Sonographic Differentiation of Benign and Malignant Cystic Lesions of the Breast

Transcription

1 Article Sonographic Differentiation of Benign and Malignant Cystic Lesions of the Breast Yun-Woo Chang, MD, PhD, Kwi Hyang Kwon, MD, Dong Erk Goo, MD, Deuk Lin Choi, MD, Hye Kyung Lee, MD, Seung Boo Yang, MD Objective. The purpose of this study was to subdivide the types of sonographic findings of benign versus malignant cystic masses and to determine appropriate patient care according to the sonographic findings with pathologic correlation. Methods. The sonographic findings of 175 symptomatic cystic breast lesions were pathologically proven and reviewed retrospectively. Cystic lesions were classified as 6 types: simple cysts (type I), clustered cysts (type II), cysts with thin septa (type III), complicated cysts (type IV), cystic masses with a thick wall/septa or nodules (type V), and complex solid and cystic masses (type VI). Sonographic findings were compared with the pathologic results and were evaluated according to the incidence of benign and malignant masses. Results. All 23 type I, 15 type II, 22 type III, and 35 type IV cases were pathologically proven to be benign. Seven (25.9%) of the 27 type V cases and 33 (62.3%) of the 53 type VI cases were proven to be malignant. We analyzed the shapes and margins of 80 cases of cystic masses with a solid component (types V and VI); 16 (44%) of 36 sonographically circumscribed masses were malignant. Conclusions. Because the sonographically detected simple cysts (type I), clustered cysts (type II), and cysts with thin septa (type III) were all benign, annual routine follow-up appears reasonable. Symptomatic complicated cysts (type IV) should be aspirated and appropriately treated according to clinical symptoms. Cystic masses with a solid component (types V and VI) should be examined by biopsy with pathologic confirmation. Key words: breast; cystic mass; sonography. Abbreviations BI-RADS, Breast Imaging Reporting and Data System Received June 21, 2006, from the Department of Radiology, College of Medicine, Soonchunhyang University Hospital, Seoul, Korea (Y.-W.C., K.H.K., D.E.G., D.L.C.); Department of Radiology, Soonchunhyang University Hospital, Bucheon, Korea (H.K.L.); and Department of Radiology, Soonchunhyang Hospital, Gumi, Korea (S.B.Y.). Revision requested July 11, Revised manuscript accepted for publication August 23, We thank Bonnie Hami (Department of Radiology, University Hospital of Cleveland, Cleveland, OH) for editorial assistance. Address correspondence to Yun-Woo Chang, MD, PhD, Department of Radiology, Soonchunhyang University Hospital, 22 Dasagwan-gil, Yongsan-ku, Seoul , Korea. ywchang@hosp.sch.ac.kr Breast cysts, including simple cysts, are common findings in women older than 40 years who have undergone sonography. Simple cysts have no potential malignancy, but solid and cystic masses may cause diagnostic dilemmas. 1,2 Various types of cystic lesions of the breast have been described by Berg et al 1 ; complex cystic lesions were proven malignant in 23% of the patients in their study. Intracystic carcinomas of the breast are rare entities that constitute 0.3% to 2.0% of all breast carcinomas, and the solid portion may be quite large at the time of diagnosis. 3 We attempted to classify the types of sonographic findings of symptomatic cystic lesions of the breast and correlated these findings with the pathologic results. We tried to evaluate the characteristic features of benign versus malignant cystic masses and to determine appropriate patient care according to the sonographic findings by the American Institute of Ultrasound in Medicine J Ultrasound Med 2007; 26: /07/$3.50

2 Benign and Malignant Cystic Lesions of the Breast Materials and Methods From June 2002 through March 2006, during which time 57,437 female patients underwent sonography of the breast at our institution, we determined that 212 symptomatic patients had cystic breast masses. One hundred seventy-five patients with breast cystic masses were pathologically confirmed, and 37 patients with cysts were followed without pathologic confirmation during the following 4 years by 3 to 4 subsequent sonographic examinations. Recording and reviewing the results of these examinations and the results of subsequent imaging, clinical, and pathologic follow-up was approved by our institutional database. Pathologic confirmation was performed by fine-needle aspiration (n = 123), core needle biopsy (n = 39), or excision (n = 13). All patients had symptoms such as a palpable mass, a lump, breast pain, nipple discharge, or a red skin change. The mean age of the patients was 44.6 years (range, years). The mean diameter of the lesions was 25 mm (range, mm). Diagnosis was established with aspiration using an 18- to 20-gauge needle in 123 lesions from which fluid was sent for cytologic examination and was followed by core needle biopsy or excision in 57 lesions because of the persistent existence of a solid component. Sonographically guided core needle biopsy was performed in 39 lesions with a 14-gauge automated biopsy gun (Pro-Mag 2.2; Manan Medical Products, Northbrook, IL), and excision biopsy was performed in 13 lesions. Sonography was performed by a physician using a broad-bandwidth linear array transducer with a center frequency of 10 MHz supplemented by a transducer with a center frequency of 7.5 MHz (LOGIQ 700 Expert Series; GE Healthcare, Milwaukee, WI; or HDI 5000; Philips Medical Systems, Bothell, WA). Cystic lesions were classified into 6 types (types I VI), which were slightly modified from those mentioned in an article by Berg et al 1 (Figure 1). Simple cysts (type I) were defined as anechoic masses with an imperceptible, circumscribed border and acoustic enhancement. Clustered cysts (type II) were defined as clustered anechoic cysts with no discrete solid components. Cysts with thin septa (type III) were defined as cysts within septa of less than 0.5 mm in thickness. Figure 1. Subtypes of cystic masses of the breast. Type I indicates simple cyst; type II, clustered cysts; type III, cyst with thin septa; type IV, complicated cyst; type V, cyst with a thick wall/septa or nodules; and type VI, complex solid and cystic mass. Complicated cysts (type IV) were defined according to the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) 4 as lesions with homogeneous low-level echoes that otherwise meet the criteria of simple cysts, including cystic lesions containing fluid-debris levels or floating echogenic debris. Cystic masses with septa or a wall greater than 0.5 mm in thickness or mixed cystic and solid masses with at least a 50% cystic component were classified as cystic masses with a thick wall/septa or nodules (type V). Primarily solid masses with eccentric cystic foci were considered complex solid and cystic masses (type VI). Two expert radiologists classified the types of cystic breast lesions according to the sonographic findings in consensus. The sonographic findings of each type of cystic breast lesion were compared with the pathologic results and were evaluated according to the incidence of benign and malignant masses. We analyzed the shapes and margins of the cystic lesions with solid components as 2 types of type V (cystic mass with a thick wall/septa or nodules) and type VI (complex solid and cystic masses) and correlated with the pathologic results. Ninety-four lesions, including 37 lesions classified as sonographic type I lesions without pathologic confirmation, were then followed 1 to 5 times during the next 45 months. Results In Table 1, the classifications and methods of examination are summarized. Among 175 lesions, 23 type I cases, 15 type II cases, 22 type III cases, 48 J Ultrasound Med 2007; 26:47 53

3 Chang et al and 35 type IV cases were proven to be benign. Seven type V cases and 33 type VI cases were proven to be malignant. All 23 type I, 15 type II, and 22 type III cases were pathologically confirmed as cysts or fibrocystic disease (Figures 2 and 3). One type I case and 1 type III case were confirmed as fat necrosis, and 1 type III case was proven to be a mucocelelike tumor. Thirty-five cases with the sonographic appearances of type IV were proven to be benign, including 4 cysts, 9 fibrocystic changes, 21 abscesses, and 1 mucocelelike tumor (Figure 4). All cases had symptoms, for example, a palpable mass, breast pain, or skin redness. Twenty-one cases were confirmed as abscesses by fine-needle aspiration or biopsy. Subsequent excision and drainage were performed after antibiotic treatment. Among the 27 cases of type V cystic masses, 20 (74.1%) were proven to be benign (3 cysts, 3 fibrocystic changes, 7 abscesses, 2 mucocelelike tumors, 1 fibroadenoma, and 4 papillomas), and 7 (25.9%) were proven to be malignant (5 invasive ductal carcinomas and 2 papillary carcinomas) (Figure 5). Among the 53 sonographic type VI cystic masses, 20 (37.7%) were proven to be benign (7 fibrocystic changes, 2 abscesses, 8 fibroadenomas, 2 papillomas, and 1 phyllodes tumor), and 33 (62.3%) were proven to be malignant (19 invasive ductal carcinomas, 2 ductal carcinomas in situ, 4 metaplastic carcinomas, 5 malignant phyllodes tumors, 1 papillary carcinoma, and 2 mucinous carcinomas) (Table 2). There were 80 cases of cystic masses with solid components, composed of types V and VI, and 40 (50%) lesions were proven malignant (Table 3). All the lesions Figure 2. Transverse sonogram from a 39-year-old woman with nipple discharge shows aggregate cysts of various diameters but without a discrete solid component, classified as type II (arrowheads). Aspiration cytologic examination revealed a fibrocystic change. were composed of palpable masses, including those in 6 patients with breast pain and in 2 patients with bloody nipple discharge. When we analyzed the shape and margin of these 80 lesions, 16 (44%) of 36 lesions composed of round, oval, or lobular masses with sonographically circumscribed margins were proven malignant (Figure 6). Surgery was performed in 32 cases (16 mastectomies and 16 breast conservation surgeries with axillary lymph node dissection). Figure 3. Radial sonogram from a 38-year-old woman shows a cyst with thin (<0.5-mm) septa, classified as type III (arrowheads), which otherwise met the criteria for a simple cyst. The cyst disappeared after aspiration, and the pathologic result was benign cyst contents. Table 1. Subclassification, Method of Sampling, and Rates of Malignancy in 175 Cystic Lesions Rate of Sonographic Malignancy, Feature Aspiration CNB Excision n (%) Type I (n = 23) Type II (n = 15) Type III (n = 22) Type IV (n = 35) Type V (n = 27) (17.5) Type VI (n = 53) (82.5) Total (n = 175) (100) CNB indicates core needle biopsy. J Ultrasound Med 2007; 26:

4 Benign and Malignant Cystic Lesions of the Breast Discussion Figure 4. Transverse sonogram from a 34-year-old woman with complicated cysts (type IV) shows well-defined oval masses with homogeneous internal echoes (arrowheads). Aspiration yielded 2 ml of milky fluid, and subsequent excisional biopsy revealed acute and chronic nonspecific inflammation. In 94 cases, follow-up sonography was performed over 45 months, during which time the cysts underwent no appreciable changes or had disappeared at the time of follow-up examination. One patient (1%) of 94 was proven to have a newly detected mass, BI-RADS category 4, separate from the cyst, which was confirmed as invasive ductal carcinoma on follow-up sonography after the 36-month follow-up. Figure 5. Transverse sonogram from a 39-year-old woman with a thick-walled cystic mass (type V) shows a well-circumscribed oval cystic mass with thick septa (arrowheads). Aspiration yielded bloody fluid with atypical cells, and subsequent core needle biopsy revealed cystic degeneration of high-nuclear-grade invasive ductal carcinoma. Breast cysts are common findings in women older than 40 years, and most can be dismissed as benign (ie, BI-RADS category 2). Type I lesions, composed of simple cysts, do not require intervention if a patient is not symptomatic because such cysts have no potential malignancy. 4 7 If patients have symptoms such as pain or palpation owing to a very large cyst, aspiration can be performed on an elective basis. 1 Because type II lesions, composed of clustered cysts without a solid component, are considered benign, routine follow-up is usually recommended. Berg 8 stated that clustered microcysts without a solid component are likely to be benign, on the basis of their study of 79 lesions with followup, which were not proven malignant. These lesions frequently occur in conjunction with apocrine metaplasia or fibrocystic changes, and apocrine metaplasia appears to be the progenitor of cyst formation in which adjacent acini unfold and fuse because of increased intraluminal pressure from secretions of forming apocrine-lined microcysts. 9 It is possible that larger type III cysts with thin septa represent the continuum of the spectrum from apocrine metaplasia to cysts as the acini fuse. In our study, 15 sonographic clustered cysts revealed cysts or fibrocystic changes and no malignancy during follow-up. Although we defined type IV cysts as complicated lesions with homogeneous low-level echoes that otherwise meet the criteria of simple cysts, with an imperceptible wall and a fluid-debris level, 21 (60%) of the 35 cases in our series were proven to be abscesses. When an abscess was suspected clinically on imaging or on inspection of the fluid, we performed aspiration for confirmation and initiated a course of antibiotics or surgical drainage. In a series by Venta et al, 7 0.3% of 308 complicated cysts proved to be malignant; however, these are usually managed with periodic follow-up imaging studies because they are probably benign lesions. Buchberger et al 10 found none of 133 such lesions to be malignant, and Kolb et al 11 found none of 126 such lesions to be malignant in their series with screening sonography. Symptomatic complicated cysts should be managed on the basis of clinical symp- 50 J Ultrasound Med 2007; 26:47 53

5 Chang et al Table 2. Correlation of Pathologic Outcome and Sonographic Features for 175 Cystic Lesions Finding Type I Type II Type III Type IV Type V Type VI Total, n (%) Benign Cyst (18.9) Fibrocystic disease (28.6) Fat necrosis (1.1) Abscess (17.1) Mucocelelike tumor (2.3) Fibroadenoma (5.2) Papilloma (3.4) Phyllodes (0.6) Total, n (%) 23 (100) 15 (100) 22 (100) 35 (100) 20 (74.1) 20 (37.7) 135 (77.2) Malignant Infiltrative ductal carcinoma (13.7) Ductal carcinoma in situ (1.1) Metaplastic carcinoma (2.3) Malignant phyllodes (2.9) Papillary carcinoma (1.7) Mucinous carcinoma (1.1) Total, n (%) (25.9) 33 (62.3) 40 (22.8) Total, n (%) 23 (100) 15 (100) 22 (100) 35 (100) 27 (100) 53 (100) 175 (100) toms and generally warrant aspiration, with abscesses, hematomas, fat necrosis, and galactoceles being included in the differential diagnosis. Type V lesions, composed of a cystic mass with a thick wall, septa, or nodules, should suggest possible malignancy, and biopsy should be performed. Berg et al 1 found that 35% of cystic masses with a thick wall or thick septa were malignant, with 86% being high-grade invasive ductal carcinomas and 33% having circumscribed margins on sonography. In the case of type V, core needle biopsy of the wall/septum or nodule is preferred to aspiration for providing a specific diagnosis because the cystic component can be necrosis or acellularity. In our study, among the 27 type V cases, 7 (25.9%) were proven to be malignant, 5 (71%) of which were invasive ductal carcinomas and 2 of which were papillary carcinomas. Abscesses, apocrine metaplasia, inflamed or ruptured cysts or ducts, and hematomas can also present as thick-walled cysts. Fat necrosis can manifest as a thick-walled cystic lesion or as a complex cystic and solid mass. 1,12 In type VI complex solid and cystic masses, eccentric cystic foci can be caused by duct dilatation, acini, or necrosis. Jackson et al 13 first described that fibroadenomas rarely have eccentric cystic foci. The consistent feature of a fibroadenoma with the presence of cystic foci may suggest a possible phyllodes tumor, although such tumors are rare. Liberman et al 14 proposed that tumors with cystic foci are more common in malignant phyllodes tumors. According to Berg et al, 1 malignancies with eccentric cystic foci have no particularly distinguishing features in lowand high-grade invasive ductal carcinoma. In our study, there was no difference in the cellular degree of invasive ductal carcinoma according to the type VI complex solid and cystic masses. However, complex solid and cystic masses were proven to be metaplastic carcinomas, malignant phyllodes tumors, and mucinous carcinomas, which may represent malignant masses with cystic components. Papillary carcinoma can be seen in cystic masses of the breast. The prognosis of cystic malignant masses is better than for other Table 3. Sonographic Findings of 80 Cystic Masses With Solid Components and Rates of Malignancy Cystic Masses With Margins Solid Components Malignant, n (%) Circumscribed (44) Round 12 6 (50) Oval 24 7 (29) Lobular 9 3 (33) Irregular 0 0 (0) Indistinct 4 2 (50) Angular 9 6 (67) Microlobulated (71) Spiculated 1 1 (100) Total (50) J Ultrasound Med 2007; 26:

6 Benign and Malignant Cystic Lesions of the Breast Figure 6. Transverse sonogram from a 57-year-old woman shows a well-circumscribed round solid mass with small internal cystic components, classified as type VI (arrowheads). The pathologic result was intermediate-nuclear-grade invasive ductal carcinoma. forms of breast cancer, and when the internal fluid is aspirated, it usually contains blood. 3,15 17 The hypotheses of malignant masses with cystic components have been followed as several research studies investigated the chance invasion of a carcinoma into an area of cystic disease and the cystic degeneration of a high-grade malignancy. 3,17,18 Papillomas or papillary carcinomas can be seen in cystic masses of the breast. Intracystic papillary carcinomas account for 0.3% of all breast cancers. 19 Hong et al 20 showed that 16 (9%) of 372 masses described as both oval and circumscribed on sonography were malignant. In our study, among the 80 cases of cystic masses with a solid component, 40 cases (50%) were proven to be malignant, and 16 (20%) sonographically detected circumscribed masses of 80 lesions were proven to be malignant. This relatively high malignancy rate was likely due to the exclusive inclusion of symptomatic patients and likely included the various pathologic types of benign-looking malignant masses (eg, metaplastic carcinomas, papillary carcinomas, mucinous carcinomas, and malignant phyllodes tumors). In our series, adequate follow-up was achieved in 97 lesions classified as cystic masses without a solid component, and 1 patient was proven to have had invasive ductal carcinoma on follow-up sonography 36 months after examination. However, on follow-up sonography, this lesion was proven to be a newly developed 5-mm solid mass separate from the cyst. This study had several limitations in that the cases were selected by retrospectively proven disease and interpretation of static images, although static images represent a common method for retrospectively interpreting clinical breast sonograms, and there were only a few cases of each of the classified types. Validation of this approach from multiple centers is needed. However, it is notable that the differentiation of cystic masses is estimated by classifying the sonographic findings of benign versus malignant masses with pathologic correlation, and patient care is subsequently recommended according to the classified types of cystic masses of the breast. In summary, because the sonographically detected simple cysts (type I), clustered cysts (type II), and cysts with thin septa (type III) were all benign in our study, interventional treatment was not needed, but annual routine follow-up appears reasonable for such lesions. Symptomatic complicated cysts (type IV) should be managed by aspiration cytology or treatment according to clinical symptoms. Cystic masses with a thick wall/septa or nodules (type V) and complex solid and cystic masses (type VI) should undergo biopsy with pathologic confirmation, even if these masses have oval and well-circumscribed margins. References 1. Berg WA, Campassi CI, Loffe OB. Cystic lesions of the breast: sonographic-pathologic correlation. Radiology 2003; 227: Bassett LW. Imaging of breast masses. Radiol Clin North Am 2000; 38: Omiri LM, Hisa N, Ohkuma K, et al. Breast masses with mixed cystic-solid sonographic appearance. J Clin Ultrasound 1993; 21: American College of Radiology. Breast Imaging Reporting and Data System (BI-RADS) Ultrasound. Reston, VA: American College of Radiology; Mendelson EB, Berg WA, Merritt CR. Toward a standardized breast ultrasound lexicon, BI-RADS: ultrasound. Semin Roentgenol 2001; 36: Hilton SV, Leopold GR, Olson LK, Willson SA. Real-time breast sonography: application in 300 consecutive patients. AJR Am J Roentgenol 1986; 147: Venta LA, Kim JP, Pelloski CE, Morrow M. Management of complex breast cysts. AJR Am J Roentgenol 1999; 173: J Ultrasound Med 2007; 26:47 53

7 Chang et al 8. Berg WA. Sonographically depicted breast clustered microcysts: is follow-up appropriate? AJR Am J Roentgenol 2005; 185: Warner JK, Kumar D, Berg WA. Apocrine metaplasia: mammographic and sonographic appearances. AJR Am J Roentgenol 1998; 170: Buchberger W, DeKoekkoek-Doll P, Springer P, Obrist P, Dunser M. Incidental findings on sonography of the breast: clinical significance and diagnostic workup. AJR Am J Roentgenol 1999; 173: Kolb TM, Lichy J, Newhouse JH. Occult cancer in women with dense breasts: detection with screening US-diagnostic yield and tumor characteristics. Radiology 1998; 207: Soo MS, Kornguth PJ, Hertzberg BS. Fat necrosis in the breast: sonographic features. Radiology 1998; 206: Jackson VP, Rothschild PA, Kreipke DL, Mail JT, Holden RW. The spectrum of sonographic findings of fibroadenoma of the breast. Invest Radiol 1986; 21: Liberman L, Bonaccio E, Hamele-Bena D, Abramson AF, Cohen MA, Dershaw DD. Benign and malignant phyllodes tumors: mammographic and sonographic findings. Radiology 1996; 198: Kersschot EA, Hoste MV, Dochez CJ, van Marck EA, De Schepper AM, Van Goethem ML. Intracystic carcinoma of the breast. Rofo 1986; 144: Reuter K, D Orsi CJ, Reale F. Intracystic carcinoma of the breast: the role of ultrasonography. Radiology 1984; 153: Czernobilsky B. Intracystic carcinoma of the female breast. Surg Gynecol Obstet 1967; 124: Ravichandran D, Carty NJ, al-talib RK, Rubin C, Royle GT, Taylor I. Cystic carcinoma of the breast: a trap for the unwary. Ann R Coll Surg Engl 1995; 77: Knelson MH, el Yousef SJ, Goldberg RE, Balance W. Intracystic papillary carcinoma of the breast: mammographic, sonographic, and MR appearance with pathologic correlation. J Comput Assist Tomogr 1987; 11: Hong AS, Rosen EL, Soo MS, Baker JA. BI-RADS for sonography: positive and negative predictive values of sonographic features. AJR Am J Roentgenol 2005; 184: J Ultrasound Med 2007; 26: