Intraductal Mass on Breast Ultrasound: Final Outcomes and Predictors of Malignancy

Transcription

1 Women s Imaging Original Research Kim et al. Intraductal Mass on Breast Ultrasound Women s Imaging Original Research Won Hwa Kim 1 Jung Min Chang 1 Woo Kyung Moon 1 Nariya Cho 1 Ann Yi 2 Hye Ryoung Koo 1 Seung Ja Kim 2 Kim WH, Chang JM, Moon WK, et al. Keywords: BI-RADS, breast, intraductal mass, papillary lesion, ultrasound DOI: /AJR Received April 18, 2012; accepted after revision June 28, Supported by grant 2012K from the Converging Research Center Program through the Ministry of Education, Science, and Technology, Korea. 1 Department of Radiology, Seoul National University Hospital, 28, Yongon-dong, Chongno-gu, Seoul, , Korea. Address correspondence to J. M. Chang (imchangjm@ gmail.com). 2 Department of Radiology, Seoul Metropolitan Government, Seoul National University, Boramae Medical Center, Seoul, Korea. AJR 2013; 200: X/13/ American Roentgen Ray Society Intraductal Mass on Breast Ultrasound: Final Outcomes and Predictors of Malignancy OBJECTIVE. The purpose of this study was to retrospectively investigate the final outcomes of intraductal masses on breast ultrasound and determine the clinical and radiologic variables associated with malignancy. MATERIALS AND METHODS. A database search ( ) was performed to find patients who had an intraductal mass on breast ultrasound. Histopathologic or ultrasound follow-up (> 24 months) data were available from 147 women (mean age, 49.8 years) with 163 intraductal masses. Clinical and radiologic variables (age, symptom, personal and family history, lesion size, and distance from the nipple) and pathologic results were collected. Ultrasound features of the intraductal masses were reviewed by two radiologists in consensus and classified into three morphologic types: mass incompletely filling the duct, mass completely filling the duct, and mass extending outside the duct. Involvement of a branch duct was also analyzed. Associations between variables and final outcomes were analyzed using chi-square tests and Student t tests. RESULTS. Thirteen (8%) of the 163 intraductal masses were malignant (10 ductal carcinomas in situ and three invasive ductal carcinomas). Malignancy was significantly associated with symptoms (p = 0.008) and personal history of breast cancer (p < 0.007). Malignant intraductal masses were larger than benign intraductal masses (1.4 cm vs 0.9 cm, p = 0.02). Malignant intraductal masses tended to fill the duct more completely or extend outside the duct (p < 0.001), and they more frequently involved the branch duct (p < 0.001) than did the benign intraductal masses. CONCLUSION. Our study showed that 8% of intraductal masses are malignant. Symptoms, personal history, lesion size, and ultrasound features can be possible predictors of malignancy. I ntraductal masses in the breast are intraluminal solid masses in the lactiferous duct of the breast. Intraductal masses are best known as a cause of pathologic nipple discharge [1, 2]. Although galactography, MRI, and ductoscopy are methods that can identify intraductal abnormalities [3 5], ultrasound is also considered a valuable tool [2, 6, 7]. Recent advances in breast ultrasound technology, along with high-resolution transducers (10 13 MHz), have enabled ultrasound to clearly visualize the ductal system, including the normal lactiferous duct in the breast. These advances and the subsequent increase in breast ultrasound examinations have allowed the detection of unprecedented numbers of masses, including intraductal masses [8]. Currently, according to the BI-RADS used in the American College of Radiology Imaging Network (ACRIN) 6666 trial, intra- ductal masses are one of the special cases. Final assessment of these masses is recommended to be in subcategory 4A, which indicates a need for biopsy [9, 10]. To our knowledge, there have been no published reports of the final outcomes of intraductal masses detected by breast ultrasound, and it is still not clear whether all intraductal masses should be classified into BI-RADS category 4A because there are no defined clinical and radiologic predictors for malignancy. Thus, the purpose of this study was to retrospectively investigate the final outcomes of intraductal masses visualized on breast ultrasound and determine which clinical and radiologic variables may be associated with malignancy. Materials and Methods This retrospective study was approved by the institutional review board of our institution, and the requirement for informed patient consent was waived. 932 AJR:200, April 2013

2 Intraductal Mass on Breast Ultrasound Patient Selection Radiologic records of all breast ultrasound examinations performed at our metropolitan hospital between January 2006 and December 2008 were searched for the term intraductal. In our institution, screening breast ultrasound examinations were performed for patients with a personal or family history of breast cancer and for those with dense breasts. Among a total of 77,952 breast ultrasound examinations in 42,394 patients performed in our hospital, this search yielded 1334 ultrasound examinations in 830 patients who had radiologic reports that included the term intraductal, and we identified 258 patients (0.6%) who had intraductal masses by breast ultrasound. Of these, 111 patients who did not undergo biopsy or surgery or had imaging follow-up for less than 24 months were excluded. There were 14 patients who had two lesions and one who had three lesions. Thus, 163 intraductal masses from 147 patients were included in this study. The mean age of the patients was 49.8 years (range, years). Ninety-two (56%) lesions of the total 163 intraductal masses were associated with symptoms, including nipple discharge (n = 79), palpable mass (n = 8), or both palpable mass and nipple discharge (n = 5). Seventy-one (44%) lesions were detected in asymptomatic patients. Mammograms were available for 133 women with 148 lesions during the ultrasound examinations. Lesions were observed as a mass in 13 cases (9%), as microcalcifications without a mass in seven (5%), as a mass with microcalcification in one (1%), as a focal asymmetry in 15 cases (10%), and as a focal asymmetry with microcalcification in two cases (1%). No mammographic abnormalities were found for 110 lesions (74%). Six patients had a history of breast cancer, and three patients had a family history of breast cancer in their first-line relatives. Ultrasound Ultrasound was performed by one of five radiologists with 5 11 years of experience in breast imaging using a 14-5 MHz linear transducer on an IU-22 scanner (Philips Healthcare) or EUB-8500 scanner (Hitachi Medical). All breast ultrasound examinations were performed using handheld real-time ultrasound. Bilateral whole-breast scanning was performed in every examination for both screening and diagnostic purposes. The scanning technique was standardized for screening ultrasound: scanning of the breast in the transverse and sagittal orientations, inner aspect of the breast in a supine position, and outer aspect in supine oblique position with the patient s arm raised above the head. When breast nodules or masses were visible, radial and orthogonal antiradial plane scanning was performed to evaluate the mass for a decision on the precise location and associated abnormality of the intraductal or periductal area. Scanning the axilla was routinely included before or after scanning the breast in our protocol. The average time to perform a screening ultrasound examination was approximately minutes. For diagnostic purpose, additional targeted scanning was performed after routine screening ultrasound. In cases of palpable abnormality, the radiologist palpated the lesion before and during scanning of the lesion, and targeted scanning to the area of palpable abnormality was performed simultaneously palpating the area. For the patients with nipple discharge, radiologists were instructed to systematically scan the periareolar region using radial/antiradial plane scanning to lay the probe in parallel with the long and short axes of the lactiferous duct. In addition, the probe was also carefully tilted beneath the nipple-areola complex to reduce the nipple shadow effects. Finally, in cases of patients referred to have suspicious findings on ultrasound, targeted scanning at the area of concern was carried out after the routine whole-breast scanning. Image Evaluation Ultrasound features of intraductal masses were retrospectively reviewed in consensus without clinical or pathologic information by two dedicated breast imaging radiologists with 5 and 11 years of experience. The morphologic type of the intraductal masses was classified into three groups depending on the degree of expansion or filling of the lactiferous duct by the mass: type 1, mass incompletely filling the duct; type 2, mass completely filling the duct; and type 3, mass extending outside the duct (Fig. 1A). Whether the mass involved the branch duct was also recorded (Fig. 1B). Retrospective reviews of the ultrasound images were performed with digitized static images on a PACS with a 5000-pixel monochrome liquid crystal display monitor (ME1i2-BC, Totoku Electric). In case of multiple lesions, the review was performed on a per-lesion basis. Biopsy Ultrasound-guided core needle biopsy was performed with high-resolution ultrasound units with 14-5 MHz linear transducer using a freehand technique. The biopsy procedures were performed by one of five radiologists with 5 11 years of experience in breast imaging. Each radiologist determined the type of biopsy to be performed, with 11-gauge vacuum-assisted biopsy preferred for intraductal lesions. When 14-gauge automated needle biopsy was selected, obtaining five core samples was recommended in most cases. Tissue samples were obtained with an automated biopsy gun with a 14-gauge needle (Pro-Mag 2.2, Manan Medical Products) for 65 lesions (mean core number, 6 ± 1.3; range, 3 10) and an 11-gauge vacuum-assisted probe (Mammotome, Ethicon-Endosurgery) for 68 lesions (mean core number, 8.7 ± 4.0; range, 2 25). The biopsy procedure took approximately minutes. Surgery For lesions that were definitely palpable on clinical examination by the surgeon, palpationguided excision was performed. In cases of nonpalpable lesions, ultrasound-guided preoperative needle localization was performed with a 21-gauge modified Kopans spring-hook localization needle, which was inserted into and through the lesions or lesions with postprocedural changes created by core needle biopsy and vacuum-assisted biopsy. A B Fig. 1 Ultrasound appearance of intraductal masses. A, Schematic drawing shows types of sonographic findings of intraductal masses according to degree of expansion or filling of duct by mass. Type 1, mass incompletely filling duct; type 2, mass completely filling duct; and type 3, mass extending outside duct. B, Schematic drawing shows absence (left) and presence (right) of branch duct involvement. Mass with involvement of branch duct is defined as mass extending into two branch ducts. AJR:200, April

3 Kim et al. For the excised lesions, surgical pathologies and core specimens were reviewed by one pathologist with 25 years of experience in breast pathology. Data Collection and Statistical Analysis Medical and radiologic records were reviewed for patient age, personal history (either synchronous or metachronous) or family history of breast cancer, associated symptoms, lesion size, and distance from the nipple. The radiologists who performed the ultrasound recorded the mammographic and ultrasound findings, including mammographic breast density, lesion size (maximal diameter as measured on ultrasound), and distance from the nipple in the radiologic reports. Mammographic breast density was recorded according to the BI-RADS grade on the available mammograms [11]. For statistical analysis, chi-square tests, Fisher exact tests, Student t tests, and Mann-Whitney U tests were performed to evaluate whether clinical variables (age, personal or familial history of breast cancer, and symptoms) and radiologic variables (size, distance from the nipple, multiplicity, and ultrasound features) were associated with malignancy. A p value < 0.05 indicated statistical significance. All statistical analyses were performed with the use of SPSS version 12.0 software. Results Of the 163 intraductal masses, 152 lesions were confirmed by histopathologic examination. Ultrasound-guided core needle biopsy (n = 65) or vacuum-assisted biopsy (n = 68) were performed in the intraductal masses with BI-RADS category 4. For 73 lesions, including 54 lesions that were confirmed histopathologically by core biopsy, surgery was performed promptly (mean, 1.5 ± 1.04 months; range, 0 6 months) because of the presence of malignancy (n = 5), atypical pathologic results (n = 11), complete excisional biopsy for papillary lesions (n = 55), or lesions of imaging-histologic discordance (n = 2). The histopathologic findings of the 152 lesions revealed 13 malignancies (eight ductal carcinoma in situ [DCIS] and five invasive ductal carcinoma [IDC]) and 139 benign diseases (84 papilloma, eight atypical ductal hyperplasia, 34 fibrocystic disease, five fibroadenoma, six ductectasia, one galactocele, and one adenosis). Eleven intraductal masses with BI-RADS category 3 were followed for more than 24 months and showed no significant interval change and thus were considered benign (mean, 42 ± 9 months; range, months). Among the 79 benign lesions diagnosed by core needle biopsy or vacuum-assisted biopsy, 72 lesions were also sonographically followed. The followed lesions were unchanged during the follow-up period (mean, 26 ± 15 months; range, 8 days to 56 months), and no malignancies developed. Thus, the malignancy rate of the intraductal masses was 8% (13/163; 95% CI, %). TABLE 1: Clinical Findings of Benign and Malignant Intraductal Masses Variable Benign (n = 150) Malignant (n = 13) p a Patient age (y) 50.0 ± ± Personal history of breast cancer Yes 3 3 No Family history Yes 3 0 No Symptoms Palpable 5 3 Nipple discharge 73 6 Both 4 1 None 68 3 Mammographic density (almost fatty) (scattered fibroglandular tissue) (heterogeneously dense) (extremely dense) 30 3 a Boldface type highlights statistically significant difference. The clinical findings between the benign and malignant intraductal masses are shown in Table 1. Patients with malignant intraductal masses more frequently had combined symptoms compared with those with a benign intraductal mass (p = 0.008). Among the 150 benign intraductal masses, 82 lesions (55%) had associated symptoms. The most common symptom was nipple discharge (73/150, 49%), followed by palpability (5/150, 3%). Among the 13 malignant intraductal masses, 10 patients (76.9%) had symptoms and the most common symptom was nipple discharge (5/13, 38.5%) followed by palpability (4/13, 30.8%). Palpability was more frequent in patients with malignant intraductal masses than in those with benign intraductal masses (5/13, 38.5% vs 9/150, 6%; p = 0.002). Thus, for the symptomatic lesions, the rate of malignancy was 10.9% (10/92), whereas the asymptomatic lesions had a rate of malignancy of 4.2% (3/71). Patients with malignant intraductal masses had a higher percentage of personal history of breast cancer than those with benign intraductal masses (3/13, 23% vs 3/150, 2%; p = 0.007). However, other clinical findings, such as age, familial history of breast cancer, and mammographic BI-RADS density, were not significantly different between the benign and malignant intraductal masses. Table 2 summarizes the ultrasound findings between benign and malignant intraductal masses. The mean size of the intraductal masses was 0.93 cm (SD, 1.3 cm; range, cm), and the mean distance from the nipple was 0.88 cm (SD, 1.3 cm; range, cm). The mean size of the malignant intraductal masses was significantly larger than the benign intraductal masses (1.4 ± 0.83 cm; range, cm vs 0.9 ± 0.46 cm; range, cm; p = 0.02). The mean distance from the nipple of the malignant intraductal masses was greater than that of benign intraductal masses; however, the difference was not statistically significant (1.27 ± 1.45 cm; range, 0 4 cm vs 0.85 ± 1.29 cm; range, 0 6 cm; p = 0.145). For morphologic type, a significant contrast was found between the benign and malignant intraductal masses (p < 0.001). All malignant intraductal masses filled the duct completely (type 2, nine of 13 [69%]) or extended outside the duct (type 3, four of 13 [31%]) (Fig. 2). In contrast, even though the majority of the benign intraductal masses (109 of 150 [73%]) were type 2, all of the intraductal masses incompletely filling the duct (type 1) (Fig. 3) were benign. No malignant intraductal masses showed type 1 morphology. For the morphologic types of in- 934 AJR:200, April 2013

4 Intraductal Mass on Breast Ultrasound traductal masses, the positive predictive values (PPVs) for type 1, 2, and 3 were 0% (0/37), 7.6% (9/118), and 50% (4/8), respectively. Involvement of the branch duct was more frequently noted in malignant intraductal masses than in benign intraductal masses (seven of 13, 54% vs 16 of 150, 11%; p < 0.001). Multiplicity showed no significant differences between the benign and malignant intraductal masses. The characteristics of the 13 malignant intraductal masses are listed in Table 3. A TABLE 2: Comparison of Ultrasound Findings Between Benign and Malignant Intraductal Masses Ultrasound Findings Benign (n = 150) Malignant (n = 13) p a Diameter (cm) 0.9 ± ± Distance from nipple (cm) 0.85 ± ± Multiplicity Yes 15 1 No Lesion appearance < Involvement of branch duct < Yes 16 7 No Note Type 1, mass incompletely filling duct; type 2, mass completely filling duct; and type 3, mass extending outside duct. a Boldface type highlights statistically significant difference. Discussion In this study, we retrospectively assessed the malignancy rate of 163 intraductal masses on breast ultrasound. On the basis of the histopathologic results and ultrasound followup, the rate of malignancy was 8% (13/163; 95% CI, %). A standardized lexicon for breast ultrasound was developed in 2003 by the American College of Radiology in light of the increasing use of ultrasound in clinical practice. Like its mammographic counterpart, the sonographic BI-RADS lexicon was intended to provide a unified language for sonographic reporting and research and to avoid ambiguity in the communication and teaching of sonographic interpretation [9, 11, 12]. Currently, palpable solid masses with benign features, complicated cysts, and some complex cystic and solid masses are included in the subgroup category 4A [13]. There have been several studies that investigated the PPV and negative predictive value (NPV) of the BI-RADS categories for solid masses [14, 15]. The reported PPVs for category 4 subcategory lesions showed the following ranges: 6 41%, 15 89%, and 53 90% for subcategories 4A, 4B, and 4C, respectively. The NPVs for BI- RADS category 3 were reported as 93 95%. Considering the malignancy rates (8% overall, 10.9% for symptomatic lesions, and 4.2% for asymptomatic lesions) of intraductal masses in our study and the reported PPVs and NPVs of the BI-RADS categories in previous studies, it is reasonable to characterize intraductal masses into BI-RADS subcategory 4A, supporting the current ACRIN Protocol Fig year-old woman with nonbloody nipple discharge in left breast. A, Ultrasound image shows 1.5-cm intraductal mass involving branch duct. B, Ultrasound image shows this intraductal mass extending outside duct (arrows) (type 3). This mass was diagnosed as low-grade ductal carcinoma in situ of papillary and noncomedo type as surgical-pathologic result. B In this study, ultrasound features (morphologic type and involvement of the branch duct) showed a significant difference between benign and malignant intraductal masses. Among 13 malignant intraductal masses, four masses (31%) showed type 3 and the other nine masses (69%) showed type 2, which were also commonly found in benign intraductal masses (109/150). Thus, for the management of intraductal masses, biopsy is warranted overall. However, there was no malignancy in 37 lesions showing type 1 appearance. A report by Stavros [16] also showed that most intraductal papillary lesions can be qualified for BI-RADS category 3, except for intraductal papillary lesions that are longer than 12 mm, are peripherally located, and expand the ectatic duct. Even though our results are limited because the number of type 1 intraductal masses was relatively small, according to these results, we can consider ultrasound follow-up for type 1 intraductal masses that incompletely fill the duct. In terms of lesion size, malignant intraductal masses were larger than benign intraductal masses in this study, and this finding is consistent with previous studies that showed that malignant papillary lesions are larger than benign papillary lesions [17 20]. The major difference between previous studies for imaging of intraductal mass and our study is that most previous studies focused solely on papillary lesions, ranging from benign papilloma to intraductal carcinomas, on the basis of histopathologic diagnosis. Because some papillary lesions presented as solid mass, both intra- AJR:200, April

5 Kim et al. ductal masses and solid masses were included in those studies [17 24]. On the other hand, our study focused on intraductal lesions, not papillary lesions. Because we included lesions manifesting with intraductal locations without pathologic correlation, overinclusion for masses mimicking intraductal masses on breast ultrasound, which were not intraductal masses in the final pathologic examination, and exclusion of papillary lesions without definite evidence of intraductal location could be possible. A total of 98 lesions (60%) among 163 were proven to be papillary lesion involving benign papilloma to DCIS from the results of surgical excision (n = 64), core needle biopsy (n = 11), and vacuum-assisted biopsy (n = 23), and the others showed various pathologic conditions, such as fibrocystic disease, fibroadenomas, and ductectasia. These findings are consistent with previous reports on intraductal masses [25 27]. Among papillary lesions, 19 lesions (19%) showed type 1, 74 lesions (76%) showed type 2, and 5 lesions (5%) showed type 3. Among these papillary lesions, no cancer was found in type 1 intraductal papillary lesions (0%; 0/19), and TABLE 3: Characteristics of 13 Malignant Intraductal Masses Patient No. Age (y) Personal History Symptom Mammographic Density and Finding Lesion Size (cm) Distance From Nipple (cm) Lesion Type PPVs for types 2 and 3 papillary lesions were 11% (8/74) and 50% (1/2), respectively. Our study has several limitations. First, our study was retrospective in design and the included population was small. Second, this study included intraductal masses on the basis of the radiologists sonographic reports. Misinclusion for masses mimicking intraductal masses on breast ultrasound, which were not intraductal masses in the final pathologic examination, is possible. However, our data likely reflect real clinical situations in which the physician would have to make decisions for those intraductal mass lesions regardless of their pathologic nature. Therefore, this study could also provide guidance for radiologists. Further studies with a larger patient population and prospective inclusion of intraductal masses with subjective criteria are needed to confirm our findings. Finally, the patients in Involvement of Branch Duct Surgical Pathology 1 47 Yes Nonbloody discharge 2; negative 1 SA 2 Yes Pure IDC 2 43 No No 3; negative No DCIS, low grade, papillary type with intraductal papilloma 3 62 Yes Palpable 2; irregular mass with microcalcification 3.5 SA 2 Yes DCIS, low grade 4 40 No No 3; clustered microcalcification Yes DCIS, low grade, noncomedo type 5 51 No No 3; negative Yes DCIS, low grade, papillary and noncomedo 6 39 No Nonbloody discharge 4; negative 0.5 SA 2 No DCIS, high grade, noncomedo 7 45 No Palpable 4; masses Yes IDC with DCIS, high grade, comedo, and noncomedo 8 45 No Palpable 4; masses Yes IDC with DCIS, high grade, comedo, and noncomedo 9 49 No Nonbloody discharge 3; subareolar mass 0.65 SA 2 No DCIS, high grade, noncomedo, and papillary Yes Bloody discharge 3; asymmetry Yes DCIS, low grade, cribriform, and papillary type No Nonbloody discharge and palpable 3; subareolar mass 1.8 SA 3 No IDC with low grade DCIS, cribriform, and papillary type No Palpable 2; negative No IDC with high-grade DCIS and atypical papilloma No Bloody discharge 1; asymmetry No DCIS, low grade, noncomedo type, with intraductal papilloma Note IDC = invasive ductal carcinoma, DCIS = ductal carcinoma in situ, SA = subareolar. A Fig year-old woman with nonbloody nipple discharge in right breast. A, Ultrasound image shows 0.5-cm intraductal mass that incompletely fills duct (type 1). B, Ultrasound image after more than 4 years of follow-up shows no significant interval change in size and appearance of intraductal mass. This mass was considered benign lesion. B 936 AJR:200, April 2013

6 Intraductal Mass on Breast Ultrasound this study underwent a variety of pathologic or imaging follow-up methods. Although 152 of 163 lesions (93%) were confirmed by histopathologic examination, surgical diagnosis was made for only 72 of the lesions (44%). There are possibilities of false-negative results from core needle biopsy or vacuum-assisted biopsy that result in underestimation of the malignancy rate. However, most of the biopsy-diagnosed benign intraductal masses were stable after ultrasound follow-up. In conclusion, our retrospective study showed that 8% of the intraductal masses were malignant. On the basis of this malignancy rate, intraductal masses should be characterized as BI-RADS 4A, warranting biopsy. However, intraductal masses that incompletely fill the duct might be followed-up by breast ultrasound, although further studies are required to confirm. Symptoms, personal history, lesion size, and ultrasound features can be possible predictors of malignancy. References 1. Morrogh M, Park A, Elkin EB, King TA. Lessons learned from 416 cases of nipple discharge of the breast. Am J Surg 2010; 200: Sakorafas GH. Nipple discharge: current diagnostic and therapeutic approaches. Cancer Treat Rev 2001; 27: Morrogh M, Morris EA, Liberman L, Borgen PI, King TA. The predictive value of ductography and magnetic resonance imaging in the management of nipple discharge. Ann Surg Oncol 2007; 14: Orel SG, Dougherty CS, Reynolds C, Czerniecki BJ, Siegelman ES, Schnall MD. MR imaging in patients with nipple discharge: initial experience. Radiology 2000; 216: Liu GY, Lu JS, Shen KW, et al. Fiberoptic ductoscopy combined with cytology testing in the patients of spontaneous nipple discharge. Breast Cancer Res Treat 2008; 108: Rissanen T, Reinikainen H, Apaja-Sarkkinen M. Breast sonography in localizing the cause of nipple discharge: comparison with galactography in 52 patients. J Ultrasound Med 2007; 26: Hild F, Duda VF, Albert U, Schulz KD. Ductal orientated sonography improves the diagnosis of pathological nipple discharge of the female breast compared with galactography. Eur J Cancer Prev 1998; 7(suppl 1):S57 S62 8. Berg WA, Blume JD, Cormack JB, et al. Combined screening with ultrasound and mammography vs mammography alone in women at elevated risk of breast cancer. JAMA 2008; 299: Mendelson EB, Baum JK, Berg WA, Merritt CRB, Rubin E. Breast Imaging Reporting and Data System, BI-RADS: Ultrasound. Reston, VA: American College of Radiology, Berg WA. Rationale for a trial of screening breast ultrasound: American College of Radiology Imaging Network (ACRIN) AJR 2003; 180: D Orsi CJ, Bassett LW, Berg WA, et al. Breast Imaging Reporting and Data System, BI-RADS: Mammography, 4th ed. 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: Raza S, Goldkamp AL, Chikarmane SA, Birdwell RL. US of breast masses categorized as BI-RADS 3, 4, and 5: pictorial review of factors influencing clinical management. RadioGraphics 2010; 30: Lazarus E, Mainiero MB, Schepps B, Koelliker SL, Livingston LS. BI-RADS lexicon for US and mammography: interobserver variability and positive predictive value. Radiology 2006; 239: Lee HJ, Kim EK, Kim MJ, et al. Observer variability of Breast Imaging Reporting and Data System (BI-RADS) for breast ultrasound. Eur J Radiol 2008; 65: Stavros AT. Breast ultrasound. Philadelphia, PA: Lippincott Williams & Wilkins, 2003: Youk JH, Kim EK, Kwak JY, Son EJ, Park BW, Kim SI. Benign papilloma without atypia diagnosed at US-guided 14-gauge core-needle biopsy: clinical and US features predictive of upgrade to malignancy. Radiology 2011; 258: Chang JM, Cho N, Moon WK, Park JS, Chung SY, Jang M. Does ultrasound-guided directional vacuum-assisted removal help eliminate abnormal nipple discharge in patients with benign intraductal single mass? Korean J Radiol 2009; 10: Kil WH, Cho EY, Kim JH, Nam SJ, Yang JH. Is surgical excision necessary in benign papillary lesions initially diagnosed at core biopsy? Breast 2008; 17: Chang JM, Moon WK, Cho N, et al. Risk of carcinoma after subsequent excision of benign papilloma initially diagnosed with an ultrasound (US)-guided 14-gauge core needle biopsy: a prospective observational study. Eur Radiol 2010; 20: Liberman L, Tornos C, Huzjan R, Bartella L, Morris EA, Dershaw DD. Is surgical excision warranted after benign, concordant diagnosis of papilloma at percutaneous breast biopsy? AJR 2006; 186: Mercado CL, Hamele-Bena D, Oken SM, Singer CI, Cangiarella J. Papillary lesions of the breast at percutaneous core-needle biopsy. Radiology 2006; 238: Page DL, Salhany KE, Jensen RA, Dupont WD. Subsequent breast carcinoma risk after biopsy with atypia in a breast papilloma. Cancer 1996; 78: Sydnor MK, Wilson JD, Hijaz TA, Massey HD, Shaw de Paredes ES. Underestimation of the presence of breast carcinoma in papillary lesions initially diagnosed at core-needle biopsy. Radiology 2007; 242: Lammie GA, Millis RR. Ductal adenoma of the breast: a review of fifteen cases. Hum Pathol 1989; 20: Cummings MC, da Silva L, Papadimos DJ, Lakhani SR. Fibroadenoma and intraduct papilloma: a common pathogenesis? Virchows Arch 2009; 455: Chung A, Scharre K, Wilson M. Intraductal fibroadenomatosis: an unusual variant of fibroadenoma. Breast J 2008; 14: AJR:200, April