经乳腺专用MRI诊断为BI-RADS 4类病灶良恶性比例及“误诊”原因分析

Proportion of benign and malignancy in BI-RADS category 4 breast lesions diagnosed by dedicated breast MRI and analysis of misdiagnosis cases

目的:探讨经乳腺专用磁共振成像(DBMRI)诊断为BI-RADS 4类病灶中的良恶性比例,并分析其误诊原因。方法:收集我院经DBMRI诊断为BI-RADS 4类的781例患者的资料,参照美国放射学BI-RADS分类标准,结合我院临床经验,认为形态学及血流动力学两项指标,一项表现为非良性,则将病灶归为BI-RADS 4类。以病理结果作为金标准,分析诊断为BI-RADS 4类病灶中的良恶性比例及"误诊"原因。结果:DBMRI诊断的781例患者的781个BI-RADS 4类病灶(BI-RADS 4A类301个,BI-RADS 4B类226个,BI-RADS 4C类254个),病理结果均提示异常,检测灵敏度达100%。其中,病理证实为良性的病灶共457个,恶性病灶共324个。DBMRI考虑为良性(BI-RADS 4A类),而病理结果为恶性的病灶(27个)主要为浸润性导管癌(37.04%)和导管原位癌(33.33%);DBMRI考虑为恶性(BI-RADS 4B、4C类),而病理结果为良性的病灶(457个)主要为纤维腺瘤(32.24%)、乳腺腺病(27.87%)及导管内乳头状瘤(25.68%)。结论:DBMRI因其较高的空间分辨率及高对比度,能为临床BI-RADS 4类亚分类病灶的不同处理提供依据。然而DBMRI仍存在其局限性,对于非肿块样病灶性质的判定仍较困难。建议非肿块样强化病灶,一定要结合乳腺X线及超声检查,提高其诊断准确率。

Objective： To analyze the proportion of benign and malignancy in BI-RADS （breast imaging reporting and data system）category 4 breast lesions diagnosed by dedicated breast magnetic resonance imaging （DBMRI）and to investigate the reason for misdiagnosis. Methods： According to the American College of Radiology （ACR） breast imaging reporting and data system （BI-RADS） and our experience, morphological characteristics and blood flow kinetic features of breast lesions from 781 patients were analyzed retrospectively. If one of the features was non-benign, the lesion would be assessed as BI-RADS 4. Using pathological diagnosis as golden standard, the proportion of benign and malignancy in BI- RADS 4 breast lesions and the reason of misdiagnosis were analyzed. Results： All 781 cases of BI-RADS 4 breast lesions （including 301 cases of category 4A lesions,226 cases of category 4B lesions, 254 cases of category 4C lesions） had abnormal pathological results, with a sensitivity of 100%. Of them, 457 were confirmed as benign and 324 were confirmed as malignancy pathologically. The BI-RADS-diagnosed as benign while pathologically diagnosed as malignant lesions included infiltrative ductal carcinoma （IDC） （37.04%）and ductal carcinoma in situ （DCIS） （33.33%）. The BI-RADS diagnosed as malignant while pathologically demonstrated as benign lesions included fibroadenoma （32.24%）, adenosis （27.87%）and intraductal papilloma （25.68%）. Conclusions： Because of its high-spatial-resolution and high-contrast resolution, DBMRI contributes a lot in the selection of management for BI-RADS 4 subcategory lesions. But it is very difficult for DBMRI to identify different types of non-mass lesions. For non-mass lesions, combined use of BI-RADS, mammography and uhrasonography may improve the diagnosis.