Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2383041953
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Rodriguez-Soto, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Rodriguez-Soto, J.
(Radiology 2006;239:61-70.)
© RSNA, 2006

Breast Imaging

Breast Microcalcifications: Retrieval Failure at Prone Stereotactic Core and Vacuum Breast Biopsy—Frequency, Causes, and Outcome1

Roger J. Jackman, MD and Jorge Rodriguez-Soto, MD, PhD

1 From the Departments of Radiology (R.J.J.) and Pathology (J.R.), Palo Alto Medical Clinic, Palo Alto, Calif. Received November 17, 2004; revision requested January 18, 2005; revision received March 4; accepted March 23; final version accepted June 17. R.J.J. is a clinical consultant to Ethicon Endo-Surgery and was formerly a shareholder in and clinical consultant to Biopsys Medical. Supported in part by an educational grant from Biopsys to the Palo Alto Medical Foundation. Address correspondence to: R.J.J., 3589 Arbutus Ave, Palo Alto, CA 94303. (e-mail: JackmanR{at}Gmail.com).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Purpose: To retrospectively determine the frequency and causes of failure to retrieve microcalcifications in nonpalpable lesions, as judged on a radiograph of the specimen, and to determine outcome in patients with those lesions.

Materials and Methods: Informed consent was obtained from each patient prior to biopsy. The institutional review board approved this HIPAA-compliant study and granted a waiver of informed consent. Retrospective review was performed of 1701 consecutive nonpalpable microcalcification lesions in 1511 women aged 29–92 years (median age, 54 years) who underwent percutaneous stereotactic biopsy on a prone biopsy table. Biopsy was successively performed with 14-gauge core, 14-gauge vacuum, and 11-gauge vacuum devices, with mild selection bias, and for each lesion, biopsy was performed with one device. Radiographs of the specimen were obtained to see whether microcalcifications were retrieved. Patient, mammographic, and biopsy variables were correlated with negative radiographs of the specimen. At repeat biopsy or mammographic follow-up, outcome was evaluated in patients with benign histologic results and negative radiographs of the specimen by using Fisher exact test P values.

Results: Radiographs of the specimen were negative in 16% (30 of 182) of lesions at 14-gauge core biopsy, in 4% (four of 96) of lesions at 14-gauge vacuum biopsy, and in 1% (19 of 1423) of lesions at 11-gauge vacuum biopsy (P < .001). Substantial bleeding was a significant factor (P < .001) in failure to retrieve microcalcifications at only 11-gauge vacuum biopsy. Histologic results in 53 lesions with negative radiographs of the specimen were malignant (n = 6), indicated atypical hyperplasia (n = 6), or were benign (n = 41). Follow-up in patients with 40 benign lesions was performed with repeat biopsy (n = 17, with malignancy in three lesions) or mammography (n = 23) for 15–128 months (median, 70 months); one patient with one lesion was lost to follow-up.

Conclusion: Failure to retrieve microcalcifications was least common with 11-gauge directional vacuum-assisted biopsy and occurred in 1% (19 of 1423) of lesions. Cancer was missed in 8% (three of 40) of benign lesions in patients who were followed up.

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
In patients with nonpalpable breast lesions, image-guided histologic biopsy can be performed either surgically or percutaneously. For decades, needle-localized breast biopsy was the standard method (1). Review articles present image-guided percutaneous biopsy with automated large-core or directional vacuum-assisted techniques as popular alternatives to surgical biopsy (2–6).

In July 1991, we switched from needle-localized breast biopsy to image-guided histologic biopsy, with a goal of performing stereotactic prone biopsy in all patients with mammographically detected lesions for which an image-guided biopsy was needed (7). Our current study was performed to retrospectively determine the frequency and causes of failure to retrieve microcalcifications in nonpalpable lesions, as judged on a radiograph of the specimen, and to determine the outcome in patients with those lesions.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Records of 1511 female patients (age range, 29–92 years; median age, 54 years) were retrospectively reviewed. These patients had 1701 consecutive nonpalpable microcalcification lesions (without an associated mass) for which they underwent a stereotactic biopsy and radiography of a specimen from April 1993 through December 2001. Radiographs of the specimen were not obtained from July 1991 through March 1993 (n = 158 lesions). Apart from mild inconsistency in the next few months (10 lesions without radiographs of the specimen), radiographs of the specimen were obtained from April 1993 through December 2001 (1701 lesions with radiographs of the specimen). At biopsy, the radiologist recorded the presence or absence of at least one microcalcification on the radiograph of the specimen. Informed consent was obtained from each patient prior to biopsy. The institutional review board approved our study and granted a waiver to perform this retrospective study without informed consent. The study was HIPAA compliant. Both authors had control of the data and information in this study.

Concordant and Discordant Criteria
Until approximately mid-1999, we considered biopsy of a microcalcification lesion to be technically adequate if microcalcifications were identified histologically and/or on the radiograph of the specimen. Only after that did we consider the biopsy results of a histologically benign lesion to be discordant if microcalcifications were seen histologically but not on the radiograph of the specimen. Radiographs of the paraffin blocks were obtained on the rare occasion when microcalcifications were seen on the radiograph of the specimen but were not seen histologically. Paraffin block radiographs were not obtained if microcalcifications were not seen on the radiograph of the specimen. Immediate postbiopsy mammograms of lesions with a negative radiograph of the specimen were usually but not always obtained, and we have incomplete data about the status of residual microcalcifications on those mammograms.

Biopsy Restrictions
Patients were restricted from scheduling a stereotactic biopsy if their weight exceeded the biopsy table limit of 300 pounds (135 kg), if they had bleeding diathesis, or if they were taking anticoagulants and could not temporarily stop taking them, but no patients met those criteria. We had no standard prebiopsy restrictions that were based on presumed inability of the patient to cooperate during the procedure; breast size; or lesion size, position, or conspicuity. Of all lesions for which an image-guided biopsy was needed and for which patients were referred from within our multispecialty clinic, a stereotactic biopsy was technically feasible in 98% (1818 of 1852) of lesions and was actually performed in 97% (1786 of 1844) of lesions for which a biopsy was performed (7). In 1844 lesions for which an image-guided biopsy was needed and for which patients were referred from within our multispecialty clinic, a surgical biopsy was performed in 3% (58 lesions), and an ultrasonographically (US) guided biopsy was performed in 0% (no lesions) (7). We do not have data about the percentages of stereotactic, surgical, and US-guided biopsies performed at outside facilities at which lesions were found. In all lesions for which patients were referred to us from outside facilities during the study period, however, stereotactic biopsy was performed. Starting in 2002, we performed US-guided biopsy for mass lesions in patients who were referred from both inside and outside our facility.

Data Recorded
Patient, mammographic, and biopsy variables were evaluated to see which ones were associated with negative radiographs of the specimen. Before biopsy, lesions were categorized for palpability (discrete, vague, or nonpalpable) by referring clinicians. The radiologist who performed the biopsy recorded the biopsy method (14-gauge core, 14-gauge vacuum, or 11-gauge vacuum biopsy), mammographic lesion type (microcalcifications without a mass, masses [including asymmetries and areas of architectural distortion] with or without associated microcalcifications, or miscellaneous), maximum mammographic lesion diameter, presence or absence of substantial bleeding during biopsy, and number of tissue specimens obtained during biopsy of each lesion. A subjective decision about whether there was substantial bleeding and a decision about the number of specimens to obtain were left to the discretion of the radiologist who performed the biopsy. Starting in 1991, we tried to obtain at least five core specimens per lesion and gradually increased that number. We tried to obtain at least 12 specimens per lesion at vacuum biopsy.

Mammographic tissue density (extremely dense, heterogeneously dense, scattered fibroglandular density, or almost entirely fat) and the Breast Imaging Reporting and Data System (BI-RADS) classification of the American College of Radiology (8) (categories 2–5) were recorded by the radiologist who performed the biopsy starting in February 1995. Density and BI-RADS classification were recorded retrospectively by one of the authors (R.J.J., who had 9 years of experience with BI-RADS classification at the time of retrospective review) for lesions for which biopsy was performed prior to that time, with the exception of information about density that was not available for 131 lesions with positive findings on radiographs of the specimen for which biopsy was performed prior to 1995. Other variables were recorded for all lesions.

Biopsy Procedures
We performed stereotactic biopsy with the patients prone on a biopsy table (Mammotest or Mammotest Plus/S; Fischer Imaging, Denver, Colo) and with three successive techniques. Biopsy was performed as follows: from July 1991 to mid-April 1995, with core biopsy, a long-throw (2.3-cm excursion) automated biopsy gun (Biopty; Bard Urological, Covington, Ga), and a variety of 14-gauge cutting needles; from mid-March 1995 to mid-June 1996, with vacuum biopsy, a directional vacuum-assisted biopsy device (Mammotome; Biopsys Medical/Ethicon Endo-Surgery, Cincinnati, Ohio), and 14-gauge probes; and from mid-May 1996 through December 2001, with vacuum biopsy, the same Mammotome device, and 11-gauge probes. During the 1-month overlap between the times the various biopsy techniques were employed, availability of the new Mammotome probe was used as a criterion to determine which method was used. For each lesion, biopsy was performed with one device. No lesion or patient variables were used to determine the method. Four radiologists, including one of the authors (R.J.J.), performed biopsy. These radiologists had 1.75, 1.75, 1.00, and 0.25 years of experience in performance of stereotactic biopsy prior to performing the first biopsy in the lesions included in this study.

Tissue Specimens and Radiographs of Specimens
Tissue specimens were moistened with sterile saline and placed in either the lid of a specimen container or a Petri dish, and radiography was performed. From April 1993 through June 2001, radiographs of specimens were obtained with analog mammographic machines (models M-III or M-IV; Lorad/Hologic, Bedford, Mass) and 20 kVp, 6 mAs, x1.8 magnification, and a 100-µm focal spot. From July 2001 through December 2001, radiographs of specimens were obtained with an analog radiographic machine (model MX-20; Faxitron X-ray, Wheeling, Ill) and 17 kVp, 3 mAs, x1.5 magnification, and a 20-µm focal spot. The tissue specimens were then placed in formalin and processed in the Department of Pathology.

Histologic Diagnoses
The original histologic diagnoses determined by a variety of pathologists, including one of the authors (J.R.), were accepted for this study. For all lesions for which a biopsy was performed, histologic diagnoses of lesions as malignant, high risk, or benign were recorded from the histology reports by one of the authors (R.J.J.). Malignant lesions included invasive carcinoma, ductal carcinoma in situ (DCIS), lymphoma, and sarcoma. We considered atypical ductal hyperplasia, atypical lobular hyperplasia, lobular carcinoma in situ, radial scar, phyllodes tumor, and papilloma to be high-risk lesions, for which the associated presence of carcinoma can be underestimated at percutaneous biopsy. Lesions that were not categorized as histologically malignant or high risk were classified as benign.

Follow-up
We had no set protocol for follow-up of benign lesions with a negative radiograph of the specimen. The radiologist who performed the biopsy, often in consultation with the pathologist who evaluated the histologic slides, decided if an immediate repeat biopsy, with a percutaneous biopsy and removal of more tissue specimens or with surgical excision, was recommended. Histologic results at repeat biopsy were compared with histologic results at initial percutaneous biopsy. For all patients with benign lesions who were not thought to need a repeat biopsy, we recommended 6- and 12-month postbiopsy mammography and then annual diagnostic mammographic follow-up for at least 3 years after biopsy. The original follow-up mammographic reports made by a variety of radiologists, including one of the authors (R.J.J.), were accepted for this study. We reported that the microcalcifications were gone, decreased, unchanged, or increased compared with findings on the prebiopsy mammograms. Delayed repeat biopsy was recommended if there was lesion progression at postbiopsy mammographic follow-up (9–12).

Histologic Review
One of the authors (J.R., who had 13 years of experience with breast histology at the time of review) retrospectively reviewed all available histologic slides for the presence or absence of microcalcifications. If no microcalcifications were initially identified, polarizing light was used to look for calcium oxalate. If microcalcifications were identified, the maximal histologic size of each microcalcification was determined by using an upright light transmission microscope (Olympus BX40; Olympus Optical, Tokyo, Japan) with an ocular micrometer (Olympus WH10X/22; Olympus Optical) that was similar to the microscope used by Dahlstrom et al (13). Histologic diagnoses were recorded for histologically calcified lesions.

Statistical Analysis
Data were analyzed with statistical software (Stata, version 8.2, 2004; Stata, College Station, Tex). The Fisher exact test and correlation coefficients with P values of less than .05 were considered significant. A preliminary check for problematic colinearity among the independent variables was performed; ordinary correlations were calculated for continuous variables, whereas rank correlations were used for the noncontinuous variables of breast density, BI-RADS category, biopsy method, substantial bleeding, and lesion histologic results. The largest correlations were between biopsy method and number of specimens (r = 0.33, P < .001), between BI-RADS category and lesion histologic results (r = 0.25, P < .001), between lesion size and number of specimens (r = 0.21, P < .001), between lesion size and lesion histologic results (r = 0.17, P < .001), and between lesion size and BI-RADS category (r = 0.15, P < .001). None of these correlations was judged to be large enough to cause colinearity problems with the multivariate analysis; in the largest correlation, that between biopsy method and number of specimens, the two variables shared only 11% (r = 0.33) of their variability.

A logistic regression analysis of all variables was performed to determine which variables were associated with retrieval of microcalcifications. Age was a variable observed per patient, and breast density was a variable observed per breast. All other variables were observed per lesion. Because 18% (311 of 1701) of the sample consisted of two or more lesions from the same breast, an approach with a population-average generalized estimating equation and an exchange correlational model was used, with the breast as the clustering variable. Because the distribution of lesion size was highly skewed, the log of the lesion size was used in the multivariate analysis.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Failure Rates
Results of univariate analysis of the rates for failure to retrieve microcalcifications according to patient, mammographic, and stereotactic biopsy variables indicate that failures were statistically most significant for four variables—biopsy method, substantial bleeding during the biopsy, number of specimens per lesion, and breast density; each had a P value of less than .001 (Table 1). Failure rates for lesion diameter, BI-RADS category, and histologic result category also were significant, and all had a P value of less than .05 (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. Results for Radiographs of the Specimen for 1701 Microcalcification Lesions according to Patient, Mammographic, and Stereotactic Biopsy Variables

 
The results of the multivariate population-average generalized estimating equation logistic regression analysis indicated that only four variables were significantly associated with failure to retrieve microcalcifications: biopsy method, substantial bleeding, number of specimens, and lesion diameter, all with P < .001. The estimated within-breast correlation coefficient among lesions was 0.013. The effect of these four variables can best be expressed by the increased or decreased odds of failure to retrieve microcalcifications.

Biopsy Method
On the basis of the biopsy method, radiographs of the specimen were negative in 16% (30 of 182) of lesions at 14-gauge core biopsy, in 4% (four of 96) of lesions at 14-gauge vacuum biopsy, and in 1% (19 of 1423) of lesions at 11-gauge vacuum biopsy (P < .001). For the biopsy method, the odds ratios were expressed relative to the 14-gauge core device. The odds of failing to retrieve microcalcifications with the 14-gauge vacuum device were only 0.01 of those for the core device (95% confidence interval: 0.002, 0.049), whereas for the 11-gauge vacuum device, the odds of failure were only 0.003 of those for the core device (95% confidence interval: 0.001, 0.007).

Bleeding
Substantial bleeding during the biopsy in 45 lesions occurred in 0.5% (one of 182) of lesions at 14-gauge core biopsy, in 2% (two of 96) of lesions at 14-gauge vacuum biopsy, and in 3% (42 of 1423) of lesions at 11-gauge vacuum biopsy (P < .15). Although substantial bleeding was a significant factor in overall failure to retrieve microcalcifications (Table 1), bleeding was a significant factor according to biopsy device only at 11-gauge vacuum biopsy (Table 2). Bleeding was noted to be arterial in 16% (seven of 45) and venous in 84% (38 of 45) of lesions with substantial bleeding. We do not have complete data about complications that required intervention; we are aware of two hematomas that required drainage and no infections that required antibiotic treatment in the study period. The effect of substantial bleeding was to increase the odds of failure to retrieve microcalcifications by a factor of 28 (95% confidence interval: 9.1, 89).


View this table:
[in this window]
[in a new window]

  		Table 2. Results for Radiographs of the Specimen for 1701 Microcalcification Lesions according to Biopsy Method versus Substantial Bleeding and Number of Specimens per Lesion

 
Number of Specimens
Although the number of specimens obtained per lesion was a significant factor in overall failure to retrieve microcalcifications (Table 1), the number of specimens was a significant variable according to biopsy device at only 11-gauge vacuum biopsy (Table 2). For each additional specimen retrieved, the odds of failure to retrieve microcalcifications actually increased slightly, with an odds ratio of 1.07 (95% confidence interval: 1.04, 1.10). Thus, for every 12 additional specimens, the odds of failure doubled. This counterintuitive result appears to have been caused by the diminishing returns of obtaining more than 24 specimens (Table 1). If analysis is restricted to samples with a maximum of 24 specimens (86% of cases), there is no effect of the number of specimens on the odds of failure (odds ratio = 1.04; 95% confidence interval: 0.91, 1.20; P < .569).

Lesion Diameter
For each doubling of lesion diameter, the odds of failure to retrieve microcalcifications decreased by almost half, with an odds ratio of 0.45 (95% confidence interval: 0.31, 0.66).

Biopsy Results
Percutaneous biopsy results revealed six malignant lesions, all DCIS. Findings at subsequent surgery confirmed DCIS in five of those lesions and failed to show residual carcinoma in one lesion, with no DCIS lesions upgraded to invasive carcinoma at subsequent surgery. Percutaneous biopsy results revealed six high-risk lesions, all atypical hyperplasia. Findings at subsequent surgery revealed carcinoma (both DCIS) in two (50%) of four atypical ductal hyperplasia lesions and in none (0%) of two atypical lobular hyperplasia lesions. No atypical hyperplasia lesions were upgraded to invasive carcinoma.

Percutaneous histologic results were benign in 41 (77%) of 53 lesions with negative radiographs of the specimen. Repeat biopsy in 17 (41%) of 41 benign lesions was performed with repeat stereotactic biopsy (four lesions; in one lesion, repeat biopsy was performed with the core method and in three, repeat biopsy was performed with the vacuum method), with needle-localized excision (11 lesions), or at mastectomy (two lesions, both with carcinoma elsewhere in the same breast). Repeat biopsy histologic results included a malignant lesion (three lesions; two lesions with DCIS and one lesion with tubular carcinoma), atypical ductal hyperplasia (one lesion), or a benign lesion (13 lesions). Repeat biopsy was performed within 4 months (n = 11, with two malignant results), 5–12 months (n = 2, with no malignant results), and more than 12 months (n = 4, with one malignant result, with mammographic progression of the lesions after initial biopsy that prompted repeat biopsy).

Follow-up Findings
Postbiopsy mammographic follow-up, without mammographic progression of the lesion and without repeat biopsy, was performed in 23 (56%) of 41 benign lesions. Findings on the initial follow-up mammograms obtained at 3–26 months (median, 6 months) after biopsy revealed that the microcalcifications were gone (n = 2), decreased (n = 10), unchanged (n = 11), or increased (n = 0), compared with the findings on prebiopsy mammograms. Follow-up was performed for 15–128 months (median, 70 months); follow-up in just one lesion was performed for less than 37 months. One (2%) of 41 lesions was in a patient lost to follow-up.

Thus, lesions with a negative radiograph of the specimen were false-negative findings in 8% (three of 40) of benign lesions with postbiopsy follow-up and in 27% (three of 11) of those study lesions that proved to be malignant at percutaneous biopsy and/or surgical excision.

Histologic Review Findings
We retain histologic slides for 10 years and had discarded slides for all 30 lesions for which a core biopsy was performed; in all of these lesions, biopsy was performed more than 10 years before histologic review. Slides were available for review in 96% (22 of 23) of lesions for which a vacuum biopsy was performed. Histologic microcalcifications were identified in 50% (11 of 22) of lesions with negative radiographs of the specimen. All 11 lesions with identified microcalcifications revealed calcium carbonate, which was visualized without polarized illumination. Histologic diagnoses in these 11 lesions included DCIS (n = 1), atypical ductal hyperplasia (n = 1), and benign lesions (n = 9). One lesion contained a single microcalcification that was 130 µm, and all other microcalcifications were smaller than 100 µm.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Radiography of the specimen for detection of microcalcifications in stereotactic core biopsy specimens was introduced by Meyer et al (14) and Liberman et al (15), and the authors of the latter article also described the technique for performing radiography of specimens (cores) of tissue. Our retrospective study of patients with nonpalpable lesions manifesting as microcalcifications was performed to see what factors were associated with failure to retrieve microcalcifications, as judged on a radiograph of the specimen, and to determine the outcome in patients with those lesions.

Three aspects of the different biopsy methods should be noted. First, acquisition of more than five specimens from the same area by using the 14-gauge core device extracts progressively less tissue and more blood (16). More specimens can be effectively obtained with the vacuum device (17). There is, thus, a small but reliable association between the biopsy method and the number of specimens obtained (r = 0.33, P < .001). The difference appears to be caused mostly by the difference between the core and vacuum methods (Table 3).


View this table:
[in this window]
[in a new window]

  		Table 3. Number of Specimens per Lesion for Each Biopsy Method

 
Second, as previously mentioned, both the biopsy method and the number of specimens were significantly associated with failure to retrieve microcalcifications. Even when failure to retrieve microcalcifications was analyzed for just the 290 lesions in which 12 specimens were obtained, the failure rate was still significantly smaller for the vacuum technique. With the vacuum technique, failures occurred in 3% (seven of 241) of lesions, but failures occurred in 12% (six of 49) of lesions with the core technique (P < .01).

Third, the distribution of the number of specimens was multimodal, with peaks at 12, 18, 24, and 30 specimens. This distribution may have been caused by the fact that 84% of the samples were obtained with 11-gauge vacuum biopsy, and the design of the device used for this method makes it natural for the operator to obtain specimens in groups of six.

If we combine data in Tables 1 and 2, we can draw three conclusions about success in retrieving microcalcifications with the three techniques that we evaluated. First, biopsy should be performed with the 11-gauge vacuum technique, for which failure occurred in only 1% (19 of 1423) of lesions.

Second, substantial bleeding should be avoided during 11-gauge vacuum biopsy, if possible. Apart from being sure that patients are not taking aspirin and other medications that delay blood clotting, we do not know how to avoid substantial bleeding. If a hematoma develops, it could obscure the breast microcalcifications and/or cause movement of the breast microcalcifications. If substantial bleeding occurred during the biopsy, the radiologist who performed the biopsy chose whether to complete the biopsy as quickly as possible in an attempt to do so before a large hematoma developed or, alternatively, to suspend the biopsy for several minutes so that firm compression of the biopsy site could be performed in an effort to stop the bleeding and prevent a large hematoma from developing. We do not know which approach is better.

Third, failures at 11-gauge vacuum biopsy were least common when 13–24 specimens were obtained. The reasons for obtaining a smaller percentage of microcalcifications with more than 24 specimens per lesion are complex and are related to the biopsy process. The radiologist who performs the biopsy usually will obtain the initial radiograph of the specimen after he or she obtains what is thought to be a representative sample of the lesion (perhaps 6–12 specimens, depending on the size of the lesion and the biopsy device used). While the radiologist waits for the radiograph of the specimen to be developed, another six samples usually are obtained. Although more than six specimens per lesion yield more accurate results, there is little need to obtain more than 12–18 specimens per lesion unless initial radiographic analysis of the specimens fails to reveal the expected microcalcifications (18). Thus, a failure to retrieve microcalcifications initially leads to the need to obtain a larger number of biopsy specimens and to the observed inverse relationship between success and the number of specimens collected for large numbers of specimens.

Acquisition of more than 24 specimens frequently, but not always, led to conversion of failure to success for retrieval of microcalcifications. Those lesions from which the largest number of specimens were removed were generally clusters of microcalcifications just faintly visible on radiographic images of the breast obtained during the stereotactic biopsy, and in these instances, the radiologist who performed the biopsy continued to extract samples until at least one sample revealed at least one microcalcification on a radiograph of the specimen, or the biopsy was terminated for other reasons. Despite extraction of a large amount of tissue, for four (12%) of 33 lesions with more than 40 specimens per lesion removed at 11-gauge vacuum biopsy, radiographs of the specimen were negative. These data suggest that failure to retrieve microcalcifications is unlikely to be improved by obtaining even more specimens per lesion. We think that definite visualization of the lesion on the radiographic images of the breast obtained during the stereotactic biopsy and precise positioning of the biopsy needle prior to extraction of tissue samples are more important to successfully retrieve microcalcifications than is obtaining a large number of samples. We think that the needle should be removed and the missed microcalcifications should be retargeted periodically during the biopsy, perhaps after radiography of each sequential group of 12 samples.

Failure to retrieve breast microcalcifications at prone stereotactic biopsy reported by other researchers (13–15,17,19–34) is summarized in Table 4. We attempted to avoid duplication of data in compiling the totals in Table 4. In the literature review, including data from our study, radiographs of the specimen were negative in 14% (169 of 1236) of lesions at 14-gauge core biopsy, in 3% (15 of 493) of lesions at 14-gauge vacuum biopsy, and in 1% (54 of 4781) of lesions at 11-gauge vacuum biopsy (P < .001) (Table 4).


View this table:
[in this window]
[in a new window]

  		Table 4. Rates for Negative Radiographs of the Specimen at Prone Stereotactic Biopsy according to Biopsy Method

 
For several decades, in virtually all nonpalpable microcalcification lesions for which an image-guided biopsy was needed, needle-localized breast biopsy was performed (1). Biopsy options became available with the introduction of stereotactic 14-gauge core biopsy (35,36) and, subsequently, stereotactic vacuum biopsy (2,3,37,38) techniques. To our knowledge, our approach has less selection bias than that in other reports. With our approach, in 97% of lesions in patients who were referred from within our multispecialty clinic and in 100% of lesions in patients who were referred from outside our multispecialty clinic for image-guided biopsy, biopsy was performed with the stereotactic technique, with no selection bias for the stereotactic biopsy method. Selection bias for biopsy method could influence failure to retrieve microcalcifications and other measurements of success.

Most authors of articles do not address the issue of possible selection bias. Those that do address this issue provide variable information about how and why biopsy of lesions was performed with different methods. Liberman et al (21) performed biopsy in 614 lesions by using the stereotactic 14-gauge vacuum method in 108 (18%) lesions, US-guided core method in 34 (6%) lesions, or needle localization and surgical biopsy in 472 (77%) lesions but did not indicate why the different methods were used. In a later article, Liberman and Sama (39) indicated that 106 (53%) of 200 lesions for which biopsy was performed with the stereotactic 11-gauge vacuum technique would have been excluded from an attempt at stereotactic 14-gauge core biopsy because of superficial lesion location, lesion size of less than 5 mm, or inadequate breast thickness. Berg et al (30) performed biopsy in 150 clusters of amorphous calcifications by using the stereotactic 11-gauge vacuum method in 102 (68%) lesions, the stereotactic 14-gauge core method in nine (6%) lesions, or the needle-localized method in 39 (26%) lesions; in the 39 lesions in which surgical biopsy was performed by using needle localization, this technique was used either because an attempt at stereotactic biopsy had failed (n = 8) or because an attempt at stereotactic biopsy had been bypassed (n = 31).

Reynolds et al (25) reported their use of a minimum lesion size restriction of 5 mm for stereotactic biopsy of microcalcifications with the 14-gauge core technique but no size restriction with the 11-gauge vacuum technique. Lee et al (20) mentioned no lesion size restriction for stereotactic core biopsy, but they recommended surgical biopsy instead of stereotactic biopsy if the lesion was close to the skin, chest wall, or nipple; the lesion was vague or poorly defined; the lesion was likely to represent a radial scar; or the breast size was small. Pijnappel et al (40), who reported data from three hospitals in the Netherlands, performed biopsy in 749 nonpalpable lesions by using the stereotactic core technique in 63 (8%) lesions, the US-guided core technique in 128 (17%) lesions, US-guided fine-needle aspiration in 242 (32%) lesions, or needle-localized surgical excision in 316 (42%) lesions but did not indicate why the different methods were used.

Liberman et al (41) initially said discordant calcific lesions included those in which no microcalcifications were identified at radiography of the specimen and/or histologic analysis, which was identical to our initial definition. In the study of Pfarl et al (42), which included Liberman as a coauthor, the definition was later amended to state that identification of histologic microcalcifications alone, and not identification on the radiographs of the specimen, usually is not adequate, and they cited the articles that are mentioned in the next paragraph here (13,43). As stated earlier, we also changed our definition to that noted by Pfarl et al (42) in the later years of this study.

Histologic visualization of microcalcifications without visualization of them on the radiograph of the specimen is not adequate because many small microcalcifications are seen in that way and do not represent the mammographic lesion that prompted biopsy. Stomper et al (43) found histologic microcalcifications in 41% (11 of 27) of malignancies that manifested as masses, with no microcalcifications evident on mammograms or on radiographs of the specimen. Dahlstrom et al (13) found histologic microcalcifications in 33% (11 of 33) of 14-gauge core samples, with no microcalcifications evident on radiographs of the specimen. They reported that microcalcifications that were smaller than 100 µm were seen only histologically. We found histologic microcalcifications in 50% (11 of 22) of lesions with negative radiographs of the specimen, and we were able to confirm that all microcalcifications but one were smaller than 100 µm.

Although visualization of just one definite microcalcification on the radiograph of the specimen is considered as positive retrieval, we think that that finding is not adequate for diagnostic accuracy. The goal of performing a percutaneous biopsy is to be able to establish an accurate histologic diagnosis by removing a representative sample from what may be a histologically heterogeneous lesion. Most important, we want to avoid a false-negative biopsy result (9,10,28,31,32,34,42,44), but we also want to minimize the chance that a biopsy-proved diagnosis of DCIS would lead to underestimation of the presence of invasive cancer (9,19,32,45) and that a biopsy-proved diagnosis of atypical ductal hyperplasia (and other high-risk lesions) would lead to underestimation of the presence of carcinoma (9,19,32,44,46).

We do not know how much tissue needs to be removed at stereotactic biopsy to achieve maximal accuracy. Even with complete removal of the mammographic lesion at 11-gauge vacuum biopsy, subsequent surgery revealed that a percutaneous histologic diagnosis of DCIS resulted in a missed diagnosis of an invasive cancer in 7% of lesions (32) and a percutaneous histologic diagnosis of atypical ductal hyperplasia resulted in a missed diagnosis of carcinoma in 19% (32) and 8% (46) of lesions.

Limitations of our study were that we relied on the histologic presence of microcalcifications despite a negative radiograph of the specimen in the early years of the study, we did not obtain a mammogram immediately after biopsy in all patients with a negative radiograph of the specimen, we did not perform repeat biopsy in all patients with benign lesions and a negative radiograph of the specimen, and we retrospectively recorded the tissue density and the BI-RADS classification for lesions in which biopsy was performed prior to February 1995. We believe that the strengths of our study are the mild selection bias in lesions for which an image-guided biopsy was needed (with 97% of lesions for which patients were referred within our multispecialty clinic undergoing prone stereotactic biopsy), lack of selection bias in the three successive stereotactic biopsy methods, and thoroughness of mammographic follow-up in patients with benign lesions for which they did not undergo repeat biopsy.

In conclusion, failure to retrieve breast microcalcifications was least common with 11-gauge directional vacuum-assisted biopsy; a benign histologic diagnosis with a negative radiograph of the specimen indicates a discordant biopsy result and the need for repeat biopsy.


   ACKNOWLEDGMENTS
 
We thank Jarrett Rosenberg, PhD, for statistical analysis.


   FOOTNOTES
 

Abbreviations: BI-RADS = Breast Imaging Reporting and Data System • DCIS = ductal carcinoma in situ

See Materials and Methods for pertinent disclosures.

Author contributions: Guarantor of integrity of entire study, R.J.J.; study concepts/study design or data acquisition or data analysis/interpretation, R.J.J., J.R.; manuscript drafting or manuscript revision for important intellectual content, R.J.J., J.R.; manuscript final version approval, R.J.J., J.R.; literature research, R.J.J.; clinical studies, R.J.J., J.R.; and manuscript editing, R.J.J., J.R.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Jackman RJ, Marzoni FA. Needle-localized breast biopsy: why do we fail? Radiology 1997;204:677–684.[Abstract/Free Full Text]
  2. Burbank F. Stereotactic breast biopsy: its history, its present, and its future. Am Surg 1996;62:128–150.[Medline]
  3. Parker SH, Burbank F. A practical approach to minimally invasive breast biopsy. Radiology 1996;200:11–20.[Abstract/Free Full Text]
  4. Bassett L, Winchester DP, Caplan RB, et al. Stereotactic core-needle biopsy of the breast: a report of the Joint Task Force of the American College of Radiology, American College of Surgeons, and College of American Pathologists. CA Cancer J Clin 1997;47:171–190.[Abstract]
  5. International Breast Cancer Consensus Conference. Image-detected breast cancer: state of the art diagnosis and treatment. J Am Coll Surg 2001;193:297–302.[CrossRef][Medline]
  6. Liberman L. Percutaneous image-guided core breast biopsy. Radiol Clin North Am 2002;40:483–500.[CrossRef][Medline]
  7. Jackman RJ, Marzoni FA. Stereotactic histologic biopsy with patients prone: technical feasibility in 98% of mammographically detected lesions. AJR Am J Roentgenol 2003;180:785–794.[Abstract/Free Full Text]
  8. American College of Radiology. Illustrated breast imaging reporting and data system (BI-RADS). 3rd ed. Reston, Va: American College of Radiology, 1998.
  9. Jackman RJ, Nowels KW, Rodriguez-Soto J, Marzoni FA, Finkelstein SI, Shepard MJ. Stereotactic, automated, large-core needle biopsy of nonpalpable breast lesions: false-negative and histologic underestimation rates after long-term follow-up. Radiology 1999;210:799–805.[Abstract/Free Full Text]
  10. Lee CH, Philpotts LE, Horvath LJ, Tocino I. Follow-up of breast lesions diagnosed as benign with stereotactic core-needle biopsy: frequency of mammographic change and false-negative rate. Radiology 1999;212:189–194.[Abstract/Free Full Text]
  11. Verkooijen HM, Peeters PHM, Buskens E, et al. Diagnostic accuracy of large-core needle biopsy for nonpalpable breast disease: a meta-analysis. Br J Cancer 2000;82:1017–1021.[CrossRef][Medline]
  12. Burns RP, Brown JP, Roe M, Sprouse LR, Yancey AE, Witherspoon LE. Stereotactic core-needle breast biopsy by surgeons: minimum 2-year follow-up of benign lesions. Ann Surg 2000;232:542–548.[CrossRef][Medline]
  13. Dahlstrom JE, Sutton S, Jain S. Histologic-radiologic correlation of mammographically detected microcalcification in stereotactic core biopsies. Am J Surg Pathol 1998;22:256–259.[CrossRef][Medline]
  14. Meyer JE, Lester SC, Frenna TH, White FV. Occult breast calcifications sampled with large-core biopsy: confirmation with radiography of the specimen. Radiology 1993;188:581–582.[Abstract/Free Full Text]
  15. Liberman L, Evans WP, Dershaw DD, et al. Radiography of microcalcifications in stereotaxic mammary core biopsy specimens. Radiology 1994;190:223–225.[Abstract/Free Full Text]
  16. Liberman L, Dershaw DD, Rosen PP, Abramson AF, Deutch BM, Hann LE. Stereotaxic 14-gauge breast biopsy: how many core biopsy specimens are needed? Radiology 1994;192:793–795.[Abstract/Free Full Text]
  17. Burbank F. Stereotactic breast biopsy: comparison of 14- and 11-gauge Mammotome probe performance and complication rates. Am Surg 1997;63:988–995.[Medline]
  18. Lomoschitz FM, Helbich TH, Rudas M, et al. Stereotactic 11-gauge vacuum-assisted breast biopsy: influence of number of specimens on diagnostic accuracy. Radiology 2004;232:897–903.[Abstract/Free Full Text]
  19. Burbank F. Stereotactic breast biopsy of atypical ductal hyperplasia and ductal carcinoma in situ lesions: improved accuracy with directional, vacuum-assisted biopsy. Radiology 1997;202:843–847.[Abstract/Free Full Text]
  20. Lee CH, Egglin TK, Philpotts L, Mainiero MB, Tocino I. Cost-effectiveness of stereotactic core needle biopsy: analysis by means of mammographic findings. Radiology 1997;202:849–854.[Abstract/Free Full Text]
  21. Liberman L, Hann LE, Dershaw DD, Morris EA, Abramson AF, Rosen PP. Mammographic findings after stereotactic 14-gauge vacuum biopsy. Radiology 1997;203:343–347.[Abstract/Free Full Text]
  22. Meyer JE, Smith DN, DiPiro PJ, et al. Stereotactic breast biopsy of clustered microcalcifications with a directional, vacuum-assisted device. Radiology 1997;204:575–576.[Abstract/Free Full Text]
  23. Park ST, Galbo C, Ghosh BC. Stereotactic breast biopsy as an alternative to excisional biopsy. World J Surg 1997;21:794–798.[CrossRef][Medline]
  24. Liberman L, Smolkin JH, Dershaw DD, Morris EA, Abramson AF, Rosen PP. Calcification retrieval at stereotactic, 11-gauge, directional, vacuum-assisted breast biopsy. Radiology 1998;208:251–260.[Abstract/Free Full Text]
  25. Reynolds HE, Poon CM, Goulet RJ, Lazaridis CL. Biopsy of breast microcalcifications using an 11-gauge directional vacuum-assisted device. AJR Am J Roentgenol 1998;171:611–613.[Free Full Text]
  26. Philpotts LE, Shaheen NA, Carter D, Lange RC, Lee CH. Comparison of rebiopsy rates after stereotactic core needle biopsy of the breast with 11-gauge vacuum suction probe versus 14-gauge needle and automatic gun. AJR Am J Roentgenol 1999;172:683–687.[Abstract/Free Full Text]
  27. Cangiarella J, Waisman J, Symmans WF, et al. Mammotome core biopsy for mammary microcalcification: analysis of 160 biopsies from 142 women with surgical and radiologic followup. Cancer 2001;91:173–177.[CrossRef][Medline]
  28. Liberman L, Benton CL, Dershaw DD, Abramson AF, LaTrenta LR, Morris EA. Learning curve for stereotactic breast biopsy: how many cases are enough? AJR Am J Roentgenol 2001;176:721–727.[Abstract/Free Full Text]
  29. Grimes MM, Karageorge LS, Hogge JP. Does exhaustive search for microcalcification improve diagnostic yield in stereotactic core needle breast biopsies? Mod Pathol 2001;14:350–353.[CrossRef][Medline]
  30. Berg WA, Arnoldus CL, Teferra E, Bhargavan M. Biopsy of amorphous breast calcifications: pathologic outcome and yield at stereotactic biopsy. Radiology 2001;221:495–503.[Abstract/Free Full Text]
  31. Apesteguia L, Mellado M, Saenz J, Cordero JL, Reparaz B, De Miguel C. Vacuum-assisted breast biopsy on digital stereotaxic table of nonpalpable lesions non-recognizable by ultrasonography. Eur Radiol 2002;12:638–645.[Medline]
  32. Liberman L, Kaplan JB, Morris EA, Abramson AF, Menell JH, Dershaw DD. To excise or to sample the mammographic target: what is the goal of stereotactic 11-gauge vacuum-assisted breast biopsy? AJR Am J Roentgenol 2002;179:679–683.[Abstract/Free Full Text]
  33. Han BK, Choe YH, Ko YH, Nam SJ, Kim JH, Yang JH. Stereotactic core-needle biopsy of non-mass calcifications: outcome and accuracy at long-term follow-up. Korean J Radiol 2003;4:217–223.[Medline]
  34. Kettritz U, Rotter K, Schreer I, et al. Stereotactic vacuum-assisted breast biopsy in 2874 patients. Cancer 2004;100:245–251.[CrossRef][Medline]
  35. Parker SH, Lovin JD, Jobe WE, et al. Stereotactic breast biopsy with a biopsy gun. Radiology 1990;176:741–747.[Abstract/Free Full Text]
  36. Parker SH, Lovin JD, Jobe WE, Burke BJ, Hopper KD, Yakes WF. Nonpalpable breast lesions: stereotactic automated large-core biopsies. Radiology 1991;180:403–407.[Abstract/Free Full Text]
  37. Parker SH, Stavros T, Dennis MA. Needle biopsy techniques. Radiol Clin North Am 1995;33:1171–1186.[Medline]
  38. Burbank F, Parker SH, Fogarty TJ. Stereotactic breast biopsy: improved tissue harvesting with the Mammotome. Am Surg 1996;62:738–744.[Medline]
  39. Liberman L, Sama MP. Cost-effectiveness of stereotactic 11-gauge directional vacuum-assisted breast biopsy. AJR Am J Roentgenol 2000;175:53–58.[Abstract/Free Full Text]
  40. Pijnappel RM, Peeters PH, van den Donk M, et al. Diagnostic strategies in non-palpable breast lesions. Eur J Cancer 2002;38:550–555.[CrossRef][Medline]
  41. Liberman L, Drotman M, Morris EA, et al. Imaging-histologic discordance at percutaneous breast biopsy: an indicator of missed cancer. Cancer 2000;89:2538–2546.[CrossRef][Medline]
  42. Pfarl G, Helbich TH, Riedl CC, et al. Stereotactic 11-gauge vacuum-assisted breast biopsy: a validation study. AJR Am J Roentgenol 2002;179:1503–1507.[Abstract/Free Full Text]
  43. Stomper PC, Davis SP, Weidner N, Meyer JE. Clinically occult, noncalcified breast cancer: serial radiologic-pathologic correlation in 27 cases. Radiology 1988;169:621–626.[Abstract/Free Full Text]
  44. Verkooijen HM. Diagnostic accuracy of stereotactic large-core needle biopsy for nonpalpable breast disease: results of a multicenter prospective study with 95% surgical confirmation. Int J Cancer 2002;99:853–859.[CrossRef][Medline]
  45. Jackman RJ, Burbank F, Parker SH, et al. Stereotactic breast biopsy of nonpalpable lesions: determinants of ductal carcinoma in situ underestimation rates. Radiology 2001;218:497–502.[Abstract/Free Full Text]
  46. Jackman RJ, Birdwell RL, Ikeda DM. Atypical ductal hyperplasia: can some lesions be defined as probably benign after stereotactic 11-gauge vacuum-assisted biopsy, eliminating the recommendation for surgical excision? Radiology 2002;224:548–554.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
J.-M. Lee, J. B. Kaplan, M. P. Murray, L. Bartella, E. A. Morris, S. Joo, D. D. Dershaw, and L. Liberman
Imaging Histologic Discordance at MRI-Guided 9-Gauge Vacuum-Assisted Breast Biopsy
Am. J. Roentgenol., October 1, 2007; 189(4): 852 - 859.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Rodriguez-Soto, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Rodriguez-Soto, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE