Atlas of Mammography
The preoperative localization of a nonpalpable breast abnormality is necessary for the accurate identification of the lesion and its removal. The procedure involves an integration of breast imaging, surgery, and pathology, with close interaction by all members of this team to assure optimization of the procedure. Careful communication of the information regarding the case is needed among the radiologist, surgeon, and pathologist to confirm that the correct area is removed and that the lesion is identified within the specimen.
The preoperative localization of a nonpalpable mammographic abnormality offers multiple advantages. Using preoperative localization, a smaller amount of tissue can be excised than without radiographic guidance. This is important not only from the patient's standpoint, because of a lesser degree of postoperative deformity, but also from the pathologist's standpoint, because the ease and accuracy of identifying a tiny focus of carcinoma in situ are improved when the amount of tissue to be sectioned is less.
Prior to the Procedure
Prior to the procedure, the radiologist must carefully review the mammogram, other imaging studies, and pathology records if the lesion already has been biopsied. As percutaneous breast biopsy has evolved, the vast majority of suspicious lesion on mammography are biopsied by core biopsy methods instead of surgical excision. Prior to this era, clinically occult, mammographically detected lesions were excised following needle localization. Now, few lesions are excised for diagnosis. Instead, the vast majority of needle-localized lesions are cancers, high-risk lesions, or lesions that were insufficiently sampled or unable to be sampled percutaneously.
Prior to the procedure, it also is important that the complete imaging evaluation be performed of the patient who has not already undergone a percutaneous biopsy. A complete diagnostic mammogram is needed to confirm that the lesion is actually suspicious and warrants biopsy. For example, magnification views and a mediolateral (ML) view are necessary in the case of microcalcifications, to evaluate them and to confirm that they do not represent milk of calcium. The ML view also is necessary to plan the approach to the lesion, because it is the orthogonal view to the craniocaudal (CC) view.
If the lesion has been already biopsied, the post-biopsy mammogram should be reviewed for several reasons. First, one must determine if the mammographic finding is still present. Second, if a clip was placed following biopsy, one should verify whether the clip is at the biopsy site or if it is deep or shallow to it. If one is planning to target the clip, it is critical to know that the clip is deployed at the biopsy site. Once the image review and plan for the procedure are complete, the procedure is explained to the patient, and informed consent is obtained.
Mammographically Guided Needle Localization
The techniques for localization are varied, and have included triangulation and needle placement (1), needle localization with dye injection (2,3), and needle localization with wire placement (4,5,6). A number of wires are available, most of which have a type of hook (4) (Fig. 15.1) or J configuration (5) when released from the needle. The spring hookwire breast lesion localizer designed by Kopans (4,7), combined with a rigid compression plate, allows for accurate and safe localization of breast lesions. The compression plate has a fenestrated window, which maintains the breast in position and allows for insertion of the needle containing the hookwire in a direction parallel to the chest wall. Once deployed, the wire is in position and cannot be repositioned or withdrawn. In the situation in which a hookwire is placed in a position suboptimal for surgical excision, a second wire should be placed rather than trying to reposition the first wire.
A J-wire was designed by Homer (8,9) to allow retraction and repositioning of the wire if needed. The curved-end wire has a memory that allows it to be retracted into the needle and readvanced into the breast tissue in the
same shape. This type of wire is also thicker and not transected during surgical dissection.
Figure 15.1 A typical needle and hookwire used for needle localization. The wire has a stiffener near the distal end that aids the surgeon in identifying it on palpation.
Many mammographic units are equipped with a localization device. Two basic designs are available: a rectangular aperture or multiple 1-cm round perforations (6) in a plastic compression plate (Fig. 15.2). The technical performance of needle localization is improved with the use of digital mammography. Dershaw et al. (10) found that the time to complete a needle localization was reduced by 50% when digital mammography was used, and the mean glandular dose was reduced by a similar amount.
Figure 15.2 A localization plate with 1-cm round perforations is in place with the breast compressed for needle localization.
Figure 15.3 Procedure for mammographically guided needle localization. The needle is placed into the breast through the appropriate aperture (A), the opposite view is made to determine the depth of the needle, the needle position is adjusted if necessary (B), and the wire is ejected (C).
Prior to the localization, a 90-degree lateral (ML) view is obtained if only CC and mediolateral oblique (MLO) projections are available. From these images, the closest skin surface to the abnormality is determined—superior, medial, lateral, or inferior. The localization plate is placed over the surface determined, and an image is obtained (Fig. 15.3). While the patient remains in position with her breast compressed, the coordinates of the aperture overlying the lesion are determined. A localization needle is placed into the breast, parallel to the chest wall at the indicated location; the depth of placement is estimated only, because the breast is compressed by the localization device. An image is obtained with the needle in place, to determine if the placement is accurate.
The localization plate is then carefully removed, leaving the needle in place, and the orthogonal view is made. If the needle is too deep, it is withdrawn until the tip is within the lesion, and then the wire is deployed. Most wires have a mark that, when positioned at the hub of the needle, indicates that the tip of the wire is at the tip of the needle. The wire is deployed by advancing this mark into the needle hub by about 1 cm. The needle is removed, the wire is taped to the skin, and two final images—CC and ML views—are made to show the position of the hook relative to the lesion (Fig. 15.4). Information regarding the procedure is communicated to the surgeon. This should include the location of the lesion targeted, the approach to the lesion used, the length of wire within the breast, the position of the hook relative to the lesion, and the need for a specimen radiograph.
Figure 15.4 Needle localization series performed for a mass in the left breast that was found to be malignant. Left ML (A) and CC (B)views show the dense, lobular, indistinct mass in the 12 o'clock position. On the CC image, the localization plate is in place over the superior surface of the breast. Because the lesion is closest to the superior aspect of the breast (12 o'clock position), the localizer plate was placed in this position to provide the shortest distance for needle insertion.
The needle has been placed into the aperture directly over the lesion on the CC view (C).
The localizer plate has been removed, and the ML view (D) obtained, showing that the needle is directly through the lesion. Focal asymmetry is present inferior to the mass, which developed during the procedure, representing a small hematoma. The needle was withdrawn slightly and the wire was deployed.
Postprocedure ML (E) and CC (F) views show that the hookwire position is accurate. The specimen film (G) shows that the lesion is contained within the specimen and that a margin of normal tissue appears to be present around it.
For a dye localization, the needle is placed in a similar manner but, instead of placing a hookwire through the needle, dye is injected. Because the dye dissipates rather quickly, the scheduling of the localization and the operating room must be coordinated carefully to avoid a prolonged lapse between the two procedures. Because of the
dissipation of the dye, a prolonged time lapse leads to the excision of an unnecessarily large specimen. Dye alone is uncommonly used today, but some surgeons prefer a combined localization using both dye and a wire in case the wire is inadvertently displaced.
The types of dye that have been used include methylene blue, isosulfan blue, and toluidine blue. Methylene blue may interfere with estrogen-receptor assays (11). Isosulfan blue has been developed as an alternative to methylene blue (12), because it does not interfere with
receptors, and it is commonly used for sentinel node mapping as well. Toluidine blue has also been suggested as an alternative to methylene blue because it diffuses less quickly (13).
Stereotactically-Guided Needle Localization
Stereotaxis affords a rapid and precise way to visualize and localize a lesion for surgical excision. Early experience with stereotactic guidance was negative (14,15), because the hookwires were deployed either too shallowly or too deeply, relative to the lesion. The challenge with stereotaxis is understanding how to handle the wire with the breast compressed along the plane of insertion, as it is in stereotactic procedures. Once this is understood, the procedure is a very accurate and fast way of localizing the lesion.
For stereotactic localization (16), the breast is compressed with the aperture placed over the skin surface at a point that represents the shortest distance to the lesion. If the lesion has already been biopsied percutaneously, the same trajectory used for the biopsy is used for the localization procedure (Fig. 15.5). The technologist positions the patient with the aperture of the compression plate over the area of the lesion. A 0-degree scout image for confirmation may be obtained, followed by a stereotactic pair of images at a fixed angle of obliquity specific to the type of unit (usually 10 or 15 degrees) (Figs. 15.6 and 15.7). The reference point on the stereotactic images is confirmed, and the target point is identified. The length of the localization needle (not the wire) is programmed into the unit, and the coordinates are transferred to the table or are entered. The needle guides are positioned over the lesion on the X and Y axes by dialing the guides to a ΔX = 0 and ΔY = 0 position. At this position, no difference is present in the position of the target point in the lesion and the needle guide on the X or Y axes. For needle localizations, the ΔZ-position of the needle is ideally placed at 10mm deep to the lesion. This is a vital step for a successful procedure. Depending on the type of stereotactic unit, this may be a plus or minus ΔZ-position.
Figure 15.5 Patient in position for needle localization with stereotactic guidance using an upright unit. She is seated because a superior approach to the lesion is being utilized for this case. She would be lying on her side if an inferior, medial, or lateral needle placement were being performed. If the localization were performed on a stereotactic prone table, the patient would be lying prone for any needle placement.
After the needle has been placed at ΔX = 0, ΔY = 0, ΔZ = (±) 10.0, a stereotactic pair of images is obtained. On this set of images, the needle tip should appear to be just through the lesion on both views, because the ΔZ is set past or deep to the lesion. Once the needle position has been confirmed, the wire is deployed. Errors can occur in this second vital step. The wire is inserted until its tip is at the needle tip. On most wires, an indicator mark on the wire is present that, when placed at the level of the needle hub, indicates that the wire tip is at the needle tip. The wire is not inserted further. Instead, the wire is grasped firmly, and the needle is withdrawn over it. Compression is released, the patient is moved from the stereotactic unit, and the final postprocedure CC and mediolateral (ML) views are obtained. Because the breast is compressed along the plane of needle insertion, any advancement of the wire can cause it to be very deep to the lesion, once the compression is released. Similarly, allowing the wire to slip back as the needle is withdrawn causes the hook to deploy very superficially to the lesion.
Ultrasound-Guided Needle Localization
Ultrasound guidance is an excellent method for needle localization (17,18). For lesions visible on sonography, guidance using ultrasound affords a fast and accurate method for localization that is associated with no radiation exposure. Sonography is particularly useful for localization of a mass that is awkward to approach using a fenestrated plate, or for patients who have difficulty sitting up.
Several companies have developed clips embedded in materials (collagen, Gelfoam) that are visible on ultrasound.
If a patient has a percutaneous breast biopsy, and one of these clips is placed, the area may be evident on ultrasound several weeks later. If surgical excision is needed, the clip may be evident on ultrasound and is used as the target point for the localization procedure using sonographic guidance.
When ultrasound guidance is used for needle localization, a vertical needle insertion (Fig. 15.8) is preferred (18). The technique, as described by Fornage (18), allows for the shortest distance to the lesion from the skin. The lesion is visualized in the middle of the scan field, and the needle is inserted at the transducer's midpoint. The degree of obliquity of the needle insertion depends on the depth of the lesion. A more vertical angle is used for deeper lesions. Using this method, the needle tip is not seen until it crosses the scan plane at the depth of the lesion. Care must be taken to not penetrate the chest wall; therefore, this technique is not for inexperienced interventionalists.
Figure 15.6 HISTORY: 31-year-old with a palpable thickening in the right upper outer quadrant.
MAMMOGRAPHY: Right MLO (A) and CC (B) views show a small indistinct mass located at 11 o'clock (arrow). This was biopsied using ultrasound guidance, and a clip was placed. Pathology showed invasive ductal carcinoma. Post biopsy images (C) and (D) show the clip at the site of the abnormality, which is no longer evident after core biopsy. Stereotactic needle localization was performed, and post–wire placement ML (E) and CC (F) views show accurate hookwire placement at the site.
IMPRESSION: Successful localization of invasive ductal carcinoma. Surgical pathology from the lumpectomy showed residual invasive ductal carcinoma.
A safer angle of insertion is an oblique (Fig. 15.9) or horizontal placement of the needle. The lesion is visualized at the edge of the scan field, and the needle is placed at the end of the transducer. The needle is visualized in its entirety when it is within the scan plane. This technique is best for lesions located near the chest wall, in patients with very thick breasts, and in patients with implants. This technique also should be used by those with less interventional experience. The disadvantage is that the length of wire within the breast is greater than it is for the vertical procedure and, therefore, the surgery may be more difficult. With either a vertical or horizontal approach, however, ultrasound guidance is a fast and accurate method of needle localization that has a high level of patient comfort and acceptance (Fig. 15.10).
Problems in localizing lesions are sometimes caused by lack of clear visibility of the lesion while the localization device is in place, a very posterior lesion that is difficult to keep in position for the localization procedure, and lesions that are evident on only one mammographic view. Prior to the procedure, it is important to completely work up the lesion with imaging to establish a clear understanding of where the lesion is expected to be, so that the localization plate is positioned correctly. The mediolateral view is used for a variety of reasons including determining the position of a lesion seen only on the MLO view. Rolled CC views help to identify whether a lesion seen only on the CC view is in the upper or lower half of the breast.
At times, the localization plate can obscure faint abnormalities. The comparison with the preprocedure films is important to focus on the area in which the lesion is expected to be located. Repositioning the breast slightly is sometimes helpful to move the lesion under an aperture and to visualize it.
For a very posterior lesion, the patient must be encouraged to stay in position, leaning forward, to avoid the breast slipping back during the procedure. Marking the skin near the aperture of the compression plate is also important to observe if any patient motion has occurred.
For lesions seen on only one view, a parallax approach, with the placement of several needles using mammographic guidance, has been described (19). A far simpler method is the use of stereotactic guidance (16). Using stereotaxis, the patient is positioned with the aperture over the breast on the view on which the lesion is seen. Stereotactic images are obtained to determine the depth for needle placement.
Localization of Known Cancers
The localization of lesions that represent known cancers must take into consideration the goal of the procedure. In this situation, the goal is to remove the entire tumor with a clear margin. This objective differs from the case
in which localization of an indeterminate lesion is being performed for biopsy, when margins are much less important and removing the lesion with the least amount of normal tissue is performed. In the cancer patient, the role of the radiologist is to map out the extent of the mammographic lesion for the surgeon, to optimize excision of all the tumor, thus reducing the likelihood of the need for re-excision.
Figure 15.7 HISTORY: 53-year-old woman with a history of right breast needle biopsy showing a questionably atypical lesion, presents for second opinion regarding her management.
MAMMOGRAPHY: Left MLO (A) and CC (B) views show a small cluster of amorphous microcalcifications (arrow) located at 5 o'clock that are suspicious and require biopsy. The clip and a few residual microcalcifications from the prior right breast biopsy are seen on the right ML (C) and CC (D) magnification views (arrows) as well. Stereotactic biopsy of the left breast calcifications was performed and revealed ductal carcinoma in situ. Because of the findings in both breasts, stereotactically guided needle localizations and excisions were performed.
Left stereotactic pair (E) shows the clip marking the biopsy site. The T at the bottom of the field is the reference point used for calculation of the Z position of the lesion. The clip was targeted, and the needle is in accurate position on the subsequent stereotactic pair (F).
Right stereotactic pair (G) shows the clip to be targeted. Post–needle placement stereotactic pair shows the needle in correct position(H).
Final wire placement ML (I) and CC (J) images show accurate hookwire position in the lesions.
IMPRESSION: Successful localization of left DCIS and right atypia. Histopathology from the excisions showed ductal carcinoma in situ bilaterally.
Depending on the type of lesion, one or more wires may be used. For the solitary mass, the placement of a single hookwire through the lesion is ideal. However, for a mass with associated surrounding calcifications or for a
segment or region of calcifications that are ductal carcinoma in situ (DCIS), one wire is not the best choice. Instead, the placement of several wires to bracket the margins of the region of mammographic abnormality will guide the surgeon better in understanding the orientation and scope of the lesion (Fig. 15.11). Marking the anterior and posterior borders and/or the medial and lateral or superior and inferior margins can be performed. The choice of which borders to bracket depends on the orientation of the lesion. For example, if the lesion is a large linear area of calcifications extending from the subareolar area posteriorly, two wires marking the anterior and posterior margins are sufficient. However, for a lesion that extends transversely as well, medial and lateral wires may
be used instead of or in addition to the anterior/posterior markers (Fig. 15.12).
Figure 15.8 Vertical needle localization with ultrasound guidance. The needle is placed vertically into the breast in the center of the long axis of the transducer (A). When viewed from the opposite direction, at the short end of the transducer, the needle tip is seen at the lesion as it bisects the scan plane (B). The angle of the needle insertion depends on the depth of the lesion.
Figure 15.9 Oblique/horizontal needle insertion using ultrasound guidance. The needle is placed horizontally into the breast along the short end of the transducer and is visualized in its entirety as it is inserted (A) toward the lesion. When viewed enface (B), it is seen to be parallel to the scan plane. This technique is required for core biopsies, but it is optional for cyst aspiration or for needle localization.
Figure 15.10 Right MLO (A) and CC (B) views show a small, dense, indistinct mass located laterally (arrow). This was biopsied using ultrasound guidance and found to be an infiltrating ductal carcinoma. Subsequently, the area was needle localized using ultrasound guidance prior to lumpectomy. Preprocedure sonography (C) shows the microlobulated mass. Using an oblique insertion, the needle has been placed through the lesion in D, and it is visualized along the path leading to the lesion. The needle tip is just through the lesion. In E, the needle has been removed, and the wire has been deployed in the mass.
Postprocedure ML (F) and CC (G) views show the hookwire in position with the hook just through the mass.
IMPRESSION: Successful needle localization of right breast cancer using ultrasound guidance.
Figure 15.11 Schematic showing the procedure for localization of a region of microcalcifications. A: If one wire is placed in the center of the region (B), the majority of calcifications may be excised, but the others that represent extension of carcinoma in the periphery may be missed. By placing two wires to bracket the anterior and posterior margins (C) the entire lesion is more accurately excised.
Figure 15.12 HISTORY: A 38-year-old referred for an abnormal baseline mammography.
MAMMOGRAPHY: Left MLO (A) and CC magnification (B) views show heterogeneously dense tissue. Segmentally oriented pleomorphic microcalcifications are located at 4 o'clock (arrows). These were biopsied stereotactically, showing ductal carcinoma in situ (DCIS), high nuclear grade. A second small focus of amorphous microcalcifications (arrowhead) laterally was also biopsied and was benign. A clip was deployed at this second site.
A bracketing procedure using stereotactic guidance was performed for the needle localization of the region of DCIS prior to lumpectomy. Preliminary stereotactic images (C, D) show the region of calcifications in the field of view. Needles were placed to mark the superior and inferior margins of the abnormality, and needle placement is shown on the stereotactic pairs (E, F and G, H).
Final wire placement ML (I) and CC (J) images show the hookwires marking the superior and inferior margins of the mammographic finding.
IMPRESSION: Successful bracketing of area of DCIS using stereotactic guidance. Clear margins were obtained at surgery.
Magnetic Resonance Imaging-Guided Localization
As magnetic resonance imaging (MRI) has evolved and is utilized for the evaluation of the extent of disease, so has the need for MRI-guided interventions. If the lesion identified on MRI is visible in retrospect on mammography or ultrasound, biopsy or needle localization is performed using the simpler methods. However, if the lesion is visible only on MRI, the biopsy or needle localization using MRI guidance is necessary.
MRI breast coils and biopsy devices that compress the breast and help to localize the lesion are used. A high-resolution study is performed and, if the lesion is visible without contrast,
the needle may be placed through the localizer grid at the proper coordinates. An MRI-visible marker, such as vitamin E, is used as a fiducial and placed on the skin to help to orient the location of the lesion. Contrast is injected to confirm the needle position at the lesion, and the wire is deployed. MRI-compatible needles and hookwires are available for the procedure (20,21,22).
Complications and Risks
The potential complications of needle localization include bleeding and hematoma formation, infection, wire migration, vasovagal reaction, pneumothorax, and allergic reaction to dye that may be used. The most common problem by far is vasovagal reactions, which may
occur in up to 7% of patients (23). Because of this, the team should be prepared by having a stretcher available and by having ammonia capsules and ice on hand. The interventional team should be experienced and able to keep the patient comfortable during the procedure, to distract her from the procedure with conversation, to perform the procedure as quickly and efficiently as possible, and to be ready to manage a vasovagal reaction if it occurs.
Prolonged bleeding is an uncommon problem during needle localization. Helvie et al. (24), in a review of 370 procedures, reported a 1% rate of bleeding for more than 5 minutes. Hematomas are infrequent and, if significant, are usually related to a vascular puncture (25). If care is used in needle placement, the risks of the procedure are minimal. Although a pneumothorax could occur, attention to placement of a localization wire parallel to the chest wall should avoid this problem. To avoid migration or a lost wire, it is important that a wire to be used of sufficient length to allow at least 10 cm to protrude from the breast at the end of the procedure (26).
The wire may be transected during the procedure, and the surgeon may lose perspective as to the location of the lesion and the hook, accounting for unsuccessful excision of the lesion. Some surgeons make an incision somewhere near the wire entry point in the direction of the lesion, intentionally clip the wire at the skin and excise the tissue en bloc, pulling the remainder of wire in. Others, who follow the wire down to the lesion, prefer a stiffener on the wire near the hook to help to locate it by palpation in case the distal wire is inadvertently transected.
In patients treated for breast cancer diagnosed by a needle-localization biopsy procedure, no documented increase has been noted in local recurrence rate to suggest seeding of tumor cells along the wire tract (27). In patients with implants, a potential exists for rupture during localization. Care must be taken during the localization of a lesion in a breast augmented with a prosthesis to avoid puncturing the prosthesis, particularly if it contains silicone. Robertson et al. (28) described wire localization in eight patients with implants, with placement of the needle using the implant displacement technique (29). Another option in these patients is to place the localization grid over the breast and mark the lesion with skin markers only, giving instructions as to the depth of the lesion from the skin surfaces.
Failure to surgically excise a localized lesion has been reported to occur in 0% to 17.9% of cases (30,31,32,33). Jackman et al. (34) reported a biopsy failure rate of 2.5% in 280 lesions. Unsuccessful localizations (34) were associated with smaller lesions, small specimens, more than one lesion in the breast, and microcalcifications. Removal of a second specimen converted the failed procedure to a successful one in 67% of cases (34). Causes of lesion missed at surgery are a poorly positioned wire, transection of the wire at surgery, wire migration in a fatty breast, movement of the wire in preparing the patient for surgery or in performing the surgery, and poor communication between the radiologist and the surgeon about the position of the hook relative to the lesion. The miss rates reported with dye localization have been 2.2% (34) and 1.3% (32). A successful localization has been described as having the wire tip within 5 mm of the lesion (35).
Specimen radiography is an integral part of the needle localization and excisional biopsy procedure and should be performed in all cases (36,37,38,39). The specimen is radiographed for several reasons including: (a) to verify that the lesion is included, (b) to verify that the hookwire has been removed, and (c) to describe the relationship of the lesion to the margins of the specimen radiographically.
The technique for specimen radiography of a mammographically visible lesion involves an analog or digital radiograph. A mammographic unit or a dedicated specimen radiographic unit may be utilized. Magnification and compression (40) are used with low kVp techniques (22,23kVp) and low mAs (39). Specimen containers are available that are ideal for transportation of the tissue; these contain a fenestrated compression plate that compresses the tissue (Fig. 15.13). Care must be taken to not overcompress the tissue, which could potentially lead to distortion of the margins (41). In addition, the fenestrated grid allows the radiologist to identify the location of the abnormality within the tissue for the pathologist (Fig. 15.14). Additional levels or cuts through the area of greatest interest may be performed. At times, radiography of the sliced specimen block is performed to localize the microcalcifications for pathologic analysis. When calcifications on the localized lesion are not evident to the pathologist on the initial slides, polarized light should be used to look for calcium oxalate crystals, which are not evident on routine hematoxylin and eosin staining. If calcium oxalate is not seen, the tissue block is radiographed and recut, with new slides made to search for the lesion (39).
For noncalcified lesions, Stomper et al. (37) found that film screen specimen radiography was a highly effective procedure for correctly identifying the presence of the mammographic abnormality. In a study of 104 specimen radiographs after excision of a clinically occult noncalcified lesion, the authors found that 93% of the lesions were visible on the specimen films. On comparison with the original mammograms, the specimen films showed better anatomic detail of the lesion in 35% of cases (37).
Figure 15.13 The patient had undergone core biopsy of a region of microcalcifications showing DCIS at two foci. Clips were placed after the biopsy. Needle localization using two hookwires was performed prior to lumpectomy. The specimen has been placed in a container with a fenestrated plate over it, compressing the tissue. The tissue is radiographed in this manner so that the region of interest can be indicated to the pathologist. The extent of the calcifications indicating the area of DCIS have been marked for the pathologist to focus on in sectioning and analyzing the tissue. The tissue is sent to the lab in the container and with the radiograph as marked.
Figure 15.14 Specimen radiography shows a small cancer contained within the specimen. It does not appear to extend to the margins. The hookwire is included.
Two orthogonal views of the specimen may be performed routinely and are particularly helpful in specific circumstances. When the lesion is not clearly seen on one view, the orthogonal view may demonstrate the abnormality. For cancers, in which the relationship of the abnormality to the surgical margins should be described to the surgeon while the patient is still in the operating room, the analysis of two views of the specimen is more complete (Fig. 15.15).
The specimen radiograph is certainly not sufficient to determine the presence or absence of tumor at the surgical margin, but it is helpful to the surgeon in deciding whether to excise more tissue at the initial lumpectomy. The false-negative rate of specimen radiography for tumor was found to be 44% and the false-positive rate was 21% (42).
For lesions localized using ultrasound and seen only on ultrasound, specimen radiography may demonstrate the lesion. However, sonography of the specimen using a high-resolution transducer and a stand-off pad may be necessary (43,44,45). Following MRI localization, specimen radiography remains a challenge (46). The deployment of a clip, if MRI biopsy is performed first, may aid in specimen radiography; however, for lesions excised for diagnosis, this is not feasible.
Most series report a true-positive rate of 10% to 35% in needle localization series (47,48,49,50,51,52,53,54,55,56,57,58,59,60). Much of this data precedes the advent of stereotactic biopsy and reflects more the positive predictive value of a suspicious mammographic lesion. The vast majority of benign lesions biopsied represent some form of fibrocystic disease, although proliferative lesions, which increase the risk of the patient to develop breast cancer by 1.9 (without atypia) to 5.3 (with atypia) times, account for a significant number of nonpalpable lesions biopsied (57). Without meticulous techniques and attention to subtle signs of malignancy, early breast cancers will be missed. Rosenberg et al. (56) found that, in a series of 927 needle-guided breast biopsies, 29% were malignant, and 30% of the patients with invasive lesions had axillary nodal metastases. Hermann et al. (60) retrospectively reviewed the mammograms of 220 women who underwent needle localization biopsy procedures and classified the lesions as “probably benign” or “probably malignant.” The radiologic diagnosis was correct in 68% of cases; 27 of the cancers were retrospectively interpreted as “probably benign” and would have been missed had there not been an aggressive approach to indeterminate lesions. A high-quality screening program coupled with percutaneous biopsy and needle localization biopsy procedures for suspicious or indeterminate lesions will yield increasing numbers of early breast cancers at biopsy.
Figure 15.15 HISTORY: A 56-year-old woman who had presented for screening mammography.
MAMMOGRAPHY: Left ML view (A) and specimen film (B). A high-density spiculated lesion is present deep in the breast (curved arrow), and two other areas of nodularity are located more superficially (straight arrows) (A). Between these nodules are extensive ductal-type microcalcifications. These findings are highly suspicious for extensive malignancy, and it is important that the extent of disease be demonstrated. In approaching the interventional procedure for this patient, it is best to place two wires for needle localization, with one marking the anterior extent of the abnormality, and the other the more posterior extent. The specimen radiograph shows the region of interest included, as demonstrated by the two hookwires that were placed for the needle localization procedure.
IMPRESSION: Two needle localization wires placed to mark extensive area of microcalcifications and nodularity, consistent with breast carcinoma.
HISTOPATHOLOGY: Infiltrating ductal and extensive intraductal carcinoma.
- Kalisher L. An improved needle for localization of nonpalpable breast lesions. Radiology1978;128:815–817.
- Edeiken S, Suer WD, Vitale SF, et al. Needle localization of nonpalpable breast lesions using methylene blue. Breast Dis1990;3:75–80.
- Raininko R, Linna MI, Rasanen O. Preoperative localization of nonpalpable breast tumours. Acta Chir Scand1976;142: 575–578.
- Kopans DB, DeLuca S. A modified needle-hookwire technique to simplify preoperative localization of occult breast lesions. Radiology1980;134:781.
- Homer MJ. Nonpalpable breast lesions localization using a curved-end retractable wire. Radiology1985;157: 259–260.
- Goldberg RP, Hall FM, Simon M. Preoperative localization of nonpalpable breast lesions using a wire marker and perforated mammographic grid. Radiology1983;146: 833–835.
- Kopans DB, Lindfors K, McCarthy KA, Meyer JE. Spring hookwire breast lesion localizer: Use with rigid-compression mammographic systems. Radiology1985;157:537–538.
- Homer MJ, Pile-Spellman RE. Needle localization of occult breast lesions with a curved-end retractable wire: technique and pitfalls.Radiology1986;161:547–548.
- Homer MJ. Nonpalpable breast lesion localization using a curved-end retractable wire. Radiology1985;157:259–260.
- Dershaw DD, Fleischman RC, Liberman L, et al. Use of digital mammography in needle localization procedures. Am J Roentgenol1993;161:559–562.
- Hirsch JI, Banks WL, Sullivan JS, Horsley JS. Effect of methylene blue on estrogen-receptor activity. Radiology1989;171: 105–107.
- Hirsch JI, Banks WL, Sullivan JS, Horsley JS. Noninterference of isosulfan blue on estrogen-receptor activity. Radiology1989; 171:109–110.
- Czarnecki DJ, Feider HK, Splittgerber GF. Toluidine blue dye as a breast localization marker. Am J Roentgenol1989;153: 261–263.
- Jackson VP. Needle localization to guide excisional biopsy. RSNA Categorical Course in Breast Imaging. RSNA Publications, Radiological Society of North America, Oak Brook, IL. 1995; 161–166.
- Jackson VP, Bassett LW. Stereotactic fine-needle aspiration biopsy for nonpalpable breast lesions. Am J Roentgenol1990; 154:1196–1197.
- Shaw de Paredes E, Langer TG, Cousins J. Interventional breast procedures. Curr Probl Diagn Radiol1998;27:133–184.
- D'Orsi CJ, Mendelson EB. Interventional breast ultrasonography. Semin Ultrasound CT MRI1989;10:132–138.
- Fornage BD, Coan JD, David CL. Ultrasound-guided needle biopsy of the breast and other interventional procedures. Radiol Clin N Am1992;30:167–185.
- Kopans DB, Waitzkin ED, Linesky L, et al. Localization of breast lesions identified on only one mammographic view. Am J Roentgenol1987;149:39–41.
- Fischer U, Vosshenrich R, Keating D, et al. MR-guided biopsy of suspect breast lesions with a simple stereotaxic add-on device for surface coils. Radiology1994;192:272–273.
- Schnall MD, Orel SG, Connick TJ. MR guided biopsy of the breast. MRI Clin North Am1994;2:585–589.
- Gorczyca DP. Interventional breast imaging resonance imaging. In: Interventional Breast Procedures, 1996. New York: Churchill Livingstone; 1996:147–153.
- Dershaw DD. Needle localization for breast biopsy. In: Dershaw DD, ed. Interventional Breast Procedures. New York: Churchill Livingstone; 1996:25–35.
- Helvie MA, Ikeda DM, Adler DD. Localization and needle aspiration of breast lesions: complications in 370 cases. Am J Roentgenol1983;41:929–930.
- Sistrom C, Abbitt PL, Shaw de Paredes E. Hematoma of the breast: a complication of needle localization. Va Med1988; 115:78–79.
- Davis PS, Wechsler RJ, Feig SA, March DE. Migration of breast biopsy localization wire. Am J Roentgenol1988;150: 787–788.
- Kopans DB, Gallagher WJ, Swann CA, et al. Does preoperative needle localization lead to an increase in local breast cancer recurrence? Radiology1988;167:677–668.
- Robertson CL, Kopans DB, McCarthy KA, Hart NE. Nonpalpable lesions in the augmented breast: preoperative localization. Radiology1989;173:873–874.
- Eklund GW, Busby RC, Miller SH, Job JS. Improved imaging of the augmented breast. Am J Roentgenol1988;151:469–473.
- Landercasper J, Gundersen SB Jr., Gundersen AL, et al. Needle localization and biopsy of nonpalpable lesions of the breast. Surg Gynecol Obstet1987;164:399–403.
- Parker SH, Lovin JD, Jobe WE, et al. Stereotactic breast biopsy with a biopsy gun. Radiology1990;176:741–747.
- Alexander HR, Candela FC, Dershaw DD, Kinne DW. Needle-localized mammographic lesions: results and evolving treatment strategy.Arch Surg1990;125:1441–1444.
- Winchester DP. Limitations of mammography in the identification of noninfiltrating carcinoma of the breast. Surg Gynecol Obstet1988;167:135–140.
- Jackman R, Marzoni FA. Needle-localization breast biopsy: why do we fail? Radiology1997;204:677–687.
- Abrahamson PE, Dunlap LA, Amamoo A, et al. Factors predicting successful needle-localized breast biopsy. Acad Radiol2003;10:610–606.
- Snyder RE. Specimen radiography and preoperative localization of nonpalpable breast cancer. Cancer1980;46:950–956.
- Stomper PC, Davis SP, Sonnenfeld MR, et al. Efficacy of specimen radiography of clinically occult noncalcified breast lesions. Am J Roentgenol1988;151:43–47.
- Rebner M, Pennes DR, Baker DE, et al. Two view specimen radiography in surgical biopsy of nonpalpable breast masses. Am J Roentgenol1987;149:283–285.
- D'Orsi CJ. Management of the breast specimen. Radiology1995;194:297–302.
- Chilcote WA, Davis GA, Suchy P, Paushter DM. Breast specimen radiography: evaluation of a compression device. Radiology1988;168:425–427.
- Clingan R, Griffin M, Phillips J, et al. Potential margin distortion in breast tissue by specimen mammography. Arch Surg2003;138:1371–1374.
- Lee CH, Carter D. Detecting residual tumor after excisional biopsy of impalpable breast carcinoma: efficacy of comparing preoperative mammograms with radiographs of the biopsy specimen. Am J Roentgenol1995;164:81–86.
- Frenna TH, Meyer JE, Sonnenfeld MR. US of breast biopsy specimens. Radiology1994;190:573.
- Laing FC, Jeffrey RB, Minagi H. Ultrasound localization of occult breast lesions. Radiology1984;151:795–796.
- Mesurolle B. El-Khoury M, Hori D, et al. Sonography of post-excision specimens of nonpalpable breast lesions: value, limitations, and description of a method. Am J Roentgenol2006;186:1014–1024.
- Morris EA, Liberman L, Dershaw DD, et al. Preoperative MR-imaging-guided needle localization of breast lesions. Am J Roentgenol2002;178:1211–1220.
- Libshitz HI, Feig SA, Fetouh S. Needle localization of nonpalpable breast lesions. Radiology1976;121:557–560.
- Hall FM, Frank HA. Preoperative localization of nonpalpable breast lesions. Am J Roentgenol1979;132:101–105.
- Gisvold JJ, Martin JK Jr. Prebiopsy localization of nonpalpable breast lesions. Am J Roentgenol1984;143:477–481.
- Oppedal BR, Drevvatne T. Radiographic diagnosis of nonpalpable breast lesions: correlation to pathology. Acta Radiol (Diagn)1983;24:259–265.
- Meyer JE, Kopans DB, Stomper PC, Lindfors KK. Occult breast abnormalities: percutaneous preoperative needle localization.Radiology1984;150:335–337.
- Hoehn JL, Hardacre JM, Swanson MK, Williams GH. Localization of occult breast lesions. Cancer1982;49:1142–1144.
- Yankaskas BC, Knelson MH, Abernathy ML, et al. Needle localization biopsy of occult lesions of the breast: experience in 199 cases.Invest Radiol1988;23:729–733.
- Hall FM, Storella JM, Silverstone DZ, Wyshak G. Nonpalpable breast lesions: recommendations for biopsy based on suspicion of carcinoma at mammography. Radiology1988;167:353–358.
- Ciatto S, Cataliotti L, Distanne V. Nonpalpable breast lesions detected with mammography: review of 512 consecutive cases.Radiology1987;165:99–102.
- Rosenberg AL, Schwartz GF, Feig SA, Patchefsky AS. Clinical occult breast lesions: localization and significance. Radiology1987;162:167–170.
- Rubin E, Visscher DW, Alexander RW, et al. Proliferative disease and atypia in biopsies performed for nonpalpable lesions detected mammographically. Cancer1988;161:2077–2082.
- Hallgrimsson P, Karesen R, Artun K, Skjennald A. Nonpalpable breast lesions: diagnostic criteria and preoperative localization. Acta Radiol1988;29:285–288.
- Meyer JE, Sonnenfeld MR, Greene RA, Stomper PC. Preoperative localization of clinically occult breast lesions: experience at a referral hospital. Radiology1988;169: 627–628.
- Hermann G, Janus C, Schwartz IS, et al. Nonpalpable breast lesions: accuracy of prebiopsy mammographic diagnosis. Radiology1987;165:323–326.