Radiological Reasoning: Imaging Characterization of Bilateral Adnexal Masses

Transcription

1 AJR Integrative Imaging LIFELONG LEARNING FOR RADIOLOGY This Radiological Reasoning article is available for SAM credit and CME credits when completed with the additional educational material provided in Imaging Evaluation of Adnexal Masses: Self-Assessment Module. See page S457 for details. Susanna I. Lee 1 Keywords: adnexa, genitourinary imaging, MRI, ovaries, pelvic imaging, sonography, uterus, women s imaging DOI: /AJR Received December 27, 2005; accepted after revision March 13, Department of Radiology, Massachusetts General Hospital, 55 Fruit St., Boston, MA Address correspondence to S. I. Lee. (slee0@partners.org). AJR 2006; 187:S460 S X/06/1873 S460 American Roentgen Ray Society Radiological Reasoning: Imaging Characterization of Bilateral Adnexal Masses Objective A 46-year-old woman presented with bilateral adnexal masses on pelvic sonography, a complex cystic mass on the right, and a homogeneously hypoechoic, solid-appearing mass on the left. Pelvic MRI showed a T1 hypointense, T2 hyperintense, nonenhancing mass in the right ovary and a homogeneously T1 hyperintense lesion with mixed T2 signal in the left ovary. Fat-saturated T1-weighted sequences showed the left ovarian lesion to be an endometrioma containing a high concentration of hemoglobin breakdown products and revealed other endometriotic implants in the right ovary, broad ligament, and cul-de-sac. Follow-up pelvic sonography 9 weeks later showed resolution of the right complex cystic ovarian mass, indicating that it was a physiologic hemorrhagic cyst. Conclusion Imaging features of benign and malignant ovarian masses overlap. If imaging is inconclusive in characterizing an adnexal mass as benign, the mass will be resected rather than followed up because of the concern for a rapidly growing ovarian cancer. Thus, the goal of imaging is to identify unequivocally the benign lesions that can be left untreated. Pelvic MRI provides a powerful adjunct to pelvic sonography in characterizing adnexal masses as benign. Case History A 46-year-old woman presented to the emergency department with right-sided pelvic pain that had been gradually increasing in intensity over the past 14 hours. She was premenopausal, and the onset of her last menstrual period had been 11 days earlier. She had no associated fevers, bowel symptoms, or vaginal discharge. Her serum β-hcg was negative. A pelvic examination revealed fullness and mild tenderness in the right adnexa. Pelvic sonography was performed. Sonography Transabdominal imaging (Fig. 1A) shows a normal uterus and two masses posterior to the uterus. Transvaginal images show a 6.6-cm cystic lesion with solid components in the right adnexa (Fig. 1B) and a 3.9-cm solid-appearing left adnexal mass (Fig. 1C). Neither ovary was seen separate from these lesions. Expert Discussion (Dr. Lee) Although acute-onset pelvic pain in a nonpregnant woman has many possible causes, the one requiring the most urgent diagnosis is ovarian torsion. Pelvic sonography is often performed to look for underlying lesions that might cause torsion, but this diagnosis must ultimately be made on clinical assessment. Acute ovarian torsion can S460 AJR:187, September 2006

2 Bilateral Adnexal Masses Fig year-old woman who presented to emergency department with rightsided pelvic pain of gradually increasing intensity. Pelvic sonography was performed first.p A, Transverse sonogram from transabdominal examination shows two adnexal lesions (arrows) posterior to normal-appearing uterus. B and C, Transvaginal sonograms reveal complex cystic mass on right (B) and a more solid-appearing hypoechoic mass on left (C). Neither ovary was seen separate from these adnexal lesions. present with normal pelvic sonographic findings [1], and Doppler flow can be detected in up to 93% of torsed ovaries [2]. Because this patient lacked any history of sudden acute onset of pain and because extreme adnexal pain was not elicited on examination, the attending gynecologist excluded ovarian torsion on clinical grounds. Pelvic sonography shows two lesions, one in each adnexum. Once emergent diagnoses have been excluded, adnexal lesions must next be evaluated for the possibility of malignancy. In this context, four characteristics on sonography should be assessed: location (of ovarian rather than extraovarian origin), size, morphology (simple cyst, mixed solid and cystic, or homogeneously solid), and persistence or resolution on follow-up after at least one menstrual cycle. Doppler sonography is rarely useful for assessing ovarian lesions. Color Doppler signal, if present, is occasionally helpful in A B distinguishing a soft-tissue component of a complex cystic epithelial tumor from a clot in a hemorrhagic cyst. However, studies on the usefulness of color and pulsed-wave Doppler sonography for distinguishing malignant from benign ovarian lesions [3, 4], although prolific, have not proven sufficiently conclusive or definitive to dictate management decisions. When an adnexal mass can be shown to be separate from the ipsilateral ovary (i.e., extraovarian), it is considered benign [5 7]. However, lesions of ovarian origin can be malignant. The masses in this patient cannot be definitively shown to be extraovarian because neither ovary is seen separate from these lesions. Thus, the lesions should be considered of ovarian origin and further characterized to exclude malignancy. Size is important because smaller lesions are much less likely to be ovarian cancer. Ovarian cancers can grow rapidly. Studies evaluating surveillance sonography to screen for ovarian cancer consis- C AJR:187, September 2006 S461

3 Lee tently reveal patients who present with cancer less than 12 months after sonography with normal findings [8 10], which indicates that ovarian cancer develops and becomes advanced in a matter of months rather than years. This observation explains why incidentally identified primary ovarian cancer is usually large, with a mean size of 6.5 cm [11]. Although consensus has not been reached as to a cutoff size above which ovarian lesions are considered suspicious for malignancy, generally ovarian masses larger than 3 or 4 A C Fig. 2 In same patient as in Figure 1, pelvic MRI was performed for further characterization of lesions. A, Axial T2-weighted image shows two lesions in right ovary, one large (solid white arrow) and the other small (black arrow), and single lesion in left ovary (dashed white arrow). B, Axial T1-weighted image shows that small right (black arrow) and left (dashed white arrow) ovarian lesions are homogeneously hyperintense, whereas large right ovarian lesion (solid white arrow) is hypointense. C, Sagittal T1-weighted image after administration of gadolinium shows no internal enhancement of large right ovarian lesion (arrow). Uterus (asterisk) and bladder (dot) are also seen. (Fig. 2 continues on next page) cm in mean diameter, such as the masses seen in this patient, are recommended for further evaluation [12]. Morphology is helpful in deciding whether a lesion should be further evaluated. Simple cysts (cysts with an imperceptible wall and no solid components such as septations, mural nodules, or papillary projections) need no further imaging evaluation when presenting in a premenopausal woman. Although such cysts could represent benign epithelial tumors (e.g., cystadenomas), they are B S462 AJR:187, September 2006

4 Bilateral Adnexal Masses usually physiologic ovarian cysts or inclusion cysts and are not malignant [13 15]. Large simple cysts are often resected because they cause pain and may predispose the ovary to torsion. Mixed solid and cystic lesions include epithelial ovarian malignancies and should be further evaluated. The right ovary in this patient is mixed solid and cystic with a papillary projection and a thin septum. Although the linear border of the solid component is strongly suggestive of a retracting clot of a hemorrhagic cyst [16, 17], this lesion was nevertheless classified as indeterminate. A follow-up sonography examination in 6 weeks was recommended to assess lesion resolution. Homogeneously solid-appearing lesions, in the absence of a history of prior malignancy, are usually benign [11, 18 20]. The most common solid adnexal lesions are pedunculated uterine or broad ligament fibroids, benign ovarian stromal tumors (e.g., fibroma or fibrothecoma), dermoids, and endometriomas. Although dermoids and endometriomas are pathologically cysts, they are protean in their sonographic appearance and can sometimes appear homogeneously solid on sonography. Each of the solid lesions listed displays a characteristic appearance on MRI. This patient s left ovarian lesion appears solid. On the basis of its low-level internal echoes [21], the diagnosis of endometrioma was suggested. However, MRI was recommended for complete characterization. MRI The right ovary shows two lesions. The larger, homogeneously T2 hyperintense lesion (Fig. 2A) corresponds in size to the 6.6-cm D Fig. 2 (continued) In same patient as in Figure 1, pelvic MRI was performed for further characterization of lesions. D and E, Axial T1-weighted images with fat saturation show that small right (black arrow, D) and left (dashed white arrow, D) ovarian lesions continue to be homogeneously hyperintense. Other tiny T1 hyperintense foci are seen in pelvis (arrowheads). Solid white arrow in D indicates large lesion in right ovary. complex cystic mass seen on sonography. On T1-weighted images (Fig. 2B), this lesion is homogeneously hypointense. After gadolinium administration (Fig. 2C), it shows a thin smooth margin without any internal enhancement, consistent with a cyst. The second right ovarian lesion is smaller and is homogeneously T2 (Fig. 2A) and T1 (Fig. 2B) hyperintense. The left ovary contains a lesion with heterogeneous T2 signal (Fig. 2A) corresponding in size to the 3.9-cm solid-appearing mass seen on sonography. On T1-weighted images, this lesion is homogeneously hyperintense (Fig. 2B). The two T1 hyperintense lesions, the smaller right and the left ovarian masses, remain hyperintense to surrounding tissues on fat-suppression sequences (Figs. 2D and 2E). Their hyperintensity is enhanced by compression of the dynamic range of signal received by the receiver coils. Thus, the observed T1 hyperintensity is not fat but probably represents products of hemoglobin breakdown. Additional extraovarian foci of T1 hyperintensity are seen in the right broad ligament (Fig. 2D) and in the cul-desac (Fig. 2E). Expert Discussion (Dr. Lee) MRI was performed to characterize the left ovarian lesion, the sonographic appearance of which was suggestive of an endometrioma. Characteristics of the lesion on MRI are diagnostic of an endometrioma. On T1-weighted sequences, one sees light-bulb bright lesions that show extremely high T1 signal because of high concentrations of paramagnetic products of E AJR:187, September 2006 S463

5 Lee hemoglobin breakdown. Endometriomas are particularly easy to spot on fat-saturated T1-weighted images because they are light-bulb bright in the surrounding dark pelvic soft tissue. Using this sequence, one can see that not only is this patient s left ovarian lesion an endometrioma, but so is the smaller right ovarian lesion. In addition, MRI identifies an extraovarian implant of endometriosis in the right adnexum and in the cul-desac. Interestingly, only the left endometrioma shows the areas of T2 hypointensity described as shading [21], whereas the right endometrioma appears homogeneously hyperintense on the T2-weighted images. The larger right ovarian lesion, suggested to be a hemorrhagic cyst on sonography, appears to be a cyst on MRI as well. It is homogeneously hyperintense on T2-weighted images and shows no gadolinium enhancement. However, it is T1 isointense, an imaging feature that is compatible with but not diagnostic for blood. Thus, this lesion is still considered indeterminate. On a cautionary note, the differential diagnosis of T1 hyperintense lesions that show persistent signal on fat-suppression sequences includes potentially malignant lesions. These include the rare epithelial ovarian tumors containing high concentrations of mucin [22] and endometriomas that have evolved into endometrial carcinoma from chronic exposure to unopposed estrogen [23]. However, these lesions, in contrast to endometriomas, show heterogeneous T1 signal and T1 isointense softtissue components that often enhance. Thus, if the light bulb is not uniformly bright, or the lesion shows darker T1 components that enhance with contrast material, a definitive diagnosis of endometrioma cannot be made on MRI. Follow-Up Sonography Nine weeks after the original sonography, the transvaginal sonogram of the right ovary (Fig. 3A) shows complete resolution of the complex cystic lesion. The small right ovarian endometrioma seen on MRI is now identified on sonography. The left ovary (Fig. 3B) again shows the endometrioma seen in the prior studies. Expert Discussion (Dr. Lee) Complex cystic ovarian masses that resolve on sonography after at least one menstrual cycle are, by definition, physiologic cysts and benign. The disappearance of the 6.6-cm right ovarian mass on follow-up imaging is consistent with the diagnosis of a physiologic hemorrhagic cyst. The persistence and stability of the smaller right ovarian and the 3.9-cm left ovarian masses are consistent with the previous MRI diagnosis of endometriomas. Clinical Management Overall, the imaging evaluation resulted in a diagnosis of endometriosis. Because the patient was past child-bearing and nearing menopause, she elected to undergo hysterectomy. Surgical findings, confirmed on pathology, showed an endometrioma in each ovary and endometriotic implants in the broad ligaments bilaterally, in the cul-de-sac, and on the appendix. Commentary The traditional first step in the evaluation of the female pelvic organs is sonography. In the evaluation of adnexal lesions in adult women, the primary concern is to not miss malignancy. A B Fig. 3. Follow-up pelvic sonography 9 weeks after initial examination of patient shown in Figures 1 and 2. A, Transvaginal sonogram of right ovary shows resolution of large complex cystic lesion. Smaller right ovarian lesion (arrow) is still present. B, Transvaginal image of left ovary shows that left ovarian lesion is persistent and unchanged. S464 AJR:187, September 2006

6 Bilateral Adnexal Masses Consequently, much literature has been devoted to evaluating sonographic criteria (e.g., size, Doppler flow, septal thickness and number, and amount of free fluid) that might predict malignancy. Some authors have evolved elaborate statistical models calculating the likelihood of malignancy [24, 25]. Although these studies have been useful in pointing out worrisome features of ovarian masses, they are not sufficiently sensitive to ensure identification of all cancers while simultaneously excluding most benign lesions, nor are the multifactorial analyses sufficiently simple to assess ovarian masses encountered in everyday sonography practice. The difficulty in trying to distinguish benign from malignant ovarian masses is that the imaging features overlap between the two categories. If the radiologist cannot definitively identify an ovarian lesion as benign, a complex ovarian mass will likely be surgically resected because of the concern for malignancy. As noted, ovarian cancer grows quickly, and one of the best prognostic indicators for survival is resection at an early stage [26]. If imaging is inconclusive in characterizing an ovarian lesion as benign, the lesion will undergo prompt resection, even if the likelihood of malignancy is low by imaging criteria. Thus, the goal of imaging should not be to identify the malignant lesions that will be taken out, but to identify and definitively characterize the benign lesions that can be left untreated. Pelvic MRI has proven to be powerful as an adjunct to pelvic sonography in characterizing adnexal masses as benign [27, 28]. With its accessibility and high resolution, transvaginal sonography is still the best method for characterizing a lesion as a simple cyst, a complex cyst, or a homogeneously solid mass. However, sonography is limited in its range of tissue characterization and field of view. Although MRI is less accessible and has lower resolution than sonography, it has the benefit of a wide range of tissue specificity and a much larger field of view. MRI can be used to definitively characterize fat, blood, and simple fluid, and the enhancement properties of solid tissues can be measured with the administration of gadolinium. The larger field of view proves useful for characterizing large masses or for evaluating lesion location relative to the ovary in patients with large uterine fibroids when the adnexal anatomy is significantly distorted. MRI can characterize two categories of lesions as definitively benign. First are the cystic lesions suspected to be of extraovarian origin. These include paratubal cyst, peritoneal inclusion cyst, and hydrosalpinx. Second are the solid-appearing lesions in patients without a history of prior malignancy in whom metastases are extremely unlikely. The common solid adnexal lesions are pedunculated uterine or broad ligament fibroid, benign ovarian stromal tumor (fibroma and fibrothecoma), dermoid, and endometrioma. Fibroids show a discrete round margin and a distinctive whirling internal structure, and they typically enhance intensely with gadolinium. Benign ovarian stromal tumors show T2 hypointensity, homogeneous internal structure, and a low level of enhancement with the administration of gadolinium. Dermoids contain fat and endometriomas contain high concentrations of blood breakdown products. If a lesion is suspected on sonography of being one of these benign lesions, the patient should be referred to MRI for further evaluation and, in many cases, definitive characterization. The diagnosis of endometriosis poses a special problem for the gynecologist. In the context of a patient presenting with chronic or recurrent pelvic pain, endometriosis is often on the list of suspected causes. However, a definitive diagnosis requires laparoscopy, an invasive procedure that is often not warranted given the degree of symptoms or the level of suspicion. In this context, a noninvasive sensitive imaging method to look for endometriosis, MRI, is a useful and popular diagnostic tool. Pelvic MRI has proven to be more sensitive than sonography in the diagnosis of endometriosis [29, 30], as has been shown by the patient presented here. MRI detected both the additional endometriotic components outside the ovary (where sonography rarely finds disease) and the smaller implant in the right ovary, which was obscured by the large hemorrhagic cyst. The detection of these small implants requires the use of a specific pulse sequence, a T1- weighted sequence with fat saturation. The addition of fat saturation to the T1-weighted sequence significantly improves the conspicuity of small implants [31, 32] and hence the sensitivity of the MRI examination for detecting endometriosis. Because most complex cystic ovarian lesions seen in premenopausal women are physiologic cysts, follow-up imaging is usually performed to document the persistence of a lesion before resection. Imaging is recommended after the passage of one menstrual cycle, typically after 6 weeks. If the lesion is likely to be one of the benign lesions that can be definitively characterized with imaging, MRI should be chosen as the technique for follow-up. However, if the lesion is just a complex cystic lesion with no features suggestive of benignity, the best way to characterize it as benign is to document that it resolves. Under these circumstances, follow-up imaging should be performed with repeated sonography. If this algorithm still fails to result in characterization of an ovarian mass as benign, surgical resection should be considered. In postmenopausal women, in whom physiologic cysts are rare, lesions should be considered persistent on first presentation. Rather than being followed, they should be triaged directly to MRI or surgical resection for definitive characterization. The only lesions that are occasionally followed up with sonography are small simple cysts because of the rare but reported possibility of their evolving into complex cystic lesions [15]. References 1. Varras M, Tsikini A, Polyzos D, Samara Ch, Hadjopoulos G, Akrivis C. Uterine adnexal torsion: pathologic and gray-scale ultrasonographic findings. Clin Exp Obstet Gynecol 2004; 31: Albayram F, Hamper UM. Ovarian and adnexal torsion: spectrum of sonographic findings with pathologic correlation. J Ultrasound Med 2001; 20: Jain KA. Prospective evaluation of adnexal masses with endovaginal grayscale and duplex and color Doppler US: correlation with pathologic findings. Radiology 1994; 191:63 67 AJR:187, September 2006 S465

7 Lee 4. Stein SM, Laifer-Narin S, Johnson MB, et al. Differentiation of benign and malignant adnexal masses: relative value of gray-scale, color Doppler, and spectral Doppler sonography. AJR 1995; 164: Kim JS, Woo SK, Suh SJ, Morettin LB. Sonographic diagnosis of paraovarian cysts: value of detecting a separate ipsilateral ovary. AJR 1995; 164: Kim JS, Lee HJ, Woo SK, Lee TS. Peritoneal inclusion cysts and their relationship to the ovaries: evaluation with sonography. Radiology 1997; 204: Levine CD, Patel UJ, Ghanekar D, Wachsberg RH, Simmons MZ, Stein M. Benign extraovarian mimics of ovarian cancer: distinction with imaging studies. Clin Imaging 1997; 21: Fishman DA, Cohen L, Blank SV, et al. The role of ultrasound evaluation in the detection of early-stage epithelial ovarian cancer. Am J Obstet Gynecol 2005; 192: van Nagell JR Jr, DePriest PD, Reedy MB, et al. The efficacy of transvaginal sonographic screening in asymptomatic women at risk for ovarian cancer. Gynecol Oncol 2000; 77: Tailor A, Bourne TH, Campbell S, Okokon E, Dew T, Collins WP. Results from an ultrasound-based familial ovarian cancer screening clinic: a 10-year observational study. Ultrasound Obstet Gynecol 2003; 21: Valentin L, Skoog L, Epstein E. Frequency and type of adnexal lesions in autopsy material from postmenopausal women: ultrasound study with histological correlation. Ultrasound Obstet Gynecol 2003; 22: Misawa T, Asai M, Higashide K. How to decrease false-positive cases of ovarian cancer screening by transvaginal sonography. J Exp Clin Cancer Res 1997; 16: Modesitt SC, Pavlik EJ, Ueland FR, DePriest PD, Kryscio RJ, van Nagell JR Jr. Risk of malignancy in unilocular ovarian cystic tumors less than 10 centimeters in diameter. Obstet Gynecol 2003; 102: Nardo LG, Kroon ND, Reginald PW. Persistent unilocular ovarian cysts in a general population of postmenopausal women: is there a place for expectant management? Obstet Gynecol 2003; 102: Castillo G, Alcazar JL, Jurado M. Natural history of sonographically detected simple unilocular adnexal cysts in asymptomatic postmenopausal women. Gynecol Oncol 2004; 92: Patel MD, Feldstein VA, Filly RA. The likelihood ratio of sonographic findings for the diagnosis of hemorrhagic ovarian cysts. J Ultrasound Med 2005; 24: Baltarowich OH, Kurtz AB, Pasto ME, Rifkin MD, Needleman L, Goldberg BB. The spectrum of sonographic findings in hemorrhagic ovarian cysts. AJR 1987; 148: Fried AM, Kenney CM 3rd, Stigers KB, Kacki MH, Buckley SL. Benign pelvic masses: sonographic spectrum. RadioGraphics 1996; 16: Stephenson WM, Laing FC. Sonography of ovarian fibromas. AJR 1985; 144: Yeh HC, Kaplan M, Deligdisch L. Parasitic and pedunculated leiomyomas: ultrasonographic features. J Ultrasound Med 1999; 18: Woodward PJ, Sohaey R, Mezzetti TP Jr. Endometriosis: radiologic pathologic correlation. RadioGraphics 2001; 21: Ghossain MA, Buy JN, Ligneres C, et al. Epithelial tumors of the ovary: comparison of MR and CT findings. Radiology 1991; 181: Tanaka YO, Yoshizako T, Nishida M, Yamaguchi M, Sugimura K, Itai Y. Ovarian carcinoma in patients with endometriosis: MR imaging findings. AJR 2000; 175: Brown DL, Doubilet PM, Miller FH, et al. Benign and malignant ovarian masses: selection of the most discriminating gray-scale and Doppler sonographic features. Radiology 1998; 208: Szpurek D, Moszynski R, Smolen A, Sajdak S. Artificial neural network computer prediction of ovarian malignancy in women with adnexal masses. Int J Gynaecol Obstet 2005; 89: Brun JL, Feyler A, Chene G, Saurel J, Brun G, Hocke C. Long-term results and prognostic factors in patients with epithelial ovarian cancer. Gynecol Oncol 2000; 78: Kinkel K, Lu Y, Mehdizade A, Pelte MF, Hricak H. Indeterminate ovarian mass at US: incremental value of second imaging test for characterization meta-analysis and Bayesian analysis. Radiology 2005; 236: Sohaib SA, Mills TD, Sahdev A, et al. The role of magnetic resonance imaging and ultrasound in patients with adnexal masses. Clin Radiol 2005; 60: Bazot M, Thomassin I, Hourani R, Cortez A, Darai E. Diagnostic accuracy of transvaginal sonography for deep pelvic endometriosis. Ultrasound Obstet Gynecol 2004; 24: Bazot M, Darai E, Hourani R, et al. Deep pelvic endometriosis: MR imaging for diagnosis and prediction of extension of disease. Radiology 2004; 232: Sugimura K, Okizuka H, Imaoka I, et al. Pelvic endometriosis: detection and diagnosis with chemical shift MR imaging. Radiology 1993; 188: Ha HK, Lim YT, Kim HS, Suh TS, Song HH, Kim SJ. Diagnosis of pelvic endometriosis: fat-suppressed T1-weighted vs conventional MR images. AJR 1994; 163: S466 AJR:187, September 2006