Small Echogenic Foci in the Ovaries

Transcription

1 Case Series Small Echogenic Foci in the Ovaries Correlation With Histologic Findings Douglas L. Brown, MD, Mary C. Frates, MD, Michael G. Muto, MD, William R. Welch, MD Objective. The purpose of this study was to determine the histologic correlate of small echogenic foci in the ovary and to assess for any association with endometriosis or endosalpingiosis. Methods. Women planning to have a normal ovary surgically removed were scanned preoperatively with transvaginal sonography. If echogenic foci were present in either normal ovary on the preoperative scan, the removed ovary was scanned in a saline bath, and the surface was marked with india ink over an echogenic focus. Histologic sections were then obtained at the marked site. Results. Echogenic foci were detected in 23 ovaries of 16 women. Possible causes were found in 17 of the 23 ovaries: hemosiderin in 6 cases, calcification in 5 cases, hemosiderin and calcification in 2 cases, clusters of inclusion cysts in 2 cases, 1 of which also had hemosiderin, and dense cortical nodules in 2 cases. Histologic findings were benign in all cases except in 1 patient who had primary peritoneal carcinoma unrelated to the echogenic foci. One ovary in another patient had both endosalpingiosis and endometriosis. One other patient had endometriosis involving a fallopian tube but not the ovary. There were no other cases of endometriosis or endosalpingiosis. Conclusions. Small echogenic foci in the ovaries are most frequently due to hemosiderin or calcification. A few small echogenic foci in the ovaries are associated with benign histologic changes and do not appear to be reliable indicators of endosalpingiosis or endometriosis. Key words: calcification; hemosiderin; ovary; sonography. Abbreviations H&E, hematoxylin-eosin Received August 11, 2003, from the Departments of Radiology (D.L.B., M.C.F.), Obstetrics and Gynecology (M.G.M.), and Pathology (W.R.W.), Brigham and Women s Hospital, Harvard Medical School, Boston, Massachusetts USA. Manuscript accepted for publication September 15, This study was supported by a grant from the William F. Milton Fund of Harvard University. Address correspondence and reprint requests to Douglas L. Brown, MD, Department of Radiology, Mayo Clinic, 200 First St SW, Rochester, MN USA. brown.douglas@mayo.edu. Small echogenic foci in the ovaries are observed in about half of women undergoing transvaginal sonography of the pelvis, with a nearly equal frequency of bilateral and unilateral occurrence. 1,2 Generally, the echogenic foci measure 1 to 3 mm, are located in the periphery of the ovary, and have no distal acoustic shadowing. 1,2 The ovary is usually otherwise normal on sonography. It was initially suggested that these echogenic foci were due to psammomatous calcifications in inclusion cysts or calcifications in adenofibromas. 2,3 Endosalpingiosis has also occasionally been reported in cases in which echogenic foci were observed sonographically. 2 On histologic evaluation, psammomatous calcifications are frequently associated with endosalpingiosis. 4,5 Endosalpingiosis, which is benign, is defined in the ovary as the occurrence of multiple cortical inclusion cysts lined by tubal-type epithelium. 4 7 Epithelial inclusion cysts are thought to be the sites of origin of most 2004 by the American Institute of Ultrasound in Medicine J Ultrasound Med 23: , /04/$3.50

2 Small Echogenic Foci in the Ovaries ovarian carcinoma. 8,9 There is evidence that endosalpingiosis might be a precursor to some cases of ovarian serous neoplasms. 4,5,9,10 Endosalpingiosis is often underdiagnosed 6 and may coexist with endometriosis, which is more common and more widely known than endosalpingiosis. Given these associations, we initially wondered whether these small echogenic foci might be an indicator of endosalpingiosis and thereby a potential indicator of an ovarian carcinoma precursor. However, given more recent studies showing that these small echogenic foci are common, 1,2 it seems unlikely that they would indicate an increased risk for ovarian carcinoma. It has also been suggested that these echogenic foci may be an indicator of endometriosis, 11 although there currently is little published evidence in this regard. The primary purpose of our study was to determine the histologic correlates of small echogenic foci seen in the ovaries on transvaginal sonography. Secondarily, we evaluated for any association of the echogenic foci with endosalpingiosis or endometriosis. Materials and Methods This prospective study was approved by our Institutional Review Board, and patients gave written informed consent. We searched the surgery schedule at our institution to identify patients who were to undergo an oophorectomy of a normal ovary, usually incidental to a nonovarian indication for surgery. Additionally, we included 1 other patient who, just before the formal study started, was found to have echogenic foci in an otherwise normal ovary on preoperative sonography, and the surgeon and pathologist requested that it be localized by sonography before histologic study. Patients who agreed to participate in the study underwent transvaginal sonography within 1 month before their scheduled surgery. If no echogenic foci were identified in either ovary, or if neither ovary was seen, the patient was excluded from further participation. If at least 1 small echogenic focus was identified in either ovary on sonography, the patient was included in the study. The sonographic characteristics of the echogenic foci were categorized by size, location, number, and extent of involvement of the peripheral portion of the ovary. The largest dimension of what was subjectively judged to be the largest echogenic focus in each ovary was measured. The location of the echogenic foci was characterized as peripheral (arbitrarily designated as within 3 mm of the ovarian surface), deep within the ovary, or both. The number of calcifications was counted. For peripheral calcifications, the extent of involvement of the ovary was arbitrarily and subjectively characterized as less than one third, one third to two thirds, or greater than two thirds of the surface area of the ovary. The echogenic foci were also evaluated to determine whether there was a distal acoustic artifact, either acoustic shadowing or a comet tail artifact (sometimes referred to as a V-shaped artifact). 12,13 The transvaginal sonography was performed with an Acuson 128XP or Sequoia system (Siemens Medical Solutions, Mountain View, CA) or an HDI 5000 system (Philips Medical Systems, Bothell, WA). The highest feasible transducer frequency was used for each patient; with transducer frequencies from 5 to 8 MHz for the Acuson systems and for the Philips system, the general or resolution mode was used. The ovaries were obtained within 1 hour of their surgical removal. Each ovary was scanned in a saline bath to identify the echogenic foci. The same sonography systems were used for the preoperative scans, although a linear transducer was used, at a frequency similar to that used preoperatively. For each ovary in which echogenic foci were seen preoperatively, an echogenic focus was arbitrarily selected and localized under sonographic guidance by placing the blunt wooden end of a cottontipped swab onto the surface of the ovary, directly over the echogenic focus. Care was taken to avoid marking any echogenic focus just external to the ovary that could have been due to adherent air bubbles. The specimen was removed from the saline bath, with the wooden end of the cotton-tipped swab kept in contact with the site localized, and india ink was applied at the site marked. The swab was then used to apply Bouin solution to the ovarian surface over the india ink mark to help fix the mark. The specimen was returned to the pathology department for histologic sections. Each ovary was sectioned in toto at approximately 3-mm intervals. Sections were processed according to the routine surgical pathology protocol, and hematoxylin and eosin (H&E)-stained slides were prepared for examination. 308 J Ultrasound Med 23: , 2004

3 Brown et al The histologic sections were reviewed to determine the cause of the echogenic foci seen on sonography, identifying the marked area by the india ink on the ovarian surface. On the initial review of the histologic sections, we expected to find calcification and had planned to characterize any associated lesion. Not always finding calcification, we rereviewed all sections looking for other possible causes of the echogenic foci. In addition to calcification, we also assessed for hemosiderin (because we had noted this in some of the studies), clusters of inclusion cysts (because that was reported in a recent study 1 ), or other focal lesions that might be possible causes of an echogenic focus. Results Nineteen patients were recruited into the study. In 3 patients, both ovaries were seen, but neither had an echogenic focus. In another 3 patients, neither ovary could be identified; in 2 of these patients, the uterus was enlarged with numerous fibroids, and 1 patient was very obese. This left 13 patients who had echogenic foci in at least 1 ovary. Additionally, we included the 1 other patient who, just before the formal study started, was found to have echogenic foci in an otherwise normal ovary, and the surgeon and pathologist requested that it be localized by sonography before histologic study. This specimen was scanned and marked in the same manner as the formal study patients. Thus, 14 patients with echogenic foci on transvaginal sonography underwent oophorectomy and constituted the study group. The echogenic foci were bilateral in 9 patients and unilateral in 5. The 5 patients with unilateral echogenic foci included the 1 patient who did not participate in the formal study but for whom the surgeon requested localization of the echogenic focus. In this case, the surgery was done for a contralateral 20-cm cystic mass (that was proven to be a serous cystadenoma); therefore, we cannot be certain that the echogenic focus was truly unilateral in this case. Thus, there were 23 ovaries with echogenic foci and histologic correlation. The mean age ± SD of the 14 patients was 49.0 ± 8.4 years (range, of years). Seven patients were premenopausal, 2 were perimenopausal, 4 were postmenopausal, and menopausal status was unknown in 1 patient who was 54 years of age. The indications for surgery were variable (Table 1). Table 1. Indications for Surgery in the 14 Patients With Echogenic Foci Indication The largest echogenic focus in each ovary measured between 1 and 3 mm in all cases. None of the echogenic foci had distal acoustic shadowing. In 2 ovaries (in 2 different patients), a few of the echogenic foci had a short distal comet tail artifact, but most echogenic foci in these ovaries had neither type of distal acoustic artifact. There was neither type of distal acoustic artifact in the remaining 21 ovaries. The mean number of echogenic foci per ovary was 8.7 ± 7.4 (range, 1 30). The echogenic foci were peripheral in 17 ovaries, central in 1, and both central and peripheral in 5. Of the 22 ovaries with peripheral echogenic foci, 20 involved less than one third of the surface area of the ovary, and 2 involved one third to two thirds of the surface area of the ovary. In no case did the echogenic foci involve greater than two thirds of the surface area of the ovary. A potential cause (Table 2) of the echogenic foci was identified in 17 (74%) of the 23 ovaries. No cause was found in the other 6 cases (26%). There were 5 cases (22%) with calcification, most frequently associated with inclusion cysts (Figure 1) but also with adenofibroma and flecks of surface calcification of unclear cause. The most frequent sole cause of an echogenic focus was a corpus albicans with hemosiderin (Figure 2), which occurred in 6 cases (26%). Both calcification and hemosiderin in the corpus albicans were observed in 2 cases (9%). A cluster of inclusion cysts alone was observed in 1 case (4%) and inclusion cysts and a corpus albicans with hemosiderin were observed in 1 case (4%). In 1 patient, both ovaries (9%) had nodules of dense cortical tissue, and no other cause for an echogenic focus was detected. No. of Patients Chronic pelvic pain 3 Family history of ovarian cancer (prophylactic oophorectomy) 2 Endometrial cancer 2 Chronic pelvic pain and urinary stress incontinence 1 Enterocele repair 1 Enlarging fibroids 1 Unexplained adenocarcinoma cells on a Papanicolaou test 1 Cervical dysplasia 1 Abnormal vaginal bleeding 1 Ovarian mass* 1 *This is the patient in whom the surgeon and pathologist requested localization of an echogenic focus in a normal ovary contralateral to an ovary that had a cystic mass. J Ultrasound Med 23: ,

4 Small Echogenic Foci in the Ovaries Table 2. Causes of Echogenic Foci Histologic Finding No. of Ovaries Corpus albicans with hemosiderin 6 Inclusion cysts with dystrophic calcification 3 Corpus albicans with calcification and hemosiderin 2 Cortical adenofibroma with calcification 1 Flecks of surface calcification, unclear cause 1 Cortical inclusion cysts 1 Cortical inclusion cysts and corpus albicans with hemosiderin 1 Dense cellular cortical nodules* 2 No cause identified 6 *Multiple dense cortical nodules measuring up to 3 mm were present in both ovaries of 1 patient. It is uncertain whether this was the actual cause of the echogenic foci or an incidental finding unrelated to the echogenic foci. The ovarian histologic evaluations revealed benign findings in 21 of the 23 ovaries. One patient with bilateral echogenic foci went to surgery because adenocarcinoma cells were identified on a Papanicolaou test. She was found to have microscopic foci of serous adenocarcinoma on both ovarian surfaces, along with serous adenocarcinoma involving the peritoneum, omentum, pelvic lymph nodes, and serosal surfaces of the fallopian tubes. Surgically and histologically, this was thought to represent a primary peritoneal carcinoma. Sonographically, the echogenic foci were not numerous or extensive. Histologically, no cause for an echogenic focus was identified in 1 ovary, whereas a corpus albicans with hemosiderin was identified in the other ovary. No foci of carcinoma were seen adjacent to the corpus albicans with hemosiderin. There were no calcifications or hemosiderin associated with the carcinoma, and there was no endosalpingiosis. Therefore, we think that the occurrence of peritoneal carcinoma in this case was likely coincidental and unrelated to the echogenic foci. None of the patients had endometriosis by surgical inspection. In 21 of the 23 ovaries, there was no histologic evidence of endometriosis. In 1 patient, there were foci of endometriosis on the serosal surface of 1 fallopian tube but no involvement of the ovary. In 1 other patient, endosalpingiosis and endometriosis were observed in 1 ovary. This was the only ovary with endosalpingiosis. Discussion As transvaginal scanning came into more common use, and perhaps related to the improved resolution of higher-frequency transducers, reports appeared of small calcifications in normal ovaries. 3,14 Only a few of these earlier reported cases had histologic evaluation. Initial reports suggested that these echogenic foci were due to calcification in adenofibromas and calcification in a mucinous cystadenoma, 3 although this latter calcification was curvilinear and measured 15 mm, which is larger than the echogenic foci typically seen in the ovaries. The same authors also reported 1 case in which a dermoid developed Figure 1. Echogenic focus in an ovary due to calcification. A, Transvaginal sonogram showing echogenic foci (some indicated by arrows) in an otherwise normal ovary. B, Histologic section (H&E, original magnification 10). Calcification (arrow) is present in an inclusion cyst. A B 310 J Ultrasound Med 23: , 2004

5 Brown et al 36 months after an echogenic focus was initially seen, although the echogenic focus may have been separate from the dermoid. 3 A subsequent study with histologic correlation in 10 ovaries found epithelial inclusions cysts with associated calcifications in all ovaries. 2 The authors concluded that sonographically identified peripheral ovarian calcifications correlated with psammomatous calcifications associated with surface epithelial inclusion cysts. The same authors also found 3 ovaries with microscopic adenofibromas, although they doubted that the echogenic foci seen on sonography were due to the adenofibromas. It is a reasonable assumption that echogenic foci are due to calcifications. It is known that calcifications in various organs appear sonographically as brightly echogenic areas. Although initial studies regarding echogenic foci in the ovaries suggested that calcifications were the cause, there was no direct sonographic-histologic correlation. A more recent study with a direct sonographic-histologic correlation did not find evidence of calcification. Instead, the authors found inclusion cysts or clusters of inclusion cysts and concluded, on the basis of the absence of identifiable calcification and on the basis of an in vitro model, that the echogenic foci were due to specular reflectors from the walls of tiny unresolved cysts. 1 Thus, several causes have now been suggested as the explanation for small echogenic foci in the ovaries. Calcification may be a cause, either in inclusion cysts (with or without endosalpingiosis), in adenofibromas, or, on the basis of our study, in corpora albicantia. Specular reflectors from the walls of tiny unresolved cysts have also been suggested as the cause, although they were not a frequent finding in our study. On the basis of our study, we think that focal deposits of hemosiderin, usually related to a corpus albicans, may also be a cause. It is not unreasonable to think that hemosiderin could cause a small echogenic area. The iron in hemosiderin, as in other minerals such as calcium, may have enough of an acoustic impedance difference to generate a bright echo. There is indirect evidence in another organ that hemosiderin may cause echogenic foci on sonography. Siderotic nodules, also known as Gamna-Gandy bodies, may occur in the spleens of patients with portal hypertension. They are due to areas of organized hemorrhage and histologically contain hemosiderin and sometimes also calcification. 15,16 A recent prospective study found that sonography had a sensitivity of 71% for Gamna-Gandy bodies. 17 On sonography, Gamna-Gandy bodies appear as small hyperechoic foci, 17 similar in appearance to the echogenic foci seen in the ovaries. Although hemosiderin seems to be universally Figure 2. Echogenic focus in an ovary due to hemosiderin. A, Transvaginal sonogram showing echogenic foci (some indicated by arrows) in an otherwise normal ovary. B, Histologic section (H&E, original magnification 5). Three general regions are shown in this section. There is a corpus albicans (CA) with fibrosis, associated with breakdown products of blood (ie, hemosiderin) in macrophages. The hemosiderin (H, roughly the central region of the area of hemosiderin-laden macrophages) appears as the darker brown areas in the center. Background ovarian cortex (O) is present in the remainder. A B J Ultrasound Med 23: ,

6 Small Echogenic Foci in the Ovaries present in these siderotic nodules, calcification may also occur in some cases; hence one might wonder whether the sonographically identified echogenic foci are due to calcification or to hemosiderin. Although not absolute proof, the presence of low-signal foci on magnetic resonance imaging in 8 cases, none of which had calcification identified on computed tomography, suggests that this imaging finding is more likely due to hemosiderin than to calcification. 16 None of the echogenic foci in our study were related to endometriosis. Echogenic foci have been reported in the walls of 35% of endometriomas. 18 The cause of the echogenic foci was not determined in that study. It has also been suggested that echogenic foci in the ovaries are predictive of diffuse endometriosis. 11 We were unable to confirm this association in our study. We found 2 (9%) of 23 cases to have endometriosis, and in 1 of these, the endometriotic foci did not even involve the ovary. It is possible that our small study population may not be representative of the full clinical spectrum of patients who have echogenic ovarian foci. Our study design did not allow us to evaluate whether echogenic foci are more frequent or more numerous in women with endometriosis than in women without endometriosis. Thus, although we cannot confirm an association of echogenic ovarian foci with endometriosis, our study design did not allow us to disprove this association. Our results also suggest that echogenic foci in the ovaries are not an indicator of endosalpingiosis. Of 5 patients with echogenic ovarian foci who underwent oophorectomy in 1 study, 1 patient had endosalpingiosis. 2 No cases of endosalpingiosis were reported in the 7 ovaries with histologic correlation in another study, 1 although it is not known whether this was specifically assessed. One problem with evaluation of the literature in this area is that pathologists vary in whether they report endosalpingiosis. Additionally, there are no universal guidelines to separate multiple cortical inclusion cysts from endosalpingiosis. In any case, because only 1 (4%) of the 23 patients in our study had endosalpingiosis, echogenic ovarian foci do not appear to be a reliable indicator of endosalpingiosis. However, we do not know how frequently echogenic foci are observed in cases of endosalpingiosis. Imaging studies infrequently identify adenofibromas. Adenofibromas are benign, usually unimportant findings, and lesions smaller than 1 cm are usually not considered neoplasms. 19 Calcification in an adenofibroma was an infrequent cause of echogenic foci in our study. Ovarian cortical stroma can be quite variable in appearance. It sometimes is organized in loosely defined nodules in otherwise unremarkable ovaries or in cases of nodular stromal hyperplasia. In 1 of our cases, this nodularity was the only histologic finding of interest. We mention it as a possible explanation but are not sure whether it could cause such a brightly echogenic appearance on sonography. Review of the pathology literature provides no clear evidence as to what may be the cause of the echogenic foci seen on sonography. Calcifications are present in some ovarian neoplasms. For ovarian calcifications unassociated with a neoplasm, most seem to be associated with cortical inclusion cysts. 20 Some of these calcifications may be too small to be identified sonographically, however. A case of extensive psammomatous calcifications replacing most of the ovarian stroma has been reported, 20 but the cause was unknown, and such extensive calcification is not typical of what was observed in our study. Limitations of our study include the relatively small number of cases, yet, to our knowledge, it is the largest number reported thus far with direct sonographic-histologic correlation. Our inability to show a single cause indicates that there likely are multiple causes. Despite obtaining direct sonographic-histologic correlation, there is a small possibility that our localization method was still not precise enough. We used india ink to mark the ovarian surface over an echogenic focus, as opposed to a prior study, which used a suture soaked in india ink. 1 Although both methods can localize the approximate area of an echogenic focus, it can still be difficult to precisely locate a 1- to 3-mm echogenic focus. Despite any shortcomings in the method of marking, our protocol for submitting the entire ovary for histologic evaluation should have uncovered any possible causes in the sections adjacent to the marked area. It is important to realize that our study only addresses small (1- to 3-mm) echogenic foci in the ovaries. Our study did not evaluate larger echogenic areas that may be calcifications. Also, none of our cases had extensive peripheral echogenic foci. In our clinical practice, we have uncommonly observed numerous echogenic foci around the periphery of the ovary such that 312 J Ultrasound Med 23: , 2004

7 Brown et al they form a nearly continuous rim of increased echogenicity, yet there were no such cases available for inclusion in our study. One case with such extensive circumferential echogenic foci in a patient with ovarian endometriosis was previously reported. 2 The clinical significance of this uncommon form of extensive circumferential echogenic foci could be different than the less extensive cases that are the subject of this report. In conclusion, our results provide evidence that there are multiple causes of echogenic foci in the ovaries. Such causes include calcification (in inclusion cysts, on the ovarian surface, in adenofibromas, or in corpora albicantia), hemosiderin (in corpora albicantia), and possibly clusters of inclusion cysts or dense cortical nodules. We conclude that several small echogenic foci in an otherwise normal ovary are unimportant and do not warrant follow-up. We also found no evidence that these echogenic foci are reliable indicators of endosalpingiosis or endometriosis. References 1. Muradali D, Colgan T, Hayeems EB, Burns PN, Wilson SR. Echogenic ovarian foci without shadowing: are they caused by psammomatous calcifications? Radiology 2002; 224: Kupfer MC, Ralls PW, Fu YS. Transvaginal sonographic evaluation of multiple peripherally distributed echogenic foci of the ovary: prevalence and histologic correlation. AJR Am J Roentgenol 1998; 171: Brandt KR, Thurmond AS, McCarthy JL. Focal calcifications in otherwise ultrasonographically normal ovaries. Radiology 1996; 198: Ryuko K, Miura H, Abu-Musa A, Iwanari O, Kitao M. Endosalpingiosis in association with ovarian surface papillary tumor of borderline malignancy. Gynecol Oncol 1992; 46: Tutschka BG, Lauchlan SC. Endosalpingiosis. Obstet Gynecol 1980; 55:57S 60S. 6. dehoop TA, Mira J, Thomas MA. Endosalpingiosis and chronic pelvic pain. J Reprod Med 1997; 42: Scully RE. Pathology of ovarian cancer precursors. J Cell Biochem Suppl 1995; 23: Resta L, Russo S, Colucci GA, Prat J. Morphologic precursors of ovarian epithelial tumors. Obstet Gynecol 1993; 82: Schuldenfrei R, Janovski NA. Disseminated endosalpingiosis associated with bilateral papillary serous cystadenocarcinoma of the ovaries. Am J Obstet Gynecol 1962; 84: Kerr L, Gabas F, Machado F, Cukier E. Transvaginal diagnosis of diffuse endometriosis [abstract]. J Ultrasound Med 2002; 21(suppl):S Lafortune M, Gariepy G, Dumont A, Breton G, Lapointe R. The v-shaped artifact of the gallbladder wall. AJR Am J Roentgenol 1986; 147: Shapiro RS, Winsberg F. Comet-tail artifact from cholesterol crystals: observation in the postlithotripsy gallbladder and an in vitro model. Radiology 1990; 177: Sherer DM, Allen TA, Abulafia O. Asymptomatic calcifications of a normal-sized ovary [letter]. J Ultrasound Med 1993; 12: Minami M, Itai Y, Ohtomo K, et al. Siderotic nodules in the spleen: MR imaging of portal hypertension. Radiology 1989; 172: Sagoh T, Itoh K, Togashi K, et al. Gamna-Gandy bodies of the spleen: evaluation with MR imaging. Radiology 1989; 172: Chan Y, Yang W, Sung J, Lee Y, Chung S. Diagnostic accuracy of abdominal ultrasonography compared to magnetic resonance imaging in siderosis of the spleen. J Ultrasound Med 2000; 19: Patel MD, Feldstein VA, Chen DC, Lipson SD, Filly RA. Endometriomas: diagnostic performance of ultrasound. Radiology 1999; 210: Czernobilsky B. Cystadenofibroma, adenofibroma, and malignant adenofibroma of the ovary. Pathol Annu 1977; 12: Clement PB, Cooney TP. Idiopathic multifocal calcification of the ovarian stroma. Arch Pathol Lab Med 1992; 116: Laufer MR, Heerema AE, Parsons KE, Barbieri RL. Endosalpingiosis: clinical presentation and followup. Gynecol Obstet Invest 1998; 46: J Ultrasound Med 23: ,