DOI: 10.1148/rg.266065025
RadioGraphics 2006;26:1719-1734
© RSNA, 2006
Patterns of Enhancement on Breast MR Images: Interpretation and Imaging Pitfalls1
Katarzyna J. Macura, MD, PhD,
Ronald Ouwerkerk, PhD,
Michael A. Jacobs, PhD and
David A. Bluemke, MD, PhD
1 From the Russell H. Morgan Department of Radiology and Radiological Sciences (K.J.M., R.O., M.A.J., D.A.B.) and Sidney Kimmel Cancer Center, Department of Oncology (M.A.J.), Johns Hopkins Medical Institutions, 600 N Wolfe St, BLA-B 179 RAD, Baltimore, MD 21287. Presented as an education exhibit at the 2005 RSNA Annual Meeting. Received March 15, 2006; revision requested April 12 and received May 23; accepted May 31. M.A.J. supported by grants NIH 1R01CA100184 and P50 CA103175. R.O. supported by grant 5R21CA095907. All authors have no financial relationships to disclose.
Address correspondence to K.J.M. (e-mail: kmacura{at}jhmi.edu).
 |
Abstract
|
|---|
The role of dynamic contrast material–enhanced magnetic resonance (MR) imaging of the breast as an adjunct to the conventional techniques of mammography and ultrasonography has been established in numerous research studies. MR imaging improves the detection and characterization of primary and recurrent breast cancers and allows evaluation of the response to therapy. The breast imaging lexicon published by the American College of Radiology allows a standardized and consistent description of the morphologic and kinetic characteristics of breast lesions; however, there are many challenges in the interpretation of breast enhancement patterns and kinetics, and many imaging and interpretation pitfalls must be considered. New breast MR imaging techniques that are based on the use of molecular markers of malignancy may help improve lesion characterization. The margin characteristics of a lesion and the intensity of its enhancement at MR imaging 2 minutes or less after contrast material injection are currently considered the most important features for breast lesion diagnosis.
© RSNA, 2006
|
LEARNING OBJECTIVES
|
|---|
After reading this article and taking the test, the reader will be able to:- Describe the different enhancement patterns in lesions on breast MR images and the predictive value of each pattern for cancer detection.
- Recognize the enhancement kinetics that help determine whether a breast lesion is benign or malignant.
- Discuss diagnostic and technical pitfalls of breast MR imaging.
|
Introduction
|
|---|
Dynamic contrast material–enhanced magnetic resonance (MR) imaging of the breast is increasingly used as an adjunct to mammography and ultrasonography (US) to improve the detection and characterization of primary and recurrent breast cancers and for evaluation of the response to therapy. MR imaging is useful for detecting multifocality and multicentricity of breast cancer; differentiating between scar tissue and recurrent cancer after breast-conserving therapy (1); screening patients in high-risk groups (eg, those with the BRCA1 gene) (2); examining breasts that contain implants; examining the breasts of patients with histologically proved metastatic breast cancer with unknown primary origin (3); and, in patients with a finding of cancer in one breast, screening the contralateral breast for occult cancer (4). The sensitivity of MR imaging for the detection of breast cancer is very high, with 90% being the value reported in most studies (5,6). However, with regard to the detection of ductal carcinoma in situ (DCIS), the sensitivity of MR imaging varies between 40% and 100% (7). The result may be false-negative in the presence of DCIS or an invasive ductal or lobular malignancy. Specificity of 37%–100% has been reported for breast MR imaging; in most studies, it has varied from 50% to 70% (5). The relatively low specificity of breast MR imaging is a disadvantage, and rigorous criteria have been proposed for the interpretation of breast MR images (8). In addition, new imaging techniques based on molecular and cellular properties of tissues have been investigated (9,10). The use of computer-assisted methods such as neural network clustering has been proposed to aid in the assessment of contrast enhancement on breast MR images (11).
The American College of Radiology breast MR imaging lexicon provides a specific standardized vocabulary for describing the morphologic and kinetic characteristics of breast lesions (8). All suspicious areas are defined as a focus or foci (with a diameter of less than 5 mm), mass (a three-dimensional space-occupying lesion with a convex margin), or nonmass. The description of a mass should include a characterization of its shape (round, oval, lobular, irregular), margin (smooth, irregular, spiculated), and internal enhancement pattern (homogeneous, heterogeneous, rimlike, central, septal). The description of a nonmass lesion should include a characterization of its distribution (focal or multifocal, linear, ductal, segmental, regional, multiregional, diffuse), internal enhancement (homogeneous, heterogeneous, stippled, clumped, reticular), and symmetry. Associated findings such as edema, adenopathy, cysts, and skin and chest wall involvement are reported as well. The kinetic curve assessment involves a description of both the initial peak (slow, medium, rapid) and the delayed phase (persistent, plateau, or washout) of contrast enhancement.
In this article, we discuss the different patterns of enhancement, each of which has a specific meaning with regard to the diagnosis of breast abnormalities. The discussion is based on the results of several research studies in which the predictive value of particular imaging features was investigated during the interpretation of contrast-enhanced breast MR images. We also focus on potential diagnostic and technical pitfalls of breast MR imaging and discuss emerging techniques that are aimed at improving lesion characterization with the use of molecular markers of malignancy.
|
Enhancement Patterns
|
|---|
Lack of Enhancement
The absence of a visible lesion on contrast-enhanced MR images that corresponds to a palpable or mammographically visible abnormality is highly predictive of a benign finding. However, the absence of observed enhancement at breast MR imaging does not exclude in situ or invasive cancer. Many invasive cancers that show no enhancement are small or have a small invasive component (12). Lack of enhancement has a high negative predictive value (NPV) for malignancy (88%–96%) (12,13). Among nonenhancing tumors, about 48% are DCIS (7,12), and 52% are invasive carcinomas (12,13).
Morphologic Criteria for Benignity
Many benign breast lesions may appear enhanced, including nonproliferative lesions (mild hyperplasia, fibroadenomas), proliferative lesions without atypia (sclerosing adenosis, radial and complexing sclerosing lesions, moderate hyperplasia, intraductal papillomas), and atypical lobular and ductal hyperplasia. The normal breast parenchyma, especially in premenopausal women, also may appear focally enhanced, and this appearance may lead to a false-positive finding (14).
A typical benign feature for masses is a smooth margin (NPV, 95%). Low-signal-intensity internal septa were demonstrated to be predictive of benignity (NPV, 98%) (13) (Fig 1). However, recent results of a study by Schnall et al revealed that 47% of malignant lesions were shown to have nonenhancing internal septa (12). If a mass is lobulated and shows no enhancement or only minimal enhancement, it is likely benign (NPV, 100%) (13). If a mass is lobulated and shows moderate to marked enhancement (NPV, 67%), further evaluation may be warranted (13). Correlation between the enhancing portion of the lesion and its T2-weighted signal intensity is helpful (15). T2-weighted signal hyperintensity in the same portion of the lesion that appears enhanced on T1-weighted images is highly suggestive of benignity, although not all masses with high signal intensity on T2-weighted images are benign. A finding of mild regional nonmass enhancement also is suggestive of a benign abnormality (NPV, 92%).
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Benign features. (a, b) Contrast-enhanced T1-weighted fat-saturated gradient-echo (GRE) (repetition time msec/echo time msec, 20/4.5; flip angle, 30°) images from a 51-year-old woman show regional micronodular (<5 mm stippled or punctate) enhancement in fibrocystic breast tissue (a) and rimlike enhancement around a cyst (arrow in b) within a region of fibrocystic breast tissue. (c) T2-weighted (4000/90) fat-saturated image (same patient as in a and b) shows the cyst (arrow). Comparison of the contrast-enhanced images with the T2-weighted image is important to recognize the benign nature of these findings. (d) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 44-year-old woman shows an oval mass with smooth and lobular margins and enhancement with dark internal septa, typical of a fibroadenoma (arrow). (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows increased signal intensity in the lesion (arrow) in comparison with that seen in d. Increased T2-weighted signal intensity is typical of a myxoid fibroadenoma in a woman of this age. In women older than 50 years, age-related sclerotic changes in fibroadenomas result in a greater prevalence of lesions with fibrotic low signal intensity, which reduces the diagnostic usefulness of the T2-weighted imaging characteristic.
|
|
View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Benign features. (a, b) Contrast-enhanced T1-weighted fat-saturated gradient-echo (GRE) (repetition time msec/echo time msec, 20/4.5; flip angle, 30°) images from a 51-year-old woman show regional micronodular (<5 mm stippled or punctate) enhancement in fibrocystic breast tissue (a) and rimlike enhancement around a cyst (arrow in b) within a region of fibrocystic breast tissue. (c) T2-weighted (4000/90) fat-saturated image (same patient as in a and b) shows the cyst (arrow). Comparison of the contrast-enhanced images with the T2-weighted image is important to recognize the benign nature of these findings. (d) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 44-year-old woman shows an oval mass with smooth and lobular margins and enhancement with dark internal septa, typical of a fibroadenoma (arrow). (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows increased signal intensity in the lesion (arrow) in comparison with that seen in d. Increased T2-weighted signal intensity is typical of a myxoid fibroadenoma in a woman of this age. In women older than 50 years, age-related sclerotic changes in fibroadenomas result in a greater prevalence of lesions with fibrotic low signal intensity, which reduces the diagnostic usefulness of the T2-weighted imaging characteristic.
|
|
View larger version (101K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c. Benign features. (a, b) Contrast-enhanced T1-weighted fat-saturated gradient-echo (GRE) (repetition time msec/echo time msec, 20/4.5; flip angle, 30°) images from a 51-year-old woman show regional micronodular (<5 mm stippled or punctate) enhancement in fibrocystic breast tissue (a) and rimlike enhancement around a cyst (arrow in b) within a region of fibrocystic breast tissue. (c) T2-weighted (4000/90) fat-saturated image (same patient as in a and b) shows the cyst (arrow). Comparison of the contrast-enhanced images with the T2-weighted image is important to recognize the benign nature of these findings. (d) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 44-year-old woman shows an oval mass with smooth and lobular margins and enhancement with dark internal septa, typical of a fibroadenoma (arrow). (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows increased signal intensity in the lesion (arrow) in comparison with that seen in d. Increased T2-weighted signal intensity is typical of a myxoid fibroadenoma in a woman of this age. In women older than 50 years, age-related sclerotic changes in fibroadenomas result in a greater prevalence of lesions with fibrotic low signal intensity, which reduces the diagnostic usefulness of the T2-weighted imaging characteristic.
|
|
View larger version (100K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1d. Benign features. (a, b) Contrast-enhanced T1-weighted fat-saturated gradient-echo (GRE) (repetition time msec/echo time msec, 20/4.5; flip angle, 30°) images from a 51-year-old woman show regional micronodular (<5 mm stippled or punctate) enhancement in fibrocystic breast tissue (a) and rimlike enhancement around a cyst (arrow in b) within a region of fibrocystic breast tissue. (c) T2-weighted (4000/90) fat-saturated image (same patient as in a and b) shows the cyst (arrow). Comparison of the contrast-enhanced images with the T2-weighted image is important to recognize the benign nature of these findings. (d) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 44-year-old woman shows an oval mass with smooth and lobular margins and enhancement with dark internal septa, typical of a fibroadenoma (arrow). (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows increased signal intensity in the lesion (arrow) in comparison with that seen in d. Increased T2-weighted signal intensity is typical of a myxoid fibroadenoma in a woman of this age. In women older than 50 years, age-related sclerotic changes in fibroadenomas result in a greater prevalence of lesions with fibrotic low signal intensity, which reduces the diagnostic usefulness of the T2-weighted imaging characteristic.
|
|
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1e. Benign features. (a, b) Contrast-enhanced T1-weighted fat-saturated gradient-echo (GRE) (repetition time msec/echo time msec, 20/4.5; flip angle, 30°) images from a 51-year-old woman show regional micronodular (<5 mm stippled or punctate) enhancement in fibrocystic breast tissue (a) and rimlike enhancement around a cyst (arrow in b) within a region of fibrocystic breast tissue. (c) T2-weighted (4000/90) fat-saturated image (same patient as in a and b) shows the cyst (arrow). Comparison of the contrast-enhanced images with the T2-weighted image is important to recognize the benign nature of these findings. (d) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 44-year-old woman shows an oval mass with smooth and lobular margins and enhancement with dark internal septa, typical of a fibroadenoma (arrow). (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows increased signal intensity in the lesion (arrow) in comparison with that seen in d. Increased T2-weighted signal intensity is typical of a myxoid fibroadenoma in a woman of this age. In women older than 50 years, age-related sclerotic changes in fibroadenomas result in a greater prevalence of lesions with fibrotic low signal intensity, which reduces the diagnostic usefulness of the T2-weighted imaging characteristic.
|
|
Morphologic Criteria for Malignancy
The description of the margin of a focal mass is the most predictive feature of the breast MR image interpretation. Irregular or spiculated margins have a positive predictive value (PPV) of 84%–91% (13) (Fig 2). Rimlike enhancement highly correlates with a cancer diagnosis (PPV, 84%) (12); however, this feature is uncommon, with a prevalence of 16%. Other features associated with malignancy include heterogeneous internal enhancement and enhancing internal septa.
View larger version (87K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Malignant features. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows two masses in the left breast of a 44-year-old woman, one with thin rimlike enhancement at the 4-o’clock position (arrowhead) and the other with heterogeneous enhancement and enhanced internal septa at the 2-o’clock position (arrow). (b) T2-weighted (4000/90) fat-saturated image (same patient as in a) shows low signal intensity in the portions of the masses that appeared enhanced in a. A central region of necrosis in the mass at the 4-o’clock position shows increased internal T2-weighted signal intensity. The masses proved to be poorly differentiated ductal carcinoma with necrosis and signet ring cell features. (c) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 42-year-old woman shows a spiculated margin in an infiltrating carcinoma with ductal and lobular features. (d) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image from a 52-year-old woman shows a retroareolar mass (arrow) with an irregular margin and heterogeneous enhancement. The results of histologic analysis indicated infiltrating ductal carcinoma. (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows the mass (arrow) with low signal intensity. Note the focal skin thickening and nipple retraction.
|
|
View larger version (80K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Malignant features. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows two masses in the left breast of a 44-year-old woman, one with thin rimlike enhancement at the 4-o’clock position (arrowhead) and the other with heterogeneous enhancement and enhanced internal septa at the 2-o’clock position (arrow). (b) T2-weighted (4000/90) fat-saturated image (same patient as in a) shows low signal intensity in the portions of the masses that appeared enhanced in a. A central region of necrosis in the mass at the 4-o’clock position shows increased internal T2-weighted signal intensity. The masses proved to be poorly differentiated ductal carcinoma with necrosis and signet ring cell features. (c) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 42-year-old woman shows a spiculated margin in an infiltrating carcinoma with ductal and lobular features. (d) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image from a 52-year-old woman shows a retroareolar mass (arrow) with an irregular margin and heterogeneous enhancement. The results of histologic analysis indicated infiltrating ductal carcinoma. (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows the mass (arrow) with low signal intensity. Note the focal skin thickening and nipple retraction.
|
|
View larger version (64K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c. Malignant features. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows two masses in the left breast of a 44-year-old woman, one with thin rimlike enhancement at the 4-o’clock position (arrowhead) and the other with heterogeneous enhancement and enhanced internal septa at the 2-o’clock position (arrow). (b) T2-weighted (4000/90) fat-saturated image (same patient as in a) shows low signal intensity in the portions of the masses that appeared enhanced in a. A central region of necrosis in the mass at the 4-o’clock position shows increased internal T2-weighted signal intensity. The masses proved to be poorly differentiated ductal carcinoma with necrosis and signet ring cell features. (c) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 42-year-old woman shows a spiculated margin in an infiltrating carcinoma with ductal and lobular features. (d) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image from a 52-year-old woman shows a retroareolar mass (arrow) with an irregular margin and heterogeneous enhancement. The results of histologic analysis indicated infiltrating ductal carcinoma. (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows the mass (arrow) with low signal intensity. Note the focal skin thickening and nipple retraction.
|
|
View larger version (72K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2d. Malignant features. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows two masses in the left breast of a 44-year-old woman, one with thin rimlike enhancement at the 4-o’clock position (arrowhead) and the other with heterogeneous enhancement and enhanced internal septa at the 2-o’clock position (arrow). (b) T2-weighted (4000/90) fat-saturated image (same patient as in a) shows low signal intensity in the portions of the masses that appeared enhanced in a. A central region of necrosis in the mass at the 4-o’clock position shows increased internal T2-weighted signal intensity. The masses proved to be poorly differentiated ductal carcinoma with necrosis and signet ring cell features. (c) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 42-year-old woman shows a spiculated margin in an infiltrating carcinoma with ductal and lobular features. (d) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image from a 52-year-old woman shows a retroareolar mass (arrow) with an irregular margin and heterogeneous enhancement. The results of histologic analysis indicated infiltrating ductal carcinoma. (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows the mass (arrow) with low signal intensity. Note the focal skin thickening and nipple retraction.
|
|
View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2e. Malignant features. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows two masses in the left breast of a 44-year-old woman, one with thin rimlike enhancement at the 4-o’clock position (arrowhead) and the other with heterogeneous enhancement and enhanced internal septa at the 2-o’clock position (arrow). (b) T2-weighted (4000/90) fat-saturated image (same patient as in a) shows low signal intensity in the portions of the masses that appeared enhanced in a. A central region of necrosis in the mass at the 4-o’clock position shows increased internal T2-weighted signal intensity. The masses proved to be poorly differentiated ductal carcinoma with necrosis and signet ring cell features. (c) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image from a 42-year-old woman shows a spiculated margin in an infiltrating carcinoma with ductal and lobular features. (d) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image from a 52-year-old woman shows a retroareolar mass (arrow) with an irregular margin and heterogeneous enhancement. The results of histologic analysis indicated infiltrating ductal carcinoma. (e) T2-weighted (4000/90) fat-saturated image (same patient as in d) shows the mass (arrow) with low signal intensity. Note the focal skin thickening and nipple retraction.
|
|
A linear or branching ductal distribution of enhancement is difficult to assess and interobserver agreement is poor, according to a report by the MR imaging lexicon committee (16), and the reported PPV for this feature has varied from 24% to 85% (5).
A moderate to marked degree of nonmass regional enhancement has a PPV of 59% (13). Regional enhancement with a micronodular (stippled) pattern was found in both benign (eg, fibrocystic disease) and malignant (eg, DCIS) abnormalities (13). Stippled enhancement is associated with a low incidence of malignancy (25%), while clumped, heterogeneous, and homogeneous enhancement are associated with a 60%, 53%, and 67% likelihood of cancer, respectively (12). Regional enhancement associated with a focal lesion was shown to have a PPV of 81%; however, this finding has a low prevalence and is seen in only 14% of patients with a focal malignancy. A segmental distribution of enhancement was associated with a 78% likelihood of cancer, while a regional distribution was associated with a 21% likelihood of cancer. Therefore, mild stippled enhancement with a regional distribution is indicative of benignity, whereas segmental or clumped enhancement is more likely to indicate malignancy (12).
|
Other Features
|
|---|
T2-weighted Signal Characteristics
T2-weighted signal hyperintensity within the viable (enhancing) portion of the lesion is highly suggestive of benign histology (15). However, T2-weighted signal intensity is not a reliable predictor of benignity in irregular or spiculated masses. Colloid (mucinous) carcinoma may be manifested as a hyperintense and minimally enhancing mass on T2-weighted images (13,17). On T2-weighted images, breast cancers are more likely (87%) to have iso- or hypointense signal, compared to the signal intensity of the normal breast parenchyma (Fig 2) (18). Hypointense internal septa on T2-weighted images are typical of benign fibroadenomas (Fig 1). Rarely, adenoid cystic carcinoma may contain low-signal-intensity septa that are nonenhanced (13).
The assessment of lesions on T2-weighted images is most helpful in younger patients. Myxoid fibroadenomas in young women typically demonstrate increased T2-weighted signal intensity. However, there are age-related sclerotic changes in fibroadenomas in older women, and the greater prevalence of fibrotic low-signal-intensity fibroadenomas in women older than 50 years reduces the diagnostic value of the T2-weighted signal intensity criterion (18). Sclerotic fibroadenomas may have the same hypointense appearance as well-circumscribed breast cancers. However, with the progression of fibrosis, the enhancement rate of fibroadenomas decreases. Thus, the primary diagnostic criterion of enhancement may be used to avoid unnecessary referral for biopsy. Medullary cancers tend to occur in younger women and may have a circumscribed appearance. The signal intensity of the lesion on T2-weighted images may not aid in the differential diagnosis of myxoid fibroadenoma and medullary cancer. For these reasons, the T2-weighted signal intensity of a breast lesion is used as a secondary criterion, to confirm a diagnosis of benignity established on the basis of morphologic features (the primary criterion), and not to rule out breast cancer.
Focal Perilesional Edema
A focal area of hyperintense T2-weighted signal near a lesion is highly suggestive of malignancy. The signal intensity increase is thought to be related to increased capillary permeability in the presence of tumor-related angiogenesis (15).
Architectural Distortion
An architectural distortion that is nonenhanced may represent a radial scar; if enhanced, it is highly suggestive of invasive cancer. Desmoplastic tethering (the hook sign) has been described as highly indicative of a malignancy in an untreated breast (15). This feature, which is thought to represent invasion of the Cooper ligaments in the direction of the pectoral muscle, is best evaluated on T2-weighted images. The hook sign is not observed in the presence of noninvasive cancers.
Skin Thickening and Edema
Skin thickening and edema in an untreated breast on MR images, as on mammograms, may be signs of malignancy, especially of inflammatory carcinoma. In a treated breast, these features are frequently observed after radiation therapy.
Lymph Nodes
The lack of visibility or the subtle visibility of axillary lymph nodes is almost as commonly associated with benign lesions as with malignant ones. Therefore, the absence of adenopathy does not help differentiate benignity from malignancy. In contrast, nodes with a diameter of more than 1 cm are more common in malignant cases (15). In addition, a loss of the fatty hilum of the lymph node is suggestive of malignant involvement.
|
Enhancement Kinetics
|
|---|
Breast lesion enhancement can be characterized qualitatively by assessing the enhancement kinetics curve obtained by plotting the signal intensity values in breast tissue intensity over time after contrast material injection. These curves show both the initial slope of increasing enhancement, from the baseline value to the peak, and the subsequent trend. Three enhancement patterns can be identified on the basis of the signal intensity–time curve.
Type I is a pattern of progressive enhancement, with a continuous increase in signal intensity on each successive contrast-enhanced image (Fig 3). This enhancement pattern is usually associated with a benign finding (83% benign, 9% malignant) (19). Its sensitivity and specificity for indication of a benign lesion are 52.2% and 71%, respectively (6). However, Schnall et al reported that readers in a multiinstitutional trial described the enhancement kinetics as persistent in 45% of lesions that proved to be cancers (12).
View larger version (93K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Type I enhancement curve in a lesion stable over 2 years. The patient, a 28-year-old woman, had contralateral breast cancer (Fig 6). (a) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image shows an 8-mm-diameter mass (arrow) with minimal lobulation and enhancement. (b) T2-weighted (4000/90) fat-saturated image shows an area of high signal intensity (arrow) within the lesion. (c) Curve indicates progressive enhancement in the lesion. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds.
|
|
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Type I enhancement curve in a lesion stable over 2 years. The patient, a 28-year-old woman, had contralateral breast cancer (Fig 6). (a) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image shows an 8-mm-diameter mass (arrow) with minimal lobulation and enhancement. (b) T2-weighted (4000/90) fat-saturated image shows an area of high signal intensity (arrow) within the lesion. (c) Curve indicates progressive enhancement in the lesion. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds.
|
|
View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. Type I enhancement curve in a lesion stable over 2 years. The patient, a 28-year-old woman, had contralateral breast cancer (Fig 6). (a) Contrast-enhanced T1-weighted GRE (20/4.5; flip angle, 30°) subtraction image shows an 8-mm-diameter mass (arrow) with minimal lobulation and enhancement. (b) T2-weighted (4000/90) fat-saturated image shows an area of high signal intensity (arrow) within the lesion. (c) Curve indicates progressive enhancement in the lesion. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds.
|
|
Type II is a plateau pattern, in which an initial increase in signal intensity is followed by a flattening of the enhancement curve (Fig 4). This pattern has a sensitivity of 42.6% and specificity of 75% for the detection of malignancy (6).
View larger version (82K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Type II curve in a multifocal and bilateral invasive lobular carcinoma in a 39-year-old woman. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows regional nonmass enhancement. (b) Color-coded map shows foci with the maximum slope of enhancement increase (red) after contrast material injection. Two foci were selected as regions of interest (ROIs). (c) Curves indicate plateau enhancement. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in minutes.
|
|
View larger version (86K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Type II curve in a multifocal and bilateral invasive lobular carcinoma in a 39-year-old woman. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows regional nonmass enhancement. (b) Color-coded map shows foci with the maximum slope of enhancement increase (red) after contrast material injection. Two foci were selected as regions of interest (ROIs). (c) Curves indicate plateau enhancement. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in minutes.
|
|
View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. Type II curve in a multifocal and bilateral invasive lobular carcinoma in a 39-year-old woman. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows regional nonmass enhancement. (b) Color-coded map shows foci with the maximum slope of enhancement increase (red) after contrast material injection. Two foci were selected as regions of interest (ROIs). (c) Curves indicate plateau enhancement. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in minutes.
|
|
Type III, a washout enhancement pattern, involves an initial increase and subsequent decrease in signal intensity (Figs 5, 6). This pattern is not usually seen in patients with benign lesions (specificity, 90.4%), but it has a sensitivity of only 20.5% (6). Schnall et al reported that 76% of curves that showed a washout pattern were associated with cancer. Both type II and type III curves should be considered suggestive of malignancy.
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Type III curve from multifocal infiltrating lobular carcinoma in a 40-year-old woman. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows regions of clumped nonmasslike enhancement in all four quadrants. (b) Color-coded map of the maximum slope of enhancement shows three ROIs selected in areas of rapid peak enhancement (red). (c) Enhancement curves indicate early washout. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in minutes.
|
|
View larger version (95K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Type III curve from multifocal infiltrating lobular carcinoma in a 40-year-old woman. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows regions of clumped nonmasslike enhancement in all four quadrants. (b) Color-coded map of the maximum slope of enhancement shows three ROIs selected in areas of rapid peak enhancement (red). (c) Enhancement curves indicate early washout. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in minutes.
|
|
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. Type III curve from multifocal infiltrating lobular carcinoma in a 40-year-old woman. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows regions of clumped nonmasslike enhancement in all four quadrants. (b) Color-coded map of the maximum slope of enhancement shows three ROIs selected in areas of rapid peak enhancement (red). (c) Enhancement curves indicate early washout. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in minutes.
|
|
View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Moderately to poorly differentiated palpable mass (infiltrating ductal carcinoma) in a 28-year-old woman, 3 months postpartum. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows an irregular mass with an irregular and spiculated margin, features typical of invasive carcinoma. (b) T2-weighted (4000/90) fat-saturated image shows a region of low signal intensity (arrows) in the portion of the mass that appears enhanced in a. (c) The enhancement curve is type III, with an early peak and delayed phase washout. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Moderately to poorly differentiated palpable mass (infiltrating ductal carcinoma) in a 28-year-old woman, 3 months postpartum. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows an irregular mass with an irregular and spiculated margin, features typical of invasive carcinoma. (b) T2-weighted (4000/90) fat-saturated image shows a region of low signal intensity (arrows) in the portion of the mass that appears enhanced in a. (c) The enhancement curve is type III, with an early peak and delayed phase washout. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds.
|
|
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6c. Moderately to poorly differentiated palpable mass (infiltrating ductal carcinoma) in a 28-year-old woman, 3 months postpartum. (a) Contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows an irregular mass with an irregular and spiculated margin, features typical of invasive carcinoma. (b) T2-weighted (4000/90) fat-saturated image shows a region of low signal intensity (arrows) in the portion of the mass that appears enhanced in a. (c) The enhancement curve is type III, with an early peak and delayed phase washout. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds.
|
|
The specificity of breast MR imaging is improved when both morphologic and kinetic features are considered in the interpretation (20). Because of the overlap in enhancement characteristics between benign and malignant lesions, reliance on a kinetics assessment alone is not recommended. The exclusion of cancer on the basis of persistent enhancement (a type I curve) alone would lead to false-negative results (12). The quantitative features most indicative of cancer were reported to be the maximum enhancement rate (percentage of enhancement per second) and the percentage of enhancement at 1 minute (12), with rapid enhancement being characteristic of a malignancy.
|
Imaging Technique
|
|---|
Various MR imaging protocols may be used to obtain breast images of acceptable quality. Most studies reported in the literature were performed at 1.5 T; fewer, at 1.0 T. A standard protocol includes a T2-weighted rapid (fast or turbo) spin-echo (repetition time msec/echo time msec, 4000/90; section thickness, 
4 mm) acquisition and three-dimensional T1-weighted GRE (20/4.5; flip angle, 30°–45°; section thickness, 3 mm) acquisitions before and after the intravenous administration of contrast material, with the usual dose of 0.1 mmol/kg injected as a bolus and often followed by a 10–20-mL saline flush (injected manually or with a power injector). For breast imaging in the sagittal plane, an image matrix of 256 x 192 can be used with zero-filled interpolation to 512 x 512, a small field of view (16–18 cm), and chemical fat suppression. For bilateral axial breast imaging, the field of view is increased to approximately 30 cm, and high-resolution matrices (between 256 and 512) are used to reduce the voxel size.
Dedicated Breast Surface Coils
A dedicated breast surface coil should always be used. Both unilateral and bilateral coils are commercially available. Unilateral breast MR imaging typically is performed when only a two-channel coil is available; one breast can be imaged on one day, and the patient may return the next day for imaging of the contralateral breast. Most contemporary breast coils have a multicoil phased-array configuration that has been optimized for bilateral imaging of the breasts, chest wall, and axillae. Bilateral breast MR imaging is currently performed with four-channel and eight-channel coils (16-channel coils are expected to be available in the near future), with signal-to-noise improvements resulting from the higher number of channels. Bilateral imaging has the advantage of being quicker and more convenient for the patient. It also allows direct comparison between the two breasts in one field of view, thereby facilitating the assessment of symmetry during screening. However, higher spatial resolution can be obtained with unilateral breast imaging, because, for a given image matrix size, in-plane spatial resolution increases with a reduction in the field of view. It is important to select the right-to-left and superior-to-inferior directions as the phase-encoding direction for axial and sagittal acquisitions, respectively, to reduce motion artifacts. Most coils are open at the side to allow lateral access for intervention and are equipped with a compression plate. The advantages of breast compression during imaging-guided biopsy are minimization of patient motion, decreased thickness of the breast, and shorter imaging time. However, excessive compression may compromise the enhancement of breast tumors, leading to a false-negative result (21); therefore, only minimal or no compression should be applied to immobilize the breast during routine breast MR examinations.
Contrast Enhancement
Most tumors can be detected only after intravenous contrast material administration. High-resolution T1-weighted images without or with fat saturation, with thin sections and no gap, are obtained for optimal sensitivity. These requirements are readily achieved by applying three-dimensional GRE sequences with an acquisition time of less than 2 minutes. Rapid imaging allows the application of several contrast-enhanced sequences in a single examination for simultaneous analysis of both the morphologic features and the enhancement kinetics of lesions. It is important to achieve optimal contrast between breast lesions and the surrounding glandular and fatty tissue; therefore, subtraction of unenhanced images from contrast-enhanced images is crucial. Morphologic features (Figs 7–9) are best evaluated on high-spatial-resolution images; thus, the acquisition of thin sections and the use of a small field of view are recommended (22). An analysis of architectural features can be performed at any point during the first 4 minutes 30 seconds after the contrast material injection (23).
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Enhancement patterns in DCIS in four patients. DCIS may be manifested as stippled and clumped regional or segmental enhancement; linear and branching ductal enhancement; focal mass enhancement with spiculated, irregular, lobulated, or smooth margins; focal enhancement with a diameter of less than 5 mm; or no enhancement. (a) Segmental linear and reticular enhancement. (b) Linear enhancement. (c) Spiculated 1.5-cm mass within extensive sclerosing adenosis. (d) Small oval mass with smooth margins.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Enhancement patterns in DCIS in four patients. DCIS may be manifested as stippled and clumped regional or segmental enhancement; linear and branching ductal enhancement; focal mass enhancement with spiculated, irregular, lobulated, or smooth margins; focal enhancement with a diameter of less than 5 mm; or no enhancement. (a) Segmental linear and reticular enhancement. (b) Linear enhancement. (c) Spiculated 1.5-cm mass within extensive sclerosing adenosis. (d) Small oval mass with smooth margins.
|
|
View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. Enhancement patterns in DCIS in four patients. DCIS may be manifested as stippled and clumped regional or segmental enhancement; linear and branching ductal enhancement; focal mass enhancement with spiculated, irregular, lobulated, or smooth margins; focal enhancement with a diameter of less than 5 mm; or no enhancement. (a) Segmental linear and reticular enhancement. (b) Linear enhancement. (c) Spiculated 1.5-cm mass within extensive sclerosing adenosis. (d) Small oval mass with smooth margins.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7d. Enhancement patterns in DCIS in four patients. DCIS may be manifested as stippled and clumped regional or segmental enhancement; linear and branching ductal enhancement; focal mass enhancement with spiculated, irregular, lobulated, or smooth margins; focal enhancement with a diameter of less than 5 mm; or no enhancement. (a) Segmental linear and reticular enhancement. (b) Linear enhancement. (c) Spiculated 1.5-cm mass within extensive sclerosing adenosis. (d) Small oval mass with smooth margins.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Enhancement patterns in invasive ductal carcinoma in four patients. (a) Lobular mass with smooth margins. (b) Mass with spiculated margins. (c) Lobular mass with irregular margins. (d) Irregular mass associated with clumped segmental enhancement.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Enhancement patterns in invasive ductal carcinoma in four patients. (a) Lobular mass with smooth margins. (b) Mass with spiculated margins. (c) Lobular mass with irregular margins. (d) Irregular mass associated with clumped segmental enhancement.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8c. Enhancement patterns in invasive ductal carcinoma in four patients. (a) Lobular mass with smooth margins. (b) Mass with spiculated margins. (c) Lobular mass with irregular margins. (d) Irregular mass associated with clumped segmental enhancement.
|
|
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8d. Enhancement patterns in invasive ductal carcinoma in four patients. (a) Lobular mass with smooth margins. (b) Mass with spiculated margins. (c) Lobular mass with irregular margins. (d) Irregular mass associated with clumped segmental enhancement.
|
|
View larger version (123K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Enhancement patterns in invasive lobular carcinoma in four patients. (a) Mass with spiculated margins. (b) Mass with rimlike enhancement and associated segmental clumped enhancement. (c) Irregular mass. (d) Nonmasslike regional moderate enhancement.
|
|
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Enhancement patterns in invasive lobular carcinoma in four patients. (a) Mass with spiculated margins. (b) Mass with rimlike enhancement and associated segmental clumped enhancement. (c) Irregular mass. (d) Nonmasslike regional moderate enhancement.
|
|
View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9c. Enhancement patterns in invasive lobular carcinoma in four patients. (a) Mass with spiculated margins. (b) Mass with rimlike enhancement and associated segmental clumped enhancement. (c) Irregular mass. (d) Nonmasslike regional moderate enhancement.
|
|
View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9d. Enhancement patterns in invasive lobular carcinoma in four patients. (a) Mass with spiculated margins. (b) Mass with rimlike enhancement and associated segmental clumped enhancement. (c) Irregular mass. (d) Nonmasslike regional moderate enhancement.
|
|
Trade-off between Spatial and Temporal Resolution
With increased spatial resolution (with an imaging matrix of 512 x 512) and decreased temporal resolution, the depiction of subtle morphologic details improves significantly, and improved depiction of breast lesions leads to an increase in diagnostic accuracy. There is some loss of kinetic information regarding enhancement rates (eg, the wash-in rate within the first 1–2 minutes), but, compared with morphologic features, enhancement rates are a weaker diagnostic criterion for a finding of malignancy. However, kinetic information regarding the enhancement time-course and enhancement pattern is preserved with acquisition times of 2 minutes or less (22).
Interpretation Pitfalls
Excessive Acquisition Time.—
If the contrast-enhanced imaging time is longer than 2 minutes and the first contrast-enhanced images are obtained too late after peak enhancement, washout may no longer be depicted in the descending part of the signal intensity–time curve.
Excessive Size and Incorrect Placement of an ROI.—
An ROI of optimal size covers only part of the lesion and should be placed in the region of strongest enhancement on the first contrast-enhanced image, a feature that is most clearly distinguishable on color maps of the maximum slope of enhancement. When an ROI is randomly placed in the mass, the enhancement curve may be variable and yield lower specificity (Fig 10).
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Infiltrating ductal carcinoma with micropapillary features in a 52-year-old woman. (a) Axial contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows an intensely enhanced 10-mm mass. The dotted vertical line indicates the sagittal plane in b. The use of multiplanar coordinates allows precise localization of masses and correlation between contrast-enhanced images and T2-weighted images. (b) Sagittal T2-weighted (4000/90) fat-saturated image shows an area of low signal intensity characteristic of a malignancy (arrow). Note that the dotted line is the same as the vertical coordinate in a. (c) Color-coded map of the maximum slope of enhancement shows that peak enhancement was off center and not covered by the ROI that was selected in the center of the mass. (d) Enhancement curve for the ROI in c shows a type II pattern. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds. (e) Color-coded map of the maximum slope of enhancement shows the ROI repositioned in the area of peak enhancement. (f) Enhancement curve for the ROI in e shows a type III washout pattern. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds. For the assessment of enhancement kinetics, it is important to select an ROI in the portion of the tumor with maximum peak enhancement. When an ROI is randomly placed in a mass, the enhancement curve may be variable and yield lower specificity.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Infiltrating ductal carcinoma with micropapillary features in a 52-year-old woman. (a) Axial contrast-enhanced T1-weighted fat-saturated GRE (20/4.5; flip angle, 30°) image shows an intensely enhanced 10-mm mass. The dotted vertical line indicates the sagittal plane in b. The use of multiplanar coordinates allows precise localization of masses and correlation between contrast-enhanced images and T2-weighted images. (b) Sagittal T2-weighted (4000/90) fat-saturated image shows an area of low signal intensity characteristic of a malignancy (arrow). Note that the dotted line is the same as the vertical coordinate in a. (c) Color-coded map of the maximum slope of enhancement shows that peak enhancement was off center and not covered by the ROI that was selected in the center of the mass. (d) Enhancement curve for the ROI in c shows a type II pattern. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds. (e) Color-coded map of the maximum slope of enhancement shows the ROI repositioned in the area of peak enhancement. (f) Enhancement curve for the ROI in e shows a type III washout pattern. The vertical axis indicates the percentage of enhancement, and the horizontal axis indicates the time in seconds. For the assessment of enhancement kinetics, it is important to select an ROI in the portion of the tumor with maximum peak enhancement. When an ROI is randomly placed in a mass, the enhancement curve may be variable and yield lower specificity.
|
|
Top
Abstract
LEARNING OBJECTIVES
