Investigation of Abnormal Uterine Bleeding in Postmenopausal Women

Chapter

11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women Linda D. Bradley

Most women spend one third of their lives after the menopause. For most women, the reproductive organs are senescent. Longevity, obesity, endogenous production of estrogen, and hormonal therapy invariably lead to episodes of postmenopausal bleeding (PMB). After years of amenorrhea, the onset of any bleeding leads to physician and patient concern. Even small amounts of bleeding need prompt and thorough evaluation.

ENDOMETRIAL CANCER Endometrial cancer is the most common genital female malignancy.1 In 2005, there were 41,000 cases of endometrial cancer.2 It is the fourth most common cancer in women, ranking behind breast, bowel, and lung cancers. Endometrial cancer is twice as common as ovarian cancer and three times more common than invasive carcinoma of the cervix. Average age of diagnosis is 65 years, and the lifetime risk of diagnosis of endometrial cancer is 2.7%. Endometrial cancer accounts for almost half of all new cases of genital cancer. It usually has an excellent prognosis and is often referred to as the “curable cancer” due to early detection and surgical intervention. Overall 5-year survival of endometrial cancer is 86%, and when the disease is confined within the uterus, survival rates peak at 97% at 5 years. A sobering statistic demonstrates the disparity between white and black patients with endometrial cancer. Although more white women develop endometrial cancer than African American women, African American women have a higher mortality rate than white women. African American women have a later diagnosis, more advanced stage, and a more lethal histologic subtype (serous tumors) of endometrial cancer than white women. The uncorrected survival rate for endometrial cancer is 75%. Moreover, the low death rate from endometrial cancer, which contributes to 23% of all gynecologic cancer deaths, is usually due to early detection coupled with prompt surgery. Currently, fewer than 4,000 women die of endometrial cancer per year. Although the total death rate for endometrial cancer has decreased since 1980s, endometrial cancer is not an indolent

disease. Earlier diagnosis leads to improved survival and better outcomes. Endometrial cancer rates peaked in 1979 and have declined since then. The increased incidence, witnessed in the 1970s, was likely due to unopposed estrogen therapy, which was commonly prescribed by the medical community. Now it is known that women who desire hormone replacement therapy (HRT) should also take a daily progestin or at least 12 days of progesterone with estrogen therapy. Prolonged unopposed estrogen therapy is not indicated in women with an intact uterus or prior endometrial ablation. Most patients with endometrial cancer are symptomatic and present with PMB or a lengthy bout of perimenopausal bleeding. Stage for stage, endometrial cancer is just as lethal as ovarian cancer; however, the tangible hallmark symptoms–namely, PMB, pelvic cramping, or unexplained chronic vaginal discharge–should prompt early investigation. Patients and physicians must not ignore subtle symptoms. With early identification and initiation of treatment, better prognosis is assured. Almost 75% of cases of endometrial cancer occur in women older than 50 years, and less than 5% of cases occur in women younger than 40 years. There are more cases of uterine cancer in the advanced geriatric population (>65 years) than in women between ages 45 and 55.3 In younger women, endometrial cancer is most often associated with obesity, chronic anovulation, long history of abnormal bleeding, and coexisting ovarian malignancy. Clinicians must evaluate abnormal bleeding aggressively in older women, because 95% of endometrial cancers occur in women 40 years and older and because endometrial hyperplasia, the precursor state, can precede the diagnosis.4 Some women are asymptomatic and fortuitously identified by a thickened endometrial echo on transvaginal ultrasound (TVUS) performed for other indications or when the uterus is removed for other indications. When the highest risk factors for endometrial cancer are grouped—diabetes (when due to obesity), nulliparity, age greater than 70 years, and vaginal bleeding—87% of those women have endometrial cancer.5 Ovarian cancer and cervical cancer occur in 3% of women with PMB.6 Fallopian tube carcinoma, a true rarity, must also

115

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women be considered when persistent vaginal bleeding occurs despite a negative evaluation of the endometrium. These statistics should not be forgotten, especially in the patient who continues to bleed after a negative evaluation of the uterine cavity. Remember to image the adnexa when endometrial evaluation is unremarkable.

POSTMENOPAUSAL BLEEDING PMB is defined as any bleeding that occurs 12 months after the last menstrual period. Noncyclic vaginal bleeding in postmenopausal women receiving HRT warrants evaluation. The average age of menopause has remained steady, occurring in 80% of women by age 51 in and 95% of women by age 55, with virtually all women experiencing cessation of menses by age 58. The highest rate of postmenopausal bleeding occurs within the first year after spontaneous amenorrhea and gradually declines thereafter. This is due to the responsiveness of the endometrium and possibility because a few remaining oocytes precipitate estrogen secretion. A prospective observational population study of 297 postmenopausal women completed 1 year of daily recording of bleeding. Postmenopausal bleeding occurred in 409 per 1000 person-years in the first year immediately after menopause. When more than 3 years elapsed, the rate plummeted to 42 per 1000 person-years.7 Luckily, only 5% to 20% of women with postmenopausal bleeding have endometrial cancer or endometrial hyperplasia.8 Currently, the American Cancer Society guidelines state that there is no proven role of screening asymptomatic women who have an average risk of endometrial cancer.9 Even among symptomatic women taking tamoxifen, no surveillance of the endometrium is advocated. Outpatient evaluation is preferable, when possible, to evaluate PMB. Outpatient evaluation, especially with office hysteroscopy or saline infusion sonography (SIS, also known as sonohysterography [SHG]) is efficient, cost-effective, and well tolerated and has comparable accuracy to inpatient evaluation.

sure that I have not missed a focal lesion?” Karlsson and colleagues reviewed 1168 cases of postmenopausal bleeding with TVUS followed by D&C.11 The study group included 351 women taking HRT, who were between ages 41 and 91 years. Karlsson’s group found that more than 59% of cases of PMB were due to endometrial atrophy; polyps (12%); endometrial hyperplasia (9.8%); endometrial cancer (10%); hormonal effect (7%); hydrometra, pyometra, and hematometra (2%); and cervical cancer (<1%). Consistently, polyps appear as the most common benign structural abnormality in women with postmenopausal bleeding who are evaluated by SIS or hysteroscopy, especially those with breast cancer treated with tamoxifen (Fig. 11–1).12 Tamoxifen has both antiestrogenic and estrogenic effects on the endometrium. The antiestrogenic effects cause atrophy leading to thin, fragile, atrophic endometrium with superficial petechial lesions that bleed easily. Because endometrial cancer is rare, it is most helpful to use less-invasive diagnostic tests that have a high negative predictive value and high sensitivity, thereby reliably excluding cancer, but that can detect intracavitary lesions that can cause bleeding.13

Clinical Symptoms Luckily, most women with endometrial cancer are symptomatic. Sometimes the symptoms are subtle, but clinicians must be astute and question postmenopausal women in detail at annual visits. Unlike ovarian cancer, the silent killer, endometrial cancer patients usually complain of abnormal bleeding, pelvic cramping, or vaginal discharge, albeit in subtle and dismissive ways. Quite often the patient minimizes the amount of bleeding, stating, “I only see a little spot of blood now and then.” As gynecologists, we too are dismissive, because the amount of bleeding seems innocuous. If the patient admits to even a smidgen of blood, gynecologists are obligated to investigate thoroughly. Be persistent. Additional complaints can include abnormal discharge,

Evaluation What modality is best to evaluate women presenting with postmenopausal bleeding? Options include office endometrial biopsy, dilation and curettage (D&C), office hysteroscopy, TVUS, SIS, computed tomography (CT), magnetic resonance imaging (MRI), and three-dimensional (3D) ultrasound. Although 5% to 15% of all gynecologic visits are for the evaluation of postmenopausal bleeding, the correct work-up and triage is still confusing for most clinicians.10 Traditionally, endometrial biopsy was considered the gold standard for the evaluation. If the biopsy was normal, then the clinician and patient were both reassured. The challenge, however, is the two thirds of patients with PMB who have an endometrial biopsy that demonstrates endometrial atrophy or tissue insufficient for diagnosis. When the clinician obtains results of atrophy, the perpetual question or query is, “Am I

116

Figure 11–1 A large hemorrhagic polyp in a patient taking tamoxifen.

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women staining, frank blood, and cramping. In fact, 80% of patients with endometrial cancer experience abnormal bleeding, and 10% have leukorrhea. Women should be taught any vaginal bleeding after the menopause needs prompt evaluation. The amount of bleeding is not correlated with the stage of endometrial cancer. Characteristics of bleeding include spotting, mild staining, or frank bright red blood. A subtle but very important clinical caveat is to inquire about abnormal vaginal discharge. The vast majority of women in menopause have vaginal atrophy and annoying vaginal dryness. Vaginal moisture (when not due to urinary incontinence) is unusual. Women subsequently found to have endometrial hyperplasia or endometrial cancer often complain of varying types of discharge from clear to faint pink, serosanguinous, or bright red. Sometimes heavy bleeding and a discharge with associated clotting occur. Pelvic cramping, another subtle complaint, may be due to cervical stenosis or cervical synechiae that prevent egress of blood. Pyometria and hematometra can coexist with postmenopausal bleeding. This may be preceded by uterine cramping or a feeling that “my period feels like it is coming on.” Those with advanced and extrapelvic disease can present with pelvic or uterine cramping, pelvic pressure, back pain, or fever of unknown origin before frank bleeding is noted. The bleeding may be episodic, and length of time ranges from days to months before the patient seeks care. The risk of endometrial cancer increases as the number of years since the last menstrual period elapses. A patient who notes vaginal bleeding more than 10 years since her last menses has a higher risk of uterine cancer than one who recently stopped menstruating. Likewise, women on unopposed estrogen therapy and tamoxifen have a higher incidence of endometrial cancer than women not taking those medications.

Risk Factors Although endometrial cancer accounts for less than 10% of the causes of postmenopausal bleeding, the work-up must be exhaustive before a benign etiology is determined. The survival of patients with endometrial cancer is influenced by stage, grade of disease, and histology. The risk of endometrial cancer increases with each decade of life, such that at age 50, 1% of cases are due to endometrial cancer, and by age 80, 25% of cases of postmenopausal bleeding are due to cancer.14 Risk factors for endometrial cancer include obesity, unopposed estrogen therapy, prolonged tamoxifen use, anovulation, polycystic ovarian syndrome, estrogen-secreting tumors of the ovary, nulliparity, late menopause, and history of complex endometrial hyperplasia with atypia. Medical illnesses associated with endometrial cancer include diabetes, hypertension, arthritis, and hypothyroidism. Obesity is clearly associated with increased risk of endometrial cancer. Women weighing more than 30 pounds over ideal weight have a threefold increased risk of developing endometrial cancer, and those 50 pounds or more over ideal weight have a tenfold increased risk. Other increased risk factors include nulliparity (double), diabetes (threefold), hypertension (1.5-fold), entering menopause after age 52 years

(2.5-fold), and increased bleeding at the time of menopause (fourfold). Patients taking unopposed exogenous estrogen therapy have a four-to 15-times greater risk, and 29% of patients with untreated complex hyperplasia with atypia develop endometrial adenocarcinoma. Newer studies estimate that 17% to 52% of women with endometrial hyperplasia with atypia concurrently have endometrial cancer. Although there are many theories for endometrial cancer, the greatest risk is found in patients in patients taking unopposed estrogen or with higher circulating levels of endogenous estrogen. The TVUS endometrial measurement (endometrial echo or endometrial stripe) is helpful in categorizing patients into a lowrisk or a high-risk group. The exact cutoff measurement chosen is a function of the sensitivity and specificity sought. Most clinicians use a cutoff of 5 mm to define a low-risk patient group, whose combined risk for cancer and atypical hyperplasia ranges from 2% to 3%. The initial goal of ultrasound was not to replace endometrial biopsy but to decrease the number of endometrial biopsies needed.15 Ultrasonographers painstakingly sought to find the magic endometrial echo number, for which the likelihood of missing endometrial cancer would be exceedingly low.16 Stratification of patient risks into low-risk groups include less than 70 years, multiparity, bleeding that occurs within 1 year of menopause, and a no diabetes. In contrast, patients with an endometrial echo of greater than 5 mm have an increased risk of endometrial cancer and atypical hyperplasia approaching 5% or more. Patients with all four risk factors—diabetes, nulliparity, age greater than 70 years, and bleeding that occurs more than 1 year after menopause—have an 80% risk for endometrial cancer or complex hyperplasia. Unfortunately, many articles evaluating the usefulness of the endometrial measurement have included asymptomatic patients, and therefore it is likely that patients presenting with bleeding have a higher detection rate than asymptomatic patients. Remain vigilant in the work-up of your patients with postmenopausal bleeding. TVUS, SIS, and office hysteroscopy are not mutually exclusive studies; they are complementary tools. When the patients’ bleeding persists despite a negative biopsy or normal transvaginal ultrasound, continue to look again with additional technology. If only TVUS and endometrial biopsy were initially used, then consider direct inspection of the endometrium with office hysteroscopy or SIS if bleeding recurs. The interface created by fluid during SIS delineates intraluminal defects well. Likewise, direct visualization of the endometrium with hysteroscopy detects intracavitary pathology.17 Several authors have found that when then initial endometrial biopsy was negative, on follow-up, 20% of women with persistent or episodic vaginal bleeding were ultimately found to have either complex atypical hyperplasia or cancer.18

CLINICAL PEARL ●

117

Vigilance and using additional technology is important to minimize missing intracavitary pathology.

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women EVALUATION History and Physical Examination A detailed history and physical examination should be obtained for all patients presenting with postmenopausal bleeding. Query the patient about current or recently discontinued medications. Inspect the vagina, vaginal fornices, vulva, and urethra thoroughly. Urethral prolapse may be the cause of the vaginal bleeding (Fig. 11–2). Genital and vulvar lesions must not be overlooked. Biopsy any palpable or visible lesions. Any vulvar lesion with chronic excoriation should be biopsied liberally. Vaginal atrophy is a common cause of abnormal bleeding. Thin, friable, tissue that bleeds easily spontaneously or with intercourse is common with each advancing decade. When vaginal atrophy is noted, excellent response to oral or topical estrogen therapy is the norm. It can take 8 to 12 weeks of continuous estrogen therapy to notice clinical improvement. Estrogen therapy increases vaginal blood flow, increases glycogen in vaginal cells, changes the pH from alkaline to acidic, increases the amount of lactobacillus, and increases collagen and elastic tissue within the vagina. These changes decrease the friability and fragility of the vaginal epithelium, thereby decreasing the frequency of vaginal bleeding associated with atrophy.

Cervical Cytology The Papanicolaou (Pap) test is not a screening method for endometrial cancer. However, the Pap test must be included in patients with PMB to exclude cervical or endocervical causes of bleeding. Liquid-based Pap smear is warranted for all women with new-onset postmenopausal bleeding, even when the patient has had a Pap test within 3 years. Even though newer cervical cytology guidelines propose less frequent Pap tests in the meno-

Figure 11–2 A 78-year-old woman presented with abnormal bleeding. A large urethral prolapse was detected. After repair and topic estrogen, her symptoms resolved.

pause, remember that there is a bimodal distribution of cervical cancer as well as its indolent endocervical adenocarcinoma counterpart. The mean age of cervical cancer diagnosis is 52 years, with bimodal peaks at age 35 years 39 and again at 60 to 64 years.2 Karlsson’s study of 1168 postmenopausal women11 found a 1% rate of cervical cancer among postmenopausal women studied. Paradoxically, some asymptomatic patients presenting for routine annual examination have endometrial cells detected on Pap. When benign endometrial cells are present on the Pap smear in asymptomatic postmenopausal women, endometrial cancer is rare. More often, atrophy or endometrial polyps are the cause of exfoliated endometrial cells detected on Pap among asymptomatic women. The ideal work-up of such patients is visualization with hysteroscopy or SIS. Although very uncommon, fallopian tube carcinoma must be considered in women with persistent postmenopausal bleeding.19 The mean age of patients with fallopian tube carcinoma is 69 years. Hysteroscopy is most often normal in women with fallopian tube carcinoma; however, a high index of suspicion should be maintained. TVUS is strongly advised for patients with persistent bleeding if hysteroscopy has been the only modality used in the evaluation. Classic findings of fallopian tube carcinoma include adnexal mass, abdominal pain, hydrosalpinx, watery vaginal discharge, and increased pelvic fluid.

ENDOMETRIAL BIOPSY AND TRANSVAGINAL ULTRASOUND Rationale Do we need to rethink the role of endometrial biopsy in evaluating the patient with postmenopausal bleeding? Is endometrial biopsy, when used alone, an outdated modality in evaluating postmenopausal bleeding? Most studies demonstrate the inadequacy of blind endometrial sampling, including D&C, in women with focal lesions.20 A blind endometrial pipelle biopsy is woefully inaccurate in detecting focal pathology. Increasingly, TVUS is proposed as the first-line, minimally invasive tool for evaluating postmenopausal bleeding. Visualization of the entire endometrial echo is imperative in order to determine the health of the endometrium. In general, if TVUS is used alone, an endometrial stripe that is well delineated in its entirety, is homogenous, and has no fluid and no irregularities is unlikely to harbor endometrial cancer when the endometrial echo is less than 5 mm and the patient has minimal risk factors.21 The adnexa should be imaged to rule out adnexal pathology. Using a threshold of less than 5 mm for endometrial echo is the most cost-effective diagnostic strategy for detecting endometrial cancer.22 However, an endometrial echo less than 5 mm can harbor other pathology that causes postmenopausal bleeding. Endometrial atrophy, polyps, endometritis, submucosal fibroids, pyometria, and proliferative and hyperplastic endometrium can be present with an endometrium less than 5 mm. Do not stop the work-up with an endometrial echo of less than

118

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women 5 mm in a symptomatic patient. Your patient had the initial test because of a complaint: bleeding. Listen to her story and evaluate thoroughly. Certainly the patient is reassured by being told that she does not have cancer. But what does she have that makes her bleed, now and then, or once in a while? A report indicating an endometrial echo less than 5 mm is very unlikely to be malignant, However, when symptoms persist or the amount of bleeding is profuse, reevaluate with direct visualization with hysteroscopy (preferably office based), hysteroscopically directed biopsy, or SIS. When a focal lesion is detected that cannot be completely removed in the office, ambulatory operative hysteroscopic surgery is indicated. Similarly, if a patient with postmenopausal bleeding has a thickened endometrial echo, especially greater than 5 mm, and has endometrial curettings obtained blindly or biopsy that is insufficient, additional evaluation with hysteroscopy or SIS is mandatory (Fig. 11–3). Most of these patients have a focally growing lesion that can be hysteroscopically resected. More than 50% of pathology will be missed using D&C alone. Benign pathology causes bleeding. Polyps and hyperplasia are culprits for bleeding. Less than 2% of polyps contain a malignancy, complex hyperplasia with atypia, or hyperplasia. Malignant changes within polyps can rarely be detected with ultrasound (including Doppler flow) or hysteroscopic visualization alone. Histologic evaluation is mandatory. Even small polyps (<2 cm) can be malignant.23 Utmost care must be taken when removing polyps in order to prevent total thermal damage to the specimen, so that it can be evaluated histologically. Ulceration, hemorrhage, increased surface vascularity, increased endometrial surface area, and progesterone insensitivity are possible causes of the bleeding. Operative hysteroscopy is crucial in all women with postmenopausal bleeding who have a focal lesion. Histologic evaluation of the focal lesion is necessary to make a diagnosis. A polypoid growth can actually be a focal area of hyperplasia, malignancy, or mesenchymal tumor.24

SIS

1 1

A

1 2

3

B

Transvaginal Ultrasonography A thorough review of the literature clearly indicates that in an endometrium that is seen entirely by TVUS and is homogenous, a thickness less than 5 mm likely does not harbor endometrial cancer. However, TVUS alone rarely can determine the presence of focal lesions. Meta-analysis of 35 studies involving 5892 women using an endometrial thickness cutoff found that a 5 mm cutoff had a greater than 92% sensitivity for detecting endometrial disease (polyps, atypical hyperplasia, or cancer) and a 96% sensitivity for detecting endometrial cancer. A postmenopausal woman with a 10% pretest probability of endometrial cancer thus had a 1% probability of cancer if her TVUS had an endometrial echo less than 5 mm.25 Tsuda and colleagues evaluated 600 postmenopausal women and used different endometrial echo cutoff values based on the number of years since menopause. If the patient was less than 5 years in the menopause, the endometrial echo was set at

C Figure 11–3 A 67-year-old woman who was taking hormone replacement therapy presented with postmenopausal bleeding. A “nondiagnostic” endometrium was noted on pipelle biopsy. A, Endometrial echo was 12 mm. B, Saline infusion sonography (SIS) demonstrates three intracavitary lesions. C, the three intracavitary lesions were confirmed at operative hysteroscopy. Polypectomy removed benign endometrial polyps.

119

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women 4 mm. If she was more than 5 years past menopause, the cutoff was 3 mm. TVUS demonstrated a 97.4% sensitivity, 75.7% specificity, 23.8% positive predictive value, and 99.7% negative predictive value.26 Gull and colleagues followed 394 Scandinavian women followed for 10 years. The subjects initially presented with PMB and had an endometrial echo less than 4 mm; and no cases of endometrial cancer were found by TVUS compared to curettage.16 A Nordic multicenter trial found no cases of endometrial cancer among 1168 women with PMB who had an endometrial echo less than 4 mm and underwent curettage.11 Fleischer and colleagues used TVUS to evaluate 1750 women (without PMB) for a selective estrogen receptor modulator (SERM) study.27 When the endometrial echo was less than 5 mm, the negative predictive value was 99.94% for excluding malignancy (1 cancer in 1750 women) and it was 99.77% for complex hyperplasia (4 in 1750 women). Gull and colleagues, using a 4-mm cutoff found only seven cancers in 1361 women.28 When Epstein and Valentin studied 97 women who had PMB and an endometrial echo less than 5 mm, no endometrial cancers were present.29 The data are very solid for use of TVUS in evaluating postmenopausal bleeding. Literature review supports a conservative approach when the endometrial echo is less than 4 mm in a postmenopausal woman. The risk of missing an endometrial cancer is low. However to rely upon TVUS, the endometrial echo should be fully visualized and should appear homogeneous and without fluid. If the endometrial echo is indistinct, not visualized, or irregular and has a heterogeneous appearance, then additional surveillance with SIS or hysteroscopy is mandatory. Additionally, if bleeding recurs despite a normal TVUS, then a direct view of the endometrium is necessary. Despite a TVUS endometrial echo of less than 4 mm, in patients with persistent bleeding, office hysteroscopy or SIS should be the next tool to evaluate for subtle intracavitary lesions. When the endometrial echo is greater than 4 mm, SIS or hysteroscopy should be performed. The distention of the endometrial cavity permits identification of focal lesions, polyps, fibroids, dyssynchrous endometrium, hyperplasia, and cancer. If a focal lesion is identified, then a targeted endometrial biopsy (with full removal of the lesion) is the most sensitive method of determining the cause of the increased endometrial thickness. If a diffusely thickened endometrium is identified, then a blind aspiration biopsy with a pipelle or electronic suction (Vabra) aspirate provides a sensitive analysis (Fig. 11–4). If endometrial biopsy yields insufficient or inadequate tissue for evaluation, if no endometrial tissue obtained, or if stenosis is encountered, then direct visualization of the endometrium should be performed and directed biopsies taken.

Figure 11–4 A 57-year-old woman presented with postmenopausal bleeding. She was not taking hormone replacement. Examination showed circumferential thickening of the endometrium. Pipelle biopsy confirmed complex endometrial hyperplasia without atypia.

review of 65 articles that evaluated 26,345 women and the role of hysteroscopy determined that a positive hysteroscopy increased the probability of endometrial cancer to 71.8%, whereas a negative result reduced the probability of cancer to 0.6%.30 De Jong and colleagues note that the sensitivity and specificity for the diagnosis of endometrial disease was 78% and 95.8%, respectively. For a positive result, the pretest probability increased from 10.6% to 55.2% and decreased to 2.8% with a negative result. All women having hysteroscopy should also have an endometrial biopsy that provides a tissue diagnosis.30 There is little evidence that there hysteroscopy poses a risk of dissemination of malignant cells and a worsening of the stage of endometrial cancer in women with uterine cancer.31 When this risk is analyzed more critically, it appears that the dissemination of malignant cells with hysteroscopy, whether performed with saline or CO2, is not the critical factor. Rather, poor prognosis is associated with tumor grade, deep myometrial invasion, and disease already present outside of the uterine cavity.32 Hysteroscopy performed in the operating room should be reserved for women who cannot tolerate office evaluation, who present with cervical stenosis, or who have recurrent unexplained bleeding despite previous evaluation. Office hysteroscopy is quick and comfortable, and it accurately details the number and size of lesions within the endometrium.33 Small 3mm flexible hysteroscopes are comfortable when hysteroscopy is performed in an office setting.34

HYSTEROSCOPY Hysteroscopy, preferably office based, is excellent in evaluating the endometrial cavity and endocervix. The false-negative rate of hysteroscopy is 3%. Hysteroscopy is more accurate than TVUS in detecting focal disease and has greater specificity. A

Principles of Hysteroscopy in the Postmenopausal Patient Hysteroscopy is more accurate in detecting intracavitary lesions, such as polyps and fibroids, than blind endometrial biopsy alone.

120

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women A study of 181 patients reported a sensitivity of 96.6% and a specificity of 100% when hysteroscopy was combined with endometrial biopsy.35 Hysteroscopists sometimes find it difficult to distinguish between proliferative, exaggerated endometrium and endometrial hyperplasia.36 This is why it is critical to also perform endometrial biopsy with hysteroscopy. Theoretically, the specificity and positive predictive value of hysteroscopy in cases of abnormal uterine bleeding is 100%. In practice, however, the false-negative rate is 2% to 4% and is the result of operator error in detecting abnormal endometrial lesions. Evaluation of the endometrium hysteroscopically in postmenopausal women must be regimented, through, and complete. The hysteroscopist should systematically evaluate the endometrial height, surface, and vascular architecture. Uterine distention media can include CO2 or saline. However, both may be complementary and used concurrently. The endometrial thickness is usually less than 5 mm in the postmenopausal woman. The endometrial thickness can be accessed by pressing the tip of the hysteroscope into the endometrium. Without estrogen, there should be no compression or indentation of the endometrium. Under the influence of estrogen, there may be a wave-like surface with pseudopolypoid projections. Whether using CO2 or saline, it is important to distend the endometrium to provide full visualization. Distention of the endometrium can lead to a falsely negative view by compressing subtle intracavitary lesions. Therefore, at the conclusion of the procedure, make it a habit to decrease the intrauterine pressure and reinspect the endometrium closely. In so doing, you will miss fewer subtle lesions. The surface of the endometrium in the menopause is normally pale and can have a porcelain appearance. Pay attention to the presence of glands, gland openings, endometrial texture, vascularized projections, cysts, cystic openings, microcalcifications, and necrosis. Look closely at the endocervix, too. Notice the appearance of the vascular architecture. Normally the blood vessels are rarely seen on the surface of normal atrophic endometrium. Take heed when U-shaped blood vessels are seen, because this is highly associated with endometrial hyperplasia or endometrial cancer. Blood vessels with variable calibers or trajectory endings, or the appearance of hemorrhagic zones, should raise the flag for a malignancy. When atypical blood vessels appear, reinspect them with a lowered intrauterine pressure.

Pathology Results Endometrial cancer begins in the inner glandular lining and has two subtypes.37 Type 1 is more common and is associated with increased estrogen exposure. Risks factors include obesity, nulliparity, polycystic ovary disease, and unopposed estrogen. Type 1 endometrial cancers occur more often in younger women and are less aggressive. Histologic subtypes most often include lower-grade endometrioid adenocarcinomas that account for 75% to 80% of cases. Type 2 cancers occur in older women without estrogen exposure, arise spontaneously, and most often

include histologic subtypes of clear, serous, adenosquamous, and grade 3 adenocarcinomas. Endometrial cancer can arise from atrophic endometrium, from hyperplastic endometrium, or within an endometrial polyp. Ovarian malignancies including granulosa cell or endometrioid tumors can secrete estradiol and increase the risk of endometrial malignancies. Mixed müllerian tumors can arise within an endometrial polyp. Expeditious treatment with hysterectomy is indicated for women with atypical endometrial hyperplasia. Trimble and colleagues prospectively evaluated 289 women who had a diagnosis of atypical endometrial hyperplasia on endometrial biopsy who underwent hysterectomy within 12 weeks of the diagnosis.38 The results were sobering: At hysterectomy, concurrent endometrial cancer was found in 123 specimens (42.6%), and of these, 30.9% had myometrial invasion, including 10.6% in which the invasion was found in the outer 50% of the myometrium. In 18.9% of patients who received a second diagnosis of less than atypical endometrial hyperplasia, 39.1% of patients who received a second diagnosis of atypical endometrial hyperplasia, and 64.3% who received a second diagnosis of carcinoma all had carcinoma in their hysterectomy specimens. Thus 42.6% of patients had endometrial cancer when they underwent hysterectomy. It has been said that atypical endometrial hyperplasia should be treated surgically, and the old dictum of “not cancer, but better out” should be reworded to “likely cancer, and definitely better out.”

Hysteroscopic Findings in Postmenopausal Women Endometrial cancer has myriad appearances. Especially among postmenopausal women, the diagnosis is less likely to be missed. Boring, pale, atrophic, monotonous endometrial architecture best describes the postmenopausal endometrium. When the hysteroscopist detects thickened, irregular endometrium, increased surface vascularity, or friable cells or encounters difficulty distending the endometrium, the chance of endometrial cancer increases. To decrease the chance of artificially creating a negative hysteroscopic view, it is imperative that the intrauterine pressure be decreased (and documented in your office note). In so doing, the true contours of the endometrium can be evaluated. A small-caliber hysteroscope is particularly advantageous in the menopausal woman. Currently available are 3-mm and 5-mm flexible or rigid hysteroscopes manufactured by a variety of companies. Cervical stenosis is more common during menopause compared to the reproductive years. Increasingly, however, misoprostol (Cytotec), when used orally or vaginally 8 to 12 hours before uterine instrumentation, softens the cervix and facilitates cervical dilation. Misoprostol 200 to 400 mcg by mouth 2 days and 8 to 12 hours before hysteroscopy is extremely helpful. Minimal side effects of include bloating, cramping, diarrhea, and occasional bleeding. Atrophy

During menopause, the uterus involutes. The endometrium is pale, thin, and fragile and has a translucent and porcelain

121

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women appearance. The surface is smooth and dull. An atrophic endometrium might have focal mucus-containing cysts covered by a thin, pale surface. The intrauterine cavity is small, constricted, and narrow, when compared to that of a reproductive patient. Diffuse or isolated petechial hemorrhages are commonly noted when patients complain of postmenopausal bleeding. Landmarks such as the tubal ostia may be less well visualized or, when seen, can appear opaque or covered with a veil of thin, wispy adhesions. Isolated calcifications are often seen. Gland openings or cystic atrophy with openings adjacent to the myometrium are also noted. These gland openings appear translucent blue-gray. Because the endometrium is thin, the underlying muscle bundles of the myometrium and interlacing columns and recesses might produce diverticula. Uterine length when measured is shorter. The endocervix is long, cylindrical, and pale. Cervical stenosis may be encountered. The ratio of the cervical length to fundal length is also different than in the reproductive years. Due to the smaller uterine size and volume, extreme care must be taken with uterine manipulation in to decrease the risk of uterine perforation. Endometrial Polyps

Endometrial polyps are a common cause of postmenopausal bleeding. Polyps vary in size, are usually single, and are usually benign (Fig. 11–5). Polyps respond to estrogen and tamoxifen with growth, but they are less sensitive to progesterone. As polyps grow, they can develop a pedicle and can even protrude into the endocervix or into the vagina. Polyps share the same surface as the surrounding endometrium. They can have a variety of colors ranging from one that resembles the endometrium to red-yellow, and at the distal tip of the polyp an ecchymotic purple-blue color may be seen. In the postmenopausal woman, a pale endometrium covers the

Figure 11–5 Hysterectomy specimen of patient with a large intracavitary polyp filling the majority of the endometrial cavity. The pathology was benign.

polyp, with fragile blood vessels coursing throughout the surface of the polyp. The surface of a polyp in a postmenopausal woman may also contain bluish translucent cysts. Because of the polyp’s mobility, the hysteroscopist can push it away and look under its surface. If saline is used for distention, then the polyp can be seen to float away from endometrium. When CO2 is used, the polyp might appear compressed against the adjacent endometrium. Using the distal tip of the hysteroscope, the gynecologist can indent or compress a polyp. Large endometrial polyps that fill the endometrial cavity can lead the hysteroscopist to a false diagnosis of endometrial hyperplasia. The hysteroscopist views one side of the endometrium and then might view the atrophic opposite wall. Moving the hysteroscope to the lower uterine cavity and evaluating the upper anatomy can lead to fewer incorrect diagnoses. Submucosal Fibroids

Intracavitary fibroids are firm and protrude from the endometrium (Figs. 11–6 to 11–9). They are usually solitary but can be accompanied by additional leiomyomas. In the menopause, the endometrium covering the fibroid is thin. Blood vessels can be seen coursing over the surface of the myoma. On occasion, the endometrium opposite the myoma is ulcerated. When the myoma is palpated with the distal tip of the hysteroscope, resistance is met. Unlike a polyp, the myoma cannot be pushed away from the hysteroscope. Intramural fibroids, which are unlikely sources of bleeding during the menopause, unless the endometrium overlying it becomes markedly atrophic, appear as protuberances from the

Figure 11–6 Vaginal hysterectomy specimen demonstrating a single intracavitary leiomyoma. The patient had persistent postmenopausal bleeding. This procedure was performed without hysteroscopy. The patient had experienced three blind dilation and curettages before her physician performed this procedure.

122

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women

Figure 11–7 Fundal pedunculated leiomyoma in patient with postmenopausal bleeding. This was hysteroscopically resected, and her bleeding fully resolved.

Figure 11–9 Operative hysteroscopic view of the lower uterine segment of a patient with a 1-cm leiomyoma; an endometrial polyp can be seen in the foreground.

Endometrial Hyperplasia

Figure 11–8 Hysterectomy specimen of patient with two intracavitary leiomyomas. Most certainly if performed today, hysteroscopic resection would have been performed.

endometrium. The best view of an intramural fibroid appears when it is visualized from the internal os. Hysteroscopy is excellent for determining the number, size, and location and to determine if the patient has an intracavitary fibroid. Merely scraping with a curette is archaic, and in fact scraping can lead to more bleeding when the myoma is merely eroded. A pedunculated fibroid can be hysteroscopically resected completely. Detecting intramural fibroids can spare the patient unnecessary surgical intervention and can exclude other intracavitary lesions that might coexist within the uterine cavity. Most intracavitary leiomyomas are benign. The risk of sarcomatous changes occurs in less than 1% of patients with uterine fibroids.

Making the diagnosis of endometrial hyperplasia in the postmenopausal woman is less difficult than in reproductive-age patients who have a variable appearance of the endometrium due to hormonal fluctuations. Even so, the diagnosis of hyperplasia is the most often missed diagnosis with hysteroscopy compared to SIS.33 The most likely reason for missing the hyperplasia might result from our technique. When distention fluid or CO2 compresses the endometrium, the endometrial projections of hyperplastic tissue are more difficult to discern. Making the diagnosis of endometrial hyperplasia requires close surveillance of the tissue thickness, color, vasculature, and consistency of the endometrium. Hyperplasia can be focal or global (Fig. 11–10). Hyperplastic tissue has no organized structure and is an outgrowth of aberrant tissue, so it is easily friable and tears easily when touched with the hysteroscope. Abnormalappearing endometrium can have focal lesions that include polyps, which may be broad-based or pedunculated. Additional suspicion for hyperplasia includes focal or papillary mucosal projections with or without gland cysts, abnormal vascular network with atypical vessels, and crowded or abnormally spaced gland openings (Figs. 11–11 and 11–12).39 Endometrial hyperplasia cannot be diagnosed with SIS alone during the reproductive years because the ranges of endometrial thickness in hyperplasia and carcinoma overlap. However, during the menopause, endometrium does not change, and an endometrial echo greater than 4 mm raises the suspicion of endometrial abnormalities. A histologic diagnosis based on a hysteroscopically obtained endometrial office biopsy specimen is also required. Most hyperplasias of the endometrium are 0.6 cm to 1.3 cm thick in postmenopausal patients, with a mean thickness of 1 cm. In most endometrial cancers, the thickness is greater than

123

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women

Figure 11–12 Complex endometrial hyperplasia without atypia near the lower uterine segment. Directed biopsy was necessary to reach this diagnosis.

Figure 11–10 Focal endometrial complex hyperplasia without atypia near the fundus and anterior wall. The surrounding endometrium is otherwise normal.

times find it difficult to distinguish between proliferative, exaggerated endometrium and endometrial hyperplasia.36 This is why it is critical to also perform endometrial biopsy with hysteroscopy. Theoretically, the specificity and positive predictive value of hysteroscopy in cases of abnormal uterine bleeding, is 100%. In practice, however, the false-negative rate is 2% to 4% and is the result of operator error in detecting abnormal endometrial lesions. Endometrial Cancer

Figure 11–11 Fundal endometrial complex hyperplasia without atypia. Notice the thin bridging tissue surrounded by normal endometrium.

4.7 mm. Most often, hyperplasia occurs diffusely; however, it can be focal or can appear as a broad-based polyp. SIS can also reveal asymmetrical or multifocal areas of endometrial irregularities in endometrial hyperplasia. The endometrial–myometrial interface is intact. Hysteroscopy is more accurate in detecting intracavitary lesions, such as polyps and fibroids, than blind endometrial biopsy alone. A recent study of 181 patients reported a sensitivity of 96.6% and a specificity of 100% when hysteroscopy was combined with endometrial biopsy.35 Hysteroscopists some-

Endometrial cancer requires histologic diagnosis; however, the index of suspicion can be increased when the topography of the endometrium is irregular or has focal lesions. Sugimoto40 have noted a high sensitivity in detecting endometrial cancer when the following features are observed: papillary, polypoid, nodular or mixed endometrial growth demonstrating friable tissue, focal necrosis, and atypical vessels (Fig. 11–13). Guruti and colleagues retrospectively reviewed 25 menopausal patients who had atypical hyperplasia found on endometrial biopsy and treated by hysterectomy.41 The pathologic findings of the uterine specimen were correlated with the diagnosis obtained hysteroscopically. They found that the sensitivity was 84.6%, specificity was 100%, and negative and positive predictive values were 87.5% and 100%, respectively, with hysteroscopy for a diagnosis of infiltrating carcinoma.43 Garuti’s study41 relays an important message about a high false-negative rate when endometrial biopsy demonstrates atypical endometrial hyperplasia yet fails to diagnose infiltrating carcinoma in 44% of patients. Even targeted biopsies did not improve the diagnosis compared to blind tissue collection (Vabra). Why would this happen? Many hysteroscopists realize that the tissue obtained with targeted biopsy is small, and the endometrial stromal tissue or myometrial tissue invaded by cancer may be

124

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women

Figure 11–14 Thickening of the endometrium caused difficult uterine distention. Endometrial biopsy was consistent with adenocarcinoma of the endometrium

Figure 11–13 Typical cerebroid thickening of the endometrium. Endometrial biopsy was consistent with endometrioid endometrial cancer.

poorly sampled with small forceps. Failure to diagnose could result from biopsying only the most superficial layers of lesions, with poor sampling of stromal tissue; most operative hysteroscopes have a small 1.5 mm working channel, which uses a 4to 5-F biopsy grasper. Tissue handling of architecturally small samples or friable endometrium can lead to suboptimal tissue given to the pathologist. Finally, friable tissue bleeds easily, and when fluid irrigant is used, the field of view may be distorted, leading to less-accurate directed biopsies. If a symptomatic patient has a highly suspicious hysteroscopic lesion, yet endometrial biopsy is negative for malignancy, strongly consider repeat endometrial sampling with a resection loop electrode (monopolar or bipolar) to obtain excellent tissue samples, improved sampling of the base and deeper layers of the endometrium, and larger specimens, for the most accurate diagnosis. Some patients with endometrial cancer have a very patulous cervix with excessive cervical mucus. This is in stark contrast to cervical stenosis, which is encountered most often. Difficult endometrial distention may be encountered with saline or CO2. Variegated endometrium and diffuse clusters of vascular polypoid lesions are routinely visualized (Fig. 11–14). Leukorrhea, pyometria, and increased amounts of transudate are also hallmarks of coexisting cancer. Focal lesions such as endometrial polyps can also harbor endometrial cancer. Although hysteroscopy cannot stage endometrial cancer, the proximity to the endometrial cavity can be discerned. To avoid a false-negative hysteroscopy, deflate the intrauterine pressure during hysteroscopy.42 High uterine pressures can artificially flatten out lesions and push them into the endometrial cavity or myometrium. Look closely at the architectural landscape. Look at the blood vessels on lesions. Look for variegated colors, hemorrhage, necrosis, and irregularity of lesions. Most benign lesions have smooth borders (Figs. 11–15 and 11–16).

Figure 11–15 This 59-year-old woman was not taking hormone replacement therapy and presented with episodic postmenopausal bleeding. She had an 18-week size uterine fibroid. Friable, polypoid, and thickened endometrium was seen in addition to an intramural leiomyoma. Directed biopsy demonstrated endometrioid cancer.

Early stage endometrial cancer has an excellent prognosis. However, when extension to the cervix occurs, survival rates decrease. Stromal invasion, stage 2 disease, alters the pattern of spread and decreases survival rates to 60% to 70%. With 10% to 20% incidence of endometrial cancer involving the cervix, some gynecologists have attempted to determine hysteroscopically if cervical disease is present. Other techniques used to preoperatively stage endometrial cancer include endocervical curettage, TVUS, and MRI imaging. Lo and colleagues used hysteroscopic inspection of the cervix in 200 consecutive patients with primary endometrial cancer to

125

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women

A Figure 11–16 Necrotic friable thickened endometrium consistent with endometrioid endometrial cancer.

determine whether the tumor had spread to the cervix.43 Tumor involvement was noted in 20.5% of patients. Hysteroscopic accuracy was 92.5%, sensitivity was 68.3%, specificity was 98.7%, positive predictive value was 93.3%, and negative predictive value was 92.4% in determining cervical involvement. Grossly inspecting the cervix for malignancy was accurate 93% of the time. Hysteroscopic findings suggesting cervical involvement included exophytic growth, irregular surface contours, and abnormal vasculature. Use of normal saline compared to CO2 improved diagnostic accuracy.

B

Tamoxifen-Induced Changes

Endometrial surveillance of asymptomatic patients taking tamoxifen does not differ from that of women routinely using estrogen replacement therapy. Although most long-term users of tamoxifen have an inactive endometrium, some show increased endometrial thickness during conventional TVUS. Goldstein and colleagues15 advocated SIS to monitor tamoxifen effects because of improved imaging. With traditional TVUS, the endometrial thickness appears highly unusual and heterogeneous and demonstrates centrally located uterine changes. If SIS is not performed, then these unusual features noted with TVUS can easily be overinterpreted. Unlike TVUS, SIS more accurately detects endometrial health and can determine if additional imaging is necessary. SIS can identify the subendometrial sonolucencies to the proximal myometrium. The abnormalities might represent abnormal adenomyomatous-like changes in the proximal myometrium that are microcysts. When viewed microscopically, the junction between the endometrium and myometrium is irregular and nonlinear, whereas in patients not receiving tamoxifen, the junction is linear. Other ultrasonographic TVUS findings observed in patients taking Tamoxifen as compared to control patients consist of increased uterine volume and depth, greater endometrial thickness, increased incidence of endometrial polyps (36% vs. 10%),

Figure 11–17 A, Saline infusion sonography (sagittal view) of a 64-yearold woman with postmenopausal bleeding. Note the very irregular thickened endometrium. Bridging tissue is noted between the anterior and posterior walls. B, The same lesion from the coronal view. Later biopsy results were consistent with cancer.

and increased endometrial atrophy (28% vs. 87%). A slight increase in the incidence of endometrial cancer of 2 to 3 cases per 1000 women has been noted among tamoxifen users (Fig. 11–17).

Summary When symptomatic focal lesions are detected by hysteroscopy, it is imperative that an operative hysteroscopic resection of the lesion be performed. Epstein and colleagues20 describe the sobering statistics when gynecologists perform blind D&C alone in the presence of focal lesions. Remember that endometrial biopsy and D&C is limited for focal, pedunculated, or regions within the uterine cavity that are difficult to access, such as tubal ostia, fundus, or broad-based lesions. A prospective study involving 105 women with postmenopausal bleeding and endometrium greater than 5 mm on TVUS examination underwent diagnostic

126

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women hysteroscopy, D&C, and hysteroscopic resection of any focally growing lesion still left in the uterine cavity after D&C. Twentyfour women also underwent hysterectomy. If the histologic diagnosis differed between specimens from the same patient, the most relevant diagnosis was considered the final one. In this study,20 80% (84 of 105) of the women had pathology in the uterine cavity, and 98% (82 of 84) of the pathologic lesions manifested a focal growth pattern at hysteroscopy. In 87% of the women with focal lesions in the uterine cavity, the whole or parts of the lesion remained in situ after D&C. D&C missed 58% of polyps (25 of 43), 50% of hyperplasias (5 of 10), 60% of complex atypical hyperplasias (3 of 5), and 11% of endometrial cancers (2 of 19). The agreement between the D&C diagnosis and the final diagnosis was excellent (94%) in women without focally growing lesions at hysteroscopy. It was less than optimal in women with focal lesions. In summary, if there are focal lesions in the uterine cavity, hysteroscopy with endometrial resection is superior to D&C for obtaining a representative endometrial sample in women with postmenopausal bleeding and endometrium greater than 5 mm.20

CLINICAL CONUNDRUMS Cervical Stenosis For the woman who presents with postmenopausal bleeding and who has marked cervical stenosis, evaluation can be challenging. Prior surgical procedures such a laser conization are associated with 0% to 25% risk of cervical stenosis, which increases to 1% to 5% risk after loop excision and 40% after cold-knife cone biopsy.44 Most patients with postmenopausal bleeding do not have endometrial cancer. However, it is imperative that the physician rule out cancer unequivocally in the patient with symptoms. If the patient has had a TVUS and the endometrial echo is greater than 5 mm, then a tissue diagnosis and visualization of the endometrial cavity to determine the presence of a focal or global pathology is of utmost importance. But what can the clinician do in cases of profound cervical stenosis? Traditionally, laminaria tents have been used for cervical dilation. But even the smallest laminaria tent cannot be placed in a markedly stenotic cervix. Oral or vaginal misoprostol is of particular assistance for cervical stenosis. Oral misoprostol 200 to 400 mcg 48 hours and 8 to 12 hours before cervical dilation is associated with increasing cervical softening and enhancing placement of a dilator.45 Patients might have transient cramps, diarrhea, or low-grade fever; however, misoprostol greatly facilitates performance of the procedure. When laminaria tents or misoprostol is not helpful, consider dilation of the cervix under ultrasound guidance. Ultrasound confirms proper placement of the instruments into the uterine cavity. Using transabdominal ultrasound greatly improves the ability to guide an intrauterine catheter or endometrial pipelle biopsy device into the uterine cavity and decreases the likelihood of creating a false tract, perforation, or abandoning the procedure. If this procedure cannot be performed in the office, it can be performed in the operating room under light sedation.46

Alternatively, flexible hysteroscope may be more advantageous when a tortuous uterine cavity is encountered. If a 3-mm flexible hysteroscope can be placed into the cervix, it may be able to be navigated through the endocervix and into the uterine cavity to visualize the endometrium. The use of fluid or CO2 also helps to mechanically dilate the cervix. When marked cervical stenosis is encountered, consider a shallow cone loop electrosurgical excision procedure (LEEP) biopsy to remove the stenotic cervical os. Resorting to this maneuver is very unlikely. Rarely will the gynecologist need to resort to hysterectomy for marked cervical stenosis and postmenopausal bleeding. However, in the presence of an abnormal ultrasound with a nonvisualized endometrium, thickened endometrium, or cervical dysplasia, hysterectomy is a reasonable option for managing cervical stenosis.47 A retrospective study of patients with postmenopausal bleeding and cervical stenosis that precluded further evaluation and was treated by hysterectomy had benign pathology (64%), cervical dysplasia (12%), or uterine cancer (4%).

Hormone Replacement Therapy and Postmenopausal Bleeding Many women use HRT to alleviate vasomotor symptoms, to decrease risk of osteoporosis, and to improve urogenital atrophy. Irregular menstruation was commonly found among 590 women who began cyclic HRT; 38.3% had one or more visits for PMB evaluation, and 12.3% had one or more endometrial biopsies. When combined HRT was used, 41.6% had one or more visits for PMB, and 20.1% had one or more endometrial biopsies.48 In the Women’s Health Initiative study of more than 8000 women randomized to HRT, about 40% of women complained of abnormal bleeding on the combined continuous HRT regimen using conjugated equine estrogen 0.625 mg and medroxyprogesterone acetate 2.5 mg daily.49 Users of estrogen and progesterone HRT have a decreased risk of endometrial cancer than nonusers. Current recommendations include watchful waiting if vaginal bleeding occurs during the first 3 months of HRT. Office hysteroscopy or TVUS can be used to evaluate bleeding if it is persistent. If patients are unwilling to continue therapy because of bleeding, investigation can be initiated earlier. If TVUS only is used and the endometrial echo is less than 5 mm, SIS should be performed next to exclude intracavitary lesions. In the 1960s and 1970s, unopposed estrogen therapy was usually prescribed, but it was later found to increase the risk of endometrial cancer or hyperplasia. We now know that progesterone reduces estrogen-receptor synthesis and increases the conversion of estradiol to the less potent metabolite estrone, and it is a necessary adjunctive therapy for HRT users. Today most women taking HRT use either daily sequential therapy or cyclic therapy. Patients taking cyclic combined estrogen–progesterone therapy should receive a minimum of 12 or 13 days of progestin therapy. Patients who bleed before day 11 of a cyclic progestin regimen should undergo endometrial biopsy. If secretory or pseudodecidual changes are not evident, then additional

127

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women progestin should be used. Patients using cyclic estrogen– progestin therapy and who experience bleeding should be reassured if their bleeding occurs during the final 12 to 14 days of the progestin regimen or the week following progestin withdrawal. If erratic bleeding occurs, evaluation with office hysteroscopy or TVUS is undertaken. Despite hormonal therapies chosen, approximately 5% to 40% of women initiating HRT experience episodic or prolonged bleeding while taking HRT. In the past, watchful waiting was recommended. Needless worry or discontinuation of hormonal therapy by patients was the rule rather than the exception. The baseline prevalence of endometrial polyps is 20% and the baseline prevalence of submucosal fibroids is 10% to 20%, and for some women, HRT can produce bleeding in women who were previously asymptomatic. Earlier intervention with office

SIS

[2D] G42/105dB FA2/P95

hysteroscopy or SIS demonstrating endometrial atrophy or the presence of intracavitary lesions can allay many fears (Fig. 11–18). If endometrial atrophy is detected, the patient can be reassured that bleeding will likely resolve with continued HRT use. Benign polyps and fibroids might continue to bleed, requiring hysteroscopic resection. Likewise, if a premalignant lesion or occult malignancy detected, prompt treatment with progesterone therapy or hysterectomy must be undertaken.

OTHER CAUSES OF ABNORMAL VAGINAL BLEEDING The proximity of the urethra to the vagina may make it difficult for patients to determine the source of vaginal bleeding. Most patients assume that the blood originates from the vagina. When persistent bleeding occurs, despite a thorough gynecologic evaluation, urine cytology and bladder imaging should be considered. Common causes of abnormal vaginal bleeding are listed in Box 11–1. The prevalence of bladder tumors was prospectively evaluated in 280 postmenopausal women who underwent TVUS and ultrasound imaging of the bladder. Abnormal findings were confirmed by urologic consultation and cystoscopy. Results demonstrated 3 of 280 subjects had bladder tumors, of which one was malignant. This serves as a reminder to clinicians to also image the bladder when the TVUS is unremarkable.50

BOX 11–1 Common Causes of Abnormal Vaginal Bleeding Foreign bodies Intrauterine devices Trichomonas Medication

⫹D 1.49 cm ⫻D 0.59 cm

A

SIS

[2D] G42/ FA2/

● ● ●

Hormone replacement therapy Unopposed estrogen therapy Tamoxifen therapy

Cancer ● ● ● ● ●

Cervical cancer Ovarian cancer Fallopian tube cancer Bladder tumors Epithelial ovarian cancer and granulosa cell tumors

Anatomy ● ● ● ●

B

●

Figure 11–18 A, Saline infusion sonography (SIS) (sagittal view) demonstrating a small focal lesion. Operative hysteroscopic myomectomy demonstrated a benign endometrial polyp. B, Saline infusion sonography (SIS) (coronal view) demonstrating small focal lesion. Biopsy was consistent with benign endometrial polyp.

Urethral prolapse Cervical lesions Diverticular uterocolonic fistula Genital prolapse Genital atrophy

Therapy ● ● ●

128

Supertherapeutic levels of anticoagulation Radiation therapy Devascularization and obliterative vaginal endarteritis

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women SUMMARY Women are spending one third or more of their lives during the menopause. Therefore, increasing gynecologic complaints, including bleeding and leukorrhea, may be encountered. The use of cervix-softening agents such as misoprostol can help to make hysteroscopy, SIS, and endometrial biopsies more comfortable, decreases the risk of uterine perforation or cervical lacerations, and facilitates cervical dilation. Hysteroscopy, particularly with small-diameter hysteroscopes, provides excellent imaging of the

endocervix and endometrium. When hysteroscopy is not available, TVUS is helpful in the initial triage of patients with abnormal uterine bleeding. When the endometrium is not well visualized with TVUS or when bleeding continues despite normal endometrial parameters, then hysteroscopy or SIS is advisable. The array of clinical tools available to physicians improves the care of women with postmenopausal bleeding. Using these tools decreases the frustration and allays fears of patients while increasing the confidence and the ability of the physician to make an accurate diagnosis.

REFERENCES 1.

2. 3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

Eitan R, Saenz CC, Venkatraman ES, et al: Pilot study prospectively evaluating the use of the measurement of preoperative sonographic endometrial thickness in postmenopausal patients with endometrial cancer. Menopause 2005;12:27-30. Jemal A, Murray T, Ward E, et al: Cancer statistics, 2005. CA Cancer J Clin 2005;55:10-30. Ozalp S, Tanir HM, Gurer H: Gynecologic problems among elderly women in comparison with women aged between 45-64 years. Eur J Gynaecol Oncol 2006;27(2):179-181. Marchetti M, Vasile C. Chiarelli S: Endometrial cancer :Asymptomatic endometrial findings. Characteristics of postmenopausal endometrial cancer. Eur J Gynaecol Oncol 2005;26(5):479-484. Feldman S, Cook EF, Harlow BL, Berkowitz RS: Predicting endometrial cancer among older women who present with abnormal vaginal bleeding. Gynecol Oncol 1995;56:376-381. Gredmark T, Kvint S, Havel G, et al: Histopathological findings in women with postmenopausal bleeding. Br J Obstet Gynaecol 1995;102: 133-136. Astrup K, Olivarius Nde F. Frequency of spontaneously occurring postmenopausal bleeding in the general population. Acta Obstet Gynecol Scand.2004;83:203-207. Richenberg J, Cooperberg P: Ultrasound of the uterus. In Callen PW (ed): Ultrasonography in Obstetrics and Gynecology, 4th ed. Philadelphia: Saunders, 2000, pp 814-846. Smith RA, Cokkinides V, Eyre HJ: American Cancer Society guidelines for the early detection of cancer, 2005. CA Cancer J Clin 2005;55(1):31-44. Medverd JR, Dubinsky TJ: Cost analysis model: US versus endometrial biopsy in evaluation of peri- and postmenopausal abnormal vaginal bleeding. Radiol 2002;222(3):619-627. Karlsson B, Granberg S, Wikland M, et al: Transvaginal ultrasongraphy of the endometrium in women with postmenopausal bleeding—A Nordic multicenter study. Am J Obstet Gynecol 1995;172:1488-1494. Berliere M, Radikov G, Galant C, et al: Identification of women at high risk of developing endometrial cancer on tamoxifen. Eur J Cancer 2000;36(Suppl 4):S35-S36. Davidson KG, Dubinsky TJ: Ultrasonographic evaluation of the endometrium in post-menopausal vaginal bleeding. Radiol Clin N Am 2003;41:769-780. Okaro E, Bourne T: Sporadic abnormal vaginal bleeding (intermenstrual, postcoital and postmenopausal bleeding). Curr Obstet Gynecol 2002;12:334-340. Goldstein SR, Nachtigall M, Snyder JR, Nachtigall L: Endometrial assessment by vaginal ultrasonography before endometrial sampling in patients with postmenopausal bleeding. Am J Obstet Gynecol 1990;163:119-123. Gull B, Karlsson B, Milsom I, Granberg S: Can ultrasound replace dilation and curettage? A longitudinal evaluation of postmenopausal

17.

18.

19.

20.

21.

22.

23. 24.

25.

26.

27.

28.

29.

129

bleeding and transvaginal sonographic measurement of the endometrium as predictors of endometrial cancer. Am J Obstet Gynecol 2003;188:401-408. Epstein E, Ramirez A, Skoog L, et al: Transvaginal sonography, saline contrast sonohysterography and hysteroscopy for the investigation of women with postmenopausal bleeding and endometrium greater than5 mm. Ultrasound Obstet Gynecol 2001;18:157-162. Twu NF, Chen SS: Five-year follow-up of patients with recurrent postmenopausal bleeding. Zhonghua Yi Xue ZaZhi (Taipei) 2000;63:628-633. Jaaback KS, Hirschowitz L: Primary fallopian tube carcinoma—the experience of a UK cancer centre and a review of the literature. J Obstet Gynaecol 2005;25(7):694-702. Epstein E, Ramirez A Skoog L, et al: Dilation and curettage fails to detect most focal lesions in the uterine cavity in women with postmenopausal bleeding. Acta Obstet Gynecol Scand 2001;80: 1131-1136. Moodley M, Roberts C: Clinical pathway for the evaluation of postmenopausal bleeding with an emphasis on endometrial cancer detection. J Obstet Gynaecol. 2004;24(7):736-741. Clark TJ, Barton PM, Coomarasamy A, et al: Investigating postmenopausal bleeding for endometrial cancer: Cost-effectiveness of initial diagnostic strategies. BJOG 2006;113:502-510. Shushan A, Revel A, Rojansky N: How often are endometrial polyps malignant? Gynecol Obstet Invest 2004;58;212-215. Savelli L, De Iaco P, Santini D, et al: Histopathologic features and risk factors for benignity, hyperplasia, and cancer in endometrial polyps. Am J Obstet Gynecol 2003;188:927-931. Smith-Bindman R, Kerlikowske K, Feldstein VA, et al: Endovaginal ultrasound to exclude endometrial cancer and other endometrial abnormalities. JAMA 1998;280:1510-1517. Tsuda H, Kawabata M, Yamamotot K, et al: Prospective study to compare endometrial cytology and transvaginal ultrasonography for identification of endometrial malignancies. Gynecol Oncol 1997;65: 383-386. Fleischer A, Wheeler JE, Lindsay I, et al: An assessment of the value of ultrasonographic screening for endometrial disease in postmenopausal women without symptoms. Am J Obstet Gynecol 2001:184:740744. Gull B, Carlsson SA, Karlsson B et al: Transvaginal ultrasonography of the endometrium in women with postmenopausal bleeding: Is it always necessary to perform an endometrial biopsy? Am J Obstet Gynecol 2000;182:509-515. Epstein E, Valentin L: Rebleeding and endometrial growth in women with postmenopausal bleeding and endometrial thickness <5 mm managed by dilatation and curettage or ultrasound follow-up: A randomized controlled study. Ultrasound Obstet Gynecol 2001;18:499-504.

Chapter 11

Investigation of Abnormal Uterine Bleeding in Postmenopausal Women 30. De Jong P, Doel F, Falconer A: Outpatient diagnostic hysteroscopy. BJOG 1990;97;299-303. 31. Vilos GA, Edris F, Al-Mubarak A, et al: Hysteroscopic surgery does not adversely affect the long-term prognosis of women with endometrial adenocarcinoma. J Minim Invasive Gynecol 2007;14:205-210. 32. Takeshima N, Nishida H, Tabata T, et al: Positive peritoneal cytology in endometrial cancer: Enhancement of other prognostic indicators. Gynecol Oncol 2001;82:470-473. 33. Widrich T, Bradley LD, Mitchinson A, Collins R: Comparison of saline infusion sonography with office hysteroscopy for the evaluation of the endometrium. Am J Obstet Gynecol 1996;174:1327-1334. 34. Shushan A, Protopapas A, Hart R: Diagnostic and therapeutic advantages of hysteroscopic surgery in management of intrauterine lesions in postmenopausal women. J Am Assoc Gynecol Laparosc 2001:8:87-91. 35. Marchetti M, Litta P, Lanza P, et al: The role of hysteroscopy in early diagnosis of endometrial cancer. Eur J Gynaecol Oncol 2002;23(2):151153. 36. Fay TN, Khanem N, Hosking D: Out-patient hysteroscopy in asymptomatic postmenopausal women. Climacteric 1999;2(4):263267. 37. Kurman RJ, Zaino RJ, Norris HJ: Endometrial carcinoma. In Kurman RJ (ed): Blaustein’s Pathology of the Female Genital Tract, 4th ed. New York: Springer-Verlag, 1994, pp 439-486. 38. Trimble CL, Kauderer J, Zaino R, et al: Concurrent endometrial carcinoma women with a biopsy diagnosis of atypical endometrial hyperplasia: A Gynecologic Oncology Group study. Cancer 2006;106:812-819. 39. Garuti G, Sambruni I, Colonnelli M, Luerti M: Accuracy of hysteroscopy in predicting histopathology of endometrium in 1500 women. J AmAssoc Gynecol Laparosc 2001;8:207-213. 40. Sugimoto O: Hysteroscopic diagnosis of endometrial carcinoma. A report of fifty-three cases examined at the women’s clinic of Hyoto University Hospital. Am J Obstet Gynecol 1975;121:105-113.

41. Guruti G, Mirra M, Luerti M: Hysteroscopic view in atypical endometrial hyperplasias : A correlation with pathologic findings on hysterectomy specimens. J Minim Invasive Gynecol 2005;12:247253. 42. Loffer FD: Hysteroscopy with selective endometrial sampling compared with D&C for abnormal uterine bleeding: The value of a negative hysteroscopic view. Obstet Gynecol 1989;73:1620. 43. Lo KWK, Cheung TH, Yim SF: Preoperative hysteroscopic assessment of cervical invasion by endometrial carcinoma: A retrospective study. Gynecol Oncol 2001:82,279-282. 44. Baldauf JJ, Dreyfus M, Ritter J, et al: Risk of cervical stenosis after large loop excision or laser conization. Obstet Gynecol 1996; 88:933-938. 45. Thomas JA, Leyland N, Durand N, et al: The use of oral misoprostol as a cervical ripening agent in operative hysteroscopy: A doubleblind, placebo-controlled trial. Am J Obstet Gynecol 2002;186:876879. 46. Weiderpass E, Adami HO, Baron JA, et al: Risk of endometrial cancer following estrogen replacement with and without progestins. J Natl Cancer Inst 1999;1131-1137. 47. Newman C, Finan M: Hysterectomy in women with cervical stenosis. J Repro Med 2003;48:672-676. 48. Ettinger B, Li DK, Klein R: Unexpected vaginal bleeding and associated gynecologic care in postmenopausal women using hormone replacement therapy: Comparison of cyclic versus continuous combined schedules. Fertil Steril 1998;69:865-869. 49. Writing Group for the Women’s Health Initiative Investigators: Risks and benefits of estrogen plus progestin in healthy postmenopausal women: Principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321-333. 50. Abdel-Fattah M, Barrington JW, Youssef M, Mac Dermott JP: Prevalence of bladder tumors in women referred with postmenopausal bleeding. Gynecol Oncology 2004;94(1):167-169.

130