| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Breast Imaging Section, Memorial Sloan-Kettering
Cancer Center, 1275 York Ave., New York, NY 10021.
2
Department of Surgery, Breast Service, Memorial Sloan-Kettering Cancer Center,
New York, NY 10021.
Received November 30, 2000;
accepted after revision March 13, 2001.
Address correspondence to L. Liberman.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. We prospectively examined mammograms of 1057 consecutive lesions that had preoperative needle localization and surgical excision and classified the lesions according to Breast Imaging Reporting and Data System (BI-RADS) final assessment categories. Bracketing wires, defined as multiple wires placed to delineate the boundaries of a single lesion, were used in 103 (9.7%) of 1057 lesions. Medical records, imaging studies, and histologic findings in these 103 lesions were reviewed.
RESULTS. Of 103 bracketed lesions, median lesion size was 3.5 cm (range, 1.5-9.5 cm). Ninety-three lesions (90.3%) contained calcifications; 65 lesions (63.1%) were BI-RADS category 5 (highly suggestive of malignancy); and 33 lesions (32.0%) were percutaneously proven cancers. The median number of wires placed was two (range, 2-5). Surgical histologic findings were carcinoma in 75 lesions (72.8%), atypical hyperplasia in eight lesions (7.8%), and benign in 20 lesions (19.4%). Of 42 calcific lesions that were bracketed and had postoperative mammograms available for review, complete removal of suspicious calcifications was accomplished in 34 (81.0%). Of 75 cancers that were bracketed, clear histologic margins of resection were obtained in 33 (44.0%).
CONCLUSION. Bracketing wires were used during preoperative needle localization primarily for larger calcific lesions that were proven cancers or were highly suggestive of malignancy (BI-RADS category 5). Bracketing wires may assist the surgeon in achieving complete excision of calcifications, but bracketing wires do not ensure clear histologic margins of resection.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The surgeon performing breast conservation must balance the goals of achieving adequate excision with maintaining a good cosmetic result. Achieving this balance can be difficult, particularly in women with relatively large areas of cancer who desire breast conservation. For nonpalpable breast cancers, delineation of the extent of the mammographic abnormality may help the surgeon to excise the entire lesion with a minimal volume of tissue. Silverstein et al. [7, 8] suggested that preoperative placement of bracketing wires, defined as multiple wires to delineate the boundaries of a single lesion, may be useful in this regard. The purpose of our study was to evaluate the outcomes of bracketing wire placement during preoperative breast needle localization.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Needle localizations were performed with modified Kopans spring-hook localization needles (Cook, Bloomington, IN), which contain a thickened or reinforced portion proximal to the wire tip. Each lesion that had needle localization was evaluated before surgery by the radiologist performing the needle localization. The radiologist recorded the breast parenchymal density and the lesion type, size, and final assessment category according to the Breast Imaging Reporting and Data System (BI-RADS) [9] lexicon as category 3 (probably benign), category 4 (suspicious), or category 5 (highly suggestive of malignancy).
Bracketing wires were placed in 103 (9.7%) of 1057 lesions that underwent needle localization. These 103 lesions, which occurred in 101 women with a median age of 56 years (range, 35-85 years), constitute the basis of this article.
Bracketing Wire Technique
The decision to place bracketing wires was made by the radiologist
performing the needle localization in consultation with the referring surgeon.
Bracketing was generally performed for larger lesions that were highly
suggestive of malignancy, larger areas of percutaneously proven carcinoma, and
other lesions for which wider excision was desired. The number of bracketing
wires to be used was also determined by consultation between the radiologist
and referring surgeon. Two wires were used in most cases, particularly for
calcifications in a clustered or linear distribution, but three or more wires
were occasionally used for large lesions of marked geometric complexity, such
as segmentally distributed calcifications (Fig.
1A,1B,1C,1D,1E,1F,1G).
|
|
|
|
|
|
|
Bracketing wire placement was performed using dedicated mammography equipment (LoRad MIV; LoRad, Danbury, CT). The approach for wire placement was determined by review of the preoperative mammograms (as is done in conventional needle localization procedures) on the basis of considerations such as the distance from the lesion to the skin surface, lesion conspicuity and accessibility, and patient positioning. In general, the longest dimension of the lesion was chosen for bracketing wire placement (e.g., if the lesion's longest dimension was anteroposterior, an attempt was made to delineate the anterior and posterior borders of the lesion). In most cases, the boundaries of the lesion were encompassed within one of two compression paddles, either a perforated compression paddle with 36 holes covering a surface area of approximately 7.5 x 6.5 cm or a rectangular compression paddle with an opening covering a surface area of 7.0 x 5.0 cm (Fig. 1A,1B,1C,1D,1E,1F,1G). Optimal wire placement was considered to be within 5 mm of the lesion border, but not necessarily beyond the confines of the lesion. No freehand localizations were performed.
After needle position was confirmed with orthogonal (craniocaudal and mediolateral) views, the wires were deployed, and the same views were obtained to confirm wire position (Fig. 1A,1B,1C,1D,1E,1F,1G). The films were labeled, and a labeled diagram was drawn that delineated the location of the lesion in relation to the localizing wires. In some instances, the referring surgeon and radiologist also discussed the positioning after needle localization was completed. In all cases, the labeled films and diagram accompanied the patient to the operating room. Specimen radiographs were obtained (Fig. 1A,1B,1C,1D,1E,1F,1G) and evaluated intraoperatively.
Postoperative Imaging and Surgical Protocols
During the study period at our institution, breast surgery was performed by
seven full-time breast surgeons and one general oncologic surgeon who had
extensive experience in breast surgery. For women with surgically proven
malignant calcifications in whom breast conservation was planned, a
postoperative mammogram was obtained no earlier than 2 weeks after surgery to
determine if all of the calcifications had been excised. If no residual
calcifications were identified on conventional views, the protocol for this
postoperative mammogram included magnification images of the surgical bed
(Fig.
1A,1B,1C,1D,1E,1F,1G).
If residual calcifications were present at the surgical site, needle
localization and reexcision were recommended. Postoperative mammograms were
not routinely performed in women with malignant masses or with benign
histologic findings. Surgical specimens were analyzed by pathologists
specializing in oncologic pathology. The histologic margins were considered
"close" if the tumor extended to within 1 mm of the inked margin
[10]. In women treated with
breast conservation, additional surgery was recommended if the tumor extended
as far as, or close to, the margins of resection.
Data Entry and Analysis
Medical records, mammograms, specimen radiographs, and histologic findings
in these 103 lesions were reviewed to determine the indications for
bracketing, the frequency of achieving complete removal of the mammographic
lesion, and the likelihood of obtaining clear histologic margins of resection
for carcinomas. Data were recorded in a computerized spread sheet (Excel;
Microsoft, Redmond, WA). Statistical analyses were performed with the
chi-square test and the Fisher's exact test using a computerized statistics
program (Epi-Info; Centers for Disease Control, Atlanta, GA).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Surgical histologic findings in 103 bracketed lesions were carcinoma in 75 lesions (72.8%), atypical hyperplasia in eight lesions (7.8%), and benign in 20 lesions (19.4%). Frequency of cancer as a function of mammographic lesion type and BI-RADS category is shown in Table 1. Among the 75 cancers, histologic findings were ductal carcinoma in situ (DCIS) in 37 and infiltrating carcinoma in 38 (including infiltrating ductal carcinoma and DCIS in 31, mixed infiltrating ductal and lobular carcinoma and DCIS in three, infiltrating lobular carcinoma and DCIS in two, infiltrating ductal carcinoma in one, and infiltrating lobular carcinoma in one). The median size of infiltrating carcinoma was 1.0 cm (range, 0.2 -6.0 cm). Axillary surgery (sentinel lymph node biopsy, axillary dissection, or both) performed in 35 patients with infiltrating carcinoma revealed axillary metastases in 12 patients (34.3%).
|
Among the eight lesions yielding atypical hyperplasia, histologic analysis showed atypical ductal hyperplasia in seven and atypical lobular hyperplasia in one. Among the 20 lesions yielding benign findings, surgical histology yielded duct hyperplasia and sclerosing adenosis in five, fibrocystic change in four, duct hyperplasia in three, sclerosing adenosis in two, and single cases of calcifications in fibroadenoma, papilloma, fibrosis, and pseudoangiomatous stromal hyperplasia; in two lesions yielding calcifications in benign tissue, lobular carcinoma in situ was also found in the surgical specimens. None of the lesions yielding atypia or having benign findings had undergone percutaneous biopsy.
Of 42 calcific lesions that were bracketed and had postoperative mammograms available for review, complete removal of calcifications was achieved in 34 (81.0%). Of 31 malignant calcific lesions that were bracketed and had postoperative mammograms available, complete removal of calcifications was achieved in 23 (74.2%). Among malignant calcific lesions, the likelihood of obtaining clear histologic margins was significantly higher if calcifications were completely removed than if they were not completely removed (16/23 = 69.6% versus 2/8 = 25.0%; p < 0.05).
Clear histologic margins of resection were obtained in 33 (44.0%) of 75 cancers (Figs. 1A,1B,1C,1D,1E,1F,1G and 2A,2B,2C,2D). The likelihood of tumor at the margins was significantly higher in breasts that were dense as compared with those that were mildly or moderately dense (10/11 = 90.9% versus 32/64 = 50.0%; p < 0.02) (Table 2). No significant difference in the frequency of tumor at the margins was observed as a function of mammographic lesion size, calcification distribution, or prior percutaneous diagnosis (Table 2). Of 42 cancers that were bracketed and had tumor at the margins, 20 (47.6%) were subsequently treated with mastectomy.
|
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At our institution, bracketing wires were placed in 103 (9.7%) of 1057 lesions that had preoperative needle localization during a 1-year period, and were primarily placed for larger areas of calcifications that were highly suggestive of malignancy (BI-RADS category 5) or those that had proven to be cancer at percutaneous biopsy. Complete excision of the mammographic finding was accomplished in most bracketed lesions and was particularly helpful for women with malignant calcifications. Gluck et al. [12] reported that the positive predictive value of residual calcifications after lumpectomy for carcinoma was 69% and suggested reexcision with preoperative needle localization for residual calcifications. We found that removal of all of the calcifications at the initial surgery spared the need for a repeated needle localization procedure and was associated with a significantly higher likelihood of achieving clear histologic margins. In eight (19.0%) of 42 calcific lesions that were bracketed, residual calcifications were present on postoperative mammograms (Fig. 3A,3B,3C,3D,3E); this may reflect the fact that two wires, as placed for most of our lesions, do not completely delineate the arrangement of the lesion in three dimensions.
|
|
|
|
|
Clear histologic margins were obtained in less than half (44%) of the cancers that were bracketed, perhaps in part as a result of case selection. The mammographic lesions bracketed were large. Some were not ideal candidates for breast conservation, but excision was being performed for diagnosis or because of the patient's strong desire to preserve her breast. Furthermore, the mammogram may underestimate the extent of disease, particularly in dense breasts and in lesions that contain DCIS (present in 97.3% of the cancers in this study) [13,14,15]. To achieve clear margins on a large lesion containing DCIS requires a wide excision that may compromise cosmesis. Some surgeons may have preferred to start with a smaller excision, particularly when there was no preoperative tissue diagnosis of cancer, with the knowledge that additional surgery could be performed later if tumor was present at the margins. We hypothesize that clear margins may have been obtained even less frequently without bracketing wires, but we cannot test this hypothesis because our study design was not randomized.
How can we increase the likelihood of achieving clear histologic margins in women with larger carcinomas while preserving as much uninvolved breast tissue as possible? First, increased use of percutaneous biopsy for preoperative diagnosis may be helpful [16]. Previous studies have shown that clear margins are more frequently obtained in lesions that are percutaneously proven cancers than in lesions without a preoperative diagnosis [17,18,19]. In addition, preoperative diagnosis of carcinoma in different quadrants may lead to mastectomy, bypassing a fruitless attempt at breast conservation [20, 21]. Second, placement of a larger number of bracketing wires may be useful. Silverstein [22] has suggested placing four or more wires to better define the often complex arrangements of malignant calcifications in the breast parenchyma (Fig. 1A,1B,1C,1D,1E,1F,1G); an attempt should be made to encompass the mammographic lesion within the confines of the bracketing wires. Third, careful intraoperative analysis of specimen radiographs can be helpful: Graham et al. [23] found that radiographic evidence of cancer at the edge of the specimen had a high (62/63 = 98%) positive predictive value for tumor at the margin, although the negative predictive value of a radiographically clear specimen margin was low (18/56 = 32%). Fourth, sonography [24, 25] and MR imaging [26, 27] may provide better delineation of local extent of disease than does mammography, particularly for women with dense breasts; perhaps using these ancillary imaging modalities (or, in the operating room of the future, real-time intra-operative MR imaging) in bracketing would facilitate attaining clear margins. Finally, case selection remains of paramount importance because some lesions may not be amenable to breast conservation, in spite of our best efforts.
Our study had several limitations. The study design was retrospective and lacked a control group of similar cases that did not undergo bracketing wire placement. Furthermore, the criteria and techniques for bracketing wire placement were not rigorously standardized. However, we used bracketing wires in approximately 10% of needle localization procedures performed during the study period. If our results were generalized to the national level, assuming that preoperative needle localization will be used in approximately 30% of the 1 million breast biopsies to be performed this year in the United States, then bracketing wires may be used in as many as 30,000 women (10% x 30% x 1,000,000). Bracketing wires may be used more often with the increased use of screening mammography for detection, percutaneous biopsy for diagnosis, and breast conserving surgery for treatment of breast cancer. Our study is the first attempt, to our knowledge, to scientifically evaluate the out-comes of bracketing wire placement, and it should be followed by rigorous prospective investigations.
In conclusion, we found that bracketing wires were placed most often for larger calcific lesions that were highly suggestive of malignancy (BI-RADS category 5) or for percutaneously proven cancers. Bracketing wires may assist the surgeon in achieving complete excision of calcifications, but bracketing wires do not ensure clear histologic margins of resection. Our results reflect the challenges of breast conservation in women with relatively large breast cancers, and they reinforce the importance of patient selection, postbiopsy mammography to assess for residual calcifications, and careful attention to histologic margins in women undergoing breast conserving surgery [5, 6]. Further work is necessary to determine the usefulness of other techniques in delineating local extent of disease and in expediting surgical treatment for women with relatively large breast cancers who desire breast conservation, to make possible complete excision of tumors with minimal cosmetic deformity.
Acknowledgments
We thank Melvin J. Silverstein for critical review of the manuscript and
David C. Perlman for invaluable assistance.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Mariscal Martinez, M. Sola, A. P. de Tudela, J. F. Julian, M. Fraile, S. Vizcaya, and J. Fernandez Radioguided Localization of Nonpalpable Breast Cancer Lesions: Randomized Comparison With Wire Localization in Patients Undergoing Conservative Surgery and Sentinel Node Biopsy Am. J. Roentgenol., October 1, 2009; 193(4): 1001 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Whelan, J. Julian, M. Levine, R. Arriagada, I. Kunkler, H. Bartelink, J.-C. Horiot, and P. Poortmans Radiotherapy for Breast Cancer N. Engl. J. Med., March 14, 2002; 346(11): 862 - 864. [Full Text] [PDF]
|
|
|
|
|
|
L. Liberman, S. L. Goodstine, D. D. Dershaw, E. A. Morris, L. R. LaTrenta, A. F. Abramson, and K. J. V. Zee One Operation After Percutaneous Diagnosis of Nonpalpable Breast Cancer: Frequency and Associated Factors Am. J. Roentgenol., March 1, 2002; 178(3): 673 - 679. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
