Other
Breast Imaging
April 2002

Sonographically Guided Biopsy of Suspicious Microcalcifications of the Breast: A Pilot Study

Abstract

OBJECTIVE. The purpose of this study is to evaluate the use of sonographic guidance for biopsy of mammographically detected suspicious microcalcifications.
SUBJECTS AND METHODS. Twenty-three patients with suspicious microcalcifications detected on mammography (15 associated with masses or distortion; eight with microcalcifications alone) underwent sonographically guided core biopsy (n = 18) or sonographically guided needle localization before excision (n = 5). Microcalcifications were targeted, and specimen radiographs were obtained for each lesion, with the success of the procedure based on identifying microcalcifications on the specimen radiograph. For core biopsies, the number of cores obtained was compared with that in 49 control patients who underwent sonographically guided core biopsy of noncalcified masses.
RESULTS. All 23 lesions (100%) were successfully biopsied under sonographic guidance, with microcalcifications seen on specimen radiographs in each case. Of 18 core biopsies, a mean of 8.7 cores was obtained compared with a mean of 5.5 cores in the control group (p<0.0001). Of 13 lesions sampled with core biopsy that subsequently underwent surgical excision, three (23%) were upgraded from atypical ductal hyperplasia to ductal carcinoma in situ (n = 1) and from ductal carcinoma in situ to invasive carcinoma (n = 2). Mammographically, most lesions contained more than 15 pleomorphic microcalcifications. On sonography, echogenic foci corresponded to microcalcifications in all but two cases in which broader echogenic regions were seen. When no mass or distortion was visible on mammography, sonography showed a mass or dilated ducts with internal echogenic foci.
CONCLUSION. Microcalcifications identifiable on sonography can be successfully biopsied under sonographic guidance. Further study is necessary to determine whether targeting microcalcifications seen sonographically in the mass or duct can improve the rate of underestimation of disease compared with stereotactic core biopsy.

Introduction

Sonography plays an important role in characterizing breast masses and in guiding needle core biopsies and wire localizations of suspicious sonographic abnormalities [1, 2]. However, sonography has not traditionally been used for detection of microcalcifications, and the role of sonography for this use has not been established [3]. In the mid to late 1990s, the advent of high-frequency transducers, increased computing power, and digital signal processing resulted in improved sonographic resolution and contrast. Using this state-of-the-art technology, radiologists are now identifying microcalcifications more frequently on sonography [4,5,6]. These microcalcifications are usually identified within hypoechoic masses that facilitate detection of the echogenic microcalcifications [6].
Although screen-film mammography remains the imaging modality of choice for detecting and characterizing microcalcifications [3], identifying suspicious microcalcifications on sonography now allows the possibility of sonographically guided biopsy of these lesions. The advantages of sonographically guided procedures over mammographically guided procedures include lack of ionizing radiation, lower cost, and improved patient comfort with supine positioning and lack of breast compression [7]. Other advantages for radiologists experienced in sonographically guided biopsy include the ease and speed of performance of the procedure compared with stereotactic guidance. Fine-needle aspiration and skin marking with an indelible marker before surgery have been attempted with sonographic guidance to evaluate microcalcifications [8,9,10]. Most recently, Teh et al. [11] reported a 78% success rate with core biopsy of microcalcifications using high-frequency and power Doppler sonography. The purpose of this study is to evaluate the use of sonographic guidance for core biopsy and needle localization of suspicious microcalcifications, with a review of imaging features and histologic results.

Subjects and Methods

Between May 1998 and April 2000, we undertook a prospective evaluation of mammographically suspicious lesions containing clustered microcalcifications in an attempt to identify and biopsy microcalcifications on sonography. The lesions identified during this time served as pilot data for an ongoing institutional review board—approved study of microcalcification detection and biopsy on sonography. In this pilot study group, we selected suspicious mammographic lesions that had a high likelihood of being identified on sonography to gain experience in identifying microcalcifications on sonography and to determine the feasibility of performing sonographically guided biopsy. Two groups of lesions were evaluated: microcalcifications within an associated mass or area of architectural distortion; and large clusters of microcalcifications with a diameter of 1 cm or greater containing numerous pleomorphic microcalcifications, without an associated mammographic abnormality other than fibroglandular density. Smaller diameter clusters containing faint, very small amorphous or punctate microcalcifications were not initially targeted in this pilot study, although at the end of the study period, one cluster smaller than 1 cm was evaluated. Each lesion was considered suspicious for or highly suggestive of malignancy (Breast Imaging Reporting and Data System [BI-RADS] [12] category 4 or 5) on the basis of the mammographic appearance. At the time of diagnosis, each patient was notified of the suspicious lesion that required biopsy and consented to undergo sonographic evaluation of the suspicious area to determine whether the lesion could be biopsied under sonographic guidance.
Overall, 38 lesions were initially evaluated with sonography. Seven lesions were excluded because they showed no corresponding sonographic abnormality. On mammography, these seven lesions were seen as clustered (n = 6) or segmental (n = 1) microcalcifications without an associated mass or architectural distortion. The grouped microcalcifications were pleomorphic (n = 6) or punctate (n = 1) in morphology and ranged from 10 to 50 mm. Thirty-one lesions had a possible (n = 4) or definite (n = 27) corresponding sonographic abnormality. The four lesions with possible corresponding sonographic abnormalities were excluded from the study because the radiologist performing the biopsy was uncertain whether the finding was an actual lesion or whether the lesion correlated with the mammographic lesion. For example, in one case, only one microcalcification was suspected to be present on sonography without a corresponding mass, and in another, a hypoechoic region without definite microcalcifications was seen. In these cases, the radiologist elected to perform stereotactic biopsy, and in one of the four cases, sonographically guided biopsy of the hypoechoic region was also performed. Sonographically guided biopsy was performed in the remaining 27 lesions and was shown to correlate with the mammographic abnormality. In one patient with two lesions (a 4-cm highly suspicious mass with microcalcifications and a satellite density with microcalcifications), both lesions were targeted with separate wires during the needle localization procedure under sonographic guidance, and each was seen within a single large specimen on the specimen radiograph. However, the smaller lesion was not described as a separate finding on the pathology report and therefore was excluded. Three of the remaining 26 lesions were seen as either a hypoechoic mass or a hypoechoic region of shadowing, but no microcalcifications were seen in the lesions; therefore, these were excluded. In the remaining 23 patients, regions of distinct microcalcifications were visible as specific targets for sonographically guided biopsy, and these 23 lesions constitute the study population.
Conventional screen-film mammography was performed using commercially available equipment (Mammomat II, Mammomat III, and Mammomat 3000; Siemens, Iselin, NJ). Standard craniocaudal and mediolateral oblique films were reviewed in all cases. Spot compression magnification images of the regions of microcalcifications were also available for review in 21 (91%) of 23 patients. Two lesions without magnification images were initially evaluated at outside institutions, and the magnification film from the diagnostic workup was unavailable at the time of the film review. Sonograms were obtained by the attending radiologist in all cases using commercially available high-resolution equipment (Elegra; Siemens, Issaquah, WA). A 9-MHz linear array transducer was used for the first six patients; then a 13-MHz 1.5-dimensional linear array transducer became available and was used in the remaining 17 patients.
Tissue sampling was performed under sonographic guidance using the free-hand technique in each case. Fifteen clusters of microcalcifications associated with a mass or architectural distortion (group 1) underwent either percutaneous core biopsy (n = 12) or needle localization with surgical excision (n = 3). Eight suspicious clusters of microcalcifications shown on mammography without an associated mammographic lesion (group 2) underwent either core biopsy (n = 6) or needle localization with surgical excision (n = 2).
For both the needle localization procedures and the core biopsy procedures, informed written consent was obtained. The patient was placed supine on the table with the ipsilateral arm extended above the head. If the lesion was located in the lateral breast, the patient was placed in an oblique position, allowing easier access to the lesion. The lesion was scanned to determine the optimal approach, and the skin was marked to denote the transducer position and needle entry site. The breast was then prepared and draped in a sterile fashion, and local anesthesia was administered at the planned needle entry site.
For sonographically guided needle localizations before surgical excision, a 21-gauge needle (Kopans spring hook localization needle; Cook, Bloomington, IN) was inserted into the region of microcalcifications under sonographic guidance using the free-hand technique. The hooked wire was then inserted through the needle and the needle removed. After the needle localization procedure and before surgery, sonographic images were taken to document the positioning of the lesion and wire, and mediolateral and craniocaudal mammograms were obtained for the surgeon's reference and to assure proper placement of the wire. Radiographs of the surgical specimen were also taken to confirm the presence of targeted microcalcifications. Accuracy of wire placement in relation to the microcalcifications was assessed on postprocedure mammograms and specimen radiographs.
For the 18 lesions sampled with core biopsy, microcalcifications were targeted, and the identification of microcalcifications on the specimen radiographs was considered the end of the biopsy. The multipass automated gun technique with a 14-gauge needle (Bard, Covington, GA) was used in 16 (89%) of 18 core biopsies, and two lesions (11%) were sampled with a hand-held 11-gauge vacuum-assisted technique (Mammotome; Biopsys/Ethicon Endo-Surgery, Cincinnti, OH). When the multipass technique was used, the needle was inserted into the breast until the tip was just proximal to the edge of the mass or duct (Fig. 1A,1B,1C,1D). The needle was aligned so that microcalcifications were in the path of the needle as it was fired. For the vacuum-assisted technique, the probe was placed below one small mass and within a 3.5-cm larger mass, positioned so that the calcifications were located superficially to the tissue-cutting notch of the probe.
Fig. 1A. 65-year-old woman with breast carcinoma. Spot compression magnification mammogram of left breast reveals new cluster of six microcalcifications (arrows).
Fig. 1B. 65-year-old woman with breast carcinoma. Sonogram in radial orientation of clustered microcalcifications shows apparent lobulated duct (arrows) in region of mammographic abnormality.
Fig. 1C. 65-year-old woman with breast carcinoma. Sonogram in antiradial orientation reveals duct (arrowheads) in cross-section with echogenic specular reflector (arrow) corresponding to microcalcification.
Fig. 1D. 65-year-old woman with breast carcinoma. Sonographically guided core biopsy with needle tip located just proximal to abnormal duct (arrowheads) and aligned to target single microcalcification (arrow). Lesion was diagnosed as atypical hyperplasia at core biopsy, which was upgraded to ductal carcinoma in situ at excisional biopsy.
Using either technique, the radiologists initially obtained five to seven cores of tissue and then radiographed the specimens together to determine whether microcalcifications were present. Additional cores were obtained and an additional specimen radiograph was obtained if microcalcifications were not identified on the initial specimen radiograph. Marker clips were not deployed during sonographic core biopsies performed during this study. The mean number of cores obtained and the complication rates were recorded. These core biopsy data were compared with data from a control group of 49 consecutive sonographically guided biopsies of noncalcified lesions performed over 4 months of the study period. A Wilcoxon—Mann—Whitney test was used to compare the number of cores retrieved in the study and control groups.
The mammographic and sonographic images were retrospectively reviewed to evaluate the features of the two groups of clustered microcalcifications amenable to sonographically guided biopsy. Mammographic images alone were evaluated initially and assigned a BI-RADS category (4 or 5) independently by two breast imaging radiologists. Mammographic microcalcifications were categorized according to lesion size, distribution, associated findings, number of microcalcifications, and microcalcification morphology using BI-RADS descriptors. Sonograms were subsequently reviewed in conjunction with the mammograms to determine whether the level of suspicion for malignancy changed when the sonographic findings were included.
The imaging findings were correlated with the histologic results from the biopsy. Concordance between the imaging findings and the histologic results was ascertained at weekly imaging—histologic review sessions. Any lesions with a histologic result that included atypical cells after core biopsy were recommended for surgical excision. For all patients undergoing core biopsy followed by surgical excision, the histology from the core biopsy was reviewed and compared with the final pathology after surgical excision. The number of lesions upgraded from atypical ductal hyperplasia to carcinoma and from ductal carcinoma in situ (DCIS) to invasive ductal carcinoma was determined.

Results

Biopsy Results

Sonographically guided needle localizations.— Sonographically guided needle localizations were successful for localizing microcalcifications before surgical excision in all (5/5) cases attempted, as confirmed on postprocedure mammograms. The wire was noted to traverse the cluster of microcalcifications in each case, and specimen radiographs confirmed the presence of microcalcifications in the specimens (Fig. 2A,2B). The imaging and histologic findings of all lesions were considered concordant on review. Microcalcifications were seen in association with carcinoma in all four lesions consisting of invasive ductal carcinoma and DCIS, and microcalcifications were associated with a radial scar in the one benign lesion.
Fig. 2A. 69-year-old woman with benign breast mass. Magnification mammogram of biopsy specimen with localizing wire shows irregular multilobulated mass with innumerable microcalcifications.
Fig. 2B. 69-year-old woman with benign breast mass. Sonogram of multilobulated mass in A before excision shows three echogenic, shadowing components of mass (arrows) that correspond to groups of innumerable microcalcifications contained in different lobulated components of mass. This lesion was localized under sonographic guidance, and excisional biopsy yielded radial scar.
Sonographically guided core biopsy.— Sonographically guided needle core biopsy attempted in 18 lesions was successfully completed in all cases. A mean of 8.7 cores was obtained (range, 5-15 cores), compared with a mean of 5.5 cores (range, 1-11 cores) in the control group of 49 patients (p < 0.0001). Specimen radiographs confirmed the presence of microcalcifications in all cases, although fewer microcalcifications were present on the specimen radiograph than were seen in the lesion on the original mammograms, suggesting that residual microcalcifications would be present in the breast. In 14 of 18 cases, postprocedure mammograms were available for review, and all 14 revealed residual microcalcifications at the site of the lesion, including the lesion that initially contained only six microcalcifications. No substantial complications were reported in either the study or the control group. The imaging and histologic findings were considered concordant in all cases except one in which atypical ductal hyperplasia without microcalcifications was reported at histology and surgical excision was recommended (Fig. 1A,1B,1C,1D), yielding DCIS. Ductal carcinoma was identified in the remaining 13 of 14 malignant lesions, with microcalcifications seen in association with DCIS in 12 of 13. Extensive comedo-type DCIS was present in one of 13 cases in which calcifications were not seen at histology. The four benign lesions included one fibroadenoma, one tubular adenoma, one fibrocystic change with microcalcifications, and one rare case of breast infarction and necrosis, which were all considered concordant. Overall, the histology findings in three (23%) of 13 lesions that were excised after core biopsy were upgraded, one from atypical ductal hyperplasia to DCIS and two from DCIS to invasive ductal carcinoma.

Imaging Findings

Group 1, microcalcifications with associated mammographic findings.—Microcalcifications on mammography were associated with 10 masses (67%), four asymmetric densities (27%), and one region of architectural distortion (7%). The mammographic features are listed in Table 1. Lesions ranged in size from 6 to 65 mm (mean size, 26 mm). The mammographic parenchyma was extremely dense (n = 4), heterogeneously dense (n = 4), or scattered (n = 7). Eight (53%) lesions were assigned a BI-RADS category 4 (suspicious), and seven (47%) were assigned a BI-RADS category 5 (highly suggestive of malignancy).
TABLE 1 Imaging Features of Clustered Microcalcifications
Imaging Features of Microcalcifications (n = 23)Microcalcification Groups
Group 1 Microcalcifications with Other Lesions (n = 15)Group 2 Microcalcifications Alone (n = 8)
Mammography  
    Number of microcalcifications  
        5-1520
        16-5084
        >5054
    Morphology of microcalcifications  
        Pleomorphic116
        Amorphous31
        Round and coarse11
    Associated findings  
        Mass100
        Architectural distortion10
        Asymmetric density40
Sonography  
    Morphology of microcalcifications  
        Specular reflectors138
        Echogenic region20
    Associated findings  
        Mass154
        Ductlike structure
0
4
On sonography, the corresponding lesions ranged in size from 9 to 50 mm (mean size, 21 mm). Microcalcifications were seen as distinct echogenic specular reflectors within an associated hypoechoic mass in 13 lesions (Table 1). In two cases, the grouped microcalcifications were seen as unusual broader echogenic regions that were the predominant sonographic findings in each case (Figs. 2A,2B and 3A,3B). Three lesions were upgraded from a BI-RADS category 4 on mammography alone to a BI-RADS category 5 after review of the sonograms.
Fig. 3A. 49-year-old woman with benign calcified breast mass. Spot compression magnification mammogram of right breast reveals small, ill-defined mass containing clustered microcalcifications.
Fig. 3B. 49-year-old woman with benign calcified breast mass. Sonogram of mass shows predominantly echogenic shadowing mass (arrows) containing central hypoechoic component. Sonographically guided core biopsy yielded papillary apocrine metaplasia and calcifications associated with benign tissue.
Group 2, microcalcifications with no associated mammographic abnormality.—The mammographic features of groups of microcalcifications without associated mammographic findings are listed in Tables 1 and 2 (Figs. 1A,1B,1C,1D and 3A,3B,4A,4B,5A,5B,6A,6B). The diameter of the groups of microcalcifications ranged in size from 10 to 100 mm (mean size, 30 mm). The distribution of microcalcifications was clustered (Figs. 1A,1B,1C,1D, 4A,4B, and 5A,5B) in five (63%), segmental in two (25%), and regional Fig. 6A,6B) in one (13%). The mammographic parenchyma was extremely dense (n = 3), heterogeneously dense (n = 4), or scattered (n = 1). Four (50%) lesions were assigned a BI-RADS category 4 (suspicious) and four (50%) were assigned a BI-RADS category 5 (highly suggestive of malignancy) on review of the mammograms alone.
TABLE 2 Lesions Seen as Microcalcifications Alone on Mammography and Biopsied Under Sonographic Guidance
Patient No.Mammographic Tissue DensityMammographic Microcalcification MorphologyaSonographic FindingsbType of BiopsyCore Biopsy ResultExcision Result
1Extremely densePleomorphic (10 mm/15-50)Oval mass, echogenic foci (10 mm)CoreDCISDCIS
2ScatteredPleomorphic (100 mm/>50)Irregular mass, ductal extension, echogenic foci (16 mm)CoreIDC/DCISIDC/DCIS
3Heterogeneously densePleomorphic (10 mm/5-15)Dilated duct, echogenic foci (10 mm)CoreADHDCIS
4Heterogeneously densePleomorphic (40 mm/>50)Irregular mass, ductal extension, echogenic foci (24 mm)CoreDCISIDC/DCIS
5Heterogeneously densePunctate (40 mm/>50)Dilated duct, echogenic foci (20 mm)CoreInfarction and necrosisNA
6Extremely densePleomorphic (14 mm/>50)Oval mass, echogenic foci (15 mm)CoreDCISDCIS
7Extremely denseAmorphous (10 mm/5-15)Dilated duct, echogenic foci (18 mm)NLNAIDC/DCIS
8
Heterogeneously dense
Pleomorphic (15 mm/15-50)
Irregular mass, echogenic foci (12 mm)
NL
NA
IDC/DCIS
Note.—DCIS = ductal carcinoma in situ, IDC = invasive ductal carcinoma, ADH = atypical ductal hyperplasia, NA = not applicable, NL = needle localization with excisional biopsy.
a
Extent of microcalcifications and range of number of microcalcifications seen on mammography.
b
Size of lesions seen on sonography.
Fig. 4A. 52-year-old woman with breast carcinoma. Spot compression magnification mammogram of right breast reveals fewer than 50 clustered pleomorphic microcalcifications (arrows) within dense fibroglandular tissue. No associated mass is identified.
Fig. 4B. 52-year-old woman with breast carcinoma. Sonogram of region of clustered microcalcifications shows hypoechoic mass (arrows) with internal echogenic specular reflectors corresponding to microcalcifications. Sonographically guided core biopsy yielded ductal carcinoma in situ.
Fig. 5A. 57-year-old woman with breast carcinoma. Mediolateral oblique mammogram of left breast shows more than 50 tightly clustered pleomorphic microcalcifications with no associated mass.
Fig. 5B. 57-year-old woman with breast carcinoma. Sonogram of clustered microcalcifications shows oval mass (arrows) with ill-defined posterior margin and central shadowing. Numerous echogenic specular reflectors correlating with microcalcifications are seen centrally within mass. Sonographically guided core biopsy yielded ductal carcinoma in situ.
Fig. 6A. 85-year-old woman with breast carcinoma. Magnification mammogram of left breast reveals large region of pleomorphic microcalcifications corresponding to carcinoma. Several benign calcifications are also present anteriorly and inferiorly.
Fig. 6B. 85-year-old woman with breast carcinoma. Sonogram of region of calcifications shows apparent ductlike structure (arrows) with internal specular reflectors corresponding to calcifications. Sonographically guided core biopsy yielded ductal carcinoma in situ and invasive ductal carcinoma.
On sonography, a mass (n = 4; Figs. 4A,4B and 5A,5B) or dilated ductlike structure (Figs. 1A,1B,1C,1D and 6A,6B) (n = 4) was identified in association with the microcalcifications in each case (Table 2). These sonographic lesions ranged in size from 10 to 24 mm (mean size, 15 mm). For microcalcifications extending over regions from 40 to 100 mm on mammography, the identified sonographic abnormality measured 24 mm or less in each case. For smaller clusters of microcalcifications on mammography (≤15 mm), the sonographic lesion was similar in size or slightly larger (range, 10 to 18 mm) (Table 2). Microcalcifications were seen as distinct echogenic specular reflectors (Figs. 1A,1B,1C,1D and 4A,4B,5A,5B,6A,6B) within the hypoechoic mass or as an apparent ductlike structure in each. With the addition of the sonographic images, both radiologists reviewing the films upgraded one of four lesions from a BI-RADS category 4 on mammography alone to a BI-RADS category 5 lesion.

Histologic Results

Group 1.—Twelve percutaneous core biopsies and three needle localizations with surgical excision were performed in this group, which showed microcalcifications with an associated finding on mammography. Eleven (73%) of 15 lesions were malignant, including invasive ductal carcinoma with DCIS in 10, and DCIS alone in one. The four benign lesions included two fibroadenomas, one radial scar, and one tubular adenoma. Of 12 lesions initially sampled with core biopsy, subsequent surgical excision or mastectomy was performed in nine (eight malignant, one benign). One patient with invasive carcinoma refused further surgery. Of two lesions initially diagnosed at core biopsy as DCIS, the histology finding in one was upgraded to invasive ductal carcinoma with DCIS after surgery.
Group 2.—Seven (88%) of eight lesions in which sonography showed a mass or ductlike structure that was not suspected on mammography were malignant, including invasive ductal carcinoma with DCIS in four, and DCIS alone in three. Of the six lesions initially sampled with core biopsy, patients with the five malignant lesions subsequently underwent needle localization and surgical excision. The histology findings in two of these cases were upgraded from atypical ductal hyperplasia to DCIS in one, and from DCIS to invasive ductal carcinoma in the other. The one benign lesion was a rare case of breast infarction and necrosis in a diabetic patient with end-stage renal disease.

Discussion

Although mammography is the modality of choice to detect and characterize clustered microcalcifications [3], some microcalcifications can now also be identified during a directed sonographic evaluation because of improvements in spatial resolution and other advancements in technology [4,5,6, 8,9,10,11, 13,14,15]. The advantage of identifying microcalcifications on sonography is that these lesions can then be biopsied under sonographic guidance. Sonographically guided percutaneous procedures require only a sonographic unit, which is more widely available in breast imaging centers than the specialized equipment required for stereotactic biopsy. Sonographically guided needle core biopsies and needle localizations are often preferred by patients over stereotactic needle core biopsies or mammographic needle localizations because of more comfortable supine positioning, lack of mammographic compression, and lack of radiation exposure [7]. The radiologist can often perform sonographically guided procedures more quickly with more flexibility for the insertion site [7]. Sonographically guided biopsy may be the only option for core biopsy of identifiable microcalcifications in patients with breasts that compress to less than 2 cm on mammography, which makes stereotactic biopsy difficult or impossible.
Microcalcifications were successfully targeted and retrieved under sonographic guidance in all cases attempted in our study. This included two very small mammographic lesions (one measuring only 6 mm and another containing only six microcalcifications), which are sometimes challenging even to biopsy using stereotactic guidance. Our success rate (100%) in retrieving microcalcifications during sonographically guided core biopsies is higher than that in the only other report of sonographically guided core biopsy of microcalcifications in which core biopsy was attempted but nondiagnostic in 22% of cases [11]. In those nondiagnostic biopsies, the authors reported that microcalcifications were not seen on specimen radiography. As with stereotactic core biopsy of microcalcifications, specimen radiography for sonographic guidance of microcalcifications is a critical component of the biopsy process. In our study, if microcalcifications were not documented on the initial specimen radiograph, the procedure was considered incomplete, and additional cores were obtained. The number of cores obtained in our study series was substantially higher than the number of cores obtained in our control group of 49 noncalcified lesions biopsied under sonographic guidance. This increase in the number of cores can be attributed to the need to retrieve microcalcifications within the biopsy specimens, which was not necessary in the control group. There was no difference in the complication rates between our study core of biopsy cases and the control group of noncalcified masses biopsied under sonographic guidance.
As in stereotactic biopsy of microcalcifications, ascertaining imaging and histologic concordance after sonographically guided core biopsy of microcalcifications is also critical to the process. The identification of microcalcifications in the histologic specimen must be documented to assure that the appropriate lesion was evaluated. In one case, two microcalcifications were identified within one core on specimen radiograph; however, a review of the pathology revealed atypical hyperplasia, and microcalcifications were not seen histologically, even on stepped sections. In this case, excisional biopsy was recommended, revealing DCIS with microcalcifications.
Eighty-eight percent of sonographically guided core biopsies in our study were performed using the multipass automated gun tachnique. Limitations of this technique for sampling microcalcifications have been reported with stereotactic biopsy [16]. Targeting the tiny specular reflectors representing microcalcifications at sonographically guided biopsy using the multipass technique required greater precision of alignment of the needle, compared with our experience in targeting a larger general area within noncalcified masses. The need for more precise alignment of the biopsy needle parallels the situation experienced with stereotactic core biopsy of microcalcifications using the automated gun technique before the development of the vacuum-assisted device. Difficulty in targeting the echogenic foci may also arise using the multipass technique if air is introduced into the biopsy cavity because the gas can produce a focal echogenic appearance or cause shadowing in the area, mimicking or obscuring the original echogenic target. Another potential limitation of the multipass automated gun technique for sampling microcalcifications under sonographic guidance is that the sampling error and underestimation of disease in cases of atypical ductal hyperplasia and DCIS may be higher, as has been documented with stereotactic biopsy of microcalcifications [16, 17].
Recently, a hand-held vacuum-assisted device (Biopsys/Ethicon Endo-Surgery) has been made commercially available for sonographically guided biopsy with the promise of larger cores and faster tissue retrieval, similar to the vacuum-assisted device used with stereotactic guidance. We had only limited experience (two lesions) with the vacuum-assisted device for sonographically guided biopsy of microcalcifications during this study period. However, as with stereotactic biopsy of microcalcifications, we anticipate that the vacuumassisted device will enable the radiologist to obtain samples more quickly and require less precision of the needle placement to retrieve microcalcifications because of the vacuum mechanism's ability to suction tissue from adjacent areas. The vacuum device might also solve the problem of air in the biopsy cavity because it can suction air and blood from the biopsy cavity during the procedure. As with stereotactic biopsy, this device might also reduce the potential for sampling errors and the underestimation of disease introduced by the multipass technique [16, 17].
During the core biopsy procedures in this study, we did not deploy marker clips to mark the biopsy site because in most cases we used the multipass automated gun device that is not designed for clip deployment. In all cases, a review of the specimen radiographs showed fewer microcalcifications than were seen in the lesion on the original mammograms, suggesting that residual microcalcifications would be present in the breast. Residual microcalcifications were present in all 14 cases in which postprocedure mammograms were available for review, and the remaining four lesions initially contained more than 50 microcalcifications, with fewer than 50 seen on specimen radiographs. However, in lesions containing only a small number of microcalcifications, it would be prudent to deploy a marker clip in case all microcalcifications were removed at core biopsy. The use of the hand-held vacuum-assisted device does allow deployment of clips, and in subsequent cases, we have chosen this device for sonographically guided biopsy of microcalcifications to facilitate retrieval of microcalcifications and to enable clip deployment.
Sonographic guidance for wire localization of microcalcifications before surgery was successfully performed in all five cases. The wire localization procedures did not require the same degree of precision of needle alignment as did the core biopsy procedures because the wire could be placed within the entire group of microcalcifications, rather than aligned with just one microcalcification. However, the use of this technique might be less advantageous than mammographic guidance when a large group of malignant microcalcifications is targeted for surgical excision. In these cases, the surgeon often requests that the radiologist place two wires at either end of the group of calcifications, to bracket the lesion. We did not attempt to use sonographic guidance for bracketing microcalcifications in our study. In our experience with sonography of microcalcifications, the extent of calcifications seen on sonography is often less than the extent of calcifications seen on mammography. In this situation, sonography may underestimate the extent of tumor and may not be reliable for guiding wires to bracket the lesion.
On the basis of our past experience with sonography, many of the masses and areas of architectural distortion seen on mammography in group 1 of our study were expected to be seen on sonography, regardless of the presence of microcalcifications. Sonographically guided core biopsies and wire localizations of masses and regions of architectural distortion seen on mammography are commonly performed when the corresponding masses are seen on sonography. Although biopsy of such targets under sonographic guidance is not new, targeting the microcalcifications within such lesions is a novel application for sonography. When we biopsy masses associated with clustered microcalcifications under stereotactic guidance, we target the microcalcifications and obtain specimen radiographs to assure that the correct region was sampled. Likewise, microcalcifications should be targeted when sonographic guidance is used to biopsy such lesions. Targeting and retrieving microcalcifications at sonographically guided biopsy immediately increases our confidence that the correct area was sampled.
On mammography, most of the lesions in group 1 contained more than 15 microcalcifications that were pleomorphic and highly suspicious. Fewer microcalcifications were seen in the corresponding sonographic examination. Most microcalcifications were seen as focal echogenic specular reflectors that seemed to flicker as the transducer moved in and out of the plane of the microcalcifications. In two cases, a broader echogenic region accounted for innumerable tightly grouped tiny punctate and pleomorphic microcalcifications. This unusual appearance proved to be benign in both cases, although histology (radial scar) of one conferred increased risk for subsequent breast cancer.
Microcalcifications seen alone on mammography in group 2 of our study were also predominantly pleomorphic in morphology. Sonography in these cases revealed a mass or dilated duct that provided a hypoechoic background to contrast with the microcalcifications, likely contributing to their identification. The size of lesions on sonography in this group was often smaller than the size of microcalcification clusters seen on mammography. However, the smaller unsuspected mass or duct containing microcalcifications targeted on sonography may represent the higher grade component of the lesion (e.g., invasive component in a large area of DCIS). In one case in which microcalcifications extended 100 mm on mammography, the biopsy of a 16-mm abnormal ductlike structure detected on sonography resulted in the diagnosis of the one small focus of invasive carcinoma within the breast. This area may not have been identified as the target site if stereotactic biopsy had been performed. Our sample size for this observation was quite small; therefore, a larger study is necessary to determine whether sonographically guided biopsies could actually reduce the underestimation of disease compared with stereotactic biopsy.
A recent study suggests that the sonographic identification of a suspicious mass in lesions where microcalcifications alone are seen on mammography might increase the likelihood that the lesion represents a malignant process [11]. Likewise, in our study, most of the lesions (88%) seen as microcalcifications alone on mammography with a mass or dilated duct seen on sonography proved to be DCIS or invasive ductal carcinoma. However, our population did not include consecutive clusters of microcalcifications undergoing biopsy; we selected many large lesions considered highly suggestive of malignancy to target for sonographically guided biopsy. A larger prospective study evaluating consecutive patients with suspicious microcalcifications is underway to determine which lesions can be reliably identified on sonography and to evaluate whether sonography can affect the pretest probability of malignancy.
In conclusion, our experience shows that microcalcifications identifiable on sonography can be successfully biopsied under sonographic guidance. Suspicious microcalcifications on mammography correlated with either suspicious hypoechoic masses or abnormal ductlike structures associated with the microcalcifications on sonography, findings that made the microcalcification target more obvious. A larger prospective evaluation is necessary to determine whether sonographic detection of a mass or duct seen as microcalcifications alone on mammography is predictive of malignancy and whether targeting these lesions with sonography can improve the rate of underestimation of disease compared with stereotactic core biopsy.

Footnotes

Presented at the annual meeting of the American Roentgen Ray Society, Washington, DC, May 2000.
Address correspondence to M. S. Soo.

References

1.
Jackson VP. The role of ultrasound in breast imaging. Radiology 1990; 177:305-311
2.
Stavros AT, Thickman D, Rapp CL, Dennis MA, Parker SH, Sisney GA. Solid breast nodules: use of sonography to distinguish between benign and malignant lesions. Radiology 1995; 196:123-124
3.
Cardenosa G, Mendelson E, Bassett L, et al. Appropriate imaging work-up of breast microcalcifications. American College of Radiology ACR Appropriateness Criteria. (abstr) Radiology 2000; 215(P):973
4.
Anderson ME, Soo MSC, Bentley RC, Trahey GE. The detection of breast microcalcifications with medical ultrasound. J Acoust Soc Am 1997; 101:29-39
5.
Yang WT, Suen M, Ahuja A, Metreweli C. In vivo demonstration of microcalcification in breast cancer using high resolution ultrasound. Br J Radiol 1997; 70; 685-690
6.
Moon WK, Im J-G, Koh YH, Noh D-Y, Park IA. US of mammographically detected clustered microcalcifications. Radiology 2000; 217:849-854
7.
Parker SH, Burbank F. A practical approach to minimally invasive breast biopsy. Radiology 1996; 200:11-20
8.
Ranieri E, D'Andrea MR, D'Alessio A, et al. Ultrasound in the detection of breast cancer associated with isolated clustered microcalcifications, mammographically identified. Anticancer Res 1997; 17:2831-2836
9.
Rickard MT. Ultrasound of malignant breast microcalcifications: role in evaluation and guided procedures. Australas Radiol 1996; 40:26-31
10.
Cleverley JR, Jackson AR, Bateman AC. Pre-operative localization of breast microcalcification using high-frequency ultrasound. Clin Radiol 1997; 52:924-926
11.
Teh WL, Wilson ARM, Evans AJ, Burrell H, Pinder SE, Ellis IO. Ultrasound guided core biopsy of suspicious mammographic calcifications using high frequency and power Doppler ultrasound. Clin Radiol 2000; 55:390-394
12.
American College of Radiology. Breast imaging reporting and data system (BI-RADS), 3rd ed. Reston, VA: American College of Radiology, 1998
13.
Shook JE, Rubin E, Pile N, Dempsey P, Bernreuter W. Breast microcalcifications: a role for ultrasound? AJR 1998; 170[American Roentgen Ray Society 98th Annual Meeting Abstract Book suppl]:104-105
14.
Mester J, Eisen CS, Keating DM, Giess CS, Rosenblatt R. The role of ultrasound for biopsy of microcalcifications. AJR 2000; 174[American Roentgen Ray Society 100th Annual Meeting Program Book suppl]:47-48
15.
Soo MS, Vo TT, Rosen EL. Breast lesions containing microcalcifications biopsied under sonographic guidance: findings at mammography and sonography. AJR 2000; 174[American Roentgen Ray Society 100th Annual Meeting Program Book suppl]:49
16.
Burbank F. Stereotactic breast biopsy of atypical ductal hyperplasia and ductal carcinoma in situ lesions: improved accuracy with directional, vacuumassisted biopsy. Radiology 1997; 202:834-847
17.
Jackman RJ, Burbank F, Parker SH, et al. Atypical ductal hyperplasia diagnosed at stereotactic breast biopsy: improved reliability with 14-gauge, directional, vacuum-assisted biopsy. Radiology 1997; 204:485-488

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: 1007 - 1015
PubMed: 11906892

History

Submitted: June 15, 2001
Accepted: September 21, 2001

Authors

Affiliations

Mary Scott Soo
Department of Radiology, Breast Imaging Division, Box 3808, Duke University Medical Center, Durham, NC 27710.
Jay A. Baker
Department of Radiology, Breast Imaging Division, Box 3808, Duke University Medical Center, Durham, NC 27710.
Eric L. Rosen
Department of Radiology, Breast Imaging Division, Box 3808, Duke University Medical Center, Durham, NC 27710.
Thuy T. Vo
Department of Radiology, Breast Imaging Division, Box 3808, Duke University Medical Center, Durham, NC 27710.
East Valley Diagnostic Imaging, 1125 E. Southern Ave., Ste. 300, Mesa, AZ 85212.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

View Options

View options

PDF

View PDF

PDF Download

Download PDF

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media