© 2016 Korean Breast Cancer Society. All rights reserved. http://www.jbd.or.kr | eISSN 2288-5560
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
JBDJournal of Breast Disease
Preoperative Breast Magnetic Resonance Imaging for the Assessment of the Size of Ductal Carcinoma In SituMusaed Rayzah, Jai Min Ryu, Jeong Eon Lee, Mansour Alramadhan, Bookyung Han1, Ha Woo Yi, Sungmin Park, Hyun-June Paik, Seok Jin NamDepartments of Surgery and 1Radiology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
Purpose: The purpose of this study was to determine whether magnetic resonance imaging (MRI) could assess the size of ductal carcinoma in situ (DCIS) more accurately compared to mammography and ultrasonography using the histopathological dimension of the surgical speci-men as the reference measurement. Methods: This was a retrospective review study using data from our institution database of breast can-cer. Preoperative contrast-enhanced MRI, mammography and ultrasonography were performed to detect and assess the size of DCIS in 131 patients. The greatest dimensions of DCIS determined by the imaging modalities were compared with the histopathological dimensions of the surgical specimens. Intraclass coefficients were calculated to examine the agreement among the MRI, mammography and ultrasonogra-phy measurements. The Wilcoxon signed-rank test was used to evaluate the statistical significance of the differences in size among MRI, mammography or ultrasonography and histopathology findings. Results: Of the 131 DCIS lesions, 126 (96.2%) were detected by MRI, 103 (78.6%) were detected by mammography, and 121 (92.4%) were detected by ultrasonography. The mean lesion size was 38.8 mm on histo-pathology, 36.0 mm on MRI, 28.8 mm on mammography, and 23.3 mm on ultrasonography, and there were no significant differences be-tween sizes determined by histopathology and MRI, while there were significant differences between histopathology and the other modali-ties. The correlation coefficient between histopathological measurement and MRI was 0.837, versus 0.461 between histopathology and mammography and 0.284 between histopathology and ultrasonography. The lesion size was correctly estimated (±5 mm), under-estimated (<5 mm), or over-estimated (>5 mm), respectively, by MRI in 52.7%, 30.5%, and 16.8% of cases; by mammography in 32.0%, 51.2%, and 16.8% of cases, respectively; and by ultrasonography in 24.4%, 62.6%, and 13.0% of cases, respectively. Conclusion: In our study, MRI was more accurate for detection and assessment the size of DCIS compared to mammography and ultrasonography.
Key Words: Breast neoplasms, Ductal carcinoma in situ, Magnetic resonance imaging
INTRODUCTION
Ductal carcinoma in situ (DCIS) is a noninvasive form of ductal car-
cinoma, limited to the confines of the basement membrane of the
duct. DCIS is a multiform disease with different growth patterns and a
heterogeneous set of clinical signs and symptoms [1]. Before the era of
screening mammography, DCIS was a relatively uncommon presen-
tation of breast cancer. However, since the introduction of mammog-
raphy screening programs for breast cancer, the incidence of DCIS has
risen consistently. DCIS currently accounts for 15% to 25% of all breast
cancers and 30% to 50% of all clinically occult cancers [2,3]. Approxi-
mately 40% of cases of DCIS evolve into invasive cancer [4].
Accurate information regarding the extent and distribution of
DCIS is important in determining the extent of surgery required.
Mammography is the primary tool of radiologists for the detection
and assessment of size of microcalcifications. However, mammogra-
phy is relatively limited in its ability to detect DCIS and assess tumor
size because it only detects the calcified portion of DCIS, and it can
underestimate or overestimate the histopathological tumor size. Con-
sequently, surgical resection is too often insufficient, resulting in a 30%
rate of re-excision because of persistent margin involvement [5], as well
as a higher risk of local recurrence [6,7].
Studies have shown that magnetic resonance imaging (MRI) is
more accurate than mammography for detecting and assessing tumor
size in patients with invasive cancer [8]. The sensitivity of MRI has
been reported to vary between 90% and 94% in larger studies and me-
ta-analyses [9,10]. However, few studies have investigated the value of
MRI for assessing the extent of DCIS; the initial results are encourag-
ORIGINAL ARTICLEJ Breast Dis 2016 December; 4(2): 77-84https://doi.org/10.14449/jbd.2016.4.2.77
Correspondence: Seok Jin NamDepartment of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 06351, KoreaTel: +82-2-3410-3478, Fax: +82-2-3410-6982, E-mail: [email protected]: Mar 16, 2016 Revised: Apr 8, 2016 Accepted: Oct 24, 2016
http://www.jbd.or.kr https://doi.org/10.14449/jbd.2016.4.2.77
78 Musaed Rayzah, et al.
ing and demonstrate a fairly good correlation between size deter-
mined with MRI and histopathological size [11-13]. However, MRI of
the breast in patients with DCIS is not routinely performed and re-
mains experimental. Few studies have examined the effects of breast
MRI on surgical management specifically for DCIS [14-16].
The aim of this study was to evaluate the value of MRI for the stag-
ing of DCIS, and particularly for the accurate assessment of tumor
size in comparison with mammography and ultrasonography, using
the histopathological measurements as a reference.
METHODS
Patient selection
This was a single-center retrospective review from a Samsung
Medical Center database of breast cancer. All patients diagnosed with
pure DCIS, proven by core needle biopsy or vacuum-assisted biopsy,
between January 2013 and December 2014, were included. Patients
were excluded if they did not undergo MRI. Patients with multifocal
or multicentric lesions were also excluded because of the difficulty in
estimating the exact tumor size; thus, 131 patients were eligible for the
study. Data obtained for each patient included age; sex; and results of
the clinical breast examination, mammography, ultrasonography,
MRI, and histopathological examination of the biopsy and surgical
specimens. Clinical examination included the patient’s history, hor-
mone receptor status, and physical breast examination.
Imaging techniques
Mammography was performed using the Senographe 2000D sys-
tem (General Electric Medical Systems, Buc, France). Spot compres-
sion and magnification views were also obtained in patients with mi-
crocalcifications visible on mammography. Assessment of mammo-
grams considered breast density, and abnormalities were scored using
the Breast Imaging-Reporting and Data System (BI-RADS) mam-
mography classification [17].
Ultrasonography was performed systematically using the HDI
5000 (Advanced Technology Laboratories, Bothell, USA), iU22
(Philips Medical Systems, Bothell, USA) or LOGIQ700 (GE Medical
Systems, Milwaukee, USA), ultrasonography scanner, respectively
equipped with commercially available 12- to 7-MHz, and 10- to
7-MHz, linear-array transducers. The examinations were interpreted
using the BI-RADS ultrasonography classification system [18].
MRI was performed using a 1.5-T (Sigma; GE Healthcare, Milwau-
kee, USA) or 3.0-T (Philips Achieva; Philips Healthcare, Best, The
Netherlands) MRI system. Three-dimensional, fat-suppressed, gradi-
ent-echo, contrast material-enhanced, and dynamic images before
and seven times after a bolus injection of 0.1 mmol/kg of gadopen-
tetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, USA)
were acquired in the sagittal or axial plane. The image acquisition
time per one dynamic scan was 60 to 90 seconds. Standard subtrac-
tion images were obtained by subtracting the precontrast images from
the early peak (or serial) postcontrast images on a pixel-by-pixel basis.
Reverse subtraction images were obtained by subtracting the last
postcontrast images from the early peak postcontrast images. Abnor-
malities were scored according to the BI-RADS MRI classification
system [19].
The findings obtained on all the imaging studies were analyzed,
and the size of the DCIS was measured as the single greatest measure-
ment by an experienced radiologist.
Biopsy, surgery, and histopathology
The diagnosis of DCIS was obtained via core needle biopsies or
vacuum-assisted biopsies, prior to the preoperative MRI and mam-
mography. All patients underwent surgery at our institution. Follow-
ing lumpectomy or mastectomy, gross specimens were reviewed by a
pathologist with expertise in breast pathology. The DCIS grade (high,
intermediate or low), presence or absence of microinvasive clusters or
an associated invasive tumor component, and nodal status were re-
corded. The histopathological size was recorded as the single greatest
measurement in one long axis dimension, regardless of the spatial
plane used.
Statistical analysis
Quantitative data are presented in a histogram, and as medians
with range and means and standard deviation (SD). The criterion
used for the accuracy of size assessment was a size within ± 5 mm
compared with the histological size. Two types of statistical analyses
were performed for correlations. Intraclass coefficients were calculat-
ed to examine the agreement between the MRI, mammography and
ultrasonography measurements and the histopathological size, with a
95% confidence interval. The Wilcoxon signed-rank test was used to
https://doi.org/10.14449/jbd.2016.4.2.77 http://www.jbd.or.kr
Preoperative Breast MRI for DCIS Assessment 79
evaluate the statistical significance of the differences in size between
MRI, mammography or ultrasonography and histopathology. A
p-value less than 0.05 was considered to indicate statistical signifi-
cance. All statistical analysis were performed using SAS version 9.4
(SAS Institute, Cary, USA). The requirement for informed consent
was waived because of the low risk posed by this investigation. This
study adhered to the ethical tenets of the Declaration of Helsinki and
was approved by the Institutional Review Board of Samsung Medical
Center in Seoul, Korea (IRB number: 2016-04-006).
RESULTS
The mean age ± SD was 48.6 ± 9.8 years overall (range, 35–78 years)
(Table 1); estrogen receptor expression was observed in 97 patients
(74.1%), while progesterone receptor expression was observed in 90
patients (68.7%). Of the 131 DCIS lesions, 126 (96.2%) were detected by
MRI, 103 (78.6%) were detected by mammography, and 121 (92.4%)
were detected by ultrasonography (Table 2).
The final histopathological study found 34 low-grade (26.0%), 59
intermediate-grade (45.0%), and 38 high-grade DCIS lesions (29.0%).
The mean histopathological size of the DCIS lesions was 38.8 mm (±
SD, 26.2 mm). The mean size of lesions determined by mammogra-
phy was 28.8 mm (± SD, 25.7 mm) (Table 1). The difference in the
mean size determined by mammography and histopathology was
-10.0 mm (± SD, 25.6 mm), a difference that was statistically signifi-
cant (p < 0.001) (Table 3). The graph showed wide variation between
sizes determined by mammography and histopathology (Figure 1).
The correlation coefficient between the histopathological size and
mammographic size was 0.461. Mammography correctly assessed the
lesion size (± 5 mm) in 33 cases (32.0%), underestimated the size in 53
cases (51.2%), and overestimated the size in 17 cases (16.8%) (Table 2).
Among our patient population, 25 patients (19.0%) had dense breasts
in which the lesions could not be visualized, while three patients
(2.3%) did not show lesions.
The mean ultrasonographic size of DCIS was 23.3 mm (± SD, 18.0
mm). The difference in mean size between ultrasonography and histo-
pathology was -15.6 mm (± SD, 23.9 mm), a difference that was statis-
tically significant (p < 0.001) (Table 3). The graph depicted a variation
between the ultrasonographic and histopathological sizes (Figure 1).
The correlation coefficient between the histopathological and ultraso-
nographic sizes was 0.284. Ultrasonography correctly assessed the le-
sion size (± 5 mm) in 29 cases (24.4%), underestimated the size in 76
cases (62.6%), and overestimated the size in 16 cases (13.0%) (Table 2).
The mean MRI size of the DCIS lesions was 36 mm (± SD, 24.3
mm). The difference in the mean size between MRI and histopathol-
ogy was -2.9 mm (± SD, 14.2 mm), a difference that was statistically
significant (p = 0.011) (Table 3). A close fit was graphically depicted
between the MRI and histopathological sizes (Figure 1). The size dis-
crepancy and span were smaller for the MRI measurements than for
the mammographic and ultrasonographic differences. The correla-
tion coefficient between the histopathological and MRI sizes was
Table 1. Mean age and mean size of DCIS as determined by ultrasonogra-phy, mammography, MRI, and histopathology
No. Mean SD
Age (yr) 131 48.6 9.8Pathology size (mm) 131 38.8 26.2Ultrasonography size (mm) 121 23.3 18.0Mammography size (mm) 103 28.8 25.7MRI size (mm) 126 36.0 24.3
DCIS = ductal carcinoma in situ; MRI = magnetic resonance imaging.
Table 2. Size assessment correlation coefficients between ultrasonographic, MRI imaging sizes and histological size
Ultrasonography Mammography MRI
No. of detection 121 103 126Sensitivity (%) 92.4 78.6 96.2Accurate estimation,
no. (%) 29 (24.4) 33 (32.0) 67 (52.7)
Under estimation ( < 5 mm), no. (%)
76 (62.6) 53 (51.2) 38 (30.5)
Over estimation ( > 5 mm), no. (%)
16 (13.0) 17 (16.8) 21 (16.8)
Correlation coefficient, all 0.284 0.461 0.837Nuclear grade Low 0.367 0.434 0.872 Intermediate 0.236 0.391 0.815 High 0.234 0.527 0.829
MRI = magnetic resonance imaging.
Table 3. Differences between ultrasonography, mammography or MRI as-sessments and histopathological assessment
Pathology difference No. Mean (mm) SD (mm) p-value
Ultrasonography 121 -15.6 23.9 < 0.001Mammography 103 -10.0 25.6 < 0.001MRI 126 -2.9 14.2 0.011
MRI = magnetic resonance imaging.
http://www.jbd.or.kr https://doi.org/10.14449/jbd.2016.4.2.77
80 Musaed Rayzah, et al.
0.837. MRI correctly assessed the lesion size (± 5 mm) in 67 cases
(52.7%), underestimated the size in 38 cases (30.5%), and overestimat-
ed the size in 21 cases (16.8%) (Table 2). In patients with dense breast
tissue, MRI showed the lesion in 24 out of 25 patients.
According to histological grade, in low-grade DCIS, the mean size
determined by mammography was 21.4 mm (Table 4); the difference
between the mean mammographic size and the mean histopathologi-
cal size was -6.1 mm (± SD, 24.4 mm), which was not statistically sig-
nificant (p = 0. 067) (Table 5). On ultrasonography, the mean size was
19.5 mm; the difference between the mean ultrasonographic size and
the mean histopathological size was -8 mm (± SD, 18.1 mm), which
was not statistically significant (p = 0.067). On MRI, the mean size
was 27.3 mm; the difference in the mean MRI and histopathological
sizes was -0.2 mm (± SD, 10.3 mm), which was not statistically signifi-
cant (p = 0.758) (Table 5).
For intermediate grade DCIS, the mean size determined by mam-
mography was 28.1 mm (Table 4). The difference between the mean
mammographic size and the mean histopathological size was -13.5
12.5
10.0
7.5
5.0
2.5
0
12.5
10.0
7.5
5.0
2.5
0
12.5
10.0
7.5
5.0
2.5
0
12.5
10.0
7.5
5.0
2.5
0
0 2.5 5.0 7.5 10.0 12.5
0 2.5 5.0 7.5 10.0 12.5
0 2.5 5.0 7.5 10.0 12.5
0 2.5 5.0 7.5 10.0 12.5
Lesio
n siz
eLe
sion
size
Lesio
n siz
eLe
sion
size
Pathology tumor size
Ultra Mammogram MRI
Ultra
Mammogram
MRI
Pathology tumor size
Pathology tumor size
Pathology tumor size
Figure 1. Correlation between the histopathological ductal carcinoma in situ (DCIS) size and corresponding DCIS size as measured by MRI, mammography and ultrasonography. Ultra=ultrasonography; MRI=magnetic resonance imaging.
Table 4. Mean size of DCIS by ultrasonography, mammography, and MRI and histopathological size according to histological grade
No.MRI size* (mm) Mammography size† (mm) Ultrasonography size‡ (mm) Pathology size (mm)
Mean SD Mean SD Mean SD Mean SD
Low 34 27.3 20.7 21.4 26.3 19.5 13.8 27.5 19.6Intermediate 59 36.4 25.0 28.1 25.3 20.9 17.8 41.6 28.5High 38 43.0 24.5 36.4 24.5 30.4 20.0 44.6 25.2
DCIS = ductal carcinoma in situ; MRI = magnetic resonance imaging.*5 Patients were excluded because they were not detected by MRI; †28 Patients were excluded because they were not detected by mammography; ‡10 Patients were excluded because they were not detected by ultrasonography.
https://doi.org/10.14449/jbd.2016.4.2.77 http://www.jbd.or.kr
Preoperative Breast MRI for DCIS Assessment 81
mm (± SD, 27.7 mm), which was statistically significant (p < 0.001)
(Table 5). On ultrasonography, the mean size was 20.9 mm (Table 4).
The difference between the mean ultrasonographic and histopatho-
logical sizes was -20.8 mm (± SD, 24.5 mm), which was statistically
significant (p < 0.001) (Table 5). On MRI, the mean size was 36.4 mm
(Table 4). The difference between the mean MRI size and the mean
histopathological size was -5.24 mm (± SD, 15.6 mm), which was sta-
tistically significant (p = 0.007) (Table 5).
In high-grade DCIS, the mean size determined by mammography
was 36.4 mm (Table 4). The difference between the mean mam-
mographic and histopathological sizes was -8.2 (± SD, 23.3 mm),
which was statistically significant (p = 0.039) (Table 5). On ultrasonog-
Table 5. Differences between ultrasonography, mammography or MRI assessments and histopathological assessment according to histological grade
No.
MRI pathology difference* (mm)
Mammography pathology difference† (mm)
Ultrasonography pathology difference‡ (mm)
Mean SD p-value Mean SD p-value Mean SD p-value
Low 34 -0.2 10.3 0.758 -6.1 24.4 0.670 -8.0 18.1 0.067Intermediate 59 -5.2 15.6 0.007 -13.5 27.7 0.000 -20.8 24.5 0.000High 38 -1.6 14.6 0.392 -8.2 23.3 0.039 -14.3 26.0 0.001
MRI = magnetic resonance imaging.*5 Patients were excluded because they were not detected by MRI; †28 Patients were excluded because they were not detected by mammography; ‡10 Patients were excluded because they were not detected by ultrasonography.
Figure 2. A 48-year-old patient with an 18-mm ductal carcinoma in situ comedo type in her left breast that was more accurately assessed by magnetic reso-nance imaging (MRI). (A) Mediolateral mammography views showed a 13-mm calcification (arrow). (B) MRI showed a 19-mm nonmass enhancement (arrow). (C) Ultrasonography showed a 15-mm microcalcification.
B
CA
http://www.jbd.or.kr https://doi.org/10.14449/jbd.2016.4.2.77
82 Musaed Rayzah, et al.
raphy, the mean size was 30.4 mm (Table 4). The difference between
the mean ultrasonographic and histopathological sizes was -14.3 mm
(± SD, 26.0 mm), which was statistically significant (p = 0.001) (Table
5). On MRI, the mean size was 43 mm (Table 4). The difference be-
tween MRI and histopathological sizes was -1.6 (± SD, 14.6 mm),
which was not statistically significant (p = 0.392). The correlation co-
efficient between the histopathological and MRI sizes was greater
than 0.8 for all grades, but this was not the case for the sizes deter-
mined by mammography or ultrasonography (Table 2).
DISCUSSION
The current study investigated the impact of MRI in preoperative
assessment of the size of DCIS compared with that of mammography
and ultrasonography. The findings showed that MRI assessed the ex-
tent of DCIS much more accurately than mammography and ultraso-
nography. Mammography is widely accepted as the most important
imaging method for the detection of DCIS, which is usually visualized
as clusters of microcalcifications [20]. However, its sensitivity was 78%
in the current study. In a prospective study of 167 cases of DCIS, Kuhl
et al. [21] reported a diagnostic sensitivity of only 56% with mammog-
raphy. In addition, the modality has some limitations in dense breasts,
which were found in 19% of participant in our study.
Ultrasonography is an adjunct modality used routinely in patients
diagnosed with DCIS. Previous studies [22,23] have reported the ul-
trasonographic features of DCIS as an architectural distortion, an in-
tracystic lesion or a bulky, hypoechoic vascular mass with ductal ex-
tension and prominent microlobules. In our study, DCIS was ob-
served as a nonmass abnormality, a round or oval-shaped, microlobu-
lated, mildly hypoechoic mass or a microcalcification (Figure 2), a re-
sult similar to those obtained in previous studies [22,23]. In addition,
microcalcification was also observed. The ability to visualize micro-
calcifications on ultrasonography has been described previously
[24,25]. In our study, ultrasonography detected DCIS lesions in 121
patients with a sensitivity of 92.4%. Lee et al. [25] reported the sensitiv-
ity of ultrasonography in DCIS as 86.5%.
MRI is not currently used as a systematic diagnostic modality in
DCIS, and controversy remains regarding its use for preoperative
planning. Previous studies have found that suspicious MRI findings
for DCIS include nonmass enhancement, especially in a ductal or seg-
mental/linear pattern, and variable perfusion patterns, including de-
layed washout, plateau, and persistent kinetics [26]. Although in our
study MRI showed enhancement in 126 patients, it showed nonmass
enhancement with an excellent sensitivity of 96% for DCIS detection.
Previous studies have reported sensitivities ranging from 60% to 100%
[12,13,27,28]. MRI is a useful adjunct modality for dense breast tissue;
in our study, it revealed the lesions in 96% of patients with dense
breasts and negative findings on mammography.
Regarding the preoperative size assessment of DCIS, mam-
mographic and ultrasonographic assessments of the lesion size were
mediocre in our study because the correlation coefficients between
these measurements and the size determined by histology were only
0.461 and 0.284, respectively. In our study, mammography tended to
underestimate the size of DCIS. This finding did not differ signifi-
cantly from those reported in the literature [29]. Furthermore, in our
study, ultrasonography tended to underestimate the size of lesions by
10% compared to mammography.
Can MRI palliate the limitations of mammography and ultraso-
nography to serve as an effective tool for the staging of DCIS? The
correlation coefficient between size determined by MRI and size de-
termined by histology in our study was very satisfactory (0.837), with
excellent reliability. MRI accurately estimated the DCIS size in 52.7%
of cases. Other studies [11-13,30], which included 30, 22, 72, and 33
patients, respectively, all reported accurate MRI estimation of size in
57% to 72% of cases. Only a few studies [11,30] considered an accurate
size assessment as one that was within ± 5 mm. In the study of
Schouten van der Velden et al. [30], size assessment with MRI was ac-
curate in 38% of cases versus 27% with mammography, with no statis-
tical significance; however, in the study of Marcotte-Bloch et al. [11],
size assessment had an accurate rate of 60% with MRI versus 38% with
mammography (this difference was statistically significant).
The results obtained in our study showed statistical significance in
favor of better size assessment for all histological grades of DCIS by
MRI compared with size determinations by mammography and ul-
trasonography. However, our study had limitations. First, it employed
a retrospective design, with data from a single institution. Second,
multicentric or multifocal cancer was excluded, possibly limiting the
reliability of the results. Third, breast images were performed after
core needle or vacuum-assisted biopsies, generating a bias in imaging
and pathological measurements. Fourth, all lesions analyzed were
https://doi.org/10.14449/jbd.2016.4.2.77 http://www.jbd.or.kr
Preoperative Breast MRI for DCIS Assessment 83
pure DCIS in the final histopathology; lesions with associated micro-
invasive or invasive clusters were also included. Despite these limita-
tions, our findings are significant and warrant further investigations.
In conclusion, our study confirmed that preoperative MRI staging
of DCIS assessed the extent of these tumors much more accurately
than mammography and ultrasonography. Precise preoperative as-
sessment of tumor size should improve surgical planning and reduce
the risk of secondary recurrence.
CONFLICT OF INTEREST
The authors declare that they have no competing interests.
REFERENCES
1. Dershaw DD, Abramson A, Kinne DW. Ductal carcinoma in situ:
mammographic findings and clinical implications. Radiology 1989;
170:411-5.
2. Ernster VL, Barclay J, Kerlikowske K, Grady D, Henderson C. Inci-
dence of and treatment for ductal carcinoma in situ of the breast.
JAMA 1996;275:913-8.
3. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Can-
cer statistics, 2003. CA Cancer J Clin 2003;53:5-26.
4. Recht A, Rutgers EJ, Fentiman IS, Kurtz JM, Mansel RE, Sloane JP.
The fourth EORTC DCIS Consensus meeting (Château Marquette,
Heemskerk, The Netherlands, 23-24 January 1998): conference re-
port. Eur J Cancer 1998;34:1664-9.
5. Cellini C, Huston TL, Martins D, Christos P, Carson J, Kemper S, et
al. Multiple re-excisions versus mastectomy in patients with per-
sistent residual disease following breast conservation surgery. Am J
Surg 2005;189:662-6.
6. Sakorafas GH, Tsiotou AG. Ductal carcinoma in situ (DCIS) of the
breast: evolving perspectives. Cancer Treat Rev 2000;26:103-25.
7. Boyages J, Delaney G, Taylor R. Predictors of local recurrence after
treatment of ductal carcinoma in situ: a meta-analysis. Cancer 1999;
85:616-28.
8. Chung A, Saouaf R, Scharre K, Phillips E. The impact of MRI on
the treatment of DCIS. Am Surg 2005;71:705-10.
9. Berg WA, Gutierrez L, NessAiver MS, Carter WB, Bhargavan M,
Lewis RS, et al. Diagnostic accuracy of mammography, clinical ex-
amination, US, and MR imaging in preoperative assessment of
breast cancer. Radiology 2004;233:830-49.
10. Medeiros LR, Duarte CS, Rosa DD, Edelweiss MI, Edelweiss M,
Silva FR, et al. Accuracy of magnetic resonance in suspicious breast
lesions: a systematic quantitative review and meta-analysis. Breast
Cancer Res Treat 2011;126:273-85.
11. Marcotte-Bloch C, Balu-Maestro C, Chamorey E, Ettore F, Raoust I,
Flipo B, et al. MRI for the size assessment of pure ductal carcinoma
in situ (DCIS): a prospective study of 33 patients. Eur J Radiol 2011;
77:462-7.
12. Kim DY, Moon WK, Cho N, Ko ES, Yang SK, Park JS, et al. MRI of
the breast for the detection and assessment of the size of ductal
carcinoma in situ. Korean J Radiol 2007;8:32-9.
13. Shiraishi A, Kurosaki Y, Maehara T, Suzuki M, Kurosumi M. Exten-
sion of ductal carcinoma in situ: histopathological association with
MR imaging and mammography. Magn Reson Med Sci 2003;2:
159-63.
14. Orel SG, Schnall MD, LiVolsi VA, Troupin RH. Suspicious breast
lesions: MR imaging with radiologic-pathologic correlation. Ra-
diology 1994;190:485-93.
15. Solin LJ, Orel SG, Hwang WT, Harris EE, Schnall MD. Relationship
of breast magnetic resonance imaging to outcome after breast-con-
servation treatment with radiation for women with early-stage in-
vasive breast carcinoma or ductal carcinoma in situ. J Clin Oncol
2008;26:386-91.
16. Itakura K, Lessing J, Sakata T, Heinzerling A, Vriens E, Wisner D, et
al. The impact of preoperative magnetic resonance imaging on sur-
gical treatment and outcomes for ductal carcinoma in situ. Clin
Breast Cancer 2011;11:33-8.
17. D’Orsi CJ, Mendelson EB, Ikeda DM. ACR BI-RADS Atlas: Breast
Imaging Reporting and Data System. Reston: American College of
Radiology; 2003.
18. Mendelson EB, Berg WA, Merritt CR. Toward a standardized breast
ultrasound lexicon, BI-RADS: ultrasound. Semin Roentgenol 2001;
36:217-25.
19. Morris EA, Comstock C, Lee C, Lehman CD, Ikeda DM, Newstead
GM, et al. ACR BI-RADS magnetic resonance imaging. In: ACR
BI-RADS Atlas: Breast Imaging Reporting and Data System. Res-
ton: American College of Radiology; 2013.
20. Cutuli B, Fourquet A, Luporsi E, Arnould L, Caron Y, Cremoux Pd,
http://www.jbd.or.kr https://doi.org/10.14449/jbd.2016.4.2.77
84 Musaed Rayzah, et al.
et al. Standards, options and recommendations for the management
of ductal carcinoma in situ of the breast (DCIS): update 2004. Bull
Cancer 2005;92:155-68.
21. Kuhl CK, Schrading S, Bieling HB, Wardelmann E, Leutner CC,
Koenig R, et al. MRI for diagnosis of pure ductal carcinoma in situ:
a prospective observational study. Lancet 2007;370:485-92.
22. Yang WT, Tse GM. Sonographic, mammographic, and histopatho-
logic correlation of symptomatic ductal carcinoma in situ. AJR Am
J Roentgenol 2004;182:101-10.
23. Stavros AT. Ultrasound of ductal carcinoma in situ. In: Silverstein
MJ, editor. Ductal Carcinoma In Situ of the Breast. Baltimore: Wil-
liams & Wilkins; 1997. p.135-58.
24. Yang WT, Suen M, Ahuja A, Metreweli C. In vivo demonstration of
microcalcification in breast cancer using high resolution ultra-
sound. Br J Radiol 1997;70:685-90.
25. Lee MH, Ko EY, Han BK, Shin JH, Ko ES, Hahn SY. Sonographic
findings of pure ductal carcinoma in situ. J Clin Ultrasound 2013;
41:465-71.
26. Soderstrom CE, Harms SE, Copit DS, Evans WP, Savino DA, Krakos
PA, et al. Three-dimensional RODEO breast MR imaging of lesions
containing ductal carcinoma in situ. Radiology 1996;201:427-32.
27. Hwang ES, Kinkel K, Esserman LJ, Lu Y, Weidner N, Hylton NM.
Magnetic resonance imaging in patients diagnosed with ductal car-
cinoma-in-situ: value in the diagnosis of residual disease, occult in-
vasion, and multicentricity. Ann Surg Oncol 2003;10:381-8.
28. Menell JH, Morris EA, Dershaw DD, Abramson AF, Brogi E, Liberman
L. Determination of the presence and extent of pure ductal carcino-
ma in situ by mammography and magnetic resonance imaging.
Breast J 2005;11:382-90.
29. Holland R, Hendriks JH, Vebeek AL, Mravunac M, Schuurmans
Stekhoven JH. Extent, distribution, and mammographic/histologi-
cal correlations of breast ductal carcinoma in situ. Lancet 1990;335:
519-22.
30. Schouten van der Velden AP, Boetes C, Bult P, Wobbes T. The value
of magnetic resonance imaging in diagnosis and size assessment of
in situ and small invasive breast carcinoma. Am J Surg 2006;192:
172-8.