November 18, 2007

Effects of Subdermal Implants of Estradiol and Testosterone on the Endometrium of Postmenopausal Women

By De Barros Filho, Adelmo M B Barbosa, Ione C; Maia, Hugo Jr; Genes, Consuelo C; Coutinho, Elsimar M

Abstract A retrospective review of the medical records of 258 postmenopausal patients using estradiol and testosterone implants as combined hormone therapy was carried out to evaluate the effects of testosterone on the endometrium after two years of continuous use. Endometrial thickness was measured by ultrasonography. Histology was performed on samples of thickened endometria obtained during hysteroscopy with biopsy. In the 44 patients in whom endometrial thickening was > 5 mm at the end of the second year of implant use, the most frequent finding at hysteroscopy was polypoid lesion in 61.3% of cases, followed by normal uterine cavity in 31.8% of cases and submucous myoma in 6.8%. Histology of the endometrial samples confirmed endometrial polyp in 38.6% of cases, a histologically normal endometrium in 31.8% of cases, simple endometrial hyperplasia in 20.4% of cases, and myoma and atrophie endometrium in 4.5%. It is possible that testosterone may exert its antiproliferative effects on the endometrium but not on polyps in an action similar to that exerted by combined estrogen/ progestin therapies. A greater incidence of simple, low-grade endometrial hyperplasia was found in our study compared with studies using continuous estrogen/progestin regimens. The use of progestins as the ideal endometrial protection should therefore be reconsidered.

Keywords: Implant, hormone therapy, testosterone, endometrium


As a woman approaches menopause, a physiological decrease in circulating levels of sex hormone-binding globulin (SHBG) occurs, together with an increase in the activity of the aromatase enzyme in certain tissues in order to encourage tissue-specific estrogen production through androgenic precursors. This mechanism is the principal pathway of estrogen production in the menopause. Therefore, the most physiological manner in which to replace estrogens and androgens in the postmenopause is to increase the bioavailability of the endogenous androgenic precursors, a preferable alternative to providing exogenous estrogens [1].

It is well known that only the free fraction of testosterone, i.e. the part not bound to SHBG (around 2%), exerts a biological role on tissue, and that oral estrogen therapy increases the production of SHBG, diminishing the effects of endogenous testosterone and accelerating the appearance of symptoms of androgen deficiency. Nevertheless, when estrogen therapy is administered in the form of subcutaneous implants, the increase in SHBG levels is less evident, thereby permitting testosterone to exert its effects on tissues. This is the rationale behind the use of estradiol and testosterone in the form of subcutaneous implants for the relief of menopausal symptoms [1].

Androgens protect against estrogen stimulation in the normal breast tissue because they inhibit aromatase, the enzyme responsible for the transformation of these androgens into estrogens in the breast. In breast cancer, on the other hand, breast tissue responds to androgens differently from the normal breast, and in this case androgens may become a source of estrogen [2].

In fact, it has been demonstrated in vitro that androgens, particularly androstenedione, are capable of inhibiting endometrial epithelial growth and its secretory activity [3]. Androgen receptors in the endometrium are localized principally in the stroma, with only a small amount being detected in the glandular epithelium. This predominantly stromal distribution of the androgen receptors is observed in various benign endometrial pathologies such as polyps and typical hyperplasia. This suggests that most of the effects of testosterone on the endometrial epithelium are exerted on the stroma. These findings suggest that testosterone may exert an antiproliferative effect on the normal endometrium when present in an appropriate ratio with estrogen. Indeed, in menopausal women using implants of estradiol and testosterone, the endometrium remains atrophie during treatment, showing few signs of endometrial proliferation [1].

When these patients develop endometrial thickening, in the majority of cases this is due to the presence of a polyp [4]. The growth of polyps in the uterine cavity appears to be unabated by testosterone in the same way that polyps also appear insensitive to the antiproliferative effects of progestins, growing in response to estrogens and provoking various degrees of hyperplasia in the glandular epithelium [5].

Aromatase expression varies according to the tissue and the age of the patient [6]. In the endometrium, aromatase expression differs according to the diverse histological types found in the various situations analyzed. The atrophie menopausal endometrium does not express or expresses very little aromatase and is rich in testosterone receptors in the stroma. According to Maia and colleagues [7], free testosterone cannot be converted into estradiol due to the lack of aromatase and, consequently, the endometrium does not proliferate.

Endometrial polyps contain testosterone receptors in the stroma just as the normal endometrium. Recent studies [8] show that polyps express the aromatase p450 enzyme in glandular tissue, stimulating the local production of estradiol from free testosterone, thereby inducing its proliferation. This explains why endometrial polyps are considered insensitive both to progestins and testosterone. Endometrial polyps may undergo malignant transformation and removal by hysteroscopy is therefore considered obligatory [9]. Thus, the effect of testosterone will depend on the presence or absence of a pre-existing lesion [8].

When reporting results on the use of different hormone regimens and the incidence of endometrial cancer, investigators of the Million Women Study concluded that combined estrogen/progestin therapies lead to a greater incidence of breast cancer than other forms of hormone therapy (HT). Considering the sum of the number of cases of breast and endometrial cancer, there is a greater increase associated with the use of continuous or cyclic combined HT compared with other regimens [10].

Following the finding that testosterone may exert a protective effect similar to that of progestins on endometrial proliferation [7], some referral centers in postmenopausal therapy initiated the use of a combination of estrogen and testosterone in nonhysterectomized women [1]. This initiative abolished the practice of the triple association of estrogen, progestin and testosterone [11,12], preventing and/or eliminating side-effects that could be attributed to the progestins such as a greater risk of breast cancer and some of the deleterious effects on the cardiovascular system [13], among other negative effects [14].

The principal objective of the present study was to evaluate the effects of subcutaneous implants of estradiol and testosterone on the endometrium by calculating the incidence of endometrial thickening after two years of continuous use. The principal endometrial findings detected at hysteroscopy and the results of histology carried out on the tissue removed at biopsy were also assessed. Finally, the incidence of the pathologies found in our sample was compared with reports published in the literature on studies in menopausal women using combined estrogen/progestin hormone regimens.

Material and methods

A retrospective study was carried out by reviewing the medical records of patients receiving postmenopausal HT between January 2000 and January 2002 at the Centro de Pesquisa e Assistencia em Reproducao Humana (CEPARH), a referral center for the treatment and study of the menopause. This project was approved by the Internal Review Board of the institution. Non-hysterectomized, postmenopausal women using HT in the form of subdermal implants containing estradiol and testosterone continuously for a period of 2 years were included in the study.

Transvaginal sonography was carried out in all patients prior to treatment and again at the end of the first and second years of treatment, always at the time of implant renewal (implants expired after one year of use). A total of 258 patients using implants of estradiol and testosterone for two consecutive years, who had undergone sonographic evaluation at the same clinic, were identified. Of these, ultrasonography detected endometrial thickening (> 5 mm) in 44 patients at the end of the second year of implant use. These 44 patients were submitted to hysteroscopy with guided biopsy, followed by histological evaluation of the endometrial sample obtained at biopsy.

The subdermal implants are fabricated in the institution using segments of surgical-grade, semipermeable Silastic(R) tubing (Technical Products, Inc., Decatur, GA, USA), steam-sterilized in an autoclave at 121[degrees]C for 10 min and allowed to dry inside the autoclave for a further 10 min. Implants are submitted to microbiological and endotoxin control. The estradiol implants, measuring 5 cm in length and 2.4 mm in external diameter, are filled with 50 mg of 17beta-estradiol, providing a release rate that ranges from 4 to 5 [mu]g estradiol/day over one year. Testosterone implants measure 4 cm in length and 2.4 mm in external diameter, and are filled with 50 mg of 17beta-testosterone, equivalent to a mean daily release rate of 0.2-0.8 [mu]g/day over one year. The implants were inserted using surgical steel trocars in the upper, outer quadrant of the gluteal region of the patients following local anesthesia with 2% procaine. All sonographic examinations were carried out by the same physician using the same equipment (Aloka-SSD-1700, color Doppler; KPI Ultrasound, Riverside, CA, USA). All patients underwent hysteroscopy in the same institution, carried out by a specialist in gynecological endoscopy. The procedure was performed using saline solution as a distension medium controlled by a Hysteromat (Storz, Tuttlingen, Germany), following intravenous sedation with propofol and paracervical block. Duration of hospitalization ranged from 3 to 6 h.

Endometrial samples obtained by biopsy were sectioned into 3 [mu]m-thick, paraffin-embedded blocks for histological evaluation following hematoxylin and eosin staining. The entire extension of the histological section was evaluated morphologically. Finally, the principal hysteroscopic findings and the respective histological evaluations were compared.

The SPSS software program, version 9.0 for Windows, was used in the statistical analysis (SPSS Inc., Chicago, IL, USA). Statistical significance was established at p


A total of 258 patients aged 40-60 years (mean +- standard deviation (SD): 53.3 +- 7.3 years), who had used estradiol and testosterone implants for two consecutive years, were identified and selected for inclusion in this study. Patients were classified into three principal groups according to the number of estradiol implants (four, five or six) used by each patient in addition to the two testosterone implants: 4E + 2T, 5E + 2T or 6E + 2T. Of these, 135 patients used 4E + 2T, 68 used 5E + 2T and 55 patients used 6E + 2T. Patients were inserted with four to six estradiol implants a year, resulting in blood levels that ranged from 35 to 50 pg/ml. Two implants of testosterone were inserted in each patient, resulting in physiological blood levels that ranged from 50 to 80 ng/dl. Of the total of 258 patients, endometrial thickening (endometrium > 5 mm) was detected by ultrasonography in 44 (17%), who were then submitted to hysteroscopy with biopsy. Endometrial thickness ranged from 6 to 28 mm (mean +- SD: 10.5 +- 3.2 mm).

Sonography detected endometrial thickening in 20 (14.8%) of the 135 patients who used 4E + 2T for two consecutive years. Hysteroscopic findings in these 20 patients were suggestive of polyps in 13 patients (65.0%), normal uterine cavity in six (30.0%) and myoma in one patient (5.0%). Of the 13 patients with hysteroscopy suggestive of polyps, histological analysis confirmed this diagnosis in six, but reported simple endometrial hyperplasia in four and histologically normal endometrium in three cases. Among the six cases with hysteroscopy suggestive of a normal uterine cavity, histology confirmed a normal endometrium in four cases, while in two cases the endometrium was atrophic and insufficient material was obtained for analysis. In one case, hysteroscopy suggested a finding of submucous myoma and this diagnosis was confirmed by histology. The incidence of endometrial polyps in the total sample of 135 users of 4E + 2T was 4.4% (Table I).

The hysteroscopic findings in 12 (17.6%) of the 68 patients who used 5E + 2T implants for two consecutive years comprised polyps in five cases (41.7%) and a normal uterine cavity in seven patients (58.3%). Histology confirmed the diagnosis of polyps in all five cases. In the seven cases of normal uterine cavity, histology confirmed these findings in six cases, while simple hyperplasia was found in one case. Therefore, in the group of 68 users of 5E + 2T, the incidence of polyps was 7.3% (Table II).

Among the 55 patients who used implants of 6E + 2T for two consecutive years, 12 (21.8%) were submitted to hysteroscopy with biopsy and histological analysis. Hysteroscopy was suggestive of polyp in nine patients (75.0%), myomas in two patients (16.7%) and a normal uterine cavity in one patient (8.3%). Histology confirmed the diagnosis of polyps in four of the nine cases, while simple hyperplasia was found in four cases and normal endometrium in one patient. Histology detected an endometrial polyp in the patient with hysteroscopy suggestive of a normal uterine cavity. In the two cases in which hysteroscopy was suggestive of submucous myoma, histology confirmed the diagnosis of myoma in one case and diagnosed an endometrial polyp in the other case. Therefore, of the 55 patients using 6E + 2T, the incidence of polyps was 10.9% (Table III).

Table I. Hysteroscopic and histological findings in patients with endometrial thickening (>5 nun) following two years of 4E + 2T implant use.

Hysteroscopy and histology findings for all 44 patients submitted to these diagnostic procedures to investigate endometrial thickening are summarized in Table IV. Of the 27 patients in whom hysteroscopy was suggestive of polyp, histology confirmed diagnosis in 15 cases, while simple endometrial hyperplasia was detected in eight cases and histologically normal endometrium in four cases. With respect to the 14 cases in which hysteroscopy was suggestive of a normal uterine cavity, histology confirmed this finding in ten cases (of which two were cases of secretory endometrium); in two cases the endometrium was atrophie, and endometrial polyp was histologically diagnosed in one case. In the three cases in which hysteroscopy was suggestive of submucous myomas, histology confirmed myoma in two cases and detected an endometrial polyp in one case. Figure 1 illustrates the incidence of polyps in relation to the total number of patients studied.

Table II. Hysteroscopic and histological findings in patients with endometrial thickening (> 5 mm) following two years of 5E + 2T implant use.

Table III. Hysteroscopic and histological findings in patients with endometrial thickening (> 5 mm) following two years of 6E + 2T implant use.

Table IV. Hysteroscopic and histological findings of the 44 patients who were investigated for endometrial thickening (endometrium > 5 mm) following the use of various doses of estrogen/ testosterone implants for two consecutive years.


There is currently a growing interest in the potential use of androgens in postmenopausal women, but the majority of studies carried out on the side-effects of androgen therapy have focused on the virilizing effects of this treatment [15]. There are few studies on the effects of testosterone on the postmenopausal endometrium [16].

The most adequate dose of implants to be used for each individual patient is decided on the basis of her medical history and physical and laboratory examinations. Although doses of several types of medication have been standardized, in a patient on HT changes in dose may be a result of various factors, including inadequate response and the appearance of minor, undesired side-effects such as mastalgia and cephalea, among others. Individual factors may justify the use of higher doses of estradiol in HT regimens, such as the concomitant use of some drugs, including smoking, that interact with this steroid, reducing its bioavailability.

Our study was limited to users of four to six implants of estradiol per year associated with two implants of testosterone per year, in agreement with the current trend towards low doses of HT, highly emphasized following reports of important side-effects with the use of estrogen/progestin regimens. A set of four implants of estradiol annually releases, proportionally, a dose equivalent to that achieved with transdermal patches (50 [mu]g/week), which is considered low-dose [1].

Studies evaluating histological results have reported a highly variable frequency of functional endometrium in postmenopausal patients [17]. Insufficient material obtained for biopsy is interpreted as a state of intense atrophy, making it difficult or impossible to obtain material for histological evaluation [18]. In our study, since patients were using HT, a practice associated with endometrial proliferation, and since the mean age of patients was low (53.3 years) indicating that menopause occurred up to a maximum of 5 years previously, there were few biopsies with insufficient material (atrophie endometrium being found in only two cases).

In the largest prospective, multicenter study carried out to evaluate uterine bleeding in the postmenopause, Karlsson and associates [17] evaluated 1168 women and found a rate of endometrial atrophy of 59%, while other causes of bleeding included polyps (12.4%), endometrial cancer (10.1%), hyperplasia (9.9%) and functional endometrium (6.8%).

Figure 1. Principal hysteroscopic findings and the respective histology results of 44 patients with endometrial thickening.

It was not an objective of the present study to evaluate the uterine bleeding profile or the rate of amenorrhea in these menopausal women using HT in the form of implants, as described above. However, it is known that endometrial bleeding may be a consequence of endometrial atrophy or may be associated with specific endometrial pathology such as polyps or hyperplasia that may appear during use of the currently available regimes of HT [19,20].

In our study, the incidence of endometrial polyps was higher than that reported by Karlsson's group. This difference is probably due to the fact that, in their study, as well as in other large studies, curettage was used to evaluate the uterine cavity, whereas in our study hysteroscopy with guided biopsy was used.

In the present study, in 15 of the 27 cases in which hysteroscopy detected polypoid lesions (55.6%), histology confirmed this diagnosis, i.e. there was lack of concordance in only 44.4% of cases. These results are compatible with results published in the literature with respect to the incidence of endome-trial pathologies in users of estrogen/progestin regimens [21]. Machado and co- workers evaluated the accuracy of hysteroscopy as a tool for studying the uterine cavity of postmenopausal patients with vaginal bleeding. These investigators compared the images obtained by hysteroscopy with histological findings, and showed a sensitivity and specificity of 85.7% and 88.7%, respectively, for this method [21]. Scavuzzi and colleagues also found a high incidence of agreement between hysteroscopic and histopathological findings using an adequate biopsy technique, and reached a correct diagnosis in 13 out of 16 cases of cancer in postmenopausal patients with vaginal bleeding [22]. Agreement between hysteroscopic and histological findings of endometrial polyp in our study was 55.6%, lower than that reported by Scavuzzi's group. One possible explanation for the lower agreement rate between the two diagnostic methods may be the fact that macroscopic images of polypoid lesions are less clear-cut than those of malignant lesions.

In a study involving 35 postmenopausal patients with a diagnosis of endometrial thickening according to transvaginal sonography, Sheth and associates [23] found a predominance of polypoid lesions in 28.6% of cases. Accorsi and co-workers carried out a study in 58 postmenopausal patients with an endometrial echo >/=4 mm, and reported the presence of polyps as the principal cause of endometrial thickening in 51.7% of cases [24]. Loizzi and collaborators also reported polyps as the principal lesions responsible for endometrial thickening, identifying this pathology in 23.2% of 155 patients investigated [25]. Our results using hysteroscopy in women with a sonographic finding of endometrial thickening are in agreement with data from the aforementioned studies. Histology revealed endometrial polyp as the most frequent pathology found in patients in this study using any of the three doses of implants evaluated.

In the group of women using 4E + 2T, the occurrence of endometrial polyps was detected at histology in 30.0% of cases of endometrial thickening, i.e. in 4.4% of all users of this dose for two consecutive years (Table I). In the group using 5E + 2T, endometrial polyps were responsible for endometrial thickening in 41.6% of cases, i.e. 7.3% of these users for two consecutive years (Table II). In the 6E + 2T group, endometrial polyps were found in 50.0% of cases of endometrial thickening, making a total of 10.9% of the initial 55 patients.

Although the incidence of endometrial polyps was higher in the 6E + 2T group compared with the 4E + 2T and 5E + 2T groups, this difference was not statistically significant according to Student's t test for independent samples or the Mann-Whitney test.

The results reported in the present study are similar to those seen in postmenopausal women using combined estrogen/progestin therapy in which the progestin protects the endometrium from the proliferative changes caused by the estrogens in the endometrium, but fails to inhibit the growth of polyps [5].

Our findings are in agreement with the data from a meta-analysis of long cycles of estrogen therapy with progestins administered at varying intervals, which showed polyps to be the most frequently found endometrial lesions [26]. Similarly, it is possible that testosterone exerts its antiproliferative effects on the endometrium but not on polyps.

Among the patients with endometrial thickening, simple endometrial hyperplasia was found in 4/20 cases (20.0%) using 4E + 2T, in 1/12 patients using 5E + 2T (8.3%) and in 4/12 users of 6E + 2T (33.3%). According to the literature, simple endometrial hyperplasias are lesions with a low potential for malignancy, and there is currently some controversy regarding whether or not they are indeed precursor lesions of endometrial carcinoma [27]. The fact that in our study we found only simple endometrial hyperplasia with no atypia may stimulate the adherence of the medical community to this option of HT. Although the occurrence of simple endometrial hyperplasia was greater in the group of users of 6E + 2T compared with the groups of users of 4E + 2T and 5E + 2T, this difference was not statistically significant.


The results of the present study show that, in the majority of postmenopausal women using implants of estradiol and testosterone continuously for two years in whom sonographic endometrial echo was > 5 mm, lesions of the uterine cavity were detected by hysteroscopy with biopsy.

Hysteroscopy was a highly accurate method for identifying intrauterine abnormalities, and should always be followed by biopsy for histological confirmation.

The cut-off point of 5 mm adopted by this institution as the criterion for the investigation of endometrial thickness in users of estradiol and testosterone implants is appropriate in view of its relationship with the presence of polyps, although this criterion will inevitably result in the investigation of a certain number of patients who will be found to have a normal endometrium.

Histology carried out in patients evaluated in our study revealed that endometrial polyps were the most common pathology at any of the three dose combinations of estradiol and testosterone used.

The association of estrogen and testosterone in the form of subcutaneous implants for two consecutive years did not lead to any greater occurrence of endometrial pathology than that reported in the classic studies on the endometrium of postmenopausal women using estrogen/progestin combinations in which polyps are also the principal pathology found.

In our study, in addition to polyps, simple endometrial hyperplasia, which has a low malignant potential, was found.

The fact that endometrial polyps and simple endometrial hyperplasia were more common in the group of women using the highest dose of estradiol implants studied, although not statistically significant, reaffirms the recommendation to use the lowest possible dose of HT.

In our study, no cases of endometrial cancer were found in the 44 patients using implants of estradiol and testosterone for two continuous years, who had endometrial thickening as diagnosed by vaginal sonography.

The possibility of using testosterone to protect the endometrium is promising, but larger population-based studies need to be carried out to obtain more data. The data reported in the present study, however, are encouraging with respect to endometrial protection.


1. Coutinho EM, Teixeira CE, Maltez A, Maia H Jr. Long-term testosterone replacement therapy with slow-release silastic implants. In: Genazzani AR, editor. Advances in gynecological endocrinology: The proceedings of the plenary sessions of the 8th World Congress of Gynecological Endocrinology. Florence, Italy, December 2000. London: Parthenon Publishing Group; 2002. pp 259- 266.

2. Tworoger SS, Missmer SA, Barbieri RL, Willett WC, Colditz GA, Hankinson SE. Plasma sex hormone concentrations and subsequent risk of breast cancer among women using postmenopausal hormones. J Natl Cancer Inst 2005;97:595-602.

3. Tuckerman EM, Okon MA, Li T, Laird SM. Do androgens have a direct effect on endometrial function? An in vitro study. Fertil Steril 2000;74:771-779.

4. Maia H Jr, Maltez A, Fahel P, Oliveira M, Coutinho EM. Hysteroscopic findings in postmenopausal patients with a thick endometrium after using implants of oestradiol and testosterone. Gynaecol Endosc 2000;9:259-265.

5. Maia H Jr, Maltez A, Calmon LC, Oliveira M, Marques D, Coutinho EM. Histopathology and steroid receptors in endometrial polyps of postmenopausal patients under hormone-replacement therapy. Gynaecol Endosc 1998;7:267-272.

6. Longcope C, Baker S. Androgen and estrogen dynamics: relationships with age, weight and menopausal status. J Clin Endocrinol Metab 1993;76:601-604.

7. Maia H Jr, Maltez A, Fahel P, Athayde C, Coutinho E. Detection of testosterone and estrogen receptors in the postmenopausal endometrium. Maturitas 2001;38:179-188.

8. Maia H Jr, Pimentel K, Silva TM, Freitas LA, Zausner B, Athayde C, Coutinho EM. Aromatase and cyclooxygenase-2 expression in endometrial polyps during the menstrual cycle. Gynecol Endocrinol 2006;22:219-224.

9. Maia H Jr, Maltez A, Fahel P, Coutinho E. Hysteroscopic and immunohistochemical findings in type I and type II endome-trial carcinomas. J Am Assoc Gynecol Laparosc 2001;8:222-230.

10. Beral V, Bull D, Reeves G; Million Women Study Collaborators. Endometrial cancer and hormone-replacement therapy in the Million Women Study. Lancet 2005;365:1543-1551.

11. Magos AL, Brincat M, O'Dowd T, Wardle PJ, Schlesinger P, Studd JW. Endometrial and menstrual response to subcutaneous oestradiol and testosterone implants and continuous oral progestagen therapy in post-menopausal women. Maturitas 1985;7:297-302.

12. Cardozo L, Gibb DM, Tuck SM, Thorn MH, Studd JW, Cooper DJ. The effects of subcutaneous hormone implants during climacteric. Maturitas 1984;5:177-184.

13. Stahlberg C, Pedersen AT, Lynge E, Ottesen B. Hormone replacement therapy and risk of breast cancer: the role of progestins. Acta Obstet Gynecol Scand 2003;82:335-344.

14. Chanson P. [Historical perspective of hormone replacement therapy.] Rev Prat 2005;55:369-375.

15. Bolour S, Braunstein G. Testosterone therapy in women: a review. Int J Impot Res 2005;17:399-408.

16. Somboonporn W, Davis S, Seif MW, Bell R. Testosterone for peri- and postmenopausal women. Cochrane Database Syst Rev 2005;(4):CD004509.

17. Karlsson B, Granberg S, Wikland M, Ylostalo P, Torvid K, Marsal K, Valentin L. Transvaginal ultrasonography of the endometrium in women with postmenopausal bleeding: a Nordic multicenter study. Am J Obstet Gynecol 1995; 172: 1488-1494.

18. Gredmark T, Kvint S, Havel G, Mattsson LA. Histopathological findings in women with postmenopausal bleeding. Br J Obstet Gynaecol 1995;102:133-136.

19. Leather AT, Savvas M, Studd JW. Endometrial histology and bleeding patterns after 8 years of combined estrogen and progestagen therapy in postmenopausal women. Obstet Gynecol 1991;78:1008-1010. 20. Vashisht A, Wadsworth F, Carey A, Carey B, Studd JW. Bleeding profiles and effects on the endometrium for women using a novel combination of transdermal oesrradiol and natural progesterone cream as a part of a continuous combined hormone replacement regime. Br J Obstet Gynaecol 2005;112:1402-1406.

21. Machado MKN, Pina H, Matos E. [Hysteroscopy accuracy in the evaluation of the uterine cavity of patients with postmenopausal bleeding.] Rev Bras Ginecol Obstet 2003,25:237-241.

22. Scavuzzi A, Amorim M, Pinho Neto JS, Santos LC. [Comparison of ultrasonographic, hysteroscopic and histopathologic findings in women with postmenopausal uterine bleeding.] Rev Bras Ginecol Obstet 2003;25:229-235.

23. Sheth S, Hamper UM, Kurman RJ. Thickened endometrium in the postmenopausal woman: sonographic-pathologic correlation. Radiology 1993;187:135-139.

24. Accorsi Neto AC, Goncalves WJ, Mancini SN, Soares JR Jr, Haidar MA, Lima JR, Baracat EC. [Comparison between hysterosonography, hysteroscopy and histopathology in the evaluation of the uterine cavity of post-menopausal women.] Rev Bras Ginecol Obstet 2003;25:667-672.

25. Loizzi V, Bettocchi S, Vimercati A, Ceci O, Rossi C, Marello F, Greco P. Hysteroscopic evaluation of menopausal women with endometrial thickness of 4 mm or more. J Am Assoc Gynecol Laparosc 2000;7:191-195.

26. Lethaby A, Suckling J, Barlow D, Farquhar CM, Jepson RG, Roberts H. Hormone replacement therapy in postmenopausal women: endometrial hyperplasia and irregular bleeding. Cochrane Database Syst Rev 2004;(3):CD000402.

27. Mutter GL. Endometrial intraepithelial neoplasia (EIN): will it bring order to chaos? The Endometrial Collaborative Group. Gynecol Oncol 2000;76:287-290.



1 Centro de Pesquisa e Assistencia em Reproducao Humana (CEPARH), Salvador, Bahia, Brazil, 2 School of Mediane,

Federal University of Bahia, Salvador, Bahia, Brazil, and 3 Matemidade Climerio de Oliveira, Teaching Hospital,

Federal University of Bahia, Salvador, Bahia, Brazil

(Received 29 November 2006; revised 28 June 2007; accepted 28 June 2007)

Correspondence: A. M. B. de Harros Filho, Rua Almirante Barroso 110 - Apt. 231, Rio Vennelho, 41950 350 Salvador, Bahia, Brazil. Tel: +55 71 3334 6127. Fax: 55 71 3331 6399. E-mail: [email protected]

Copyright Taylor & Francis Ltd. Sep 2007

(c) 2007 Gynecological Endocrinology. Provided by ProQuest Information and Learning. All rights Reserved.