Diagnosis and Management of Endometrial Cancer MICHAEL M

Diagnosis and Management of Endometrial Cancer MICHAEL M. BRAUN, DO, Madigan Army Medical Center, Tacoma, Washington ERIKA A. OVERBEEK-WAGER, DO, Evans Army Community Hospital, Fort Carson, Colorado ROBERT J. GRUMBO, MD, Madigan Army Medical Center, Tacoma, Washington

Endometrial cancer is the most common gynecologic malignancy. It is the fourth most common cancer in women in the United States after breast, lung, and colorectal cancers. Risk factors are related to excessive unopposed exposure of the endometrium to estrogen, including unopposed estrogen therapy, early menarche, late menopause, tamoxifen therapy, nulliparity, infertility or failure to ovulate, and polycystic ovary syndrome. Additional risk factors are increasing age, obesity, hypertension, diabetes mellitus, and hereditary nonpolyposis colorectal cancer. The most common presentation for endometrial cancer is postmenopausal bleeding. The American Cancer Society recommends that all women older than 65 years be informed of the risks and symptoms of endometrial cancer and advised to seek evaluation if symptoms occur. There is no evidence to support endometrial cancer screening in asymptomatic women. Evalua- tion of a patient with suspected disease should include a pregnancy test in women of childbearing age, complete blood count, and prothrombin time and partial thromboplastin time if bleeding is heavy. Most guidelines recommend either transvaginal ultrasonography or endometrial biopsy as the initial study. The mainstay of treatment for endometrial cancer is total hysterectomy with bilateral salpingo-oophorectomy. Radiation and chemotherapy can also play a role in treatment. Low- to medium-risk endometrial hyperplasia can be treated with nonsurgical options. Survival is generally defined by the stage of the disease and histology, with most patients at stage I and II having a favorable prognosis. Controlling risk factors such as obesity, diabetes, and hypertension could play a role in the prevention of endometrial cancer. (Am Fam Physician. 2016;93(0):468-474. Copyright © 2016 American Academy of Family Physicians.)

CME This clinical content ndometrial cancer is the most com- tumors are generally low grade. Type II conforms to AAFP criteria mon gynecologic malignancy. It is tumors are more likely to be high grade and for continuing medical edu- cation (CME). See CME Quiz the fourth most common cancer of papillary serous or clear cell histologic Questions on page 449. in women after breast, lung, and type. They carry a poor prognosis and have Ecolorectal cancers. Projections from the a high risk of relapse and metastasis. Type II Author disclosure: No relevant financial affiliations. American Cancer Society (ACS) for 2015 accounts for only 10% of endometrial can- ▲ Patient information: estimated 54,870 new cases of endometrial cers, but it is associated with 40% of related A handout on this topic, cancer and 10,170 deaths from the disease.1 deaths.2,3 Familial tumors are commonly written by the authors of The death rate for endometrial cancer has found in association with Lynch syndrome this article, is available at http://www.aafp.org/ increased more than 100% during the past (hereditary nonpolyposis colorectal cancer). afp/2016/0315/p468-s1. 20 years, rising by 8% since 2008. The mean Genetic disease represents 10% of cases of html. age of patients at the time of diagnosis is 63 endometrial cancer.3 years, with 90% of cases occurring in women Endometrial hyperplasia represents a pre- older than 50 years. Only 20% of patients cursor lesion to endometrial cancer. Hyper- with endometrial cancer receive a diagnosis before menopause.2 WHAT IS NEW ON THIS TOPIC: Histopathology ENDOMETRIAL CANCER Endometrial cancer is generally classified The 2009 update of the International into two types.2 Type I is the most common Federation of Gynecology and Obstetrics form, representing more than 70% of cases. tumor-node-metastasis staging system for endometrial cancer better predicts disease Type I tumors are associated with unop- prognosis compared with the previous posed estrogen stimulation and are known system (Table 1). as endometrioid adenocarcinoma.3 These

468Downloaded American from the Family American Physician Family Physician website at www.aafp.org/afp.www.aafp.org/afp Copyright © 2016 American AcademyVolume of Family 93, Physicians. Number For 6 the◆ March private, 15,noncom 2016- mercial use of one individual user of the website. All other rights reserved. Contact [email protected] for copyright questions and/or permission requests. Endometrial Cancer SORT: KEY RECOMMENDATIONS FOR PRACTICE

Evidence Clinical recommendation rating References Comments

Women older than 65 years should be informed of the risks and C 4 Recommendation based symptoms of endometrial cancer and advised to seek evaluation if on consensus guidelines symptoms occur. Women with abnormal uterine bleeding should be evaluated for C 2, 3, 19 Recommendation based endometrial cancer if they are older than 45 years or if they have a on consensus guidelines history of unopposed estrogen exposure. In postmenopausal women, the endometrial thickness on transvaginal C 17, 18 Recommendation based ultrasonography should be less than 4 to 5 mm. With thickness on consensus guidelines above this level, biopsy should be considered to rule out endometrial hyperplasia or cancer.

A = consistent, good-quality patient-oriented evidence; B = inconsistent or limited-quality patient-oriented evidence; C = consensus, disease-oriented evidence, usual practice, expert opinion, or case series. For information about the SORT evidence rating system, go to http://www.aafp.org/afpsort. plasia carries a 1% to 3% risk of progression asymptomatic women, with the exception to cancer. Atypical hyperplasia is associated of those who have or are at increased risk of with greater cancer risk than simple or com- Lynch syndrome.4 Although the recommen- plex hyperplasia; 30% to 40% of patients dation is controversial, these patients should with atypical hyperplasia have concomitant be screened annually with an endometrial adenocarcinoma.2 biopsy starting at age 35 because of a 22% to 50% lifetime risk of endometrial cancer.4,12 Risk Factors Patients with Lynch syndrome should be Risk factors for type I endometrial cancer advised to keep a menstrual calendar and are related to unopposed exposure of the report abnormal bleeding. Women older endometrium to estrogen, including unop- than 40 years who have the mutation and do posed estrogen therapy, early menarche, late not wish to become pregnant in the future menopause, tamoxifen therapy, nulliparity, can consider a prophylactic hysterectomy.13,14 infertility or failure to ovulate, and polycys- There are no recommendations on screen- tic ovary syndrome. Other risk factors not ing for endometrial cancer in patients who involving unopposed estrogen include fam- are taking tamoxifen; however, those who ily history of endometrial cancer, age older present with abnormal uterine bleeding than 50 years, hypertension, diabetes melli- should be considered for diagnostic workup. tus, obesity, thyroid disease, and Lynch syn- Use of the levonorgestrel-releasing intrauter- drome.2-6 Although they are less common ine system (Mirena) has not been proven to overall, type II tumors are found predomi- protect against endometrial hyperplasia or nantly in black women older than 50 years.2 cancer in patients who take tamoxifen.15 Nearly 70% of patients with early stage Management of risk factors such as obesity, endometrial cancer are obese.7,8 The relative diabetes, and hypertension could play a role risk of death increases with rising body mass in the prevention of endometrial cancer. For index.8 Patients undergoing treatment with women on hormone therapy, the addition of tamoxifen are at increased risk of endome- progesterone has been shown to decrease the trial cancer.9-11 risk of endometrial cancer.16 Protective factors include prior use of combined oral contraceptives for one or History more years and grand multiparity.2,3 Vaginal bleeding is the most common clinical presentation of endometrial cancer in Screening and Prevention postmenopausal women.2,17,18 Approximately The ACS recommends that all women older 75% of postmenopausal women who are than 65 years be informed of the risks and diagnosed with endometrial cancer are diag- symptoms of endometrial cancer and advised nosed at an early stage, which improves the to seek evaluation if symptoms occur. There chances of successful treatment.2 However, is no evidence to support the screening of only 10% to 20% of postmenopausal women

March 15, 2016 ◆ Volume 93, Number 6 www.aafp.org/afp American Family Physician 469 Endometrial Cancer

who are evaluated for uterine bleeding are abnormal vaginal bleeding, including post- diagnosed with endometrial cancer because menopausal bleeding, and the pretreatment the most common cause of postmenopausal evaluation and follow-up of endometrial bleeding is endometrial atrophy.17,18 All post- cancer (https://acsearch.acr.org/list). The menopausal bleeding should be investigated, type of initial study depends on the avail- especially if risk factors for endometrial ability of options and their level of invasive- hyperplasia or cancer are present. Abnormal ness, and patient and physician preference. uterine bleeding can also be a sign of endometrial cancer in premenopausal women, TRANSVAGINAL ULTRASONOGRAPHY who comprise 20% of cases of endometrial Transvaginal ultrasonography is often the cancer.17 The American College of Obstetri- initial diagnostic study of choice when eval- cians and Gynecologists (ACOG) recom- uating for endometrial cancer because of mends that women with abnormal uterine its availability, cost-effectiveness, and high bleeding be evaluated for endometrial can- sensitivity.17 Transvaginal ultrasonography cer if they are older than 45 years, or if they can be used to measure endometrial thick- are younger than 45 years and have a history ness. There is some uncertainty regarding of unopposed estrogen exposure.2,3,19 Evalu- the optimal cutoff for endometrial thick- ation can be done with endometrial tissue ness. Several meta-analyses that have used a sampling or ultrasonography.19 cutoff measurement of 5 mm or less had a 96% sensitivity and a posttest probability of Physical Examination 2.5% for endometrial cancer in postmeno- There are few physical examination findings pausal women.20,21 A recent ACOG commit- in women with endometrial cancer. A pelvic tee opinion notes that the cutoff value for a examination should be performed to evaluate normal transvaginal ultrasonography result for other sources of abnormal bleeding, such should be 4 mm or less.18 Postmenopausal as the vagina or cervix. The uterus and adnexa patients with endometrial thickness greater should be palpated for unusual masses. than 5 mm should be evaluated with a tissue Abnormal physical examination findings may sample, especially if bleeding is present. The be suggestive of more advanced disease. American College of Radiology uses a cutoff of 5 mm or less.17 The optimal cutoff for Laboratory Evaluation evaluating premenopausal women has not There are no specific laboratory tests for the been defined, but recommendations include evaluation of endometrial cancer. Labora- a cutoff of 16 mm or less. In all patients, if tory tests should include a pregnancy test bleeding persists despite a normal transvagi- in patients of childbearing age. A complete nal ultrasonography result, a tissue biopsy is blood count and prothrombin time and warranted.2 partial thromboplastin time may also be considered for patients with heavy bleed- ENDOMETRIAL SAMPLING ing. Papanicolaou smears are not a required The definitive diagnosis of endometrial can- part of the evaluation, but occasionally a Pap cer requires an endometrial tissue sample.6 smear result can suggest endometrial cancer Curettage has been considered the preferred (i.e., atypical glandular cells). method for obtaining a tissue sample, but the newer Pipelle method offers an alter- Diagnostic Studies native.2,22 When an adequate sample is Most guidelines recommend either trans- obtained, the Pipelle method has high diag- vaginal ultrasonography or endometrial nostic accuracy, with a positive predictive biopsy as the initial study for the evalua- value of 81.7% and a negative predictive tion of endometrial cancer.6,17-19 The Ameri- value of 99.1%.6 However, adequate samples can College of Radiology Appropriateness can be difficult to obtain using the Pipelle Criteria include tables outlining the pre- method. One study showed that only 34% ferred imaging studies for the evaluation of of patients had an adequate sample.22 This

470 American Family Physician www.aafp.org/afp Volume 93, Number 6 ◆ March 15, 2016 Endometrial Cancer

percentage rose to 60% when evaluating women with an endometrial thickness of at BEST PRACTICES IN ONCOLOGY: least 5 mm. If an adequate sample cannot be RECOMMENDATIONS FROM THE CHOOSING WISELY CAMPAIGN obtained, a referral for dilation and curet- Recommendation Sponsoring organization tage should be considered. Additional evaluation is needed if symptoms persist despite Do not perform Papanicolaou tests for Society of Gynecologic 6 surveillance of women with a history of Oncology a benign biopsy result. endometrial cancer. SALINE INFUSION SONOHYSTEROGRAPHY Source: For more information on the Choosing Wisely Campaign, see http:// Saline infusion sonohysterography can also www.choosingwisely.org. For supporting citations and to search Choosing Wisely recommendations relevant to primary care, see http://www.aafp.org/afp/ be used to evaluate the endometrial cavity. recommendations/search.htm. This study technique uses saline infused into the endometrial cavity, followed by ultrasonography to allow better visualiza- combining former stages IA and IB, elimina- tion of structural changes, particularly when tion of stage IIA, and stratification of stage patients have focal irregularities such as pol- IIIC into pelvic nodes only or para-aortic yps, submucosal fibroids, or endometrial nodal involvement26 (Table 15,24-26). hyperplasia.17 Saline infusion sonohysterography is rarely used, but it can be considered ENDOMETRIAL HYPERPLASIA when endometrial biopsy or transvaginal Management of endometrial cancer is bro- ultrasonography is inadequate.17 ken down into surgical and nonsurgical therapies. All patients with endometrial HYSTEROSCOPY hyperplasia should have testing to rule out Hysteroscopy is commonly used to evaluate concurrent adenocarcinoma. abnormal uterine bleeding and offers direct The definitive treatment for complex visualization of the endometrial cavity.2 atypical endometrial hyperplasia is hyster- Hysteroscopy can be performed in conjunc- ectomy. Surgical options include abdomi- tion with a focal biopsy or curettage. A sys- nal and minimally invasive procedures tematic review found that hysteroscopy had a such as laparoscopy. Hysterectomy can sensitivity of 99.2% and specificity of 86.4% be performed with or without bilateral in the diagnosis of endometrial cancer.23 salpingo-oophorectomy.27 ACOG does not recommend supracervical procedures as ADDITIONAL DIAGNOSTIC IMAGING treatment because these procedures can Magnetic resonance imaging may be able to leave behind residual disease.28 Abdominal provide additional information on endome- surgery is associated with more pain, longer trial thickening or structural abnormalities recovery, and longer hospital stay compared such as fibroids or adenomyosis when trans- with laparoscopy.29 Additional procedures vaginal ultrasonography is not adequate and may be warranted if carcinoma is identified. saline infusion sonohysterography is not Lymphadenectomy at the time of surgery is tolerated.17 Computed tomography and posi- not recommended, as long as there are no tron emission tomography are generally not intra-abdominal findings suggestive of inva- useful in the initial evaluation.17 sive processes. Most patients with endometrial hyperplasia will not have carcinoma.27 Treatment Patients with low-risk endometrial hyper- The International Federation of Gynecol- plasia (without atypia) or multiple comor- ogy and Obstetrics tumor-node-metastasis bidities precluding surgery, and those who staging system of endometrial cancer desire continued fertility, can be treated with was updated in 2009 and appears better nonsurgical options. The most common able to predict prognosis compared with treatment option is progesterone therapy to the previously published 1988 system.24,25 stabilize the disease and prevent progression Major changes to the new system included to endometrial cancer.

March 15, 2016 ◆ Volume 93, Number 6 www.aafp.org/afp American Family Physician 471 Endometrial Cancer

Use of the levonorgestrel-releasing intra- samplings in the posttreatment surveillance uterine system and oral progesterone (e.g., period have also been recommended.27 medroxyprogesterone [Provera], 10 mg daily for 10 to 14 days per month) for the treatment ENDOMETRIAL CANCER of low- to medium-risk endometrial hyper- Surgical Approaches. The mainstay of treat- plasia showed a reduction in hyperplasia ment is total hysterectomy with bilateral six months after treatment.27,30 The optimal salpingo-oophorectomy, para-aortic and pel- route, dose, and duration of therapy have not vic lymphadenectomy, and pelvic washing to been well defined. General consensus is to stage the disease. Laparoscopy has been asso- treat patients for six months, with tissue sam- ciated with fewer postoperative complica- ples obtained every three months to evaluate tions than laparotomy. Vaginal hysterectomy for disease regression. Multiple endometrial is generally not recommended because it

Table 1. International Federation of Gynecology and Obstetrics Tumor-Node-Metastasis Staging for Uterine Carcinoma

Five-year Stage Description Treatment options survival (%)

I Tumor limited to uterine corpus Bilateral salpingo-oophorectomy Hysterectomy Pelvic and para-aortic lymphadenectomy Pelvic washings IA Tumor limited to endometrium or < 50% myometrial invasion 90 IB ≥ 50% myometrial invasion 78

II Tumor invades cervix but not beyond the uterus Bilateral salpingo-oophorectomy 74 Hysterectomy Pelvic and para-aortic lymphadenectomy Pelvic washings Radiation

III Local and/or regional spread Bilateral salpingo-oophorectomy Hysterectomy Pelvic and para-aortic lymphadenectomy Pelvic washings Radiation IIIA Tumor involves serosa and/or adnexa 56 IIIB Vaginal or parametrial involvement 36 IIIC Metastases to pelvic and/or para-aortic lymph nodes Systemic adjuvant therapy IIIC1 Regional lymph node metastasis to pelvic lymph node 57 IIIC2 Regional lymph node metastasis to para-aortic lymph nodes 49 with or without pelvic lymph nodes

IV Intra-abdominal or extra-abdominal metastasis Surgery and tumor debulking Systemic adjuvant therapy IVA Invasion of bladder or bowel mucosa 22 IVB Distant metastasis (including intra-abdominal and inguinal 21

lymph node spread)

Adapted with permission from Buchanan EM, et al. Endometrial cancer. Am Fam Physician. 2009;80(10):1079, with additional information from references 24 through 26.

472 American Family Physician www.aafp.org/afp Volume 93, Number 6 ◆ March 15, 2016 Endometrial Cancer

precludes abdominal survey and lymphad- Data Sources: Searched Cochrane databases, DynaMed, enectomy.3 Most patients who have endo- PubMed, PEPID, Clinical Evidence, National Guideline Clearinghouse, Essential Evidence Plus, UpToDate, and metrial cancer will have stage I carcinoma. OVID using key terms endometrial cancer, tamoxifen, Need for further treatment is based on intra- cancer, diagnosis, endometrial hyperplasia, biopsy, and operative and histologic findings.2 treatment. The search included meta-analyses, random- Pelvic and para-aortic lymphadenectomy ized controlled trials, clinical trials, and reviews. Search dates: November 1, 2014, through February 20, 2015; remain controversial. Several studies have August 1, 2015; and December 1, 2015. noted an associated improvement in sur- 31 The views expressed in this article are those of the vival, whereas others have not. There is no authors and do not reflect the official policy of the consensus about which patients will require Department of the Army, the Department of Defense, or lymph node staging.2 the U.S. government. Adjuvant Radiotherapy. Radiation therapy does not affect overall survival in patients The Authors with low-grade carcinoma. It is associ- MICHAEL M. BRAUN, DO, FAAFP, is associate residency ated with a reduction in quality of life and director of the Family Medicine Residency at Madigan increased morbidity when used in patients Army Medical Center, Tacoma, Wash., and clinical fac- with low-risk endometrial cancer.3,32 Radia- ulty of family medicine at the University of Washington, tion therapy is an option for patients who are Seattle. medically inoperable.33 ERIKA A. OVERBEEK-WAGER, DO, is a staff family physi- Chemotherapy and Hormone Therapy. cian at Evans Army Community Hospital, Fort Carson, Colo. Cytoreduction therapy (debulking with surgery and chemotherapy or radiation) appears ROBERT J. GRUMBO, MD, FAAFP, is adjunct faculty in the Family Medicine Residency at Madigan Army Medical to improve survival time in patients with Center. intra-abdominal disease by increasing survival and decreasing recurrence.34 Evidence Address correspondence to Michael M. Braun, DO, Madigan Army Medical Center, 9040 Fitzsimmons Dr., to support the use of adjuvant progesterone Tacoma, WA 98431 (e-mail: [email protected]). therapy to prevent endometrial cancer recur- Reprints are not available from the authors. rence is lacking.35 Progesterone is a treatment option for patients with stage I endometrial REFERENCES cancer who wish to preserve fertility.36 There 1. American Cancer Society. Cancer Facts and Figures 2014. is a 30% chance that the patient whose biopsy http://www.cancer.org/acs/groups/content/@research/ noted a grade I carcinoma may have a grade documents/webcontent/acspc-042151.pdf. Accessed II or III carcinoma instead.37 Patients should October 27, 2014. be counseled on immediate hysterectomy 2. Sorosky JI. Endometrial cancer. Obstet Gynecol. 2012; once childbearing is completed. 120(2 pt 1):383-397. 3. Practice Bulletin No. 149: Endometrial cancer. Obstet Gynecol. 2015; 125(4):1006-1026. Prognosis 4. Smith RA, et al.; ACS Prostate Cancer Advisory Com- Survival is based on the stage and histology mittee, ACS Colorectal Cancer Advisory Committee, of the diagnosis. Most patients with stage I ACS Endometrial Cancer Advisory Committee. Ameri- can Cancer Society guidelines for the early detection of and II endometrial cancer will have a favor- cancer: update of early detection guidelines for prostate, able prognosis, whereas patients with stage colorectal, and endometrial cancers [published correction III or IV endometrial cancer will have a appears in CA Cancer J Clin. 2001;51(3):150]. CA Cancer J Clin. 2001;51(1):38-75. worse likelihood of survival24 (Table 15,24-26). 5. Buchanan EM, et al. Endometrial cancer. Am Fam Physi- Posttreatment surveillance is recommended cian. 2009; 80(10):1075-1080. for detection of recurrent disease. The Soci- 6. Saso S, et al. Endometrial cancer. BMJ. 2011;343:d3954. ety of Gynecologic Oncology recommends 7. Courneya KS, et al. Associations among exercise, body follow-up symptom surveillance and pel- weight, and quality of life in a population-based sample vic examinations every three to six months of endometrial cancer survivors. Gynecol Oncol. 2005; 97(2):422-430 . for two years posttreatment, then every 8. Calle EE, et. Overweight, obesity, and mortality from six months for three years, and annually cancer in a prospectively studied cohort of U.S. adults. thereafter.3 N Engl J Med. 2003;348(17):1625-1638.

March 15, 2016 ◆ Volume 93, Number 6 www.aafp.org/afp American Family Physician 473 Endometrial Cancer

9. Fisher B, et al. Tamoxifen for prevention of breast can- 22. Elsandabesee D, et al. The performance of Pipelle endo- cer: report of the National Surgical Adjuvant Breast and metrial sampling in a dedicated postmenopausal bleed- Bowel Project P-1 Study. J Natl Cancer Inst. 1998;90(18): ing clinic. J Obstet Gynaecol. 2005;25(1):32-34. 1371-1388. 23. Clark TJ, et al. Accuracy of hysteroscopy in the diagno- 10. Davies C, et al.; Adjuvant Tamoxifen: Longer Against sis of endometrial cancer and hyperplasia: a systematic Shorter (ATLAS) Collaborative Group. Long-term quantitative review. JAMA. 2002;288(13):1610-1621. effects of continuing adjuvant tamoxifen to 10 years 24. Lewin SN, et al. Comparative performance of the 2009 versus stopping at 5 years after diagnosis of oestro- International Federation of Gynecology and Obstetrics’ gen receptor-positive breast cancer: ATLAS, a ran- staging system for uterine corpus cancer. Obstet Gyne- domised trial [published correction appears in Lancet. col. 2010 ;116 (5):1141-1149. 2013;381(9869):804]. Lancet. 2013;381 (9869): 25. Edge SB; American Joint Committee on Cancer. AJCC 805-816. Cancer Staging Manual. 7th ed. New York, NY: Springer; 11. Nelson HD, et al. Systematic review: comparative 2010. effectiveness of medications to reduce risk for primary 26. Pecorelli S. Revised FIGO staging for carcinoma of the breast cancer. Ann Intern Med. 2009; 151(10):703-715, vulva, cervix, and endometrium [published correction W-226-W-235. appears in Int J Gynaecol Obstet. 2010;108(2):176]. Int 12. Lindor NM, et al. Recommendations for the care of individ- J Gynaecol Obstet. 2009;105(2):103-104. uals with an inherited predisposition to Lynch syndrome: 27. Trimble CL, et al.; Society of Gynecologic Oncology Clini- a systematic review. JAMA. 2006;296(12):1507-1517. cal Practice Committee. Management of endometrial 13. Vasen HF, et al.; Mallorca group. Revised guidelines for precancers. Obstet Gynecol. 2012;120 (5):1160 -1175. the clinical management of Lynch syndrome (HNPCC): 28. American College of Obstetricians and Gynecologists. recommendations by a group of European experts. Gut. Supracervical hysterectomy. ACOG Committee Opin- 2013;62(6): 812-823. ion No. 388, November 2007. Obstet Gynecol. 2007; 14. Schmeler KM, et al. Prophylactic surgery to reduce the 110 (5):1215-1217. risk of gynecologic cancers in the Lynch syndrome. N 29. Ghezzi F, et al. Postoperative pain after laparoscopic and Engl J Med. 2006;354(3):261-269. vaginal hysterectomy for benign gynecologic disease: 15. Chin J, et al. Levonorgestrel intrauterine system for a randomized trial. Am J Obstet Gynecol. 2010;203(2): endometrial protection in women with breast cancer on 118.e1-8. adjuvant tamoxifen. Cochrane Database Syst Rev. 2009; 30. Orbo A, et al. Levonorgestrel-impregnated intrauter- (4):CD007245. ine device as treatment for endometrial hyperplasia: a 16. Grady D, et al. Hormone replacement therapy and endo- national multicentre randomised trial. BJOG. 2014;121 metrial cancer risk: a meta-analysis. Obstet Gynecol. (4):477-486. 1995;85(2):304-313. 31. Aalders JG, et al. Endometrial cancer—revisiting the 17. Khati NJ, et al.; Expert Panel on Women’s Imaging. ACR importance of pelvic and para aortic lymph nodes. Gyne- Appropriateness Criteria: abnormal vaginal bleeding. col Oncol. 2007;104(1):222-231. Reston, Va.: American College of Radiology; 2014:1- 32. Kong A, et al. Adjuvant radiotherapy for stage I endo- 13. http://www.guideline.gov/content.aspx?id=48294. metrial cancer. Cochrane Database Syst Rev. 2012;(3): Accessed February 27, 2015. CD003916. 18. American College of Obstetricians and Gynecologists. 33. Podzielinski I, et al. Primary radiation therapy for medi- The role of transvaginal ultrasonography in the evalu- cally inoperable patients with clinical stage I and II endo- ation of postmenopausal bleeding. ACOG Committee metrial carcinoma. Gynecol Oncol. 2012;124(1):36-41. Opinion No. 440, August 2009. Obstet Gynecol. 2009; 34. Barlin JN, et al. Cytoreductive surgery for advanced or 114:409 - 411. recurrent endometrial cancer: a meta-analysis. Gynecol 19. American College of Obstetricians and Gynecologists. Oncol. 2010 ;118 (1):14-18. Management of acute abnormal uterine bleeding in non- 35. Martin-Hirsch PP, et al. Adjuvant progestagens for endo- pregnant reproductive-aged women. ACOG Committee metrial cancer. Cochrane Database Syst Rev. 2011;( 6): Opinion No. 557, April 2013. Obstet Gynecol. 2013; CD001040. 121(4):891-896. 36. Gunderson CC, et al. Oncologic and reproductive out- 20. Smith-Bindman R, et al. Endovaginal ultrasound to comes with progestin therapy in women with endo- exclude endometrial cancer and other endometrial metrial hyperplasia and grade 1 adenocarcinoma: a abnormalities. JAMA. 1998;280(17): 1510-1517. systematic review. Gynecol Oncol. 2012;125(2):477-482. 21. Gupta JK, et al. Ultrasonographic endometrial thickness 37. Eltabbakh GH, et al. Surgical stage, final grade, and sur- for diagnosing endometrial pathology in women with vival of women with endometrial carcinoma whose pre- postmenopausal bleeding: a meta-analysis. Acta Obstet operative endometrial biopsy shows well-differentiated Gynecol Scand. 2002;81(9):799-816. tumors. Gynecol Oncol. 2005;99(2):309-312.

474 American Family Physician www.aafp.org/afp Volume 93, Number 6 ◆ March 15, 2016