AUTHOR AND EDITOR INFORMATION

Section 1 of 8  

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

INTRODUCTION

Section 2 of 8  

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

Background

Carcinoma of the endometrium is among the most common female pelvic malignancies and may develop in normal, atrophic, or hyperplastic endometrium. Most of the cancers are detected at an early stage, with the tumor confined to the uterine corpus in 75% of patients. The prognosis generally is favorable.

Multiple risk factors associated with endometrial cancer include conditions associated with disorders of menstruation, increased perimenopausal bleeding, menopause after than age 52 years, long time period between menarche and menopause, estrogen replacement therapy, tamoxifen therapy for breast cancer, endometrial hyperplasia, obesity, nulliparity, diabetes mellitus, and hypertension. Genetic predisposition appears to play a role since risk factors also include a family history of endometrial or breast cancer and a personal history of ovarian or breast cancer.

For excellent patient education resources, visit eMedicine's Cancer and Tumors Center and Women's Health Center. Also, see eMedicine's patient education articles Cervical Cancer and Menopause.

Pathophysiology

The most frequent risk factor contributing to the development of endometrial cancer is protracted exposure to endogenous or exogenous estrogen that is unopposed by progesterone. Overexposure to endogenous estrogen has been reported in patients with chronic anovulation (eg, secondary to Stein-Leventhal syndrome). Overexposure to estrogen may be the result of an estrogen-producing neoplasm (eg, granulosa or theca cell tumor of the ovary). There may be an association with increased peripheral conversion of androstenedione to estrone in the adipose tissue of obese women.

Tamoxifen is a nonsteroidal drug that has a therapeutic anti-estrogen effect on the breast and an estrogenic effect on the endometrium of postmenopausal women. Patients with breast cancer on prolonged tamoxifen therapy are reported to have increased risk of developing endometrial polyps, hyperplasia, and cancer.

In estrogen-related conditions, endometrial cancer most likely progresses from endometrial hyperplasia, tends to be well differentiated, and is associated with a generally favorable prognosis. In other conditions with unknown cause, the cancer typically develops de novo in the setting of an atrophic or inert endometrium, tends to have more aggressive or undifferentiated cell types, and usually has a poorer prognosis than do the estrogen-related endometrial cancers.

Endometrial cancer usually arises from the glandular component of the endometrium in the upper uterus. It may grow in a focal circumscribed pattern presenting as a friable mass protruding into the endometrial cavity. The cancer may be diffuse, involving multiple regions of the endometrium or the whole endometrial surface. It can occasionally arise within an endometrial polyp.

Spread of the disease occurs initially within the endometrium and/or myometrium, as well as from the fundus towards the isthmic portion of the uterus and the cervix. Progression beyond the uterus occurs through lymphatic pathways into pelvic and abdominal lymph nodes. The cancer eventually may metastasize hematogenously to the lungs and, in decreasing frequency, to the liver, brain, or bones. Transmural or transtubal tumor spread into the peritoneal cavity occurs with more aggressive cancers.

The rate of extrauterine tumor spread and lymph node metastasis increases with the depth of myometrial invasion, the degree of endocervical extension, and the presence of poor prognostic histologic factors. Approximately one half of patients with metastatic lymphadenopathy in the pelvis also have metastasis to the paraaortic lymph nodes. Solitary paraaortic lymph node metastasis rarely occurs.

Histopathologic types

Endometrioid adenocarcinoma: This is the histologic prototype and the most common form of endometrial carcinoma, occurring in as many as 75-80% of affected patients. Tumors vary from well differentiated to undifferentiated and are graded from 1-3 based on architectural and cytologic features. The most aggressive and least differentiated cell types are categorized as grade 3. Well-differentiated tumors occur most commonly and include a low-grade endometrioid cell type, which is usually associated with estrogen-related endometrial hyperplasia. It tends to occur in younger perimenopausal patients. Variants of endometrioid adenocarcinoma, including villoglandular or papillary carcinoma, secretory carcinoma, ciliated carcinoma, and adenocarcinoma with squamous differentiation, also occur.

Uterine serous papillary carcinoma comprises 5-10% of endometrial carcinomas, occurs in women older than those seen with the endometrioid prototype, and commonly arises in atrophic endometrium. This type of carcinoma is aggressive with a predilection for early deep myometrial and vascular invasion that frequently is associated with early extrauterine spread and, in some patients, with transtubal peritoneal dissemination.

Additional aggressive tumors include clear cell carcinoma, comprising 3-5% of endometrial carcinomas, and the less common undifferentiated carcinomas. Similar to the serous papillary variety, clear cell carcinoma and undifferentiated carcinoma tend to occur in older women, have dismal prognoses, and are usually unrelated to protracted unopposed exposure to estrogen.

Miscellaneous rare endometrial carcinomas include mucinous adenocarcinoma, squamous cell carcinoma, mixed cell type carcinoma, postmenopausal-occurring choriocarcinoma, endodermal sinus tumor, small cell carcinoma, and metastatic carcinoma.

Frequency

United States

Cancer of the endometrium is the most common genital malignancy in the United States and ranks as the fourth most common malignancy in women after breast, bronchopulmonary, and colorectal cancers. Approximately 1 in 100 women may develop the disease in the United States. The American Cancer Society estimated 36,100 newly diagnosed US cases in 2000, comprising 6% of the estimated total new cancers in women.

International

Worldwide, cancer of the endometrium ranks as the third most common genital malignancy after cancers of the cervix and ovary. The estimated total number of new cases of cancer of the corpus uteri is 142,000 per year worldwide or 3.7% of cancers in women.

Reported global cancer statistics show the incidence to be highest in North America, followed by Europe and temperate South America. The increased prevalence of the disease in the United States relative to other gynecologic cancers is influenced by the earlier diagnosis and decreasing incidence of cervical cancer, prolonged life expectancy, and dietary factors (perhaps related to increased obesity). The incidence of endometrial carcinoma is low in southern and eastern Asia, as well as in most of Africa.

Mortality/Morbidity

Endometrial carcinoma usually has a favorable prognosis. Good survival rates, 84% in the United States and 72% in Europe, have been reported. However, worldwide, approximately 42,000 deaths occur annually from cancer of the corpus uteri, which represents 1.9% of cancer deaths in women. Approximately 6500 deaths were estimated to occur from this disease in the United States in 2000, which represented 2% of cancer deaths in women.

Race

The prevalence and survival rates of endometrial carcinoma appear to be higher in whites than in blacks. A review of the Gynecologic Oncology Group database showed a relatively higher number of African American patients were older than 70 years at the time of diagnosis and had more aggressive histologic cell types and more advanced local and metastatic disease.

Age

Endometrial carcinoma is primarily a disease of menopausal and postmenopausal women with the peak incidence in women aged 55-65 years. Approximately 75% of patients are aged 50 years and older, and 5% are younger than 40 years. Endometrial carcinoma is rare in patients younger than age 30 years.

Anatomy

The endometrium is the mucosal lining of the uterine cavity. It consists of a columnar surface epithelium overlying a stroma of round to ovoid cells intermixed with glands and a characteristic vascular system. The glands open into the surface epithelium and are surrounded by a network of fibrillar elements.

The thickness and sonographic appearance of the normal endometrium vary with the menstrual cycle. Endometrial thickness, as reported in the ultrasound (US) literature, is measured on a midline sagittal image of the uterus and is a summation of the anteroposterior (AP) width of both the anterior and posterior endometrial layers, exclusive of possible intracavitary content.

During menses, the endometrium is thin, patchy, and not clearly delineated. In the proliferative phase, the endometrial stripe develops a multilayered sonographic appearance; the wider inner portion of the proliferative endometrium is hypoechoic and the outer portion is relatively hyperechoic. In the secretory phase, the endometrial stripe thickens to a mean of 14-16 mm in AP width and becomes diffusely hyperechoic secondary to accumulation of mucus and glycogen in the increasingly tortuous glands.

The endometrium progressively atrophies following menopause. The rate at which this process occurs is influenced by the variable presence of adrenal sources of genital hormones and by the degree of ovarian activity. Many authorities will report as abnormal measurements of 5 mm or greater in postmenopausal patients not on hormone replacement therapy (HRT). Some will allow as great as 8-mm thickness if the patient is on HRT. Others will limit the AP measurement, even in that group, to up to 5 mm.

CT does not depict the endometrium consistently and is not reliable for accurate evaluation of its thickness. Immediate postcontrast dynamic CT scans of the uterus often show central hypo-attenuation that may be related to secretions in the cavity or a lag in contrast enhancement of the endometrium compared to myometrium; however, the endometrium is not visualized distinctly as separate from the myometrium, and accurate measurement of its thickness is not feasible. This is because the endometrium and myometrium have similar attenuation and cannot be distinguished either on CT scans obtained without intravenous contrast or on routine or delayed postcontrast CT scans.

MRI depicts the endometrium as a central zone of high signal intensity on T2-weighted images while the myometrium is depicted at its inner aspect as a zone of low signal intensity (junctional zone) and at its outer aspect as a wider zone of intermediate signal intensity. On T1-weighted images, the endometrium has intermediate signal intensity similar to the myometrium; therefore, the endometrium is not visualized distinctly as separate from the myometrium.

As measured on MRI, endometrial thickness reportedly is almost always less than that measured on US. Endometrial thickness varies in menstruating women from 4 mm in the early proliferative phase to 13 mm in the late secretory phase. The upper limit of normal thickness in the asymptomatic postmenopausal women has been suggested at 8 mm, regardless of hormonal therapy.

Clinical Details

Early endometrial cancer is usually asymptomatic. Eventually, 80% of patients present with vaginal bleeding, mostly postmenopausal. Ten percent of patients present with purulent vaginal discharge, which sometimes is tinged with blood.

In patients with endometrial cancer, 5% or fewer cases are diagnosed while the patient still is asymptomatic, with the cancer being discovered after a hysterectomy is performed for benign indications or during a diagnostic workup for abnormal Papanicolaou smear results. Pain and pelvic pressure are usually manifestations of advanced disease.

Although only 10-20% of postmenopausal vaginal bleeding is the result of gynecologic malignancy, the probability that it is caused by endometrial carcinoma progressively increases with age.

Preferred Examination

Medical procedures

Endometrial biopsy, usually using an aspiration-type curet or other device, is generally accepted as the first-step office procedure for the diagnosis of endometrial cancer and should be coupled with endocervical curettage. The procedure is definitive if results are positive for malignancy. The reported accuracy of the procedure is approximately 90%.

When endometrial sampling does not yield a conclusive histologic diagnosis and when adequate evaluation is precluded by cervical stenosis or by limited patient tolerance, traditional fractional curettage may be necessary because it provides the largest amount of endometrial tissue. This procedure is performed under anesthesia in the operating room. The procedure includes initial circumferential scraping of the endocervical canal and subsequent systemic curettage of the entire endometrial surface. Cervical and endometrial specimens are evaluated separately to determine if endometrial cancer invaded the cervix and to allow detection of occult endocervical cancer.

Endometrial curettage may be falsely negative in 2-6% of cases because the endometrium is sampled randomly and incompletely. Some clinicians advocate hysteroscopy with biopsy as the preferred procedure following negative biopsy results, because this allows the surgeon to perform the biopsy directly on focal abnormalities that may be missed during curettage. Other clinicians caution against the use of hysteroscopy unless absolutely necessary, since cancer cells may be pushed through the fallopian tubes into the peritoneal cavity.

The thickness of the endometrial stripe, as measured by US, has been advocated as a factor in determining the need for dilatation and curettage. Although considerable overlap exists in the endometrial thickness and appearance between the various benign and malignant histologic types, several reports suggest that the presence of a thin stripe is usually associated with atrophic endometrium and may indicate that a histologic diagnosis is not necessary.

In 1991, Granberg et al reported that if a cutoff of 5-mm was used to define the thickness of the normal endometrial stripe in symptomatic postmenopausal patients, approximately 70% of patients with vaginal bleeding could have avoided curettage. In 1995, a multicenter study reported by Karlsson et al revealed that dilatation and curettage found cancer in 2 of 88 patients with a 5-mm stripe, while cancer was detected in none of 518 women with a stripe of 4-mm or less; therefore, the authors concluded that a 4-mm cutoff may be more appropriate. Women on estrogen replacement therapy typically may have endometrial stripe thickness measuring up to 8-10 mm.

No consensus has been reached regarding the cutoff for the US measurement of endometrial thickness that definitively eliminates the need for histologic evaluation of patients with abnormal uterine bleeding. In addition, many clinicians still prefer endometrial sampling as the initial diagnostic procedure used in evaluating symptomatic postmenopausal patients.

Radiologic procedures

US is the modality of choice for the initial imaging evaluation of female pelvic organs. US is widely available in many regions of the world, relatively inexpensive, noninvasive, and does not use ionizing radiation. Typical examinations include transabdominal sonography (TAS) and transvaginal sonography (TVS), which are supplemented by color Doppler imaging as needed.

TAS is performed through subcutaneous fat and abdominal wall muscles and uses the full urinary bladder as an acoustic window. TAS transducers, needed in most patients to penetrate the abdominal wall and adequately visualize pelvic organs, have lower frequency and resolution than TVS probes.

TVS has the advantage of using high-frequency transducers that are placed close to the regions of interest and produce high-resolution images of significantly better quality than transabdominal images. While both TAS and TVS allow visualization of the endometrium, exquisitely finer endometrial details are possible to depict transvaginally rather than transabdominally.

TVS is clinically established as the preferred technique for evaluation of endometrial disorders and is especially useful in the workup of abnormal uterine bleeding. Hysterosonography can be used to identify the cause of endometrial stripe thickening in some patients. The procedure consists of TVS performed with sterile fluid placed within the endometrial cavity and may help show a thick endometrial stripe as secondary to diffuse or focal endometrial thickening, endometrial polyp, submucosal leiomyoma, or synechiae. This may help further diagnostic planning.

TVS is superior to CT and approaches MRI in its ability to depict endometrial carcinoma and to provide information regarding myometrial, cervical and, perhaps, parametrial tumor invasion. However, US is unable to depict the entire intrapelvic or intraabdominal anatomic regions adequately; therefore, US is not suitable for the comprehensive staging of endometrial carcinoma. US has significantly lower sensitivity than CT in detecting enlarged abdominal or pelvic lymph nodes and in depicting intraperitoneal, omental, or mesenteric metastases. In addition, US is inferior to CT in assessing pelvic sidewall extension and adjacent organ invasion.

CT and MRI are more accurate staging modalities than US. Both techniques allow survey of the entire pelvis, abdomen, thorax, and brain. CT is available more widely, is less costly than MRI, provides rapid image acquisition, and has high spatial resolution. The advantages of CT also include the availability of GI and intravenous (IV) contrast materials. Opacification of the GI tract with oral and rectal contrast facilitates optimal evaluation of the bowel and helps distinguish intraperitoneal and retroperitoneal masses from bowel. IV contrast injection improves evaluation of vascular structures and detection of mass lesions in parenchymatous organs. The recent advent of spiral/helical and multidetector technology has improved the multiplanar capability of CT.

The advantages of MRI include superior spatial and tissue contrast resolution, multiplanar capabilities, lack of exposure to ionizing radiation, and availability of noniodinated, nonnephrotoxic IV contrast material.

Preferred staging modality

Histopathologic features of the tumor and clinical findings at presentation influence the choice of imaging modality for preoperative staging of endometrial cancer. Kinkel et al provided clinical practice guidelines for staging based on a meta-analysis of the usefulness of MRI, CT, and US in imaging patients with endometrial cancer.

• Patients with grade 1 tumor, a clinically normal-sized uterus, and no clinical evidence of coexisting pelvic disease generally require no preoperative imaging because the risk for myometrial, cervical, or lymph node disease is low. If the clinical evaluation is inconclusive or coexisting pelvic disease is suggested, then US, CT, or MRI may be used for the initial imaging evaluation.
• In patients at risk for disease dissemination and lymph node involvement at presentation (because of tumor grade, histologic cell type, or clinical findings) CT or MRI of the abdomen and pelvis should be performed to determine the extent of tumor spread.
• Patients in whom cervical invasion is suggested clinically or in whom endocervical curettage was inconclusive benefit in particular from MRI, because MRI can depict cervical and myometrial invasion most accurately and is approximately equivalent to CT in detecting enlarged lymph nodes.

MRI, with its exquisite soft tissue contrast and multiplanar capability, is superior to US and CT in helping assess the depth of myometrial invasion, cervical invasion, and early parametrial invasion. MRI is approximately equivalent to CT in detecting enlarged lymph nodes, but CT is considerably superior to MRI in detecting and distinguishing intraperitoneal, omental, and mesenteric metastases from bowel.

Although MRI is superior to CT in evaluating myometrial and cervical invasion and is the best alternative for patients with significant contrast allergies or renal malfunction, CT is more sensitive than MRI in the overall detection of tumor spread outside the uterus. In addition, CT remains the imaging modality used most frequently in clinical practice for comprehensive preoperative evaluation of the extent of disease.

CT is clinically advocated in the evaluation of patients with poorly differentiated or high-grade tumor, serous papillary carcinoma, or clear cell carcinoma because of the high risk for advanced disease and metastatic lymphadenopathy at the time of presentation. CT also is advised for patients who have abnormal liver function test results, elevated serum cancer antigen 125 levels, clinical suggestion of advanced disease, or inconclusive clinical evaluation.

Limitations of Techniques

US is operator-dependent, has relatively poor spatial and tissue contrast resolution compared with MRI and CT, image quality is degraded by large body habitus, and visualization of portions of the pelvis and abdomen is precluded by bowel gas and bony structures. The transabdominal approach also is influenced by the degree of bladder filling and is impeded by the presence of surgical incision, dressings, drains, or skin lesions. Transvaginal probes have inherent limitations including small field of view, short range of penetration of high-frequency transducers, and occasional patient intolerance or lack of acceptance of the transvaginal approach.

CT uses ionizing radiation and has inferior soft tissue contrast resolution, making it less capable than MRI of distinguishing between tumor and normal soft tissues in the uterine corpus and cervix. CT image quality is degraded by metallic prostheses, an extremely large body habitus, and patient or respiratory motion. The iodinated IV contrast available for CT is associated with a risk of significant allergic reactions (including fatal anaphylaxis), nephrotoxicity, and complications of contrast extravasation.

MRI is contraindicated in patients who have vital metallic biomedical devices or metallic objects in strategic anatomic regions. It is more costly and less readily available than CT and requires long image acquisition times. MRI image quality is degraded by artifacts related to respiratory motion and bowel peristalsis, which are likely to occur during the long image acquisition time. No effective GI contrast material is currently available for MRI. Claustrophobia deters some patients from undergoing MRI.

Gadolinium-based contrast agents (gadopentetate dimeglumine [Magnevist], gadobenate dimeglumine [MultiHance], gadodiamide [Omniscan], gadoversetamide [OptiMARK], gadoteridol [ProHance]) have recently been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD). For more information, see the eMedicine topic Nephrogenic Fibrosing Dermopathy. The disease has occurred in patients with moderate to end-stage renal disease after being given a gadolinium-based contrast agent to enhance MRI or MRA scans. As of late December 2006, the FDA had received reports of 90 such cases. Worldwide, over 200 cases have been reported, according to the FDA.

NSF/NFD is a debilitating and sometimes fatal disease. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness. For more information, see the FDA Public Health Advisory or Medscape.

DIFFERENTIALS

Section 3 of 8  

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

Other Problems to be Considered

Endometrial hyperplasia
Endometrial polyp
Tamoxifen-related endometrial changes
Hematometra or pyometra
Endometrial sarcoma
Myometrial malignancy

CT SCAN

Section 4 of 8  

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

Findings

CT is the imaging modality used most commonly in clinical practice to determine the extent of spread of endometrial cancer. Oral, rectal, and IV administration of contrast material is necessary for optimal CT evaluation. IV contrast injection is particularly useful to increase the conspicuity of the endometrial tumor and to facilitate the evaluation of myometrial and cervical invasion.

Endometrial carcinoma and a normal myometrium demonstrate approximately similar attenuation on CT images obtained without IV contrast enhancement; therefore, the 2 cannot be reliably distinguished on noncontrast CT images. Following IV contrast administration, nonuniform contrast enhancement of the tumor may occur but to a much lesser degree than the usually intense and uniform enhancement of the normal myometrium. Thus, the endometrial tumor may become apparent on postcontrast images as a lesion with relatively low attenuation.

In a prospective study of 10 patients reported by Hamlin et al in 1980, none of the endometrial cancers were diagnosed unequivocally on precontrast CT images, and 30% of the cancers were not diagnosed on postcontrast images (even retrospectively in both situations). In 24 patients with primary endometrial malignancy reported by Walsh and Goplerud in 1982, CT showed normal uterine attenuation findings in 50% of patients and 1- to 5-cm low-attenuation tumor areas in the uterus in the other 50% of patients. Dimensions of the identified tumors were not evaluated accurately because of volume averaging and the uterine position relative to the axial plane of CT imaging.

CT findings in endometrial cancer include the following:

• Relatively hypoattenuated mass in the region of the endometrial cavity (see Images 3-4), which may show uniform attenuation or may be heterogeneous, with or without a contrast-enhanced component
• Polypoid mass surrounded by endometrial fluid
• Heterogeneous soft tissue mass/masses and fluid expanding the endometrial cavity (see Images 5-8)
• Tumor involving multiple regions of the endometrium or the entire endometrial surface (see Images 9-10)
• Fluid-filled uterine cavity marginated by mural tumor implants (see Image 11)
• Fluid-filled uterine cavity secondary to obstruction of the endocervical canal by tumor that is not depicted or delineated clearly (see Images 12-13)

CT staging

CT staging of endometrial cancer is based on the 1988 surgical classification of the International Federation of Gynecology and Obstetrics (FIGO).

• Stage I - Cancer confined to the uterine corpus
• Stage II - Cancer involving the corpus and cervix, without extrauterine spread
• Stage III - Cancer extending outside the uterus but confined to the true pelvis
• Stage IV - Cancer invading the bladder or bowel mucosa and/or spreading outside the true pelvis
• In addition, each stage is categorized further as follows:
  • Stage IA: The tumor is limited to the endometrium and the CT appearance of the uterus may be essentially normal (particularly in patients with noninvasive cancer), or the tumor can be depicted as described in CT Findings.
  • Stage IB: Tumor extends into less than one half the width of the myometrium.
  • Stage IC: Tumor extends into one half or more of the myometrial width.

    Invasion of the myometrium results in eccentric or focal myometrial thinning. The cancer usually appears hypoattenuated relative to the surrounding uniformly contrast-enhanced normal myometrium, and the demarcation between the invasive tumor and the thinned myometrium is ill-defined (see Images 5-7, Images 9-10, Image 14).

  • Stage IIA: In stage II, the cancer extends from the uterine corpus into the cervix, involving only the endocervical glandular region in stage IIA.
  • Stage IIB: The fibromuscular stroma of the cervix is involved.

    Cervical invasion limited to the endocervical glandular region is not usually identifiable on CT. Tumor involvement of the cervical stroma typically results in an enlarged cervix, which shows areas of heterogeneous hypoattenuation or is hypoattenuated diffusely (see Image 8, Image 15).

  • Stage IIIA: Cancer extends outside the uterus into the parametria, pelvic sidewall, fallopian tube, or ovary.

    CT features of parametrial and pelvic sidewall invasion include the following (in order of increasing positive predictive value):

    • Loss of definition or irregularity of the uterine contours
    • Increased attenuation (see Images 9-10) and prominent linear soft tissue stranding in the parametrial and periureteral fat
    • Confluent soft tissue replacing the periureteral fat
    • Three-dimensional parametrial soft tissue mass (see Image 14)
  • Stage IIIB: Vaginal metastases characterize this stage.
  • Stage IIIC: Enlarged pelvic and/or para-aortic lymph nodes characterize this stage.
  • Stage IVA: Tumor invasion spreads into the urinary bladder or bowel mucosa. CT findings include the following:
    • Focal obliteration of the perivesical or perirectal fat
    • Eccentric or asymmetric wall thickening, which can be uniform, nodular, or serrated
    • Intraluminal extension of soft tissue mass
  • Stage IVB: Metastases outside the true pelvis characterize this stage (see Image 16), including metastasis into the intra-abdominal and/or inguinal lymph nodes.

Degree of Confidence

The reported overall accuracy of CT staging ranges from 84-88%. CT scans help confirm stage I or II disease correctly in 83-92% of patients and help identify extrauterine tumor extension correctly in 83-86% of patients with stage III or IV disease.

CT limitations in staging endometrial carcinoma include the following:

• Understaging cancers presenting with microscopic tumor spread
• Diminished accuracy in evaluating myometrial invasion in elderly women, particularly in the setting of atrophic myometrium and polypoid endometrial tumors
• Overlap with the appearances of other malignant uterine masses and leiomyomata
• Similarity in the appearances of the irregular uterine contours and adjacent soft tissue stranding produced by parametrial tumor invasion and by parametritis secondary to fractional curettage or endometrial biopsy
• Inability to differentiate lymph node enlargement resulting from benign causes from that resulting from metastases

False Positives/Negatives

CT findings are not specific for endometrial carcinoma and can be simulated by other conditions including the following:

• Endometrial extension of cervical cancer
• Endometrial polyp
• Leiomyomata
• Intrauterine fluid collections, including blood and/or necrotic material (eg, from recent biopsy or curettage)

MRI

Section 5 of 8  

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

Findings

State-of-the-art MRI requires the use of a high field-strength system with a torso phased-array coil. Administering antiperistaltic agents such as glucagon to patients is advocated. High-resolution sagittal and axial T2-weighted fast spin-echo sequences are obtained in all patients and usually depict the uterine zonal anatomy to greatest advantage. Axial T1-weighted spoiled gradient-echo images also are obtained and are helpful to detect enlarged pelvic lymph nodes. Sagittal T1-weighted spoiled gradient-echo sequences (with or without fat suppression) may be performed following IV injection of paramagnetic contrast material if a need exists to improve tumor detection, distinguish tumor from debris in the endometrial cavity, or facilitate the evaluation of myometrial invasion by increasing the contrast between the tumor and the normal myometrium.

MRI reveals variable abnormal endometrial findings in 81-84% of patients with endometrial carcinoma. The endometrium may be thickened focally or diffusely, demonstrated as irregular in thickness and configuration, or widened by polypoid tumor (see Images 17-20). Furthermore, the signal intensity of the tumor has variable patterns on T1-weighted and T2-weighted images. However, the endometrium may appear entirely normal.

A review of the findings depicted by unenhanced MRI of 45 patients with endometrial cancer, as reported by Hricak et al, revealed the following:

• No abnormal endometrial signal in 15.6% of patients, although hysterectomy specimens showed tumors measuring as large as 1-3 cm.
• No abnormal signal in 31.1% of patients on T1-weighted images, while T2-weighted images depicted the tumor with a uniform signal intensity lower than adjacent normal endometrium.
• No abnormal signal in 20% of patients on T1-weighted images, while T2-weighted images showed heterogeneous tumor with low and high signal intensities compared to normal endometrium.
• In 15.6% of patients, endometrial blotches of medium or high signal intensity were noted on T1-weighted images, while on T2-weighted images, the tumor was of uniformly high signal intensity.
• In 17.8% of patients, tumor was isointense with normal endometrium, and the only abnormal finding was a measurably thickened endometrium.

Thus, unenhanced T1-weighted MRI demonstrated endometrial tumor as follows:

• Isointense with the remainder of the uterus in 84.4% of patients
• Having medium or high signal intensity in 15.6% of patients

Unenhanced T2-weighted MRI demonstrated endometrial tumor as follows:

• Isointense with the normal endometrium in 49% of patients
• Having uniform signal intensity lower than normal endometrium in 31.1% of patients
• Heterogeneous with low and high signal intensity in 20% of patients

Endometrial tumor was isointense with the surrounding endometrium on both T1-weighted and T2-weighted images in 33.4% of patients.

MRI staging

Similar to CT staging, MRI staging of endometrial carcinoma is based on the surgical FIGO classification of 1988. Overall accuracy is reported to be as high as 85-92%.

• Stage IA: The tumor is confined to the endometrium, and the endometrial-myometrial interface is continuous and smooth. When visible, the myometrial junctional zone is preserved and intact.
• Stage IB: The tumor extends to less than one half of the myometrial width, resulting in irregularity and indistinctness of the endometrial-myometrial interface, segmented disruption or discontinuity of the myometrial junctional zone when present, and/or high signal intensity of tumor evident in the inner half of the myometrium.
• Stage IC: The tumor extends to one half or more of the myometrial width, and the uterine serosal surface is preserved and intact.
• Stage II: The tumor extends into the cervix, resulting in widening of the internal cervical os or the endocervical canal associated with heterogeneous signal intensities that also may involve the cervical fibrous stroma.
• Stage IIIA: Disruption of the uterine serosal surface is demonstrated, which should be confirmed in 2 planes. Extrauterine extension of the tumor is demonstrated into the parametria, pelvic sidewall, fallopian tube(s), or ovary.
• Stage IIIB: Vaginal metastases characterize this stage and may be evidenced by segmental loss of the hypointense vaginal wall signal.
• Stage IIIC: Enlarged pelvic and/or para-aortic lymph nodes characterize this stage.
• Stage IVA: Invasion of the bladder or bowel mucosa is demonstrated, resulting in disruption of the tissue planes between the uterus and bladder or rectum and focal alteration or loss of low signal intensity of the walls of these structures.
• Stage IVB: This stage is manifested by metastatic masses outside the true pelvis or by enlarged intra-abdominal and/or inguinal lymph nodes.

Degree of Confidence

MRI reveals variable abnormal endometrial findings only in approximately 81-84% of patients with endometrial carcinoma.

Overall accuracy of MRI staging is reported to be as high as 85-92%. MRI limitations in staging endometrial carcinoma include the following:

• Understaging cancers presenting with microscopic tumor spread
• Difficulty in consistently differentiating endometrial cancer from benign pathologic findings
• Diminished accuracy in evaluating myometrial invasion in the setting of polypoid tumor or blood distending the uterine cavity, resulting in the appearance of a thin myometrium
• Difficulty in assessing the depth of myometrial invasion in elderly/postmenopausal patients with an ill-defined or nonvisible myometrial junctional zone or in those who have atrophic myometrium
• Relatively low accuracy in identifying metastases to lymph nodes, peritoneum, or adnexal structures

False Positives/Negatives

MRI findings of endometrial cancer are not specific and can be simulated by the following:

• Adenomatous hyperplasia
• Polyp(s)
• Degenerating submucosal leiomyoma
• Endometrial secretions
• Blood clots

IV injection of paramagnetic contrast material may result in enhancement of the endometrial tumor, thus may improve tumor detection and allow differentiation between the contrast-enhanced tumor and the unenhanced blood clots or fluid in the endometrial cavity.

ULTRASOUND

Section 6 of 8  

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

Findings

Transvaginal sonography is advocated as the imaging modality of choice for screening for endometrial cancer. US findings are as follows:

• The earliest depicted finding is a thickened endometrium that is usually heterogeneous and poorly defined or has irregular contours. Although echogenicity is variable, in most patients the endometrium is either diffusely or partially more echogenic than the myometrium. These findings are nonspecific as there is considerable overlap in the US appearances and thickness of the endometrium in patients with carcinoma, hyperplasia, or polyp. Histopathologic evaluation is necessary to determine the diagnosis.
• The endometrium is typically measured at its maximal thickness on a midline sagittal image of the uterus obtained by transvaginal US. Endometrial thickness, as reported in the literature, is a bilayer measurement combining the AP width of both the anterior and posterior layers of the endometrium, exclusive of possible intracavitary content. The reported normal range for this bilayer endometrial thickness is up to 5 mm in asymptomatic postmenopausal women not receiving hormones and up to 8 mm in asymptomatic postmenopausal women on HRT. Some advocate calling all measurements greater than 5 mm abnormal with or without HRT.
• To minimize the risk of missing premalignant or malignant lesions, most experts advocate endometrial sampling when the US endometrial thickness is greater than 8 mm in asymptomatic postmenopausal women or 5 mm or greater in postmenopausal patients having abnormal uterine bleeding, regardless of whether the women are receiving hormones. Abnormal uterine bleeding in women with endometrial thickness of less than 5 mm has been thought to be almost always associated with endometrial atrophy; therefore, a histologic diagnosis may not be necessary in such patients. Some clinicians still prefer performing endometrial sampling as the initial diagnostic procedure for all symptomatic postmenopausal patients, without attempting to determine the endometrial thickness. Increasing work with hysterosonography reveals a greater number of cases having true causes for the hemorrhage other then "endometrial atrophy."
• Women on sequential estrogen and progesterone therapy develop cyclic endometrial changes resulting in considerable variation of the endometrial thickness, which is at its maximal dimension on days 13-23 of the cycle. The endometrium in these women should be evaluated at the beginning or at the end of the sequential hormone cycle when the endometrium is likely to be at its thinnest dimension. Women on unopposed estrogen therapy, continuous estrogen and progesterone therapy, or tamoxifen therapy do not undergo as much cyclic endometrial variation, and the timing of the measurement of endometrial thickness is not critical.
• Hysterosonography may help determining further diagnostic planning in some patients, particularly in the setting of thickened endometrium and negative office biopsy. The procedure, consisting of transvaginal sonography performed with sterile fluid placed in the uterine cavity, may show a thick endometrial stripe as secondary to diffuse endometrial thickening, focal endometrial thickening, polyp or other polypoid endometrial mass, or submucosal leiomyoma or other sessile or broad-based mass. Diffusely thickened endometrium can be sampled by office biopsy or fractional curettage. Focal endometrial thickening and polypoid endometrial mass are candidates for sampling with the aid of hysteroscopy. However, some clinicians caution against the use of hysteroscopy unless absolutely necessary because of the potential for cancer cells to be pushed through the fallopian tubes into the peritoneal cavity. Whether this risk applies to a similar degree to hysterosonography remains to be determined.
• In 25 cases of endometrial cancer reviewed by Atri et al, the findings were as follows:
  • Tumors were hyperechoic relative to the myometrium in 76%, isoechoic in 12%, and contained both hypoechoic and hyperechoic regions in 12% of patients.
  • Approximately 40% of patients in this series had well-defined echogenic endometrial thickening that was not distinguishable from benign endometrial abnormalities such as hyperplasia or polyp.
  • Cystic changes within thickened endometrium, previously suggested to be indicative of benign pathology (eg, hyperplasia or polyp), were found in 24% of patients with endometrial cancer.
• Other sonographic findings include the following:
  • Broad-based polypoid mass expanding the endometrial cavity with or without surrounding fluid
  • Endometrial polypoid mass protruding through the endocervical canal
  • Large mass centrally in the uterus replacing the endometrial stripe (see Images 1-2)
  • Complex fluid collection in the endometrial cavity (secondary to malignant stenosis of the endocervical or endometrial canal) that may represent pyometra or hematometra
  • Calcifications (uncommon in endometrial cancer)
• TVS is useful for the evaluation of myometrial invasion by endometrial carcinoma. The demonstration of partial or total disruption of the hypoechoic subendometrial halo, formed by the compact and vascular inner myometrial region, usually reflects the presence of myometrial invasion, although depiction of intact subendometrial halo may be associated with superficial invasion or lack of invasion. The depth of myometrial invasion can be estimated by measuring, on sagittal images, the thickness of the tumor extending beyond the outer margin of the endometrium and dividing it by the entire thickness of the adjacent intact myometrium.
• Patients with breast cancer on prolonged tamoxifen therapy have increased risk of developing endometrial polyps, hyperplasia, and cancer. Some of these patients may actually have mixed histopathologic abnormalities. The most common endometrial US abnormality associated with tamoxifen is a thickened endometrium showing multiple cystic foci. Other findings include homogeneous hyperechoic endometrial thickening and heterogeneous endometrial thickening. Tamoxifen-related endometrial changes are nonspecific and overlap in appearance with findings encountered in hyperplasia, polyps, and carcinoma detected in postmenopausal patients receiving tamoxifen or hormonal replacement therapy as well as in patients not receiving therapy.

Degree of Confidence

TVS is excellent for the evaluation of the endometrium and the contents of the endometrial cavity. TVS is superior to CT and approaches MRI in its ability to provide information regarding myometrial, cervical and, perhaps, parametrial invasion by endometrial carcinoma. US has significantly lower sensitivity than CT in detecting enlarged abdominal and pelvic lymph nodes (particularly in the obese patient) and in depicting intraperitoneal, omental, and mesenteric metastases. CT also is superior to US in assessing pelvic sidewall extension and adjacent organ invasion.

No US feature can differentiate reliably between malignant and benign endometrial pathology in the absence of identifiable myometrial invasion. The reported overall accuracy of US in detecting myometrial invasion and distinguishing between superficial and deep myometrial invasion by endometrial carcinoma ranges from 69-90%. The reported sensitivity and specificity of US evaluation of the depth of myometrial invasion are 50-100% and 65-100%, respectively. The reported sensitivity and specificity of US evaluation of cervical invasion by endometrial carcinoma are 66.7-80% and 95.2-100%, respectively.

US limitations in the evaluation of endometrial carcinoma include the following:

• Understaging cancers presenting with microscopic tumor spread
• Difficulty consistently differentiating endometrial cancer from benign pathologic findings
• Overestimation of the degree of myometrial invasion when a large exophytic endometrial tumor or a large amount of hemorrhagic debris expands the uterine cavity and considerably effaces or stretches the overlying myometrium
• Interpretation errors in the assessment of myometrial invasion in patients who lack a demonstrable subendometrial halo, in elderly patients who typically have a thin myometrium, and in the setting of a leiomyomatous uterus
• Diminished accuracy in assessing pelvic sidewall extension and adjacent organ invasion
• Low sensitivity in detecting enlarged lymph nodes and in depicting extrapelvic metastasis

False Positives/Negatives

US findings of endometrial cancer are not specific and can be simulated by the following:

• Adenomatous hyperplasia
• Polyp(s)
• Tamoxifen-related endometrial changes
• Degenerating submucosal leiomyoma
• Endometrial extension of cervical cancer
• Blood clots

MULTIMEDIA

Section 7 of 8  

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

Media file 1:  A 76-year-old woman with poorly differentiated endometrial adenocarcinoma. Sagittal transvaginal ultrasound image of the uterus shows a central mass replacing the endometrial stripe, with hyperechoic and hypoechoic regions.
	View Full Size Image
Media type:  Image

Media file 2:  A 76-year-old woman with poorly differentiated endometrial adenocarcinoma (same patient as Images 1, 3, and 4). Transverse transvaginal ultrasound image of the uterus shows a central mass replacing the endometrial stripe, with hyperechoic and hypoechoic regions.
	View Full Size Image
Media type:  Image

Media file 3:  A 76-year-old woman with poorly differentiated endometrial adenocarcinoma (same patient as Images 1, 2, and 4). CT image shows a relatively hypoattenuated mass in the region of the endometrial cavity. Diffuse myometrial thinning is evident. Surgical pathology revealed approximately 4.0 cm of pedunculated endometrial tumor associated with only superficial myometrial invasion (limited to inner one third).
	View Full Size Image
Media type:  CT

Media file 4:  A 76-year-old woman with poorly differentiated endometrial adenocarcinoma (same patient as Images 1, 2, and 3). CT image shows a relatively hypoattenuated mass in the region of the endometrial cavity. Diffuse myometrial thinning is evident. Surgical pathology revealed approximately 4 cm of pedunculated endometrial tumor associated with only superficial myometrial invasion (limited to inner one third).
	View Full Size Image
Media type:  CT

Media file 5:  A 63-year-old woman with moderately differentiated endometrial adenocarcinoma, nuclear grade 3. CT image through the upper uterus shows tumor masses and minimal fluid expanding the endometrial cavity and causing marked myometrial thinning. Surgical pathology revealed predominantly exophytic polypoid tumor in the endometrial cavity and full-thickness myometrial invasion without extension to the serosal surface.
	View Full Size Image
Media type:  CT

Media file 6:  A 63-year-old woman with moderately differentiated endometrial adenocarcinoma, nuclear grade 3 (same patient as Image 5). CT images through the upper uterus show tumor masses and minimal fluid expanding the endometrial cavity and causing marked myometrial thinning. Surgical pathology revealed predominantly exophytic polypoid tumor in the endometrial cavity and full-thickness myometrial invasion without extension to the serosal surface.
	View Full Size Image
Media type:  CT

Media file 7:  A 64-year-old woman with poorly differentiated endometrioid adenocarcinoma of the endometrium with focal squamous differentiation. CT image shows tumor expanding the entire endometrial cavity. The myometrium is markedly thinned and its inner margins are ill defined. Surgical pathology confirmed tumor involvement of the cervix and the outer third of the myometrium.
	View Full Size Image
Media type:  CT

Media file 8:  A 64-year-old woman with poorly differentiated endometrioid adenocarcinoma of the endometrium with focal squamous differentiation (same patient as Image 7). CT image shows tumor expanding the cervix. Surgical pathology confirmed tumor involvement of the cervix.
	View Full Size Image
Media type:  CT

Media file 9:  A 45-year-old woman with poorly differentiated endometrial carcinoma of mixed cell type. CT image shows tumor replacing the entire endometrial cavity and extending into the endocervix. The myometrium is predominantly thinned at the fundus and its inner margins are ill defined. The serosal surface of the uterus is surrounded by soft tissue attenuation suggestive of parametrial tumor extension. Hysterectomy performed after chemotherapy confirmed tumor involvement of the lower uterine segment, the endocervix, and the outer third of the myometrium.
	View Full Size Image
Media type:  CT

Media file 10:  A 45-year-old woman with poorly differentiated endometrial carcinoma of mixed cell type (same patient as Image 9). CT image shows tumor replacing the entire endometrial cavity and extending into the endocervix. The myometrium is predominantly thinned at the fundus and its inner margins are ill defined. The serosal surface of the uterus is surrounded by soft tissue attenuation suggestive of parametrial tumor extension. Hysterectomy performed after chemotherapy confirmed tumor involvement of the lower uterine segment, the endocervix, and the outer third of the myometrium.
	View Full Size Image
Media type:  CT

Media file 11:  A 73-year-old woman with well-differentiated endometrioid endometrial adenocarcinoma. CT image through the uterus shows fluid-filled cavity marginated by tumor involving most of the endometrial surface and the endocervical region. An enlarged lymph node is evident at the right pelvic sidewall.
	View Full Size Image
Media type:  CT

Media file 12:  A 74-year-old woman with moderately differentiated endometrioid adenocarcinoma with squamous differentiation. CT image shows a markedly distended endometrial cavity containing soft tissue attenuation material in the dependent region and fluid superiorly, resulting in a fluid/debris level. The myometrium in the fundus is thin but well defined. Surgical pathology revealed hemorrhagic fluid in the distended endometrial cavity and stenosis at the junction of the endocervical canal and the endometrium secondary to the endometrial tumor predominantly involving the lower uterine segment and extending deep into the cervical stroma.
	View Full Size Image
Media type:  CT

Media file 13:  A 74-year-old woman with moderately differentiated endometrioid adenocarcinoma with squamous differentiation (same patient as Image 12). CT image shows a markedly distended endometrial cavity containing soft tissue attenuation material in the dependent region and fluid superiorly, resulting in a fluid/debris level. The myometrial and endometrial regions in the lower uterus are ill defined and slightly heterogeneous. Surgical pathology revealed hemorrhagic fluid in the distended endometrial cavity and stenosis at the junction of the endocervical canal and the endometrium secondary to the endometrial tumor predominantly involving the lower uterine segment and extending deep into the cervical stroma.
	View Full Size Image
Media type:  CT

Media file 14:  A 57-year-old woman with stage IVB poorly differentiated endometrial carcinoma. CT images through the uterus and the lower thorax (same patient as Images 15 and 16). The upper uterus shows confluent soft tissue masses and fluid expanding the endometrial cavity, diffuse myometrial thinning consistent with deep myometrial invasion, and parametrial tumor extension through the disrupted myometrium at the right fundal region.
	View Full Size Image
Media type:  CT

Media file 15:  A 57-year-old woman with stage IVB poorly differentiated endometrial carcinoma (same patient as Images 14 and 16). CT image through the uterus. The cervix is markedly enlarged and replaced by tumor.
	View Full Size Image
Media type:  CT

Media file 16:  A 57-year-old woman with stage IVB poorly differentiated endometrial carcinoma (same patient as Images 14 and 15). CT image through the lower thorax shows multiple pulmonary metastases, small pericardial effusion, and small bilateral pleural effusions.
	View Full Size Image
Media type:  CT

Media file 17:  An 83-year-old woman with moderately differentiated endometrioid endometrial adenocarcinoma. Axial T1-weighted MR image of the uterus shows a multilobulated, low signal intensity, polypoid tumor arising from the right side of the endometrium and surrounded by fluid. The fluid markedly distending the endometrial cavity is hyperintense on T1- and T2-weighted images, consistent with hematometra. Examination under anesthesia revealed significant benign stenosis of the atrophic cervix. Dilatation and curettage confirmed the fluid to be old blood and established the diagnosis of endometrial malignancy. Note that the marked myometrial thinning, due to both the elderly age and the marked distension of the uterine cavity, limits the assessment of myometrial invasion.
	View Full Size Image
Media type:  MRI

Media file 18:  An 83-year-old woman with moderately differentiated endometrioid endometrial adenocarcinoma (same patient as Images 17, 19, and 20). Sagittal T1-weighted MR image of the uterus shows a multilobulated, low signal intensity, polypoid tumor arising from the anterior endometrial layer and surrounded by fluid. The fluid markedly distending the endometrial cavity is hyperintense on T1- and T2-weighted images, consistent with hematometra. Examination under anesthesia revealed significant benign stenosis of the atrophic cervix. Dilatation and curettage confirmed the fluid to be old blood and established the diagnosis of endometrial malignancy. Note that the marked myometrial thinning, due to both the elderly age and the marked distension of the uterine cavity, limits the assessment of myometrial invasion.
	View Full Size Image
Media type:  MRI

Media file 19:  An 83-year-old woman with moderately differentiated endometrioid endometrial adenocarcinoma (same patient as Images 17, 18, and 20). Sagittal T1-weighted image of the uterus, following the administration of gadolinium, shows mild contrast enhancement of polypoid tumor arising from the anterior endometrial layer. The fluid markedly distending the endometrial cavity is hyperintense on T1- and T2-weighted images, consistent with hematometra. Examination under anesthesia revealed significant benign stenosis of the atrophic cervix. Dilatation and curettage confirmed the fluid to be old blood and established the diagnosis of endometrial malignancy. Note that the marked myometrial thinning, due to both the elderly age and the marked distension of the uterine cavity, limits the assessment of myometrial invasion.
	View Full Size Image
Media type:  MRI

Media file 20:  An 83-year-old woman with moderately differentiated endometrioid endometrial adenocarcinoma (same patient as Images 17, 18, and 19). Sagittal T2-weighted image of the uterus without gadolinium shows slightly hyperintense polypoid tumor arising from the anterior endometrial layer. The tumor size is underestimated secondary to the surrounding isointense fluid markedly distending the endometrial cavity; the fluid is hyperintense on T1- and T2-weighted images and is consistent with hematometra. Examination under anesthesia revealed significant benign stenosis of the atrophic cervix. Dilatation and curettage confirmed the fluid to be old blood and established the diagnosis of endometrial malignancy. Note that the marked myometrial thinning, due to both the elderly age and the marked distension of the uterine cavity, limits the assessment of myometrial invasion.
	View Full Size Image
Media type:  MRI

REFERENCES

Section 8 of 8

• Authors and Editors
• Introduction
• Differentials
• CT SCAN
• MRI
• Ultrasound
• Multimedia
• References

• Alcazar JL, Errasti T, Zornoza A. Saline infusion sonohysterography in endometrial cancer: assessment of malignant cells dissemination risk. Acta Obstet Gynecol Scand. Apr 2000;79(4):321-2. [Medline].
• Anciaux D, Lawrence WD. Malignant uterine tumors. In: Shingleton HM, Fowler WC Jr, Jordan JA, Lawrence WD, eds. Gynecologic Oncology: Current Diagnosis and Treatment. London: WB Saunders Co;1996:130-45.
• Ascher SM, Brown JJ, Walsh JW. Diagnostic imaging techniques in gynecologic oncology. In: Hoskins WJ, Perez CA, Young RC, eds. Principles and Practice of Gynecologic Oncology. 2nd ed. Philadelphia: Lippincott-Raven;1997:587-625.
• Ascher SM, Takahama J, Jha RC. Staging of gynecologic malignancies. Top Magn Reson Imaging. Apr 2001;12(2):105-29. [Medline].
• Atri M, Nazarnia S, Aldis AE, et al. Transvaginal US appearance of endometrial abnormalities. Radiographics. May 1994;14(3):483-92. [Medline].
• Bakour SH, Dwarakanath LS, Khan KS, et al. The diagnostic accuracy of ultrasound scan in predicting endometrial hyperplasia and cancer in postmenopausal bleeding. Acta Obstet Gynecol Scand. May 1999;78(5):447-51. [Medline].
• Baldwin MT, Dudiak KM, Gorman B, Marks CA. Focal intracavitary masses recognized with the hyperechoic line sign at endovaginal US and characterized with hysterosonography. Radiographics. Jul-Aug 1999;19(4):927-35. [Medline].
• Balfe DM, Van Dyke J, Lee JK, et al. Computed tomography in malignant endometrial neoplasms. J Comput Assist Tomogr. Aug 1983;7(4):677-81. [Medline].
• Barakat RR. Contemporary issues in the management of endometrial cancer. CA Cancer J Clin. Sep-Oct 1998;48(5):299-314. [Medline].
• Barakat RR, Park RC, Grigsby PW, et al. Corpus: epithelial tumors. In: Hoskins WJ, Perez CA, Young RC, eds. Principles and Practice of Gynecologic Oncology. 2nd ed. Philadelphia: Lippincott-Raven;1997:859-96.
• Briley M, Lindsell DR. The role of transvaginal ultrasound in the investigation of women with post-menopausal bleeding. Clin Radiol. Jul 1998;53(7):502-5. [Medline].
• Buyuk E, Durmusoglu F, Erenus M, Karakoc B. Endometrial disease diagnosed by transvaginal ultrasound and dilatation and curettage. Acta Obstet Gynecol Scand. May 1999;78(5):419-22. [Medline].
• Cacciatore B, Lehtovirta P, Wahlstrom T, Ylostalo P. Preoperative sonographic evaluation of endometrial cancer. Am J Obstet Gynecol. Jan 1989;160(1):133-7. [Medline].
• Cohen I. Endometrial pathologies associated with postmenopausal tamoxifen treatment. Gynecol Oncol. Aug 2004;94(2):256-66. [Medline].
• Coleman BG, Arger PH, Grumbach K, et al. Transvaginal and transabdominal sonography: prospective comparison. Radiology. Sep 1988;168(3):639-43. [Medline].
• Conte M, Guariglia L, Benedetti Panici P, et al. Transvaginal ultrasound evaluation of myometrial invasion in endometrial carcinoma. Gynecol Obstet Invest. 1990;29(3):224-6. [Medline].
• Creasman WT, DeGeest K, DiSaia PJ, Zaino RJ. Significance of true surgical pathologic staging: a Gynecologic Oncology Group Study. Am J Obstet Gynecol. Jul 1999;181(1):31-4. [Medline].
• DelMaschio A, Vanzulli A, Sironi S, et al. Estimating the depth of myometrial involvement by endometrial carcinoma: efficacy of transvaginal sonography vs MR imaging. AJR Am J Roentgenol. Mar 1993;160(3):533-8. [Medline].
• DiSaia PJ, Creasman WT. Adenocarcinoma of the uterus. In: Clinical Gynecologic Oncology. 5th ed. St Louis: Mosby-Year Book;1997:134-67.
• Dore R, Moro G, D''Andrea F, et al. CT evaluation of myometrium invasion in endometrial carcinoma. J Comput Assist Tomogr. Mar-Apr 1987;11(2):282-9. [Medline].
• Dubinsky TJ, Stroehlein K, Abu-Ghazzeh Y, et al. Prediction of benign and malignant endometrial disease: hysterosonographic-pathologic correlation. Radiology. Feb 1999;210(2):393-7. [Medline].
• Farghaly SA. Can transvaginal sonographic measurement of endometrium alone predict endometrial cancer?. Am J Obstet Gynecol. Aug 2004;191(2):677; author reply 678. [Medline].
• Fistonic I, Hodek B, Klaric P, et al. Transvaginal sonographic assessment of premalignant and malignant changes in the endometrium in postmenopausal bleeding. J Clin Ultrasound. Oct 1997;25(8):431-5. [Medline].
• Fleischer AC, Dudley BS, Entman SS, et al. Myometrial invasion by endometrial carcinoma: sonographic assessment. Radiology. Feb 1987;162(2):307-10. [Medline].
• Frei KA, Kinkel K. Staging endometrial cancer: role of magnetic resonance imaging. J Magn Reson Imaging. Jun 2001;13(6):850-5. [Medline].
• Goldstein SR, Zeltser I, Horan CK, et al. Ultrasonography-based triage for perimenopausal patients with abnormal uterine bleeding. Am J Obstet Gynecol. Jul 1997;177(1):102-8. [Medline].
• Gompel C, Silverberg SG. The corpus uteri. In: Pathology in Gynecology and Obstetrics. 4th ed. Philadelphia: Lippincott-Raven;1994:163-283.
• Gordon AN, Fleischer AC, Dudley BS, et al. Preoperative assessment of myometrial invasion of endometrial adenocarcinoma by sonography (US) and magnetic resonance imaging (MRI). Gynecol Oncol. Aug 1989;34(2):175-9. [Medline].
• Granberg S, Wikland M, Karlsson B, et al. Endometrial thickness as measured by endovaginal ultrasonography for identifying endometrial abnormality. Am J Obstet Gynecol. Jan 1991;164(1 Pt 1):47-52. [Medline].
• Grasel RP, Outwater EK, Siegelman ES, et al. Endometrial polyps: MR imaging features and distinction from endometrial carcinoma. Radiology. Jan 2000;214(1):47-52. [Medline].
• Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics, 2000. CA Cancer J Clin. Jan-Feb 2000;50(1):7-33. [Medline].
• Hamlin DJ, Burgener FA, Beecham JB. CT of intramural endometrial carcinoma: contrast enhancement is essential. AJR Am J Roentgenol. Sep 1981;137(3):551-4. [Medline].
• Hann LE, Gretz EM, Bach AM, Francis SM. Sonohysterography for evaluation of the endometrium in women treated with tamoxifen. AJR Am J Roentgenol. Aug 2001;177(2):337-42. [Medline].
• Hann LE, Giess CS, Bach AM, et al. Endometrial thickness in tamoxifen-treated patients: correlation with clinical and pathologic findings. AJR Am J Roentgenol. Mar 1997;168(3):657-61. [Medline].
• Hardesty LA, Sumkin JH, Nath ME, et al. Use of preoperative MR imaging in the management of endometrial carcinoma: cost analysis. Radiology. Apr 2000;215(1):45-9. [Medline].
• Hardesty LA, Sumkin JH, Hakim C, et al. The ability of helical CT to preoperatively stage endometrial carcinoma. AJR Am J Roentgenol. Mar 2001;176(3):603-6. [Medline].
• Hasumi K, Matsuzawa M, Chen HF, et al. Computed tomography in the evaluation and treatment of endometrial carcinoma. Cancer. Sep 1 1982;50(5):904-8. [Medline].
• Hricak H, Stern JL, Fisher MR, et al. Endometrial carcinoma staging by MR imaging. Radiology. Feb 1987;162(2):297-305. [Medline].
• Hricak H, Rubinstein LV, Gherman GM, Karstaedt N. MR imaging evaluation of endometrial carcinoma: results of an NCI cooperative study. Radiology. Jun 1991;179(3):829-32. [Medline].
• Hulka CA, Hall DA. Endometrial abnormalities associated with tamoxifen therapy for breast cancer: sonographic and pathologic correlation. AJR Am J Roentgenol. Apr 1993;160(4):809-12. [Medline].
• Jorizzo JR, Riccio GJ, Chen MY, Carr JJ. Sonohysterography: the next step in the evaluation of the abnormal endometrium. Radiographics. Oct 1999;19 Spec No:S117-30. [Medline].
• Karlsson B, Granberg S, Wikland M, et al. Transvaginal ultrasonography of the endometrium in women with postmenopausal bleeding--a Nordic multicenter study. Am J Obstet Gynecol. May 1995;172(5):1488-94. [Medline].
• Kier R. Magnetic resonance imaging of the uterus. Magn Reson Imaging Clin N Am. May 1994;2(2):189-210. [Medline].
• Kim SH, Kim HD, Song YS, et al. Detection of deep myometrial invasion in endometrial carcinoma: comparison of transvaginal ultrasound, CT, and MRI. J Comput Assist Tomogr. Sep-Oct 1995;19(5):766-72. [Medline].
• Kinkel K, Kaji Y, Yu KK, et al. Radiologic staging in patients with endometrial cancer: a meta-analysis. Radiology. Sep 1999;212(3):711-8. [Medline].
• Lee EJ, Byun JY, Kim BS, et al. Staging of early endometrial carcinoma: assessment with T2-weighted and gadolinium-enhanced T1-weighted MR imaging. Radiographics. Jul-Aug 1999;19(4):937-45; discussion 946-7. [Medline].
• Levine D, Gosink BB, Johnson LA. Change in endometrial thickness in postmenopausal women undergoing hormone replacement therapy. Radiology. Dec 1995;197(3):603-8. [Medline].
• Lyons EA, Gratton D, Harrington C. Transvaginal sonography of normal pelvic anatomy. Radiol Clin North Am. Jul 1992;30(4):663-75. [Medline].
• Mendelson EB, Bohm-Velez M. Transvaginal ultrasonography of pelvic neoplasms. Radiol Clin North Am. Jul 1992;30(4):703-34. [Medline].
• Mendelson EB, Bohm-Velez M, Joseph N, Neiman HL. Gynecologic imaging: comparison of transabdominal and transvaginal sonography. Radiology. Feb 1988;166(2):321-4. [Medline].
• Morrows CP, Curtin JP. Tumors of the endometrium. In: Schmitt W, ed. Synopsis of Gynecologic Oncology. 5th ed. New York: Churchill Livingstone;1998:151-85.
• Olaya FJ, Dualde D, Garcia E, et al. Transvaginal sonography in endometrial carcinoma: preoperative assessment of the depth of myometrial invasion in 50 cases. Eur J Radiol. Feb 1998;26(3):274-9. [Medline].
• Parkin DM, Pisani P, Ferlay J. Global cancer statistics. CA Cancer J Clin. Jan-Feb 1999;49(1):33-64, 1. [Medline].
• Prat J. Female reproductive system. In: