美国放射学会（ACR）乳腺影像报告与数据系统（BI-RADS）
这篇概述是关于乳腺X射线摄影的BI-RADS分类的，总体来说，也适用于乳腺超声和磁共振。近期有中国医生问了我一些BI-RADS分类的相关问题，这篇概述也许可以帮到大家。
关于BI-RADS的发展简史
在美国，20世纪80年代开始乳腺X射线摄影筛查，虽然FDA会监督筛查的质量，但从来没有一个由政府指导、中央协调的整体组织来监督筛查。
在20世纪80年代早中期，有大约20%的女性进行了首次乳腺X射线摄影筛查，这比例一直在稳步上升，到20世纪90年代末，大约有70%的40岁及以上的女性进行了至少一次的乳腺X射线摄影筛查 ¹。
随着越来越多的女性参加筛查，乳腺X射线摄影报告质量差的问题就变得非常明显。放射科诊断医生“意识流”式的描述报告，使得临床医生经常不确定报告中描述了什么，以及接下来该为患者做什么样的临床处理。有一份白皮书更是直接对于低质量的乳腺X射线摄影报告提出了强烈批评 ²。
20世纪80年代末，美国放射学会（ACR）组织了一批乳腺影像专家，并邀请了顶尖的乳腺外科专家等人一起设法解决报告问题。Carl D'Orsi医生和我是联合主席。之前我在麻省总院（MGH）乳腺影像部 ³开发了乳腺影像系统性阅片报告方法，D'Orsi医生曾经和一家名为Bolt, Beranek, and Newman公司合作开发了乳腺X射线摄影检查报告方面的词汇，Burnside等人对BI-RADS发展史做了概述 ⁴。
BI-RADS的目的是制定一个明确的专业术语词典，用来描述乳腺X射线摄影上的发现。这本词典主要基于D'Orsi医生的工作。BI-RADS的分类和架构是基于我的报告系统。BI-RADS阐明了一份报告应如何构成，并定义了每种类型的发现、重要性以及应如何分类：
BI-RADS 0：
尽管有时在筛查中可以确定病灶为恶性，但最好将有任何可疑的发现归类到BI-RADS 0，说明有不确定的发现，需要将患者召回进行更明确的诊断评估。
BI-RADS 1：
乳腺X射线摄影检查中无任何异常发现。
BI-RADS 2：
放射科医生描述一个明显良性的发现时使用，这样其他人在看这个病例图像时就不会感到困惑。
BI-RADS 3：
用于几乎可以确定的良性病变（恶性的可能性低于2%）。放射科医生通常期望进行6个月的短期随访（基于为所有女性推荐的年度间隔），BI-RADS指南建议对该发现进行6个月、12个月的跟踪随访，然后每年一次，总共3年。根据在麻省总院的经验，我们对这部分女性进行6个月一次，持续2年的随访。
BI-RADS 4：
被分为4a，4b，4c，三个选项均建议通过活检进行组织学诊断，但是有一部分放射科医生希望使用BI-RADS 4对这些发现进行分类。
BI-RADS 5：
最初是为了农村地区创建的，那里的患者很难被召回进行额外的评估，病变的外观符合典型的乳腺癌特征，因此立即进行活检是合理的。
BI-RADS 6：
已经被病理证实存在乳腺癌，但是需要通过乳腺X射线摄影检查评估乳房的其它区域。
BI-RADS指南的主要目的是减少解读乳腺X射线摄影图像时的困惑，确定每一份报告都有清晰简洁的结论和下一步方案建议，根据需要，明确关注的程度和下一次检查的建议。
其次，BI-RADS还有一个很重要的目的是让放射科医生收集数据来跟踪他们自己的报告和在筛查项目中的报告，以确定每个患者的实际情况。Edward Sickles医生 ⁵ 将“审计”纳入指南并予以规范，这样我们就能跟踪患者情况，从每次的检查结果中学习，明确了解自己的工作结果，并不断提升乳腺X射线摄影图像的解读能力。
报告结构
每个报告都应该分类为“筛查”或“诊断”。
筛查针对的是没有乳腺癌征象女性。这个人群接受常规定期的乳腺X射线摄影检查，目的是在乳腺癌产生临床征象之前将其发现。从40岁开始进行每年一次的筛查挽救了很多生命 ⁶。
进行“诊断”评估的女性往往在临床检查中有明显的症状，如乳房局灶性疼痛，肿块，增厚，乳头改变和溢液，或者是在筛查中发现有问题需要召回接受评估。
报告要点
报告的开头要写明检查原因，我通常会写明患者的年龄和是否绝经。例如：这是一位年龄57岁，绝经后的女性进行乳腺X射线摄影筛查。
乳腺腺体密度会影响隐匿性病灶的发现（一些腺体致密的老年女性，风险会增高）这一点很重要，应该写明腺体分型。
我通过写明“乳腺腺体密度分布不均”，表示敏感性降低。“这在一定程度上降低了乳腺X射线摄影检查的敏感性”。
如果无任何异常发现，那么我们在报告中就写明没有明显恶性发现的结论：“没有发现明显肿块或成簇微钙化”。
由于有些癌症可以被致密的腺体组织所掩盖，因此即使乳腺X射线摄影检查结果看起来是阴性的，我们也应该告诉临床医生，“任何进一步评估的决定都应以临床评估为基础”。
如果患者在临床上有明显的发现并进行了临床诊断评估，但是乳腺X射线摄影检查没有明确发现，那么为了不影响临床医生采取进一步的措施，最后写明这句话就很重要。
结论：乳腺X射线摄影检查没有发现恶性肿瘤的征象。我之所以写明是乳腺X射线摄影检查，是因为如果临床上有明显发现，医生对这个患者应该有进一步措施。
报告的最后写明是：BI-RADS 1
其他最终评估
如果在乳腺X射线摄影检查里发现一个可能会令人困惑的良性发现，那么我会描述：双侧乳房可见多发的线状钙化，属于典型的良性“分泌性”钙质沉积，乳腺X射线摄影检查没有发现恶性肿瘤征象。
结论：良性钙化 乳腺X射线摄影检查没有发现恶性征象。
BI-RADS 2。
如果有可疑发现，那么报告可以写：左乳外上发现大小约1cm毛刺状肿块，腺体纠集，未发现其他肿块和簇状微钙化，建议补充乳腺X射线摄影检查和超声检查进一步评估。
结论：左乳乳腺癌可能，建议进一步检查。
BI-RADS 0。
当患者召回进行诊断评估时，可能需要写明：左侧乳房外上后部1点钟位置发现一大小约1cm，边缘不规则，呈毛刺状肿块。超声显示病变为实性，后方有声影。这应该高度怀疑为恶性。可以在超声引导下进行核芯针活检。
结论：左侧乳房肿块，高度怀疑恶性，建议活检。
BI-RADS 4。
多个病灶发现
有人曾问，对于多个病灶的情况，如何在BI-RADS范畴内报告。在报告的正文中，需要对每个发现进行描述和分类。
例如，右侧乳房发现多个病灶：
1. 右侧乳房中央发现一直径8mm，边界不清的肿块，建议进一步影像学检查。
2. 第二个发现，右侧乳房中部3点钟位置发现直径12mm的致密影，边界不清，建议对病灶进行额外（其它）的影像学检查。
3. 双侧乳房多发中心透亮的良性钙化灶。
结论：右侧乳房发现两个可疑病变，建议进一步影像学检查。
BI-RADS主要目标是为影像提供清晰简洁的准确分析。报告正文中的详细描述应使用清晰、简洁的描述性术语，并主要使用BI-RADS词典里的专业术语，这样就不会产生混淆。
如果需要进一步评估，结论应该明确无误， 并给出具体建议。
在美国，我们把BI-RADS分类放在报告的最后。如果有多个发现，我们选择最重要的那个病灶的BI-RADS结果记录于报告。这并不表示只有单个病灶发现，只是说明是否存在需要额外关注的征象。
中英翻译
BI-RADS
The American College of Radiology Breast Imaging Reporting and Data System
This overview relates to the BI-RADS system for Mammography, but it applies, in general, to BI-RADS for Ultrasound and Magnetic Resonance Imaging of the breast. I have, recently, been asked specific questions about BI-RADS so I thought this overview might be helpful.
A BRIEF HISTORY OF BI-RADS
Mammography screening began, haphazardly, in the U.S. in the1980’s. There has never been a government guided, centrally coordinated, overall organization to oversee screening in the U.S. although the Food and Drug Administration (FDA) does monitor quality issues.
In the early to mid-1980’s approximately 20% of women had their first screening mammogram and this increased steadily until it plateaued in the late 1990’s when approximately 70% of women ages 40 and over had had at least one mammogram ¹.
As more and more women participated in screening, the poor quality of the mammography reporting became evident. Radiologists were dictating reports that were “stream of consciousness” and often left our clinical colleagues uncertain as to what the study showed and what should be done for the patient. This became apparent in a white paper that was highly critical of the low quality of mammography reporting ².
In the late 1980’s the American College of Radiology assembled a group of radiologists who were expert in Breast Imaging and also invited a leading breast surgeon and others to address the reporting issues. Dr. Carl D’Orsi and I were co-chairmen. I had developed a reporting system for my Breast Imaging Division at the Massachusetts General Hospital ³ and Dr. D’Orsi had worked with a company called Bolt, Beranek, and Newman in developing a vocabulary for mammography reporting. The evolution of BI-RADS is explained in an overview by Burnside et al ⁴.
The goal of BI-RADS was to develop a clearly defined dictionary of terms for describing findings on a mammogram. This is known as the “Lexicon” and was based on the work done by Dr. D’Orsi. The categories and organization of BI-RADS were based on my reporting system. BI-RADS explains how a report should be organized, and defines each type of finding, its significance, and how it should be categorized ranging from:
BI-RADS 0:
Although there are times when a lesion can be definitively determined to be malignant on a screening examination, it is best to have the patient who has any findings that raise suspicion to be categorized as “BI-RADS 0” which indicates there are findings that are uncertain and the patient should return for more definitive diagnostic evaluation.
BI-RADS 1:
is used when the mammogram is negative with nothing to report.
BI-RADS 2:
is used when there is a clearly benign finding that the radiologist wants to describe so that anyone else looking at the mammogram will not be confused.
BI-RADS 3:
is used for findings on the mammogram that have a low probability of being cancer (less than 2% likelihood), but that the radiologist wants to follow at a shorter interval based on the annual interval that is recommended for all women. The shorter interval is six months and BI-RADS suggests that the finding be followed up at 6 months, 12 months and then annually for a total of 3 years from the study that first identified the finding. Our experience at the Massachusetts General Hospital led us to follow-up these women every 6 months for a total of 2 years.
BI-RADS 4 ：
has been divided into 4a, 4b, and 4c. Biopsy (tissue diagnosis) is recommended for all 3 subcategories, but some radiologists wanted to grade these findings within BI-RADS 4.
BI-RADS 5 ：
was originally created for rural locations where getting patients to return foe additional evaluation was difficult. The appearance of the lesion is so characteristic of a breast cancer that it was not unreasonable to do an immediate biopsy.
BI-RADS 6 ：
is used when there is a known breast cancer (pathologically proven) that is still evident on the mammogram, but the mammogram is being done to evaluate other areas of the breast.
The primary reason for BI-RADS was to reduce any confusion about the interpretation of the mammogram and to be certain that every report had a clear, action defined, conclusion and, if needed, clear levels of concern and clear guidance for the next test if needed.
A secondary, but very important goal for BI-RADS was to allow radiologists to collect data to track their own reports and those in their screening program to determine what actually happened to each patient. This was ultimately codified by the “Audit” that was described by Dr. Edward Sickles ⁵ which allowed us to track our patients and learn from the outcomes of each examination and improve future mammographic analyses. By tracking our results, we were able to understand the results of our efforts and to improve our mammographic interpretations.
ORGANIZATION
Each study should be categorized as “Screening” or “Diagnostic”.
“Screening” involves women who have no signs or symptoms of breast cancer. They have no clinically evident findings and are being routinely, and periodically having their mammograms with the goal of discovering breast cancer earlier than when it becomes clinically evident. The most lives are saved by annual screening starting at the age of 40 ⁶
Women having “Diagnostic” evaluation have either something evident on clinical examination or breast symptoms such as focal pain, a “lump”, thickening, nipple changes or discharge, or are women who have had something found on their screening mammogram and are returning to have the finding evaluated.
OUTLINE OF THE REPORT
The reason for the study should be provided at the start of the report. I always include the menopausal status of the individual and like to include the age of the patient.
Eg: “This is a 57 year old, post-menopausal female having a screening study.”
Given the importance of breast density as potentially obscuring a lesion (with some slight increase in risk for older women with the densest pattern) the tissue pattern should be reported.
“The breast tissues are heterogeneously dense” and then I indicate that sensitivity is reduced by stating “this, somewhat, lowers the sensitivity of mammography”.
If there is nothing to report, then our report provides a simple summary of the absence of the major malignant criteria:“No mases or clustered microcalcifications are evident.”
Since there are cancers that can be hidden by dense breast tissue, the clinician should always be told that, even if the mammogram is negative, “Any decision for further evaluation should be based on the clinical evaluation”.
This last statement is also important if the patient has had a diagnostic evaluation for a clinically evident finding that is not defined by the mammogram so that the clinician is not dissuaded from further evaluation of the finding.
Then:“CONCLUSION: There is no mammographic evidence of malignancy.”I state “mammographic” since if there are clinically evident findings the referring physician should act on those.
At the bottom of the report is:“BI-RADS 1”
OTHER FINAL ASSESSMENTS
If there is a benign finding on the mammogram that might confuse someone, then I will describe it. For example:“There are numerous linear calcifications that project over both breasts that are typical of benign “secretory” deposits. There is no mammographic evidence of malignancy.”
“CONCLUSION: Benign calcifications with no mammographic evidence of malignancy.”
“BI-RADS 2”
If there is a suspicious finding, then the report might read:“There is a 1 cm, spiculated mass distorting the architecture in the upper outer left breast. No other masses or clustered microcalcifications are identified. Additional evaluation with mammography and ultrasound is recommended.”
“CONCLUSION: Suspicious lesion left breast, additional evaluation is recommended.
BI-RADS 0
When the patient returns for “Diagnostic evaluation”, then the report might read:“There is a 1 cm. in diameter irregular shaped mass with spiculated margins in the upper outer left breast posterior at 1:00. Ultrasound was performed and showed that the lesion is solid with posterior acoustic shadowing. This should be viewed with a high degree of suspicion and likely represents a malignant lesion. Ultrasound guided core needle biopsy could be performed.
CONCLUSION:Very suspicious mass left breast, biopsy is recommended.
BI-RADS 4
MULTIPLE FINDINGS
I have been asked how to manage multiple findings within BI-RADS. In the body of the report, each finding should be described and categorized.
There are multiple findings in the right breast:
1. There is an ill-defined, 8 mm. in diameter mass in the center of the right breast. Additional imaging is recommended.
2. There is a second, ill-defined 12 mm in diameter density in the 3:00 mid portion of the right breast. Additional imaging of this lesion is recommended.
3. There are numerous lucent centered, benign calcifications scattered throughout both breasts.
CONCLUSION: There are two suspicious lesions in right breast. Additional imaging is recommended.
BI-RADS 4
Again, the main goal of BI-RADS is to provide a clear and accurate analysis of the imaging study. A detailed description in the body of the report should be provided using descriptive terms that are clear and concise. The BI-RADS Lexicon of terms should be used for most descriptions so that there is no confusion.
The CONCLUSION should be concise and accurate with a clear recommendation if further evaluation is needed.
In the U.S. we put the BI-RADS code at the bottom of the report. If there are multiple findings we chose the one that defines the most consequential finding in the analysis. This doesn’t mean that there is only a single finding but indicates whether or not there is an indication that requires additional attention.
REFERENCES
1. Breen N, A Cronin K, Meissner HI, Taplin SH, Tangka FK, Tiro JA, McNeel TS. Reported drop in mammography : is this cause for concern? Cancer. 2007 Jun 15;109(12):2405-9.
2. Scott, W. Establishing mammographic criteria for recommending surgical biopsy. Chicago, Ill: American Medical Association; 1989.
3. Kopans DB. Breast Imaging. J.B.Lippincott Co. Philadelphia, Pa. 1989. p. 351
4. Burnside ES, Sickles EA, Bassett LW, Rubin DL, Lee H, Ikeda M, Mendelson EB, Wilcox P, Butler F, D’Orsi CJ. The ACR BI-RADS® Experience: Learning From History. J Am Coll Radiol Volume 6, Issue 12, December 2009, Pages 851-860.
5. Sickles EA, Ominsky SH, Sollitto RA, Galvin HB, Monticciolo DL. Medical audit of a rapid-throughput mammography screening practice: methodology and results of 27,114 examinations. Radiology. 1990 May;175(2):323-7.
6. Mandelblatt JS, Cronin KA, Bailey S, Berry DA, de Koning HJ, Draisma G, Huang H, Lee SJ, Munsell M, Plevritis SK, Ravdin P, Schechter CB, Sigal B, Stoto MA, Stout NK, van Ravesteyn NT, Venier J, Zelen M, Feuer EJ; Breast Cancer Working Group of the Cancer Intervention and Surveillance Modeling Network. Effects of mammography screening under different screening schedules: model estimates of potential benefits and harms. Ann Intern Med. 2009 Nov 17;151(10):738-47.
