Other
Breast Imaging
January 2004

Sonographic, Mammographic, and Histopathologic Correlation of Symptomatic Ductal Carcinoma In Situ

Abstract

OBJECTIVE. The purpose of this study was to describe the features of symptomatic ductal carcinoma in situ (DCIS) of the breast shown on high-resolution sonography and to correlate them with findings from mammography and histopathology to evaluate the prognostic ability of sonographic findings.
MATERIALS AND METHODS. We retrospectively reviewed mammographic and sonographic images of 60 DCIS lesions from 55 symptomatic women. Images were reviewed by a radiologist who knew that the patients had DCIS but had no other information regarding pathology. Lesions were evaluated pathologically and classified using the Van Nuys classification system. Statistical comparisons were made using Fisher's exact test.
RESULTS. Of the 60 lesions, 33 were classified as Van Nuys group 1, 19 as Van Nuys group 2, and eight as Van Nuys group 3. Six (10%) of the 60 lesions were not visible on sonography, and 12 lesions (20%) were not visible on mammography. Sonography revealed a mass in 43 cases (72%), ductal changes in 14 cases (23%), and architectural distortion in four cases (7%). Eight lesions had more than one of these features. A sonographically visualized, irregularly shaped mass with indistinct or angular margins and no posterior acoustic shadowing or enhancement was associated with a high Van Nuys classification (p < 0.05). Microcalcifications were visible on sonography in 13 (22%) of the 60 lesions or on mammography in 25 lesions (42%). Both findings were associated with a high Van Nuys classification (p < 0.05).
CONCLUSION. Although sonography can reveal microcalcifications within masses, it is unreliable in depicting and characterizing the morphology and extent of microcalcifications, particularly when they are in isolation. Therefore, sonography should not be used to replace mammography but instead as an adjunctive tool to increase the sensitivity of mammography in breast diagnosis.

Introduction

Ductal carcinoma in situ (DCIS) is a malignant proliferation of ductal epithelium that is confined by the basement membrane of the involved breast ducts. Encountered increasingly because of the widespread use of mammographic screening in asymptomatic women, DCIS accounts for up to 30% of breast cancers in screened populations and approximately 5% of breast carcinomas in symptomatic women [13]. The malignant potential of DCIS [4] and its behavior after treatment [5, 6] are varied and depend on its architectural pattern and histopathologic grade. The Van Nuys classification system [7] defines three distinct and easily recognizable groups of DCIS determined by the presence or absence of high nuclear grade and comedo necrosis: group 1, low to intermediate nuclear grade DCIS without comedo necrosis; group 2, low to intermediate nuclear grade DCIS with comedo necrosis; group 3, high nuclear grade DCIS with or without comedo necrosis. Research has shown that nuclear grade, tumor size, and margin width are significant predictors of local recurrence, with the presence of comedo necrosis also approaching multivariate significance [8]. The Van Nuys Prognostic Index predicts clinical recurrence in women who have been treated with breast conservation therapy on the basis of three factors: Van Nuys group, margin status, and tumor size [79].
The mammographic features of DCIS have been well described in the literature, with microcalcifications being the dominant feature [1012]. Other findings such as masses, nodular abnormalities, architectural distortions, dilated retroareolar ducts, and developing densities have also been reported [13, 14]. Little systematic description of the gray-scale and color power Doppler sonographic features of DCIS is found in the literature [1517], although several recent studies have examined the role of sonography in the evaluation of mammographically detected microcalcifications [1820].
The purpose of this study was to describe the spectrum of sonographic features of symptomatic DCIS and to evaluate the ability of sonography to predict the grade of DCIS on the basis of imaging features. Documenting these sonographic features may help us to increase the sensitivity and specificity of sonography as an adjunctive tool in breast diagnosis, and comparing these features with the histopathologic results will help us to further evaluate the ability of this technique to determine prognosis.

Materials and Methods

From 1996 to 2001, 507 cases of breast cancer lesions were diagnosed in 490 women at a single institution with a primary diagnostic breast practice. Of these, 92 cases were DCIS and 415 were invasive carcinoma. Sixty cases of histopathologically determined DCIS found in 55 symptomatic women who underwent preoperative imaging, including mammography and sonography, were included in the study cohort. Thirty-two cases were excluded because they were asymptomatic, had no preoperative sonography performed, or both. We performed a retrospective review of these 60 DCIS cases to document the spectrum of sonographic features of DCIS lesions found in a symptomatic population and correlated these findings with those from mammographic and histopathologic evaluations. The study was performed in accordance with the recommendations of the Declaration of Helsinki [21], which provides guidelines for physicians in biomedical research for studies involving human subjects.

Clinical Features

The following clinical features were obtained from the medical histories: presence of a palpable mass, nipple discharge, Paget's disease, and mastalgia. Thirty-two women (58%) had a palpable mass, 21 (35%) had spontaneous uniorificial nipple discharge, two (3%) had Paget's disease, and two (3%) had palpable thickening. Five women had more than one clinical finding. Three women who presented with unilateral symptoms had incidental abnormalities detected in the contralateral asymptomatic breast.

Sonography

Physicians have performed bilateral whole-breast real-time sonography on all symptomatic patients at this institution. In 1998, color power Doppler sonography was included in the scanning protocol for all solid and complex cystic masses. Findings of color power Doppler sonography, including the presence or absence of intralesional vascularity, were routinely documented on hard-copy images and in reports of physicians. Lesions that did not fulfill the criteria of benignity as defined by Stavros et al. [22] were evaluated using a sonographically guided biopsy after informed consent was obtained from the patients. Sonography was performed on 60 DCIS lesions using a high-resolution unit with a 7.5- or 12.5-MHz linear array transducer with a 6-MHz Doppler operating frequency on a 700 Loqic Expert/Pro series (General Electric Medical Systems, Milwaukee, WI) or a Sonoline Elegra scanner (Siemens Medical Solutions, Erlangen, Germany). Color power Doppler sonography was performed on 32 lesions with the following optimized parameters: pulse repetition frequency of 750–1,000 Hz, low wall pass filter, medium persistence, high sensitivity, and dynamic motion differentiation. Color power Doppler gain was optimized with an increase in gain until the color box was filled with uniform low-level blue noise with minimal yellow power signal (≈ 75–85% gain) [23]. Sonograms were reviewed for masses, architectural distortion, ductal extension and dilatation, microcalcifications, and color-flow signal. We documented the following features of masses: size, nature (solid or cystic), shape, margin, echogenicity, and posterior acoustic phenomena.

Mammography

Standard two-view mammography was performed with a Senographe DMR+ unit (General Electric Medical Systems) with additional views obtained as necessary. Mammograms were reviewed for masses, calcifications, architectural distortion, tubular ductal opacity, and asymmetric density. The shape, margin, and density of masses were noted. The morphology and distribution of microcalcifications were also recorded.

Galactography

Galactography was performed successfully in eight (six with bloody and two with clear fluid) of the 21 patients who presented with nipple discharge and unsuccessfully in four (all with bloody fluid) because of difficult cannulation. The procedure was not attempted in nine (four with bloody and five with clear fluid) patients in whom no discharge was expressible at the time of examination. This relatively high success rate was likely related to the preselected study population of women with pathologic diagnoses of DCIS, resulting in a high pretest probability of cancer.

Histopathology

Histopathologic findings in excisional biopsy or mastectomy specimens were used as the gold standard. We analyzed the following histologic parameters: nuclear grade, presence and extent (by percentage area) of comedo necrosis [8], architectural pattern [4], and presence and extent of microinvasion. Lesions were classified using the Van Nuys classification system [7, 8]. Lesions with pure DCIS and DCIS with microinvasion (invasive focus of ≤ 1 mm) as defined by previously published criteria [9] were included in this study. Available long-term follow-up data on patients with microinvasive carcinoma suggest that the prognosis after surgery is excellent, with no difference in local recurrence, disease-free survival, and overall survival rates when compared with those of patients with pure DCIS [24]. In general, the therapeutic algorithm for patients with pure DCIS and DCIS with microinvasion is similar [25]. Cases of DCIS associated with minimal invasion and infiltrative ductal cancer were excluded from the study.
For all cases, sonographic and mammographic features were compared with the histopathologic findings. Statistical comparisons were performed using the Fisher's exact test for nonparametric inference with the statistical software package StatXact (Cytel Software, Cambridge, MA).

Results

Fifty-five women (age range, 27–90 years; mean, 54 years) with 60 DCIS lesions shown at histopathologic examination were included in the study group. Two of the five women with bilateral DCIS were symptomatic bilaterally. The other three women presented with unilateral breast symptoms; disease was detected sonographically and pathologically confirmed in the contralateral asymptomatic breast. Thirty-three lesions were classified as Van Nuys group 1, 19 as Van Nuys group 2, and eight as Van Nuys group 3. The histopathologic features of the 60 DCIS lesions, including nuclear grade, the percentage area of comedo necrosis, architectural pattern, and presence of microinvasion, are provided in Table 1. The mean size, as determined by histopathologic evaluation, was 2.1 cm (range, 0.4–8.0 cm). The mean lesion sizes for Van Nuys groups 1, 2, and 3 were 1.9, 2.3, and 2.6 cm, respectively. The most frequent clinical presentation was a palpable mass; the second most common presentation was nipple discharge that was typically uniorificial and blood-stained. The correlation between histopathologic results and visibility on imaging is provided in Table 2. Six (10%) of the 60 and 12 DCIS lesions (20%) were not visible using sonography and mammography, respectively. The false-negative sonographic results consisted of five cases of mammographically visible microcalcifications ranging in size from 0.5 to 2 cm and one case of a mammographically occult lesion shown as a 0.5-cm intraluminal mass at galactography. The 12 false-negative mammographic results consisted of seven solid masses (size range, 0.4–1.2 cm), one cystic mass (1 cm), and three cases of ductal dilatation and low-level intraductal echoes (two of which had positive findings at galactography). The remaining false-negative case also had negative findings on sonography but positive findings on galactography.
TABLE 1 Histopathologic Findings in 60 Cases of Ductal Carcinoma In Situ
Histopathologic FindingsVan Nuys Classification [7]
Group 1 (n = 33)Group 2 (n = 19)Group 3 (n = 8)Total (n = 60)
Nuclear grade    
    1255030
    2814022
    30088
Percentage of necrosis    
    0330033
    1-20014216
    21-500257
    51-900011
    91-1000303
Architectural pattern    
    Papillary54110
    Solid2114641
    Cribriform5016
    Cystic2103
Microinvasion (≤ 1 mm)    
    Yes 268
    No
0
17
2
52
TABLE 2 Correlation Between Histopathology and Visibility at Imaging of 60 Cases of Ductal Carcinoma In Situ
Histopathologic Findings (Van Nuys Classification [7])Visibility at Imaging
SonographyMammography
VisibleNot VisibleVisibleNot Visible
Group 1 (n = 33)30 (91)3 (9)24 (73)9 (27)
Group 2 (n = 19)16 (84)3 (16)16 (84)3 (16)
Group 3 (n = 8)
8 (100)
0
8 (100)
0
Total (n = 60)
54 (90)
6 (10)
48 (80)
12 (20)
Note.—Numbers in parentheses are percentages.

Sonography

The correlation between sonographic and histopathologic findings is provided in Table 3. Masses were the most frequent sonographic finding in 43 (72%) of the 60 cases (Fig. 1A, 1B), followed by ductal dilatation and low-level intraductal echoes (n = 14, 23%) (Fig. 2A, 2B, 2C) and architectural distortion (n = 4, 7%) (Fig. 3A, 3B, 3C). Eight lesions had more than one of these features. Microcalcifications were visible using sonography in 13 (22%) of the 60 cases; they were characteristically noted within a mass or duct (Fig. 4A, 4B) but were not visible in isolation. Color power Doppler sonography revealed a positive signal in 22 (69%) of the 32 patients in whom it was performed (Figs. 2A, 2B, 2C and 5A, 5B, 5C). Masses visible on sonography were typically solid, hypoechoic, irregularly shaped with indistinct margins and showed the presence of microcalcifications. Typically, no posterior acoustic phenomenon was present (Table 3). Positive ductal findings included a solitary nodule within a fluid-filled duct, a solid-appearing distended duct (Fig. 2A, 2B, 2C), and an isolated elongated nodule oriented in the direction of the subareolar duct. No abnormal skin, nipple retraction, or axillary lymphadenopathy was noted.
TABLE 3 Correlation Between Sonographic and Histopathologic Findings of 60 Cases of Ductal Carcinoma In Situ
Histopathologic Findings (Van Nuys Classification [7])
Imaging FindingsGroup 1 (n = 33)Group 2 (n = 19)Group 3 (n = 8)Total (n = 60)
Not visible3 (9)3 (16)0 (11)6 (10)
Mass23 (70)13 (68)7 (88)43 (72)
    Nature    
        Solid14 (61)11 (85)7 (100)32 (74)
        Cystic9 (39)2 (15)011 (26)
    Shapea    
        Ovoid11 (48)3 (23)014 (33)
        Lobular3 (13)01 (14)4 (9)
        Irregular9 (39)10 (77)6 (86)25 (58)
    Margin    
        Circumscribed0000
        Microlobulated11 (48)3 (23)014 (33)
        Indistinct12 (52)10 (77)6 (86)28 (65)
        Angular001 (14)1 (2)
    Echogenicity    
        Hypoechoic22 (96)12 (92)7 (100)41 (95)
        Isoechoic1 (4)1 (8)02 (5)
    Echogeneity    
        Heterogeneous23 (100)13 (100)7 (100)43 (100)
    Posterior phenomenaa    
        Nil11 (48)11 (85)6 (86)28 (65)
        Enhanced10 (43)1 (8)011 (26)
        Shadowing2 (9)1 (8)1 (14)4 (9)
Architectural distortion2 (6)1 (5)1 (13)4 (7)
Ductal changes8 (24)3 (16)3 (38)14 (23)
    Dilatation1 (3)1 (5)02 (3)
    Extension1 (3)1 (5)2 (25)4 (7)
    Dilatation and extension6 (18)1 (5)1 (13)8 (13)
Microcalcificationsa3 (9)7 (37)3 (38)13 (22)
    Mass2 (6)6 (32)3 (38)11 (18)
    Ductal1 (3)1 (5)02 (3)
Color flow    
    Present14 (42)5 (26)3 (38)22 (37)
    Absent8 (24)2 (11)010 (17)
    Not performed
11 (33)
12 (63)
5 (63)
28 (47)
Note.—Numbers in parentheses are percentages. All 60 lesions had more than one imaging finding.
a
Significant association between microcalcifications and Van Nuys classification (p < 0.05).
Fig. 1A. 69-year-old woman with Van Nuys group 1 [7] ductal carcinoma in situ who presented with palpable mass. Right mediolateral oblique mammogram with spot magnification shows irregular isodense mass with indistinct margins (arrow).
Fig. 1B. 69-year-old woman with Van Nuys group 1 [7] ductal carcinoma in situ who presented with palpable mass. Sonogram of same lesion as seen in A shows irregularly shaped hypoechoic solid mass with angular margins (arrowheads) and no posterior acoustic phenomena.
Fig. 2A. 73-year-old woman with Van Nuys group 1 [7] ductal carcinoma in situ who presented with spontaneous uniorificial bloody nipple discharge. Left galactogram with spot magnification shows poor filling and abrupt truncation of main lactiferous duct (arrowheads). Findings on mammography (not shown) were negative.
Fig. 2B. 73-year-old woman with Van Nuys group 1 [7] ductal carcinoma in situ who presented with spontaneous uniorificial bloody nipple discharge. Sonogram of same region as in seen in A shows solid-appearing distended duct with irregular margins (arrowheads).
Fig. 2C. 73-year-old woman with Van Nuys group 1 [7] ductal carcinoma in situ who presented with spontaneous uniorificial bloody nipple discharge. Color power Doppler sonogram of same lesion as seen in B shows multiple intralesional vessels.
Fig. 3A. 34-year-old woman with solid-type Van Nuys group 3 [7] ductal carcinoma in situ who presented with palpable mass. Left craniocaudal mammogram with spot magnification shows regional asymmetric opacity (arrow) with segmental amorphous microcalcifications (arrowheads).
Fig. 3B. 34-year-old woman with solid-type Van Nuys group 3 [7] ductal carcinoma in situ who presented with palpable mass. Sonogram shows branching ductal pattern involving terminal ductal lobular units (arrows) causing parenchymal architectural distortion in left breast.
Fig. 3C. 34-year-old woman with solid-type Van Nuys group 3 [7] ductal carcinoma in situ who presented with palpable mass. Sonogram of same region as seen in B in right breast shows normal parenchymal pattern.
Fig. 4A. 40-year-old woman with solid-type Van Nuys group 3 [7] ductal carcinoma in situ who presented with palpable mass. Left mediolateral oblique mammogram shows clustered pleomorphic microcalcifications in segmental distribution (arrowheads) with associated increased density (arrow).
Fig. 4B. 40-year-old woman with solid-type Van Nuys group 3 [7] ductal carcinoma in situ who presented with palpable mass. Sonogram shows irregularly shaped solid mass (white arrowheads) with internal microcalcifications (black arrowheads).
Fig. 5A. 73-year-old woman with solid-type Van Nuys group 1 [7] ductal carcinoma in situ who presented with palpable mass. Right craniocaudal mammogram shows ovoid mass with microlobulated margins (arrow).
Fig. 5B. 73-year-old woman with solid-type Van Nuys group 1 [7] ductal carcinoma in situ who presented with palpable mass. Sonogram shows solid ovoid mass (arrows) with microlobulated margins and posterior enhancement (arrowheads).
Fig. 5C. 73-year-old woman with solid-type Van Nuys group 1 [7] ductal carcinoma in situ who presented with palpable mass. Color power Doppler sonogram shows multiple intralesional vessels.

Mammography and Galactography

The correlation between mammographic and histopathologic findings is provided in Table 4. Microcalcifications were the dominant finding, noted in 25 (42%) of the 60 cases (Figs. 3A, 3B, 3C and 4A, 4B), followed by the presence of a mass in 24 cases (40%) (Figs. 5A, 5B, 5C and 6A, 6B). Architectural distortion or asymmetric density was noted in eight lesions (13%) (Figs. 1A, 1B and 3A, 3B, 3C). Microcalcifications were predominantly pleomorphic and linear with a segmental, clustered, or linear distribution. Masses were either irregular or round or had high or medium density and had indistinct margins. The findings of all eight galactograms available for review were positive and showed intraductal filling defects (Fig. 2A, 2B, 2C).
TABLE 4 Correlation Between Mammographic and Histopathologic Findings of 60 Cases of Ductal Carcinoma In Situ
Histopathologic Findings (Van Nuys Classification [7])
Imaging FindingsGroup 1 (n = 33)Group 2 (n = 19)Group 3 (n = 8)Total (n = 60)
Not visible9 (27)3 (16)012 (20)a
Mass16 (48)5 (26)3 (38)24 (40)
    Shape    
        Irregular5 (31)1 (20)1 (33)7 (29)
        Lobular2 (13)2 (40)1 (33)5 (21)
        Round6 (38)1 (20)07 (29)
        Ovoid3 (19)1 (20)1 (33)5 (21)
    Margin    
        Spiculated2 (13)01 (33)3 (13)
        Indistinct8 (50)3 (60)1 (33)12 (50)
        Microlobulated4 (25)1 (20)05 (21)
        Circumscribed2 (13)1 (20)1 (33)4 (17)
    Density    
        High8 (50)3 (60)2 (67)13 (54)
        Medium8 (50)2 (40)1 (33)11 (46)
Microcalcificationsb9 (27)10 (53)6 (75)25 (42)
    Morphology    
        Linear, casting1 (11)5 (42)2 (29)8 (29)
        Pleomorphic5 (56)4 (50)4 (71)13 (57)
        Amorphous3 (33)1 (8)04 (14)
    Distribution    
        Clustered4 (44)3 (42)2 (43)9 (43)
        Segmental4 (44)7 (58)4 (57)5 (54)
        Linear1 (11)001 (4)
Architectural distortion or asymmetric density
5 (15)
2 (11)
1 (13)
8 (13)
Note.—Numbers in parentheses are percentages. All 60 lesions had more than one imaging finding.
a
Four cases not visible on mammography showed a lesion on galactography.
b
Significant association between microcalcifications and Van Nuys classification (p < 0.05).
Fig. 6A. 67-year-old woman with solid-type Van Nuys group 1 [7] ductal carcinoma in situ who presented with palpable mass. Left craniocaudal mammogram shows circumscribed round high-density mass (arrowhead).
Fig. 6B. 67-year-old woman with solid-type Van Nuys group 1 [7] ductal carcinoma in situ who presented with palpable mass. Sonogram corresponding to A shows ovoid cystic mass (curved white arrowheads) with posterior enhancement (black arrowheads) and internal fluid–fluid layer (straight white arrowhead).

Histopathology Correlation

The visibility of microcalcifications on sonography and mammography was significantly associated with the Van Nuys groups (p < 0.05). Sonographic analysis of visible masses showed shape, margin, and posterior acoustic phenomena to be also significantly associated with the Van Nuys groups (p < 0.05). An irregularly shaped mass with indistinct or angular margins and no posterior acoustic phenomena was more likely to be associated with Van Nuys group 3 (p < 0.05) (Fig. 4A, 4B). In contrast, a cystic, ovoid mass with circumscribed margins and posterior enhancement was more likely to be associated with Van Nuys group 1 (p < 0.05) (Fig. 6A, 6B).

Discussion

Mammography is the most important imaging technique for the detection of DCIS, assessment of disease extent, and facilitation of radiographically guided biopsies to confirm the diagnosis of intraductal carcinoma. Despite the varied mammographic appearance of DCIS, which reflects disease heterogeneity, a cluster of microcalcifications is the most common finding on a screening mammogram [10]. Sonography has traditionally had a relatively small role in the diagnosis and evaluation of DCIS because of the unsurpassed efficacy of mammography in showing microcalcifications in most DCIS cases. Nonetheless, the emergence of newer high-resolution transducers and the increasing experience of physicians with breast sonography have resulted in the improved sensitivity and specificity of sonography, as well as in confidence in the technique [15, 16, 20, 22, 26, 27]. Although sonography can reveal microcalcifications within masses, it is unreliable in depicting and characterizing the morphology and extent of microcalcifications, particularly when they are in isolation. Therefore, sonography should not be used to replace mammography but as an adjunctive tool to increase the sensitivity of mammography in breast diagnosis.

Histopathology Categorization

The grouping of DCIS into subtypes is based on a variety of microscopic observations, including the architectural pattern of cell proliferation within the involved duct, the extent of tumor necrosis, and the degree of cytonuclear differentiation. The best mammographic–histologic correlation seems to be between the pattern of microcalcifications and the cytonuclear differentiation [28]. Most classification schemes define three histologic grades: high, intermediate, and low. High-grade DCIS exhibits large, variably sized nuclei with prominent nucleoli and clumped chromatin. Extensive (comedo-type) tumor necrosis is also usually present. Low-grade DCIS shows small uniform nuclei with inconspicuous nucleoli and a diffuse homogeneous chromatin pattern. Intermediate-grade DCIS is largely a “wastebasket” category with tumors displaying intermediate nuclear features and variable necrosis. We chose to compare imaging findings using the Van Nuys classification system because of the more reliable prognosis it provides.

Sonography

The most frequent sonographic feature of DCIS in this study was a focal mass, followed by low-level intraductal echoes, ductal extension or dilatation, and architectural distortion. The finding of microcalcifications on sonography and mammography was associated with a high Van Nuys grouping. Masses visible on sonography that were irregular in shape and had an indistinct or angular margin without associated posterior enhancement were more likely to be associated with poorer prognosis. Although color power Doppler sonography revealed the presence of color flow in 22 (69%) of the 32 lesions in which it was performed, this finding was not discriminating in terms of the Van Nuys classification.
Other authors have reported the sonographic features of DCIS as architectural distortion; an intracystic lesion; or a bulky, hypoechoic vascular mass with ductal extension and prominent microlobules [29, 30]. More recently, Moon et al. [15] analyzed the sonographic findings in 70 patients with DCIS and correlated them with mammographic and histologic findings. The most common sonographic findings in DCIS in this series were a microlobulated mass, mild hypoechogenicity, ductal extension, and no acoustic enhancement or shadowing. Although the presence of spiculated margins, marked hypoechogenicity, a thick echogenic rim, and posterior acoustic shadowing suggested the presence of invasion, it remained difficult to differentiate pure DCIS from DCIS with microinvasion or invasive carcinoma. Although pure DCIS and DCIS with microinvasion based on the analysis of the results of surgical pathology may not differ significantly in terms of management or prognosis, this distinction may be important in core biopsy specimen analysis. Axillary nodal dissection is sometimes performed in patients diagnosed with DCIS with microinvasion found at core biopsy because of possible underestimation of disease extent. This topic however remains controversial.
The findings in our study concur with those of Moon et al. [15] and confirm that when evident on sonography, DCIS appears most frequently as a solid, irregular mass with indistinct margins and without posterior acoustic enhancement or shadowing. Most cases that showed posterior acoustic enhancement in this study were cystic masses. Virtually all solid masses in this study did not show posterior acoustic enhancement or shadowing. A possible explanation may be that DCIS or intraductal cancer lacks the characteristics of invasive cancer that incite desmoplastic reaction (fibrosis, cicatrization, scarring) leading to posterior acoustic shadowing and also lacks the aggressive, rapid cellular turnover that accompanies high-grade invasive cancers leading to acoustic through-transmission. The reason that masses were so commonly found in this study may be in part because our study population consisted of symptomatic women, many of whom presented with a palpable mass. Ductal change was the next most frequent sonographic feature observed in our study. Solitary nodules within fluid-filled ducts, solid-appearing distended ducts, and isolated elongated nodules oriented in the direction of the subareolar ducts were the ductal changes most frequently encountered in this study. Similar features may be found in patients with benign fibrocystic changes, but these fibrocystic changes are usually present in both breasts and are not usually found in DCIS. Ductal extension and branching patterns are further important sonographic features that aid in differentiating benign from malignant processes. Another recent review of the sonographic features of nine cases of DCIS showed that the number of malignant features and the mean size of the lesions increased with the grade of DCIS [17]. We also found such a trend.
Microcalcification, which was noted predominantly within masses and ducts in this study, is another sonographic feature that deserves mention. The ability to visualize microcalcifications using state-of-the-art sonographic equipment has been described elsewhere in the literature [22, 31, 32]. Several recent reports have further described the ability of sonography to depict mammographically detected microcalcifications [15, 16, 18]. These authors have also reported a higher sonographic visibility of malignant versus benign microcalcifications because most malignant calcifications occur in the mass [15, 16, 18]. Moon et al. [20] reported that with sonography, 77% of DCIS cases were visible as a breast mass associated with microcalcifications. They reported that calcifications associated with malignant tumors were more likely to be seen on sonography. However, the sonographic findings in DCIS with calcifications were nonspecific, in that they were also seen in benign lesions such as sclerosing adenosis, atypical ductal hyperplasia, and radial scars.
Other researches have stressed the importance of performing sonography with good-quality equipment, high-frequency transducers in the 10-13–MHz range and an optimal technique, and preferably by a physician [15, 20, 22]. They suggest that sonography, when optimally performed, can complement mammography in detecting and evaluating DCIS, particularly in detecting DCIS without calcifications and in evaluating disease extent in women with dense breasts. The sonographic findings of DCIS without calcifications, which may consist of single or multiple cystic or solid hypoechoic masses with no pseudocapsule, are nonspecific. These features are also seen in benign conditions such as papillomas, duct ectasia, fibrocystic change, and atypical ductal hyperplasia. This study, however, only addresses the sensitivity of sonography in depicting DCIS, without evaluating specificity, because the study cohort consists of DCIS cases only.

Mammography

Mammographic abnormalities were noted in 80% of the cases in this study. These comprised microcalcifications (42%), dominant masses (40%), and architectural distortion or asymmetric density (13%). These proportions differ from previous descriptions of predominantly screening-detected DCIS in which mammographic abnormalities were present in 95% of cases and included microcalcifications (76%); asymmetric density (10%); dominant masses (8%); and, in patients with nipple discharge, abnormal findings on galactograms (6%) [10, 11]. These findings suggest that there are differences in the mammographic features of screening-detected versus symptomatic DCIS. The lower percentage of mammographically visible microcalcifications and the higher percentage of dominant masses in this study correspond to the lower percentage of comedo or high-grade (Van Nuys group 3) lesions in this symptomatic cohort of DCIS. Mammography of low-grade DCIS without comedo necrosis (Van Nuys group 1) has been reported to be less likely to show microcalcifications and more likely to either be mammographically normal or show noncalcified mammographic abnormalities [33]. Bellamy et al. [4] also noted an increased proportion of the comedo carcinoma subtype in mammographically detected lesions compared with symptomatic lesions. Researchers have hypothesized that the larger percentage of comedo carcinoma found in the screening group may reflect the greater invasive potential of these lesions, with relatively few patients experiencing symptoms before the progression to invasive disease [3, 11]. Screening-detected DCIS is reportedly less extensive, has a smaller average calcification cluster size, is less diffuse, and has less frequent retroareolar involvement [3]. About 10% of all DCIS lesions manifest as soft-tissue masses or asymmetric densities on mammography [13], and up to 16% of DCIS cases are mammographically occult.
When the ability of sonography and mammography to visualize DCIS was compared in this study, sonography proved superior in visualizing more cases. These findings reflect, in part, the distribution in a symptomatic population and also serve to reemphasize the increasing role of high-resolution sonography as an adjunctive tool in the evaluation of symptomatic breast disease, including symptomatic DCIS.
The primary limitation of this study is that only one radiologist and one pathologist reviewed the findings of radiology and pathology, respectively. Therefore, intra- and interobserver variability was not evaluated, and the potential lack of reproducibility both in the performance and interpretation of the examinations was not addressed. The second limitation is that this population was symptomatic so that only sensitivity was evaluated and specificity was not addressed. We believe that the relatively small sample size of symptomatic women in this study reflects true clinical practice, in which most DCIS cases are diagnosed on mammography in asymptomatic women and symptomatic DCIS is uncommon. It is nonetheless incumbent on sonographers to learn about this disease because although it is usually detected on mammography, it can also be detected on sonography.
The findings in this small study suggest that sonography may have a role in cases of mammographically detected abnormalities in symptomatic women and possibly in every symptomatic patient. This possibility is exemplified in the three women who presented with unilateral symptoms but had bilateral DCIS diagnosed with the aid of sonography.
The optimum role of sonography in this group of patients may be to evaluate palpable abnormalities when findings of mammography are negative or nonspecific, patients with nipple discharge in whom galactography is not possible, and nonspecific mammographic soft-tissue densities without calcification. The role of bilateral whole-breast sonography as a screening tool has also recently been addressed in the literature [34, 35]. Additionally, galactography has been shown in this study to be a useful adjunctive diagnostic tool in the subset group of patients with DCIS who present with spontaneous uniorificial nipple discharge, particularly in those women who present with negative findings on mammography and sonography in whom galactography remains the imaging technique of choice.

Acknowledgments

We thank Peggo K. W. Lam, Center for Clinical Trials and Epidemiological Research, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, NT, Hong Kong, for statistical assistance.

Footnote

Address correspondence to W. T. Yang ([email protected]).

References

1.
Tabar L, Vitak B, Chen HH, et al. The Swedish two-county trial twenty years later: updated mortality results and new insights from long-term follow-up. Radiol Clin North Am 2000; 38:625 –651
2.
Feig SA. Ductal carcinoma in situ: implications for screening mammography. Radiol Clin North Am 2000; 38:653 –668
3.
Evans AJ, Pinder S, Ellis IO, et al. Screening-detected and symptomatic ductal carcinoma in situ: mammographic features with pathologic correlation. Radiology 1994; 191:237 –240
4.
Bellamy CO, McDonald C, Salter DM, Chetty U, Anderson TJ. Noninvasive ductal carcinoma of the breast: the relevance of histologic categorization. Hum Pathol 1993; 24:16 –23
5.
Lagios MD, Margolin FR, Westdahl PR, Rose MR. Mammographically detected ductal carcinoma in situ. Cancer 1989; 63:618 –624
6.
Solin LJ, Yeh IT, Kurtz J, et al. Ductal carcinoma in situ (intraductal carcinoma) of the breast treated with breast conserving surgery and definitive irradiation. Cancer 1993; 71:2532 –2542
7.
Silverstein MJ, Poller DN, Waisman JR, et al. Prognostic classification of breast ductal carcinoma-in-situ. Lancet 1995; 345:1154 –1157
8.
Silverstein MJ, Lagios MD, Craig PH, et al. A prognostic index for ductal carcinoma in situ of the breast. Cancer 1996; 77:2267 –2274
9.
Lagios MD. Microinvasion in ductal carcinoma in situ. In: Silverstein MJ, ed. Ductal carcinoma in situ of the breast. Baltimore, MD: Williams & Wilkins, 1997:241 –246
10.
Tabar L, Gad A, Parsons WC, Neeland DB. Mammographic appearances of in situ carcinomas. In: Silverstein MJ, ed. Ductal carcinoma in situ of the breast. Baltimore, MD: Williams & Wilkins,1997 : 95–117
11.
Stomper PC, Connolly JL, Meyer JE, Harris JR. Clinically occult ductal carcinoma in situ detected with mammography: analysis of 100 cases with radiologic-pathologic correlation. Radiology 1989; 72:235 –241
12.
Evans A, Pinder S, Wilson R, et al. Ductal carcinoma in situ of the breast: correlation between mammographic and pathologic findings. AJR 1994; 162:1307 –1311
13.
Ikeda DM, Andersson I. Ductal carcinoma in situ: atypical mammographic appearances. Radiology 1989; 172:661 –666
14.
Mitnick JS, Roses DF, Harris MN, Feiner HD. Circumscribed intraductal carcinoma of the breast. Radiology 1989; 170:423 –425
15.
Moon WK, Myung JS, Lee YJ, Park IA, Noh DY, Im JG. US of ductal carcinoma in situ. RadioGraphics 2002; 22:269 –280
16.
Hashimoto BE, Kramer DJ, Picozzi VJ. High detection rate of breast ductal carcinoma in situ calcifications on mammographically directed high-resolution sonography. J Ultrasound Med 2001; 20:501 –508
17.
Schoonjans JM, Brem RF. Sonographic appearance of ductal carcinoma in situ diagnosed with ultrasonographically guided large core needle biopsy. J Ultrasound Med 2000; 19:449 –457
18.
Ranieri E, D'Andrea MR, D'Alessio A, et al. Ultrasound in the detection of breast cancer associated with isolated clustered microcalcifications, mammographically identified. Anticancer Res 1997; 17:2831 –2836
19.
Gufler H, Buitrago-Teliez CH, Madjar H, Allmann K-H, Uhl M, Rohr-Reyes A. Ultrasound demonstration of mammographically detected microcalcifications. Acta Radiol 2000; 41:217 –221
20.
Moon WK, Im JG, Koh YH, Noh DY, Park IA. US of mammographically detected clustered microcalcifications. Radiology 2000; 217:849 –854
21.
World Medical Association. Declaration of Helsinki: ethical principles for medical research involving human subjects, revised. Edinburgh, Scotland: World Medical Association,2000
22.
Stavros AT, Thickman D, Rapp CL, Dennis MA, Parker SH, Sisney GA. Solid breast nodules: use of sonography to distinguish between benign and malignant lesions. Radiology 1995; 196:123 –134
23.
Rubin JM, Bude RO, Carson PL, Bree RL, Adler RS. Power Doppler US: a potentially useful alternative to mean frequency-based color Doppler US. Radiology 1994; 190:853 –856
24.
Silverstein MJ. Microinvasion in ductal carcinoma in situ. In: Silverstein MJ, ed. Ductal carcinoma in situ of the breast. Baltimore, MD: Williams & Wilkins, 1997:557 –562
25.
Morrow M, Schnitt SJ, Harris JR. Ductal carcinoma in situ and microinvasive carcinoma. In: Harris JR, ed. Diseases of the breast. Philadelphia, PA: Lippincott, Williams & Wilkins,2000 : 383–401
26.
Zonderland HM, Coerkamp EG, Hermans J, van de Vijver MJ, van Voorthuisen AE. Diagnosis of breast cancer: contribution of US as an adjunct to mammography. Radiology 1999; 213:413 –422
27.
Rahbar G, Sie AC, Hansen GC, et al. Benign versus malignant solid breast masses: US differentiation. Radiology 1999; 213:889 –894
28.
Holland R, Peterse JL, Millis RR, et al. Ductal carcinoma in situ: a proposal for a new classification. Semin Diagn Pathol 1994; 11:167 –180
29.
Tohno E, Cosgrove DO, Sloane JP, eds. Ultrasound diagnosis of breast diseases. London, England: Churchill Livingstone, 1994:178 –179
30.
Stavros AT. Ultrasound of ductal carcinoma in situ. In: Silverstein MJ, ed. Ductal carcinoma in situ of the breast. Baltimore, MD: Williams & Wilkins, 1997:135 –158
31.
Kasumi F. Can microcalcifications located within breast carcinomas be detected by ultrasound imaging? Ultrasound Med Biol 1988; 14[suppl]:175 –182
32.
Yang WT, Suen M, Ahuja A, Metreweli C. In vivo demonstration of microcalcification in breast cancer using high resolution ultrasound. Br J Radiol 1997; 70:685 –690
33.
Wright B, Shumak R. Medical imaging of ductal carcinoma in situ. Part II. Curr Probl Cancer 2000; 24:112 –124
34.
Kolb TM, Lichy J, Newhouse JH. Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: an analysis of 27,825 patient evaluations. Radiology 2002; 225:165 –175
35.
Kaplan SS. Clinical utility of bilateral whole-breast US in the evaluation of women with dense breast tissue. Radiology 2001; 221:641 –649

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: 101 - 110
PubMed: 14684521

History

Submitted: February 14, 2003
Accepted: July 16, 2003

Authors

Affiliations

Wei Tse Yang
Department of Diagnostic Radiology, Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, NT, Hong Kong SAR, China.
Present address: Department of Diagnostic Radiology, University of Texas M. D. Anderson Cancer Center, Unit 57, 1515 Holcombe Blvd., Houston, TX 77030.
Gary M. K. Tse
Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, Shatin, NT, Hong Kong SAR, China.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

View Options

View options

PDF

View PDF

PDF Download

Download PDF

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media