counting if the haematoxylin and eosin sections contained both glands and stroma, and at least 10 (maximum) or 5 (minimum) random unit areas could be counted at a magnification of x400 (Song et al., 1995Go). All the slides were counted with a grid eyepiece pre-calibrated with a slide micrometer. Two individuals counted the slides using a consistent technique. The results were averaged and calculated as vessels per mm2.

Endothelial cells (EC staining for CD34) were counted individually and the number of blood vessels was also assessed and recorded. Groups of EC organized as a collection of single cells grouped around a lumen were recorded as blood vessels. When the EC were not organized in this fashion but were distributed in apparent random areas throughout the tissue, this was recorded as EC not in blood vessels. In each sample, the total number of EC and the total number of blood vessels were counted.

Statistical methods
The main objective of this analysis was to explore the endometrial changes associated with HRT exposure. Two groups of patients were investigated: post- and peri-menopausal HRT users and post-menopausal women who did not use HRT were examined. Descriptive summary statistics for continuous data utilized means and SEM or medians, interquartile ranges (IQ) and ranges, as appropriate depending on the data normality. Group differences between the HRT users and women who did not use HRT were investigated using the Fisher’s exact or Mann–Whitney tests as appropriate. Analysis of the CD34 staining was carried out by analysis of variance. All tests used were two-sided and P ≤ 0.05 was considered significant. SPSS (SPSS Inc., USA) and S-Plus (MathSoft Engineering & Education, Inc., USA) statistical software were used for data analysis.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Ffty-nine HRT users were recruited for this study and endometrial tissue sufficient to perform full histological and immunohistochemical studies was obtained from 34 subjects (58%). In eight subjects outpatient biopsies could not be obtained (largely due to cervical stenosis) and in 17 subjects outpatient biopsies were taken but insufficient tissue was obtained for the full analysis described in this study.

Forty-two control subjects were recruited who were not taking sex steroids or tamoxifen and with no post-menopausal bleeding. Endometrial tissue sufficient to perform histological and immunohistochemical studies was obtained from 29 women (69%). This sample represented 26 subjects undergoing hysterectomy and three outpatient biopsies from post-menopausal women not taking HRT.

Twenty-four out of 34 HRT users (71%) were post-menopausal when they commenced HRT and 10/34 (29%) were peri-menopausal when they commenced HRT. It is not possible to comment on the menopausal status of these patients when the biopsy was taken. No statistically significant differences were found in any variables studied between these groups, so where indicated the findings have been combined.

The median age of HRT users was 54 years (range 42–71, IQ range 50–56). The median duration of HRT use was 12 months (range 1–348, IQ range 3–41). 24/34 (71%) women were receiving a sequential HRT preparation and 10/34 (29%) continuous combined HRT. For the groups of 24 sequential HRT users, 13 (54%) were biopsied during the estrogen and progesterone phase, 10 (42%) were biopsied during the progestin phase and one case had no information recorded. The median age of onset of menopause in post-menopausal HRT users was 50 years (range 29–54, IQ range 45–51) and the median time since menopause was 7.5 years (range 2–25, IQ range 3–13).

As anticipated from clinical practice and from our previous studies (Hickey et al., 2001aGo) HRT users were taking a wide range of preparations (Table I).


View this table:
[in this window]
[in a new window]
 
Table I. Hormone replacement therapy (HRT) regimes, doses and dose schedules used by study subjects
 
Bleeding pattens
On direct questioning of HRT users about bleeding patterns in the previous 3 months, 14/34 (44%) of women described regular bleeding (all sequential HRT users), 16/34 (47%) described irregular bleeding (sequential and continuous combined HRT users) and 4/34 (12%) described amenorrhoea (all continuous combined HRT users). None of those with irregular bleeding had an endometrial thickness of >4 mm or other uterine pathology detected on transvaginal ultrasound and no abnormalities were found in endometrial histological appearance. Six biopsies (17%) were taken during a bleeding episode.

The median age of control subjects was 65 years (range 48–92, IQ range 58–72), significantly older than HRT users (P = 0.041). The median time since menopause was also longer than study subjects at 15 years (range 1–42, IQ range 6–23) (P = 0.013).

Significant differences in histological appearance were seen between HRT users and controls. Not surprisingly, an atrophic or inactive appearance was more common in the control group (P < 0.001; Table II). Amongst HRT users, those who did not have atrophic endometrium were more likely to report regular bleeding patterns (P < 0.001).


View this table:
[in this window]
[in a new window]
 
Table II. Endometrial histological appearance in HRT users compared with controls
 
Endometrial vascular density
CD34+ (endothelial) cells were observed in all samples investigated. Two different patterns of staining were observed. In control samples, CD34+ cells were generally found in blood vessels (Figure 1a,b), whereas in HRT samples many of the CD34+ cells were not in readily identifiable vessels (Figure 1c,d). Counting of the cells showed that in the control population 83% of CD34+ cells were clearly located in blood vessels. In the post-menopausal HRT users 71% of CD34+ cells were scattered throughout the tissues with only 29% apparently arranged in blood vessels (P > 0.001). This phenomenon was also apparent in the peri-menopausal HRT users, where 68% of CD34+ cells were scattered in the tissues (P < 0.001) (Figure 1d shows scattered cells)



View larger version (115K):
[in this window]
[in a new window]
 
Figure 1. Immunohistochemistry for CD34 and for smooth muscle {alpha} actin (SMA); positive cells are brown. (a) Normal post-menopausal CD34 immunostaining in control tissue, x200; (b) normal post-menopausal CD34 immunostaining in control tissues, x400; (c) reduced CD34 immunostaining in hormone replacement therapy (HRT) tissue, x200; (d) reduced CD34 immunostaining in HRT tissue, x400; (e) normal post-menopausal SMA immunostaining in control tissue x200; reduced (f) SMA immunostaining in HRT tissue, x200; (g) normal post-menopausal SMA immunostaining in control tissue, x400; (h) reduced SMA immunostaining in HRT tissue, x400. Typical data are shown in each case. Blood vessels are shown by arrowheads.

 
The number of blood vessels visualized in each section was recorded. Significantly more endometrial blood vessels were seen in control compared with HRT users (P < 0.001). A median of 12 per mm2 (range 1–22, IQ range 9–14) in controls compared with four vessels per mm2 was seen in HRT users (range 0–14, IQ range 2–7). Typical results can be seen in Figure 1b and d.

Significantly greater numbers of CD34+ cells were observed per mm2 in controls (mean: 85.0 ± 11.9) compared with HRT users (mean: 59.2 ± 9.6) (P < 0.001). This finding remained statistically significant when only post-menopausal HRT users (55.8 ± 11 per mm2) were compared with controls (P = 0.03).

No relationship was observed between EC density or distribution and bleeding patterns, though the highest CD34 counts were observed amongst women with irregular bleeding (mean 65 ± 12 per mm2, range 40–90) and the lowest in women with amenorrhoea (mean 51 ± 22 per mm2, range 5–98).

Endothelial cell or blood vessel density was not related to type of HRT preparation (sequential or continuous combined), to recent bleeding patterns or to whether the subject was bleeding on the day of the biopsy.

Endometrial vascular basal lamina
In view of the potential limitations of semiquantitative assessment of immunohistochemical staining, the endometrial vascular basement lamina components were analysed as ‘absent or present in only very small quantities’ or as ‘present in moderate or large quantities’. A small reduction in post-menopausal perivascular laminin was observed in (non- bleeding) control subjects compared with HRT users (P = 0.049). No differences in collagen IV immunostaining were seen.

Smooth muscle {alpha} actin
Adequate endometrial tissue for the detection of SMA was obtained from 32 HRT users (of whom 23 were post-menopausal). These samples were compared with 29 controls. Cells staining for SMA were detected in most samples (Figure 1e–h). The intensity of the staining and the numbers of positive cells were significantly greater in the control samples (Figure 1e,g) than in those obtained after HRT use (Figure 1f,h). Quantification (see Table III) showed that this was most marked in post-menopausal HRT users but reached statistical significance in both post-menopausal subjects and all HRT users.


View this table:
[in this window]
[in a new window]
 
Table III. Endometrial perivascular smooth muscle {alpha} actin (SMA) immunostaining in hormone replacement therapy (HRT) users versus controls
 
No relationship was observed between vascular basement lamina components and vaginal bleeding patterns in the previous 30 days. In endometrial biopsies from HRT users obtained during a bleeding episode, significantly weaker SMA immunostaining was seen (P = 0.03). Vascular SMA immunostaining was moderate or strong in all HRT users with amenorrhoea.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Superficial endometrial vessels consist only of endothelial cells, and their supporting structures of basal lamina and pericytes. The results from this study suggest that exposure to HRT alters the density and distribution of endothelial cells in the endometrium and also compromises endometrial vascular integrity. This implies similarities in the mechanism of HRT-related bleeding to that seen in progestogen-related breakthrough bleeding.

Microvascular density was reduced in HRT users compared with controls. In contrast, exposure to low dose progestogen implants (Norplant) leads to a marked increase in endometrial vascular density (Rogers et al., 1993Go; Hickey et al., 1999bGo). Wahab et al. (2000Go) observed that progestogen differentially affects endometrial vascularity, hence it is possible that the range of HRT preparations used may have influenced our observations.

Microvascular fragility has been proposed as a mechanism for breakthrough bleeding in progestogen users (Fraser and Peek, 1992Go; Hickey et al., 2000Go). Perivascular SMA is found in pericytes and smooth muscle cells and both of these cell types are known to add to the structural integrity and strength of blood vessels (Rogers et al., 2000Go). Thus the finding of reduced perivascular SMA in HRT users is the first evidence for a correlation between structural changes that increase vascular fragility and HRT use. This supports the findings from our pilot data indicating altered endometrial expression of proteolytic matrix metalloproteinases-9 (MMP-9) and its specific tissue inhibitor TIMP-1 in favour of MMP-9 in HRT users compared with controls (Hickey et al., 2001bGo).

Microvascular pericyte association is associated with vascular maturity and stability and endometrial pericyctes also play an important role in angiogenesis. Angiopoietin-1 (Ang-1) promotes vascular maturation via the Tie-2 receptor, while angiopoietin-2 (Ang-2) is its natural antagonist that destabilizes vessels and initiates neovascularization in the presence of vascular endothelial growth factor (Hewett et al., 2002Go). Angiopoietin-2 activity is prominent in the endometrium where vascular remodelling commonly occurs. The recruitment of pericytes is regulated by Ang-1 as part of the vascular stabilization process following angiogenesis (Maisonpierre et al., 1997Go). It is possible that the increased percentage of ‘immature’ vessels with no perivascular SMA in HRT users indicates increased vascular remodelling in these tissues (Hanahan, 1997Go). This would be supported by the changes observed in endothelial cell distribution in HRT users, with endothelial cells dispersed throughout the tissue rather than organized in recognizable vascular structures. ‘Double staining’ for CD34 and SMA might have provided further information about the precise relationship between endometrial blood vessels and SMA in these subjects.

We did not find a relationship between perivascular SMA and bleeding patterns in HRT users, suggesting that alterations in vascular density and stability do not provide a full explanation for irregular bleeding. A simple classification system was used for bleeding patterns in HRT users. Inconsistency in reporting bleeding patterns in HRT users has led to confusion in the literature. Archer and Pickar (2002Go) suggest standardized definitions to improve comparisons between HRT preparations.

To our knowledge, this is the first study to use control data of post-menopausal endometrium in assessing the endometrial effects of HRT. Other published studies have used pre-menopausal controls or have compared one population of HRT users with another. In view of our findings we contend that this is not a valid comparison, and that the substantial differences in the endogenous sex steroid environment between pre- and post-menopausal women mean that it is essential to compare ‘like with like’. This is supported by the major histological differences seen between HRT users and controls.

The age difference between HRT users and controls was a limitation of this study, and partly reflects the difficulty in obtaining control tissue from ‘normal’ post-menopausal women. It is possible that patient age, or time since the menopause may represent an independent variable affecting the endometrial vasculature, though our pilot studies did not find this (Hickey et al., 2001bGo).

As is common in clinical practice, the women in this study were taking a wide variety of HRT preparations and it is certainly possible that different sex steroids affect the endometrium in different ways. However, since all HRT preparations and all regimens are associated with irregular bleeding, and there are no clear advantages of one preparation over another with respect to bleeding patterns, we consider it reasonable to investigate a cross-section of HRT users in the first instance to look for common mechanisms of endometrial vascular breakdown (Thomas et al., 2000Go). Longitudinal studies assessing the endometrium before and after exposure to specific HRT preparations are indicated to address the different mechanisms of action of differing estrogens and progestogens.

In summary, irregular bleeding in HRT users presents a major clinical challenge. Unless the mechanisms underlying this bleeding are understood, it is unlikely that effective preventative or therapeutic strategies will be developed. The findings from this prospective controlled study suggest that HRT use is associated with changes in endometrial blood vessels which could increase vascular fragility and hence lead to irregular bleeding.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Agostini, A., Shojai, R., Cravello, L., Rojat-Habib, M.C., Roger, V., Bretelle, F. and Blanc, B. (2001) Endometrial biopsy during outpatient hysteroscopy: evaluation and comparison of two devices. Eur. J. Obstet. Gynecol. Reprod. Biol., 97, 220–222.[CrossRef][ISI][Medline]

Al-Azzawi, F. and Habiba, M. (1994) Regular bleeding on hormone replacement therapy: a myth? Br. J. Obstet. Gynaecol., 101, 661–662.[ISI][Medline]

Archer, D.F. and Pickar, J.H. (2002) The assessment of bleeding patterns in postmenopausal women during continuous combined hormone replacement therapy: a review of methodology and recommendations for reporting of the data. Climacteric, 5, 45–59.[ISI][Medline]

Dudley, E.C., Hopper, J.L., Taffe, J., Guthrie, J.R., Burger, H.G. and Dennerstein, L. (1998) Using longitudinal data to define the perimenopause by menstrual cycle characteristics. Climacteric, 1, 18–25.[Medline]

Ettinger, B., Li, D.K. and Klein, R. (1998) Unexpected vaginal bleeding and associated gynecologic care in postmenopausal women using hormone replacement therapy: comparison of cyclic versus continuous combined schedules. Fertil. Steril., 69, 865–869.[CrossRef][ISI][Medline]

Fillit, H.M. (2002) The role of hormone replacement therapy in the prevention of Alzheimer disease. Arch. Intern. Med., 162, 1934–1942.[Abstract/Free Full Text]

Fraser, I.S. (1986) A review of the role of progestogens in hormone replacement therapy: influence on bleeding patterns. Maturitas, 8, 113–121.[ISI][Medline]

Fraser, I.S. and Peek, M.J. (1992) Effects of exogenous hormones on endometrial capillaries. In Alexander, N.J. and d’Arcangues, C. (eds), Steroid Hormones and Uterine Bleeding. AAAS Press, Washington, DC, pp. 67–79.

Genazzani, A.R. and Gambacciani, M. (2000) Controversial issues in climacteric medicine. I. Cardiovascular disease and hormone replacement therapy. International Menopause Society Expert Workshop, October 13–16, 2000, Royal Society of Medicine, London, UK. Climacteric, 3, 233–240.[Medline]

Gull, B., Carlsson, S., Karlsson, B., Ylostalo, P., Milsom, I. and Granberg, S. (2000) Transvaginal ultrasonography of the endometrium in women with postmenopausal bleeding: is it always necessary to perform an endometrial biopsy? Am. J. Obstet. Gynecol., 182, 509–515.[CrossRef][ISI][Medline]

Hanahan, D. (1997) Signalling vascular morphogenesis and maintenance. Science, 277, 48–50.[Free Full Text]

Hewett, P., Nijjar, S., Shams, M., Morgan, S., Gupta, J. and Ahmed, A. (2002) Down-regulation of angiopoietin-1 expression in menorrhagia. Am. J. Pathol., 160, 773–780.[Abstract/Free Full Text]

Hickey, M., Lau, T., Fraser, I.S. and Rogers, P.A.W. (1996) Endometrial vascular density in conditions of endometrial atrophy. Hum. Reprod., 9, 2009–2013.

Hickey, M., Simbar, M., Markham, R., Russell, P., Young, L. and Fraser, I.S. (1999a) Changes in endothelial cell basement lamina in Norplant users. Hum. Reprod., 14, 716–721.[Abstract/Free Full Text]

Hickey, M., Simbar, M., Markham, R., Russell, P., Young, L. and Fraser, I.S. (1999b) Changes in endometrial vascular density in Norplant users. Contraception, 59, 123–129.[CrossRef][ISI][Medline]

Hickey, M., Dwarte, D. and Fraser, I.S. (2000) Superficial endometrial vascular fragility in Norplant users and in women with ovulatory dysfunctional uterine bleeding. Hum. Reprod., 15, 1509–1514.[Abstract/Free Full Text]

Hickey, M., Higham, J.M. and Fraser, I.S. (2001a) Hormone replacement therapy and irregular bleeding. Climacteric, 4, 95–101.[Medline]

Hickey, M., Higham, J., Sullivan, M., Miles, L. and Fraser, I.S. (2001b) Endometrial bleeding in HRT users: preliminary findings regarding the role of MMP-9 and TIMP-1. Fertil. Steril., 75, 288–296.[CrossRef][ISI][Medline]

Kaunitz, A.M. (2001) Oral contraceptive use in perimenopause. Am. J. Obstet. Gynecol., 185 (Suppl. 2), S32–37.[ISI][Medline]

Limouzin-Lamothe, M.A. (1996) What women want from hormone replacement therapy: results of an international survey. Eur. J. Obstet. Gynecol. Reprod. Biol., 64, S21–24.[CrossRef][ISI][Medline]

Maisonpierre, P.C., Suri, C., Jones, P.F., Bartunkova, S., Wiegand, S.J., Radziejewski, C., Compton, D., McClain, J., Aldrich, T.H. Papadopoulos, N. et al. (1997) Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science, 277, 55–60.[Abstract/Free Full Text]

Newton, K.M., LaCroix, A.Z., Leveille, S.G., Rutter, C., Keenan, N.L. and Anderson, L.A. (1997) Women’s beliefs and decisions about hormone replacement therapy. J. Women’s Health, 6, 459–465.[ISI][Medline]

Noyes, R.W., Hertig, A.T. and Rock, J. (1950) Dating the endometrial biopsy. Fertil. Steril., 1, 3–25.[ISI][Medline]

Rees, M., Hague, S., Oehler, M.K. and Bicknell, R. (1999) Regulation of endometrial angiogenesis. Climacteric, 2, 52–58.[Medline]

Rogers, P.A., Au, C.L. and Affandi, B. (1993) Endometrial microvascular density during the normal menstrual cycle and following exposure to long-term levonorgestrel. Hum. Reprod., 8, 1396–1404.[Abstract]

Rogers, P.A.W., Plunkett, D. and Affandi, B. (2000) Perivascular smooth muscle {alpha}-actin is reduced in the endometrium of women with progestin-only contraceptive breakthrough bleeding. Hum. Reprod., 15 (Suppl. 3), 78–85.[Medline]

Song, J.Y., Markham, R., Russell, P., Wong, T., Young, L. and Fraser, I.S. (1995) The effect of high dose medium- and long-term progestogen exposure on endometrial vessels. Hum. Reprod., 10, 797–800.[Abstract]

Thomas, A.M., Hickey, M. and Fraser, I.S. (2000) Disturbances of bleeding with hormone replacement therapy. Hum. Reprod., 15, 7–18.[Medline]

Wahab, M., Thompson, J., Hamid, B., Deen, S. and Al-Azzawi, F. (1999) Endometrial histomorphometry of trimegestone-based sequential hormone replacement therapy: a weighted comparison with the endometrium of the natural cycle. Hum. Reprod., 14, 2609–2618.[Abstract/Free Full Text]

Wahab, M., Thompson, J. and Al-Azzawi, F. (2000) Effect of different cyclical sequential progestins on endometrial vascularity in post-menopausal women compared with the natural cycle: a morphometric analysis. Hum. Reprod., 10, 2075–2081.[CrossRef]

Women’s Health Initiative (2002) Risks and benefits of estrogen plus progestin in health post-menopausal women. J. Am. Med. Assoc., 288, 321–333.[Abstract/Free Full Text]

Submitted on July 23, 2002; resubmitted on November 1, 2002; accepted on January 6, 2003.