Stereotactic 11-Gauge Vacuum- Assisted Breast Biopsy: A Validation Study

Transcription

1 Georg Pfarl 1 Thomas H. Helbich 1 Christopher C. Riedl 1 Teresa Wagner 2 Michael Gnant 3 Margaretha Rudas 4 Laura Liberman 5 Received March 11, 2002; accepted after revision May 17, Department of Radiology, University of Vienna Medical School, AKH Wien, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Address correspondence to G. Pfarl. 2 Department of Gynecology, University of Vienna Medical 3 Department of Surgery, University of Vienna Medical 4 Department of Pathology, University of Vienna Medical 5 Breast Imaging Section, Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., New York, NY AJR 2002;179: X/02/ American Roentgen Ray Society Stereotactic 11-Gauge Vacuum- Assisted Breast Biopsy: A Validation Study OBJECTIVE. The aim of our study was to determine the false-negative rate of stereotactic 11- gauge vacuum-assisted biopsy in a validation study of lesions that had subsequent surgical excision. MATERIALS AND METHODS. Retrospective review was performed of 318 lesions that underwent stereotactic 11-gauge vacuum-assisted biopsy and subsequent surgical excision. A false-negative case was defined as a pathologically proven cancer in which stereotactic biopsy yielded benign results without atypia. Medical records, imaging studies, and histologic findings were reviewed. RESULTS. False-negative findings were encountered at stereotactic 11-gauge vacuum-assisted biopsy in 3.3% (7/214) of pathologically proven cancers. False-negative findings occurred in 3.5% (4/115) of malignant lesions versus 3.0% (3/99) of malignant masses ( p = 1.0). The seven false-negative findings included five Breast Imaging Reporting and Data System (BI-RADS) category 5 lesions that yielded benign results at biopsy, one BI- RADS category 4 mass that yielded benign breast tissue, and one BI-RADS category 4 cluster of s in which no s were retrieved. The false-negative rate was 10.0% (6/60) for radiologists who performed 15 or fewer previous stereotactic vacuum-assisted biopsy procedures versus 0.6% (1/154) for radiologists who performed more than 15 previous stereotactic vacuum-assisted biopsy procedures ( p = 0.002). CONCLUSION. Stereotactic 11-gauge vacuum-assisted biopsy had a false-negative rate of 3.3% that diminished to 0.6% with experience. All false-negative findings could be prospectively identified because of failure to sample s or imaging histologic discordance. F or the past decade, stereotactic biopsy has been increasingly used for the histologic diagnosis of breast lesions. In the 1990s, several investigators reported the results of validation studies in which consecutive lesions underwent stereotactic automated core biopsy and subsequent surgical excision [1 8]. These validation studies showed concordance between results of stereotactic 14-gauge automated core biopsy and surgical biopsy in % of cases and reported that stereotactic 14-gauge automated core biopsy had a false-negative rate of 3 8% [1, 2, 5 8]. Stereotactic 11-gauge vacuum-assisted biopsy technology is now available and has several advantages compared with automated core biopsy, including better retrieval [9], fewer histologic underestimates [10 12], and a lower rebiopsy rate [13]. Although numerous clinical investigations of stereotactic 11-gauge vacuum-assisted biopsy have been conducted, there are no (to our knowledge) published validation studies of this method. We present the results of a large series of lesions that underwent stereotactic 11-gauge vacuum-assisted biopsy followed by subsequent surgical excision that can provide validation data for the biopsy technique. Materials and Methods Biopsy Technique Retrospective review was performed of medical records of the first 332 lesions that had stereotactic biopsy with an 11-gauge vacuum-assisted probe (Mammotome; Ethicon Endo-Surgery, Cincinnati, OH) at our institution, the University of Vienna Medical School, a referral center, from September 1997 to December These 332 lesions occurred in 325 patients having a median age of 56 years (range, years). Stereotactic biopsy was offered for histologic diagnosis of lesions that were suspicious for or highly suggestive of malignancy provided that the patient was able AJR:179, December

2 Pfarl et al. TABLE 1 to cooperate with the procedure and did not have a bleeding diathesis. Informed consent was obtained for all stereotactic biopsy procedures. Stereotactic biopsy was introduced at our institution in From 1994 to 1997, all stereotactic biopsies were performed with automated core needles. The 14-gauge vacuum-assisted probe (Mammotome, Ethicon Endo-Surgery) was introduced in July 1997, and the 11-gauge vacuum-assisted biopsy probe was introduced in September During the study period, stereotactic 14-gauge vacuum-assisted biopsy was performed in 521 lesions; the choice of 14- or 11-gauge vacuum-assisted biopsy was made at the discretion of the radiologist performing the biopsy. Among the 332 lesions that underwent stereotactic 11-gauge vacuum-assisted biopsy, 251 had biopsy performed by one of seven attending radiologists specializing in breast imaging, who had performed an average of 2.6 (range, 0 18) stereotactic 14-gauge vacuum-assisted biopsy procedures before the beginning of this study. The remaining 81 lesions underwent biopsy by a resident in training who had not performed stereotactic biopsy before the study period, under the supervision of an attending radiologist. Biopsies were performed with patients prone on a dedicated table (Mammotest; Fischer Imaging, Denver, CO). An average of specimens were obtained per lesion. Among the 332 lesions that had stereotactic 11-gauge biopsy during this period, 318 had subsequent surgical excision and constitute the basis of this article. Mammographic findings in these 318 lesions were s without mass in 166 (52.2%), mass without s in 84 (26.4%), mass with s in 44 (13.8%), asymmetric densities in 19 (6.0%), and architectural distortions in five (1.6%). Median mammographic lesion size was 14.9 mm (range, 3 51 mm). Management Protocol In our institution, our surgeons preferred to use stereotactic biopsy as a preoperative investigation that was performed in addition to (not in lieu of) surgery. If cancer was found at stereotactic biopsy, the surgeon performed a therapeutic operation, including axillary surgery if indicated. If stereotactic biopsy yielded atypia or other specific lesions such as a radial scar or possible phyllodes tumor, the surgeon performed a wide excision. If stereotactic biopsy yielded benign findings concordant with imaging features, the surgeon usually performed a surgical excision but would take less tissue than in an excision performed after stereotactic diagnosis of cancer or atypia. For women with benign lesions, the use of stereotactic biopsy in addition to, rather than in lieu of, surgical biopsy resulted in more procedures and higher cost, but that was the preference of our referring clinicians and was a practice of which patients were aware before undergoing the stereotactic biopsy procedure. Data Collection and Analysis Correlation of Stereotactic Biopsy and Surgical Histology Biopsy Benign Atypical Ductal Hyperplasia Surgery DCIS Note. DCIS = ductal carcinoma in situ. a Calcifications without mass. b Masses with or without s, asymmetric densities, and areas of architectural distortion. Infiltrating Cancer Calcifications a Benign Atypical ductal hyperplasia DCIS Infiltrating cancer Masses b Benign Atypical ductal hyperplasia DCIS Infiltrating cancer All lesions Benign Atypical ductal hyperplasia DCIS Infiltrating cancer Medical records were reviewed by a radiologist specializing in breast imaging to determine the prebiopsy classification of lesion type, size, and level of suspicion according to the Breast Imaging Reporting and Data System (BI-RADS) [14] as probably benign (BI-RADS category 3), suspicious (category 4), or highly suggestive of malignancy (category 5). Medical records and histologic findings were reviewed to determine surgical outcomes. Stereotactic biopsy results were considered discordant if the histologic findings did not provide a sufficient explanation for the imaging features [15]. A pathologically proven cancer was a lesion that yielded cancer at either stereotactic biopsy, surgery, or both, confirmed at subsequent pathology review. A false-negative finding was defined as a pathologically proven cancer in which stereotactic biopsy yielded benign findings with no atypia [16]. Data were entered into a computerized spreadsheet (Excel; Microsoft, Redmond, WA) for analysis. Statistical analyses were performed with statistical software (Epi-Info; Centers for Disease Control, Atlanta, GA) using the chi-square and Fisher s exact tests. Results Stereotactic and Surgical Histology In these 318 lesions, stereotactic biopsy yielded benign findings in 100 (31.4%), atypical ductal hyperplasia in 17 (5.3%), ductal carcinoma in situ (DCIS) in 91 (28.6%), and infiltrating cancer in 110 (34.6%) (Table 1). Of 201 stereotactically proven cancers, surgery revealed cancer in 199; in two stereotactically proven cancers, stereotactic biopsy yielded DCIS but surgery showed atypical ductal hyperplasia and biopsy site change. Histologic review of these two cases confirmed the presence of DCIS in the stereotactic biopsy material, suggesting that the DCIS was completely removed at stereotactic biopsy. No false-positive findings were encountered. Among 17 lesions yielding atypical ductal hyperplasia at stereotactic 11-gauge vacuumassisted biopsy, surgery revealed carcinoma in six (35.3%). Among 91 lesions yielding DCIS at stereotactic 11-gauge vacuum-assisted biopsy, surgery revealed invasive carcinoma in 11 (12.1%). Among two lesions yielding papilloma, two lesions yielding phyllodes tumor, and one lesion yielding a radial scar at stereotactic biopsy, all were confirmed at surgery to be benign. Among 13 lesions yielding discordant results, surgery revealed carcinoma in seven (53.8%). The final histologic diagnosis, based on review of stereotactic and subsequent surgical pathology, was cancer in 214 lesions, including 201 lesions that yielded cancer at stereotactic biopsy, six lesions that yielded atypical ductal hyperplasia at stereotactic biopsy, and seven lesions in which stereotactic biopsy yielded benign findings without atypia (Table 1) AJR:179, December 2002

3 Stereotactic Breast Biopsy False-Negative Findings False-negative findings were encountered in seven lesions (Table 2). The median mammographic size of these seven lesions was 0.9 cm (range, cm). Mammographic findings in these seven cases were s in four, mass in two, and both in one. Five (71%) of the seven missed cancers were prospectively classified as BI-RADS category 5 and two (29%), as BI-RADS category 4. The median number of specimens obtained in these seven lesions was 16 (range, 15 20). Stereotactic biopsy histology in these seven false-negative cases was benign breast tissue in four and fibrocystic mastopathy in three (including two with duct hyperplasia). Surgical histology in these missed cancers was infiltrating ductal carcinoma in four, infiltrating lobular carcinoma in one, and DCIS in two. The median histologic size of infiltrating carcinoma was 1.1 cm (range, cm). In all patients with infiltrating carcinoma, the lymph nodes were negative for cancer. In two missed cancers that were evident as clusters of pleomorphic s (BI- RADS category 5), s were present on specimen radiographs but were not seen histologically. Stereotactic biopsy revealed benign breast tissue without s in both cases; surgical excision yielded DCIS in one and infiltrating ductal carcinoma in the other. In one missed cancer that was evident as a cluster of amorphous s (BI-RADS category 4), specimen radiography was not performed (a deviation from our protocol); no s TABLE 2 were seen at histologic analysis. Stereotactic biopsy revealed fibrocystic mastopathy; surgical excision showed infiltrating ductal carcinoma. In one missed cancer that was evident as a cluster of pleomorphic s (BI- RADS category 5), s were identified at specimen radiography and at histologic analysis; stereotactic biopsy yielded benign breast tissue with s, and surgery revealed DCIS. Review of stereotactic images in that case suggested that the s that were targeted did not correspond to the most suspicious s in that region. In one missed cancer evident as a spiculated mass with s (BI-RADS category 5), s were not identified at either specimen radiography or histologic analysis; stereotactic biopsy yielded fibrocystic change without s, and surgical excision yielded infiltrating ductal carcinoma. In one missed cancer evident as a spiculated mass (BI-RADS category 5), stereotactic biopsy revealed benign breast tissue, and surgery showed infiltrating ductal carcinoma. Review of the stereotactic images suggests that needle placement may have been suboptimal. In one missed cancer evident as an indistinct mass (BI-RADS category 4), the lesion was subtle and difficult to target; on the postbiopsy mammogram, the area sampled appeared to be slightly caudad to the lesion. Stereotactic biopsy revealed benign breast tissue; surgery revealed infiltrating lobular carcinoma. False-negative findings were encountered in seven of 214 histologically proven cancers Cancers Missed at Stereotactic 11-Gauge Vacuum-Assisted Biopsy in this study, for a false-negative rate of 3.3%. False-negative findings were encountered in 3.5% (4/115) of malignant lesions versus 3.0% (3/99) of malignant masses ( p = 1.0). In six (85.7%) of the seven false-negative cases, the biopsy was performed by an operator who had previously performed 15 or fewer stereotactic vacuumassisted biopsy procedures (Table 2). The false-negative rate was 10.0% (6/60) for radiologists who performed 15 or fewer previous stereotactic vacuum-assisted biopsy procedures versus 0.6% (1/154) for radiologists who performed more than 15 previous stereotactic vacuum-assisted biopsy procedures ( p = 0.002). Discussion In pioneering studies in the 1990s, Parker et al. [1, 2] and others [3 8] reported the results of lesions that underwent stereotactic automated core biopsy and subsequent surgical excision (Table 3). These validation studies included 1043 lesions, of which 387 were malignant [1 8]. Among the studies that used only 14-gauge automated needles [2, 5, 6], false-negative rates ranged from 3% to 8% (mean, 6%) (Table 3). When stereotactic vacuum-assisted biopsy was introduced [17], stereotactic automated core biopsy was in widespread clinical use. Numerous clinical investigations of stereotactic vacuum-assisted biopsy were performed, but no validation study in consecutive lesions with surgical correlation has yet (to our knowledge) been reported. Case Previous Biopsies BI-RADS Specimen Lesion Type Size (cm) Description Category c Histology d Diagnosis All a 11-Gauge b Radiograph Biopsy Surgery Calcification 0.7 Amorphous 4 Not done No Fibrocystic mastopathy IFDC Calcification 2.0 Pleomorphic 5 Calcification Yes Benign breast tissue, DCIS Calcification 0.5 Pleomorphic 5 Calcification No Benign breast tissue DCIS Calcification 1.5 Pleomorphic 5 Calcification No Fibrocystic mastopathy, IFDC duct hyperplasia Mass 0.9 Indistinct 4 NA NA Benign breast tissue ILC Mass, 1.0 Spiculated, amorphous 5 No Note. IFDC = infiltrating ductal carcinoma, DCIS = ductal carcinoma in situ, NA = not applicable, ILC = infiltrating lobular carcinoma. a Total number of previous stereotactic vacuum-assisted biopsies with either 14- or 11-gauge probes performed by the radiologist before performing biopsy in this case. No Fibrocystic mastopathy, duct hyperplasia Mass 0.5 Spiculated 5 NA NA Benign breast tissue IFDC IFDC, fibrocystic mastopathy, b Number of previous stereotactic 11-gauge vacuum-assisted biopsies performed by the radiologist before performing biopsy in this case. Note that all missed cancers (false-negatives) occurred among radiologists who had performed six or fewer previous stereotactic 11-gauge vacuum-assisted biopsy procedures. c Breast Imaging Reporting and Data System [14]. d Whether s were identified at histologic analysis. AJR:179, December

4 Pfarl et al. In our study, we found false-negative cases in seven (3.3%) of 214 pathologically proven cancers that underwent stereotactic 11-gauge vacuum-assisted breast biopsy. Among radiologists who had performed more than 15 previous stereotactic biopsies, the false-negative rate was 0.6%. These false-negative rates, which include our learning experience for the 11-gauge method, compare favorably with the 3 8% range reported in the validation studies of stereotactic 14-gauge automated core biopsy [2, 5, 6], as well as the 0 8% (mean, 2%) range of false-negative rates reported in studies of needle localization and surgical biopsy [18]. All missed cancers could be recognized prospectively because of failure to identify calcium on specimen radiographs, failure to identify calcium at histologic analysis, imaging histologic discordance, or a combination of these features. Imaging histologic discordance occurred in all seven missed cancers at stereotactic 11- gauge vacuum-assisted biopsy. These seven lesions included five BI-RADS category 5 lesions that yielded benign results, one BI- RADS category 4 mass that yielded benign breast tissue, and one BI-RADS category 4 cluster of s in which no s were retrieved. Imaging histologic discordance has been reported in 0 6% TABLE 3 Investigator Validation Studies of Stereotactic Biopsy Year No. of Patients Needle No. Malignant No. (%) a False-Negative Parker et al. [1] Variable automated b 16 1 (6.3) Parker et al. [2] gauge automated 23 1 (4.3) Dowlatshahi et al. [3] gauge automated (25.0) c Dronkers [4] gauge automated 45 2 (4.4) d Elvecrog et al. [5] gauge automated 35 e 1 (2.9) Gisvold et al. [6] gauge automated 64 f 5 (7.8) Helbich et al. [7] Variable automated g 98 3 (3.1) Helbich et al. [8] Variable automated h 30 0 (0) This study gauge vacuum-assisted (3.3) a Proportion of pathologically proven cancers in which the stereotactic biopsy histology was benign without atypia. b Gauge was 18 in 65 lesions, 16 in nine lesions, and 14 in 29 lesions. c In an additional 11 (14.5%) of 76 cancers, stereotactic biopsy material was insufficient for diagnosis [3]. Every lesion in this study underwent fine-needle aspiration biopsy before core biopsy. d In an additional two (4.4%) of 45 cancers, stereotactic biopsy was nondiagnostic, with no representative material obtained because of insufficient sampling [4]. e Excludes one case of lobular carcinoma in situ. f Excludes three cases of lobular carcinoma in situ. g 16-gauge needle in 21 lesions and 14-gauge needle in 189 lesions. Guidance was stereotaxis in 149 (with the patient prone in 82 cases, sitting in 67) lesions and sonography in 61 lesions. All false-negative findings occurred in biopsies performed under stereotactic guidance (sitting in two, prone in one). h Six passes were performed for each lesion, two with 14-gauge, two with 16-gauge, and two with 18-gauge needles, in random order. Although no false-negative diagnoses were obtained with 14-gauge needle, 16-gauge needle gave false-negative results in two (6.7%) of 30 cancers, and 18-gauge needle gave false-negative results in two (6.7%) of 30 cancers. (mean, 4%) of lesions that had percutaneous imaging-guided biopsy in previous studies; among discordant lesions, cancer has been found in 0 64% (mean, 18%) [13, 15, 16, 19 21]. Imaging histologic discordance is an indication for repeating biopsy because of the high prevalence of carcinoma in these discordant lesions [15]. Failure to identify s on specimen radiographs, which occurred in one of our false-negative cases, has been previously shown to correlate with nondiagnostic results at stereotactic biopsy. Liberman et al. [22] found that the likelihood of obtaining diagnostic material at stereotactic biopsy was significantly greater if s were present on specimen radiographs than if they were absent (81% vs 38%, p < 0.001). If the stereotactic biopsy is performed for s and yields benign findings with no s on specimen radiography, a repeated biopsy is generally warranted. Identification of s at histologic analysis in the absence of s on the specimen radiographs is usually not adequate, because s may be seen microscopically in lesions that lack s on the mammogram [23, 24]. Failure to identify s at histologic analysis occurred in four calcific lesions that yielded false-negative results. In two of these four cases, s were identified at specimen radiography but not at histologic analysis. In this scenario, which can also occur at surgical biopsy, the pathologist should obtain deeper levels from the biopsy material and should look at the tissue with polarized light to search for calcium oxalate (weddellite) crystals [22, 25]. If these maneuvers fail to yield s, the need for rebiopsy is determined on the basis of the specific imaging findings before and after biopsy and the histologic diagnosis. If review of the pre- and postbiopsy mammograms and specimen radiographs indicates that the s were well sampled and the diagnosis is otherwise concordant, rebiopsy may not be necessary; however, if the sampling was poor or the diagnosis is discordant, rebiopsy is indicated. False-negative cases were significantly more frequent among radiologists who had performed 15 or fewer previous stereotactic biopsy procedures than among radiologists who had performed more than 15 procedures (10.0% vs 0.6%, p = 0.002). Liberman et al. [26] described a learning curve for stereotactic 11-gauge vacuum-assisted breast biopsy, with a greater false-negative rate for the first 15 cases than for subsequent cases (7.4% vs 0%, p < 0.06). In that study, the learning curve existed for stereotactic 11-gauge vacuum-assisted biopsy among radiologists who had a prior collective experience of more than 500 cases with stereotactic 14-gauge automated core or 14-gauge vacuum-assisted biopsy [26]. Our findings lend further support to the importance of operator experience in achieving optimal results for stereotactic biopsy and reinforce the need for training and supervision early in one s experience performing the procedure. In conclusion, we found that stereotactic 11-gauge vacuum-assisted breast biopsy had a false-negative rate of 3.3%, which decreased to 0.6% with experience. All falsenegative findings could be prospectively identified because of failure to sample s or imaging histologic discordance. These findings, which show accuracy comparable to that of surgical biopsy, validate the use of stereotactic 11-gauge biopsy as an alternative to needle localization and surgical excision, as currently practiced in many centers throughout the world. Our data indicate the importance of experience in optimizing the outcome of stereotactic 11-gauge vacuum-assisted biopsy. Careful attention to 1506 AJR:179, December 2002

5 Stereotactic Breast Biopsy identification of s at specimen radiography and at histologic analysis for calcific lesions and imaging histologic correlation for all lesions can enable prospective identification of missed cancers and avoid a delay in diagnosis. References 1. Parker SH, Lovin JD, Jobe WE, et al. Stereotactic breast biopsy with a biopsy gun. Radiology 1990;176: Parker SH, Lovin JD, Jobe WE, Burke BJ, Hopper KD, Yakes WF. Nonpalpable breast lesions: stereotactic automated large-core biopsies. Radiology 1991;180: Dowlatshahi K, Yaremko ML, Kluskens LF, Jokich PM. Nonpalpable breast lesions: findings of stereotaxic needle-core biopsy and fine-needle aspiration cytology. Radiology 1991;181: Dronkers DJ. Stereotaxic core biopsy of breast lesions: correlation of stereotaxic large-core needle biopsy and surgical biopsy results. Radiology 1992;183: Elvecrog EL, Lechner MC, Nelson MT. Nonpalpable breast lesions: correlation of stereotaxic large-core needle biopsy and surgical biopsy results. Radiology 1993;188: Gisvold JJ, Goellner JR, Grant CS, et al. Breast biopsy: a comparative study of stereotaxically guided core and excisional techniques. AJR 1994;162: Helbich TH, Mayr W, Schick S, et al. Coaxial technique: approach to breast core biopsies. Radiology 1997;203: Helbich TH, Rudas M, Haitel A, et al. Evaluation of needle size for breast biopsy: comparison of 14-, 16-, and 18-gauge biopsy needles. AJR 1998;171: Reynolds HE, Poon CM, Goulet RJ, Lazaridis CL. Biopsy of breast micros using an 11-gauge directional vacuum-assisted device. AJR 1998;171: Jackman RJ, Burbank F, Parker SH, et al. Atypical ductal hyperplasia diagnosed by 11-gauge, directional, vacuum-assisted breast biopsy: how often is carcinoma found at surgery? (abstr) Radiology 1997;205(P): Meyer JE, Smith DN, Lester SC, et al. Large-core needle biopsy of nonpalpable breast lesions. JAMA 1999;281: 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: 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 1999;172: American College of Radiology. Illustrated breast imaging reporting and data system (BI- RADS), 3rd ed. Reston, VA: American College of Radiology, Liberman L, Drotman MB, Morris EA, et al. Imaging-histologic discordance at percutaneous breast biopsy. Cancer 2000;89: Liberman L, Dershaw DD, Glassman J, et al. Analysis of cancers not diagnosed at stereotactic core breast biopsy. Radiology 1997;203: Burbank F, Parker SH, Fogarty TJ. Stereotactic breast biopsy: improved tissue harvesting with the Mammotome. Am Surg 1996;62: Jackman RJ, Marzoni FA. Needle-localized breast biopsy: why do we fail? Radiology 1997;204: Dershaw DD, Morris EA, Liberman L, Abramson AF. Nondiagnostic stereotaxic core breast biopsy: results of rebiopsy. Radiology 1996;198: Meyer JE, Smith DN, Lester SC, et al. Large-core needle biopsy: nonmalignant breast abnormalities evaluated with surgical excision or repeat core biopsy. Radiology 1998;206: Liberman L, Feng TL, Dershaw DD, Morris EA, Abramson AF. Ultrasound-guided core breast biopsy: utility and cost-effectiveness. Radiology 1998;208: Liberman L, Evans WP III, Dershaw DD, et al. Radiography of micros in stereotaxic mammary core biopsy specimens. Radiology 1994;190: Stomper PC, Davis SP, Weidner N, Meyer JE. Clinically occult, noncalcified breast cancer: serial radiologic-pathologic correlation in 27 cases. Radiology 1988;169: Dahlstrom JE, Sutton S, Jain S. Histologic-radiologic correlation of mammographically detected micro in stereotactic core biopsies. Am J Surg Pathol 1998;22: D Orsi CJ, Reale FR, Davis MA, Brown UJ. Is calcium oxalate an adequate explanation for nonvisualization of breast specimen s? Radiology 1992;182: Liberman L, Benton CL, Dershaw DD, Abramson AF, LaTrenta LR, Morris EA. Learning curve for stereotactic breast biopsy: how many cases are enough? AJR 2001;176: AJR:179, December