J Korean Soc Ultrasound Med 2013;32:279-283 pissn 1015-7085 / eissn 2288-5056 Original Article http://www.ultrasound.or.kr 손윤미 1, 2 김은경 1 김민정 1 문희정 1 1 연세대학교의과대학영상의학과 2 차의과학대학교부속구미차병원영상의학과 Complex Cysts of the Breast: Analysis using Sonographic Breast Imaging Reporting and Data System Youn Mi Son, MD 1, 2, Eun-Kyung Kim, MD 1, Min Jung Kim, MD 1, Hee Jung Moon, MD 1 1 Department of Radiology, Yonsei University College of Medicine 2 Department of Radiology, CHA Gumi Medical Center, CHA University Received August 27, 2013 Revised September 9, 2013 Accepted October 14, 2013 Address for reprints : Eun-Kyung Kim, MD, Department of Radiology, Yonsei University College of Medicine, 250 Yonsei-ro, Seodaemoongu, Seoul 120-752, Korea. Tel. 82-2-2228-7400 Fax. 82-2-393-3035 E-mail: ekkim@yuhs.ac Purpose: The purpose of this study was to evaluate the usefulness of sonographic BI-RADS in differentiation between benignity and malignancy of complex breast cysts. Materials and Methods: Among 855 female patients who underwent ultrasoundguided core needle biopsy at our hospital from March 2012 through August 2012, 82 pathologically proven complex cysts detected on ultrasonography were included. The BI-RADS classification was made by radiologists prospectively and the imaging characterization of cystic lesions were classified retrospectively as cystic masses with a thick septa or thick wall, mixed solid and cystic masses, or solid masses with eccentric cystic foci. All lesions examined by ultrasonography were classified prospectively according to the BI-RADS lexicon. The pathologic findings were compared with sonographic appearance. We reviewed and compared the radiologic features between malignant lesions and benign lesions. Results: Finally, according to biopsy results, there were 71 benign and 11 (13.4%) malignant lesions. The frequency of malignancy was significantly higher in solid masses with eccentric cystic foci (p value <0.025). The size of sonographic lesions was significantly larger in malignant lesions than in benign lesions (41.82 mm±34.36 vs 18.41 mm±16.8 respectively, p=0.049). No statistically significant difference was observed in the sonographic findings according to the BI-RADS lexicon, however, statistically significant difference in final assessment categories was observed between benign and malignant lesions (p value <0.001). Conclusions: The malignancy rate was higher in cystic lesions with predominantly solid masses and classified as suspicious malignancy using the BI-RADS lexicon. Key words : Breast; BI-RADS, Breast cysts, Ultrasound (US) 서 유방의낭종은초음파검사에서흔히관찰되는질환으로단순 론 낭종은양성으로취급되나복합낭종 (complex cyst) 의경우악성률이증가하여 23-31% 까지보고되었다 [1-3]. 2003년유방초음파에대한 BI-RADS가발표된이후로유방병변에대한초음파 BI-RADS 용어에따른초음파 BI-RADS에대한 - 279-
손윤미외 양성예측도및음성예측도, 특이도, 관찰자간의일치도들이보고되었으며주로고형병변에국한되었다 [4-6]. 낭성결절의초음파소견을분류하고병리와비교하여양성과악성의특징적소견을알아보려는연구들은보고되었지만 [1-3, 7], BI-RADS 분류체계를이용한유방낭성결절의연구는많지않다 [8]. 본연구에서는유방의복합낭종을초음파 BI-RADS 로분류하고조직학적결과와비교하여복합낭종의감별에있어서 BI-RADS 용어와분류의유용성에대해평가하고자하였다. 대상과방법 2012년 3월부터 2012년 8월까지본원에서초음파유도하핵생검를시행한 885명의 1023개의병변중초음파에서복합낭종이었던 82명의 82개결절을대상으로하였다. 대상환자들은모두여자이고나이는 19-82세 ( 평균 45.3세 ) 이었다. 82명의환자중만져지는병변은 44예, 혈성유두분비 2예, 국소부위통증은 2예였고나머지 34명은증상이없었다. 초음파검사와조직검사는 8명의유방영상전문의가시행하였다. 초음파기기는 GE Logic 9 (GE Medical systems, Milwaukee, Wisconsin, U.S.A.) 또는 HDI 5000 및 IU-22 (Philips- Advanced Technology Laboratories, Bothell, Wash., U.S.A.) 으로 7-12 MHz 선형탐촉자를이용하였다. 조직검사는초음파유도하에핵생검을시행하였으며 14게이지생검침 (Stericut, TSK Laboratory, Japan) 을사용하였다. 조직검사후양성으로진단되었지만환자가원하는경우, 영상과병리소견이불일치하는경우, 고위험병변이나악성의경우, 수술을시행하였다. Berg 등 [1] 의분류에따른평가는후향적으로 2명의영상의학전문의가협의하에시행하였다. Berg 등에따르면복합낭종의경우 1) 두꺼운벽혹은격벽을가지는낭종 (cystic masses with a thick septa or thick wall), 2) 낭성과고형성분을가지는종괴 (mixed solid and cystic masses), 3) 편심적낭성부위를가진고형종괴 (solid masses with eccentric cystic foci) 로분류하였다. 초음파영상의 BI-RADS 용어와판독체계에따라분류는검사당시전향적으로 8명의영상의학전문의가평가하였으며 category 4의경우 Kim 등 [9] 의분류에따라주소견 ( 불규칙모양, 침상변연, 미세석회화 ) 과부소견 ( 난원형, 미세소엽성변연, 흐린변연, 각진변연, 비평행방향성, 주변유관과의연결, 복합에코, 후방음영 ) 중하나이상의부소견을나타내는것을 category 4a, 세개이상의부소견을나타내는병변을 4b, 부소견의유무와상관없이주소견하나이상을보이는경우를 4c로세분하였고, 주소견두개이상을보이는경우는 category 5로분류하였다 [12]. 조직학적결과에따라 Berg 등의분류법, BI-RADS분류법에따른악성율을구하였다. 통계분석은 chi-square나 t-test 를사용하였고 p 값은 0.05이하인경우유의한것으로하였다. 통계적인분석은 SPSS 소프트웨어를사용하였다 (SPSS version 20, Chicago, IL, U.S.A.). 결과최종적으로양성이 71예, 악성이 11예로악성율은 13.4% 였다. 양성과악성결절을가진환자의나이는통계적으로차이가없었고 ( 양성 : 44.52±11.3, 악성 : 51.45±15.38, p=0.177), Fig. 1. Fibroadenoma in a 31-year-old woman. The sonogram reveals about 1.4 cm sized, oval, parallel oriented, microlobulated, complex echoic mass. (BI-RADS class 4a). It shows solid mass with eccentric cystic foci. Fig. 2. Diffuse large B cell lymphoma in a 36-year-old woman. On sonography, the mass shows about 3.6 cm sized oval, parallel oriented, microlobulated, complex echoic features and posterior acoustic enhancement in left breast. (BI-RADS class 4b). It shows mixed cystic and solid mass. - 280-
악성결절이양성에비해유의하게크기가컸다 ( 양성 : 18.41 mm±16.8, 악성 41.82 mm±34.36, p=0.049). 양성중가장흔한것은섬유낭성변화였고, 그다음이유두종이었다. 악성중가장흔한것은침윤유방암이었다 (Table 1). Berg 등의분류에따른악성과양성의소견은 Table 2에정리하였다. 고형성분이증가할수록악성률이증가하는양상이며편심적낭성부위를가진고형성에가까운종괴의악성률이유의하게높았다 (p value < 0.025). BI-RADS 분류에따른양성, 악성의분류는표 3와같다. 각각의 BI-RAD 용어에따른양성과악성의소견은차이가없었으나최종분류는차이가있었다 (Figs. 1, 2). 범주 3으로분류된 10예중악성은없었다. 범주 4a, 4b, 4c, 5의악성율은각각 5.3%(3/57), 25%(1/4), 50% (3/6), 80% (1/5) 로범주가올라갈수록악성율이증가하였다 (Table 3). 고 유방촬영술이유방암진단을위한선별검사로널리사용되고있으나최근에유방초음파의발달에따라유방암진단을위해 Table 1. Pathologic Outcome for 84 Complex Cystic Lesions of the Breast Benign Fibrocystic change 21 (25.6) Papilloma 12 (14.6) Fibroadenoma 8 (9.8) Stromal fibrosis 6 (7.3) Abscess 5 (6.1) Mucocele like lesion 4 (4.9) Adenosis 3 (3.7) Fat necrosis 2 (2.4) Granulomatous mastitis 2 (2.4) Phyllodes tumor 2 (2.4) Miscellaneous* 6 (7.3) 찰 Malignant Invasive ductal carcinoma 5 (6.1) Ductal carcinoma in situ 4 (4.9) Lymphoma 2 (2.4) Total Abscess 82 (100). Note. Data are the number of lesions. Data in parentheses are percentages. *Hematoma, Fibroadipose tissue, Xanthogranulomatous mastitis, Galoctocele, Atypical ductal hyperplasia, LCIS 초음파사용이증가하였다. 이에따라병변을보고하는데있어체계화된기술방법이필요하게되어 2003년유방초음파 BI-RADS가만들어져사용되었다 [6]. 유방초음파 BI-RADS 는초음파에서보이는병변의모양에따른용어를표준화하고, 각각의용어와이에따른최종분류는유방의양성과악성병변을분류하는데유용한것으로보고되고있다 [4, 5]. 그러나이러한분류는주로고형혹에해당되는분류이고, 낭성병변의기술혹은분류에대해더세분화하지는않았고, 낭성병변의 BI-RADS 최종분류에관한연구는드물다 [6, 8]. Berg 등에따르면단순낭종은악성잠재력은없어 BI- RADS 범주 2로, 합병낭종 (complicated cyst) 은초음파소견에따라범주 2 혹은 3로, 그리고군집된미세낭종 (clustered microcyst) 의경우범주 3 으로분류하고그에따른처치를각각기술하였다 [1, 10]. 낭성병변중내부에고형성분이있는복합낭종은조직검사가필요한병변으로 BI-RADS 범주 4에해당하는것으로받아들여지고있다 [8, 10, 11]. 하지만 BI- RADS 카테고리 4의악성률은 3-95% 로그범위가넓고판독자변동성이존재하여 BI-RADS에서 4a, 4b 및 4c로세분하여 Table 3. Correlation of Pathologic Outcome and Sonographic Features based BI-RADS classification Sonographic Feature Benign (%) Malignant (%) P value Shape 24 1 Oval or round 26 3 Irregular Margin 0.151 Circumscribed 44 (62.0) 5 (45.5) Not-circumscribed 27 (38.0) 6 (54.5) Orientation 0.307 Parallel 61 (85.9) 9 (81.8) Not-parallel 10 (14.1) 2 (18.2) Lesion boundary 0.235 Thin 70 (98.6) 10 (90.9) Echogenic halo 1 (1.4) 1 (9.1) Posterior feature 0.280 Enhancement 18 (25.4) 3 (27.3) Absent 53 (74.6) 8 (72.7) Final assessment 0.001 Category 3 10 (14.1) 0 (0) Category 4a 54 (76.1) 3 (27.3) Category 4b 3 (4.2) 1 (9.1) Category 4c 3 (4.2) 3 (27.3) Category 5 1 (1.4) 4 (36.4) Table 2. Correlation of Pathologic Outcome and Sonographic Features based on Berg s classification Sonographic Feature Benign Malignant P value Cystic mass with thick wall and/or thick septations 23 (32.4) 1 (9.1)0 Mixed cystic and solid 23 (32.4) 3 (27.3) Solid mass with eccentric cystic foci 25 (35.2) 7 (63.6) Total 71 11 0.025-281-
손윤미외 정의하고있으나이에대한정확한평가는아직까지보고된바없다. 본연구에서각각의 BI-RAD 용어에따른악성과양성의소견은차이가없어 BI-RADS 용어가복합낭종을설명하는데는한계가있음을알수있다. 그러나 BI-RADS 최종분류는악성을잘예측할수있어서범주 3은 0% (0/10), category 4a는 3/57 (53%), category 4b는 1/4 (25%), category 4c는 3/6 (50%), category 5는 4/5 (80%) 로여전히유용함을알수있다. 이에비해 Berg 등의분류법에따른분류법은악성을예측할수있으며편심적낭성부위를가진고형종괴의경우 21.9% (7/32) 가악성인반면, 두꺼운벽혹은격벽을가진낭종의경우 4.2% (1/24) 으로고형성분이증가할수록악성률이증가하는양상이며편심적낭성부위를가진고형성에가까운종괴의악성률이유의하게높았다 (p value < 0.025). Kim 등 [8] 은악성과양성의복합성에코종괴의 BI-RADS 용어를고려하여비교시불규칙모양, 비평행방향성, 불규칙경계, 병변주위고에코 halo, 증가된혈류, 그리고큰크기에서통계적으로유의하여우리의결과와는차이가있었다. 그이유는본연구가전향적으로 8명의영상의학전문의가초음파영상의 BI-RADS 용어와판독체계에따라복합성에코종괴를평가하였는데다수에판독자로인해일관성있는병변의평가가힘들었기때문이라고생각되나본연구는전향적으로소견을분석한데의의가있다고생각된다. 이에비해최종분류는양성과악성을예측할수있어복합낭종의분류에도우수함을알수있었다. Hsu 등 [11] 은 BI-RADS용어를고려하여병변의크기, 모양, 경계, 액와임파절이상에따라악성률을구하였을때복합낭종중 2 cm이상크기의경계가잘지어지지않은병변의악성률이통계적으로유의하다고하였다. 본연구에서도악성결절이양성에비해유의하게크기가컸고 ( 양성 : 18.41 mm± 16.8, 악성 41.82 mm±34.36, p=0.049) Kim 등 [8] 에서도큰크기에서통계적으로유의함을보고하고있어복합낭종의크기는악성도예측에도움을줄수있다. 본연구에서양성이 71예 (84.5%), 악성이 11예로악성율은 13.4% 로 Berg 등 [1] 의연구에서악성율 12%(18/150) 과비슷하였다. 그러나복합낭종에대한 Pongrattanaman 등 [2] 의연구에서악성률은 25.2%(26/103), Chang 등 [7] 에서 22.8%(40/175) 로더높았다. 이러한차이는복합낭종중고형성분이많은병변을얼마나포함하고있는지에따라악성율이다르기때문으로생각되며 Berg 등 [1] 의연구에는단순낭종도포함하였으며본연구는 90.2% 의병변이범주 3과 4a에해당되었기때문일것으로생각된다. 본연구의제한점으로첫째, 연구대상수가적고조직검사로확인된복합낭종에대한연구이므로대부분이범주 4에포함되었고범주 2와범주 3의예가작았다. 둘째, 판독자간의차이가본연구에서는고려되지않아판독자간에분류설정에따른 변이가있을수있다는점이다. 그렇지만본원에서시행한여러연구자들의범주 4( 세부범주포함 ), 5의악성율은권장범주안에있으므로문제가되지는않을것으로생각된다 [12, 13]. 결론적으로, 고형성에가까운복합성고형낭종에서악성률이높으며초음파 BI-RADS 용어를사용한최종분류법은병변의악성도를예측하는데유용하다. 향후낭성유방종괴에대한 BI-RADS 초음파판독체계에대한다양한연구가필요하다. 요 목적 : 유방복합낭종의양성과악성의감별진단에있어서초음파 Breast Imaging Reporting and Data System(BI-RADS) 을이용한평가의유용성을알아보고자하였다. 대상과방법 : 2012년 3월부터 2012년 8월까지본원에서초음파유도하핵생검를시행한 885명의환자중초음파소견이복합낭종이었고병리학적으로확진된 82예를대상으로하였다. 병변은전향적으로 BI-RADS에의거하여용어와최종분류가기록되었고낭성병변의영상소견은두꺼운벽혹은격벽을가지는낭종, 낭성과고형성분을가지는종괴, 편심적낭성부위를가진고형성에가까운종괴등으로후향적으로구분하였다. 병리소견에따른초음파소견을검토하였고양성과악성으로진단된결절간의영상소견을비교분석하였다. 결과 : 최종적으로양성이 71예, 악성이 11예로악성율은 13.4% 였다. 편심적낭성부위를가진고형성에가까운종괴의악성률이유의하게높았다 (p value <0.025). 악성결절이양성에비해유의하게크기가컸다 ( 양성 : 18.41 mm±16.8, 악성 : 41.82 mm±34.36, p=0.049). BI- RADS용어에따른초음파소견은양성과악성의소견은통계적으로차이가없었지만 BI-RADS 최종분류는통계학적으로차이가있었다 (p value < 0.001). 결론 : 유방복합낭종은고형성에가까운복합성고형낭종과악성의위험도가높은 BI-RADS 최종분류에서악성률이높았다. 약 References 1. Berg WA, Campassi CI, Ioffe OB. Cystic lesions of the breast: sonographic-pathologic correlation. Radiology 2003;227:183-191 2. Pongrattanaman S, Prueksadee J. Sonographic-pathologic correlation of complex cystic breast lesions. Asian Pac J Trop Dis 2013;3:51-55 - 282-
3. Doshi DJ, March DE, Crisi GM, Coughlin BF. Complex cystic breast masses: diagnostic approach and imagingpathologic correlation. Radiographics 2007;27:S53-S64 4. Hong AS, Rosen EL, Soo MS, Baker JA. BI-RADS for sonography: positive and negative predictive values of sonographic features. AJR Am J Roentgneol 2005;184: 1260-1265 5. 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;239:385-391 6. American College of Radiology. Breast Imaging Reporting and Data System, Breast Imaging Atlas. 4th ed. Reston, VA: American College of Radiology 2003 7. Chang YW, Kwon KH, Goo DE, Choi DL, Lee HK, Yang SB. Sonographic differentiation of benign and malignant cystic lesions of the breast. J Ultrasound Med 2007;26:47-53 8. Kim HY, Chang YW. Analysis of breast imaging reporting and data system category4 complex cystic masses of the breast: do all the complex cystic breast masses merit a biopsy? J Korean Soc Radiol 2013;69:71-77 9. Kim E, Ko KH, Oh KK, Kwak JY, You JK, Kim MJ, et al. Clinical application of the BI-RADS final assessment to breast sonography in conjunction with mammography. A- JR Am J Roentgneol 2008;190:1209-1215 10. Berg WA, Sechtin AG, Marques H, Zhang Z. Cystic breast masses and the ACRIN 6666 experience. Radiol Clinics N Amer 2010;48:931-987 11. Hsu HH, Yu JC, Lee HS, Lin WC, Chang WC, Tung HJ, et al. Complex cystic lesions of the breast on ultrasonography: feature analysis and BI-RADS assessment. Eur J Radiol 2011;79:73-79 12. Yoon JH, Kim MJ, Moon HJ, Kwak JY, Kim E. Subcategorization of ultrasonographic BI-RADS category 4: positive predictive value and clinical factors affecting it. Ultrasound Med Biol 2011;37:693-699 13. Park HY, Ko SY, Park WJ, Seo JY, Son YM, Youn I, et al. Positive predictive value of BI-RADS US category 4 and 5 lesions. J Korean Soc Breast Screening 2012;9:138-141 - 283-