I. Introduction to the Breast Imaging Reporting Data System
In 1992, the American College of Radiology published a guiding document: the Breast Imaging Reporting Data System (BIRDS), which has been revised three times since then, and by 2003 not only guides mammography diagnosis (4th edition), but also adds ultrasound and MRI diagnosis. The system standardizes the diagnostic reporting of all normal and abnormal imaging conditions of the breast as a whole organ, using a unified terminology, standard diagnostic categories and examination procedures, so that the radiologist’s diagnosis can be followed, and at the same time, it also strengthens the coordination and tacit understanding between the radiology department and other related clinical departments, so that the clinical treating physicians know what to do next once they read the radiologist’s report. Note that the word Category should not be translated as “level”, but as “class”, and the BI-RADS evaluation is divided into indeterminate and final categories. The indeterminate category requires further imaging, such as additional x-ray projections of the body, comparison of old films, ultrasound or MRI (note: infrared thermography or CT is not recommended). Once the additional imaging studies are performed, the final category evaluation should be completed and the contents of these imaging studies should be integrated to arrive at a comprehensive diagnostic evaluation classification.
Mammography quality specifications require a single analysis report for mammography examinations. The hospital or clinician wishes to provide the BI-RADS classification for each breast separately, which is mentioned in the Diagnostic Findings section or the Diagnostic Description section of the report. And provide a single overall diagnostic report, expressing the BI-RADS classification at the end of the entire report. Of course, the overall final report should be based on the presence of the most worrisome things. For example, if one breast is noted as a probable benign finding and the contralateral breast is suspected of having a malignant lesion, the overall diagnostic report should be recorded as “BI-RADS Category 4 (suspicious for malignancy)”. Similarly, if one breast requires immediate additional evaluation (e.g., the patient cannot wait for ultrasound) and the contralateral breast may have a benign finding, the overall classification should be “BI-RADS category 0, indeterminate”.
The clinical finding of a lesion with a negative image is a question that many hospitals are puzzled by. The final evaluation of the diagnostic report should be based on the imaging findings. When the interpretation of the imaging findings is influenced by the clinical findings, the final evaluation should consider both aspects in a holistic manner. Clinical findings should be carefully described in the report.
II. BI-RADS classification (not grading)
BI-RADS Category 0: Need to combine with other examinations.
BI-RADS Category 1: Negative.
BI-RADS Category 2: benign.
BI-RADS category 3: benign possible, requiring short-term follow-up.
BI-RADS category 4: suspicious for malignancy, biopsy recommended.
4A: Low suspicion.
4B: Moderately suspicious.
4C: High but not certain.
BI-RADS Category 5: Highly malignant.
BI-RADS Category 6: Malignancy has been pathologically confirmed.
Category 1, 0
Category 0 is used after routine mammography. Further diagnostic evaluation (e.g., additional projection positions or ultrasonography) or recall of the old film for analysis is required for a limited time. Control of old films may reduce the need for patient recall. However, controls are not always necessary. In the absence of any findings, previous photographs are helpful in only about 3.2% (35/1093) of cases. Only mammograms that identified some alteration requiring comparison of older films were classified as category 0. This often includes limited asymmetric changes that may represent normal variation or radiographs showing well-margined masses that may have been present on previous images. If, however, there are no older films to compare, then further workup (such as additional x-rays and/or performing ultrasound) is warranted. In our country, some women with less fatty breasts, abundant parenchyma and lack of natural contrast in the breast tissue also require further examination using other imaging methods (e.g. ultrasound, MRI) and can also be evaluated as category 0.
2. Category 1
Mammography shows clear breast structures without lesions showing. Note that the so-called cystic hyperplasia, lobular hyperplasia, and adenopathy (collectively referred to as fibrocystic changes or structural dysplasia) often used in our country are all classified in this category according to the BI-RADS description. A clinically detected mass with limited asymmetric changes, despite a final diagnosis of sclerosing adenopathy, cannot be classified in this category and may be classified in category 3 or 4A. Intramammary lymph nodes and anterior axillary lymph nodes showing a hypodense lymph node gate (lateral view) or a central hypodensity (axial view of the lymph node gate) are considered normal lymph nodes and belong to category 1.
3.Class 2
Definite benign breast masses (e.g. fibroadenoma, fibrolipid adenoma, lipoma, simple cyst, cumulative breast cyst, cumulative oil cyst) and definite benign calcifications (e.g. circumferential calcifications, short strips of calcifications with clear borders, coarse speckled calcifications, sparse, single sized dotted calcifications, crescent-shaped sedimentary calcifications, etc.) belong to this category. However, clear margins of the mass are not a necessary condition to exclude malignant lesions, and in women older than 35 years of age, attention should be paid to search and recall old films for comparison or follow up to observe their changes, and therefore, may be evaluated as category 0 or 3, respectively.
4. Category 3
Category 3 (probably benign) is reserved, and its findings are almost certainly benign. It must be emphasized that this category is not an indeterminate type, but for mammography it has a less than 2% chance of being malignant (i.e. almost always benign). The presentation is gradually recognized, based on images with or without previous screening results. Evaluation with additional projections and/or ultrasound of the breast in other orientations requires a category 3 (probably benign). This type of lesion includes well-defined masses that cannot be visualized on conventional radiographs (unless they are cysts, intramammary lymph nodes, or other benign lesions), limited asymmetric changes with thin sections on spot compression films, and fine clusters of calcifications. Short-term follow-up is performed using unilateral radiographs 6 months after routine mammographic findings. If there is no change in the lesion, a further bilateral breast follow-up at 6 months is recommended (i.e., 12 months after the initial finding). If no other suspicion is observed at the second bilateral breast follow-up, a category 3 is reported and a typical post-12-month bilateral breast follow-up (i.e., 24-month follow-up after the initial examination) is recommended. If no change is still observed at the next follow-up (24th month follow-up), the final assessment may be a category 2 (benign) or of course a category 3 (probably benign) in conjunction with clinical caution. After 2-3 years of stabilization according to the literature, the final diagnosis may change to category 2 (benign), but still requires follow-up and, if necessary, magnification photography.
Perhaps a less experienced physician would insist on a smaller limited asymmetric change, thus defining it as a category 3. The more experienced physician with 6, 12, or 24 month follow-up may determine that this change is a normal variant and for that reason define it as a category 1 (negative).
Category 3 may be immediately biopsied for reasons such as clinician or patient fear of tumor and unwillingness to follow up, and in these cases the final diagnostic assessment classification should be based on the risk of malignancy rather than on the management offered. Lesions that are judged by ultrasound as possibly benign include complex cysts that cannot be palpated. The malignancy rate of non-palpable ovoid hypoechoic nodules that cannot be distinguished from complex cysts has been reported to be less than 2%. Clusters of microcysts without discrete solid components may also be rated as category 3.
Proper classification of category 3 requires review of the physician’s practice. The malignancy rate in such cases should be less than 2%. For ultrasound, the malignancy rate should also be less than 2%, but this has not been confirmed in a wide range of articles. For MRI, cases classified in this category have only been followed up for a short period of time and their malignancy rate needs further study.
5. Category 4
Category 4 is used to indicate interventional radiology procedures that require biopsy from complex cyst aspiration to polymorphic calcifications. Many units subcategorize category 4 to account for differences in interventional management and risk of malignancy. This uses subject work characteristic curve analysis, subject to greater clinical scrutiny, to help clinicians and radiologists. 4 categories are divided into three subcategories to facilitate help in achieving these purposes.
Category 4A.
Category 4A is used to represent lesions that require interventional management but are less malignant. The pathology is not expected to be malignant and routine follow-up for 6 months after benign biopsy or cytology findings is appropriate. This category includes some solid masses that are palpable with partially clear margins, such as ultrasound suggestive of fibroadenoma, palpable complex cysts, or suspicious abscesses.
Category 4B.
Category 4B includes lesions that are moderately suggestive of malignancy. The correlation between radiographic diagnosis and pathologic findings is close to unanimous. In this scenario, benign follow-up depends on this concordance. A partially well-defined, partially indistinct mass that may be a fibroadenoma or fatty necrosis is acceptable; however, papilloma requires excisional biopsy.
Category 4C.
Category 4C lesions represent moderately slightly more malignant lesions that do not yet have the typical malignant features of category 5. This category includes, for example, solid masses with indistinct borders, irregular shape, or new, microscopic, pleomorphic clusters of calcifications. Such pathologic findings are often malignant.
These finer classifications of category 4 should encourage pathologists to proceed with further analysis of lesions reported as benign in category 4C, and clinicians should be made aware of the need for follow-up review of cases diagnosed as category 4 but reported as benign on biopsy.
6. Category 5
Category 5 is used to express lesions that are almost certainly breast cancer. In earlier versions of BI-RADS, when puncture biopsies to obtain a histologic or cytologic diagnosis were not widely available, category 5 predicted that the lesion would eventually be disposed of without a prior tissue specimen. Now, specimens with such findings must be retained to detect typical breast cancers with a 95% probability of malignancy. Irregularly shaped masses with burrs, thin strips of calcifications in a segmental or linear pattern, or irregularly shaped masses with burrs and accompanying irregular and polymorphic calcifications are classified as category 5. Normative biopsies without typical malignant lesions are classified as category 4.
7. Category 6
Category 6 is a new category added to describe cases with biopsy-proven breast cancer but only limited prior treatment (e.g., surgical excision, radiation, chemotherapy, or mastectomy). Unlike BI-RADS categories 4 and 5, category 6 does not require interventional management to determine if the lesion is malignant. A second diagnosis found in a prior specimen and shown to be malignant, or to detect the effects of neoadjuvant chemotherapy administered prior to surgery can be assessed as category 6.