Keywords
Breast imaging - cancer - galactogram - mammography - nipple discharge
Introduction
A conventional X-ray galactogram (CG), also referred to as breast ductogram, is a
specialized X-ray procedure used to view abnormalities in lactiferous ducts. CG has
been traditionally considered the imaging modality of choice in the evaluation of
pathological nipple discharge (PND). However, it has been largely replaced by high-resolution ultrasound (HR-USG)
as the initial imaging modality in patients with nipple discharge as it is semi-invasive
and requires ionizing radiation. However, CG still offers the best spatial resolution
among the existing imaging modalities and has a fairly high sensitivity.
In this article, we demonstrate the usefulness of CG in clinching the diagnosis in
two cases of PND. We also present a brief overview of the literature and compare CG
with other imaging modalities.
Case Reports
Case 1
A 44-year-old woman presented with a 2-week history of intermittent spontaneous painless
discharge from the right nipple. Physical examination did not reveal any mass or axillary
lymphadenopathy. Hemorrhagic discharge from one duct was observed in the right breast
on compression during mammography.
A CG was performed after injecting approximately 3.5 cc of iodinated contrast medium
into the discharging duct, and standard mediolateral oblique (MLO) and craniocaudal
(CC) views were obtained using full-field digital mammography [Figure 1]. The galactogram showed focal-on-diffuse duct irregularity in the upper and outer
quadrants of the right breast along with faint micro filling defects, periductal extravasation,
and areas of irregular ductal narrowing associated with sacculations and microcysts
in the distal subsegmental ducts. Abrupt cutoff of segmental duct was seen immediately
inferior to the first branching point of the main duct. Using the Galactogram Imaging
Classification System (GICS), the findings were classified as GICS 5, indicating a
high suspicion of malignancy. Targeted USG did not reveal any focal lesion or segmental
duct ectasia. Cytology of the nipple discharge showed malignant cells. The patient
was referred to the oncology services for staging and further management.
Figure 1: Magnified craniocaudal view of galactogram of Case 1 showing diffuse duct
irregularities in the upper and outer quadrants of the right breast, with particularly
prominent findings in a small area of peripheral portion of the breast (between horizontal
arrows). Key features include faint microdefects with ductal wall irregularities,
periductal extravasation of contrast material, and areas of ductal narrowing and sacculation
located in distal subsegmental ducts with microcysts. Abrupt cutoff of segmental duct
is seen immediately distal to the first branching point of the main duct without any
filling defect (vertical arrow)
Case 2
A 39-year-old woman presented with a 1-month-long history of persistent nipple discharge
from the left breast. On examination, serosanguinous discharge was expressed from
a single duct. No mass was palpable. Conventional mammography revealed no abnormality
[Figure 2]a. Galactography was performed after injection of approximately 2 cc of nonionic
iodinated contrast media through a 30G plastic cannula inserted in the culprit duct.
CG showed a dilated major duct in the subareolar region with at least two well-defined
filling defects. The smaller filling defect showed clear concave termination and a
subtle extension to one of the subsegmental ducts [Figure 2]b. These findings were consistent with a suspicious pathology (GICS 4). USG showed
an echogenic elongated nodular mass within a dilated subareolar duct. The mass showed
internal vascularity on power Doppler [Figure 2]c. The smaller filling defect was not localized on USG. Duct excision was done and
histopathology revealed multiple duct papillomas.
Figure 2: Mammogram, galactogram, and high-resolution USG of the left breast of a
woman with pathological nipple discharge. (a) CC view of the left breast shows no
obvious abnormality; (b) Galactogram, CC view reveals dilated duct with intraluminal
filling defect in retroareolar tissue in the superior quadrant of the left breast.
Notice the small filling defect extending in the subsegmental duct in superior quadrant
(arrow); (c) USG shows dilated duct with echogenic intraluminal mass in radial (i)
and antiradial (ii) views in the subareolar region. Notice nipple (N). Power Doppler
image shows vascularity within the mass (iii). USG: Ultrasound; CC: Craniocaudal
Discussion
Nipple discharge has a prevalence of about 3%–10% and is the third most common breast-related
complaint after mass and breast pain.[1],[2] PND is defined as persistent, spontaneous, nonlactational, unilateral, and single-duct
discharge.[2] PND is caused most commonly by benign lesions such as duct papillomas. Nearly 6%–20%
cases of PND are due to underlying malignancy such as ductal carcinoma in situ (DCIS) and invasive intraductal carcinoma. Risk of malignancy is higher in women
of age> 50 years and in cases of hemorrhagic discharge.[2],[3] Therefore, proper evaluation of every case of PND is critical.
The imaging modalities that are most commonly used in cases of PND are mammography,
HR-USG, CG, and contrast-enhanced magnetic resonance imaging (CE-MRI). The gold standard
for diagnosis in PND is major duct excision (MDE) or surgical pathology.[4]
Conventional galactogram is still considered the gold standard of ductal imaging by
some authors.[5],[6] CG is an X-ray mammography-based procedure. It is performed by cannulating the abnormal
duct with a thin plastic cannula and injecting 2–5 mL iodinated nonionic contrast
media to opacify the ductal system. After contrast injection, standard MLO and CC
mammograms are taken.
The main indication for performing a galactogram is PND. Findings that are suggestive
of a benign pathology include duct ectasia, macrodefects with concave meniscus, and
cystic changes with ductal communication [Case 2, [Figure 2]. Abrupt or irregular duct obstruction, duct wall irregularity, ductal stenosis and
sacculations, micro filling defects, and contrast extravasation in a woman> 50 years
of age are suggestive of a malignant etiology [2] [Case 1, [Figure 1]. Duct obstruction due to benign pathology is more often restricted to the periareolar
region and involves the main or segmental ducts whereas carcinomas are more commonly
located in the peripheral portions of the breast. GICS classifies CG findings into
five categories in an ascending order of suspicion for malignancy.[7]
CG has moderate negative predictive value ranging from 26% to 63% for diagnosis of
ductal lesions.[4],[8] CG has several disadvantages. It is semi-invasive and involves compression of the
breast tissue and radiation exposure. Failure to cannulate the duct can result in
incomplete study in up to 15% of cases.[1],[2] CG also lacks specificity and has a low positive predictive value in the diagnosis
of ductal lesions.[4] While some authors have demonstrated a statistically significant correlation between
histopathology and CG findings,[2] others have found that CG fares poorly in differentiating between benign and malignant
lesions.[8],[9] Combined use of digital breast tomosynthesis and CG may improve its diagnostic performance.[10]
Galactography plays an important role in the localization of breast neoplasms and
in guiding the choice and extent of surgical therapy.[11],[12] Accurate localization allows focused ductal excision. The use of CG has been shown
to improve the diagnostic yield of surgical biopsy and MDE.[3],[13] CG-guided vacuum-assisted breast biopsy is also a potential diagnostic tool.[14]
Mammogram has a very low sensitivity (20%–25%) for PND.[1] A negative mammogram does not exclude malignant pathology in patients with PND [9] as it can be negative in as many as 94% of cases of PND.[2]
HR-USG is replacing CG as the initial investigation of choice for PND in many institutes.
USG has moderate sensitivity (65%) and moderate-to-high specificity (75%–85%) for
the diagnosis of ductal lesions.[1] However, USG may be negative in up to 80% cases of PND of a malignant etiology.[15] In Case 1, USG failed to localize any lesion, while in Case 2, it failed to localize
the smaller papilloma that was readily seen on CG. Therefore, USG may not be reliable
for excluding malignant causes of nipple discharge. USG is a valuable tool as a “second-look”
modality after a suspicious area is identified on mammography, CG, or MRI. CG is superior
to USG for diagnosis of intraductal pathology.[9]
Contrast-enhanced MR mammography has a high sensitivity for detection of pathology
in cases of nipple discharge; however, it has a high false positive rate.[16] MRI is free of ionizing radiation and offers simultaneous imaging of the ductal
and extraductal pathology in a three-dimensional (3D) view. The patterns of contrast
enhancement on CE-MRI have been shown to correlate with different ductal pathologies.[8] Mass-like enhancement is more commonly seen in duct papillomas whereas segmental
enhancement is more commonly seen with DCIS. Various methods may be employed for ductal
imaging on MRI. Injection of gadolinium-based contrast media into the pathological
duct and obtaining postinjection T1-weighted 3D sequences provides images analogous
to that on CG. However, this technique is still experimental.[16] Another technique employs 3D heavily T2-weighted fat-suppressed technique for imaging
of dilated ducts in a way that is analogous to MR cholangiopancreaticography.[17]
There are very few guidelines for imaging evaluation of PND. The Institute for Clinical
Systems Improvement guidelines recommends the use of mammogram and HR-USG as the initial
imaging modality in patients with PND. In cases that are negative on conventional
imaging studies, CG or CE-MRI may be done.[18] It is pertinent to note that CG does not require any additional setup beyond a standard
mammography machine. On the other hand, the use of breast MR is limited by high cost
and limited availability of equipment and specialized breast coils. The European Society
of Breast Cancer Specialists does not recommend the routine use of MRI in PND. It
suggests that in countries where CG is considered a routine test for PND, MRI should
be considered only if CG fails for technical reasons or if the patient refuses the
procedure.[19]
Conclusion
CG is fairly simple to perform yet is an underused radiological technique in the current
clinical practice. It is especially suitable for resource-poor settings due to poor
access to MRI as CG is superior to both conventional mammogram and HR-USG in the diagnosis
of intraductal lesions. It has the added advantage of superior localization which
is useful for planning of biopsy and surgery. This article underlines the usefulness
of galactogram in clinical practice and highlights the need to bring it back to the
forefront of imaging of PND in all diagnostic mammography departments.
