Original Investigation

Contrast Enhancement in Breast

Cancer and Background Mammary-
Gland Tissue During the Super-Early
Phase of Dynamic Breast Magnetic
Resonance Imaging
Tetsuya Tomida, Atsushi Urikura, PhD, Takayoshi Uematsu, MD, PhD,
Kensei Shirata, Yoshihiro Nakaya

Rationale and Objectives: We aimed to compare the contrast enhancement between tumor and mammary-gland tissue to distin-
guish lesions in the super-early phase, during which minimal contrast media uptake is observed in mammary-gland tissue.
Materials and Methods: Dynamic magnetic resonance imaging, including the super-early phase with bolus tracking (BT) method (to
determine the optimal imaging start time), was performed by using identical parameters to obtain transverse fat-suppressed T1-
weighted images of both breasts. The percent enhancement (PE) and the contrast ratio (CR) indicators for tumor and mammary-gland
tissue were assessed in each dynamic phase.
Results: The PE values of the tumor were 62.4% and 151.6%, and those of the mammary gland were 0.3% and 20.7% in the super-
early and early phases, respectively. Therefore, virtually no background parenchymal enhancement was observed in the super-early
phase. The variation in the PE values during the super-early phase was significantly smaller when the values were determined with the
BT method (P < .05). The CR was highest in the early phase, and the CR in the super-early phase was lower than in the other phases.
Early-phase PE and CR were significantly higher in invasive cancer cases than in noninvasive cancer cases (P < .01). A significant dif-
ference in the imaging start time was observed for the anatomic side factor by the BT method.
Conclusion: Background parenchymal enhancement almost never appeared in the super-early phase, but the CR was lower in the
super-early phase than in the early phase. The BT method allowed for an optimal imaging start time for the super-early phase and
yielded images with less deviation of contrast enhancement.
Key Words: Breast; magnetic resonance imaging; background parenchymal enhancement; breast cancer; mammary gland; contrast
media.
© 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.

INTRODUCTION MRI screening for patients with a lifetime breast cancer risk
of >20% (1). High-field MRI facilitates higher resolution and

T
he increase in breast cancer-related morbidity has led
faster imaging. When combined with the use of contrast media
to the widespread use of mammography, ultrasonog-
(CM), MRI can be used to detect breast cancer and examine
raphy, and diagnostic imaging modalities in screenings
its spread as well (2–4). The use of CM in MRI for breast
and clinical examinations. Magnetic resonance imaging (MRI)
cancer diagnosis is considered essential by many guidelines (5–7).
is particularly considered useful for breast cancer examina-
In particular, dynamic MRI is recommended to differenti-
tion. The American Cancer Society guidelines recommend
ate benign from malignant tumors. For breast cancer, images
Acad Radiol 2017; 24:1380–1386
are obtained at the early phase, wherein contrast enhance-
From the Radiation and Proton Therapy Center, Shizuoka Cancer Center, 1007
ment peaks approximately 60–120 seconds after CM injection,
Shimonagakubo, Nagaizumi-cho, Shizuoka 411-8777 (T.T.); Department of and at the delay phase to confirm CM washout (5,7–9).
Diagnostic Radiology, Shizuoka Cancer Center, Nagaizumi-cho, Shizuoka, (A.U., The menstrual cycle and other factors impact contrast en-
K.S., Y.N.); Breast Imaging and Breast Intervention Section, Shizuoka Cancer
Center, Nagaizumi-cho, Shizuoka, Japan (T.U.). Received December 29, 2016; hancement in normal mammary-gland tissue (10–12). However,
revised May 30, 2017; accepted May 30, 2017. Address correspondence to: the number of contrast-enhanced breast MRI sessions that can
T.T. e-mail: te.tomida@scchr.jp
be conducted daily is limited in some institutions; hence, it
© 2017 The Association of University Radiologists. Published by Elsevier Inc.
All rights reserved.
may not be possible to schedule examinations based on the
https://doi.org/10.1016/j.acra.2017.05.018 menstrual cycle. In such cases, during conventional dynamic

1380
Downloaded for Mahasiswa FK UKDW 01 (mhsfkdw01@gmail.com) at Universitas Kristen Duta Wacana from ClinicalKey.com by Elsevier on January 29, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Academic Radiology, Vol 24, No 11, November 2017 SUPER-EARLY PHASE OF BREAST MRI

MRI, the CM is also taken up by mammary-gland tissue, some- The ethics committee of our institution approved the anal-
times even outside the menstrual cycle, which decreases the ysis of these clinical images. All patients gave their informed
contrast with tumor tissue because of background parenchy- consent to undergo breast MRI.
mal enhancement (BPE). Uematsu et al. reported that BPE
negatively affects the detection, diagnosis, and staging of
Imaging Sequences
breast cancer (13,14). BPE is expected to be lower in the super-
early phase because normal mammary-gland tissue becomes All data acquisitions were performed by using a 3-Tesla
enhanced more slowly than tumors do (15). Based on this ex- MRI (Achieva 3.0T TX, Philips Healthcare, Amsterdam, The
pectation, super-early-phase images would yield data that could Netherlands) with a dedicated coil (Mammo Track SENSE
assist in both the detection and diagnosis of breast cancer. Breast Coil 16 elements, Philips Healthcare, Amsterdam,
Therefore, previous studies have attempted to improve the The Netherlands). A power injector (Sonic Shot GX, Nemoto
diagnostic capability of dynamic images by adding imaging Kyorindo, Tokyo, Japan) was used to inject the CM. All images
during the super-early phase to the standard protocol (16,17). were obtained with the patient in the prone position, with
In the present study, we focused on the contrast enhance- both arms raised if possible. Transverse fat-suppressed gradi-
ment between breast cancer and normal mammary-gland ent echo (GRE) three-dimensional (3D) T1-weighted images,
tissue to investigate lesion depiction in invasive cancers at the called “e-Thrive,” were obtained during four phases: precontrast,
super-early phase, during which minimal CM uptake is ob- super-early phase, early phase, and delay phase. The dynamic
served in mammary-gland tissue, that would yield good results phases followed fat-suppressed T2-weighted images and
in both the detection and diagnosis of breast cancer. As many diffusion-weighted images. Table 1 describes the detailed imaging
cases of noninvasive cancer do not exhibit the classic patterns parameters for the fat-suppressed GRE 3D T1-weighted images.
of contrast enhancement in invasive cancer (18,19), we sep- Fat suppression was performed through frequency-selective
arately analyzed invasive and non-invasive cancers. Moreover, fat suppression (spectral attenuated with inversion recovery
we compared the superiority of two methods for initiating [SPAIR]; Philips Healthcare). We used an imaging protocol
imaging: the bolus tracking (BT) method, wherein varia- for high-resolution bilateral mammography recommended by
tions in contrast enhancement due to different sites of CM the breast MRI guidelines (5–7,20), along with fat suppression.
injection or circulatory dynamics can be reduced, and the fixed-
time method. We also examined the differences in the imaging
Contrast Imaging Protocol
start times for the super-early phase based on the injection
site by using the BT method. All the patients were injected with a gadolinium CM
(0.1 mmol/kg body weight; 0.2 ml/kg) at 2 ml/s, followed
MATERIALS AND METHODS by a 25-ml saline flush at the same injection rate. Meglumine
Subjects gadopentetate (Magnevist, Bayer, Leverkusen, Germany) or
gadodiamide hydrate (Omniscan, Daiichi-Sankyo, Tokyo,
The subjects who underwent imaging with the fixed-time and Japan) was randomly selected as the CM, whereas gadoteridol
BT methods were selected retrospectively from among women
who underwent MRI for suspected malignant breast tumors
from August 2012 to December 2012. In our institute, the TABLE 1. Acquisition Parameters of Fat-Suppressed GRE
scheduling of MRI examinations does not consider the men- 3D T1-Weighted Images for Dynamic Breast MRI
strual cycle. In total, 113 women (mean age, 53 years; range,
Parameters Values
28–87 years) with 124 malignant tumors were selected; pa-
tients who had received chemotherapy (13 people, 15 tumors) Field of view (mm) 330 × 330
were excluded. Imaging of the first 55 consecutive tumors Matrix size 400 × 320
was performed with the fixed-time method, including 40 in- TR/TE (ms) 3.42/1.79
Flip angle (degree) 10
vasive cancer cases, 12 non-invasive cancer cases, and 3 other
Echo train length 33
pathology cases. Imaging of the remaining 69 tumors was per-
NEX 1
formed with the BT method, including 54 invasive cancer Slice thickness/gap (mm) 1.8/−0.9
cases, 10 non-invasive cancer cases, and 5 other pathology Phase encoding order Centric order
cases. Acceleration factor; R–L/H–F 2/1
To investigate the differences in the imaging start time for Half scan factor; R–L/H–F 0.675/0.8
the super-early phase based on the injection site with the BT Shimming SmartExam Breast (Philips
method, we examined 234 patients who had undergone breast Healthcare, Amsterdam,
MRI from October 2012 to March 2013. Five patients with The Netherlands)
the slowest CM injection ratio were excluded. In these five Time per phase (s) 62
patients, standard injection ratio was not achieved because a GRE, gradient echo; H–F, head-foot; MRI, magnetic resonance
narrow needle was used or because the CM was injected into imaging; NEX, number of excitations; R–L, right-left; TE, echo time;
a fragile vein. TR, repetition time.

1381
Downloaded for Mahasiswa FK UKDW 01 (mhsfkdw01@gmail.com) at Universitas Kristen Duta Wacana from ClinicalKey.com by Elsevier on January 29, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
TOMIDA ET AL Academic Radiology, Vol 24, No 11, November 2017

Figure 1. The schema shows the process of dynamic imaging with the BT method in the present study. Before injecting contrast media,
precontrast images were acquired as reference. According to the BT method, the trigger for starting the super-early phase was when the
contrast media reached the aortic arch. The early phase, which followed the super-early phase, was started 1–2 minutes after the injection.
The scanning of the diseased breast in the sagittal plane was followed by the delay phase at approximately 5 minutes after the injection.
BT, bolus tracking.

(ProHance, Eisai, Tokyo, Japan) was used for the patients whose
renal function had declined.
The imaging start of the super-early phase was determined
by two methods: the fixed-time method, wherein imaging
was initiated 20 seconds after the start of CM injection, and
the BT method. The dynamic imaging sequences by both
methods consisted of the same parameters, as described in
Table 1. Boetes et al. reported that contrast enhancement of
malignant tumors can be observed within 11.5 seconds of the
CM reaching the aorta (16). Hence, we set the super-early-
phase imaging at 20 seconds after CM injection and the early-
phase imaging at 1–2 minutes to conform to the guidelines
(5,7). For the BT method, successive coronal images of the
thoracic aorta were obtained at approximately 0.7-second in-
tervals by using fast GRE sequence. Imaging was initiated when
the CM was visually observed to have reached the aortic arch.
Figure 1 illustrates the flow of dynamic imaging with the BT
method. The fat-suppressed GRE 3D T1-weighted images
on the sagittal plane for the affected breast were acquired between
the early and delay phases.
Comparisons of Contrast Values and Images
The percent enhancement (PE) representing the contrast en-
hancement in the tissue (8,9) and the contrast ratio (CR) (21)
between tumor and mammary-gland tissue were calculated
during the super-early phase, early phase, and delay phase by
using the fixed-time and BT methods. The PE and CR were
calculated with the following equations:
PE = 100 × (SI CE − SI pre ) SI pre [ % ]
CR = (SI t − SI m .g ) (SI t + SI m .g )
In Equation 1, SICE and SIpre are the signal intensity at the
postcontrast and the precontrast stage, respectively, which were
estimated for both the tumor and mammary-gland tissue. In
Equation 2, SIt and SIm.g are the signal intensity of tumor and
mammary-gland tissue, respectively. These mean signal intensities
were calculated within a square region of interest (ROI; 7 × 7
pixels) on the tumor and at an ipsilateral site distal from the
tumor that was considered to be the mammary-gland tissue
on post-contrast images (selected at the discretion of a single
rater). Only one ROI was selected regardless of the tumor
size. The PE and CR were calculated for each imaging phase.
Comparison of Contrast by Pathologic Type
The 69 tumors that underwent dynamic breast MRI with the
BT method were categorized as either invasive or noninva-
sive. The pathologic types were confirmed by using surgical
excision. The PE and CR for both cases were calculated using
Equations 1 and 2, respectively, for each imaging phase.
Analysis of the Imaging Start Time for the Super-Early
Phase with the BT Method
The differences in the imaging start time for the super-early
phase, based on the injection site, were statistically analyzed.
Accordingly, the subjects were assigned to left or right group
(side factor), and proximal forearm including the cubital fossa,
mid-forearm, distal forearm, or dorsum manus group (distance
factor).
Statistical Analysis
R version 3.0.2 (http://www.r-project.org/) was used for all
statistical analyses. The significance level was set at <5%. Welch
(1) t test was used to analyze the PE and CR obtained with the
fixed-time and BT methods. An F test was used to evaluate
(2) the variation in contrast enhancement with the fixed-time and
BT methods. In the analysis of contrast by pathologic type,
as the PE and CR for invasive and noninvasive cancers ex-
hibited homoscedasticity in all phases, Student t test (nonpaired)
was used. In the analysis of the imaging start time, two-way
analysis of variance (ANOVA) was followed by t test cor-
rected with the Tukey–Kramer method for the distance factor.

1382
Downloaded for Mahasiswa FK UKDW 01 (mhsfkdw01@gmail.com) at Universitas Kristen Duta Wacana from ClinicalKey.com by Elsevier on January 29, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Academic Radiology, Vol 24, No 11, November 2017 SUPER-EARLY PHASE OF BREAST MRI

TABLE 2. Background Characteristics of the Examined TABLE 3. Mean and Standard Deviation Values of PE
Patients in This Study During the Dynamic Phases, Obtained With the Bolus-
Tracking and Fixed-Time Methods
Scanning Method
Scanning Method
Characteristics Bolus Tracking Fixed-Time P-Value Dynamic
Phase Bolus Tracking Fixed-Time P-Value
Number of patients 62 50
Number of tumors 69 55 Super-early Mean 62.4 72.8 0.189
Mean age, years (SD) 55.3 (12.3) 55.3 (11.1) 0.991 SD 37.1 47.8 0.049
Mean weight, kg (SD) 53.8 (8.1) 53.5 (9.0) 0.984 Early Mean 151.6 158.2 0.426
SD 48.6 43.1 0.366
SD, standard deviation.
Delay Mean 141.4 153.0 0.095
SD 38.9 37.5 0.775

RESULTS SD, standard deviation.

The P-value for the mean and SD were calculated by Welch t test
PE and F test, respectively.

Table 2 displays the patient characteristics, which did not

significantly differ between the fixed-time and BT groups.
Figure 2 shows the time-signal intensity curves for tumor and
mammary-gland PE. The tumor PE was highest during the
early phase with the fixed-time and BT methods, and de-
clined because of washout during the delay phase. The normal
mammary-gland PE showed almost no increase during the
super-early phase with both methods, but exhibited gradual
increases in the early and delay phases by 20.7% and 36.1%,
respectively. No significant differences between the fixed-
time and BT methods were observed for tumor and mammary-
gland PE during any phase (P > 0.095).
Table 3 describes mean and standard deviation (SD) values
of the tumor PE obtained with both methods. F tests of the
PE for each phase, obtained with both methods, indicated that
200
the variation in the tumor PE values during the super-early
phase were significantly smaller when determined with the
BT method (P < .05). The coefficients of variance (CV) in
the tumor PE values during the super-early phase by the BT
and fixed-time methods were 0.59 and 0.66, respectively.
CR
The super-early phase CR was found to be 0.27 ± 0.14
(mean ± SD) with the fixed-time method and 0.28 ± 0.14
with the BT method. The early-phase CR was recorded as
0.39 ± 0.10 with both methods. Moreover, the delay-phase
CRs were 0.34 ± 0.10 and 0.33 ± 0.10 with the fixed-time
and BT methods, respectively. The CR was highest in the
early phase and lowest in the super-early phase. No signifi-
cant differences in the CR between the two methods were

Tumor (BT) noted for any phase (P > .63).

Tumor (fixed−time)
Mammary gland (BT)
Mammary gland (fixed−time)
150

Comparison of Contrast Enhancement by

Pathologic Type
PE (%)

Table 4 shows the PE and CR values for each phase, and

100

Figure 3 shows the time-signal intensity curves of the PE,

obtained with the BT method, according to the presence of
50

invasive and noninvasive cancer. The early-phase CR was sig-

nificantly higher in invasive cancer cases than in noninvasive
cancer cases (P < .01). Moreover, the PE tended to be higher
0

in invasive cancer cases during all the phases. In fact, in the

Pre−contrast Super−early Early
Dynamic phase
Figure 2. PE for tumor and mammary-gland tissue with the BT and
fixed-time methods. For the tumor tissue, the PE showed the
maximum value on the time-signal intensity curve during the early
phase with both methods, whereas the “washout” was observed only
during the delay phase. All the PEs of the mammary-gland tissue
were considerably lower than those of the tumor tissue. Although
the PE gradually increased, only minimal enhancement of the
mammary-gland tissue was observed during the super-early phase.
All error bars show 95% confidence interval. BT, bolus tracking; PE,
percent enhancement.
Delay invasive cancer cases, the PE was highest during the early phase
and then declined during the delay phase, exhibiting washout.
Differences in the Imaging Start Time for the
Super-Early Phase by the BT Method
Two-way ANOVA of the mean imaging start times did not
show any clear interaction between the side factor and dis-
tance factor (P = .23). Differences with levels of significance
<0.1% with the two-way ANOVA were observed for each
factor. Consequently, a significant difference in the imaging

1383
Downloaded for Mahasiswa FK UKDW 01 (mhsfkdw01@gmail.com) at Universitas Kristen Duta Wacana from ClinicalKey.com by Elsevier on January 29, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
TOMIDA ET AL Academic Radiology, Vol 24, No 11, November 2017

TABLE 4. Categorization of PE and CR According to Pathologic Type in the Different Dynamic Phases

PE CR
Dynamic Phase Pathologic Type Number of Tumors Mean t Test Mean t Test
Super-early IDC 54 67.7 P = .054 0.29 P = .064
DCIS 10 42.8 0.20
Early IDC 54 161.9 P = .00018 0.41 P = .0011
DCIS 10 101.0 0.30
Delay IDC 54 146.9 P = .0047 0.33 P = .34
DCIS 10 110.2 0.30

CR, contrast ratio; DCIS, ductal carcinoma in situ; IDC, invasive ductal carcinoma; PE, percent enhancement.
200

Invasive ductal carcinoma

Ductal carcinoma in situ
150
PE (%)
100
50
0

Pre−contrast Super−early Early Delay

Dynamic phase
Figure 3. The time-intensity curves of the PE for invasive and non-
invasive cancer. The PE tended to be higher in invasive cancer cases
during all the phases. All error bars show 95% confidence interval.
PE, percent enhancement.
start time was observed for the anatomic side factor. The imaging
start time during the super-early phase was 16.9 ± 2.2 seconds
for the right side and 18.6 ± 2.9 seconds for the left side. With
regard to the injection site, a significant delay in the imaging
start time was noted for injections performed at the dorsum
manus (P < .001). The imaging start time was 20.1 ± 3.1 seconds
for the dorsum manus group.
DISCUSSION
Dynamic MRI involving CM use is regularly performed for
breast cancer diagnosis. To our knowledge, however, only a
few studies have reported on the usefulness of the super-
early phase. We expect BPE to be low in the super-early-
phase images that were obtained very early during dynamic
breast MRI examination because normal mammary-gland tissue
becomes enhanced more slowly than tumors do (15), and thus
the super-early-phase images would increase the likelihood
of obtaining high CR between breast cancer and normal
mammary-gland tissue. However, in this study, both the PE
and the CR in the super-early phase were lower than those
in the early phase. In contrast, BPE was almost never seen
in the super-early-phase images. In case of strong BPE or weak
Figure 4. A 10-mm invasive ductal carcinoma in a 46-year-old
woman. (a) A bilateral fat-suppressed dynamic breast MR image ob-
tained during the super-early phase shows very good contrast between
the tumor tissue (arrow) and normal mammary-gland tissue. (b) A
bilateral fat-suppressed dynamic breast MR image obtained during
the early phase shows good contrast between the tumor tissue (arrow)
and normal mammary-gland tissue. However, moderate BPE appears
at the early phase. (c) A bilateral fat-suppressed dynamic breast MR
image obtained during the delay phase shows poor contrast between
the tumor tissue (arrow) and normal mammary-gland tissue. The lesion
was obscured by marked BPE. BPE, background parenchymal en-
hancement; MR, magnetic resonance.
tumor enhancement, it may be easier to detect and diagnose
tumors by using super-early-phase images than by using early-
phase images, if dynamic breast MRI examinations include
super-early-phase images (Figs 4 and 5). Uematsu et al. reported
that the signal intensity of BPE in dynamic breast MRI
affects breast cancer diagnosis (13,14). The authors found that
increased signal intensity of BPE led to a decrease in diag-
nostic precision and that the ability to detect lesions decreased
in cases with strong BPE. In addition, the present study shows
that normal breast tissue was less enhanced in the super-
early phase, which may allow accurate identification of cancer
using shorter MRI protocols. Recently, Abe et al. reported

1384
Downloaded for Mahasiswa FK UKDW 01 (mhsfkdw01@gmail.com) at Universitas Kristen Duta Wacana from ClinicalKey.com by Elsevier on January 29, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Academic Radiology, Vol 24, No 11, November 2017
Figure 5. A 15-mm noninvasive ductal carcinoma in a 51-year-
old woman. (a) A bilateral fat-suppressed dynamic breast MR image
obtained during the super-early phase shows good contrast between
the tumor tissue (arrow) and normal mammary-gland tissue. (b) A
bilateral fat-suppressed dynamic breast MR image obtained during
the early phase shows poor contrast between the tumor tissue (arrow)
and normal mammary-gland tissue because moderate BPE makes
ill-defined margins between the two types of tissue. (c) A bilateral
fat-suppressed dynamic breast MR image obtained during the delay
phase shows poor contrast between the tumor tissue (arrow) and
normal mammary-gland tissue. The lesion was completely ob-
scured by marked BPE. BPE, background parenchymal enhancement;
MR, magnetic resonance.
that ultrafast dynamic contrast-enhanced breast MRI imaging
is useful for discriminating benign from malignant lesions (22).
Further studies to evaluate the diagnostic utility of super-
early-phase images are needed.
Previous studies showed that many noninvasive cancer
cases do not exhibit the typical contrast enhancement (18,19),
consistent with the present findings. Moreover, we observed
typical results for invasive cancers, wherein contrast enhance-
ment was highest during the early phase and washout was
observed during the delay phase. These findings indirectly
suggest that there was no selection bias in our subject pop-
ulation and that there were no problems with our research
methods or results. The CR, which expresses contrast between
tumor and background mammary-gland tissue, was signifi-
cantly higher during the early phase in invasive cancer cases
than in noninvasive cancer cases. Moreover, the contrast during
the super-early phase tended to be higher in cases with in-
vasive cancer than in cases with noninvasive cancer (P = .064).
Thus, the super-early phase with the BT method could be
used to effectively identify invasive cancers.
In our study, we did not observe any significant differ-
ences in PE or CR for any phase between the BT method
and the fixed-time method, indicating that both methods pro-
vided similar image information. The imaging start time of
the super-early phase was sometimes greater than 20 seconds.
Downloaded for Mahasiswa FK UKDW 01 (mhsfkdw01@gmail.com) at Universitas Kristen Duta Wacana from ClinicalKey.com by Elsevier on January 29, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
SUPER-EARLY PHASE OF BREAST MRI
Thus, some super-early-phase images obtained by using the
fixed-time method, wherein imaging was initiated 20 seconds
after the start of CM injection, may have been acquired too
early. The BT method may help reduce variabilities relating
to circulation and different injection anatomic sites. We believe
that such cases led the SD and CV values of PE during the
super-early phase to be significantly higher with fixed-time
method than with the BT method; however, the BT method
and the fixed-time method provided similar enhancement
levels. Furthermore, the BT with visual observation, in which
the start of the super-early phase was triggered off when the
CM reached the aortic arch, may cause variability and lead
to complexity in dynamic breast MRI examinations. The eval-
uation for the BT technique used routinely should be explored
further. In the future, automatic imaging start for the BT
method will probably be developed.
In the 1990s, MRI devices were not able to provide high
levels of both temporal and spatial resolution; hence, imaging
parameters that prioritized one over the other had to be se-
lected when performing breast MRI. In previous studies,
imaging sequences that prioritized increased temporal reso-
lution over spatial resolution allowed for the enhanced
evaluation of kinetics through multiphase imaging immedi-
ately after and before CM injection (16,17). However, Kuhl
et al. reported that the diagnostic certainty increased when
imaging sequences that prioritized spatial resolution over tem-
poral resolution were selected (23). Recent advancements
in MRI equipment have allowed for both good temporal
and spatial resolution (24). In this study, we were able to obtain
contrast-enhanced dynamic breast MRI images with a high
spatial resolution, including the super-early phase, without
changing the conventional parameters related to spatial
resolution.
Nevertheless, the present study has certain limitations. First,
the method for setting the ROI to calculate PE and CR was
not completely objective. Although the normal mammary-
gland tissue regions were selected based on postcontrast images,
the results of the CR could have been affected by inade-
quate selection. Moreover, we did not investigate the imaging
parameters, imaging time, or CM injection ratio, and did not
perform any comparisons with results from other vendors or
devices. Furthermore, although the arrival of the CM at the
aortic arch was considered as the trigger for the start of imaging
for the super-early phase with the BT method, we did not
determine whether this was the most optimal imaging start
time, which would suppress the increase in signal intensity
in BPE while increasing tumor contrast. Furthermore, many
cases of noninvasive cancer do not present with contrast en-
hancement in the early phase (18), and hence, the usefulness
of the super-early phase in diagnosing these cancers requires
further investigation.
CONCLUSIONS
We showed that the super-early phase can be added to the
imaging sequence for contrast-enhanced dynamic breast MRI
1385
TOMIDA ET AL
without deteriorating the spatial resolution. BPE almost never
appeared in the super-early phase, but the CR was lower in
the super-early phase than in the early phase. The BT method
allowed for an optimal imaging start time for the super-
early phase and yielded images with less deviation of contrast
enhancement.
REFERENCES
1. Saslow D, Boetes C, Burke W, et al. American Cancer Society guide-
lines for breast screening with MRI as an adjunct to mammography. CA
Cancer J Clin 2007; 57:75–89.
2. Uematsu T, Yuen S, Kasami M, et al. Comparison of magnetic reso-
nance imaging, multidetector row computed tomography, ultrasonography,
and mammography for tumor extension of breast cancer. Breast Cancer
Res Treat 2008; 112:461–474.
3. Orel SG, Schnall MD. MR imaging of the breast for the detection,
diagnosis, and staging of breast cancer. Radiology 2001; 220:13–30.
4. Heywang-Köbrunner SH, Bick U, Bradley WG, Jr, et al. International
investigation of breast MRI: results of a multicentre study (11 sites)
concerning diagnostic parameters for contrast-enhanced MRI based on
519 histopathologically correlated lesions. Eur Radiol 2001; 11:531–
546.
5. Mann RM, Kuhl CK, Kinkel K, et al. Guidelines from the European Society
of Breast Imaging. Eur Radiol 2008; 18:1307–1318.
6. American College of Radiology. ACR practice parameter for the perfor-
mance of contrast-enhanced Magnetic Resonance Imaging (MRI) of the
breast. Available at: http://www.acr.org/~/media/ACR/Documents/PGTS/
guidelines/MRI_Breast.pdf. Accessed December 7, 2015. 2013. Amended
2014.
7. Nakajima Y, Okuda I, Tozaki M, et al. Guidelines for the optimal use of
breast magnetic resonance imaging (MRI) for screening of breast cancer
in high risk women, version 1.2. J Jpn Assoc Breast Cancer Screen 2013;
22:155–176.
8. Kaiser WA, Zeitler E. MR imaging of the breast: fast imaging se-
quences with and without Gd-DTPA. Preliminary observations. Radiology
1989; 170:681–686.
9. Kuhl CK. MRI of breast tumors. Eur Radiol 2000; 10:46–58.
10. Delille JP, Slanetz PJ, Yeh ED, et al. Physiologic changes in breast mag-
netic resonance imaging during the menstrual cycle: perfusion imaging,
signal enhancement, and influence of the T1 relaxation time of breast
tissue. Breast J 2005; 11:236–241.
1386
Downloaded for Mahasiswa FK UKDW 01 (mhsfkdw01@gmail.com) at Universitas Kristen Duta Wacana from ClinicalKey.com by Elsevier on January 29, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Academic Radiology, Vol 24, No 11, November 2017
11. Kuhl CK, Bieling HB, Gieseke J, et al. Healthy premenopausal breast
parenchyma in dynamic contrast-enhanced MR imaging of the breast:
normal contrast medium enhancement and cyclical-phase dependency.
Radiology 1997; 203:137–144.
12. Müller-Schimpfle M, Ohmenhaüser K, Stoll P, et al. Menstrual cycle and
age: influence on parenchymal contrast medium enhancement in MR
imaging of the breast. Radiology 1997; 203:145–149.
13. Uematsu T, Kasami M, Watanabe J. Does the degree of background en-
hancement in breast MRI affect the detection and staging of breast cancer?
Eur Radiol 2011; 21:2261–2267.
14. Uematsu T, Kasami M, Watanabe J. Background enhancement of
mammary glandular tissue on breast dynamic MRI: imaging features and
effect on assessment of breast cancer extent. Breast Cancer 2012; 19:259–
265.
15. Jansen SA, Fan X, Medved M, et al. Characterizing early contrast uptake
of ductal carcinoma in situ with high temporal resolution dynamic contrast-
enhanced MRI of the breast: a pilot study. Phys Med Biol 2010; 55:N473–
N485.
16. Boetes C, Barentsz JO, Mus RD, et al. MR characterization of suspi-
cious breast lesions with a gadolinium-enhanced TurboFLASH subtraction
technique. Radiology 1994; 193:777–781.
17. Heiberg EV, Perman WH, Herrmann VM, et al. Dynamic sequential 3D
gadolinium-enhanced MRI of the whole breast. Magn Reson Imaging 1996;
14:337–348.
18. Jansen SA, Newstead GM, Abe H, et al. Pure ductal carcinoma in situ:
kinetic and morphologic MR characteristics compared with mammo-
graphic appearance and nuclear grade. Radiology 2007; 245:684–691.
19. Lehman CD. Magnetic resonance imaging in the evaluation of ductal car-
cinoma in situ. J Natl Cancer Inst Monogr 2010; 2010:150–151.
20. Mann RM, Balleyguier C, Baltzer PA, et al. Breast MRI: EUSOBI recom-
mendations for women’s information. Eur Radiol 2015; 25:3669–3678.
21. Jiang L, Zhou Y, Zhou C, et al. Dual-source parallel radiofrequency trans-
mission for magnetic resonance breast imaging at 3T: any added clinical
value? Magn Reson Imaging 2014; 32:523–528.
22. Abe H, Mori N, Tsuchiya K, et al. Kinetic analysis of benign and malig-
nant breast lesions with ultrafast dynamic contrast-enhanced MRI:
comparison with standard kinetic assessment. AJR Am J Roentgenol 2016;
207:1159–1166.
23. Kuhl CK, Schild HH, Morakkabati N. Dynamic bilateral contrast-
enhanced MR imaging of the breast: trade-off between spatial and
temporal resolution. Radiology 2005; 236:789–800.
24. Uematsu T, Kasami M. High-spatial-resolution 3-T breast MRI of
nonmasslike enhancement lesions: an analysis of their features as
significant predictors of malignancy. AJR Am J Roentgenol 2012;
198:1223–1230.