Detection and Characterization of Ovarian Tumors Role of Ultrasound

Ultrasound is considered the modality of choice for initial evaluation of an ovarian mass.19 It has been reported that in a setting similar to day-to-day clinical practice, in which the readers were given a brief clinical history of the patients (i.e., age, menstrual status, family history of ovarian cancer, previous pelvic surgery, and presenting symptoms), the experienced readers reached a prospective diagnostic accuracy of 92%, with excellent interobserver agreement (k 0.85). The less experienced observers obtained an accuracy that ranged between 82% and 87%, with moderate to good interobserver agreement (k 0.52 to 0.76).20

The features suggestive of ovarian malignancy on ultrasound include septations greater than 3 mm, mural nodularity, and papillary projections. Unilocular or multi-locular ovarian cystic lesions without solid parts are more likely to be benign.21,22 In other words, the most significant feature predictive of ovarian malignancy is the presence of solid components within the mass.23 When solid excrescences or solid portions of the tumor demonstrate vascular flow with color Doppler sonography (conventional or power), the likelihood of malignancy is even greater.23,24 Some benign lesions, such as endometriomas and hemorrhagic cysts, may mimic ovarian neoplasms on ultrasound (Fig. 5-1). Therefore, for premenopausal women, it may be prudent to obtain short-term follow-up on ovarian lesions to exclude transient physiologic changes.22

The role of spectral Doppler analysis using parameters such as resistive index (RI), pulsatility index (PI), and peak systolic velocity (PSV) in the evaluation of ovarian masses has been controversial. On spectral Doppler, ovarian cancer often

Complex Mass Ovarian Cancver

Figure 5-1. A 38-year-old woman with thyroid cancer. A, Transverse ultrasound Imaging of the left ovary demonstrates a mass of relatively homogeneous internal echoes. B, Color Doppler ultrasound image of this mass demonstrates no obvious internal flow. C, Axial T2-weighted MRI demonstrates "shading" phenomenon within this mass (arrow), a finding frequently seen in endometriomas. D, Axial Tl-weighted fat-saturated MRI demonstrates high signal intensity within the left ovarian mass (arrow) consistent with a blood-containing lesion. The patient underwent subsequent left ovarian cystectomy, which revealed an endometrioma.

Figure 5-1. A 38-year-old woman with thyroid cancer. A, Transverse ultrasound Imaging of the left ovary demonstrates a mass of relatively homogeneous internal echoes. B, Color Doppler ultrasound image of this mass demonstrates no obvious internal flow. C, Axial T2-weighted MRI demonstrates "shading" phenomenon within this mass (arrow), a finding frequently seen in endometriomas. D, Axial Tl-weighted fat-saturated MRI demonstrates high signal intensity within the left ovarian mass (arrow) consistent with a blood-containing lesion. The patient underwent subsequent left ovarian cystectomy, which revealed an endometrioma.

demonstrates low-resistance waveforms because tumor neovasculature lacks smooth muscle and arteriovenous shunting often occurs.25 Lower RI, lower PI, and higher PSV are thought to be associated with lower impedance flow and higher vascularity in the adnexal mass and thus may be indicative of malignancy. Early research suggested high sensitivity and specificity using an RI less than 0.4 and a PI less than 1 as cut-off values for diagnosing malignant ovarian masses.26-28 However, a later study by Levine and colleagues29 showed that although use of the RI might improve specificity in the assessment of possibly malignant lesions, a significant number of malignant lesions could be miscategorized on the basis of the RI.29 Other studies found that although PI and RI tended to be lower in malignant ovarian masses, they overlapped considerably in benign and malignant lesions, and no discriminatory cut-off value could be found.21,30 Therefore, Doppler sonography has severe limitations in the differentiation of benign from malignant adnexal disease on the basis of low-impedance flow alone, although combining the flow velocity patterns obtained at pulsed Doppler ultrasound with a detailed analysis of the internal architectural appearance of the adnexal mass may increase the specificity and overall accuracy of the diagnosis.31-36 In addition, lack of detectable flow on color Doppler ultrasound does not exclude ovarian malignancy.36 It should be noted, however, that benign lesions such as corpora lutea with low-impedance flow are more common in premenopausal women; in a postmenopausal woman, low-impedance flow in an ovarian lesion is highly suggestive of malignancy.22

A study by Buy and colleagues37 compared the accuracy of three ultrasonographic techniques in characterizing adnexal masses: conventional gray-scale sonography, conventional sonography combined with color Doppler, and spectral Doppler analysis only using RI, PI, or PSV as diagnostic criteria for malignancy. They showed that conventional sonography alone had an accuracy of 83%, sensitivity of 88%, and specificity of 82%. Using conventional ultrasound combined with color Doppler ultrasound, accuracy was 95%, sensitivity was 88%, and specificity was 97%. Using spectral Doppler analysis only with RI less than or equal to 0.4 as the indication for malignancy, accuracy was 77%, sensitivity was 18%, and specificity was 98%. Using PI less than or equal to 1 as the criterion for malignancy, accuracy was 68%, sensitivity was 71%, and specificity was 67%. For a PSV greater than or equal to 15 cm/s, accuracy was 72%, sensitivity was 47%, and specificity was 81% in diagnosing ovarian malignancy. These results indicated that adding color Doppler to conventional sonography produced higher specificity and positive predictive value than did conventional sonog-raphy alone, whereas RI, PI, and PSV were of limited value as stand-alone diagnostic tests.37 A meta-analysis of 89 studies using receiver operating characteristic curve analysis found that accuracy (as measured by the area under the ROC curve [AUC]) was significantly higher for the combination of gray-scale ultrasound and Doppler ultrasound (0.92) than for morphologic information (0.85), Doppler ultrasound indexes (0.82), or color-flow Doppler imaging alone (0.73) (P < .01 for all).38

It is worth mentioning that technologic advances in diagnostic ultrasonography have led to the development of three-dimensional transvaginal gray-scale volume imaging and power Doppler imaging by ultrasound, which reportedly allows better visualization of the internal architecture of adnexal masses than does conventional two-dimensional transvaginal imaging.39,40 A few studies have demonstrated that three-dimensional power Doppler imaging better defines the morphologic and vascular characteristics of ovarian lesions and significantly improves specificity in the diagnosis of ovarian malignancy.39,41

Role of Magnetic Resonance Imaging

MRI is considered a problem-solving technique in the assessment of adnexal masses.19 For adequate evaluation of adnexal masses, T1- and T2-weighted images of the pelvis are fundamental in the delineation of pelvic anatomy and tumor, and subsequent gadolinium-enhanced sequences, typically done with fat saturation, can be helpful.22 Fat saturation technique enables the reader to distinguish between fat and blood products. Gadolinium-enhanced MRI further improves characterization of the internal architecture of ovarian lesions and has been shown to be more accurate than ultrasound in the assessment of adnexal masses.42-45 Transvaginal ultrasound cannot reliably differentiate blood products, debris, or fibrofatty tissue from neoplastic projections, whereas on gadolinium-enhanced MRI, neoplastic tissue enhances, and clot or debris do not. Gadolinium-enhanced MRI has been shown to have sensitivity, specificity, and accuracy up to 100%, 98%, and 99%, respectively, in the identification of solid components within an adnexal mass, and just as on ultrasound the presence of enhancing solid tissue on MRI is highly sensitive and specific in predicting malignancy.44 A study by Hricak and colleagues46 showed that gadolinium-enhanced MRI was highly accurate in the detection and characterization of complex adnexal masses, with excellent inter- and intraobserver agreement. Gadolinium-enhanced MRI depicted 94% of adnexal masses, with an overall accuracy of 93% for the diagnosis of malignancy.46 The MRI imaging findings that were most predictive of malignancy were necrosis in a solid lesion (odds ratio 107) and vegetations in a cystic lesion (odds ratio 40).46,47 Use of gadolinium-based contrast material contributed significantly to lesion characterization. Other features suggestive of malignancy on

MRI include enhancing septations thicker than 3 mm and septal nodularity, wall irregularity, large lesion size, early tumor enhancement on dynamic contrast-enhanced images, and the presence of ascites, peritoneal disease, or adenopathy.47 The accuracy of MRI (AUC 0.91) has been shown to be superior to that of Doppler ultrasound (AUC 0.78) and of CT (AUC 0.87) in the diagnosis of malignant ovarian masses.45 In addition, a meta-analysis showed that for women with an indeterminate ovarian mass at gray-scale ultrasound, MRI results contributed more to a change in the probability of ovarian cancer in both pre- and postmenopausal women than did CT or combined gray-scale and Doppler ultrasound results.48

Role of Computed Tomography

The advent of multislice CT, which allows faster acquisition times and higher spatial resolution, has led to a great increase in the number of CT examinations performed. For a number of clinical indications such as renal colic, appendicitis, and diverticulitis, CT has become a primary imaging approach. However, although CT has been shown to be the modality of choice in staging and preoperative planning for ovarian cancer,49,50 it is generally not considered helpful for primary characterization of adnexal masses. When an adnexal mass is detected on CT, it is common practice not to characterize the mass based on its appearance on CT, but to refer patients to ultrasound or MRI for further characterization of the mass and management guidance.

As a matter of fact, CT can probably yield more diagnostic information than is generally believed.51 One study found that the rates of detection of ovarian tumors were comparable for CT and ultrasound (87% and 86%). The study also found that CT had a higher accuracy than ultrasound in characterizing tumors as benign or malignant (94% versus 80%) and that there was no significant difference in specificity (99% for CT versus 92% for US).52 Another study from the same group found no significant difference in the overall accuracy levels of CT (92%) and MRI (86%) in characterizing tumors as benign or malignant.53 Another study showed that preopera-tive CT in patients with ovarian abnormalities had an accuracy of 87%, a sensitivity of 90%, and a specificity of 85% for predicting malignant disease.50

Role of FDG-PET

PET scanning with 2-[fluorine 18] fluoro-2-deoxy-D-glucose (18fluorodeoxyglucose; FDG) is based on uptake of FDG by functionally active tissue such as neoplasm, which has a higher glucose metabolism. The role of FDG-PET in combination with CT in the evaluation of pelvic malignancies has been growing rapidly in recent years. FDG-PET has proved valuable in the evaluation of a variety of pelvic malignancies, including colorectal cancer, uterine cervical cancer, and endometrial cancer.54 Fusion of PET and CT images obtained simultaneously allows combined anatomic and functional imaging55 and offers higher specificity than PET alone. It has been suggested that CT images of optimal diagnostic quality can be obtained using both oral and intravenous contrast material without interfering with PET, thus offering a "one-stop shopping" imaging protocol.56,57 In addition, CT images can be used for attenuation correction for PET, thus decreasing the overall cost and acquisition time.55

PET, however, has recognized limitations. The major disadvantages include misinterpretation of normal physiologic activity in the abdomen or pelvis, and limited intrinsic image resolution leading to a failure to detect small lesions (less than 0.5 cm).55 Since FDG-PET does not have the resolution needed to characterize the primary adnexal mass, the sensitivity and specificity of FDG-PET for diagnosis and assessment of adnexal masses are inferior to those of ultrasound, CT, and MRI; therefore, PET has a limited role in the evaluation of primary ovarian cancer.58

Physiologic uptake of FDG in ovaries during different phases of the menstrual cycle may hinder detection of primary ovarian cancer.54 In addition, a variety of benign lesions, such as serous and mucinous cystadenomas, corpus luteum cysts, and dermoid cysts, are known to accumulate FDG and cannot be reliably differentiated from malignant lesions22,58 (Fig. 5-2). Generally speaking, increased FDG uptake in a solid component of an ovary that does not correspond to one of the above-mentioned benign lesions should be considered suggestive of malignancy. The suggestion of malignancy is stronger when ovarian FDG uptake is present in a postmenopausal woman.

In addition, misregistration cannot be completely eliminated because the acquisition time is relatively long for PET and because physiologic patient activity, such as respiratory motion, bowel peristalsis, and bladder distention, occur during it.55 It also should be noted that both physiologic uptake and tumor activity may appear larger on PET than on the corresponding CT image owing to the "blooming" effect if the uptake activity is intense on PET.55

Bilateral Ovarian Solid Cystic Lesion

Figure 5-2. A 43-year-old woman referred to FDG-PET for suspicion of ovarian tumor.

A, Axial contrast-enhanced CT Image of the pelvis demonstrates complicated cystic structures In bilateral adnexae (arrows). B, Transverse PET image of the pelvis demonstrates increased metabolic activity in the adnexae up to 3.5 in maximum standardized uptake value (SUV) on the right (arrow) and 2.9 on the left (not shown). Subsequent surgical pathology showed acute-on-chronic salpingitis bilaterally. (We thank Dr. Pek Lan Khong at the Department of Diagnostic Radiology, University of Hong Kong for providing these images.)

Figure 5-2. A 43-year-old woman referred to FDG-PET for suspicion of ovarian tumor.

A, Axial contrast-enhanced CT Image of the pelvis demonstrates complicated cystic structures In bilateral adnexae (arrows). B, Transverse PET image of the pelvis demonstrates increased metabolic activity in the adnexae up to 3.5 in maximum standardized uptake value (SUV) on the right (arrow) and 2.9 on the left (not shown). Subsequent surgical pathology showed acute-on-chronic salpingitis bilaterally. (We thank Dr. Pek Lan Khong at the Department of Diagnostic Radiology, University of Hong Kong for providing these images.)

Specific Tumor Characterization with Imaging

Preoperative characterization of an ovarian mass is of substantial clinical value for treatment planning. Although ovarian tumors may overlap in their clinical and radiologic features, certain specific imaging features may be present in certain types of ovarian tumors, and identification of these features may enable the reader to indicate a specific diagnosis or at least narrow the differential diagnosis considerably.59

Imaging findings for specific tumor characterization follow gross pathologic features. For example, epithelial tumors typically are primarily cystic—either unilocular or multilocular—and when malignant, they are associated with varying proportions of solid tissue.51 Although the two most common types of epithelial tumors—serous and mucinous tumors—cannot always be differentiated based on their imaging appearances, certain features may suggest one diagnosis over the other. For example, a unilocular or multilocular cystic mass with a thin regular wall and septum, no soft tissues vegetations, and homogeneous CT attenuation or MRI signal intensity of the locules is most likely a benign serous cystadenoma.59 A benign mucinous cystadenoma can be similar in appearance to a benign serous cystadenoma except that the muci-nous tumor may contain liquids of different CT attenuation or MRI signal intensity, thus giving a "mosaic" pattern.59 Mucinous cystadenomas also tend to be larger than their serous counterparts at presentation.

Although it has been suggested that lesions smaller than 4 cm in diameter are more likely to be benign,60 a significant overlap in size exists between benign and malignant lesions, limiting the value of size criteria. Large benign ovarian tumors are occasionally seen and can remain clinically silent as they grow,52 whereas the primary serous cystadenocarcinomas can be quite small and manifest as peritoneal carcinomatosis. Generally speaking, a greater amount of soft tissue components (e.g., irregular cystic wall and septum greater than 3 mm in thickness, or endocystic or exocystic vegetations) is suggestive of a greater likelihood of malignancy. A large soft tissue component with necrosis is also suggestive of malignancy.46 However, benign masses such as cystadenofibromas could contain solid components and cannot be differentiated from malignant tumors by either the size of the solid portion or the intensity of contrast enhancement, and occasionally an ovarian epithelial tumor of low malignant potential may appear purely cystic on imaging53,61 (Fig. 5-3).

Figure 5-3. A 57-year-old woman with lymphoma. Sagittal T2-weighted MRI demonstrates a multiloculated cystic mass with internal soft tissue nodularity (arrow) and multiple septations. Subsequent surgical pathology demonstrated a serous cystadenofibroma.

Ovarian Lymphoma Ultrasound

Figure 5-3. A 57-year-old woman with lymphoma. Sagittal T2-weighted MRI demonstrates a multiloculated cystic mass with internal soft tissue nodularity (arrow) and multiple septations. Subsequent surgical pathology demonstrated a serous cystadenofibroma.

Papillary projections found on an imaging study are thought to constitute an important predictor of the diagnosis of epithelial ovarian tumors and may even correlate with the aggressiveness of the tumor.59 These projections are usually absent in benign epithelial tumors, and, if present, they are generally small. Papillary projections can be profuse in epithelial tumors with low malignant potential. Although they can be in invasive epithelial carcinomas, their gross appearance is often dominated by a solid component.59 However, these features do not allow confident differentiation of epithelial tumors with low malignant potential from invasive tumors.62

Ancillary findings such as ascites, tumor implants in the abdomen or pelvis, ade-nopathy, and adjacent organ invasion further increase confidence in the diagnosis of an ovarian malignancy. Bilateral primary ovarian tumors with peritoneal carcinomato-sis are seen more frequently in serous than in mucinous cystadenocarcinomas, whereas mucinous adenocarcinomas can rupture and lead to pseudomyxoma peritonei.59

Endometrioid and clear cell carcinomas are the most common malignant neoplasms arising from endometriosis. Therefore, an endometrioma with solid components suggests malignancy and must be removed.52,63,64 The imaging features of endometrioid ovarian carcinomas are nonspecific and include a large complex cystic mass with solid components.59 However, 15% to 30% of endometrioid carcinomas are associated with synchronous endometrial carcinoma or hyperplasia, which manifests as concurrent endometrial thickening on imaging.51,52,63 The imaging features of clear cell carcinomas are also nonspecific; however, clear cell carcinoma commonly presents as a large unilocular cystic lesion with solid protrusions, which tend to be round and sparse65 (Fig. 5-4). Although benign endometriomas are typically T1 hyperintense and T2 hypointense on MRI, the signal intensity of clear cell carcinomas is variable.65

Fat in an adnexal mass is typically diagnostic of a mature teratoma. On ultrasound, the sebaceous material in a mature teratoma typically manifests as diffuse or partial echogenicity within the mass. However, increased echogenicity is not specific, and the presence of fat should be definitively confirmed with CT or MRI. On CT, a negative attenuation is indicative of fat. On MRI, fat is hyperintense on T1-weighted images and suppressed in signal on chemically selective fat suppression sequences. Mature teratomas often contain a protuberance projecting into the cystic cavity known as the Rokitansky nodule, which may contain hair, bone, or teeth66 (Fig. 5-5). Rokitansky nodules are typically densely echogenic on ultrasound and may be associated with calcifications on CT. In addition, calcifications may also be present in the septations or wall of the mass. However, benign mature teratomas have a broad

Ovarian Cancer Ultrasound

Figure 5-4. A 62-year-old woman with a history of breast cancer, presenting with an adnexal mass. Axial contrast-enhanced CT image demonstrates a large complex cystic mass in the right adnexa, with large heterogeneous and irregular soft tissue components (arrow). In addition, the endometrial stripe also appears thickened on CT (arrowhead). Subsequent surgical pathology demonstrated a clear cell carcinoma of the right ovary, and endometrial polyp in the uterus.

Figure 5-4. A 62-year-old woman with a history of breast cancer, presenting with an adnexal mass. Axial contrast-enhanced CT image demonstrates a large complex cystic mass in the right adnexa, with large heterogeneous and irregular soft tissue components (arrow). In addition, the endometrial stripe also appears thickened on CT (arrowhead). Subsequent surgical pathology demonstrated a clear cell carcinoma of the right ovary, and endometrial polyp in the uterus.

Ovarian Cancer Ultrasound

Figure 5-5. A 52-year-old woman with a history of breast cancer. A, Unenhanced axial Tl-weighted MRI demonstrates a Tl-hyperintense mass in the right ovary with a small mural nodule (arrowhead). B, Unenhanced axial T1-weighted MR image through the lower portion of the mass demonstrates amorphous material at the dependent portion of the mass (arrow). C, Axial unenhanced Tl-weighted fat-saturated image demonstrates suppression of the signal within the right ovarian mass (arrow), indicating that this is a fat-containing lesion. D, Transverse ultrasound image of the right ovary illustrates the cystic mass in the right ovary with mural nodule (arrowhead) and echogenic material in the dependent portion (arrow). Subsequent surgical pathology confirmed that this was a mature cystic teratoma. C, an incidental simple cyst in the right ovary.

Figure 5-5. A 52-year-old woman with a history of breast cancer. A, Unenhanced axial Tl-weighted MRI demonstrates a Tl-hyperintense mass in the right ovary with a small mural nodule (arrowhead). B, Unenhanced axial T1-weighted MR image through the lower portion of the mass demonstrates amorphous material at the dependent portion of the mass (arrow). C, Axial unenhanced Tl-weighted fat-saturated image demonstrates suppression of the signal within the right ovarian mass (arrow), indicating that this is a fat-containing lesion. D, Transverse ultrasound image of the right ovary illustrates the cystic mass in the right ovary with mural nodule (arrowhead) and echogenic material in the dependent portion (arrow). Subsequent surgical pathology confirmed that this was a mature cystic teratoma. C, an incidental simple cyst in the right ovary.

spectrum of imaging findings, which may overlap with those of a malignant mass. For example, a mature teratoma may manifest as a mixed mass with all the components of the three germ cell layers and demonstrate a complex appearance, or it may be purely cystic in nature, containing locules of fluid with septations, mimicking an epithelial tumor.59

A malignant immature teratoma may contain mature tissue elements similar to those seen in mature cystic teratomas, including small foci of fat. However, unlike mature teratomas, immature teratomas typically have prominent solid components with internal necrosis or hemorrhage, may contain scattered (rather than localized) calcifications, may demonstrate rapid growth, and may have a capsule that is not well defined, or may be perforated or ruptured.66,67

Dysgerminoma is the ovarian counterpart of seminoma of the testis.67 Characteristic findings such as multilobulated solid ovarian masses with fibrovascular septa that demonstrate prominent flow on ultrasound or significant enhancement on CT or MR have been reported.68,69 Calcification may be present in a speckled pattern. Necrotic and hemorrhagic areas may also be in the tumor.59

In addition to endometrioid carcinomas, sex cord-stromal tumors may also be associated with endometrial abnormality due to their estrogen-producing capabilities. The hyperestrogenemia may produce combined endometrial hyperplasia, polyps, or carcinoma.59 The most common malignant sex cord-stromal tumors are granulosa cell tumors of the ovary, which have variable appearances on imaging, ranging from homogeneous solid masses, to heterogeneous tumors with varying degrees of hemorrhagic or fibrotic changes, to multilocular cystic lesions, to thick-walled or thin-walled unilocular cystic tumors.59,70 In contrast to epithelial tumors, granulosa cell tumors do not have endocystic papillary projections, are less likely to cause peritoneal seeding, and are more likely to be confined to the ovary at the time of diagnosis.59 The associated estrogenic effects on the uterus may manifest as uterine enlargement or as endometrial thickening or hemorrhage.71,72

Ovarian masses with fibrous components include fibroma, fibrothecoma, cystade-nofibroma, Brenner tumor, and Sertoli-Leydig cell tumor.59 Sertoli-Leydig cell tumor is rare and occurs most often in young adults. Approximately one third of female patients with Sertoli-Leydig cell tumors present with progressive masculinization owing to excess testosterone secreted by the tumor (Fig. 5-6). Fibroma is the most

Figure 5-6. A 15-year-old girl presenting with amenorrhea, obesity, and hirsutism.

A, Transverse ultrasound image of the pelvis demonstrates a heterogeneous mass that is mostly solid in the right ovary. B, Color Doppler image demonstrates blood flow within the solid components of the ovarian mass. Subsequent surgical pathology showed a right ovarian Sertoli-Leydig cell tumor. (We thank Dr. Pek Lan Khong at the Department of Diagnostic Radiology, University of Hong Kong for providing these images.)

Sertoli Leydig Ovarian Tumor Ultrasound
10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Responses

Post a comment