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


     

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 Google Scholar
Google Scholar
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Shepard, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Shepard, M. J.
(Radiology. 1999;210:799-805.)
© RSNA, 1999

Breast Imaging

Stereotactic, Automated, Large-Core Needle Biopsy of Nonpalpable Breast Lesions: False-Negative and Histologic Underestimation Rates after Long-term Follow-up

Roger J. Jackman, MD1, Kent W. Nowels, MD4, Jorge Rodriguez-Soto, MD2, Francis A. Marzoni, Jr, MD3, Solon I. Finkelstein, MD1 and Michael J. Shepard, MD1

1 Departments of Radiology (R.J.J., S.I.F., M.J.S.)
2 Pathology (J.R.S.)
3 Surgery (F.A.M.), Palo Alto Medical Clinic, 300 Homer Ave, Palo Alto, CA 94301
4 Department of Pathology, Stanford University Medical Center, Stanford, Calif (K.W.N.).


   Abstract
TOP
 Abstract
Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To determine the rate and causes of false-negative findings and histologic underestimates at stereotactic biopsy of nonpalpable breast lesions.

MATERIALS AND METHODS: Stereotactic, 14-gauge, automated, large-core needle biopsy (LCNB) was performed in 483 consecutive nonpalpable breast lesions. Excision was advised for the 143 carcinomas, 25 atypical ductal hyperplasia (ADH) lesions, and five radial scars. Mammographic follow-up was advised for the benign lesions without a repeat biopsy.

RESULTS: Of the 310 benign lesions, 259 underwent mammographic follow-up at 6–85 months (median, 55 months) without repeat biopsy, 48 underwent repeat biopsy and three were lost to follow-up. On the basis of the histologic diagnosis of carcinoma at surgical biopsy, diagnosis with LCNB was not correct (ie, disease was underestimated at histologic examination) in 14 (58%) of 24 ADH lesions and two (40%) of five radial scars. Two (1.2%) of 161 lesions with a final diagnosis of carcinoma were benign at LCNB but malignant at repeat biopsy (ie, false-negative findings at LCNB). Repeat biopsy was prompted by mammographic progression at 6 and 18 months after LCNB.

CONCLUSION: The false-negative rate with LCNB was 1.2% in this study and 4.0% in the literature. The presence of carcinoma in ADH and radial scar lesions was often underestimated.

Index terms: Biopsies, technology, 00.126 • Breast, biopsy, 00.126 • Breast, diseases, 00.31, 00.32 • Breast, ducts, 00.3119 • Breast, radial scar, 00.3119 • Breast neoplasms, diagnosis, 00.126


   Introduction
TOP
 Abstract
 Introduction
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
A recent review of needle-localized breast biopsy (NLBB) found an overall miss rate of 0%–18% (mean, 2.6%) and a cancer miss rate (ie, false-negative rate) of 0%–8% (approximate mean, 2.0%) (1). Percutaneous biopsy with automated, large-core needle biopsy (LCNB) (2–4) is a popular alternative to NLBB for mammographically detected lesions. The findings of standardized 14-gauge LCNB (performed with stereotactic guidance, prone table, long-throw gun, and extraction of at least five specimens per lesion) (5) have been correlated with those from NLBB immediately after LCNB. The LCNB overall miss rate was 2%–4% (mean, 3.3%) (5–7), and the false-negative rate was 2.9%–6.7% (mean, 4.4%) (5–8).

Several investigators have reported false-negative rates of 0.3%–8.2% (mean, 4.2%) in LCNB series with incomplete surgical and mammographic follow-up (9–16). Mammographic follow-up for a minimum of 2–3 years and repeat biopsy of lesions showing progression are needed before actual false-negative rates can be determined.

The purpose of our study was to determine the rate and causes of false-negative findings and histologic underestimates at stereotactic LCNB of 483 consecutive nonpalpable breast lesions in which repeat biopsy or long-term mammographic follow-up was performed.


   MATERIALS AND METHODS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
We retrospectively reviewed the histologic slides from 483 consecutive, nonpalpable, mammographically detected lesions in which percutaneous stereotactic biopsy was performed from July 1991 through December 1993. The 483 lesions were in 410 patients (age range, 29–89 years; median age, 55 years). One discretely palpable lesion was excluded from analysis. Percutaneous stereotactic biopsy was offered as an alternative to NLBB. After 2 months of experience, no lesions were rejected for attempt at percutaneous biopsy because of lesion size, breast size, or position of the lesion in the breast. During the same 21/2-year period, NLBB was performed for the initial diagnosis in an additional 17 (3%) of the 500 nonpalpable mammographically detected lesions that were sampled for biopsy. NLBB was performed because of suggestion by the radiologist in seven lesions, preference by the patient in four lesions, and preference by the surgeon in three lesions. The other three lesions could not be adequately visualized during attempted LCNB and, thus, NLBB was performed instead. None of the patients underwent percutaneous biopsy of a nonpalpable lesion by means of fine-needle aspiration or with ultrasonographic (US) guidance. Each patient gave informed consent before undergoing a biopsy.

Our LCNB technique has been previously described (17). In brief, stereotactic biopsy was performed with the patients on a prone table (Mammotest; Fischer Imaging, Denver, Colo) by using a 14-gauge needle (ACN 1416; Manan Medical Products, Northbrook, Ill) and a long-throw (2.3 cm excursion) biopsy gun (Biopty; Bard Urological, Covington, Ga). At least five specimens were obtained in 480 of the 483 lesions. Overall, we obtained a mean of 8.1 specimens (range, two to 20 specimens) per lesion. We obtained mammographic images on radiographic film rather than on a digital receptor during this study. We did not obtain radiographs of the specimens to verify extraction of calcifications until June 30, 1993. Radiographs of the specimen were obtained in 70 (30%) of the 234 lesions detected as calcifications.

Mammographic lesions were divided into two groups: calcifications (without masses) and masses. Masses included asymmetric densities, areas of architectural distortion, and other space-occupying lesions (all with or without associated calcifications). The maximum diameter of each lesion was measured on the mammogram.

Histologic diagnoses of the core biopsy specimens were divided into three categories: carcinoma, high-risk lesions, and benign lesions. Carcinomas were subcategorized as invasive ductal carcinoma, invasive lobular carcinoma, and ductal carcinoma in situ (DCIS). High-risk lesions were those known to have a high prevalence of carcinoma at excision, and the diagnosis was based on the histologic characteristics seen at LCNB alone, without regard to correlation of mammographic and histologic findings. Our high-risk lesions were ADH and radial scars. Benign lesions (ie, those that were not carcinoma or high-risk lesions) were subcategorized as fibrocystic change, fibrosis, fibroadenoma, lymph node, and nonspecific benign lesion. Any patient with a histologic diagnosis of carcinoma, high-risk lesion, benign lesion of concern to the pathologist, or benign lesion thought to be discordant with the mammographic findings was advised to undergo lesion excision. Two pathologists (K.W.N., J.R.S.) retrospectively reviewed the histologic slides from lesions diagnosed as either high-risk or benign at LCNB and in which subsequent surgical excision was performed. The diagnosis was reached by consensus. At the retrospective review, the pathologists knew each lesion was later excised but did not know the excisional diagnosis. Whenever a discrepancy existed between the histologic features of the LCNB specimen and those of the excisional biopsy specimen, the slides were reviewed a second time to ensure interpretive consistency. The same histologic criteria were used in both the prospective and retrospective reviews (17). All slides were examined for the histologic presence of calcifications.

The mammographic findings were compared with the retrospective histologic findings from LCNB for carcinoma (n = 143), high-risk lesion (n = 30), and benign lesion (n = 310); the histologic subtypes were included for each category (Table 1).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Comparison of Histologic Findings at LCNB and Mammographic Findings
 
A diagnosis of high-risk lesion based on LCNB findings was considered to be a "histologic underestimate" if the excisional specimen contained carcinoma. The underestimation rate for each histologic type of high-risk lesion was determined by dividing the total number of lesions in which excisional biopsy was performed into the number of those lesions found to be carcinoma at excision. A benign diagnosis based on LCNB findings was considered to be a false-negative finding if the specimen from a repeat biopsy of the same lesion contained carcinoma. The false-negative rate was determined by dividing the total number of carcinomas found with any biopsy method into the number of those lesions that were originally benign at LCNB. A diagnosis of carcinoma based on LCNB findings was considered to be a true-positive finding with or without confirmation by means of excisional biopsy.

Any patient with a benign histologic diagnosis who did not subsequently undergo repeat biopsy was advised to undergo mammographic follow-up. Patients with a percutaneous diagnosis of fibroadenoma or lymph node were thought to be very unlikely to have a sampling error and were advised to undergo bilateral mammography at 12 months. Patients with other benign lesions were advised to undergo unilateral mammography of the biopsied breast 6 months after biopsy (17). Patients with all benign lesions were advised to undergo bilateral mammography at 12, 24, 36, 48, 60, and 72 months after biopsy. If mammographic follow-up showed a lesion to have grown or contain more calcifications, the patient was advised to undergo repeat biopsy with either LCNB or NLBB.

Our protocol for mammographic follow-up evolved as we gained experience. From the start, patients were informed that mammographic follow-up was necessary for a histologically benign lesion, and referring physicians received the radiologist's written biopsy report requesting mammographic follow-up at either 6 or 12 months. Because of poor compliance in the 1st year, we adopted more vigorous efforts to achieve the desired follow-up. The potential of a biopsy sampling error and the attendant need for mammographic follow-up were more clearly emphasized to each patient before the biopsy was performed, both orally and in writing.

With the agreement of all the referring physicians in our multispecialty clinic, the primary responsibility for contacting patients in need of mammographic follow-up was shifted from the referring physicians to one individual trained in medical record procedures. This "follow-up analyst" was also responsible for contacting patients in need of follow-up because of a variety of abnormal imaging studies (primarily mammograms) and cytologic studies (mainly Papanicolaou smears).

Patients in need of follow-up mammography were sent a reminder letter telling them when it should be performed. If no appointment was made within 2 weeks, a reminder phone call was made offering assistance with scheduling. If no appointment was made within 1 more week, a certified letter (with return receipt) was sent as the third reminder. If there was still no response, the referring physician called or wrote the patient. Occasionally, one radiologist (R.J.J.) who performed the LCNBs also directly contacted the patient. All contacts were documented in the medical record. If there was still no compliance with our request, the four- or five-step contact procedure was usually repeated every 6–12 months until the process was completed. With the agreement of most referring physicians outside our multispecialty clinic, that same process was extended to their patients. The process was also repeated each time follow-up mammography (usually done annually) was due. Each contact was considered to be complete when mammography was performed, the lesion underwent repeat biopsy, the referring physician documented that follow-up mammography was no longer medically indicated, the patient's firm "informed refusal" was documented, the patient was deceased, or the patient had moved to an unknown location and was lost to follow-up.


   RESULTS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
In the retrospective review of histologic slides of specimens obtained with LCNB, which was performed to ensure interpretive consistency, four lesions were downgraded from ADH to fibrocystic change, three were upgraded from fibrocystic change to ADH, and two were upgraded from ADH to DCIS. The histologic diagnosis was not changed for any radial scars.

Repeat biopsy was performed before mammographic follow-up in 12 of the 310 benign lesions 1–17 weeks after LCNB: Surgical biopsy was performed in 11 lesions, and vigorous repeat LCNB was performed in one lesion. Six of the 12 benign lesions were incidentally resected when a carcinoma (proved with LCNB) in the same breast underwent a therapeutic operation (mastectomy in five lesions and lumpectomy in one lesion). In each of those six breasts, the benign lesion and the malignant lesion were discretely separate at mammography. Diagnostic repeat biopsy in the other six lesions was performed because of hyperplasia with histologic features bordering on ADH early in our experience (n = 4), histologic slides negative for calcifications (n = 1), and a hemangioma with histologic features bordering on angiosarcoma (n = 1). All 12 lesions that underwent repeat biopsy were diagnosed as benign at histologic examination. When these 12 lesions are eliminated, 298 benign lesions were eligible for mammographic follow-up.

Mammographic follow-up was performed in 295 (99%) of the 298 benign lesions without immediate repeat biopsy. The initial post-LCNB mammogram was not obtained for 13–18 months in 33 (11%) of the 295 lesions and for more than 18 months in 11 (4%).

Repeat biopsy was performed after mammographic follow-up in 36 (12%) of the 295 benign lesions: Surgical biopsy was performed in 19 lesions, and vigorous repeat LCNB was performed in 17. Five of the 36 benign lesions were incidentally resected when a carcinoma in the same breast (proved with LCNB of a new mammographic lesion in four cases and by means of diagnostic lumpectomy of a new palpable mass in one case) underwent a therapeutic operation: Mastectomy was performed in two cases, and therapeutic lumpectomy was performed in three cases. No progression was evident in those five benign lesions on mammograms obtained 36–48 months after LCNB that preceded incidental resection. There was mammographic progression in 26 (9%) of the 295 benign lesions for which mammographic follow-up was performed. This occurred in 19 (12%) of 157 calcification lesions and seven (5%) of 138 mass lesions. All 26 lesions underwent repeat biopsy, nine with NLBB and 17 with vigorous repeat LCNB.

Two of the 26 lesions with mammographic progression were diagnosed as carcinoma at repeat biopsy: One was invasive ductal carcinoma, and one was DCIS. These were the only false-negative findings obtained with LCNB in the retrospective study, and the patient, lesion, and procedural variables are shown in Table 2. Surgical excision was performed in the five remaining benign lesions with mammographic follow-up despite mammographic stability because of physician concern (n = 3), patient concern (n = 1), and incidental resection during reduction mammoplasty (n = 1). All five lesions were diagnosed as benign at repeat biopsy.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Patient, Lesion, and Procedural Variables for Five Radial Scar Lesions, Two False-Negative Lesions, and Three Lesions that Were Retrospectively Upgraded from Fibrocystic Change to ADH
 
Mammographic follow-up was performed in 259 (99%) of the 262 benign lesions without repeat biopsy (Table 3). The most recent mammogram was obtained 6–85 months (median, 55 months) after LCNB. Patients who underwent only short-term follow-up had specific problems preventing long-term follow-up (Table 3). Mammographic follow-up was not performed in two patients (with a total of three lesions) because they were lost to follow-up.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Months of Post-LCNB Mammographic Follow-up and Follow-up Problems in 259 Benign LCNB Lesions without a Repeat Biopsy
 
Repeat biopsy was performed in 24 of the 25 patients with ADH lesions. One patient with ADH died after LCNB from unrelated causes and, thus, did not undergo repeat biopsy. Repeat biopsy of the other 24 ADH lesions revealed carcinoma in 14 (58%): Nine lesions were DCIS and five were invasive ductal carcinoma (with associated DCIS). The patient, lesion, and procedural variables for the three lesions upgraded in the retrospective histologic review from fibrocystic change to ADH are shown in Table 2 and are included in the 14 ADH lesions with carcinoma at repeat biopsy. Repeat biopsy was performed in these three lesions because of mammographic progression of calcifications (n = 2) or a mass with calcifications (n = 1) 10–38 months after LCNB. If these three lesions had not been upgraded at retrospective histologic review, the false-negative rate would have been 3.1% (five of 161 lesions) instead of 1.2% (two of 161 lesions).

Excisional biopsy was not initially recommended after an LCNB diagnosis of ADH in the other 21 lesions until we became aware of the high-risk potential. In nine such lesions, repeat biopsy was not performed for 9–53 months after LCNB. The mammogram obtained immediately before repeat biopsy showed the lesion to have mammographically progressed (n = 7) or mammographically decreased (n = 2) compared to the lesion on the pre-LCNB mammogram. In the other 12 lesions, repeat biopsy was performed (either because the high-risk potential was known or because of incidental resection with a carcinoma in the same breast) 1–8 weeks after LCNB without a post-LCNB mammogram.

Repeat biopsy of the five radial scar lesions, the variables of which are shown in Table 2, revealed carcinoma in two (40%): One lesion was DCIS, and one was invasive ductal carcinoma. The high-risk potential became evident during the course of this study, and repeat biopsy was performed 6–34 months after LCNB.

Therapeutic surgery was performed in 75 of the 76 invasive ductal carcinomas (with one patient found to have metastatic disease after LCNB), 10 of 11 invasive lobular carcinomas (with one elderly patient treated with tamoxifen only), and 54 of 56 DCIS lesions (with one elderly patient with two areas of DCIS treated with tamoxifen only). Therapeutic surgery of the 54 DCIS lesions revealed invasive ductal carcinoma (in addition to DCIS) in eight (15%).


   DISCUSSION
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
It is important to compare the accuracies of various biopsy methods. The accuracy of NLBB, the standard of reference for biopsy of nonpalpable breast lesions, was recently reviewed (1). The overall miss rate (ie, failure to remove any portion of the mammographically detected lesion) was 0%–18% (mean, 2.6%). The cancer miss rate (ie, failure to diagnose a malignancy in a lesion later proved to be malignant) was 0%–8% (approximate mean, 2.0%). Because of inadequate follow-up in most series, the cancer miss rate was harder to determine and may be underestimated.

Compared to NLBB, percutaneous LCNB is a less invasive, less costly alternative for histologic diagnosis of mammographically detected breast lesions. Parker et al (18) tested various core biopsy methods and then established a standardized approach (5) that included a prone biopsy table, stereotactic guidance system, long-throw (2.3 cm excursion) biopsy gun, 14-gauge cutting needles, and removal of five or more tissue samples per lesion (Parker SH, written communication, 1991). The accuracy of that standardized 14-gauge LCNB approach has been evaluated with NLBB immediately after core biopsy. The overall miss rate for LCNB was 2%–4% (mean, 3.3%) (5–7), and the false-negative rate was 2.9%–6.7% (mean, 4.4%) (5–8) (Table 4).


View this table:
[in this window]
[in a new window]
  		TABLE 4. LCNB Miss Rates in Lesions with NLBB Correlation
 
Herein, we report on the accuracy of stereotactic LCNB performed in 483 consecutive mammographically detected nonpalpable breast lesions. These lesions represent 97% of the 500 nonpalpable breast lesions sampled for biopsy at our institution during the 21/2-year study period.

Among the 483 lesions were 298 histologically benign lesions eligible for mammographic follow-up. With a costly, time-consuming process that is probably not practical to adopt as a standard approach, we accomplished at least one mammographic follow-up in 295 (99%) of the 298 lesions.

The terminology for the histologic accuracy of LCNB is evolving. We agree with a recent joint task force report that a carcinoma diagnosed with LCNB should be categorized as a true-positive, not false-positive, finding if residual carcinoma is not found in the surgical specimen (4). As previously stated, "If one applies the same histologic criteria, a false-positive diagnosis based on LCNB findings should be no more likely than one based on findings at excision" (17).

A few years after standardized LCNB was introduced, it became evident that a high percentage of lesions diagnosed as ADH at LCNB were cancers at excision (17,19). Burbank (20,21) introduced the term "underestimate" to apply to such cases, and that terminology has been adopted by others (13,22,23). We categorize LCNB histologic diagnoses that are not carcinoma at LCNB but are frequently carcinoma at excision, without regard to correlation between mammographic and histologic findings, as high-risk lesions. Excision is required. If carcinoma is found at excision, the high-risk lesion has become a histologic underestimate. The histologic findings from LCNB are not accurate enough to be true-positive findings or erroneous enough to be false-negative findings. High-risk histologic underestimate cases do not fit into the categories needed to determine sensitivity and specificity.

ADH diagnosed with LCNB is the prototypic high-risk LCNB lesion. Our histologic underestimation rate of 58% for lesions with an LCNB diagnosis of ADH is similar to the 31%–88% rate found in other series (10–17,19,20,24–27).

We think that radial scars diagnosed with LCNB should also be categorized as high-risk lesions requiring excision. Radial scars detected with mammography and diagnosed with excisional biopsy often have involvement with invasive carcinoma, in situ carcinoma, and/or atypical hyperplasia (28–31). Controversy exists, however, about whether radial scar is a risk factor for the development of invasive carcinoma and about its role as a precursor to carcinoma (32–36). Our underestimation rate of 40% for lesions with an LCNB diagnosis of radial scar is based on five lesions, with two of the five diagnosed as malignant at NLBB (Table 2). Lee et al (12) found malignancy at excision in one of four LCNB radial scar lesions. Only with routine excision of lesions said to be radial scar at LCNB will the true underestimation rate be determined.

If DCIS is diagnosed with LCNB, but invasive carcinoma is found at excision, we consider the LCNB to be both a true-positive finding (for the presence of carcinoma) and a histologic underestimate (for the severity of the carcinoma). Our histologic underestimation rate of 15% for LCNB lesions diagnosed as DCIS is similar to the 15%–36% rate found in other series (11,14,16,17,20,37). Histologic underestimation has also been found with DCIS at NLBB and invasive carcinoma at reexcision (38,39).

A benign diagnosis made on the basis of histologic findings at LCNB is considered to be a false-negative finding if the specimen obtained at repeat biopsy contains carcinoma. With LCNB of 483 consecutive, nonpalpable breast lesions, we evaluated false-negative findings in the 307 benign lesions for which either repeat biopsy (n = 48) or long-term mammographic follow-up (n = 259) was performed. Three (1%) of the 310 benign lesions were lost to follow-up.

Diagnostic repeat biopsy of a lesion is expensive and performed in two circumstances. First, immediate diagnostic repeat biopsy, without mammographic follow-up, is recommended if a high-risk lesion is found, if the histologic diagnosis is not definitively benign or malignant, if there is discordance between a benign histologic diagnosis and the mammographic findings, or (sometimes) if the radiograph of the specimen from a lesion detected as calcifications does not reveal calcifications. The repeat biopsy will reveal the carcinomas underestimated as high-risk lesions and the immediate false-negative lesions. Excluding the six benign lesions that had immediate incidental resection along with a carcinoma in the same breast, we did an immediate diagnostic repeat biopsy in six benign lesions and an immediate or delayed diagnostic repeat biopsy in 29 high-risk lesions, for an initial diagnostic repeat biopsy rate of 7% (35 of 483 lesions).

Second, delayed diagnostic repeat biopsy is performed if a histologically benign lesion exhibits substantial mammographic progression. The repeat biopsy will reveal delayed false-negative lesions, which are the most crucial biopsy problem. When the missed malignancy is eventually diagnosed with repeat biopsy after mammographic progression, there is potential for the cancer to have progressed in size, stage, or both. If we exclude the six benign lesions in which delayed incidental resection was performed (five along with a carcinoma in the same breast and one during reduction mammoplasty), we had a delayed diagnostic repeat biopsy rate of 6% (30 of 483 lesions). Two of those 30 lesions were malignant at repeat biopsy. Thus, two (1.2%) of 161 lesions with a final diagnosis of carcinoma were false-negative findings at LCNB (Table 2). The false-negative calcification lesion showed mammographic progression at the initial follow-up performed at 6 months. The false-negative mass lesion exhibited mammographic stability at 12 months and progression at 18 months after LCNB. Fajardo (9) reported malignant diagnoses 6, 8, and 13 months after false-negative findings were obtained at LCNB. All were stage I invasive ductal carcinomas. Lee et al (12) reported two delayed false-negative lesions. A mass showed progression on a mammogram obtained 6 months after LCNB, and repeat biopsy revealed a stage I invasive ductal carcinoma. A cluster of calcifications was initially stable and showed mammographic progression 24 months after LCNB. Repeat biopsy showed DCIS with microinvasion.

The histologic diagnosis from the initial biopsy can be incorrect because of either removal of inadequate material or histologic misinterpretation. Because we were trying to evaluate the accuracy of the biopsy sampling and not the pathologists' interpretation, we reinterpreted the histologic slides of difficult cases to ensure interpretive consistency. In our study, the three lesions in which the LCNB diagnosis was retrospectively upgraded from fibrocystic change to ADH and that were subsequently diagnosed as carcinoma at repeat biopsy were followed-up mammographically after LCNB (Table 2). If the histologic diagnosis of these three lesions had not been retrospectively upgraded, the false-negative rate would have been 3.1% and not 1.2%.

When we combine our experience with reports in the literature, we think the most important post-LCNB mammographic follow-up is done 6 months after the LCNB. We think mammographic follow-up is needed for at least 36 months and are currently pursuing it for 72 months after LCNB.

In the 310 lesions diagnosed as benign at LCNB, false-negative findings occurred in one of 162 calcification lesions and one of 148 masses. Correlation of the histologic findings from NLBB with those from LCNB showed false-negative findings in two of 26 nonspecific benign lesions and none of the benign lesions with specific diagnoses of fibrocystic change (n = 150), fibrosis (n = 88), fibroadenoma (n = 39), or lymph node (n = 7).

Other investigators, using the LCNB technique standardized by Parker et al (5), have preliminary false-negative rates of 0.3%–8.2% (mean, 4.2%) in series without complete surgical or mammographic follow-up (9–16) (Table 5). A more accurate determination of the false-negative rate in those series will require mammographic follow-up of all histologically benign lesions for at least 2–3 years and repeat biopsy of lesions with mammographic progression.


View this table:
[in this window]
[in a new window]
  		TABLE 5. Comparison of Histologic Diagnoses of Cancer, High-Risk Lesions, and Benign Lesions at LCNB and Final Diagnosis of Cancer in Studies with Incomplete Surgical or Mammographic Follow-up
 
The false-negative rate with LCNB found by combining series with NLBB correlation (5–8), series with incomplete follow-up (9–16), and this report is 0.3%–8.2% (mean, 4.0%). This compares with the NLBB false-negative rate of 0%–8% (approximate mean, 2.0%) (1). High-risk LCNB lesions found to be carcinoma at repeat biopsy are called histologic underestimates and not false-negative findings. To our knowledge, lesions diagnosed as ADH or radial scar at NLBB have not been evaluated with reexcision. If such lesions were to be reexcised after NLBB, as they often are after LCNB, we suspect some would also prove to be carcinomas. Indeed, a lesion diagnosed as ADH after excision with the Advanced Breast Biopsy Instrumentation system (United States Surgical, Norwalk, Conn) was found to be DCIS at reexcision (40).

Our own data support including ADH and radial scars as high-risk LCNB lesions. Because carcinoma was found at excision of lesions diagnosed at LCNB as atypical lobular hyperplasia in one of three lesions by Lee et al (12), one of three lesions by Liberman et al (13), and two of four lesions by Fuhrman et al (16), we have included atypical lobular hyperplasia as a high-risk lesion in Table 5. There were no atypical lobular hyperplasia lesions in our series. As more experience is gained, it is possible that other LCNB histologic diagnoses (eg, lobular carcinoma in situ, papilloma, and/or phyllodes tumor) will be added to the list of high-risk lesions requiring excision.

In conclusion, the false-negative rate for 14-gauge stereotactic LCNB was 1.2% in our series (of 483 consecutive unselected cases) and 0.3%–8.2% (mean, 4.0%) in the literature. Experience suggests that post-LCNB mammographic follow-up of most histologically benign lesions should begin at 6 months and continue for at least 36 months. LCNB findings of ADH and radial scar often underestimate the presence of carcinoma, and excisional biopsy is required.


   Acknowledgments
 
We thank Crystal Burton, Marilyn A. Spears, ART, CTR, and Christy W. White, RDH, ART, for considerable effort in arranging follow-up mammography; and Julie C. Clark, BA, for manuscript preparation.


   Footnotes
 
Address reprint requests to R.J.J.

From the 1995 RSNA scientific assembly.

Abbreviations: ADH = atypical ductal hyperplasia DCIS = ductal carcinoma in situ LCNB = large-core needle biopsy NLBB = needle-localized breast biopsy

Author contributions: Guarantors of integrity of entire study, R.J.J., K.W.N., J.R.S.; study concepts and design, R.J.J.; definition of intellectual content, R.J.J.; literature research, R.J.J., J.R.S.; clinical studies, all authors; data acquisition, R.J.J., K.W.N., J.R.S.; data analysis, R.J.J.; manuscript preparation, editing, and review, all authors.

Received June 10, 1998; revision requested July 27, 1998; revision received August 27, 1998; accepted October 17, 1998.
   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. Cancer J Clin 1997; 47:171-190.[Abstract]
  5. Parker SH, Lovin JD, Jobe WE, Burke BJ, Hopper KD, Yakes WR. Nonpalpable breast lesions: stereotactic automated large-core biopsies. Radiology 1991; 180:403-407.[Abstract/Free Full Text]
  6. Elvecrog EL, Lechner MC, Nelson MT. Nonpalpable breast lesions: correlation of stereotaxic large-core needle biopsy and surgical biopsy results. Radiology 1993; 188:453-455.[Abstract/Free Full Text]
  7. Gisvold JJ, Goellner JR, Grant CS, et al. Breast biopsy: a comparative study of stereotaxically guided core and excisional techniques. AJR 1994; 162:815-820.[Abstract/Free Full Text]
  8. Brenner RJ, Fajardo L, Fisher PR, et al. Percutaneous core biopsy of the breast: effect of operator experience and number of samples on diagnostic accuracy. AJR 1996; 166:341-346.[Abstract/Free Full Text]
  9. Fajardo LL. Cost-effectiveness of stereotaxic breast core needle biopsy. Acad Radiol 1996; 3:S21-S23.
  10. Dahlstrom JE, Jain S, Sutton T, Sutton S. Diagnostic accuracy of stereotactic core biopsy in a mammographic breast cancer screening programme. Histopathology 1996; 28:421-427.[Medline]
  11. Nguyen M, McCombs MM, Ghandehari S, et al. An update on core needle biopsy for radiologically detected breast lesions. Cancer 1996; 78:2340-2345.[Medline]
  12. 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]
  13. Liberman L, Dershaw DD, Glassman JR, et al. Analysis of cancers not diagnosed at stereotactic core breast biopsy. Radiology 1997; 203:151-157.[Abstract/Free Full Text]
  14. Acheson MB, Patton RG, Howisey RL, Lane RF, Morgan A. Histologic correlation of image-guided core biopsy with excisional biopsy of nonpalpable breast lesions. Arch Surg 1997; 132:815-821.[Abstract]
  15. Meyer JE, Smith DN, Lester SC, et al. Large-needle core biopsy: nonmalignant breast abnormalities evaluated with surgical excision or repeat core biopsy. Radiology 1998; 206:717-720.[Abstract/Free Full Text]
  16. Fuhrman GM, Cederbom GJ, Bolton JS, et al. Image-guided core-needle breast biopsy is an accurate technique to evaluate patients with nonpalpable imaging abnormalities. Ann Surg 1998; 6:932-939.
  17. Jackman RJ, Nowels KW, Shepard MJ, Finkelstein SI, Marzoni FA. Stereotaxic large-core needle biopsy of 450 nonpalpable breast lesions with surgical correlation in lesions with cancer or atypical hyperplasia. Radiology 1994; 193:91-95.[Abstract/Free Full Text]
  18. Parker SH, Lovin JD, Jobe WE, et al. Stereotactic breast biopsy with a biopsy gun. Radiology 1990; 176:741-747.[Abstract/Free Full Text]
  19. Liberman L, Cohen MA, Dershaw DD, Abramson AF, Hann LE, Rosen PP. Atypical ductal hyperplasia diagnosed at stereotaxic core biopsy of breast lesions: an indication for surgical biopsy. AJR 1995; 164:1111-1113.[Abstract/Free Full Text]
  20. 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]
  21. Burbank F. Stereotactic breast biopsy: comparison of 14- and 11-gauge Mammotome probe performance and complication rates. Am Surg 1997; 63:988-995.[Medline]
  22. 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.[Abstract/Free Full Text]
  23. Liberman L, Derhsaw DD, Rosen PP, Morris EA, Abramson AF, Borgen PI. Percutaneous removal of malignant mammographic lesions at stereotactic vacuum-assisted biopsy. Radiology 1998; 206:711-715.[Abstract/Free Full Text]
  24. Tocino I, Garcia BM, Carter D. Surgical biopsy findings in patients with atypical hyperplasia diagnosed by stereotaxic core needle biopsy. Ann Surg Oncol 1996; 3:483-488.[Abstract]
  25. Moore MM, Hargett CW, Hanks JB, et al. Association of breast cancer with the finding of atypical ductal hyperplasia at core breast biopsy. Ann Surg 1997; 225:726-733.[Medline]
  26. Gadzala DE, Cederbom GJ, Bolton JS, et al. Appropriate management of atypical ductal hyperplasia diagnosed by stereotactic core needle breast biopsy. Ann Surg Oncol 1997; 4:283-286.[Abstract]
  27. Brown TA, Wall JW, Christensen ED, et al. Atypical hyperplasia in the era of stereotactic core needle biopsy. J Surg Oncol 1998; 67:168-173.[Medline]
  28. Vega A, Garijo F. Radial scar and tubular carcinoma: mammographic and sonographic findings. Acta Radiol 1993; 34:43-47.[Medline]
  29. Sloane JP, Mayers MM. Carcinoma and atypical hyperplasia in radial scars and complex sclerosing lesions: importance of lesion size and patient age. Histopathology 1993; 23:225-231.[Medline]
  30. Frouge C, Tristant H, Guinebretiere JM, et al. Mammographic lesions suggestive of radial scars: microscopic findings in 40 cases. Radiology 1995; 195:623-625.[Abstract/Free Full Text]
  31. Patel A, Steel Y, McKenzie J, Letcher M, Querci Della Rovere G, Morgan MWE. Radial scars: a review of 30 cases. Eur J Surg Oncol 1997; 23:202-205.[Medline]
  32. Fenoglio C, Lattes R. Sclerosing papillary proliferations in the female breast: a benign lesion often mistaken for carcinoma. Cancer 1974; 33:691-700.[Medline]
  33. Fisher ER, Palekar AS, Kotwall N, Lipana N. A nonencapsulated sclerosing lesion of the breast. Am J Clin Pathol 1979; 71:240-246.[Medline]
  34. Linell F, Ljungberg O, Andersson I. Breast carcinoma: aspects of early stages, progression and related problems. Acta Pathol Microbiol Scand 1980; 272(suppl):14-62.
  35. Andersen JA, Gram JB. Radial scar in the female breast: a long-term follow-up study of 32 cases. Cancer 1984; 53:2557-2560.[Medline]
  36. Wellings SR, Alpers CE. Subgross pathologic features and incidence of radial scars in the breast. Hum Pathol 1984; 15:475-479.[Medline]
  37. Liberman L, Dershaw DD, Rosen PP, et al. Stereotaxic core biopsy of breast carcinoma: accuracy at predicting invasion. Radiology 1995; 194:379-381.[Abstract/Free Full Text]
  38. Thompson WR, Bowen JR, Dorman BA, Pricolo VE, Shahinian TK, Soderberg CH. Mammographic localization and biopsy of nonpalpable breast lesions: a 5-year study. Arch Surg 1991; 126:730-734.[Abstract]
  39. Tartter PI, Bleiweiss IJ, Levchenko S. Factors associated with clear biopsy margins and clear reexcision margins in breast cancer specimens from candidates for breast conservation. J Am Coll Surg 1997; 185:268-273.[Medline]
  40. Leibman AJ, Frager D. Experience with breast biopsy using the Advanced Breast Biopsy Instrumentation (ABBI) system (abstr). AJR 1998; 170(P):85.[Free Full Text]




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 Google Scholar
Google Scholar
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Shepard, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jackman, R. J.
Right arrow Articles by Shepard, M. J.

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