| 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 August 3, 2001;
accepted after revision September 6, 2001.
Supported by grant C015709 from the New York State Department of
Health.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Retrospective review was performed of records of 350 consecutive women who had therapeutic surgery after percutaneous imaging-guided core biopsy diagnosis of nonpalpable breast cancer. Records were reviewed to determine the frequency of performing one operation and associated factors. Statistical analysis was performed.
RESULTS. One operation was performed in 283 (80.9%) of 350 women, including 106 (95.5%) of 111 women who had mastectomy and 177 (74.1%) of 239 women who had breast conserving surgery. At bivariate analysis, one operation was significantly more likely in women who had no underestimation (p < 0.001), mastectomy rather than breast conservation (p < 0.001), axillary dissection during the first operation (p < 0.001), percutaneous diagnosis of infiltrating carcinoma (p = 0.001), or mammographic mass (p = 0.006). At multivariate analysis, one operation was significantly more likely if underestimation was absent (odds ratio [OR] = 10.1, 95% confidence interval [CI] = 4.2-24.7) or if mastectomy was performed (OR = 8.7, 95% CI = 3.2-23.5); for women who had breast-conserving surgery, one operation was significantly more likely if underestimation was absent (OR = 11.4, 95% CI = 3.9-33.2) or if a mammographic mass was present (OR = 2.4, 95% CI = 1.3-4.6).
CONCLUSION. One operation was performed in 80.9% of women with percutaneously proven nonpalpable breast cancer, including 74.1% of women who had breast-conserving surgery and 95.5% of women who had mastectomy. Among women who had breast conservation, one operation was significantly more likely if histologic underestimation was absent or if a mammographic mass was present.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tissue was acquired with an 11-gauge vacuumassisted biopsy probe (Mammotome; Biopsys/Ethicon Endo-Surgery, Cincinnati, OH) in 216 women (61.7%), with a 14-gauge automated needle (Manan, Manan Medical Products, Northbrook, IL; Biopty-Cut, Bard Urological, Covington, GA; Ultra-Core, Medical Device Technologies, Gainesville, FL) in 128 women (36.6%), with both methods in three women (0.9%), and with an 8-gauge vacuum-assisted device (Minimally Invasive Breast Biopsy; United States Surgical, Norwalk, CT) in three women (0.9%).
Stereotactic biopsies were performed with patients prone on a dedicated table with digital imaging (StereoGuide with Digital Spot Mammography; LoRad, Danbury, CT). Tissue was acquired for stereotactic biopsies with the 11-gauge vacuum-assisted biopsy probe (Mammotome) in 218 women and with an 8-gauge vacuum-assisted biopsy probe (Minimally Invasive Breast Biopsy) in three women. The range of specimens acquired was seven to 42 (median, 15). Specimen radiography was performed for all lesions evident as calcifications. A localizing clip was placed if the mammographic lesion was removed.
Sonographically guided biopsies were performed with patients in the supine or supine-oblique position, using high resolution sonographic equipment (128XP, Acuson, Mountain View, CA; Ultramark 4 Plus, Advanced Technology Laboratories, Bothell, WA). Tissue was acquired with a 14-gauge automated core biopsy needle (Manan, Biopty-Cut, or Ultra-Core) in 131 women and with an 11-gauge vacuum-assisted biopsy probe in one. The range of specimens obtained was one to 10 (median, four).
Needle-Localization Technique
For patients with percutaneously proven nonpalpable breast cancer who
underwent breast conserving surgery, needle localization was performed on the
day of the surgery with modified Kopans spring hook localizer wires (Cook,
Bloomington, IN). Bracketing wires (two or more wires to delineate the
boundaries of the mammographic lesion) were placed at the discretion of the
radiologist performing the needle localization in consultation with the
surgeon, in general for larger (e.g., 
1.5 cm) lesions
[11]. After localization,
labeled films and a diagram were sent with the patient for use during
surgery.
Breast and Axillary Surgery
All surgeries were performed by one of 10 oncologic surgeons specializing
in breast surgery. The decisions regarding breast conserving surgery versus
mastectomy, the need for axillary surgery, and the choice of axillary surgery
(sentinel lymph node biopsy vs axillary dissection) were made by the surgeon
in consultation with the patient, on the basis of clinical, imaging, and
pathologic considerations and patient preferences. Absolute contraindications
to breast conservation included history of therapeutic irradiation to the
breast and two or more gross tumors in separate quadrants
[12].
Axillary surgery was performed for women with infiltrating carcinoma, unless the patient had a prior axillary dissection or comorbid or other conditions that precluded axillary surgery. Axillary surgery was also performed selectively for a subset of patients with ductal carcinoma in situ (DCIS), such as some women with extensive DCIS requiring mastectomy or lesions yielding DCIS with microinvasion [13].
Pathologic Analysis
All percutaneous biopsy and surgical specimens underwent pathologic
analysis at our institution. Histologic underestimation was considered present
for lesions yielding DCIS at percutaneous biopsy and infiltrating carcinoma at
surgery [14]. For surgical
specimens, margins were considered positive if tumor was identified at the
inked margin of the specimen, and margins were considered close if tumor
extended to within 1 mm of the inked margin
[15].
Sentinel lymph nodes were evaluated by frozen section, paraffin section, and immunohistochemistry [16]. In general, if sentinel lymph nodes were tumor-free, no further axillary surgery was performed. If sentinel lymph nodes contained tumor evident on frozen section, axillary dissection was performed during the same operative procedure. If frozen section of sentinel nodes did not reveal tumor but tumor was identified at paraffin section or immunohistochemistry, the patient usually returned for axillary dissection at a later date.
Indications for Additional Surgery
At our institution, additional surgery was recommended after surgical
excision of breast cancer in several scenarios. If tumor extended to or close
to the margins of resection, the patient usually underwent additional surgery,
either reexcision or mastectomy based on the consensus of the surgeon and
patient in consideration of individual preferences of patients, imaging
studies, and histologic findings
[12,
17]. For women with
calcifications shown to be carcinoma, additional surgery was recommended if
postoperative mammography performed approximately 2 weeks after surgery showed
residual calcifications at the biopsy site
[18]; this additional surgery
was either reexcision with preoperative needle localization or mastectomy. For
women who had sentinel lymph node biopsy, axillary dissection was usually
performed as a separate procedure if sentinel node metastases were identified
at paraffin or immunohistochemical analysis
[16].
Data Collection and Entry
Medical records were reviewed to determine the number of operations
undergone by each patient. The frequency of performing a single operation was
calculated, and reasons for performing two or more operations were determined.
Factors recorded included initial breast cancer treatment (mastectomy vs
breast conserving surgery), mammographic findings, core biopsy histology,
mammographic lesion size, guidance modality, tissue-acquisition device,
patient age, surgeon, histologic underestimation, and axillary surgery
(axillary dissection or sentinel lymph node biopsy) during the initial
procedure.
Prospective lesion classification was recorded according to the final assessment categories of the Breast Imaging Reporting and Data System (BIRADS) [19] lexicon as 4 (suspicious) or 5 (highly suggestive of malignancy). Breast density was recorded as 1 (fatty), 2 (scattered fibroglandular densities), 3 (heterogeneously dense), and 4 (dense); for purposes of statistical analysis, class 1 and 2 were considered "not dense," and class 3 and 4 were considered "dense." Data were entered into a computerized spreadsheet (Excel; Microsoft, Redmond, WA) for analysis.
Statistical Analysis
Statistical analysis of data from all 350 women was performed using
statistical software, SAS version 8.01 (SAS Institute, Cary, NC). The LOGISTIC
procedure in SAS was used for unconditional logistic regression analyses
[20]. Bivariate and
multivariate analyses were performed for all women, for women who had
breast-conserving surgery, and for women who had mastectomy.
Factors considered potential predictors of performing one operation included mass appearing on mammography (yes, no), percutaneous diagnosis (infiltrating carcinoma, DCIS without invasion), therapeutic operation (mastectomy, breast conservation), axillary surgery during first operation (yes, no), and histologic underestimation (yes, no). Forward, backward, and stepwise selection procedures resulted in the same model for the respective analyses. The Hosmer and Lemeshow goodness of fit test [20] revealed that the models adequately fit the data.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Final surgical treatment was breast conserving in 225 (64.3%) and mastectomy in 125 (35.7%). Cancer histology was DCIS in 114 (32.6%) of 350 women and infiltrating carcinoma in 236 women (67.4%) (including 193 with DCIS). The range in histologic size of infiltrating carcinoma was 0.1-3.5 cm (median, 1.0 cm). Axillary surgery, performed in 221 women, revealed metastases in 48 (21.7%).
Number of Operations
Among 350 women with percutaneously proven nonpalpable breast cancer, one
operation was performed in 283 (80.9%) (Figs.
1A,1B,1C,1D
and
2A,2B),
and two or more operations were performed in 67 (19.1%). The range of
operations was one to four (median, one; mean, 1.2). One operation was
performed in 106 (95.5%) of 111 women who had mastectomy versus 177 (74.1%) of
239 women who attempted breast conservation (p < 0.001). Among 239
women who attempted breast conservation, the final treatment was breast
conserving in 225 (94.1%) and mastectomy in 14 (5.9%).
|
|
|
|
|
|
Reasons for performing two or more operations in 67 women included tumor at or close to the margins in 54 (80.6%; including 11 with histologic underestimation, five with residual calcifications identified at postoperative mammography, and two with tumor in sentinel nodes), histologic underestimation of DCIS in six (9.0%), tumor in sentinel nodes in five (7.5%), and tumor in a nonsentinel lymph node palpable intraoperatively in one (1.5%). In one patient, the surgical excision confirmed the percutaneous biopsy diagnosis of recurrent carcinoma after prior breast conserving surgery; the patient had subsequent mastectomy.
Of 54 women who had tumor at or close to the margins of the first excision, the second surgery revealed residual carcinoma in 14 (25.9%). Of 17 women who had axillary surgery during the second operation due to histologic underestimation of DCIS, two (11.8%) had sentinel node metastases. Of seven women who had axillary dissection during the second operation due to tumor in sentinel nodes, one (14.3%) had an additional axillary nodal metastasis. In the woman who had tumor in a nonsentinel node, subsequent axillary dissection revealed an additional nodal metastasis. In the patient with recurrent carcinoma after prior breast-conserving surgery, mastectomy revealed no additional tumor.
Bivariate Analysis
All women.—At bivariate analysis of data from our 350 women,
one operation was significantly more likely in women who had no histologic
underestimation (p < 0.001), mastectomy rather than breast
conserving surgery (p < 0.001), axillary dissection during the
first operation (p < 0.001), percutaneous diagnosis of
infiltrating carcinoma rather than pure DCIS (p = 0.001), and mass
rather than calcifications without mass on mammography (p = 0.006)
(Table 1). The likelihood of
performing one operation did not depend on the treating surgeon and did not
differ significantly as a function of breast parenchymal density (p =
0.118), biopsy guidance modality (p = 0.118), tissue acquisition
device (p = 0.130), BI-RADS category (p = 0.161), patient
age (p = 0.242), or mammographic lesion size (p =
0.370).
|
Women who had breast conserving surgery as the first operation.—At bivariate analysis of data from 239 women who had breast conserving surgery as the first operation, one operation was significantly more likely in women who had no histologic underestimation (p < 0.001), mass rather than calcifications without mass on mammography (p < 0.001), percutaneous diagnosis of infiltrating carcinoma rather than pure DCIS (p < 0.001), axillary surgery during the first operation (p = 0.003), low breast density (p = 0.006), and age 50 years and older (p = 0.038) (Table 2). The likelihood of undergoing one operation did not depend on the treating surgeon and did not differ significantly as a function of biopsy guidance modality (p = 0.119), tissue acquisition device (p = 0.138), BI-RADS category (p = 0.237), or mammographic lesion size (p = 0.241).
|
Women who had mastectomy as the first operation.—Among 111 women who had mastectomy as the first operation, one therapeutic operation was performed in 106 (95.5%). In these 111 women, bivariate analysis revealed no factors associated with a significantly higher frequency of performing one operation as opposed to performing two or more surgeries.
Multivariate Analysis
At multivariate analysis, women were significantly more likely to have one
operation if there was no histologic underestimation (odds ratio [OR] = 10.1,
95% confidence interval [CI], 4.2-24.7) or if they had a mastectomy (OR = 8.7,
95% CI, 3.2-23.5) (Table 3). When the analysis was limited to those who had breast-conserving surgery,
women were significantly more likely to have one operation if there was no
histologic underestimation (OR = 11.4, 95% CI, 3.9-33.2) or if they had a
mammographic mass (OR = 2.4, 95% CI, 1.3-4.6)
(Table 3).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
In our study, one operation was performed in 80.9% of women with percutaneously proven cancer, including 95.5% of women who had a mastectomy and 74.1% of women who had breast-conserving surgery. At multivariate analysis, women were significantly more likely to have one surgery if there was no histologic underestimation or if they had a mastectomy. Although performing mastectomy as the initial procedure in all cases would increase the likelihood of one operation, preserving the breast (if feasible) is a higher priority for many women, even if it may entail more surgical procedures. Among women who had breast-conserving surgery, one operation was significantly more likely if there was no histologic underestimation or if there was a mammographic mass.
Tumor at or close to the margins of resection was the most common reason for performing more than one operation. Tumor at or close to margins led to additional surgery in 80.6% of women who required more than one operation and in 15.4% of all women in our study. Obtaining clear margins diminishes the chance of local recurrence [23, 24]. The likelihood of tumor at the margins is significantly lower in women with cancers diagnosed percutaneously rather than surgically (Table 4) because the surgeon, aware that the diagnosis is cancer, intentionally resects more tissue [2, 3, 5, 7]. Achieving clear margins is more difficult in DCIS (usually evident as calcifications on mammography) than in infiltrating carcinoma (usually evident as a mass lesion) [25, 26]. Furthermore, the mammogram may provide a more accurate assessment of the extent of disease in patients with mammographic masses rather than with calcifications [27, 28]. These factors may contribute to the higher frequency of one operation in breast conservation patients who had a mass rather than calcifications on mammography.
Histologic underestimation of DCIS was the second most common reason for performing more than one operation. Underestimation contributed to the need for additional surgery in 25.3% of women who required more than one operation and in 4.9% of all women in the study. Histologic underestimation is most commonly encountered during percutaneous biopsy of calcifications [14]. The frequency of histologic underestimation is significantly lower for 11-gauge vacuum-assisted biopsy rather than for 14-gauge automated core biopsy. In published reports of lesions yielding DCIS at 14-gauge automated core biopsy, surgery yielded infiltrating carcinoma in 16-35% [14, 29,30,31,32]; of lesions yielding DCIS at 11-gauge vacuum-assisted biopsy, surgery revealed infiltrating carcinoma in 6-15% [14, 30, 32].
We found that histologic underestimation was associated with a significantly higher likelihood of more than one operation. If percutaneous biopsy yields DCIS and surgery yields infiltrating carcinoma, the patient may require a second operation to evaluate the axilla. The need for a second surgery due to histologic underestimation is diminished if selected patients with percutaneously proven DCIS (such as those requiring mastectomy or with possible microinvasion) undergo axillary surgery during the first operation. Sentinel lymph node biopsy may be particularly valuable in this setting: Klauber-DeMore et al. [13] found sentinel node metastases in 12% of selected patients with DCIS at high risk for invasion and in 10% of patients who had DCIS with microinvasion.
The finding of tumor in sentinel nodes was the third most common reason for more than one operation. Sentinel node metastases contributed to the need for additional surgery in 10.4% of women who required more than one operation and in 2.0% of all women in the study. Previous studies of sentinel node biopsy have revealed 93% sensitivity, 100% specificity, and 97% accuracy [16]. Axillary metastases are present in approximately half of women who have tumor in sentinel nodes at paraffin analysis and in approximately 15-20% of women who have tumor in sentinel nodes at immunohistochemical analysis [16]. The treatment of women with sentinel node metastases (particularly those detected by immunohistochemistry) remains controversial, but many researchers consider the presence of tumor in sentinel nodes to be an indication for axillary dissection [16]. Although sentinel node biopsy has the advantage of low morbidity, it carries with it a chance of requiring a second operation if sentinel node metastases are found.
How can we maximize the likelihood of performing one operation in women with percutaneously proven breast cancer, particularly those who desire breast conservation? Patient selection is paramount: percutaneous diagnosis of cancer in more than one quadrant may indicate the need for mastectomy, avoiding an unsuccessful attempt to conserve the breast [33]. Use of 11-gauge vacuum-assisted technology for percutaneous biopsy significantly decreases the likelihood of histologic underestimation [14]. Sonography [34] or MR imaging [35] may provide better preoperative assessment of the extent of disease, particularly for women with dense breasts, and may increase the likelihood of obtaining clear margins during the initial surgery. Bracketing wire placement during preoperative needle localization may help achieve complete excision of calcifications in women with percutaneously proven carcinoma but does not ensure clear margins particularly for larger lesions [11]. Finally, appropriate patient selection for sentinel node biopsy (reserving the procedure for women with a low prior probability of axillary metastases) and use of intraoperative histologic analysis of sentinel nodes may diminish the need for a second surgery to evaluate the axilla [16].
In conclusion, we found that one operation was performed in 80.9% of women with percutaneously proven nonpalpable breast cancer. The American Cancer Society estimates that breast cancer will be diagnosed in 200,000 women this year [36], at least 60,000 (30%) of whom will have nonpalpable lesions identified on screening mammography. Percutaneous imaging-guided core biopsy may enable most of these women to undergo a single operative procedure for definitive surgical treatment. Our data may assist in preoperative counseling of women with cancers diagnosed at percutaneous biopsy. Optimization of patient selection for breast conservation, improvements in preoperative assessment of the extent of disease, advances in percutaneous biopsy technology to minimize histologic underestimation, and improved use of sentinel lymph node biopsy may allow more women to achieve therapeutic results in one operation.
Acknowledgments
We thank David C. Perlman for invaluable assistance.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J.-m. Lee, J. B. Kaplan, M. P. Murray, M. Mazur-Grbec, T. Tadic, D. Stimac, and L. Liberman Underestimation of DCIS at MRI-Guided Vacuum-Assisted Breast Biopsy Am. J. Roentgenol., August 1, 2007; 189(2): 468 - 474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Mesurolle, M. El-Khoury, D. Hori, J.-P. Phancao, S. Kary, E. Kao, and D. Fleiszer Sonography of postexcision specimens of nonpalpable breast lesions: value, limitations, and description of a method. Am. J. Roentgenol., April 1, 2006; 186(4): 1014 - 1024. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Newman and H. M. Kuerer Advances in Breast Conservation Therapy J. Clin. Oncol., March 10, 2005; 23(8): 1685 - 1697. [Full Text] [PDF]
|
|
|
|
 |
|
 J.-H. Yang, W.-S. Lee, S.-W. Kim, S.-U. Woo, J.-H. Kim, and S.-J. Nam Effect of Core-Needle Biopsy vs Fine-needle Aspiration on Pathologic Measurement of Tumor Size in Breast Cancer Arch Surg, February 1, 2005; 140(2): 125 - 128. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
