AJR 2005; 184:S49-S52
© American Roentgen Ray Society
In Vivo Sonography Through an Open MRI Breast Coil to Correlate Sonographic and MRI Findings
Joshua Lilienstein,
Bruce L. Daniel and
Debra M. Ikeda
Department of Radiology, Stanford University Medical Center, 875 Blake
Wilbur Dr., Stanford, CA 94305-5826.
Received April 9, 2004;
accepted after revision May 4, 2004.
Address correspondence to D.M. Ikeda
(dikeda{at}stanford.edu).
Introduction
Correlation of mammographic and sonographic findings to those of MRI
may be challenging because of changes in the breast configuration during each
examination: The patient is upright for mammography, supine for sonography,
and prone for MRI. Accurate correlation is especially important if a
nonpalpable finding on MRI prompts a biopsy and appears to correlate to a
finding on a second technique. Piron et al.
[1] demonstrated a technique of
correlating sonography and MRI findings in a breast phantom using a
fenestrated plate and an ultrasound membrane for compression. We describe an
in vivo method for sonographic scanning of an uncompressed breast through an
open MRI coil to correlate sonographic and MRI findings without special
equipment. We confirmed that a mass seen on MRI was a benign lymph node using
MRI-guided sonography.
Case Report
A nonpalpable spiculated mass was present in the lower left breast of a
middle-aged woman. It was initially found on mammography and was shown to be
an invasive lobular cancer on a subsequent sonographically guided biopsy. The
diagnostic mammogram also showed an incidental 8-mm mass in the upper outer
breast near a blood vessel that was thought to be a lymph node because of the
presence of a fatty hilum. Sonography in this location showed a typical lymph
node with a fatty hilum.
To stage the known cancer, we performed contrast-enhanced MRI on a 1.5-T
scanner (Signa, GE Healthcare) as previously described
[2], including a T1-weighted
axial sequence (TR/TE, 300/16) with a 5-mm thickness to evaluate for
adenopathy. A comprehensive diagnostic breast MRI examination
[3] was performed, beginning
with a dynamic scanning sequence consisting of rapid 3D volume spiral MRI
during administration of an IV gado-pentetate dimeglumine (Magnevist, Berlex
Laboratories) bolus at a dose of 0.1 mmol/kg to characterize initial contrast
enhancement, followed by a contrast-enhanced high-resolution 3D acquisition
(3D acquisition, spectral-spatial radiofrequency pulses, on-resonance
magnetization transfer) to characterize the lesion morphology on thin sections
[2]. Finally, a second dynamic
spiral sequence was used to record contrast washout from abnormally enhancing
lesions.
On contrast-enhanced MRI, the known invasive lobular cancer appeared
irregular in shape in the lower breast with a rapid initial enhancement and a
late plateau. MRI also revealed an 8-mm oval smooth upper breast mass near a
blood vessel with rapid initial enhancement and late washout on the dynamic
scans, possibly representing a lymph node. Because MRI did not conclusively
show a fatty hilum, the resulting differential diagnosis was a second
carcinoma [4]. When correlated
to the diagnostic mammogram, this finding was thought to correspond to the
lymph node seen on the diagnostic mammogram and sonogram. It was thought that
the mass found on MRI was a lymph node because it correlated in size,
location, and surrounding tissue with the mammography and sonography. However,
an absolute correlation was requested because invasive lobular cancer had been
diagnosed in the lower breast, and invasive lobular cancer is more difficult
to diagnose on MRI than invasive ductal cancer.
To correlate the sonography and MRI findings, we placed the patient prone
in a vertically open 0.5-T MRI scanner (Signa-FP, GE Healthcare) and
positioned in a dedicated open breast coil. Using methods previously described
for MRI-guided preoperative freehand needle localization
[5], areas of architecture on
the axial T1-weighted fast spin-echo images were used to show the location of
the upper breast mass and were correlated directly to the previous diagnostic
axial T1-weighted and reformatted contrast-enhanced images
(Fig. 1B). While the patient
remained in the coil, a fiducial marker was placed on the skin at the location
that would have been selected for needle placement had the mass been targeted
for biopsy (Fig. 2A). A repeat
unenhanced axial T1-weighted fast spin-echo image confirmed the location of
the fiducial marker relative to the expected location of the mass
(Fig. 1C).
View larger version (117K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —Middle-aged woman with a biopsy-proven invasive lobular
carcinoma. Reformatted axial contrast-enhanced 3DSSMT image shows enhancing
mass in A near blood vessel and close to pectoralis muscle (solid
arrow). Another enhancing focus in mid breast had benign dynamic
enhancing characteristics and was thought to be benign (open
arrow).
|
|
With the patient remaining motionless and prone in the breast coil, the
patient, table, and coil were moved together out of the magnet room into an
anteroom. The location of the fiducial marker was noted, marked on the skin
with ink, and removed (Figs. 2A
and 2B). A sonography scanner
was brought in, and a 7.5-MHz linear array transducer was placed through the
open breast coil to scan directly over the marked location in the axial plane
(Fig. 2C). The breast was
stabilized by a hand placed on its medial side, directly opposite the
transducer as needed. Sonography showed a normal intramammary lymph node, with
a hypoechoic cortex and an echogenic fatty hilum. The lymph node was at the
expected depth and was the same size as the MRI-detected finding, and was at
the border of fat and strands of glandular tissue as seen on MRI. It was
adjacent to a blood vessel, also seen on the MR image and mammogram, showing
that the sonographic, MRI, and mammographic findings were the same, excluding
the presence of a second cancer.
View larger version (84K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —Photographs of model showing technique of sonography/MRI
correlation in an open MRI coil. Fiducial marker is removed, and model's skin
is marked with "X" over expected location of mass.
|
|
View larger version (85K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C. —Photographs of model showing technique of sonography/MRI
correlation in an open MRI coil. Sonography is performed through open MRI coil
in axial plane. In this case, transducer is oriented along axial plane,
although it is possible to scan along coronal plane.
|
|
Excisional biopsy of the invasive lobular cancer and an axillary sentinel
lymph node dissection were performed. Six months later, follow-up mammography
and sonography again showed the lymph node in the upper breast. No other
masses were identified.
Discussion
We have described a successful in vivo application of a technique using
direct sonographic scanning of the breast in the axial plane through an open
MRI breast coil to correlate MRI and sonographic findings. Because the patient
remains motionless in the prone position within the breast coil, this
technique allows direct correlation of MRI and sonographic findings by
eliminating error caused by changes in breast orientation. In our case, we
were fairly certain that the findings correlated to a lymph node on all
imaging studies and performed this study because absolute correlation was
necessary.
Non-MRI-guided sonography has been shown to have a low (23%) rate of
detecting MRI-identified lesions. In addition, the frequency of cancer
diagnosis after biopsy is significantly higher in lesions visible on MRI and
sonography than on MRI alone (43% vs 14%)
[6], highlighting the need for
accurate MRI-guided sonography when correlation of MRI and sonographic
findings is unclear.
View larger version (176K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —Middle-aged woman with a biopsy-proven invasive lobular
carcinoma. Contrast-enhanced image from 3D acquisition spectral-spatial
radiofrequency pulse on-resonance magnetization transfer (3DSSMT) shows 8-mm
enhancing mass in upper left breast near blood vessel (solid arrow).
Invasive lobular carcinoma is seen in lower left breast (open arrow).
Upper breast mass showed rapid initial enhancement and late washout (not
shown), resulting in differential diagnosis of lymph node or second
carcinoma.
|
|
In our facility, MRI-guided preoperative needle localization of suspicious
findings is always followed by preoperative craniocaudal and mediolateral
mammograms that are marked for the surgeon. This is especially important for
MRI findings thought to correlate to mammographic or sonographic findings,
because the post-MRI-localization mammogram or sonogram occasionally yields
unexpected results. In rare cases, the sonographic or mammographic finding
does not correlate to findings at the tip of the wire deployed under MRI
guidance, and a second needle or hookwire must be placed into the sonographic
or mammographic finding by conventional methods. A more worrisome scenario
would arise if an MRI finding was mistakenly thought to correlate to a
sonographic finding and was excised after a songraphically guided preoperative
needle localization, leaving the MRI-detected finding in the breast. Our
technique of direct correlation of MRI and sonography by scanning through the
open coil prevents such occurrences. However, this technique does not
eliminate the need for MRI-guided biopsy, because some MRI-detected lesions
may not be detected on sonography.
Another way to correlate mammographic, sonographic, and MRI findings before
excisional biopsy would be to localize the sonographic or mammographic finding
with an MRI-compatible wire and to perform contrast-enhanced breast MRI after
localization. The problem with this scenario is in scheduling the
postlocalization MRI to include enough time and personnel for an add-on
MRI-guided preoperative needle localization if the conventionally localized
findings do not correlate to the suspicious abnormally enhancing
lesion(s).
Our technique is an improvement of that described by Obdeijn et al.
[7], in which sonography was
performed after an abnormal MRI finding. Our technique performs both
examinations without changing the orientation of the breast within the breast
coil, allowing direct correlation of the findings.
A technique to correlate mammographic and sonographic findings was
described by Meyer et al. [8].
They placed a suspect lesion in the aperture of a fenestrated mammographic
compression plate used for preoperative needle localization and performed
sonography though the aperture at a location shown on the mammogram. Piron et
al. [1] are developing an
analogous technique for use in correlating imaging techniques during tissue
biopsy, because some fenestrated compression plates designed for MRI-guided
preoperative needle localization or percutaneous biopsy have apertures wide
enough to accommodate the ultrasound transducer.
In contrast to that described by Piron et al.
[1], our technique does not use
a fenestrated compression plate to hold the breast but instead scans the
uncompressed breast through the open MRI breast coil. The advantages of this
method are that no new equipment or materials are needed and that the lesions
that might be blocked by the fenestrated plate are more easily accessed. A
limitation shared by both methods is the inability to visualize lesions high
in the axilla or close to the chest wall because the breast coil housing
blocks access to these portions of the breast. The ultrasound beam might
access some of these difficult lesions by the operator's angling of the
transducer around the breast coil housing toward the lesion. A second
limitation is evaluation of lesions near the nipple, where transducer contact
may be limited by the small amount of breast tissue located anteriorly.
In summary, we have described an in vivo technique for correlating
sonographic and MRI findings when results of traditional visualization
techniques are uncertain. The need for this technique may be rare and
performing sonographic scanning through the MRI coil is costly because of use
of magnet time. However, when the correlation of the findings of MRI and other
imaging techniques are uncertain, direct correlation is less invasive and less
costly than performing a biopsy when the mass is benign, as in this case. When
the mass is malignant, this technique may help avoid miscorrelation of MRI and
sonographic findings.
References
- Piron CA, Causer P, Jong R, Shumak R, Plewes DB. A hybrid breast
biopsy system combining ultrasound and MRI. IEEE Trans Med
Imaging 2003;22:1100
-1110[Medline]
- Daniel BL, Yen YF, Glover GH, et al. Breast disease: dynamic spiral
MR imaging. Radiology1998; 209:499
-509[Abstract/Free Full Text]
- Agoston AT, Daniel BL, Herfkens RJ, et al. Intensity-modulated
parametric mapping for simultaneous display of rapid dynamic and
high-spatial-resolution breast MR imaging data.
RadioGraphics2001; 21:217
-226[Abstract/Free Full Text]
- Kinoshita T, Yashiro N, Yoshigi J, Ihara N, Fukuma E, Narita M.
Inflammatory intramammary lymph node mimicking the malignant lesion in dynamic
MRI: a case report. Clin Imaging2002; 26:258
-262[Medline]
- Daniel BL, Birdwell RL, Ikeda DM, et al. Breast lesion
localization: a freehand, interactive MR imaging-guided technique.
Radiology1998; 207:455
-463[Abstract/Free Full Text]
- LaTrenta LR, Menell JH, Morris EA, Abramson AF, Dershaw DD,
Liberman L. Breast lesions detected with MR imaging: utility and
histopathologic importance of identification with US.
Radiology2003; 227:856
-861[Abstract/Free Full Text]
- Obdeijn IM, Brouwers-Kuyper EM, Tilanus-Linthorst MM, Wiggers T,
Oudkerk M. MR imaging-guided sonography followed by fine-needle aspiration
cytology in occult carcinoma of the breast. AJR2000; 174:1079
-1084[Abstract/Free Full Text]
- Meyer JE, Greenes RA, Sonnenfeld MR. Breast immobilization for
occult mass aspiration. Radiology1988; 169:266[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Introduction
Case Report
