DOI: 10.1148/rg.271065029
RadioGraphics 2007;27:79-94
© RSNA, 2007
Missed Breast Cancers at US-guided Core Needle Biopsy: How to Reduce Them1
Ji Hyun Youk, MD,
Eun-Kyung Kim, MD,
Min Jung Kim, MD,
Ji Young Lee, MD and
Ki Keun Oh, MD
1 From the Department of Diagnostic Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, Seodaemun-ku, Shinchon-dong 134, Seoul 120-752, South Korea. Presented as an education exhibit at the 2005 RSNA Annual Meeting. Received March 17, 2006; revision requested May 26; final revision received September 5; accepted September 6. All authors have no financial relationships to disclose.
Address correspondence to E.-K.K. (e-mail: ekkim{at}yumc.yonsei.ac.kr).
 |
Abstract
|
|---|
Ultrasonographically (US) guided core needle biopsy is currently recognized as a reliable alternative to surgical biopsy for the histopathologic diagnosis of breast lesions. However, despite advances in biopsy devices and techniques, false-negative diagnoses are unavoidable and may delay the diagnosis and treatment of breast cancer. The most common reasons for false-negative diagnosis are (a) technical or sampling errors, (b) failure to recognize or act on radiologic-histologic discordance, and (c) lack of imaging follow-up after a benign biopsy result. Technical difficulties (eg, poor lesion or needle visualization, deeply located lesions, dense fibrotic tissue) cause inaccurate sampling but can be reduced by using modified standard techniques. Radiologic-histologic correlation is also of critical importance in US-guided core needle biopsy. Radiologic-histologic discordance occurs when the histologic results do not provide a sufficient explanation for the imaging features and indicates that the lesion may not have been sampled adequately, so that repeat biopsy is warranted. Appropriate follow-up imaging is invaluable; even patients with concordant benign findings after US-guided core needle biopsy are directed to undergo follow-up imaging because there may be delays in the recognition of false-negative findings. Optimization of technique, radiologic-histologic correlation, and postbiopsy follow-up protocols are recommended to reduce the occurrence of false-negative diagnosis at US-guided core needle biopsy performed by radiologists.
© RSNA, 2007
|
LEARNING OBJECTIVES
|
|---|
After reading this article and taking the test, the reader will be able to:- List the common causes of false-negative results at US-guided core needle breast biopsy.
- Describe the potential technical pitfalls of US-guided core needle breast biopsy and solutions to these pitfalls.
- Discuss ways to reduce missed breast cancer at US-guided core needle biopsy, including optimization of technique, radiologic-histologic correlation, and follow-up imaging.
|
Introduction
|
|---|
Since its introduction in the early 1990s, percutaneous image-guided core needle biopsy has become a widely used technique for evaluating breast abnormalities (1). This technique revolutionized the practice of breast imaging and is now accepted as a reliable alternative to surgical biopsy for histopathologic diagnosis of breast lesions (2–4).
Percutaneous image-guided core needle biopsy is performed under either stereotactic or ultrasonographic (US) guidance. US guidance offers a number of advantages over stereotactic guidance: (a) lack of ionizing radiation; (b) readily available equipment; (c) accessibility of all areas of the breast; (d) real-time visualization of the needle; (e) multidirectional sampling; (f) lower cost; and (g) greater patient comfort, since US-guided biopsy does not require compression and can usually be performed more quickly (5,6). For these reasons, US guidance may be preferable in lesions that are amenable to core needle biopsy with either stereotactic or US guidance. Since Parker et al (7) first reported the use of US-guided automated core needle biopsy with a 14-gauge needle, investigators have demonstrated that this procedure is fast, safe, accurate, and economical (8–15), although there have been few large studies describing the use of this procedure in the breast. More recently, the directional vacuum-assisted method has become available for US-guided biopsy with 8- and 11-gauge needles. Vacuum-assisted devices provide larger core samples and allow more contiguous sampling than do automated guns, potentially leading to more complete sampling of lesions and reducing the chances of sampling error (15).
As with any percutaneous sampling procedure, however, sampling errors (eg, targeting error, poor tissue acquisition) can result in a false-negative diagnosis, which may delay treatment of the cancer. In this article, we briefly review the results of US-guided core needle biopsy; describe possible causes of sampling error; and discuss steps that should be taken to diminish potential technical pitfalls, giving particular attention to the optimization of technique, careful radiologic-histologic correlation, and follow-up imaging at appropriate intervals.
|
Review of Biopsy Results
|
|---|
US-guided core needle biopsy has shown high sensitivity for the diagnosis of breast cancer. Parker et al (7) first reported the use of US-guided 14-gauge core needle biopsy and found 100% correlation with surgical results for 49 excised masses, including 34 cancers. Results across eight series (Table 1) showed that 1456 (96%) of 1518 cancers were diagnosed at initial core biopsy. Most of the cancers that were not initially diagnosed were immediately reevaluated with repeat biopsy, and only three cancers were found in these series at delayed follow-up (12,14). In published series for which details were available, 10% of lesions required repeat biopsy, with 17% of this subset of lesions proving malignant (Table 2). Most repeat biopsies were performed due to equivocal pathologic findings (eg, lobular neoplasm, radial sclerosing lesion, papillary lesions, possible phyllodes tumors) or discordance between radiologic and histologic findings (16). Schoonjans and Brem (12) found eight carcinomas after repeat biopsy performed for these reasons in 69 lesions in which initial core biopsy had yielded benign results. Liberman et al (8) recommended repeat biopsy or later surgical biopsy in five women with equivocal pathologic findings, and a low-grade malignant phyllodes tumor was found at surgery in one patient. In another five cases of radiologic-histologic discordance, two carcinomas were found at surgery. In a series by Berg et al (11), the correct diagnosis was believed to have been missed altogether at only three (0.4%) of 687 biopsies.
US-guided 11-gauge directional vacuum-assisted core needle biopsy is also a fast and accurate method for the diagnosis of breast lesions, although, unlike for stereotactic biopsy, few data are available. The reported rate of false-negative diagnosis is 0%–2% (15,17–19). One missed cancer was reported in studies by Philpotts et al (15) and Simon et al (18); in both cases, repeat biopsy was performed owing to discordance between radiologic and histologic findings. Cho et al (17) also reported one false-negative result that led to repeat biopsy due to progression of a suspicious imaging finding during follow-up.
Therefore, radiologists performing US-guided core needle biopsy need to be aware of the possibility of a false-negative diagnosis and should be prepared to perform repeat biopsy or suggest surgical excision, particularly in cases with equivocal pathologic findings or discordance between radiologic and histologic findings.
|
Sampling Error: Causes and Solutions
|
|---|
Lesion Visualization
Use of a high-frequency, correctly focused 10–12-MHz probe can improve the resolution and contrast of the lesion. In addition, US techniques that optimize performance are essential. Sufficient gel should be used, and the focal zone should be placed just below the lesion. Adjustments in the dynamic range or postprocessing gray scales may improve contrast so that lesions are more visible (6).
Although mammography is the modality of choice for detecting and characterizing clustered microcalcifications, some microcalcifications can now also be identified during a directed US evaluation due to improvements in spatial resolution and other technical advances (20,21). In cases in which stereotactic biopsy is not available, the patient is unable to undergo proper positioning for stereotactic biopsy, or the location of targeted microcalcifications does not permit stereotactic localization, US-guided core needle biopsy may be an effective alternative. However, with use of the multipass automated gun technique, difficulties in targeting the echogenic foci of microcalcifications may arise. If air is introduced into the biopsy cavity after the pass of the biopsy needle, the gas in the biopsy track can have a focal echogenic appearance or cause shadowing, thereby mimicking or obscuring the original echogenic target (Fig 1). The vacuum-assisted device might help solve this problem because it can suction air and blood away from the biopsy cavity during the procedure (22).
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Poor lesion visualization at US-guided biopsy performed with the multipass automated gun technique. (a) US image of the breast shows numerous echogenic specular reflectors (arrow) corresponding to clustered microcalcifications against a hypoechoic background. (b) On a US image obtained after the first pass of the core biopsy needle through the microcalcifications, bright hyperechoic air in the biopsy track (arrows) mimics and obscures the targeted microcalcifications.
|
|
View larger version (123K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Poor lesion visualization at US-guided biopsy performed with the multipass automated gun technique. (a) US image of the breast shows numerous echogenic specular reflectors (arrow) corresponding to clustered microcalcifications against a hypoechoic background. (b) On a US image obtained after the first pass of the core biopsy needle through the microcalcifications, bright hyperechoic air in the biopsy track (arrows) mimics and obscures the targeted microcalcifications.
|
|
Needle Visualization
Poor needle visualization is a common difficulty that may seem surprising, given the large caliber of the needle. Even large-gauge needles are difficult to visualize if a steep angle is used, which results in less reflective echoes. When the needle is parallel to the transducer, it is seen in its entirety because the number of needle-generated reflected echoes that are perpendicular to the ultrasound beam is maximized. If the needle is angled, localization of its tip becomes more difficult (Fig 2).
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Visualization of the core biopsy needle with respect to needle angle. (a) US image obtained with a steep angle between the ultrasound beam and the needle does not clearly depict the needle (arrow) due to the generation of fewer reflected echoes. (b) US image obtained with the needle perpendicular to the ultrasound beam allows optimal visualization of the needle (arrow).
|
|
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Visualization of the core biopsy needle with respect to needle angle. (a) US image obtained with a steep angle between the ultrasound beam and the needle does not clearly depict the needle (arrow) due to the generation of fewer reflected echoes. (b) US image obtained with the needle perpendicular to the ultrasound beam allows optimal visualization of the needle (arrow).
|
|
Entry at the periphery of the breast orients the needle perpendicular to the ultrasound beam, and the lesion is targeted correctly (Fig 3). By holding the gun upside down, one can advance the needle without improper angulation relative to the ultrasound beam and avoid firing toward the chest wall in cases involving deep lesions (Fig 4). If the needle is not visualized, delicate adjustments can be made until it is brought in line with the lesion (Fig 5) (6,23).
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Effect of needle placement during core needle biopsy. Drawings illustrate how a skin entry site close to the center of the breast (a) makes the biopsy needle angled, resulting in poor needle visualization, whereas inserting the needle at the periphery of the breast (b) allows the needle to be oriented parallel to the chest wall and perpendicular to the ultrasound beam.
|
|
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Effect of needle placement during core needle biopsy. Drawings illustrate how a skin entry site close to the center of the breast (a) makes the biopsy needle angled, resulting in poor needle visualization, whereas inserting the needle at the periphery of the breast (b) allows the needle to be oriented parallel to the chest wall and perpendicular to the ultrasound beam.
|
|
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Effect of gun positioning during core needle biopsy for a deep lesion. (a) Drawing illustrates that, when the gun is held rightside up, the needle must be angulated toward the chest wall. (b) Drawing illustrates that, with the gun flipped upside down, the lesion can be approached with the needle parallel to both the chest wall (for greater safety) and the transducer face (for better needle visualization).
|
|
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Effect of gun positioning during core needle biopsy for a deep lesion. (a) Drawing illustrates that, when the gun is held rightside up, the needle must be angulated toward the chest wall. (b) Drawing illustrates that, with the gun flipped upside down, the lesion can be approached with the needle parallel to both the chest wall (for greater safety) and the transducer face (for better needle visualization).
|
|
View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Drawings illustrate how misalignment of the core biopsy needle can be corrected by adjusting its position relative to the lesion. As shown in A, the long axis of the transducer should be aligned with the lesion and the skin incision site. Drawings B and C illustrate the proper technique, should the needle be advanced to the expected location of the lesion and the needle is not visualized in the field of view: In B, stop moving the needle, and in C, gently rock the transducer to localize the needle (1). In D, the needle is withdrawn slightly (2). In E, the orientation of the long axis of the transducer is corrected by aligning the lesion and the skin incision site (3) and moving the needle slightly toward the lesion (4). The needle should then be "walked" to the lesion (5) by using fine, delicate, and slow movements.
|
|
Deep Lesions
In a medium-sized or large breast, deep lesions cannot be approached from the periphery due to the distance from the periphery to the lesion. If the standard technique proves unsuccessful, deep lesions can be approached by using the biopsy needle as a lever. Entry is gained about 2 cm from the edge of the transducer, and the lesion is manually lifted away from the chest wall using the needle (Fig 6) (6). If the skin entry site is located at the periphery of the breast, the needle can be moved down in a controlled manner to the expected depth of the lesion. Levering the needle to a more horizontal plane relative to the transducer face allows the needle to be more clearly visualized, and the needle excursion upon firing will be parallel to the chest wall rather than toward it. A pool of saline solution can be deposited deep to the lesion to help raise the lesion; however, this technique is not very helpful because the fluid generally dissipates after the first pass, precluding a second pass without more saline solution being deposited (23).
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Approaching a deep lesion. Drawing illustrates how the biopsy needle is used to elevate the lesion away from the chest wall. The tip of the needle is wedged into the lesion and the gun-needle mechanism is levered down, thereby raising the needle and the lesion.
|
|
Correlation with Mammographic Findings
Before undertaking US-guided core needle biopsy of a mammographically detected lesion, one should review mammograms and US images to ensure that the lesion seen at US correlates with the mammographic finding. After US examination, a metallic marker is placed on the skin over the target lesion and mammography is performed (Fig 7). Alternatively, after obtaining a mammogram with an open-grid compression paddle, one may mark the location of the target on the skin. While compression of the breast is being maintained, the transducer is placed in the opening over the marked area, and correlation with the US image of the target lesion is possible. US-guided core needle biopsy can be performed with insertion of the needle into the side of the breast (6).
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Correlation of mammographic and US findings. Mammography performed at another hospital revealed fine, pleomorphic, clustered microcalcifications in the left breast. (a) US image reveals a cluster of tiny hyperechoic foci (arrowheads) in the left subareolar area. A BB marker was positioned on the skin overlying this cluster. (b) Left-sided coned compression magnification view shows a suspicious cluster of microcalcifications in the area of the BB marker. (c) Specimen mammogram obtained after US-guided biopsy shows the targeted microcalcifications (arrows), which were diagnosed as ductal carcinoma in situ.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Correlation of mammographic and US findings. Mammography performed at another hospital revealed fine, pleomorphic, clustered microcalcifications in the left breast. (a) US image reveals a cluster of tiny hyperechoic foci (arrowheads) in the left subareolar area. A BB marker was positioned on the skin overlying this cluster. (b) Left-sided coned compression magnification view shows a suspicious cluster of microcalcifications in the area of the BB marker. (c) Specimen mammogram obtained after US-guided biopsy shows the targeted microcalcifications (arrows), which were diagnosed as ductal carcinoma in situ.
|
|
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. Correlation of mammographic and US findings. Mammography performed at another hospital revealed fine, pleomorphic, clustered microcalcifications in the left breast. (a) US image reveals a cluster of tiny hyperechoic foci (arrowheads) in the left subareolar area. A BB marker was positioned on the skin overlying this cluster. (b) Left-sided coned compression magnification view shows a suspicious cluster of microcalcifications in the area of the BB marker. (c) Specimen mammogram obtained after US-guided biopsy shows the targeted microcalcifications (arrows), which were diagnosed as ductal carcinoma in situ.
|
|
Dense Tissue
In some patients, breast tissue may be difficult to traverse because of dense fibrosis. In such cases, a coaxial system can facilitate biopsy (24). With the coaxial needle in one hand and the transducer in the other, the radiologist can guide the needle through the skin incision and advance it just up to the lesion under US guidance. Once the coaxial needle is in place, the inner trocar can be removed and replaced by the biopsy needle. During this step, the proximal end of the coaxial needle should be held steady to avoid any displacement. The needle position can be confirmed prior to firing by indenting the lesion with the needle tip. The greatest definable diameter of the lesion in this position is sampled on the first pass. For the second and all subsequent passes, some angulation or rotation of the transducer is performed to encompass different parts of the lesion. On rare occasions, a needle may be damaged during biopsy owing to dense fibrotic tissue. The inner trocar may bend, so that the outer cannula cannot close. If this problem persists, it is helpful to take the needle out of the gun while the needle is still in the breast and manually advance the outer cannula over the inner cannula, which is held steady and not allowed to advance.
Confirmation of Lesion Retrieval
During breast biopsy, it is essential to confirm whether the target lesion has been sampled. During US-guided 14-gauge automated core needle biopsy, postfire longitudinal visualization of the needle within a lesion can indicate proper placement and retrieval but may be somewhat subjective and can be influenced by partial volume averaging effects toward the periphery of the lesion, especially if the lesion is small. Turning the probe 90° to obtain an orthogonal image before the needle is removed from the breast allows more precise evaluation of the needle position (Fig 8).With the recent development of a three-dimensional US-guided biopsy apparatus, a multiplanar display can be used and defines the precise post-firing position of the core needle biopsy track more easily than does two-dimensional US guidance (25).
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Confirmation of lesion penetration. (a) Prefire US image shows the relationship of the biopsy needle (arrow) to the lesion. (b) On a postfire US image, the needle (arrow) is seen penetrating through the lesion. (c) Orthogonal US image helps confirm that the needle (arrow) has penetrated the lesion.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Confirmation of lesion penetration. (a) Prefire US image shows the relationship of the biopsy needle (arrow) to the lesion. (b) On a postfire US image, the needle (arrow) is seen penetrating through the lesion. (c) Orthogonal US image helps confirm that the needle (arrow) has penetrated the lesion.
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8c. Confirmation of lesion penetration. (a) Prefire US image shows the relationship of the biopsy needle (arrow) to the lesion. (b) On a postfire US image, the needle (arrow) is seen penetrating through the lesion. (c) Orthogonal US image helps confirm that the needle (arrow) has penetrated the lesion.
|
|
Unlike with the automated gun technique, during US-guided directional vacuum-assisted core biopsy, the probe is positioned posterior to the lesion so that it does not overshadow and thereby obscure the lesion; the progress of lesion removal can then be monitored in real time (26). The target lesion should be located in the center of the probe aperture. The aperture can be clearly visualized on the US monitor and is indicated by the discontinuity of the anterior wall of the probe and the ring-down artifact caused by the vacuum holes on the opposite side of the aperture (Fig 9).
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Proper positioning of the probe during US-guided directional vacuum-assisted core biopsy. (a) US image shows that the probe (small arrows) is posterior to the lesion, which is nevertheless within the limits of the probe aperture. The aperture can be clearly visualized owing to the discontinuity of the anterior wall of the probe and the ring-down artifact on the opposite side of the aperture (large arrows). (b) US image shows that the cutter (arrow) has been advanced through the lesion.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Proper positioning of the probe during US-guided directional vacuum-assisted core biopsy. (a) US image shows that the probe (small arrows) is posterior to the lesion, which is nevertheless within the limits of the probe aperture. The aperture can be clearly visualized owing to the discontinuity of the anterior wall of the probe and the ring-down artifact on the opposite side of the aperture (large arrows). (b) US image shows that the cutter (arrow) has been advanced through the lesion.
|
|
The appearance and behavior of the core samples when placed in formalin for fixation is informative. Fishman et al (27) reported that intact specimens and those that sank showed correlation with diagnostic yield. These characteristics are easily assessed at biopsy. If the core sample is yellow, floats, and has fat locules dispersing from its edges onto the surface of the formalin, additional samples are needed. If the core sample is stiff, predominantly white, and sinks or dips as soon as it is put in the formalin, it is likely diagnostic (Fig 10). These processes help us determine the number of samples that we obtain to establish a diagnosis. Some authors suggest that the number of samples needed to make a definitive diagnosis of a solid mass at US-guided 14-gauge automated core needle biopsy may be less than the standard five samples that are currently obtained (28). However, recent studies support the proposal that four or five specimens should be obtained to achieve a diagnostic accuracy of more than 99% (14). For US-guided directional vacuum-assisted core biopsy, the optimal amount of core tissue has not yet been well established; however, this procedure is capable of yielding more tissue in a shorter amount of time than is 14-gauge automated biopsy (15).
View larger version (77K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Appearance of the specimen in formalin. (a) Photograph shows a core sample that is yellow and floats on the surface of the formalin, findings that indicate that the sample is insufficient for diagnosis. (b) Photograph obtained in a different patient shows a core sample that is predominantly white and sinks in the formalin, findings that indicate that the sample is probably diagnostic.
|
|
View larger version (62K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Appearance of the specimen in formalin. (a) Photograph shows a core sample that is yellow and floats on the surface of the formalin, findings that indicate that the sample is insufficient for diagnosis. (b) Photograph obtained in a different patient shows a core sample that is predominantly white and sinks in the formalin, findings that indicate that the sample is probably diagnostic.
|
|
|
Radiologic-Histologic Correlation
|
|---|
Any of the core needle biopsy procedures may fail to sample a cancer, resulting in a benign, often nonspecific histologic diagnosis, despite optimization of the technique. Specimen radiography performed during percutaneous core needle biopsy to confirm the retrieval of tissue from the target lesion can be useful for evaluating microcalcifications but is less helpful for evaluating masses (29). Postbiopsy mammography may reveal changes at the biopsy site, but there is often no substantial alteration in the target lesion, particularly if the biopsy is performed with a smaller tissue acquisition device (30). Therefore, radiologic-histologic correlation, which has traditionally played an integral role in surgical biopsy of nonpalpable breast lesions, is critical in percutaneous image-guided breast biopsy because of the limitations of other methods of confirming lesion sampling (31). The radiologic and histologic findings are considered to be concordant when the histologic findings provide an acceptable explanation for the imaging features and discordant when they do not. According to Parikh and Tickman (32), radiologic-histologic correlation can have five possible outcomes.
Concordant Malignancy.—
In concordant malignancy, a lesion that is radiologically suspicious for malignancy is diagnosed as malignant at core biopsy (eg, a spiculated mass that is diagnosed as an infiltrating carcinoma, malignant-appearing calcifications that are diagnosed as ductal carcinoma in situ). The radiologist should communicate these results to the referring physician, and both individuals should be clear as to which of them will assume responsibility for contacting the patient. After being informed of the results, the patient should be referred to a surgeon or oncologist.
Discordant Malignancy.—
In discordant malignancy, a lesion appears radiologically benign but proves to be malignant at core biopsy. Case management by the pathologist, radiologist, and referring physician should be identical to that for concordant malignancy, without any added delay.
Concordant Benignity.—
In concordant benignity, a lesion that was initially thought to be benign on the basis of prebiopsy imaging findings also demonstrates benignity at histologic analysis. Verbal or written communication between the radiologist and the referring physician is sufficient, although the patient also needs to be notified of these results. In addition, a protocol should be in place for imaging follow-up, which is necessary to monitor for delayed false-negative diagnoses at biopsy, although some benign lesions are surgically or percutaneously excised after core biopsy because of patient anxiety, patient choice, or physician preference.
Discordant Benignity.—
The most "frightening" lesion to correlate and manage is the discordant benign lesion, which is suspicious for malignancy at imaging but demonstrates benign histologic findings. The communication between the radiology and pathology departments in the setting of a possible false-negative diagnosis at core biopsy is the cornerstone of a successful core biopsy program. If there is concern regarding a discordant benign lesion after radiologic-histologic correlation has been made, it is prudent for the radiologist to immediately contact the interpreting pathologist and communicate his or her concerns. If there is persistent concern for a discordant benign core biopsy, the radiologist needs to communicate this concern to the referring physician and then to the patient.
Borderline or High-Risk Findings.—
Borderline entities are not malignant but are considered "risk marker" lesions (eg, atypical ductal hyper-plasia, lobular neoplasm, radial sclerosing lesion, papillary lesions, possible phyllodes tumors) (16). Many of these lesions have been shown to confer an increased lifetime risk for the development of breast cancer. Controversy exists regarding the increased local risk for the development of breast cancer and, therefore, regarding the surgical and oncologic treatment of these patients. If there is concern about a radiologically malignant lesion that demonstrates borderline histologic features, treatment should be identical to that for a discordant benign lesion.
Discordance between radiologic findings and (benign) histologic findings indicates that the lesion may not have been sampled adequately, which could result in a failure to diagnose carcinoma. Discordant benign lesions include (a) lesions that were radiologically suspicious for malignancy (ie, Breast Imaging Reporting and Data System [BI-RADS] category 4 lesions) in which the histologic findings did not account for the imaging features (eg, a discrete mass yielding only benign breast tissue) (Figs 11, 12); and (b) lesions that were highly suspicious for malignancy (ie, BI-RADS category 5 lesions) in which sampling yielded benign results, unless the lesion was largely or completely excised percutaneously or the specific histologic findings accounted for the imaging features (Figs 13, 14). If a discordant benign lesion is promptly recognized, repeat biopsy is warranted, and a missed cancer can be identified prospectively, thereby avoiding delay in diagnosis. Such a delay occurs when benign histologic findings are mistakenly regarded as representative of the lesion in question (33). In published reports, radiologic-histologic discordance rates for US-guided 14-gauge core needle biopsy have ranged from 2.0% to 7.7% (9,14,25,34,35), with repeat biopsy demonstrating carcinoma in 0%–50% of discordant lesions (36). For US-guided 11-gauge directional vacuum-assisted core needle biopsy, Philpotts et al (15) noted discordant results in 11 (11%) of 100 lesions, and one cancer was detected after repeat biopsy of these 11 lesions. Therefore, careful radiologic-histologic correlation made immediately after US-guided core needle biopsy will allow the detection of a substantial number of false-negative results, thereby avoiding delay in diagnosis. Liberman (5,37) recommended several steps for minimizing the likelihood of delayed diagnosis of breast cancer after percutaneous biopsy: The radiologist should not finalize the biopsy report until the histologic results have been obtained. He or she should then add an addendum to the report, discussing the histologic findings and stating whether they are concordant or discordant with the radiologic findings. If discordance exists, the radiologist should suggest that repeat biopsy (usually surgical excision) be performed.
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. Radiologic-histologic discordance in a 36-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows an oval, ill-defined hypoechoic mass in the left breast that was graded as a BI-RADS category 4 lesion. (b) Postfire US image shows the mass. The histologic diagnosis was dense intralobular fibrosis and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
View larger version (105K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. Radiologic-histologic discordance in a 36-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows an oval, ill-defined hypoechoic mass in the left breast that was graded as a BI-RADS category 4 lesion. (b) Postfire US image shows the mass. The histologic diagnosis was dense intralobular fibrosis and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12a. Radiologic-histologic discordance in a 41-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows a palpable, irregular, markedly hypoechoic mass in the left breast. The mass was graded as a BI-RADS category 4c lesion. (b) Postfire US image shows the mass. The histologic diagnosis was fibrocystic disease and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12b. Radiologic-histologic discordance in a 41-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows a palpable, irregular, markedly hypoechoic mass in the left breast. The mass was graded as a BI-RADS category 4c lesion. (b) Postfire US image shows the mass. The histologic diagnosis was fibrocystic disease and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. Radiologic-histologic discordance in a 41-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows an irregular, spiculated, markedly hypoechoic mass with posterior shadowing in the left breast. The mass was graded as a BI-RADS category 5 lesion. (b) Postfire US image shows the mass. The histologic diagnosis was fibrocystic disease with ductal epithelial hyperplasia and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. Radiologic-histologic discordance in a 41-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows an irregular, spiculated, markedly hypoechoic mass with posterior shadowing in the left breast. The mass was graded as a BI-RADS category 5 lesion. (b) Postfire US image shows the mass. The histologic diagnosis was fibrocystic disease with ductal epithelial hyperplasia and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14a. Radiologic-histologic discordance in a 65-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows an irregular, ill-defined hypo-echoic mass in the right breast (arrows) that was graded as a BI-RADS category 5 lesion. (b) Postfire US image shows the mass. The histologic diagnosis was ductal epithelial hyperplasia and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
View larger version (104K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14b. Radiologic-histologic discordance in a 65-year-old woman with invasive ductal carcinoma. US-guided 14-gauge core needle biopsy was performed. (a) Prefire US image shows an irregular, ill-defined hypo-echoic mass in the right breast (arrows) that was graded as a BI-RADS category 5 lesion. (b) Postfire US image shows the mass. The histologic diagnosis was ductal epithelial hyperplasia and was considered discordant with the US findings. Surgical excision was performed immediately, and the mass was diagnosed as invasive ductal carcinoma.
|
|
To identify radiologic-histologic discordance, the radiologist performing the biopsy must be familiar with the imaging features of a vast array of pathologic breast lesions and be able to correlate these features with histologic findings. At the time that biopsy is recommended, the radiologist should ask him- or herself what he or she is willing to accept as a diagnosis for the finding in question and, given the diagnostic considerations, select the next step. Thus, significant dedicated study of texts and journals, continuing medical education, and radiologic-pathologic conferences or correlation at one’s home institution are required to perform biopsy effectively.
|
Follow-up after a Benign Result
|
|---|
Follow-up is essential after a concordant benign diagnosis at percutaneous breast biopsy (Figs 15, 16). In studies of stereotactic core needle biopsy, Jackman et al (38) reported two missed cases in 161 cancers (1.2%) that were detected on the basis of disease progression on mammograms obtained 6 and 18 months after biopsy. Similarly, Lee et al (39) reported a false-negative diagnosis in two (2%) of 105 cancers, with the missed cancers identified on the basis of changes on mammograms obtained 6 and 24 months after biopsy. In studies of US-guided 14-gauge core needle biopsy, Crystal et al (14) reported that delayed diagnosis was made in three cases (0.9%) at 16, 23, and 27 months after biopsy, respectively. In retrospect, changes were notable in two cases at follow-up US performed 6 and 16 months after biopsy, respectively. All of these false-negative findings appeared as interval growth at follow-up US. Schoonjans and Brem (12) reported that one additional cancer was diagnosed at 6-month imaging follow-up due to an increase in size. As the follow-up time increases, additional cancers may be found that were missed at core biopsy, but there is currently no standard follow-up interval after stereotactic or US-guided core needle biopsy. For benign stereotactic biopsy results that are concordant with the imaging findings, Lee et al (39) suggest annual mammography if the histologic diagnosis is specific (eg, fibroadenoma, lymph node, cyst) and short-interval follow-up mammography of the ipsilateral breast at 6 months and of both breasts at 12, 24, and 36 months if the histologic diagnosis is nonspecific (eg, fibrocystic changes, apocrine metaplasia, benign or fibrous breast tissue). Jackman et al (38) recommend that the first follow-up study be performed 6 months after percutaneous biopsy for all lesions in which biopsy yields benign findings that are concordant with the imaging findings. According to Comstock (36), given the fact that US-guided core needle biopsy remains a sampling procedure with the potential for false-negative results, short-interval follow-up 6 months after a concordant benign diagnosis is reasonable. A follow-up interval of 1 year for lesions in which needle biopsy yields a specific benign diagnosis that is concordant with the imaging findings should be recommended only with sufficient US-guided biopsy experience and careful auditing of long-term outcomes.
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15a. Invasive ductal carcinoma in a 60-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a focal asymmetry in the right breast (arrow). Note also the circumscribed, hyperdense nodule in the outer portion of the left breast (arrowhead), a finding that represents a benign cyst. (b) US image shows an ill-defined hypoechoic lesion with a geographic pattern in the upper medial portion of the right breast corresponding to the focal asymmetry seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. Two years later, the patient presented with a palpable lump. (c) Craniocaudal mammogram shows a round, ill-defined hyperdense mass (arrow) corresponding to the earlier imaging findings (cf a, b). (d, e) US images reveal a palpable, 1.5-cm cystic mass with irregular wall thickening and increased blood flow in the upper medial portion of the right breast. Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (65K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15b. Invasive ductal carcinoma in a 60-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a focal asymmetry in the right breast (arrow). Note also the circumscribed, hyperdense nodule in the outer portion of the left breast (arrowhead), a finding that represents a benign cyst. (b) US image shows an ill-defined hypoechoic lesion with a geographic pattern in the upper medial portion of the right breast corresponding to the focal asymmetry seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. Two years later, the patient presented with a palpable lump. (c) Craniocaudal mammogram shows a round, ill-defined hyperdense mass (arrow) corresponding to the earlier imaging findings (cf a, b). (d, e) US images reveal a palpable, 1.5-cm cystic mass with irregular wall thickening and increased blood flow in the upper medial portion of the right breast. Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (158K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15c. Invasive ductal carcinoma in a 60-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a focal asymmetry in the right breast (arrow). Note also the circumscribed, hyperdense nodule in the outer portion of the left breast (arrowhead), a finding that represents a benign cyst. (b) US image shows an ill-defined hypoechoic lesion with a geographic pattern in the upper medial portion of the right breast corresponding to the focal asymmetry seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. Two years later, the patient presented with a palpable lump. (c) Craniocaudal mammogram shows a round, ill-defined hyperdense mass (arrow) corresponding to the earlier imaging findings (cf a, b). (d, e) US images reveal a palpable, 1.5-cm cystic mass with irregular wall thickening and increased blood flow in the upper medial portion of the right breast. Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15d. Invasive ductal carcinoma in a 60-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a focal asymmetry in the right breast (arrow). Note also the circumscribed, hyperdense nodule in the outer portion of the left breast (arrowhead), a finding that represents a benign cyst. (b) US image shows an ill-defined hypoechoic lesion with a geographic pattern in the upper medial portion of the right breast corresponding to the focal asymmetry seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. Two years later, the patient presented with a palpable lump. (c) Craniocaudal mammogram shows a round, ill-defined hyperdense mass (arrow) corresponding to the earlier imaging findings (cf a, b). (d, e) US images reveal a palpable, 1.5-cm cystic mass with irregular wall thickening and increased blood flow in the upper medial portion of the right breast. Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (163K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15e. Invasive ductal carcinoma in a 60-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a focal asymmetry in the right breast (arrow). Note also the circumscribed, hyperdense nodule in the outer portion of the left breast (arrowhead), a finding that represents a benign cyst. (b) US image shows an ill-defined hypoechoic lesion with a geographic pattern in the upper medial portion of the right breast corresponding to the focal asymmetry seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. Two years later, the patient presented with a palpable lump. (c) Craniocaudal mammogram shows a round, ill-defined hyperdense mass (arrow) corresponding to the earlier imaging findings (cf a, b). (d, e) US images reveal a palpable, 1.5-cm cystic mass with irregular wall thickening and increased blood flow in the upper medial portion of the right breast. Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16a. Invasive ductal carcinoma in a 53-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a small, isodense nodule in the left breast (arrow). (b) US image shows an ill-defined hypoechoic lesion (arrow) corresponding to the nodule seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. (c, d) Follow-up mammogram (c) and US image (d) obtained 6 months later show an interval increase in the size of the nodule (arrow in c). Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (175K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16b. Invasive ductal carcinoma in a 53-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a small, isodense nodule in the left breast (arrow). (b) US image shows an ill-defined hypoechoic lesion (arrow) corresponding to the nodule seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. (c, d) Follow-up mammogram (c) and US image (d) obtained 6 months later show an interval increase in the size of the nodule (arrow in c). Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16c. Invasive ductal carcinoma in a 53-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a small, isodense nodule in the left breast (arrow). (b) US image shows an ill-defined hypoechoic lesion (arrow) corresponding to the nodule seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. (c, d) Follow-up mammogram (c) and US image (d) obtained 6 months later show an interval increase in the size of the nodule (arrow in c). Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
View larger version (183K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16d. Invasive ductal carcinoma in a 53-year-old woman. The cancer was missed at US-guided core biopsy but was detected at mammographic-US follow-up. (a) Craniocaudal mammogram shows a small, isodense nodule in the left breast (arrow). (b) US image shows an ill-defined hypoechoic lesion (arrow) corresponding to the nodule seen at mammography (cf a). The lesion was graded as a BI-RADS category 4a lesion, and biopsy was performed. The histologic diagnosis was fibrocystic disease and was considered concordant with the imaging findings. (c, d) Follow-up mammogram (c) and US image (d) obtained 6 months later show an interval increase in the size of the nodule (arrow in c). Biopsy was performed, and the lesion was diagnosed as invasive ductal carcinoma.
|
|
Follow-up data may be difficult to obtain. In a study of 160 breast lesions in which percutaneous biopsy had been performed, Goodman et al (40) reported only 74% compliance with recommended surgical excision of lesions and 54% compliance with recommended follow-up imaging surveillance. In a study by Smith et al (9), follow-up was available for 74% of all benign lesions. Patient compliance with the recommended follow-up schedule is essential if further delay in diagnosing missed malignant lesions is to be avoided. Follow-up requires a substantial commitment of time and resources but is necessary for patient care and for impro
Top
Abstract
LEARNING OBJECTIVES
