DIAGNOSING AUB

Diagnosing abnormal uterine bleeding in perimenopausal women

By Steven R. Goldstein, MD

A diagnostic work up holds the challenge of ruling out uterine malignancy and hyperplasia, as well as pinpointing the non-cancerous cause of the bleeding. An expert in the field outlines the most appropriate assessment tools to help accomplish these goals.

About one out of every five visits to a gynecologist's office involves abnormal uterine bleeding, a problem that accounts for almost 25% of gynecologic operations.^1,2 When a woman presents with bleeding during the perimenopause—defined as the 2 to 8 years preceding menopause and the year after a woman's final menses—a clinician is faced with at least two related issues: How do I rule out cancer and endometrial hyperplasia; and assuming these conditions are excluded, how can I accurately evaluate the other causes of abnormal bleeding?

Of course, even when cancer and endometrial hyperplasia have been ruled out, the decision to treat other forms of abnormal uterine bleeding or simply reassure patients that nothing is seriously wrong is a complex one. "Symptomatic" bleeding in perimenopausal women is inextricably bound up with the fear that such bleeding is a sign of impending doom. For many such patients, simply explaining that cancer or endometrial hyperplasia have been excluded provides them with the security to accept the inconveniences and quality-of-life issues that their condition produces. It is often their fears about what the bleeding may have signified that caused them to seek medical attention in the first place.

The physiologic changes accompanying perimenopause

The first clinical sign of perimenopause is a break in cyclicity in a woman who has previously had a regular menstrual pattern. Some women will have irregular spotting while others will skip several cycles and then return to a normal pattern.

Much of this results from accelerated depletion of oocytes, which eventually causes ovulation to cease. These changes are accompanied by significant changes in hormonal levels, especially estrogen.³ As ovarian estrogen production decreases, the pituitary increases its production of FSH to stimulate the ovary to secrete more estrogen. Unfortunately FSH levels can fluctuate from month to month and from woman to woman during perimenopause, which limits their usefulness as a predictor of perimenopause.⁴ In a landmark longitudinal study, Dr. Santoro and her colleagues found that perimenopausal women tended to be relatively hyperestrogenic and to also have lower progestogen levels.⁵ Shortening of the menstrual cycle and an accelerated follicular phase in ovulatory cycles alter-nate sporadically with anovulation. This results in such perimenopausal women being more likely to suffer from anovulatory bleeding and puts them at increased risk for endometrial hyperplasia.

Aging and the risk of cancer

The incidence of endometrial carcinoma increases with a woman's age. The rate of endometrial carcinoma in women between 30 and 39 is 2.3/100,000; in women 35 to 39, it increases to 6.1/100,000; and in women between 40 and 49 itis 36.2/100,000.⁶ Based on age alone, the American College of Obstetricians and Gynecologists recommends endometrial assessment to exclude cancer in any woman older than 35 who is suspected of having anovulatory uterine bleeding.⁷ One study done among 433 women between age 39 and menopause with any break in menstrual cyclicity revealed no carcinoma and an incidence of hyperplasia of 2.5% (11/433).⁸ Of these, six were contained within polyps (focal lesions).

The term "menometrorrhagia" is often used to refer to any abnormality in menstrual flow, but the expression has been so overused as to lose much ofits meaning or provide a real clue about underlying causes. Ovulatory cycles are usually regular in nature but variability is most commonly due to variations in the proliferative phase. Although intermenstrual bleeding (metrorrhagia) likely represents dysfunctional anovulatory bleeding, clinicians must exclude nonuterine causes (cervical, vaginal, inflammatory, or neoplastic). And, of course, an antecedent pregnancy must be ruled out as well.

Leiomyomata are also quite common in midlife women. The exact mechanism by which fibroids result in abnormal uterine bleeding is not entirely clear, but those of the submucous variety are most often responsible. In addition, as the size of the uterine cavity increases with some, but not all, growing fibroids, the concurrent menorrhagia may simply be the result of an increase in the surface area of the endometrium. The same will be true of the increasing uterine size seen with increasing parity.

Diagnostic approachesto abnormal bleeding

In the vast majority of patients, dysfunctional uterine bleeding is associated with episodes of anovulation that can best be managed hormonally or expectantly with patient reassurance. The proper diagnostic work up will distinguish such patients from those with organic pathology.

Several diagnostic tools have been used over the years to determine the cause of abnormal uterine bleeding.

Curettage. Initially, curettage was the diagnostic gold standard. First described in 1843, its performance in the hospital becamethe most common operation performed on women in the world.⁹ A study of curettage before hysterectomy found that less than one quarter of the cavity was curetted in 16% of specimens, less than one half of the cavity was curetted in 60%, and less than three quarters of the endometrial cavity was effectively curetted in 84%.¹⁰

In the 1970s, vacuum-suction curettage devices allowed sampling without anesthesia in an office setting. The most popular was the Vabra aspirator. It was 86% accurate in diagnosingcancer.¹¹ Subsequently, cheaper, smaller, less painful plastic catheters with their own internal pistons to generate suction became popular. One of these, the Pipelle device, was found to have similar efficacy but better patient acceptance when compared with the Vabra.¹²

However, one group of researchers who did a pathologic study of 25 hysterectomy specimens found that the percentage of endometrial surface sampled by the Pipelle device was 4% versus 41% for the Vabra aspirator.¹³

In one widely publicized study the Pipelle had a 97.5% sensitivity for detecting endometrial cancer in 40 patients undergoing hysterectomy.¹⁴ The shortcoming of that study was that the diagnosis of malignancywas known before the performance of the specimen collection.

In another study, Pipelle aspiration biopsy was performed in 135 premenopausal patients before curettage.¹⁵ Thirteen patients (10%) had different histologic results on Pipelle biopsy as compared with curettage. It is interesting that only five of these patients had polyps, of which Pipelle sampling missed three. In total, 18 patients had hyperplasia, of which Pipelle sampling missed the diagnosis in seven (39%), thus underscoring the often focal nature of that pathologic process.

Finally in yet another investigation, Guido and associates studied the Pipelle biopsy in patients with known carcinoma undergoing hysterectomy and found that, among 65 patients, Pipelle biopsy provided adequate tissue for analysis in 63 (97%).¹⁶ Malignancy was detected in only 54 patients (83%). Of the 11 with false-negative results, five (8%) had disease confined to endometrial polyps and three (5%) hada tumor localized to less than 5% of the surface area. The surface area of the endometrial involvement in that study comprised5% of the cavity or less in three of 65 (5%); 5% to 25% of the cavity in 12 of 65 (18%), of which the Pipelle missed four; 26% to 50% of the cavity in 20 of 65 (31%), of which the Pipelle missed four; and more than 50% of the cavity in 30 of 65 patients (46%), of which the Pipelle missed none. These results provide a real insight into the way endometrial carcinoma can be distributed over the endometrial surface on the one hand or confined to a polyp. Because tumors localized ina polyp or a small area of endometrium may go undetected,the researchers concluded thatthe "Pipelle is excellent for detecting global processes in the endometrium."

These data suggest that undirected sampling, whether through curettage or various types of suction aspiration, will often be fraught with error, especially in cases in which the abnormalityis not global but focal (polyps, focal hyperplasia, or carcinoma involving small areas of theuterine cavity).

Hysteroscopy. With directed biopsy, hysteroscopy has been shown to be superior to curettage, although this procedure requires specialized equipment and is very operator-dependent.¹⁷ It often requires ambulatory surgery centers and general anesthesia. When it is performed as an office procedure with local or no anesthesia, it can cause significant patient discomfort.

Transvaginal ultrasonography (TVS). This has been explored as an inexpensive, noninvasive, convenient way to indirectly visualize the endometrial cavity. It has been used most extensively in postmenopausal patients with abnormal bleeding and has beenshown to effectively exclude significant abnormality when the endometrial echo is 4 to 5 mm or less.^18,19 Premenopausal women have also been evaluated with TVS to exclude endometrial abnormalities when abnormal bleeding is present.²⁰ The addition of saline infusion sonohysterography (SIS) can reliably distinguishperimenopausal patients withdysfunctional abnormal bleeding (no anatomic abnormality) from those with globally thickened endometria or those with focal abnormalities.²¹

A clinical algorithm was proposed and studied in perimenopausal women with abnormal bleeding using unenhanced TVS, followed by SIS for selected patients, and then either no endometrial sampling, undirected endometrial sampling, or visually directed endometrial sampling, depending on whether the ultrasonographically-based triage revealed no anatomic abnormality, globally thickened endometrium, orfocal abnormalities, respectively.⁸

In that study, 280 patients (65%) displayed a thin, distinct, symmetric endometrial echo of 5 mm or less on day 4 through 6, and dysfunctional uterine bleeding was diagnosed (Figure 2). One hundred fifty-three (35%) had SIS. Of these procedures, 44 (29%) were performed because of the inability to adequately characterize and measure the endometrium (Figures 3 and 4) and 109 (71%) were done for endometrial measurement of 5 mm or more. Sixty-one of those patients then had both anterior and posterior endometrial thickness that was symmetric and less than 3 mm, compatible with dysfunctional uterine bleeding. Of the original 433 patients in the study, 58 patients (13%) had focal polypoid masses (Figure 5) that were removed hysteroscopically and confirmed pathologically. Twenty-two patients (5%) had submucous myomas although 148 patients (34%) had clinical and ultrasonographic evidence of fibroids. Ten patients had single-layer measurements of endometrium of more than 3 mm (range 3–9 mm) during SIS. Of these, five patients had a proliferative endometrium and five had a hyperplastic endometrium. SIS was technically inadequate in two patients, who then underwent hysteroscopy with curettage. Undirected office biopsy alone without imaging would have potentially missed the diagnosis of focal lesions such as polyps, submucous myomas, and focal hyperplasia in up to 80 patients (18%).

A reliable assessment with ultrasonography (U/S) requires that the endometrial echo be homogeneous, surrounded by an intact hypoechoic junctional zone, and that the operator constantly remember that the en- dometrial cavity is a three-dimensional structure. This may account for why Dijkhuizen and associates had four cases that supposedly measured less than 10 mm (some as little as 2 mm) and yet at hysteroscopy displayed polyps.²⁰ Such cases underscore the importance of the three-dimensional character of the endometrial cavity and the occasional propensity of ultrasonographers to obtain a limited number of two-dimensional views and assume that these represent the entire endometrial cavity. Any one "frozen" U/S image is nothing more than a two-dimensional "snapshot," and failure to meticulously recreate three-dimensional anatomy will result in errors (Figures 6 and 7).

Another potential pitfall of SIS is the irregular topographic surface of the endometrial cavity as it proliferates. The sonohysterography procedure is very time sensitive and should be performed as the bleeding episode ends, when the endometrial thickness can be expected to be at its thinnest. Delaying the procedure can re-sult in development of a sizable amount of endometrial tissue that can be mistaken for small polyps or focal hyperplasia leading to inappropriate further interventions. Such a sonographic appearance is reminiscent of "moguls" on a ski slope (Figure 8).

Conclusion

Abnormal uterine bleeding in midlife women is common. Ob/gyns need to exclude uterine malignancy or hyperplasia, but once that's accomplished, they also need to properly diagnose the cause of the bleeding in order to best tailor various therapeutic options for each patient. This is best accomplished with a thorough understanding of the physiologic and pathologic conditions that result in such bleeding as well as the various diagnostic and therapeutic modalities currently available.

REFERENCES

1. Shwayder JM. Patholphysiology of abnormal uterine bleeding. Obstet Gynecol Clin North Am. 2000;27:219-234.

2. Nesse RE. Abnormal vaginal bleeding in perimenopausal women. Am Fam Physicians. 1989;40:185-192.

3. Richardson SJ, Senikas V, Nelson JF. Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion. J Clin Endocrinol Metab. 1987;65;1231-1237.

4. Stellato RK, Crawford SL, McKinlay SM, et al. Can follicle-stimulating hormone be used to define menopausal status? Endocr Pract. 1998;4:137-141.

5. Santoro N, Brown JR, Adel T, et al. Characterization of reproductive hormonal dynamics in the perimenopause. J Clin Endocrinol Metab. 1996;81:1495-1501.

6. SEER Cancer Statistics Review, 1973-1996. Available at http://seer.cancer.gov/ csr/1973_1996/.

7. Management of Anovulatory Bleeding. American College of Obstetricians and Gynecologists, ACOG Practice Bulletin No. 14, 2000.

8. Goldstein SR, Zeslter I, Horan CK, et al. Ultrasonography-based triage for perimenopausal patients with abnormal uterine bleeding. Am J Obstet Gynecol. 1997;177:102-108.

9. Ricci JV. Gynaecologic surgery and instruments of the nineteenth century prior to the antiseptic age. In: The Development of Gynaecological Surgery and Instruments. Philadelphia, Pa: Blakiston; 1949:326-328.

10. Stock RJ, Kanbour A. Prehysterectomy curettage. Obstet Gynecol. 1975;45:537-541.

11. Vuopala S. Diagnostic accuracy and clinical applicability of cytological and histological methods for investigating endometrial carcinoma. Acta Obstet Gyneocl Scand. 1977(suppl 70);56:1-72.

12. Kaunitz AM, Masciello A, Ostrowski M, et al. Comparison of endometrial biopsy with the Pipelle and Vabra aspirator. J Reprod Med. 1988;33:427-431.

13. Rodriguez MH, Platt LD, Medearis AL, et al. The use of transvaginal sonography for evaluation of postmenopausal size and morphology. Am J Obstet Gynecol. 1988;159:810-814.

14. Stovall TG, Photopulos GJ, Poston WM, et al. Pipelle endometrial sampling in patients with known endometrial carcinoma. Obstet Gynecol. 1991;77:954-956.

15. Goldchmit R, Katz Z, Blickstein I, et al. The accuracy of endometrial Pipelle sampling with and without sonographic measurement of endometrial thickness. Obstet Gynecol. 1993;82:727-730.

16. Guido RS, Kanbour-Shakir A, Ruhn MC, et al. Pipelle endometrial sampling sensitivity in the detection of endometrial cancer. J Reprod Med. 1995;40:553-555.

17. Gimpelson RJ, Rappold HO. A comparative study between panoramic hysteroscopy with directed biopsies and dilatation and curettage. Am J Obstet Gynecol. 1988;158:489-492.

18. Goldstein SR, Nachtigall M, Snyder JR, et al. Endometrial assessment by vaginal ultrasonography before endometrial sampling in patients with postmenopausal bleeding. Am J Obstet Gynecol. 1990;163:119-123.

19. Langer RD, Pierce JJ, O'Hanlan KA, et al. Transvaginal ultrasonography compared with endometrial biopsy for the detection of endometrial disease. N Engl J Med. 1997;337:1792-1798.

20. Dijkhuizen FP, Brolmann HA, Potters AE, et al. The accuracy of transvaginal ultrasonography in the diagnosis of endometrial abnormalities. Obstet Gynecol. 1996;87:345-349.

21. Goldstein SR. Use of ultrasonohysterography for triage of perimenopausal patients with unexplained uterine bleeding. Am J Obstet Gynecol. 1994;170;565-570.

Dr. Goldstein is Professor of Obstetrics and Gynecology at New York University School of Medicine, New York, N.Y.

Key points

• Transvaginal ultrasonography, which has been extensively used to visualize the endometrial cavity in postmenopausal patients with abnormal bleeding, can effectively exclude significant abnormalities when the endometrial echo is 4 to 5 mm or less.

• Undirected endometrial sampling, whether through curettage or various types of suction aspiration, is often fraught with errors, especially in situations in which the abnormality is not global but focal, (polyps, focal hyperplasia, or carcinoma involving small areas of the uterine cavity).

• Although intermenstrual bleeding likely represents dysfunctional anovulatory bleeding, physicians must exclude cervical, vaginal, inflammatory, or neoplastic causes. An antecedent pregnancy must be ruled out as well.

Steven Goldstein. Diagnosing abnormal uterine bleeding in perimenopausal women.

Contemporary Ob/Gyn

May 1, 2003;48:96-109.