Bleeding Problems in Midlife
Brinda N. Kalro and Sarah L. Berga
Table Of Contents
Brinda N. Kalro, MD
Sarah l. Berga, MD
ETIOLOGY OF ABNORMAL UTERINE BLEEDING|
EVALUATION OF THE PATIENT
MEDICAL MANAGEMENT OF ABNORMAL UTERINE BLEEDING
SURGICAL OPTIONS IN THE MANAGEMENT OF ABNORMAL UTERINE BLEEDING
Abnormal uterine bleeding (AUB) is one of the most common presenting complaints encountered in a gynecologist’s office or primary care setting. An estimated 60% of visits to a physician’s office are from women seeking treatment for menstrual problems. Unpredictable and unscheduled bleeding often lead to psychological, medical, and sexual problems requiring pharmacologic and surgical interventions. Hysterectomy is the second most common surgery performed in the United States, with 500,000 to 600,000 hysterectomies performed per annum at a cost of over $ 2 billion.1 Forty percent of women over the age of 55 years have undergone a hysterectomy,2 about 22% having been performed for AUB.3 The intent of this chapter is to review the etiology, evaluation, diagnosis, and medical and surgical management of AUB in women during midlife.
Dysfunctional uterine bleeding (DUB) is defined as AUB with no demonstrable organic genital or extragenital cause, and the diagnosis is often one of exclusion. It is common at extremes of reproductive life when anovulatory cycles and luteal insufficiency are more prevalent.
The causes for AUB are listed in Table 1. AUB can be associated with ovulatory and anovulatory cycles. Bleeding disorders associated with ovulatory cycles include heavy or prolonged bleeding and intermenstrual bleeding, whereas women with the anovulatory cycles often experience irregular, unpredictable, and variable amounts of bleeding. Luteal phase insufficiency may manifest as shorter intervals between cycles. In addition, there are iatrogenic causes, structural lesions, systemic diseases, and endocrinopathies that may lead to AUB.
Gestational trophoblastic disease
Hormone replacement therapy
Hypothyroidism and hyperthyroidism
Functional hypothalamic amenorrhea
Congenital adrenal hyperplasia
Polycystic ovary syndrome
Premature ovarian failure/premature menopause
IUD, intrauterine device; PID, pelvic inflammatory disease.
|ETIOLOGY OF ABNORMAL UTERINE BLEEDING|
A detailed history and examination are often helpful in guiding the physician to the possible causes and in ordering specific and relevant tests.
Menorrhagia, or hypermenorrhea, is defined as excessive menstrual bleeding and is usually associated with ovulatory cycles (Table 2). The prevalence of menorrhagia in the general population is approximately 9 to 11%. About 5% of women between ages 30 and 49 years consult their general practitioners and 12% of referrals to the gynecologist are for menorrhagia.4 For diagnostic purposes, menorrhagia is defined as objective blood loss exceeding 80 mL/cycle5 or menses lasting longer than 7 days, although the term is often used more generally to describe excessive bleeding as perceived by the patient. Quantifying the amount of menstrual loss objectively is impractical and unnecessary.5,6,7 Typically, one has to rely on the patient’s subjective complaint and proceed with evaluation. It has been estimated by spectrophotometric methods that the average loss during a menstrual period is approximately 40 mL. Several laboratory techniques have been described attempting to quantify blood loss, but it is a cumbersome and time-consuming process.8,9,10,11 Blood loss in excess of 80 mL/cycle significantly increases a woman’s risk for iron deficiency anemia.
Dysfunctional uterine bleeding
Intermenstrual bleeding (IMB) is defined as bleeding or spotting between menstrual periods. IMB is commonly associated with the use of oral contraceptive pills (OCPs). This type of breakthrough bleeding (BTB) often occurs during the first few months of use and usually resolves thereafter. Both continuous and cyclic regimens of OCP administration have been associated with BTB. Causes include missed pills, drug interactions with anticonvulsants, rifampicin, and certain antibiotics, but BTB most commonly results from an imbalance between the estrogen and progestin that makes the endometrial lining unstable or fragile. Therefore, changing the formulation to a higher or lower dose of estrogen or a different progestin may be required to resolve the BTB. It is also important to counsel patients to take their pills daily and at approximately the same time everyday so as to not trigger “withdrawal bleeding.” IMB may be a manifestation of endometritis and may also result from structural lesions such as endometrial polyps or submucosal fibroids. Postcoital bleeding usually indicates cervical or vaginal lesions that may be either benign or malignant. Causes include cervical ectopy and polyps, cervicitis, cervical malignancy, vaginal lacerations, postmenopausal atrophic vaginitis, and vaginal neoplasms.
Fibroids, also known as leiomyomas, are a common incidental finding. They may be identified in up to 20% to 30% of women during their reproductive years,12 and their peak incidence is in the early perimenopause at approximately age 44 years.13 They are composed of fibrous connective tissue and smooth muscle and are classified according to their location. They may be within the myometrium (intramural), may be within the endometrial cavity (submucosal), or may protrude into the abdominal cavity (subserosal); a single fibroid often extends into more than one location. Ninety-five percent of fibroids are intramural or subserosal; fewer than 5% are submucosal or cervical in location.14 African-American women have a threefold to ninefold greater incidence of these tumors than do Caucasians.12 In a study evaluating 1.7 million hysterectomies performed for reasons other than fibroids, more than a third of the uteri contained fibroids.14 About a third of all hysterectomies performed in the United States are for symptoms resulting from leiomyomas.15 AUB is the most common symptom associated with fibroids, but their discovery during evaluation may also be incidental. Usually, larger fibroids that cause overall uterine enlargement or submucosal fibroids that encroach on or lie in close proximity to the endometrial cavity are more likely to be associated with excessive or irregular bleeding. Fibroids are hormone-dependent and tend to regress, with or without calcification, during the menopausal years as estrogen levels decline. With increasing use of hormone replacement therapy (HRT), fibroids tend not to regress and can cause AUB patterns in postmenopausal women. Individuals from certain ethnic backgrounds and families exhibit a genetic predisposition to fibroid formation.16
Polyps of the endometrium are a notorious and common cause for irregular bleeding in both perimenopausal and menopausal women. The etiology of polyps is poorly understood, but hormones, particularly unopposed estrogen exposure and tamoxifen treatment, may play a role in their pathogenesis. Interestingly, but not surprisingly, obese women have been reported to have a higher incidence of polyps, perhaps secondary to a relatively hyperestrogenic milieu that results from peripheral aromatization of androgens to estrogens in adipose tissues or from increased insulin action. Polyps in middle-aged women tend to be benign, but they should not be ignored in the presence of symptoms. Two morphologic types of endometrial polyps have been identified in the reproductive age group. The hyperplastic type arises from the basal layer and is dependent on estrogens but not progestogens for stimulation. The second type of polyp consists of a glandular component that responds to cyclic hormonal changes and its morphology resembles that of the surrounding endometrium. Late menopause, owing to longer duration of exposure to estrogens, has been implicated as a risk factor for development of endometrial polyps. The majority of polyps in the postmenopausal years are atrophic with fibrotic transformation of the stroma that may explain incomplete removal by curettage and offer one explanation for their persistence or recurrence in such women.17 It is likely that with increasing use of HRT, the incidence of endometrial polyps will rise. Menorrhagia may occur in approximately 12% of women with endometrial polyps18; the mechanism responsible is unknown. Reslova and colleagues studied a cohort of 245 pre- and postmenopausal women with endometrial polyps and observed AUB in 44% of postmenopausal and 82% of premenopausal women17; 8% were taking tamoxifen as treatment for breast cancer.
Bleeding problems associated with anovulatory cycles are often encountered at the extremes of reproductive life, during the perimenarchial and perimenopausal years. The character, amount, and duration of bleeding are typically unpredictable. This type of bleeding pattern results from a disordered endometrium as a consequence of anovulation and unopposed estrogen effects, and absence of the stabilizing influence of progesterone. Anovulatory bleeding increases the risk of endometrial hyperplasia, particularly in perimenopausal women, owing to the relative increase in ovarian estrogen production and relative decrease in progesterone production characteristic of the aging ovary. Endocrine dysfunction such as polycystic ovary syndrome, hyperprolactinemia, hyperthyroidism and hypothyroidism, uncontrolled diabetes, Cushing’s syndrome, and rarely, androgen-secreting tumors of the ovary and adrenal gland can be associated with anovulatory menstrual cycles. Other causes of anovulation include the use of the progesterone only pill (the “mini-pill”), major tranquilizers, tricyclic antidepressants, and heroin use, although the latter is more often associated with amenorrhea.19 Systemic illnesses such as liver and renal disease may also suppress central rhythms of gonadotropin-releasing hormone (GnRH) secretion and result in disorganized patterns of follicle-stimulating hormone (FSH) or luteinizing hormone (LH) release and ovarian stimulation or adversely affect estrogen metabolism with the end result of luteal phase deficiency or anovulation, although at this stage they are usually fairly advanced. Women with chronic renal failure have low estrogen levels and are often amenorrheic secondary to uremia. Menses often resume following dialysis and improvement of uremia. Vigorous exercise regimens, eating disorders, obesity, and stressful life events or lifestyle with secondary disruption of the hypothalamic-pituitary-ovarian axis can also be associated with anovulation.
Tamoxifen-Induced Endometrial Effects
Tamoxifen, a selective estrogen receptor modulator, is a triphenylethylene derivative that has been used to treat women with breast cancer for over 25 years. It appears to be well tolerated with few adverse effects. However, its prolonged use has been associated with significant uterine abnormalities, both benign and malignant. Tamoxifen was initially thought to possess antiestrogenic properties exclusively, but observations in postmenopausal women have revealed that it exerts estrogen-like effects on genital tract tissues and the endometrium.20,21,22,23 Although the majority of information available derives primarily from experience with tamoxifen use in postmenopausal women, increasing information now relates to the effects of tamoxifen in premenopausal women.
The endometrial effects of tamoxifen are largely proliferative and vary from benign polyps24,25,26,27,28,29 and hyperplasia to frank endometrial malignancy.27,28,29,30,31 These effects appear to be more pronounced in postmenopausal women than in premenopausal women.25,32 Tamoxifen-treated endometrium is difficult to biopsy with poor tissue yield despite ultrasound and hysteroscopic evidence of thickened and abnormal endometrium.33 Microscopic examination reveals tamoxifen-induced diffuse endometrial thickening with dilated cystic glands of endometrial origin.34 Focally, the glandular epithelium displays mitotic activity with mucinous, clear cell, oxyphilic, and other epithelial metaplasias often noted. The endometrial stroma is characteristically fibrotic with occasional stromal mitosis, structural characteristics that might explain the difficulty that may be encountered when attempting to obtain a satisfactory biopsy specimen.34,35,36
Postmenopausal women on tamoxifen are seven times more likely to develop endometrial polyps than those not taking tamoxifen.26,28 Tamoxifen-induced polyps have distinct macroscopic and microscopic features with background endometrial hyperplasia.34,37,38 Polyps may undergo ulceration, infarction, and occasional myxoid change and may be mistaken for malignancy. The prevalence of endometrial cancer in tamoxifen-induced endometrial polyps is significantly higher than in polyps that develop in women not exposed to the drug.34,37,39
It is now well established that tamoxifen use is associated with increased risk of endometrial cancers, including rare invasive varieties such as uterine leiomyosarcoma, endometrial stromal sarcoma, and malignant mixed müllerian tumor.40,41,42,43 The pathologic features of tamoxifen-induced endometrial cancer are not well understood. Whereas some studies have suggested they are most likely to be high-grade endometrioid adenocarcinomas or histologic subtypes associated with a poor prognosis,29,44,45 others have concluded that the majority of tamoxifen-induced endometrial cancers are well differentiated and less aggressive than those that develop in women not exposed to the drug.42,46,47
Perimenopause is the time interval between declining ovarian function and the complete cessation of ovarian function. The age of onset and duration of this transition varies greatly and is unpredictable. As defined by the World Health Organization, the early perimenopause is marked by the onset of a break in regular menstrual cyclicity in a previously normally cycling woman who experiences no more than 3 months of amenorrhea. In the late perimenopause, women often experience amenorrhea for 3 to 11 months. Amenorrhea lasting 12 months or longer is classically defined as menopause. The perimenopausal transition can begin as early as 10 years prior to menopause. Fewer than 10% of women experience abrupt cessation of menses prior to the menopause without prior irregularity of menstrual cycles.48 Menstrual changes appear to be the most common symptom of the perimenopause, with 90% of women experiencing more frequent or less frequent periods, missed periods, or a variation in menstrual flow. Often, in the early perimenopause, there is shortening of the follicular phase with little change in luteal phase duration, resulting in shorter cycles spanning 21 to 24 days. However, ovulation still occurs and pregnancy is therefore possible. Contraception is often overlooked in this age group, with a significant number of unintended pregnancies being the consequence. Once oligo-ovulation sets in, menstrual cycle length typically increases and FSH levels increase progressively. Ovarian aging is associated with declining sensitivity and responsiveness to FSH and LH, with resultant perturbations in estrogen levels and anovulatory cycles. Estrogen levels are typically higher than in ovulatory cycles and exhibit wide fluctuations, with concentrations ranging from as high as 700 pg/mL to as low as 20 pg/mL heralding the waxing and waning ovarian function. Endometrial stimulation from unopposed fluctuating estrogen exposure results in a vascular and friable endometrium that bleeds intermittently and sloughs at unpredictable intervals.
Intrauterine and ectopic pregnancy are important considerations in formulating the differential diagnosis in a perimenopausal woman who presents with an AUB pattern. Idiopathic hyperprolactinemia was frequently observed in one study of premenopausal women presenting with bleeding problems. The role of prolactin per se in inducing AUB in such women is not clear, but it was an interesting observation worthy of note.49 Women in the study were not screened for thyroid disease; hypothyroidism is fairly common in this age group, and elevated thyroid-stimulating hormone (TSH) levels often result in mild elevations in serum prolactin concentrations.
Ovarian follicular depletion and the concomitant decline in ovarian estradiol production results in menopause. In developed nations, the average age of menopause is approximately 51 years. The increasing use of HRT in menopausal women for relief of vasomotor symptoms in the short term, and for potential long-term benefits on bone and the brain, has led to the increased incidence of bleeding problems in this age group. Typically, a cyclic regimen of progestational agents with continuous estrogen replacement induces regular withdrawal bleeding at the end of progestin/progesterone exposure and continuous combined estrogen/progestin treatment regimens result in amenorrhea. Most women prefer not to continue experiencing cyclic bleeding indefinitely, and opt for the latter.
New-onset or persistent bleeding patterns that differ from the expected pattern, on or off HRT warrant evaluation to exclude endometrial pathology. AUB patterns are common in the initial months of HRT use with both cyclic and combined continuous treatment regimens. Prolonged unopposed estrogen use in women with intact uteri significantly increases the risk of endometrial hyperplasia, a precursor of endometrial cancer,50 necessitating regular endometrial surveillance. Inadequate progestogen exposure in cyclic treatment regimens could also increase the risk of hyperplasia. Awareness of bleeding patterns on HRT is therefore critical and avoids both excessive and inadequate evaluation.
Lesions such as endometrial polyps and fibroids, in particular submucosal fibroids, can pose a problem in menopausal women on HRT. Townsend and coworkers51 evaluated a series of postmenopausal women with persistent bleeding problems on or off HRT and found that 90% had either polyps or submucous fibroids.
Thyroid disease is more common in women in their midlife years and 2% to 5% of women exhibit clinical disease. Thyroid antibodies are present in about 5% of women. The American College of Physicians recommends that all women over age 50 years be screened for thyroid disorders every 2 to 5 years. Obtaining a baseline TSH is a good starting point. Subclinical or frank hypothyroidism can present with easy bruising and metrorrhagia or menorrhagia52,53,54 that usually resolves with thyroid replacement therapy within 3 to 6 months.55 Hypothyroidism has protean clinical manifestations, one of which is a von Willebrand–like bleeding disorder.56 Also, hypothyroidism alters steroid metabolism and clearance, making endometrial dysfunction more likely.57 Menorrhagia is considered an early manifestation of hypothyroidism and generally responds to thyroxine replacement.
Endometrial cancer is the most common gynecologic malignancy in women in the United States. Risk factors include obesity, nulliparity, late menopause, tamoxifen use, diabetes (more so with inadequate glycemic control), and unopposed estrogen exposure. The average age for adenocarcinoma appears to be the 6th decade, but about 20% to 25% of endometrial cancers are diagnosed before menopause.58 AUB is often the initial presenting symptom. Endometrial evaluation should therefore be considered, particularly in women with anovulatory menstrual cycles. Women with hyperplasia without cytologic atypia are at low risk for endometrial cancer. In contrast, women with hyperplasia and atypia have a 23% likelihood of developing endometrial cancer in the next decade.59
Inherited Bleeding Diathesis
The hemostatic mechanisms that control menstrual bleeding are complex and involve constriction of endometrial arterioles and/or myometrial arteries and the formation of platelet and fibrin hemostatic plugs within endometrial vessels. Therefore, any process that disrupts either mechanism can result in menorrhagia.62,63,64,65 Other factors such as tissue regeneration, prostaglandins (PGs), inflammation, and myometrial contractility also influence the amount of blood loss. Coagulopathies are generally uncommon, but may occasionally manifest as menorrhagia, usually at menarche.
In 1926, von Willebrand first described a family with a bleeding disorder in which the index patient died during a menstrual period. Von Willebrand’s disease (vWD), a condition characterized by defective platelet adhesion and factor VIII deficiency, is the most common bleeding disorder and affects approximately 1% of the population. Menorrhagia dating from menarche predicts an inherited bleeding disorder in 65% of women with vWD and in 67% of women with factor XI deficiency.4 In a retrospective survey, a pictorial bleeding assessment chart revealed menorrhagia in 74% of patients with vWD, in 57% of hemophilia A or B carriers, and in 57% of patients with factor XI deficiency, compared with 29% of controls.4 As the concentration of von Willebrand factor (vWF) tends to increase with age, the severity of bleeding disorders decreases as does the prevalence of menorrhagia.66 Menorrhagia can be a valuable predictor of bleeding disorders.67
A family history of excessive menstrual blood loss or other bleeding disorders is often helpful and prompts evaluation. However, milder forms can often go undiagnosed. Apart from the medical implications, menorrhagia can pose a social problem for women in today’s society and can adversely affect their quality of life. Other hematologic causes for AUB include thrombocytopenia, idiopathic or secondary to leukemia, and aplastic anemia.
|EVALUATION OF THE PATIENT|
Important considerations when eliciting a history include age at onset, nature, and duration of AUB pattern, gynecologic and obstetric history, sexual and contraceptive history, use of medications, symptoms suggestive of endocrinopathies, chronic medical illness, coagulopathies, excessive exercise and stress patterns, and eating disorders. A careful family history is also important.
General examination for physical evidence of systemic diseases, endocrine disorders, signs of hirsutism or virilization, galactorrhea, and coagulopathies is vital. A pelvic examination is essential to exclude the possibility of infections and to detect an obvious pelvic lesion.
The diagnostic evaluation of AUB in premenopausal women can be more challenging than in postmenopausal women. Identifying the cause for AUB is essential in planning appropriate treatment. Having excluded bleeding disorders and pregnancy, endocrine abnormalities account for most cases. Other causes include endometritis, endometrial polyps, and submucous fibroids. Evaluation is guided by patient’s age and history. Gonadotropin levels (FSH, LH), sex steroid hormone levels (estrogen, progesterone, and androgens), thyroid function tests, a prolactin level, complete blood count, platelet count, coagulation profile, and pregnancy test are useful initial tests.
Endometrial monitoring can be achieved by various means with varying degrees of invasiveness and risk. Listed in the following sections are methods in current use, including their advantages, risks, and limitations.
Transvaginal ultrasonography (TVUS) is a noninvasive and relatively inexpensive method for direct assessment of endometrial thickness and certain anatomic features of the pelvic organs. It does not require a distended bladder, is well tolerated by patients, and has a diagnostic sensitivity of greater than 90% in experienced hands.68,69 Despite a smaller field of view when compared with transabdominal ultrasound examination, the higher resolution of TVUS permits better tissue characterization.70,71 The proximity of the vaginal transducer to the pelvic organs allows good definition of endometrial thickness and architecture. The endometrium may be homogeneous or inhomogeneous with localized cystic or other features. An obese body habitus, low level of operator skill, and low signal-to-noise ratio pose limitations on the usefulness of this method. Several studies have documented the usefulness of TVUS as a diagnostic tool prior to proceeding with more invasive procedures such as endometrial biopsy (EMB) and hysteroscopy.72 In postmenopausal women, endometrial thickness and morphology are useful parameters for identifying those who require endometrial tissue sampling.69 As many as 30% to 40% of dilation and curettage (D&C) procedures performed for evaluation of the endometrium can be avoided by first evaluating the endometrium with TVUS.73,74
TVUS is slightly less useful in premenopausal women because endometrial thickness and architecture vary widely across the menstrual cycle. These parameters are therefore less reliable indices of pathology, and interpretation is more difficult for the less skilled or experienced operator. In one retrospective study of premenopausal women who underwent an EMB, D&C, or hysterectomy for DUB, ultrasound findings of a thickened endometrium correlated well with pathologic findings of polyps and submucous fibroids.75
In postmenopausal women, an endometrial thickness of 4 to 5 mm or less excludes significant endometrial pathology with a positive predictive value of 87.3% and sensitivity of 90%.76,77,78,79 Three-dimensional ultrasonography with Doppler velocimetry appears to be more effective for differentiating certain types of fallopian tube carcinomas from adnexal lesions.80,81,82,83
Introduction of sterile saline solution into the uterine cavity and simultaneous TVUS can clearly identify the size and location of any intracavitary lesion and has proved to be an extremely useful diagnostic technique.84 Sonohysterography (SHG) is being applied in the evaluation of AUB in women with increasing frequency. TVUS can detect a thickened endometrial lining but has limitations in that it cannot, by itself, exclude the presence of an intracavitary lesion such as a polyp or a fibroid within the thickened or asymmetric endometrial “stripe.” A study of asymptomatic postmenopausal women seeking HRT revealed a high incidence of intracavitary pathology as detected by SHG. Intracavitary lesions were more prevalent (63.6%) in those with a thick endometrial stripe on TVUS, but were also noted in 36.8% of those with a normal endometrial thickness.85 The authors concluded that a 5-mm or smaller endometrial stripe on TVUS was useful for excluding endometrial hyperplasia but could not exclude an endometrial polyp.
SHG has proved to be as effective as hysteroscopy and superior to HSG and TVUS for diagnosis of intrauterine anatomic lesions (see later). In one series, saline solution SHG avoided as many as 50% of hysteroscopies performed.86 In women with endometrial cancer, intraperitoneal spread of endometrial cancer cells following SHG or hysterosalpingography (HSG) is a concern but is debatable. In one series, survival rates were similar in patients with endometrial carcinoma who had intraperitoneal spill from a preoperative HSG to those in women who did not.87 The timing of SHG in relation to menstrual flow does not appear to be as important in postmenopausal women as it is in premenopausal women. In menstruating women with predictable cyclicity, mid and late luteal phase SHG has a higher false-positive rate (27%) for detection of structural endometrial lesions than does SHG in the early and mid follicular phase (0%).88 Premenopausal women with AUB have a higher incidence of polyps (33% versus 10%), intracavitary myomas (21% versus 1%), and intramural myomas (58% versus 13%) when compared with asymptomatic age-matched women.89
HSG is another technique used to evaluate the uterine cavity, but it is now used less frequently for this purpose. Although it remains the gold standard for evaluating fallopian tube patency, other methods have gained popularity over HSG for the evaluation of the uterine cavity. Studies comparing HSG with SHG and hysteroscopy have documented that the latter two techniques yield more information than does HSG.90,91,92
Although the technique of visualizing the uterine cavity was introduced over a century ago, hysteroscopy has gained popularity only over the last 2 decades. The first hysteroscopy was performed in 1869, when Pantaleoni attempted to visualize the uterine cavity with a lead pipe and candle.93 Improvement in optics, video monitoring systems, distention media, and smaller telescope size have made hysteroscopy a safe, popular, and acceptable method of directly visualizing and assessing the uterine cavity. AUB appears to be the most common indication for a diagnostic hysteroscopy.94 However, only 28% of gynecologists perform office hysteroscopies.95 Most diagnostic hysteroscopies are performed with the patient under anesthesia, increasing the patient’s risk from anesthesia and at an increased cost. Widrich and associates performed outpatient office hysteroscopies on pre- and postmenopausal women without anesthesia; their findings correlated well with those from SHG performed immediately before, and procedures were generally well tolerated.96
Hysteroscopy is best performed immediately after a menstrual period in the early proliferative phase for optimal visualization of the endometrial surface and any lesions that may be present. Light anesthesia is required and paracervical block with a local anesthetic and vasopressin provides both vasoconstriction during the procedure and postoperative pain relief. Complications are rare in experienced hands, but include infection, bleeding, and risk of uterine perforation. Advantages include direct visualization of the endometrium and endocervix, detection of minute focal lesions, and the ability to perform directed biopsies. In women with AUB, Loffer showed that hysteroscopy alone had better diagnostic accuracy than a routine D&C and that, when the two were combined, diagnostic specificity was 100%.97 The majority of lesions that escape detection by a routine D&C are endometrial polyps and submucous fibroids.98
Even in experienced hands, false-positive identification of endometrial pathology during hysteroscopy can be as high as 20%. An EMB at the time of a hysteroscopy increases the diagnostic accuracy to 100% for diagnosis of endometrial neoplasia and its precursors.99 Although hysteroscopy appears superior to HSG, TVUS, and endometrial sampling for diagnosis of uterine intracavitary lesions,100 it is invasive, more expensive, usually requires anesthesia, and carries risks. Hysteroscopy should therefore not be the initial choice for evaluation, but should be reserved for those patients whose initial workup does not establish a diagnosis and those who require further evaluation with D&C.
In the early 1900s, EMB became an accepted gynecologic office procedure and was performed in certain instances without anesthesia. It is now a safe, quick, convenient, well-tolerated, and inexpensive diagnostic office procedure often employed to evaluate the endometrium in women with AUB. Relative contraindications to this procedure include endometritis, cervicitis, profuse bleeding, and coagulopathy. Various sampling devices have been described and compared over the years. The Pipelle aspiration canula is a safe, rapid, and inexpensive method of obtaining endometrial tissue for evaluation.101 Several studies have compared the efficacy of the Pipelle sampler with the Novak biopsy instrument, Vabra aspirator, and D&C.102,103,104,105 An adequate endometrial tissue specimen was obtained with the Pipelle in 85-95% of premenopausal women and approximately 80% of postmenopausal women. Results from sampling with the Novak and Vabra devices were comparable. Nondirected EMB alone is not adequate for determining the cause of AUB and can fail to detect polyps and fibroids.106,107 Although still useful, EMB is further limited by an inability to identify abnormalities outside the endometrial cavity (adenomyosis, intramural fibroids).108
Among women presenting with AUB, 30% to 50% will have benign anatomic pathology such as polyps, myomas, endometrial hyperplasia, or adenomyosis. Less than 1% of women under the age of 50 years, and 10% to 15% of those over age 50 years will have a malignancy.109,110 In approximately 50% of women, bleeding is dysfunctional in nature, and surgical intervention in all women with AUB is therefore unnecessary.
Dilation and curettage
Operative D&C was first introduced in 1843 by Recamier in the preantiseptic era and was fraught with complications, including death.111 At one time, D&C was the most common diagnostic and therapeutic surgical procedure performed in the world. Today, after several decades of experience, D&C is a relatively safe procedure. Complications include perforation, infection, postoperative intrauterine adhesions, tubal damage, and infertility, the latter two being more relevant for younger women. D&C is several times more costly than office EMB and is associated with a higher risk of complications.112 D&C should therefore not be routine practice. Rather, it should be reserved for specific circumstances such as cervical stenosis in which it may be the preferred method of sampling the endometrium. The amount of endometrial surface sampled by D&C is less than 15%, the prevalence of cervical stenosis that precludes office EMB is approximately 4-10%,113 and D&C requires anesthesia.
|MEDICAL MANAGEMENT OF ABNORMAL UTERINE BLEEDING|
More than 50% of the estimated 650,000 hysterectomies performed annually in the United States are for AUB, at a cost in excess of $1.5 billion. Nearly half of the uteri removed are without disease, indicating the need for less invasive therapies. It is estimated that four of every five women treated for ovulatory menorrhagia will have no anatomic lesion, and over a third of hysterectomies performed for this indication reveal no pathology.116 A thorough understanding of the physiology of the normal menstrual cycle and of the etiology and pathogenesis of AUB during the various stages of reproductive life and beyond is essential in formulating a management plan and selecting the most appropriate treatment strategy.
Management options include medical and surgical measures. Anatomic lesions, including polyps and submucous fibroids, usually require surgical intervention such as hysteroscopy with or without D&C. Other structural anomalies such as adenomyosis and intramural leiomyomas may respond to medical interventions. Menorrhagia, anovulatory bleeding, and perimenopausal bleeding disorders can be treated medically before resorting to more invasive surgical procedures.
The endometrium produces prostaglandins throughout the menstrual cycle but more so in the late luteal and menstrual phases when systemic and local progesterone levels are declining. Phospholipases convert fatty acid precursors to arachidonic acid, which is subsequently converted to PGs by cyclooxygenases.117 Endometrial phospholipases are abundant, particularly in the luteal phase when PGF2α, a potent vasoconstrictor, predominates in the glandular epithelium.118,119 PGE2 and PGI2 (prostacyclin) have vasodilating properties and are found in the stromal cells120 and superficial endometrium,121,122 respectively. Greater amounts of PGI2123 and higher ratios of PGE2:PGF2αhave been observed in women with menorrhagia than in women with normal volumes of menstrual blood loss.122,124
The role of PGs in mediating menstrual blood loss provides the rationale for treatment of menorrhagia associated with ovulatory cycles with cyclooxygenase inhibitors such as nonsteroidal anti-inflammatory drugs (NSAIDs). In the 1970s and early 1980s, mefenamic acid and flufenamic acid were observed to be effective in decreasing menstrual blood loss in ovulatory menorrhagia. Since then, other NSAIDs, including diclofenac, flurbiprofen, ibuprofen, indomethacin, naproxen, and naproxen sodium have been used for the same purpose. The Cochrane Review observed that when compared with placebo, mean blood loss was decreased by treatment with NSAIDs in five of the seven randomized controlled trials (RCTs) examined.125 Subjective perceptions of reduced menstrual blood loss have also been documented in placebo-controlled trials.126,127 All NSAIDs studied to date appear equally efficacious, and there is no evidence to indicate that any one is superior to others. The typical treatment regimen involves starting medication on the first day of flow and continuing for approximately 5 days or until cessation of menses. Doses of mefenamic acid and naproxen prescribed are 250 – 500 mg two to four times per day; ibuprofen can be used in doses up to 600 to 1200 mg/day.
Fibrinolysis is an important physiologic process by which a vascular clot is broken down. This process is mediated by activation of plasminogen by a group of enzymes called plasminogen activators (PAs) to yield plasmin. Pathologic processes that increase local fibrinolytic activity may increase menstrual loss. Women with ovulatory menorrhagia have higher levels of PAs.128,129 PA inhibitors such as tranexamic acid have been shown to decrease menstrual blood loss by approximately 60% in women with increased endometrial PA activity.130,131 Interestingly, women taking tranexamic acid in one study experienced a 45% decrease in menstrual flow, whereas those taking norethisterone during the luteal phase experienced a 20% increase in menstrual loss.132 Tranexamic acid also appears to be more effective than NSAIDs in this respect.133,134
In North America, antifibrinolytics are rarely used owing to concerns about their adverse effects–including gastrointestinal side effects and possible increased risk of thromboembolism–and poor marketing by pharmaceutical companies for lack of a patent.135
High-dose parenteral estrogens have been used effectively to treat acute episodes of excessive uterine bleeding.136,137 Although estrogen does promote rapid endometrial regrowth over denuded epithelial surfaces, the exact mechanism responsible for the effectiveness of parenteral estrogen is not clearly understood. When such treatment is ineffective, D&C may be required to control bleeding. Parenteral estrogens have no place in the long-term management of bleeding disorders. When there are contraindications to estrogen treatment (smoking, history of thromboembolism), high-dose progestational agents can be used,138 although there are no RCTs to support their effectiveness.
Progestins have been used for decades in the treatment of DUB. However, the agent, route of administration, and dosing schedule have varied widely. In the United States, medroxyprogesterone acetate (MPA) is popular, whereas norethindrone is more widely used in other countries. Most of the world’s literature addresses the use of norethindrone. Not surprisingly, progestins appear to be more effective for the treatment of DUB than for AUB associated with anatomic lesions or menorrhagia. When compared with placebo, danazol, NSAIDs, the progesterone-containing intrauterine device (IUD), and antifibrinolytics, the effects of oral progestins range from none to an increase in menstrual loss in women with menorrhagia.125,132,139,140
Continuous systemic dosing may be more effective for women with ovulatory menorrhagia than cyclic progestin therapy, but to be most effective, the dose must be sufficient to suppress ovarian cyclicity. Partial ovarian suppression may desynchronize endometrial development and result in more rather than less bleeding. Studies of depo-MPA used for contraceptive indications have observed that approximately 80% of treated women are amenorrheic by the end of the first year, but it is important to remember that such studies were not designed to evaluate the efficacy of depo-MPA in women with excess menstrual loss.141 Incidentally, 50% of women using depo-MPA for contraception experienced irregular bleeding during the first year. Prolonged irregular bleeding in Norplant users accounts for the high rate of discontinuation of therapy. Runic and colleagues observed fragile capillaries and increased tissue factor expression at sites of endometrial bleeding in hysteroscopically directed EMB specimens in women using Norplant, when compared with specimens obtained from nonbleeding sites.142
PROGESTIN INTRAUTERINE CONTRACEPTIVE DEVICE.
IUDs have been used for decades for contraception. The concept of using an IUD impregnated with a progestational agent to deliver continuous local progestin therapy is not a novel one and was first studied nearly 2 decades ago.143 However, only recently has this approach gained popularity in the treatment of AUB. Menstrual blood loss 3 months after insertion of a progestin IUD was reduced by 94% from baseline, compared with an 87% reduction observed in women treated with norethindrone.144 Despite comparative decreases in blood loss, women who received the IUD more often chose to continue treatment. The progestin IUD rivals surgical endometrial resection in its efficacy for reducing blood loss and achieving amenorrhea.145 In a group of women awaiting hysterectomy for DUB, 41 of 50 treated with the progestin IUD elected not to proceed with surgery.146 Another study from Sweden reported similar results.147
Administering progestational agents via a local route to the endometrium is logical and offers advantages to the patient because systemic progestins may have undesirable effects on lipids and the central nervous system, the latter being the most frequent reason for poor compliance. Although an IUD insertion is somewhat invasive, the long-term benefits, efficacy and reduced need for hysterectomy, appear to justify its use. Interestingly, the cost of an IUD that is effective for 5 years is comparable with that of oral progestin treatment for the same length of time. Moreover, progestin IUDs are associated with low risks of ectopic pregnancy and pelvic inflammatory disease,148,149 and therefore offer further advantages to ovulatory women.
In the United States, the only currently available progestin IUD contains levonorgestrel, is marketed under the name Mirena, and is effective for 5 years. The plastic T-shaped frame of Mirena contains 52 mg of levonorgestrel impregnated in its vertical stem of polydimethylsiloxane, of which approximately 20 μg is released each day. The drug is distributed in silicone fluid with a rate-limiting membrane that allows slow diffusion of the drug into the endometrium. The device is also impregnated with barium sulfate that renders it radiopaque.150 Plasma levels of levonorgestrel with the system in place are negligible when compared with those in women receiving treatment with a progestin-only pill (POP), Norplant, or a combined OCP.151,152,153,154 Systemic progestational side effects are rare, and there is no demonstrable effect on the hypothalamic-pituitary-ovarian axis.155,156
COMBINED ESTROGEN AND PROGESTIN PREPARATIONS.
OCPs are commonly used to treat dysmenorrhea and menorrhagia, and significantly reduce blood loss comparable with that of NSAIDs and danazol.157 However, OCPs are less useful for the treatment of structural lesions such as polyps and submucosal fibroids.
Danazol, a synthetic derivative of testosterone, has been used to treat DUB effectively, the proposed mechanisms of action being suppression of ovulation, decreased ovarian production of estradiol, and direct effects on estrogen receptors in the endometrium. Chimbara and colleagues demonstrated that higher doses were more effective than lower doses in treating ovulatory menorrhagia.158 However, the androgenic and often persistent side effects, including oily skin, acne, weight gain, deepening of voice, have limited its use.
GONADOTROPIN-RELEASING HORMONE AGONISTS.
GnRH agonists down-regulate and desensitize the anterior pituitary, thereby reducing gonadotropin synthesis and release and inducing a reversible hypogonadal state. Agonists have been used effectively for treatment of bleeding disorders, particularly those associated with leiomyomas and adenomyosis.159,160 Both total uterine volume and myoma volume decrease; the extent of shrinkage is significant but highly variable. Within months after cessation of therapy, uterine and myoma volumes return to pretreatment levels.161 Consequently, GnRH agonists are used primarily as a preoperative adjunct to control bleeding and reduce the size of myomas and the uterus, facilitate subsequent surgery, and reduce the need for blood transfusion.
The antiprogestational class of agents, although used primarily as abortifacients, has also been shown to significantly reduce fibroid volume without affecting circulating gonadotropin or gonadal steroid levels,162 achieving amenorrhea in all women. Mifepristone (RU-486) has been available in Europe for many years, but has been approved for use as an early first trimester abortifacient in the United States only recently. Its use for the treatment of menorrhagia remains largely unexplored.
GONADOTROPIN-RELEASING HORMONE ANTAGONISTS.
GnRH antagonist agents are used primarily in patients undergoing ovulation induction, but they may have a potential role in the treatment of bleeding disorders in the future. Currently, GnRH antagonists are considerably more expensive than GnRH agonists and are not available in depot form, making them impractical for use in the management of bleeding disorders.
Cyclic progestin administration in perimenopausal women with anovulatory cycles can be extremely effective.163,164 Such treatment also effectively prevents the endometrial consequences of otherwise unopposed estrogen stimulation. The progestin-only pill (POP) may offer an alternative approach to providing both contraception and endometrial protection, but continuous progestin treatment can induce BTB by promoting overexpression of tissue factors and increased vascular fragility in the endometrium.142 There is also concern that continuous systemic progestin treatment may be detrimental to maintenance of bone mass in the long term. In women who present with complaints of mood changes or depression during the perimenopausal transition or who have a prior history of depression or severe premenstrual syndrome (PMS), it is important to exercise caution when prescribing hormonal treatment because they may not tolerate progestin therapy or progestin-dominant OCPs. Progestins are also likely to aggravate insulin resistance in susceptible individuals.165
Oral micronized progesterone appears to be better tolerated than synthetic progestins, with fewer mood and PMS-like symptoms. The most common reported side effects are fatigue, sedation, and dizziness. Micronization decreases particle size, enhances dissolution, prolongs the half-life of progesterone, and decreases its breakdown in the gastrointestinal tract. Progesterone can be administered orally, topically, intravaginally, or intramuscularly, the former three being the most popular routes for HRT.
COMBINED ORAL CONTRACEPTIVES.
Combined oral contraceptives (COCs) suppress ovarian function and thereby provide contraception. Ovarian suppression is useful in other contexts as well. COC use is advocated in perimenopausal women to eliminate the highly dynamic and fluctuating hormone levels that characterize the perimenopausal years. COCs may be used to avoid vasomotor symptoms associated with transient declines in circulating estrogen levels that typically occur in the perimenopause or to temper worsening mood changes. In addition to cycle regulation, they also provide safe and effective contraception for women who are at risk of spontaneous ovulation and pregnancy. Pregnancy is often undesirable in this age group and can be fraught with complications, including an increased risk of chromosomal abnormalities and spontaneous abortion. Other benefits of COCs include conservation of bone mass, decrease in benign breast disease, reduced risk for ovarian and endometrial cancer, and reduced need for surgical sterilization procedures. COCs also reduce the need for invasive surgical procedures such as hysteroscopies, D&Cs, endometrial ablations, and hysterectomies for persistent AUB due to hormone dysregulation. It is important to exercise caution when using OCs in smokers, hypertensive women, and those who may harbor an unrecognized thrombophilia. Women on COCs may experience BTB and may require an alternate formulation. It is interesting to note that the risk of BTB during treatment with COCs is twofold higher in smokers than in nonsmokers.166
The pill is not a panacea for all women. It may not be easy to convince a woman who has previously experienced side effects on the pill to try another formulation. In older women, prudence dictates the use of preparations containing 20 μg or less ethinyl estradiol to avoid unwanted side effects and risks such as venous thromboembolism. Standard doses of estrogens and progestogens used in conventional HRT are relatively ineffective in suppressing the hypothalamic-pituitary-ovarian axis and may aggravate the underlying hormone chaos.
COCs can be used in cyclic or continuous fashion.167 Some women feel more secure having regular monthly cycles. Others prefer having less frequent menses or none at all. It is therefore important to educate and counsel patients carefully to allow them to select their preferred regimen. Women in this age group generally are very receptive to this approach and find this concept attractive. The two regimens are interchangeable if one is not well tolerated by the patient. The continuous combined regimen is often associated with some BTB in the initial months of use. Abdominal bloating, breast tenderness, weight gain, and loss of libido can be associated with COC use and may hinder compliance.
COCs and progestins have a very limited role in treating women with AUB that is associated with structural abnormalities, such as polyps and submucous fibroids, although there is no evidence that their use makes matters worse.
Unopposed estrogen effects on the endometrium include endometrial hyperplasia that may be simple or complex in histologic pattern. Complex hyperplasia may be observed with or without associated cytologic atypia. The presence of atypia increases the risk for endometrial cancer. Risk for developing endometrial carcinoma is approximately 25% in patients with atypia, but only 2% in patients without atypia.168 Hyperplasia without atypia can be treated with higher doses of progestins with a repeat EMB 4 to 6 months after initiation of treatment to ensure that the abnormality has resolved. Hyperplasia with atypia is best treated surgically. Persistence of atypia has been demonstrated in as many as 50% of women despite treatment with high doses of progestational agents.169
|SURGICAL OPTIONS IN THE MANAGEMENT OF ABNORMAL UTERINE BLEEDING|
Hysteroscopy is a useful diagnostic and therapeutic tool for patients with and without intracavitary lesions. Diagnostic hysteroscopy can be performed as an office procedure without anesthesia. Most gynecologists, however, prefer to perform this procedure with the patient under anesthesia, particularly if they plan to biopsy or excise polyps/fibroids.
Polyps are usually pedunculated, vary in size, and can be excised by twisting and avulsion with a grasping forceps. Sessile polyps are best excised by D&C or monopolar loop excision under direct hysteroscopic visualization.
Hysteroscopic treatment of fibroids depends on the size, number, nature, and location of lesions. Small pedunculated submucosal fibroids can be excised by techniques similar to those described for polyps. Larger pedunculated fibroids or fibroids having an intramural component that does not extend deeply into the myometrium require resection with a loop electrode. For fibroids with a deeper intramural component, laparoscopic guidance during hysteroscopic resection is advisable.
In women with unexplained menorrhagia and normal endometrial histology, an endometrial ablation can be considered using electrosurgical or laser energy systems. With improvement in fiberoptic and video monitoring techniques, the safety and efficacy of distention media, operative equipment, and energy systems, this procedure is relatively safe in experienced hands. Transcervical endometrial resection was first described in 1981 by Goldrath who used the neodymium:yttrium-aluminum-garnet (Nd:YAG) laser energy system. Laser energy systems vary in their depth penetration, extent of endometrial destruction, beam scatter, and absorption by distention media. Amenorrhea and complication rates vary widely and are largely operator and technique dependent. Initial reports were encouraging, with over 90% initial eumenorrhea or amenorrhea rates,170,171 but longer follow-up studies revealed increasing failure rates requiring additional surgery.172 Failure rates vary between 5% and 10%. Contraindications to endometrial ablation include endometrial pathology such as hyperplasia and malignancy, a grossly enlarged uterus, multiple fibroids, and the desire to preserve fertility. Several cases of endometrial adenocarcinoma have been reported in women who have previously undergone endometrial ablation.173 Therefore, any subsequent HRT should include progestogens to prevent endometrial hyperplasia in residual islands of functional endometrium, even in amenorrheic women. Complications of the procedure include cervical trauma during dilation, particularly with a stenotic cervix, uterine perforation with or without trauma to other intra-abdominal organs, excessive bleeding, infection, air embolism, fluid and electrolyte disturbances, and those that may be associated with laser and electrosurgical instruments or anesthesia. Late complications include hematometra, recurrence of bleeding problems, retrograde menstruation, and pelvic pain.
Laparoscopic or conventional myomectomy by laparotomy is an option for women with large fibroids causing menorrhagia. The size and location of myomas often dictates the surgical approach. Myomectomies are more frequently performed in women of reproductive age who wish to conserve their fertility and in those who prefer not to undergo a hysterectomy for cultural or other reasons. Postoperative recurrence and adhesions can pose subsequent problems.
Uterine artery embolization of symptomatic myomas can avoid the need for a surgical procedure. However, this procedure requires a highly skilled and experienced interventional radiologist and is not widely available.
Hysterectomy is the second most commonly performed surgical procedure in the United States. Approximately half a million hysterectomies are performed annually, at a cost of approximately $5 billion. Between 1988 and 1990, the overall hysterectomy rate was 56.8 per 10,000 women.174 Women in the United States are twice as likely as British women and four times more likely than Swedish women to undergo a hysterectomy. Women who fail to respond to conservative measures clearly need and will benefit from a hysterectomy. Satisfaction rates following surgery, immediate and long term, vary widely and largely depend on the indication for surgery.
Hysterectomies can be performed via the vaginal, abdominal, or laparoscopic route. Appropriate patient selection is essential to reduce the risk of complications. In women who undergo hysterectomy for benign indications, the overall mortality rate is approximately 6 per 10,000, and is 3 per 10,000 in those under age 45 years. In one series, the perioperative death rate was 0.1% for abdominal hysterectomy and 0.2% for vaginal hysterectomy.175 Complications include wound or urinary tract infection, injury to intra-abdominal organs (bladder, ureter, intestines, rectum), and hemorrhage. Laparoscopically assisted vaginal hysterectomy and total laparoscopic hysterectomy are gaining popularity and involve a shorter duration of hospital stay. Cost differences between laparoscopic and traditional approaches can be an issue and vary widely between hospitals and providers.
AUB is a frequently encountered presenting complaint in perimenopausal and postmenopausal women and accounts for a substantial proportion of office visits and gynecologic procedures. The etiology of AUB is varied. Establishing a diagnosis often takes time and requires specific procedures to exclude neoplasms and frank malignancies. Treatment options are many, and choice depends on patient preference and the underlying cause(s). Counseling and education on the various medical and surgical options available, and their side effects, is essential to ensure compliance and successful treatment. Given the many effective medical therapies available, surgery should be reserved for those who cannot tolerate or fail to respond to medical therapy and those whose pathology is not amenable to medical treatment.
1. Bernstein SJ, McGlynn EA, Siu AL, et al: The appropriateness of hysterectomy. A comparison of cases in seven health plans. Health Maintenance Organization Quality of Care Consortium JAMA 269:2398, 1993
16. Thurmond AS, Jones MK, Cohen DJ: Leiomyoma-related bleeding: A classic hypothesis updated for the molecular era. Gynecological, Obstetrics and Breast Radiology—A Text/Atlas of Imaging in Women. Cambridge, MA, Blackwell Science Inc, 1996 http://www.fibroids.net/html/frameset.html
25. Cheng WF, Lin HH, Torng PL, et al: Comparison of endometrial changes among symptomatic tamoxifen-treated and nontreated premenopausal and postmenopausal breast cancer patients. Gynecol Oncol 66:233, 1997
41. Fisher B, Costantino JP, Redmond CK, et al: Endometrial cancer in tamoxifen-treated breast cancer patients: Findings from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14. J Natl Cancer Inst 86:527, 1994
45. Evans MJ, Langlois NEI, Kitchener HC, et al: Is there an association between long-term tamoxifen and the development of carcinosarcoma (malignant mixed müllerian tumor) of the uterus. Int J Gynecol Cancer 5:310, 1995
46. Barakat RR, Wong G, Curtin JP, et al: Tamoxifen use in breast cancer patients who subsequently develop corpus cancer is not associated with a higher incidence of adverse histologic features. Gynecol Oncol 55:164, 1994
72. Emanuel MH, Verdel MJ, Wamsteker K, et al: A prospective comparison of transvaginal ultrasonography and diagnostic hysteroscopy in the evaluation of patients with abnormal uterine bleeding: clinical implications. Am J Obstet Gynecol 172:547, 1995
74. Taipale P, Tarjanne H, Heinonen U-M: The diagnostic value of transvaginal sonography in the diagnosis of endometrial malignancy in women with peri- and postmenopausal bleeding. Acta Obstet Gynecol Scand 73:819, 1994
75. Dubinsky T, Ab-Gazzeh Y, Stroehlein K: Role of transvaginal sonography and endometrial biopsy in the evaluation of dysfunctional uterine bleeding in premenopausal women. J Clin Ultrasound 26:180, 1998
77. Goldstein SR, Nachtigall M, Synder JR, et al: Endometrial assessment by vaginal ultrasonography before endometrial sampling in patients with postmenopausal bleeding. Am J Obstet Gynecol 163:119, 1990
86. Bronz L, Suter T, Rusca T: The value of transvaginal sonography with and without saline instillation in the diagnosis of uterine pathology in pre- and postmenopausal women with abnormal bleeding or suspect sonographic findings. Ultrasound Obstet Gynecol 9:53, 1997
105. Rodriguez GC, Yaqub N, King ME: A comparison of the Pipelle device and the Vabra aspirator as measured by endometrial denudation in hysterectomy specimens: The Pipelle device samples significantly less of the endometrial surface than the Vabra aspirator. Am J Obstet Gynecol 168:55, 1993
111. Ricci JV: Gynaecologic surgery and instruments of the nineteenth century prior to the antiseptic age. The Development of Gynaecological Surgery and Instruments. pp 326, 328 Philadelphia, Blakiston, 1949
123. Makarainen L, Ylikorala O: Primary and myoma associated menorrhagia: Role of prostaglandins and effects of ibuprofen. Br J Obstet Gynaecol 93:974, 1986
130. Gleeson NC, Buggy F, Sheppard BL, et al: The effect of tranexamic acid on measured menstrual loss and endometrial fibrinolytic enzymes in dysfunctional uterine bleeding. Acta Obstet Gynecol Scand 73:274, 1994
133. Andersch B, Milsom I, Rybo G: An objective evaluation of flurbiprofen and tranexamic acid in the treatment of idiopathic menorrhagia. Acta Obstet Gynecol Scand 67:645, 1988
136. DeVore GR, Owens O, Kase N: Use of intravenous Premarin in the treatment of dysfunctional uterine bleeding—a double-blind randomized controlled study. Obstet Gynecol 59:285, 1982
139. Cameron IT, Leask R, Kelly RW, et al: The effects of danazol, mefenamic acid, norethisterone, and a progesterone-impregnated coil on endometrial prostaglandin concentrations in women with menorrhagia. Prostaglandins 34:99, 1987
140. Higham JM, Shaw RW: A comparative study of danazol, a regimen of decreasing doses of danazol, and norethindrone in the treatment of objectively proven unexplained menorrhagia. Am J Obstet Gynecol 169:1134, 1993
143. Bergqvist A, Rybo G: Treatment of menorrhagia with intrauterine release of progesterone. Br J Obstet Gynaecol 90:255, 1983
144. Irvine GA, Campbell-Brown MB, Lumsden MA, et al: Randomized comparative trial of the levonorgestrel intrauterine system and norethisterone for treatment of idiopathic menorrhagia. Br J Obstet Gynaecol 105:592, 1998
145. Crosignani PG, Vercellini P, Mosconi P, et al: Levonorgestrel-releasing intrauterine device versus hysteroscopic endometrial resection in the treatment of dysfunctional uterine bleeding. Obstet Gynecol 90:257, 1997
150. Panay N, Studd J: Place of intra-uterine delivery of progestogen and progesterone. In Sheth SS, Sutton CJG (eds): Menorrhagia. pp 111, 126 Oxford, Isis Medical Media, 1999
152. Kuhnz W, Al-Yakoub G, Fuhrmeister A: Pharmacokinetics of levonorgestrel and ethinylestradiol in 9 women who received a low-dose oral contraceptive over a treatment period of 3 months, and, after a wash-out phase, a single oral administration of the same contraceptive formulation. Contraception 46:455, 1992
153. Weiner E, Victor A, Johansson EDB: Plasma levels of d-norgestrel after oral administration. Contraception 14:563, 1976
155. Tillson SA, Marian M, Hudson R, et al: The effect of intrauterine progesterone on the hypothalamic-pituitary-ovarian axis in humans. Contraception 11:179, 1975
158. Chimbara TH, Anderson AB, Cope E, et al: Preliminary results on clinical and endocrine studies in the treatment of menorrhagia with danazol. Postgrad Med J 55(Suppl 5):90, 1979
159. Aleem FA, Predanic M: The hemodynamic effect of GnRH agonist therapy on uterine leiomyoma vascularity: A prospective study using transvaginal color Doppler sonography. Gynecol Endocrinol 9:253, 1995
161. Scahlaff WD, Zerhouni EA, Huth JA, et al: A placebo-controlled trial of a depot gonadotropin-releasing hormone analogue (leuprolide) in the treatment of uterine leiomyomata. Obstet Gynecol 74:856, 1989
169. Ferenczy A, Gelfand M: The biologic significance of cytologic atypia in progesterone-treated endometrial hyperplasia. Am J Obstet Gynecol 160:126, 1989