DOI:10.2214/AJR.05.1944
AJR 2006; 187:W528-W537
© American Roentgen Ray Society
Evaluation of Tumor Angiogenesis of Breast Carcinoma Using Contrast-Enhanced Digital Mammography
Clarisse Dromain1,
Corrine Balleyguier1,
Serge Muller2,
Marie-Christine Mathieu3,
France Rochard4,
Paule Opolon3 and
Robert Sigal1
1 Department of Radiology, Institut Gustave Roussy, 39 rue Camille Desmoulin,
Villejuif Cedex, France 94805.
2 GE Healthcare, Buc, France.
3 Department of Pathology, Institut Gustave Roussy, Villejuif Cedex,
France.
4 Department of Surgery, Institut Gustave Roussy, Villejuif Cedex, France.
Received November 3, 2005;
accepted after revision March 30, 2006.
This work was supported by a grant from the Ministère
Français de la Jeunesse, de l'Education et de la Recherche.
Address correspondence to C. Dromain
(dromain{at}igr.fr).
WEB
This is a Web exclusive article.
Abstract
OBJECTIVE. The purpose of this article is to assess the accuracy of
contrast-enhanced digital mammography in the detection of breast carcinoma and
to correlate the findings on the images with those of histologic analysis
using microvessel quantification.
SUBJECTS AND METHODS. Twenty patients with a suspicious breast
abnormality underwent contrast-enhanced digital mammography using a full-field
digital mammography unit that was modified to detect iodinated enhancement.
For each patient, a total of six contrast-enhanced craniocaudal views were
acquired from 30 seconds to 7 minutes after the injection of a bolus of 100 mL
of an iodinated contrast agent. Image processing included a logarithmic
subtraction and the analysis of enhancement kinetic curves. Contrast-enhanced
digital mammography findings were compared with histologic analysis of
surgical specimens, including intratumoral microvessel density quantification
evaluated on CD34-immunostained histologic sections obtained from all
patients.
RESULTS. An area of enhancement was depicted on contrast-enhanced
digital mammograms in 16 of the 20 histologically proven breast carcinomas.
Excellent correlation was seen between the size of enhancement and the
histologic size of tumors, which ranged from 9 to 22 mm. Early enhancement
with washout was observed in four cases, early enhancement followed by a
plateau in four cases, gradual enhancement in seven cases, and unexpected
decrease of enhancement in one case. Intratumoral microvessel density ranged
from 11.7 to 216.6 microvessels per square millimeter. A poor correlation was
found between data measured on contrast-enhanced digital mammography and
intratumoral microvessel density measured on CD34-immunostained histologic
sections.
CONCLUSION. Contrast-enhanced digital mammography is able to depict
angiogenesis in breast carcinoma. Breast compression and projective images
acquisition alter the quantitative assessment of enhancement parameters.
Keywords: angiogenesis • breast • breast cancer • contrast media • digital images • mammography • microvessel
Introduction
Angiogenesis has been a common prognostic indicator for breast carcinoma in
the last decade. Indeed, previous studies have shown that higher intratumoral
microvessel density is statistically correlated with a greater incidence of
metastases, and that intratumoral microvessel density is an independent
prognostic indicator for overall and relapse-free survival in early-stage
invasive breast carcinoma [1,
2]. During the past few years,
many methods for imaging angiogenesis in vivo have been developed. Digital
subtraction angiography of the breast has been performed using a radiograph
image intensifier system [3,
4]. Subtracted images of
malignant tumors showed rapid and strong enhancement followed by a washout,
whereas benign tumors showed less or no enhancement. At present, contrast
medium is used with both CT and MRI techniques to explore angiogenesis in
breast carcinoma. Both techniques improve detection and characterization of
breast carcinomas
[5-9];
however, MRI is limited by its specificity, its high cost, and the restricted
access of MRI time. Full-field digital mammography systems offer new
capabilities not provided by conventional film-screen radiography.
Contrast-enhanced digital mammography is a new breast imaging technique using
full-field digital mammography in conjunction with the injection of an
iodinated contrast medium. The purpose of the present study was to assess the
accuracy of contrast-enhanced digital mammography in the detection of breast
carcinoma and to correlate the technique's findings with those of histologic
analysis using microvessel quantification.
Subjects and Methods
Patients
From September 2000 to October 2002, 20 consecutive patients with 22
suspicious breast abnormalities, who were referred to the institution for
surgery, were enrolled in this prospective study. The study was approved by
the institutional review board and all patients gave their written informed
consent. Inclusion criteria were patients with suspicious breast lesions
detected at physical examination, breast sonography, conventional mammography,
or a combination; patients with lesions that were classified as American
College of Radiology (ACR) BI-RADS
[10] category 4 or 5; and
patients who were scheduled for surgery. Exclusion criteria were women who
were pregnant or believed they could become pregnant and those with a history
of allergic accident with an iodinated contrast agent.
The 20 women selected for the study had a median age of 63 years (age
range, 42-80 years). Suspicious breast lesions included palpable nodules in
eight patients, mammographic opacities in 17 patients (including one bifocal),
and sonographic nodules in 19 patients (including one bifocal lesion different
from the bifocal opacity detected on mammography). (Bifocal is defined as two
different foci of carcinoma in the same breast.) Mammogram features were
spiculated lesions in 13 patients, masses with clustered microcalcifications
in three patients, and a bifocal mass in one patient. Breast densities were
classified as BI-RADS category 2 in 11 patients, category 3 in eight patients,
and category 4 in one patient. Sonographic features were hypoechoic nodules in
18 patients and a bifocal nodule in one patient.
Contrast-Enhanced Digital Mammography Technique
All contrast-enhanced digital mammography examinations were performed
within 15 days before breast surgery with a digital mammography system
(Senographe 2000D, GE Healthcare) equipped with a cesium iodine-amorphous
silicon flat-panel detector.
To perform contrast-enhanced digital mammography, it was necessary to adapt
the digital mammography unit to maximize the sensitivity of the imaging
technique to low concentrations of iodine. The radiograph spectrum was shaped
to be just above the K-edge of iodine (33.2 keV). The digital mammography
system was modified accordingly, adding a copper filter specifically used for
contrast-enhanced digital mammography in addition to the usual molybdenum and
rhodium filters used for standard mammography. Moreover, a high voltage range
of 45-49 kVp was used (instead of 26-32 kVp used for conventional digital
mammography) [11].
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —Kinetic curves of contrast enhancement. Type 1 = early
enhancement followed by washout, type 2 = early enhancement followed by
plateau, type 3 = gradually increasing enhancement, type 4 = decrease of
enhancement with time.
|
|
Patients were comfortably settled to avoid motion. A catheter was inserted
into the antecubital vein of the arm contralateral to the breast of concern.
Each breast under examination was compressed to limit motion. Light breast
compression (5 daN) was used for all images, which was strong enough to limit
motion but not to reduce blood flow. All images were acquired in the same
position during a single breast compression. The milliampere-second value was
chosen according to the thickness and the composition of the breast.
A single mask mammogram was first taken. A one-shot IV injection of 100 mL
of nonionic contrast agent (iohexol [Omnipaque 300, GE Healthcare]) was given,
using a power injector (Vistron CT, Medrad), at a rate of 3 mL/s with bolus
chase. The first contrast-enhanced mammogram was obtained 30 seconds after
starting the injection, and subsequent mammograms were obtained after 90, 150,
240, 330, and 420 seconds. Therefore, a total of seven mammograms including
six contrast-enhanced mammograms were obtained for each patient.
The mean examination duration was approximately 15 minutes (ranging from 12
to 25 minutes). The total radiograph dose of the procedure ranged between 1
and 4 mGy, which is similar to a conventional single-view mammogram.
Image Analysis
A research workstation was used for image analysis. First, compensation of
breast motion and logarithmic subtraction between unenhanced and
contrast-enhanced images were performed. All focal areas of enhancement
depicted on subtracted images were considered to be abnormalities. The maximum
diameter of the tumor measured on subtracted contrast-enhanced digital
mammography images was recorded. Regions of interest were placed at areas of
early enhancement and adjacent breast tissues to analyze the uptake and the
washout of the contrast agent. To minimize the effect of breast thickness on
density values, the region of interest for the lesion and the healthy breast
tissue had the same size and were located at the same distance from the
posterior aspect of the breast. Values of differential contrast enhancement
between lesions and healthy breast tissues were then plotted versus time. The
time-intensity curves were classified into four types based on the wash-in and
washout of contrast medium (Fig.
1): type 1, early enhancement followed by washout; type 2, early
enhancement followed by plateau; type 3, gradually increasing enhancement;
type 4, decrease of enhancement with time. The enhancement was considered
early if the peak of enhancement was before 1 minute 30 seconds. The following
quantitative values extracted from the images were recorded: peak of
enhancement, value of enhancement at 90 seconds, area under the enhancement
curve, and maximum gradient of enhancement. Each quantitative value was
normalized with an estimate of the tumor thickness (considered as the maximum
diameter of the tumor on subtracted images) so that the quantitative value
became independent of the tumor size and examinations could be compared.
Histologic Analysis
The surgical specimen from each patient was histologically analyzed. For
evaluation of intratumoral vascular density, anti-CD34 immunohistochemistry,
detecting human endothelial cells was performed for all tumors on a large
section that included the total tumor. All histologic slides were processed
during the same experiment and validated by an expert breast pathologist.
Evaluation of intratumoral vascular density was achieved by scanning the whole
immunostained specimen using a dedicated Nikon Super Coolscan 8000 film
scanner. Digitalized images were analyzed using Pix Cyt software
(François Baclesse Center)
[12,
13]. This automatic procedure
included five steps: background correction, tissue detection and necrosis
elimination, immunostained structure detection, hot spot identification, and
measurements. For each image, intratumoral mean microvessel density expressed
as the number of microvessels per square millimeter (mm2) of tumor
tissues, and intratumoral mean microvessel surface proportion were determined.
Microvessel counts were determined without knowledge of contrast-enhanced
digital mammography results.
Statistical Analysis
The correlation of the size of the tumor from contrast-enhanced digital
mammography images and from histology sections was analyzed. To compare the
image findings with histologic findings, intratumoral microvessel density and
mean intratumoral microvessel surface between the group of true-positive
results (tumor enhancement depicted on contrast-enhanced digital mammography
images corresponding to a malignant tumor that was histologically proven) and
the group of false-negative results (absence of tumor enhancement detected on
contrast-enhanced digital mammography images but the presence of malignant
tumor at histology) were compared using the two-tailed Student's t
test. Pearson's and Spearman's correlation tests were used to compare the
quantitative data evaluated on contrast-enhanced digital mammography images
and intratumoral microvessel density and mean intratumoral microvessel surface
from immunostained histologic sections.
Results
Results for each patient are summarized in
Table 1. Contrast-enhanced
digital mammography examinations were successful in 18 patients and failed in
two patients. For these two patients, the lesions were located in the deep
part of the breast close to the pectoral muscle and went out of the field of
view between the unenhanced image and the contrast-enhanced images. For the
remaining 18 patients, contrast-enhanced digital mammography showed a focal
enhancement in 15 patients (positive result) and showed no enhancement in
three patients (negative result). The median maximum diameter of the lesions
measured on the subtracted image at 90 seconds after injection was 18 mm,
ranging from 9 to 30.4 mm. The kinetic curves of enhancement were classified
as type 1 in four tumors (Figs.
2A,
2B,
2C,
2D, and
2E), type 2 in four tumors
(Figs. 3A,
3B,
3C,
3D, and
3E), type 3 in seven tumors
(Figs. 4A,
4B,
4C,
4D, and
4E), and type 4 in one tumor
(Figs. 5A,
5B,
5C,
5D, and
5E).
View larger version (85K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —62-year-old woman with nonpalpable mass at physical
examination. Subtraction image derived from 1.30-second contrast-enhanced
digital mammography image shows two adjacent areas of enhancement, one with
strong and spiculated enhancement (arrow) and one with moderate and
less-circumscribed enhancement (arrowheads).
|
|
View larger version (7K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —62-year-old woman with nonpalpable mass at physical
examination. Kinetic curves of enhancement derived from regions of interest
drawn in strong area of enhancement (•) shows early enhancement followed
by washout, whereas moderate area of enhancement ( ) shows gradually
increasing enhancement.
|
|
View larger version (178K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —62-year-old woman with nonpalpable mass at physical
examination. At histopathologic examination, lesion with strong and early
contrast enhancement proved to be ductal carcinoma in situ, whereas lesion
with gradually increasing enhancement proved to be invasive ductal
carcinoma.
|
|
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2E —62-year-old woman with nonpalpable mass at physical
examination. Immunohistochemically stained section (anti-CD34 stain) shows
intratumoral microvessel density values higher for in situ (167.04
microvessels/mm2) than for invasive (69.03
microvessels/mm2) component of tumor.
|
|
View larger version (91K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —43-year-old woman with palpable mass at physical examination.
Subtraction image derived from 1.30-second contrast-enhanced digital
mammography image well depicts round enhancing mass in deep part of breast
(arrow).
|
|
View larger version (7K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —43-year-old woman with palpable mass at physical examination.
Kinetic curve of contrast enhancement derived from regions of interest drawn
in this lesion shows early enhancement followed by plateau.
|
|
View larger version (167K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3E —43-year-old woman with palpable mass at physical examination.
Immunohistochemically stained section (anti-CD34 stain) shows numerous
microvessels (brown markings) in tumor stroma with intratumoral microvessel
density value of 104.33/mm2.
|
|
View larger version (53K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —45-year-old woman with palpable mass at physical examination.
Subtraction image derived from 1.30-second contrast-enhanced digital
mammography image shows homogeneous enhancement of lesion
(arrow).
|
|
View larger version (7K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4C —45-year-old woman with palpable mass at physical examination.
Kinetic curve of contrast enhancement derived from regions of interest drawn
in this lesion shows gradually increasing enhancement.
|
|
View larger version (166K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4E —45-year-old woman with palpable mass at physical examination.
Immunohistochemically stained section (anti-CD34 stain) shows high density of
microvessels (brown markings) in tumor with intratumoral microvessel density
value of 103.78 microvessels/mm2.
|
|
View larger version (6K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C —60-year-old woman with palpable nodule at physical
examination. Kinetic curve of contrast enhancement derived from regions of
interest drawn in this lesion shows unexpected decrease of enhancement with
time, probably caused by motion artifacts.
|
|
View larger version (117K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5E —60-year-old woman with palpable nodule at physical
examination. Immunohistochemically stained section (anti-CD34 stain) shows
high density of microvessels (brown markings) in tumor with intratumoral
microvessel density value of 85.61/mm2.
|
|
Histologic analyses of surgical specimens are summarized in
Table 1. It shows malignant
tumors in all patients including two cases of bifocal carcinoma. Therefore, a
total of 20 tumors were analyzed, including 16 invasive ductal carcinomas, one
invasive lobular carcinoma, one ductal carcinoma in situ with microinvasion
(with an invasive component of less than 5%), and two sets of bifocal tumors.
In one patient, the bifocal tumors consisted of one invasive ductal carcinoma
and one ductal carcinoma in situ located in the same quadrant (Figs.
2A,
2B,
2C,
2D, and
2E). Another patient had one
lobular ductal carcinoma and one invasive lobular carcinoma located in
different quadrants. The median maximum diameter of lesions measured at
histology was 14 mm (range, 5-25 mm). The median values of the intratumoral
microvessel density and the intratumoral microvessel surface were 79.2
microvessels per square millimeter (range, 11.7-216.6 microvessels per square
millimeter) and 2.6% (range, 0.5-6.9%), respectively.
Globally, the comparison between the 18 contrast-enhanced digital
mammography sequences and the histology results showed 16 true-positives and
four false-negatives for contrast-enhanced digital mammography. Consequently,
the sensitivity of this technique for detecting breast carcinomas was 80% (95%
CI, 56-94%).
A good correlation was found between the size of lesions measured on
contrast-enhanced digital mammography images and those measured on histologic
sections, with a coefficient of correlation of 95%
(Fig. 6). Histologic analysis
of the tumor with an unexpected decrease of enhancement (patient 8) showed no
necrosis. When comparing the group of true-positives with the group of
false-negatives, a trend was observed toward higher intratumoral microvessel
density in true-positives with a median value of 79.2 microvessels per
mm2 compared with a median value of 56.5 microvessels per
mm2 for the false-negatives, but the difference was not
statistically significant (p = 0.72).
Table 2 shows the results of
correlation studies between the different enhancement parameters from
contrast-enhanced digital mammography and microvessel density determination. A
poor correlation was found between the area under the curve (the value at 90
seconds) and the peak of enhancement without significant statistical
correlation.
View this table:
[in this window]
[in a new window]
|
TABLE 2: Pearson's Correlation Coefficient Comparing Contrast-Enhanced Digital
Mammography Images with Mean Intratumoral Microvessel Density and Mean
Intratumoral Microvessel Surface from Immunostained Sections
|
|
Discussion
The present study indicates that contrast-enhanced digital mammography is
able to show breast carcinomas. The sensitivity of this technique for the
detection of breast carcinomas was 80%. This result is consistent with
previous studies
[14-16].
In a recent study on dynamic contrast-enhanced digital mammography, Jong at
al. [14] showed that
enhancement was observed in eight of 10 patients with histologically proven
breast carcinoma and five of 12 patients with a histologically proven benign
lesion corresponding to three fibroadenomas and two fibrocystic changes with
focal intraductal hyperplasia. Lewin at al.
[16], in a study on
dual-energy contrast-enhanced digital mammography, reported strong
enhancement, moderate enhancement, and weak enhancement in, respectively,
eight, three, and two of 14 malignant lesions and weak enhancement in two of
12 patients with benign lesions, corresponding to one atypical ductal
hyperplasia and one fibrocystic change. The specificity of contrast-enhanced
digital mammography was not evaluated in this study because all patients had
malignant tumors and the negative predictive value was unknown. Actually,
because the goal of our study was to compare contrast-enhanced digital
mammography data with histologic data of surgical specimens, our inclusion
criteria required suspicious breast abnormalities referred for surgery,
creating selection bias.
The contrast-enhanced digital mammography results reported in this study
were independent of the histologic type of carcinoma. Indeed, in the two cases
of ductal carcinoma in situ (DCIS) and microinvasive ductal carcinoma and the
two cases of invasive lobular carcinoma, a focal brightly enhancing lesion was
seen after contrast medium administration, whereas our four false-negative
results were invasive ductal carcinoma. These results suggest that the rate of
contrast medium enhancement is not related to the histologic type of the tumor
and reflects the heterogeneity that exists between different tumors in their
ability to induce angiogenesis. Similar results have been already shown using
dynamic enhanced MRI and explain limitations observed to differentiate benign
from malignant breast tumors
[17,
18]. Furthermore, Gilles et
al. [19] observed angiogenesis
in 34 of 36 cases of DCIS with early contrast medium enhancement.
Gradual increase of enhancement was the most common kinetic curve observed
in malignant lesions during contrast-enhanced digital mammography
examinations. A typical contrast MRI curve for malignancy
[20,
21] with rapid enhancement
followed by a decrease during the delayed phase was observed in only four of
the 20 malignant lesions of this study. These results differ from the typical
kinetic curves observed using MRI in malignant tumors, which usually tend to
exhibit a washout pattern, whereas benign tumors usually exhibit either
persistent or plateau-type enhancements
[17,
22]. These differences between
kinetic curves observed using dynamic contrast-enhanced digital mammography
and MRI are probably because of the breast compression, even low breast
compression, which may alter blood flow.
We found poor correlation between the intratumoral mean vascular density
evaluated on CD34-immunostained histologic sections and quantitative
characteristics of kinetic curves of enhancement. This could be because of the
difficulty of quantitative assessments on projection images acquired with
breast compression. Indeed, contrast-enhanced digital mammography images are
projections through the entire breast, and enhancement depends on the size of
the tumor. Moreover, enhancement is not related exclusively to the number of
vessels, but is also likely related to functional parameters such as vessel
permeability, particularly when using a contrast agent that is migrating to
the extracellular fluid space.
In dynamic contrast-enhanced digital mammography, patient motion is the
leading cause of artifacts observed on subtracted images that negatively
impact quantitative measurements. Two contrast-enhanced digital mammography
examinations failed in our series for this reason. We had two cases of deep
tumors located close to the chest wall where the lesions went out of the field
of view between unenhanced and contrast-enhanced images. In one patient, we
observed an unexpected decrease in the enhancement inside the tumor after
contrast medium injection. This case of so-called "black
carcinoma" was not associated with histologic intratumoral necrosis. In
this case, we observed major motion artifacts during image acquisition that
could probably explain this abnormal decrease of density values.
One of the limitations of the contrast-enhanced digital mammography as
performed in this study is the restriction to a unilateral craniocaudal view.
We chose the craniocaudal projection rather than the mediolateral oblique
projection to minimize motion artifacts. Because we wanted to analyze the
wash-in and the washout of the contrast material in the tumor and to quantify
it, patients were under breast compression for 7 minutes; we therefore more
comfortably placed them on the craniocaudal view rather than on the
mediolateral oblique view. However, the craniocaudal position is questionable
because it does not allow as much breast tissue to be visualized as does the
mediolateral oblique view. In the future, we expect to optimize the technical
protocol of contrast-enhanced digital mammography with a higher temporal
resolution on craniocaudal projections (six image acquisitions every 30
seconds during the first 3 minutes after initiation of contrast agent
administration, rather than six image acquisitions during 7 minutes) and with
the introduction of an additional mediolateral oblique view at the end of the
examination (one image acquisition at 5 minutes).
Other limitations of contrast-enhanced digital mammography in breast
imaging compared with MRI are irradiation of the breast, lower contrast
resolution, and the use of iodinated contrast material. However, the technique
is weakly irradiating, easily implemented, inexpensive, fast, and practical.
It should be considered a complementary tool to conventional digital
mammography. One of its potential applications should be the detection of
lesions occult on conventional mammography, particularly in dense breasts.
Another application of interest should be the clarification of equivocal
lesions on conventional imaging, particularly in follow-up after
breast-conserving therapy. Furthermore, in the near future, contrast-enhanced
digital mammography will probably have the benefit of other digital
mammography improvements such as tomosynthesis
[23]. Also, tomosynthesis is
still an investigational method. Technical improvements and additional studies
are necessary to assess the specificity of this new imaging technique and its
diagnostic accuracy in a larger series of patients.
Acknowledgments
We thank Fanny Jeunehomme for her technical assistance and statistical
analysis.
References
- Weidner N, Semple JP, Welch WR, et al. Tumor angiogenesis and
metastasis: correlation in invasive breast carcinoma. N Engl J
Med 1991; 324:1
-8[Abstract]
- Chu JS, Lee WJ, Chang TC, Chang KJ, Hsu HC. Correlation between
tumor angiogenesis and metastasis in breast cancer. J Formos Med
Assoc 1995; 94:373
-378[Medline]
- Ackerman LV, Watt AC, Shetty P, et al. Breast lesions examined by
digital angiography: work in progress. Radiology1985; 155:65
-68[Abstract/Free Full Text]
- Watt AC, Ackerman LV, Windham JP, et al. Breast lesions:
differential diagnosis using digital subtraction angiography.
Radiology 1986;159
: 39-42[Abstract/Free Full Text]
- Hagay C, Cherel PJ, de Maulmont CE, et al. Contrast-enhanced CT:
value for diagnosing local breast cancer recurrence after conservative
treatment. Radiology 1996;200
: 631-638[Abstract/Free Full Text]
- Chang CH, Nesbit DE, Fisher DR, et al. Computed tomographic
mammography using a conventional body scanner. AJR1982; 138:553
-558[Abstract/Free Full Text]
- Harms SE, Flamig DP. MR imaging of the breast. J Magn
Reson Imaging 1993; 3:277
-283[Medline]
- Heywang SH, Wolf A, Pruss E, et al. MR imaging of the breast with
Gd-DTPA: use and limitations. Radiology1989; 171:95
-103[Abstract/Free Full Text]
- Orel SG, Schnall MD. MR imaging of the breast for the detection,
diagnosis, and staging of breast cancer. Radiology2001; 220:13
-30[Abstract/Free Full Text]
- American College of Radiology (ACR). BI-RADS:
mammography, 4th ed. Breast imaging reporting and data system: BI-RADS
atlas. Reston, VA: ACR, 2003
- Skarpathiotakis M, Yaffe MJ, Bloomquist AK, et al. Development of
contrast digital mammography. Med Phys2002; 29:2419
-2426[CrossRef][Medline]
- Kim NT, Elie N, Plancoulaine B, et al. An original approach for
quantification of blood vessels on the whole tumour section. Anal
Cell Pathol 2003; 25:63
-75[Medline]
- Elie N, Plancoulaine B, Signolle JP, Herlin P. A simple way of
quantifying immunostained cell nuclei on the whole histologic section.
Cytometry A 2003;56
: 37-45[Medline]
- Jong RA, Yaffe MJ, Skarpathiotakis M, et al. Contrast-enhanced
digital mammography: initial clinical experience.
Radiology 2003;228
: 842-850[Abstract/Free Full Text]
- Diekmann F, Diekmann S, Taupitz M, et al. Use of iodine-based
contrast media in digital full-field mammography: initial experience.
Rofo 2003; 175:342
-345[Medline]
- Lewin JM, Isaacs PK, Vance V, et al. Dual-energy contrast-enhanced
digital subtraction mammography: feasibility.
Radiology 2003;229
: 261-268[Abstract/Free Full Text]
- Buadu LD, Murakami J, Murayama S, et al. Breast lesions:
correlation of contrast medium enhancement patterns on MR images with
histopathologic findings and tumor angiogenesis.
Radiology 1996;200
: 639-649[Abstract/Free Full Text]
- Frouge C, Guinebretiere JM, Contesso G, et al. Correlation between
contrast enhancement in dynamic magnetic resonance imaging of the breast and
tumor angiogenesis. Invest Radiol 1994;29
: 1043-1049[Medline]
- Gilles R, Zafrani B, Guinebretiere JM, et al. Ductal carcinoma in
situ: MR imaging-histopathologic correlation.
Radiology 1995;196
: 415-419[Abstract/Free Full Text]
- Heywang-Kobrunner SH, Haustein J, Pohl C, et al. Contrast-enhanced
MR imaging of the breast: comparison of two different doses of gadopentetate
dimeglumine. Radiology 1994;191
: 639-646[Abstract/Free Full Text]
- Kaiser WA, Zeitler E. MR imaging of the breast: fast imaging
sequences with and without Gd-DTPA—preliminary observations.
Radiology 1989;170
: 681-686[Abstract/Free Full Text]
- Kuhl CK, Mielcareck P, Klaschik S, et al. Dynamic breast MR
imaging: are signal intensity time course data useful for differential
diagnosis of enhancing lesions? Radiology1999; 211:101
-110[Abstract/Free Full Text]
- Smith AP, Hall PA, Marcello DM. Emerging technologies in breast
cancer detection. Radiol Manage 2004;26
: 16-27[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
