| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Departments of Radiology (C.C.R, M.M., T.H.H.), Gynecology (T.W.), Surgery (F.F.), and Clinical Pathology (M.R.), Medical University Vienna, Waehringer Guertel 18-20, 1080 Vienna, Austria; and Department of Radiology, Danube Hospital, Vienna, Austria (G.P.). From the 2001 RSNA Annual Meeting. Received August 10, 2004; revision requested October 19; revision received December 20; accepted January 21, 2005. Address correspondence to C.C.R. (e-mail: riedlc{at}mskcc.org).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: The ethical review board approved the study; the need for informed consent was waived. A total of 1115 stereotactic NLOBBs performed in 1068 women aged 22–90 years (mean age, 54 years) were tracked to determine outcomes. In cases of malignancy, NLOBB was considered to be diagnostically successful. The mammographic outcomes in all patients with benign results at NLOBB were tracked for at least 2 years. Cases without such mammographic follow-up were cross-referenced with a tumor registry after at least 54 months. The lesion miss rate was based on all malignant lesions and all lesions with long-term mammographic follow-up. Results from all 1115 NLOBBs were used to report the false-negative rate.
RESULTS: Of 1115 NLOBBs, 472 (42%) had malignant results. Mammographic follow-up data were available for 535 (83%) of 643 NLOBBs with benign results. Mammographic follow-up revealed 11 lesions that were missed with NLOBB, of which five were malignant and six benign. Thus, the lesion miss rate with NLOBB was 1.1% (11/[472 + 535]). Among the 643 cases, in 108 of them without mammographic follow-up (17%), cross-referencing with a tumor registry did not reveal missed cases of breast carcinoma. The false-negative rate was therefore 1.0% (5/[472 + 5]).
CONCLUSION: On the basis of the results of long-term follow-up, the authors noted a lesion miss rate of 1.1% and a false-negative rate of 1.0%, which indicate that NLOBB with stereotactic guidance is an accurate method for diagnosing breast lesions.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Several investigators addressed the diagnostic accuracy of NLOBB, reporting lesion miss rates (rate of all missed lesions, benign and malignant) of 0%–18% (mean, 2.6%) and a mean false-negative rate (cancer miss rate) of 2% (1,2,7–12). Most of the guidance systems that were used in these studies, however, do not represent those commonly used in today's practice. Currently, NLOBB with stereotactic guidance is the most common guiding technique next to ultrasonography (US) for making a definitive diagnosis of nonpalpable breast lesions. Few data exist, however, about the reliability of NLOBB performed with stereotactic guidance. Thus, the purpose of our study was to retrospectively determine the lesion miss rate and false-negative rate of NLOBB with stereotactic guidance in a large study population.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Open Biopsy and Postintervention Management
Stereotactically guided NLOBB was performed by three radiologists (one an author of this study, T.H.H.), who have experience with more than 500 localizations each, on a dedicated prone table (Mammotest; Fischer Imaging, Denver, Colo). The stereotactic guidance method has been previously described (13,14). Briefly, after targeting a lesion on stereotactic views of the compressed and fixed breast, a dedicated software program calculates the coordinates of the target and a guiding needle is percutaneously inserted into the target lesion. In all instances, the position of the needle was checked with further stereotactic views so that the needle depth could be readjusted if necessary (15). In 53 of the 1115 cases (4.7%), large–core-needle biopsy was performed before the localization procedure by using the same guiding needle (9–12). In 642 cases (58%), a hook wire (X-Reidy; Cook, Bjaeverskov, Denmark) was used for localization. In the remaining 473 cases (42%), carbon dye prepared by the institution (0.08-mL suspension of activated charcoal powder, 0.02 mL polysorbitant-80-oleinatum, and 0.018 mg sodium chloride in 1.9 mL distilled water) was injected into the lesion as described by Svane (13). The choice for the applied marker was not influenced by the type or position of the lesion but depended only on the preference of the surgeon.
Open surgical breast biopsy was performed within 1 day after needle localization. To ensure the adequacy of resection, radiography of the specimen and/or histologic examination of frozen slices of the specimen was performed (16–18). In cases with malignant findings at histologic examination, NLOBB was considered successful in the diagnosis and the patient was treated accordingly. In cases with benign findings at histologic examination, management recommendations were dependent on the type of histopathologic result and its concordance with mammographic findings. Histologic results considered to be specific benign findings included fibroadenoma, intramammary lymph node, and cyst. Nonspecific benign findings included fibrocystic changes, apocrine metaplasia, benign breast tissue, and fibrous breast tissue. High-risk lesions included atypical ductal hyperplasia, lobular carcinoma in situ, and radial scarring. In cases with specific benign histologic findings and concordance with mammography (eg, a circumscribed mass yielding a fibroadenoma), mammographic follow-up after 12 months was recommended. In cases with unspecific benign histologic findings and concordance with mammography (eg, a lesion with microcalcifications yielding fibrocystic changes with microcalcifications), mammographic follow-up after 6 months was recommended. In cases of discordance with mammography (eg, calcific lesions in which no calcifications were identified at histologic analysis, a discrete mass lesion yielding only benign breast tissue, or lesions that were highly suggestive of malignancy or Breast Imaging Reporting and Data System [BI-RADS] category 5 and were sampled with benign results), immediate mammographic follow-up (after 6–12 weeks) was recommended (19).
Data Collection
Mammographic lesions were divided into three groups: (a) calcifications, (b) masses including asymmetric density and areas of architectural distortion, and (c) a combination of masses and calcifications.
As noted earlier, NLOBB was considered diagnostically successful if malignant lesions were diagnosed at histologic examination. The remaining lesions with a benign histopathologic diagnosis were tracked for outcome with mammographic surveillance for at least 24 months. This time interval was chosen because missed lesions are not always detected on the initial mammograms; they are also detected later on follow-up mammograms (8). Follow-up examinations were performed at our institution or at outside facilities. Whenever follow-up mammography revealed a lesion (BI-RADS category 2–5 [20]), a panel comprising two radiologists (T.H.H., C.C.R.), a gynecologist experienced in breast surgery (T.W.), and a pathologist (M.R.) reviewed medical records, histologic findings, and pre- and postoperative mammograms. The panel established whether the lesion found during follow-up was the target lesion of the initial open biopsy on the basis of lateralization, localization, size, appearance, and radiologic compatibility with the histopathologic diagnosis. If it was determined that the target lesion was seen on the follow-up mammograms, the case was classified as missed. Thus, only patients with data on the mammographic follow-up of 24 months were included in the calculation of the lesion miss rate. If one of the missed lesions was shown to be cancer at subsequent surgery, the case was classified as false-negative.
To enable calculation of the false-negative rate on the basis of all 1115 NLOBBs, patients without follow-up mammograms were cross-referenced with a tumor registry by one author (C.C.R.) in an attempt to identify any possible cases of breast cancer that may have been diagnosed subsequent to NLOBB. The panel determined whether the reported case corresponded with the target lesion and could be classified as a false-negative finding. Cross-referencing with the tumor registry was performed 54 months after the last patient was included in this study.
Data Analysis
A radiologist (T.H.H.) who had specialized in breast imaging for 5 years analyzed all cases of NLOBB classified as missed or false-negative to determine the prebiopsy classification of lesion type, lesion size, and level of suspicion according to BI-RADS (20). In addition, the distance of the needle tip to the target was calculated with the method described by Liberman et al (21). A distance within 1 cm was arbitrarily chosen as a correct position. Also recorded was the type of localization marker (carbon dye or hook wire), as well as whether, in the case of calcifications, radiography of the specimen was performed, and, if so, whether the calcifications could be identified. Further, the histopathologic diagnosis of the initial NLOBB specimen, concordance with the radiologic findings (concordant, discordant) as mentioned previously (19), and the histopathologic diagnosis of the second NLOBB specimen were noted. Finally, the time between the initial and definitive diagnoses was determined.
Statistical Analysis
Data were analyzed by using statistical software (Statistical Package for the Social Sciences, release 8.0.0; SPSS, Chicago, Ill). To compare the number of missed lesions between the three groups of lesion types (ie, calcification lesions, mass lesions, and combination lesions) and between the two types of localization material (carbon dye and hook wire), P values were determined by using the 
2 test with Yates correction, if appropriate, and were considered to indicate a statistically significant difference if they were less than .05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The 1115 cases of NLOBB represent the study population and were used to calculate the false-negative rate. At initial NLOBB, 472 cases (42%) were malignant and 643 (58%) benign. Of the 643 benign cases, information from mammographic follow-up of 24–86 months (mean, 47 months) was available in 535 (83%). No mammographic follow-up information was available in the remaining 108 benign cases (17%); thus, 1007 cases (535 + 472) reflect the actual number for the calculation of the lesion miss rate.
All 472 cases of NLOBB with malignant histologic results were considered successful at diagnosis. Mammographic follow-up of the 535 cases with benign histologic results revealed 28 lesions (5.2%) that necessitated further analysis by the panel. The panel decided that 17 of the 28 lesions (61%) did not correspond to the target lesion. The remaining 11 lesions were classified as missed. The lesion miss rate of NLOBB in this study was 1.1% (11 of 1007 lesions).
The histologic and imaging findings from the 11 lesions missed at the initial NLOBB are presented in the Table. Five of the 11 missed lesions were malignant at final diagnosis (Fig 1), and six were benign (Fig 2). Missed lesions occurred in six of 312 calcification lesions (1.9%), five of 744 mass lesions (0.7%), and none of the 59 combination lesions (P > .05). Stereotactic views showed inappropriate needle placement in one case and correct needle placement in 10. In one case, core-needle biopsy had been performed previously (2% of a total of 53 core-needle biopsies) and revealed the missed malignancy (case 4). Radiography of the specimen was performed in two of the six missed lesions classified as calcifications. In one case with pleomorphic calcifications (case 5), calcifications were identified at radiography of the specimen and histologic analysis. NLOBB yielded lobular carcinoma in situ. At follow-up mammography performed after 7 months, calcifications could be identified in the area of the prior surgery. Subsequent NLOBB performed after 11 months revealed invasive ductal carcinoma. In the second case (case 7), no calcifications were identified at radiography of the specimen or histologic analysis. NLOBB and repeat biopsy performed after 4 months revealed fibrocystic changes. In the remaining four lesions with calcifications, radiography of the specimen was not performed and no calcifications were identified at histologic analysis. In these cases, repeat biopsy was performed 4, 11, 27, and 40 months after NLOBB. In all 11 missed cases, the histologic result was considered a nonspecific finding (3). With regard to the concordance of the histologic results with the mammographic findings, two of the 11 cases were considered to be concordant and nine were discordant (19). The diagnostic delay, defined as the time between the first and second open breast biopsies, was 0.3–40.0 months (mean, 14.5 months). Six of the 11 missed lesions were marked with a hook wire; the other five were marked with carbon dye (six of 576 wire localizations [1.0%], five of 431 dye localizations [1.2%], 2 = 0.00, P > .99).
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
NLOBB with stereotactic guidance is a common procedure for the definitive diagnosis of nonpalpable breast lesions. Some studies have reported the lesion miss rate and/or false-negative rate of this procedure. However, it is difficult to compare the results of these studies because procedures and follow-up vary. In 1997, Jackman and Marzoni (7) determined the lesion miss rate of NLOBB in a well-detailed literature review comparing 49 studies (published between 1976 and 1997) and reported a mean lesion miss rate of 2.6% (0%–18%) on the basis of 12 563 NLOBBs. None of those 49 studies, however, reported the lesion miss rate at stereotactic NLOBB with an adequate quality control. Preoperative localization was performed with stereotactic guidance in only four of the 49 studies (9–12). All four of these studies were designed to correlate the findings at preceding stereotactically guided, large–core-needle biopsy with the histologic findings at subsequent NLOBB. Hence, the lesion miss rates of NLOBB were based only on discordance with immediately preceding core-needle biopsy. In these studies, a lesion miss rate of 2.0%–3.9% (12 of 464 lesions; mean lesion miss rate, 2.6%) was reported.
There may be several reasons for the higher rate of misses in these studies compared with our results (1.1%). First, complete removal of mammographic lesions after stereotactic 14-gauge large–core-needle biopsy has been reported in 9% of cases (22). Thus, it seems possible that some such lesions might not be found at subsequent surgery. Second, Parker et al (9) reported a lesion miss rate of 3.9%, which could be explained by the inexperience of the authors in using two different stereotactic units. In contrast, in our study, stereotactic marker placement was performed by three experienced radiologists, two of whom have worked with stereotactic systems since 1987. This lends further support to the importance of operator experience in achieving optimal results for stereotactic procedures and reinforces the need for training and supervision early in one's experience when performing the procedure. Studies by Liberman et al (23) and Pfarl et al (24) describe a learning curve for stereotactic large–core-needle biopsy procedures with a greater false-negative rate for the first 15 cases than for subsequent cases.
Finally, it should be stated that long-term follow-up of at least 24 months was not performed in any of the 49 studies. Verkooijen et al (8) reported a false-negative rate of 4% and emphasized the importance of long-term follow-up to estimate the diagnostic performance of NLOBB because half of the missed cancers were detected more than 3 years after NLOBB. The need for adequate long-term follow-up to estimate the diagnostic performance of NLOBB is also supported by the results of our study. Seven of the 11 missed cases remained undetected for 9–50 months: four because the patient did not return for the recommended early follow-up mammography (Fig 2) and three because of nondetection at early follow-up mammography (Fig 1).
Several factors contribute to the failure of NLOBB. Inaccurate needle placement has frequently been reported as a source of error (7,8,25). One of the 11 lesions in the present series was missed because of inappropriate needle placement on stereotactic images. In addition, it should be reconsidered whether stereotactic control images are sufficient (13,15) to help confirm the correct position of the localization needle. Studies have shown that a small error of needle position in the z-axis with compression can lead to a substantial dislocation at decompression of the breast. Rosen and Vo (26) reported a dislocation of the clip of more than 1 cm in 28% of cases. Thus, in addition to stereotactic control images, two-view mammograms after marker placement are recommended by these authors to verify the correct position. This approach is simple for hook-wire marking. In the case of carbon localization, however, iodine must be mixed with carbon to make the localization site visible on mammograms (27–29). It has also been reported that insufficient removal of the target lesion can be avoided if the lesion is penetrated by the localization needle (7).
Regardless of the guiding method used, Peeters et al (30) reviewed false-negative open biopsy results in screening-detected lesions and found that another reason for missed lesions was that radiography of the specimen, which is today considered to be of great importance in reducing the number of missed lesions (7,8,13,19,31), was not performed on a routine basis. In our study, radiography of the specimen was not performed in all cases because some of our surgeons preferred to use a frozen slice of the specimen for margin examination. Radiography of the specimen was not performed in four of six missed lesions with calcifications. Because no calcifications could be identified at histologic analysis in all four cases, one could assume that the use of radiography of the specimen might have avoided these failures.
Nevertheless, the failure to identify a missed cluster of calcifications owing to a false-positive finding at radiography of the specimen, which occurred in one of our lesions, has been reported previously (8,18). The reason for a false-positive finding at radiography of the specimen may be removal of adjacent calcifications next to the target calcifications or distortion of the specimen during radiography. The possibility of false-positive findings at radiography of the specimen strongly supports the need for postoperative mammography (18).
Previous studies have reported imaging-histologic discordance in 0%–6% (mean, 4%) of lesions in which percutaneous imaging-guided biopsy had been performed; among discordant lesions, cancer was found in 0%–64% (mean, 18%) of cases (19). Thus, in cases of imaging-histologic discordance, mammography after 6–12 weeks is indicated. In our study, imaging-histologic discordance occurred in eight of the 11 missed lesions. These included four BI-RADS category 5 lesions that yielded benign results; three BI-RADS category 4 clusters of calcifications, in which no calcifications were retrieved; and one BI-RADS category 4 mass that yielded benign breast tissue. In three of the eight missed lesions with imaging-histologic discordance, repeat biopsy was performed within a 4-month period. Despite the fact that, at the time of this study, referring physicians at our institution usually made individual management recommendations on the basis of the concordance of histologic and mammographic findings, repeat biopsy was performed after 11, 22, 27, 30, and 40 months in the remaining five cases. Weekly multidisciplinary case conferences, before and after intervention, should have avoided the delayed detection in these cases (32).
In addition, we should consider that the failure of NLOBB is influenced also by the experience of other team members in a breast unit besides the radiologist. Thus, a minimum of 150 newly diagnosed cases of breast cancers per year and unit are recommended. The reason for recommending a minimum number is to ensure a caseload sufficient to maintain an expertise for each team member and to ensure cost-effective working of a breast unit (31).
NLOBB with stereotactic guidance can be performed with the patient in the prone or sitting position. In our study, NLOBB was performed with patients in the prone position. Thus, the transferability of our data to NLOBB in the sitting position might be seen as a limitation. As demonstrated in two studies of stereotactic core-needle biopsy, however, only vasovagal reactions, and no other disadvantages, are seen slightly more often with patients in the sitting position than with patients in the prone position (33,34). Furthermore, digital mammography systems, in combination with stereotactic units in the sitting position, which have now become commercially available, decrease the time necessary to perform the procedure and lower the vasovagal reaction rate (33).
Our overall malignancy rate of 42% with open biopsy is low, and it could be questioned why more core-needle biopsies were not performed. At the time of this study (1994–1997), the implementation of core-needle biopsies at our institution had just started. Because initially the acceptance by the surgeons was low, needle biopsies were performed rarely and in addition to (not instead of) open biopsy. This has since changed, and we now perform core-needle biopsies in accordance with the European Society of Mastology guidelines (32).
In conclusion, NLOBB with stereotactic guidance is an accurate and reliable method with a low lesion miss rate and false-negative rate in the diagnosis of nonpalpable breast lesions. We believe the following quality assurance measures should be performed routinely to avoid any failure in the treatment of these patients: multidisciplinary case conferences before and after NLOBB, two-view mammography after marker placement (in case of stereotactic guidance, mammograms should be obtained in addition to stereotactic control images) (26), and radiography of the fresh specimen or the paraffin sections. In cases with imaging-histologic discordance, follow-up mammography is indicated 6–12 weeks after open biopsy (8). Even when successful excision of the lesion is reported, failure of NLOBB should be reconsidered and excluded at subsequent follow-up and routine mammography. Performance and audit figures must be produced yearly to verify lesion miss rates of less than 5% at the first open breast biopsy (32).
|
FOOTNOTES |
|---|
Abbreviations: BI-RADS = Breast Imaging Reporting and Data System • NLOBB = needle-localized open breast biopsy
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, C.C.R., T.H.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, G.P., F.F., T.H.H.; clinical studies, C.C.R., M.M., T.W., M.R., T.H.H.; statistical analysis, M.M., F.F.; and manuscript editing, C.C.R., G.P., T.H.H.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. C. Riedl, L. Ponhold, D. Flory, M. Weber, R. Kroiss, T. Wagner, M. Fuchsjager, and T. H. Helbich Magnetic Resonance Imaging of the Breast Improves Detection of Invasive Cancer, Preinvasive Cancer, and Premalignant Lesions during Surveillance of Women at High Risk for Breast Cancer Clin. Cancer Res., October 15, 2007; 13(20): 6144 - 6152. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
