2013
140 American College of Radiology
MAGNETIC RESONANCE
IMAGING
biopsy, given less than robust data that support a watchful waiting approach. If surveillance
imaging is undertaken, an increase in size of the mass should prompt immediate biopsy.
Follow-up of NME
NME that is unique and separate from the overall background enhancement should undergo
assessment based on morphology and kinetics. Bright-uid imaging sequences can be helpful
in these instances to demonstrate any associated cysts, which may support a diagnosis of fo-
cal brocystic change and a benign (category 2) assessment. However, limited data on linear,
clumped, and segmental enhancement indicate that these ndings should not be followed,
as the malignancy rate appears to be greater than 2%.
4
At this time, the literature is not suf-
ciently robust to endorse the use of a category 3 assessment for NME.
Timing of Follow-up
Final assessment category 3 is best used for a unique focal nding and managed with an initial
short-interval follow-up (6 months) examination followed by additional examinations until long-
term (2- or 3-year) stability is demonstrated. For category 3 assessments, the initial short-term
follow-up interval is usually 6 months, involving the breast(s) containing the probably benign
nding(s). Assuming stability at this 6-month examination, a category 3 assessment again will
be rendered with a management recommendation for a second short-interval follow-up exami-
nation in 6 months, but now involving both breasts if the opposite breast will be due for routine
annual screening. Again assuming stability at this second short-interval follow up, the examina-
tion is once more assessed as category 3, but now the recommended follow-up interval usually
is lengthened to 1 year due to the already-observed 12-month stability. The recommended 2- or
3-year follow-up in these cases is: 6 months, 6 months, 1 year, and, optionally, 1 more year to es-
tablish stability. After 2 to 3 years of stability, the nding should be assessed as benign (category
2). It should be emphasized that this approach is borrowed from mammography. While the vast
majority of probably benign ndings are managed with follow-up, there may be occasions in
which biopsies are done instead (patient preference or overriding clinical concern). As with any
interpretive examination, a less experienced reader may conclude that a nding such as benign
BPE, for example, should be classied as category 3. A more experienced reader may recognize
this as normal or benign at 6 or 12 months and classify it as category 1 or 2. With a properly
worded report, the assessment category may then be changed to one that the current reader
feels is appropriate, even though long-term stability has not been demonstrated.
It is imperative that surveillance imaging does not alter the stage at diagnosis or prognosis
of the few patients with malignancies who are given category 3 assessments. Because this
has not yet been demonstrated for MRI, as it has been for mammography, careful audit-
ing of the use of category 3 assessments is necessary, and publication of outcomes data
is strongly recommended. Although the data are not robust, it appears the ≤ 2% malig-
nancy rate already demonstrated at mammographic follow-up also may be achieved at
MRI. Several recent publications have demonstrated that focal lesions assigned to category
3 had a ≤ 2% malignancy rate, albeit without use of specic BI-RADS® MRI descriptors for
the lesions included in the studies.
5,6,7
Publication of outcomes data for specic category
3 lesions, using BI-RADS® MRI, is strongly recommended. It should be noted that a ≤ 2%
malignancy rate may be dicult to achieve due to the high-risk population that usually is
studied by MRI (higher than average prior probability of cancer).
A desirable goal for the frequency of making category 3 assessments at MRI is less than 10%.
Over time, this rate should decrease to the point that a mature program should demonstrate