AJR 2001; 176:1261-1266
© American Roentgen Ray Society
Artifacts and Pitfalls in Sonographic Imaging of the Breast
Jay A. Baker1,
Mary Scott Soo and
Eric L. Rosen
1
All authors: Department of Radiology, Division of Breast Imaging, Box 3808,
Duke University Medical Center, Durham, NC 27710.
Received August 22, 2000;
accepted after revision September 19, 2000.
Address correspondence to J. A. Baker.
Introduction
Sonography is an increasingly important tool to the breast imager both for
evaluation of breast lesions and for guidance of percutaneous interventions.
Although sonography has historically been used for differentiating cystic from
solid breast masses, there has been growing interest in using sonography to
differentiate benign from malignant solid masses
[1,
2] and for screening the
breasts of women with dense tissue
[3].
Sonographic image quality has improved dramatically since its early use for
breast imaging during the 1970s because of advances in technology such as the
use of all-digital high-frequency transducers of up to 13 MHz, color and power
Doppler imaging, and harmonic imaging. Despite these advances, the potential
remains for encountering artifacts or unexpected, spurious findings during
breast sonography.
The artifacts and pitfalls associated with breast sonography are important
to recognize because they may result in the failure to recognize a breast
cancer or prompt unnecessary biopsy of a benign finding. Common sources of
error include the failure to recognize the sonographic appearance of normal
breast anatomy, the improper setting of instrumentation controls,
peculiarities of sonography physics, and the presence of iatrogenic air and
foreign bodies in the breast. To minimize confusion and misdiagnosis, we
review and illustrate characteristic artifacts and pitfalls encountered during
sonographic imaging of the breast.
Normal Anatomy
Rib
Novice breast imagers may mistake a rib for a solid breast mass. In cross
section, the cartilaginous portion of a rib has the appearance of a suspicious
breast mass (Fig. 1A). This
portion of rib can be recognized by its position posterior to pectoralis
muscle and, when imaged longitudinally, by its elongated shape
(Fig. 1B).
View larger version (121K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —Cartilaginous rib mimics solid breast mass on sonography of
normal breast in 46-year-old woman. Sonogram obtained transverse to rib
reveals apparent oval, circumscribed, markedly hypoechoic mass (M) with
posterior acoustic shadowing. Position of this finding posterior to pectoralis
muscle (arrows) confirms actual identity as rib.
|
|
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —Cartilaginous rib mimics solid breast mass on sonography of
normal breast in 46-year-old woman. Sonogram obtained after transducer was
rotated perpendicular to its location in A shows long axis of structure
(r), confirming its identity as rib.
|
|
Nipple
Although the normal nipple is easily recognized at real-time sonography, on
static images an inverted nipple may project beneath the skin of the areola
giving the false appearance of a markedly hypoechoic, solid parenchymal breast
mass with extensive acoustic attenuation
(Fig. 2). Awareness of the
precise location of the transducer relative to the nipple is sufficient to
identify this normal structure. Sonographic imaging of submamillary tissue
should be performed from an angled or coronal projection because of the
intense posterior acoustic shadow produced by the whirled smooth muscle
bundles of the nipple.
View larger version (127K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2. —Inverted nipple mimics suspicious solid breast mass at
sonography of 69-year-old woman. Sonogram shows that inverted nipple projects
beneath skin surface with appearance of ill-defined markedly hypoechoic mass.
Striking posterior acoustic attenuation is present because of dense connective
tissue and smooth muscle bundles comprising nipple.
|
|
Fat Lobule
The normal breast is composed of numerous fat lobules mixed with dense
fibroglandular tissue. At sonography, cross section of a fat lobule can be
mistaken for a solid mass that is isoechoic to surrounding adipose tissue
(Figs.
3A,3B
and 4). The normal appearance
of the structure can be revealed by rotating the transducer until the apparent
mass is observed merging with surrounding normal fat tissue
(Fig. 3B). On occasion, a small
lobule of normal fat tissue may be completely isolated from surrounding
adipose tissue by echogenic fibroglandular tissue
(Fig. 4). This configuration
can be particularly confusing and may infrequently require biopsy to confirm
its benign etiology.
View larger version (131K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A. —Normal fat lobule mimics isoechoic solid mass on breast
sonogram of 44-year-old woman. Sonogram obtained in radial plane reveals
circumscribed isoechoic mass (m) separated from surrounding adipose
tissue.
|
|
View larger version (134K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —Normal fat lobule mimics isoechoic solid mass on breast
sonogram of 44-year-old woman. Sonogram obtained after transducer was rotated
90° into antiradial plane shows mass apparent in A
"merged" with surrounding fat tissue, confirming identity of
structure as fat lobule.
|
|
View larger version (160K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4. —Isolated fat lobule that was mistaken for solid mass on
breast sonogram of 36-year-old woman. Radial projection reveals apparent mass
is isoechoic to nearby fat tissue. Apparent mass failed to merge with
surrounding fat despite rotating transducer. Sonographically guided
percutaneous needle core biopsy confirmed pseudomass as normal breast tissue.
Subsequent follow-up examinations showed no change over 35 months.
|
|
Lactiferous Duct
Between 15 and 20 lactiferous ducts converge at the nipple of the normal
breast. Normal ducts measure up to 3 mm in diameter, whereas ectatic ducts can
dilate significantly. Even with the excellent resolution and contrast
available with modern high-resolution sonography equipment, normal-caliber
ducts imaged in cross section may mimic a small cyst. To avoid
misidentification of tubular or elongated structures, imaging should always be
performed in both radial and antiradial planes.
Acoustic Shadowing
Although invasive breast cancer is usually identified at sonography as a
visible mass, a focus of acoustic attenuation or shadowing without a definable
mass may be the only feature identified
(Fig. 5). This finding is an
important one to recognize and distinguish from several other causes of
isolated acoustic shadowing.
Acoustic shadowing without a mass may be seen as an incidental finding in
the normal breast. A column of shadowing often identifies Cooper's suspensory
ligament—normal suspensory fibrous strands of the breast
[4]. Shadowing due to Cooper's
suspensory ligament is often faint and narrow but can be markedly hypoechoic
and wide enough to simulate a malignant tumor (Fig.
6A,6B).
When Cooper's suspensory ligament are the cause, the shadowing often resolves
with mild increased pressure on the transducer or small changes in the angle
of insonation.
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A. —Acoustic shadowing from Cooper's suspensory ligament mimics
breast neoplasm at sonography of normal breast in 43-year-old woman. Breast
sonogram reveals focus of intense acoustic attenuation without mass lesion.
Acoustic shadowing originates from Cooper's ligament between normal fat
lobules.
|
|
View larger version (124K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B. —Acoustic shadowing from Cooper's suspensory ligament mimics
breast neoplasm at sonography of normal breast in 43-year-old woman. Sonogram
obtained after mild additional pressure was applied to sonography transducer
shows acoustic attenuation has resolved. Cooper's ligament that is causing
attenuation is visible (arrows).
|
|
Poor contact between the transducer and skin can also cause acoustic
shadowing without an underlying mass (Fig.
7A,7B).
Short echogenic parallel lines at fixed intervals starting at the skin and
leading to the acoustic shadowing are pathognomonic for poor skin contact.
Postsurgical scar may also cause substantial acoustic shadowing without an
underlying mass (Fig. 8).
Significant shadowing at the site of prior lumpectomy may be impossible to
differentiate from residual or recurrent tumor, limiting the usefulness of
sonography in this setting.
View larger version (167K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7A. —Poor transducer—skin contact mimics suspicious acoustic
shadowing at sonography in normal breast of 39-year-old woman. Sonogram
reveals marked acoustic shadowing (sh) that appears similar to invasive
carcinomas shown in Figure 5.
Parallel echogenic lines between skin and superficial aspect of shadowing
(arrows) indicate poor contact between transducer and skin as cause
of apparent lesion.
|
|
View larger version (161K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7B. —Poor transducer—skin contact mimics suspicious acoustic
shadowing at sonography in normal breast of 39-year-old woman. Sonogram
obtained after additional acoustic gel was applied to skin reveals that
apparent lesion and parallel echogenic lines have resolved.
|
|
Sonography Physics and Instrumentation
Edge Shadowing
Although acoustic shadowing identified behind a mass is suggestive of
malignancy [5], a thin line of
shadowing seen only behind the peripheral edge of a mass should not be
mistaken for a suspicious feature. This edge shadowing—caused by the
complex combination of absorption and refraction along the mass border
[6]—can be seen with
cysts and benign or malignant solid masses and has no diagnostic
significance.
Reverberation Artifact
Parallel echogenic lines caused by reflection of the ultrasound beam back
and forth between the transducer and tissue interface can give the artifactual
appearance of solid or complex material along the nondependent wall of a cyst
(Fig.
9A,9B).
When characteristic, this artifact is easily recognized. Less characteristic
reverberation can be confirmed by changing the angle of insonation.
Reverberation is always perpendicular to the direction of the ultrasound beam
and will appear to shift position within the cyst as the angle of insonation
is changed.
View larger version (164K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9A. —Reverberation artifact in simple cyst on sonography of normal
breast in 63-year-old woman. Sonogram reveals curvilinear echogenic lines
parallel anterior wall of simple cyst (arrows). Lines result from
multiple reflections of acoustic beam caused by impedance mismatch between
breast tissue and cyst. Simple cyst may be mistaken for complex cyst because
of apparent internal echoes.
|
|
View larger version (177K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9B. —Reverberation artifact in simple cyst on sonography of normal
breast in 63-year-old woman. Sonogram obtained after angle of insonation was
changed confirms finding is reverberation artifact. Curvilinear lines are now
absent at position (arrowhead) where they are visible in
A.
|
|
Gray-Scale Gain
The gray-scale gain setting determines the amplitude of the returning
sonographic signal. If the gain is set inappropriately high, spurious echoes
may be displayed in a simple cyst resulting in the appearance of a complex
cyst or solid mass (Fig.
10A,10B).
The surrounding tissue also appears inappropriately echogenic. The appropriate
gray-scale gain setting—measured in decibels—varies depending on
the tissue type and sonography equipment. This parameter should be set so that
fluid in a sonography cyst phantom appears black, whereas fat lobules in the
breast parenchyma vary from dark gray to light gray.
View larger version (177K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10A. —Artifactual echoes in simple cyst caused by improper setting
of gray-scale gain on sonography of normal breast in 38-year-old woman.
Sonogram shows breast cyst appears complex with extensive internal echoes.
Surrounding normal tissue is inappropriately echogenic. Gray-scale gain
setting is 42 db.
|
|
View larger version (127K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10B. —Artifactual echoes in simple cyst caused by improper setting
of gray-scale gain on sonography of normal breast in 38-year-old woman.
Sonographic image of same cyst as that shown in A obtained with
gray-scale gain set at 14 db shows that, with appropriate gain setting, cyst
is anechoic with well-circumscribed margins and increased
through-transmission. Surrounding normal tissue varies from light to dark
gray.
|
|
Dynamic Range
Dynamic range settings determine the range of echo amplitudes detected by
the sonography system; echo amplitudes are displayed using the available
number of gray levels of the video monitor
[7]. Dynamic range settings
directly affect contrast resolution and function similarly to the window
portion of window level settings on CT scans.
Setting the dynamic range too low increases image contrast but may cause
the low-level echoes in a solid mass to be displayed as black pixels so that
the appearance of the mass mimics that of a simple cyst (Fig.
11A,11B).
Setting the dynamic range too high results in an image with little contrast,
which hinders differentiation of fat lobules from subtle masses.
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11A. —Fibroadenoma mimics simple cyst as result of inappropriate
setting of dynamic range in 41-year-old woman. Sonogram obtained with dynamic
range set at 40 db shows solid breast mass. Mass appears anechoic with
questionable increased posterior acoustic sound transmission and appears
similar to simple cyst. Surrounding tissue is conspicuously high contrast.
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11B. —Fibroadenoma mimics simple cyst as result of inappropriate
setting of dynamic range in 41-year-old woman. Sonogram obtained with dynamic
range set at 55 db of same breast mass as shown in A clearly reveals
solid nature of mass. Surrounding tissue has appropriate contrast.
|
|
Focal Zone
The width of the ultrasound beam is narrowest and resolution greatest in a
narrow range—the focal zone—that can be set to a particular depth
by the user. The focal zone should be set to match the depth of the object
being imaged. Because resolution and beam width deteriorate outside the focal
zone, imaging with an inappropriately positioned focal zone may cause a subtle
mass to be less visible, sharp edges to appear ill-defined, and an anechoic
simple cyst to appear to have internal echoes due to increased partial volume
effects (Fig.
12A,12B).
View larger version (161K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12A. —Sonograms of simple breast cyst in 59-year-old woman were
degraded by inappropriate placement of focal zone. Sonogram obtained with
focal zone inappropriately placed at bottom of field of view (arrow)
reveals small ill-defined hypoechoic mass with normal through-transmission
suggestive of small complex cyst or, less likely, solid mass. Definite echoes
are visible within mass.
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12B. —Sonograms of simple breast cyst in 59-year-old woman were
degraded by inappropriate placement of focal zone. Sonogram obtained with
appropriate positioning of focal zone at depth of mass (arrow)
reveals characteristic simple cyst. Mass is completely anechoic with sharp
margins and increased through-transmission of acoustic beam.
|
|
Air and Foreign Bodies
Foreign Bodies
Foreign bodies in the breast may have a variety of appearances at breast
sonography. Solid foreign bodies such as a retained Dacron (DuPont,
Wilmington, DE) cuff from a central venous catheter may look like a solid mass
with remarkably intense and sharp posterior acoustic shadowing (Fig.
13A,13B).
The presence of a foreign body at sonography is easily confirmed by
correlation with mammograms.
View larger version (132K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13A. —Central venous catheter Dacron (DuPont, Wilmington, DE)
retention cuff mimics solid breast mass at sonography of 59-year-old woman.
Sonogram shows focus of intense posterior acoustic attenuation similar to that
of invasive carcinoma shown in Figure
5. Echogenic focus anterior to acoustic shadow (arrows)
represents fibrosis adjacent to cuff. Hard palpable mass corresponded to
sonographic abnormality.
|
|
View larger version (79K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13B. —Central venous catheter Dacron (DuPont, Wilmington, DE)
retention cuff mimics solid breast mass at sonography of 59-year-old woman.
Mediolateral oblique mammogram confirms that sonographic and palpable
abnormality correspond to retained Dacron retention cuff overlying pectoralis
muscle (arrow).
|
|
A small amount of extracapsular silicone due to implant rupture may
initially be mistaken for a cyst or hypoechoic solid mass with posterior
acoustic enhancement (Fig.
14). However, recognition of short echogenic lines paralleling the
back wall of the structure allow a confident diagnosis of extracapsular
silicone and, therefore, implant rupture.
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 14. —Extracapsular silicone deposit in 44-year-old woman. Sonogram
reveals small solid circumscribed mass near silicone breast implant.
Curvilinear echogenic lines paralleling posterior margin (arrow)
confirm identity as extracapsular silicone.
|
|
Air
Air introduced into the breast parenchyma during percutaneous needle
procedures can cause several pitfalls during sonography. Considerable air may
be introduced along the path of the needle, particularly during
sonographically guided needle core biopsy. This air may substantially obscure
visualization of the needle. In fact, the needle may be so severely obscured
that it must be withdrawn and reinserted at a second site away from the
obscuring air to allow simultaneous visualization of the mass and needle
during tissue sampling. The introduction of air can be limited by purging air
from all syringes and needles before local anesthesia is administered.
Air persisting along a biopsy tract after removal of the biopsy needle may
look like a thin echogenic line. This echogenic line may closely mimic the
appearance of a needle remaining in the mass
(Fig. 15). When the needle is
reinserted for subsequent sampling, it may be difficult to differentiate a
prior needle tract from an actual needle. Rapid back-and-forth motion of the
needle tip may allow differentiation of the two.
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 15. —Air tract mimics in situ needle during sonographically guided
needle core breast biopsy. Sonographic image shows needle traversing breast
mass (M) has appearance of linear echogenic focus (arrows). However,
needle was removed from breast before image was obtained as indicated by
"needle out" notation at bottom of image. Echogenic line results
from residual air in needle tract.
|
|
Conclusion
To avoid overlooking signs of breast cancer while preventing unnecessary
biopsies, it is important to recognize the artifacts and pitfalls commonly
encountered at breast sonography. Knowledge of normal breast anatomy, an
understanding of the appropriate setting of instrument controls, and the
performance of routine imaging in two orthogonal planes are essential for
accurate interpretation of breast sonograms. Because the true cause of
pseudolesions is often more evident at real-time imaging than on static films,
it is often important for the breast imaging radiologist to actually perform
the examination in real time or, at least, to recognize the sometimes subtle
clues that identify a finding as an artifact. Finally, evaluating available
mammograms in association with the sonographic images is essential to accurate
breast sonography.
References
-
Stavros AT, Thickman D, Rapp CL, et al. Solid breast nodules: use
of sonography to distinguish between benign and malignant lesions.
Radiology
1995;196:123
-134[Abstract/Free Full Text]
-
Rahbar G, Sie AC, Hansen GC, et al. Benign versus malignant solid
breast masses: US differentiation. Radiology
1999;213:889
-894[Abstract/Free Full Text]
-
Kohl TM, Lichy J, Newhouse JH. Occult cancer in women with dense
breasts: detection with screening US—diagnostic yield and tumor
characteristics. Radiology
1998;207:191
-199[Abstract/Free Full Text]
-
Lanfranchi ME. Breast ultrasound. Madrid:
Marban Books, 2000:219
-
Kopans DB. Breast imaging, 2nd ed.
Philadelphia: Lippincott-Raven, 1998:438
-
Hedrick WR, Hykes DL, Starchman DE. Ultrasound physics
and instrumentation, 3rd ed. St. Louis: Mosby,
1995: 318
-
Hedrick WR, Hykes DL, Starchman DE. Ultrasound physics
and instrumentation, 3rd ed. St. Louis: Mosby,
1995: 66-68
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
M. Costantini, P. Belli, R. Lombardi, G. Franceschini, A. Mule, and L. Bonomo
Characterization of solid breast masses: use of the sonographic breast imaging reporting and data system lexicon.
J. Ultrasound Med.,
May 1, 2006;
25(5):
649 - 659.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Introduction
Normal Anatomy
