| 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
Breast Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center,
New York, NY 10021.
Received January 30, 2001;
accepted after revision April 3, 2001.
Supported by a grant from the New York State Department of Health
(C015709).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Retrospective review revealed 200 consecutive nonpalpable breast cancers diagnosed by percutaneous imaging-guided large-core biopsy and treated with surgery that included sentinel lymphadenectomy. Percutaneous breast biopsy was performed with stereotactic or sonographic guidance with an automated core needle or vacuum-assisted biopsy probe. Sentinel lymphadenectomy was performed with intradermal injection of a radioisotope and intraparenchymal injection of blue contrast agent. Technical success was defined as identification of sentinel nodes at surgery. Medical records were reviewed.
RESULTS. Technical success rate was 200 (100%) of 200. In 158 (79%) of 200 cancers, sentinel nodes were tumor-free, and axillary dissection was avoided. In three (2%) of 200 carcinomas, the sentinel nodes were negative for tumor, but nonsentinel nodes suspicious on intraoperative palpation were excised and found by frozen section analysis to contain tumor. Tumor was found in sentinel nodes in 39 (20%) of 200 carcinomas; axillary dissection, performed in 31 of these 39 women, revealed additional tumor in nonsentinel nodes in seven (23%). A single surgical procedure was performed for 164 (82%) of 200 carcinomas; the breast was preserved in 191 (96%) of these 200 carcinomas.
CONCLUSION. Percutaneous imaging-guided large-core breast biopsy and sentinel lymphadenectomy provide a minimally invasive approach to the diagnosis and treatment of women with nonpalpable invasive breast cancers.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent advances allow us to reassess the traditional two-stage surgical approach. Percutaneous large-core biopsy with stereotactic or sonographic guidance is an alternative to surgical biopsy for the diagnosis of nonpalpable breast lesions [1,2,3,4,5]. Sentinel lymphadenectomy with a radioisotope [6], blue contrast agent [7], or a combination of methods [8] can be used to determine the histologic status of the axilla. Although a previous small series suggested that the combination of percutaneous breast biopsy and sentinel lymphadenectomy was potentially useful in the diagnosis and treatment of nonpalpable breast cancers [9], some investigators have excluded women with nonpalpable breast cancer from sentinel lymphadenectomy [10,11,12]. Our study was undertaken to evaluate the usefulness of percutaneous imaging-guided large-core breast biopsy and sentinel lymphadenectomy in women with nonpalpable breast cancer.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All women had mammographically detected nonpalpable carcinomas and findings negative for tumor in axillae on preoperative physical examination. Percutaneous biopsy was performed with sonographic (n = 108) or stereotactic (n = 92) guidance, with a 14-gauge automated needle (n = 136), vacuum-assisted biopsy probe (n = 63), or large biopsy cannula (n = 1). The median time from percutaneous biopsy to surgery was 24 days (range, 5-172 days).
Sentinel lymph nodes were identified with a combination of intradermal radioisotope and intraparenchymal blue contrast agent. Radioisotope injection was accomplished with the following technique: for women who underwent lumpectomy and sentinel lymph node biopsy, preoperative needle localization was performed with a modified Kopans wire (Cook, Bloomington, IN). After needle localization, 0.1 mCi (3.7 MBq) of unfiltered technetium-99m sulfur colloid in 0.05 mL of saline solution was injected intradermally via a 28-gauge needle superolateral to the site of the primary tumor in the breast (Fig. 1). For women who underwent mastectomy and sentinel lymph node biopsy, we injected the radioisotope intradermally superolateral to the primary tumor site before surgery, using the same technique, without preoperative needle localization.
|
We performed preoperative lymphoscintigraphy, using a large field-of-view dual-head gamma camera with a low-energy high-resolution collimator. Transmission images in the anterior and lateral projections were obtained with a cobalt-57 flood source. Lymphoscintigraphic imaging began approximately 20 min after injection (Fig. 2). The time between injection and surgery ranged from 2 to 4 hr.
|
The handheld C-Trak gamma probe (Care-Wise Medical, Morgan, CA) was used intraoperatively to obtain counts of room background, the breast injection site, and the "hottest" spot in the axilla. Four milliliters of isosulfan blue contrast agent (Lymphazurin; Zenith Parenterals, Rosemont, IL) was injected in one to three aliquots around the tumor site into the parenchyma of the breast. Within 5-10 min after administration of blue dye contrast agent, an axillary incision was made, and sentinel node biopsy was performed, guided by the gamma probe and by the appearance of blue contrast agent in lymphatic channels and nodes (Fig. 3). In situ and ex vivo node counts were taken, and axillary background counts were taken after removal of the node. Technical success was defined as identification of the sentinel nodes at surgery. A node was considered to be a sentinel node if it exhibited blue staining at surgery, its removal resulted in a fourfold or greater reduction in axillary counts, or both. If sentinel nodes were not found, axillary dissection was performed.
|
All sentinel nodes were excised and analyzed by frozen and paraffin sections. If frozen section and the initial H and E-stained paraffin section of the sentinel node were negative for tumor, immunohistochemical stains and deeper sections were obtained as per the following protocol: two immediately adjacent sections were evaluated with cytokeratin immunohistochemistry with AE1/3 and CAM 5.2 antibodies, and three deeper levels 50 µm apart were evaluated by H and E stain. The protocol was to perform axillary lymph node dissection if frozen section, paraffin section, or immunohistochemical analysis of the sentinel nodes yielded carcinoma. Nonsentinel axillary lymph nodes were examined with a single H and E section, according to the standard procedure used for axillary dissections at our institution. If the sentinel nodes were free of tumor by all methods and the axilla was clinically negative for tumor on intraoperative examination, no further axillary surgery was performed.
Medical records and histologic findings were reviewed to determine the technical success rate of sentinel lymph node biopsy and to determine the frequency of performing a single surgical procedure, sparing axillary lymph node dissection and preserving the breast. Exact 95% confidence intervals (CIs) were calculated around relevant frequencies with statistical software (version 8.0; SAS Institute, Cary, NC). Results were entered on a computerized spreadsheet (Excel 97; Microsoft, Redmond, WA) for analysis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Sentinel nodes were found at surgery in 200 of 200 carcinomas, indicating a technical success rate of 100% (95% CI, 99-100%). The median number of sentinel nodes was 2.0, and the mean was 2.0 (range, 1-9). In 158 (79%; 95% CI, 73-84%) of 200 carcinomas, the sentinel nodes were free of tumor, and axillary lymph node dissection was averted. In three (2%) of 200 carcinomas, the sentinel lymph nodes were negative for tumor, but nonsentinel nodes, suspicious on intraoperative palpation, were excised and found to contain tumor by frozen section analysis. Axillary lymph node dissection revealed additional axillary metastases in all three of these women.
Tumor was present in sentinel nodes in 39 (20%; 95% CI, 14-25%) of 200 carcinomas. Tumor was found in sentinel nodes by frozen section analysis in 22 (56%) of these 39 patients, by paraffin analysis in three (8%), and by immunohistochemical analysis after frozen and paraffin sections were negative for tumor in 14 (36%). Axillary lymph node dissection, performed in 31 of 39 women with sentinel node metastases, revealed additional tumor in nonsentinel nodes in seven (23%), including four (19%) of 21 women with sentinel node metastases found by frozen section, one (33%) of three women with sentinel node metastases found by paraffin section, and two (29%) of seven women with sentinel node metastases found by immunohistochemistry.
In the 200 nonpalpable breast cancers in this study, axillary dissection was spared in 158 (79%; 95% CI, 73-84%) with no tumor in sentinel nodes. Axillary dissection was performed for 34 carcinomas (17%) on the day of the first surgery in 24 (because of sentinel node metastases identified on frozen section in 22 and suspicious nonsentinel node on intraoperative palpation that contained tumor at frozen section in two) and at a later date in 10 (because of tumor found on immunohistochemical analysis of sentinel nodes in six, tumor found at paraffin analysis of sentinel nodes in three, and tumor found at frozen and paraffin analysis of a suspicious nonsentinel node found at intraoperative palpation in one). Axillary dissection was not performed because of patient or physician choice for eight carcinomas with tumor in sentinel nodes, including seven with sentinel node metastases found by immunohistochemistry and one with sentinel node metastases found at frozen and paraffin sections.
A single surgical procedure was performed for 164 (82%; 95% CI, 76-87%) of 200 carcinomas. The median number of surgeries performed was one, and the mean was 1.2 (range, 1-4). The initial breast surgery performed at the time of sentinel lymphadenectomy was wide excision in 197 women (99%) and mastectomy in three women (2%). Two or more surgical procedures were performed for 36 carcinomas; subsequent surgeries included reexcision (n = 22), mastectomy (n = 4), axillary dissection (n = 6), reexcision and axillary dissection (n = 2), mastectomy and axillary dissection (n = 1), and two reexcisions followed by mastectomy and axillary dissection (n = 1). Among these 200 carcinomas, the breast was preserved in 191 (96%; 95% CI, 92-98%); of these, 167 (87%; 95% CI, 82-92%) achieved clear histologic margins of resection during the first operation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Percutaneous imaging-guided large-core breast biopsy was pioneered by Parker et al. [1] in 1990. Early validation studies showed 87-100% concordance between the results of 14-gauge automated percutaneous core biopsy and surgery, with the best results achieved by obtaining multiple specimens with a long-excursion gun [1,2,3,4,5]. Percutaneous biopsy is less invasive and less expensive than surgery, can be performed quickly, and does not deform the breast [17]. For women with benign breast lesions, who account for 70-80% of all women undergoing breast biopsy, percutaneous biopsy can obviate surgery. Percutaneous biopsy is also advantageous for women with breast cancer. Women whose cancers are diagnosed by percutaneous rather than surgical biopsy undergo fewer surgical procedures and are more likely to obtain clear histologic margins of resection at the first operation [18,19,20,21].
Sentinel lymphadenectomy was first applied to breast cancer by Krag et al. [6] using a radioisotope in 1993 and by Giuliano et al. [7] using blue contrast agent in 1994. The sentinel node hypothesis states that the sentinel nodes are the first nodes draining a tumor and that the histologic status of the sentinel nodes predicts the status of the regional lymphatic basin. Studies correlating the results of sentinel lymphadenectomy with axillary lymph node dissection in more than 3000 women with breast cancer have shown a technical success rate of 88%, sensitivity of 93%, and accuracy of 97% [22]. On the basis of these data, some medical centers have begun to offer sentinel lymphadenectomy as an alternative to axillary dissection for women with small infiltrating breast carcinomas and lymph nodes clinically negative for tumor. Axillary dissection is performed if sentinel nodes are not found, if tumor is present in sentinel nodes, or if axillary nodes are suspicious on palpation.
In our study, we found the combination of percutaneous breast biopsy and sentinel lymphadenectomy provided a minimally invasive approach to the diagnosis and treatment of nonpalpable invasive breast cancer. Ninety-six percent of the women in our study had breast-conserving surgery. Eight-two percent of these women had a single surgical procedure. The high frequency of achieving clear histologic margins at the first surgery likely reflects the fact that the surgeon, knowing that the diagnosis was cancer from the percutaneous biopsy results, intentionally excised a larger volume of tissue to resect the cancer with clear histologic margins [18, 20, 21]. For 158 (79%) of the 200 carcinomas analyzed in our study, sentinel nodes were free of tumor, and axillary dissection was averted.
The high technical success of sentinel lymphadenectomy in our study likely reflects several factors. First, we used a combination of radioisotope and blue contrast agent, an approach shown to be successful more often than either method individually [11]. Second, the radioisotope was injected intradermally. Linehan et al. [23] showed a significantly higher technical success rate for sentinel lymph node biopsy using intradermal rather than intraparenchymal radioisotope injection, perhaps because the higher interstitial pressure that can be obtained with the intradermal method favors lymphatic uptake. In addition to having a high technical success rate, intradermal radioisotope injection has high accuracy. In a validation study, Boolbol et al. [24] reported an accuracy of 100% in 44 women who had sentinel lymphadenectomy after intradermal injection of radioisotope and intraparenchymal blue contrast agent. Other researchers have shown that intraparenchymal and intradermal injections resulted in labeling the same sentinel nodes in 95-100% of patients [10].
A third reason accounting for our high technical success rate may be the use of percutaneous biopsy rather than surgical biopsy for diagnosis. Some investigators have reported higher technical success rates for sentinel lymphadenectomy in women who did not have prior surgical biopsy, perhaps due to disruption of normal lymphatic drainage pathways at surgical biopsy [16]. Finally, our previous experience with sentinel lymph node biopsy likely contributed to our high success rate. In the patients in this study, percutaneous biopsy was performed after a preliminary validation study of 60 patients with backup axillary dissection and after additional clinical experience with more than 500 patients. The literature indicates that sentinel lymph node biopsy has a learning curve, with higher technical success and lower false-negative rates achieved with experience [25].
Our study does not address the false-negative rate of either percutaneous biopsy or sentinel lymphadenectomy. Other investigators have reported that an average of 2.8% (range, 0.3-8.2%) of cancers were missed at percutaneous biopsy, comparable to the average cancer miss rate of 2.0% (range, 0-8%) reported at needle localization and diagnostic surgical biopsy [26, 27]. The false-negative rate of percutaneous breast biopsy can be reduced by use of appropriate tissue-acquisition devices such as directional vacuum-assisted biopsy for calcifications [28, 29] and by meticulous attention to technique, particularly with respect to lesion targeting [30]. Furthermore, the impact of a false-negative percutaneous diagnosis can be minimized by careful correlation of imaging and histologic findings after biopsy; imaging-histologic discordance may allow prospective identification of a missed cancer, avoiding deleterious delay in diagnosis [31].
Studies of sentinel lymphadenectomy have reported an average false-negative rate of 7% (range, 0-22%), but there are several factors that can reduce this. First, the false-negative rate of sentinel lymph node biopsy decreases with experience. Cody et al. [25] reported that among 104 patients who had sentinel node biopsy with planned "backup" axillary dissection, the false-negative rate was 10.6%; eliminating the first six or the first 15 patients of each surgeon reduced the false-negative rate to 5.2% and 2%, respectively. Second, the false-negative rate of sentinel lymphadenectomy is lower if the combination of radioisotope and blue contrast agent is used rather than either method individually [32]. Third, intraoperative palpation is important: a node replaced by tumor may not take up contrast agent or radioisotope, as illustrated in three (2%) of the 200 carcinomas in our study and in four (3%) of 130 cases reported by Borgstein et al. [10]. If nodes are of clinical concern, they should be excised. Finally, immunohistochemical analysis of sentinel nodes significantly reduces the false-negative rate from 8% to 3% [33].
In 1998, McMasters et al. [34] suggested that sentinel lymphadenectomy was not yet the standard of care. A paucity of long-term follow-up data is an inherent limitation of all new technologies, and that of sentinel lymphadenectomy is no exception. However, validation studies in more than 3000 women suggest that sentinel lymphadenectomy is a viable alternative to axillary dissection for women with small invasive breast cancers and clinically negative nodes. It can be estimated that of approximately 200,000 women who will be diagnosed with invasive breast cancer in the year 2001 [35], at least 50,000 (25%) will have nonpalpable lesions identified at screening mammography. Percutaneous biopsy combined with sentinel lymphadenectomy can provide a minimally invasive approach to the diagnosis and treatment of nonpalpable breast cancer in such women. Continued refinement of these techniques may allow us to transcend the traditional two-stage surgical approach and set a new minimally invasive standard.
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:
|
|
 |
|
  L. Liberman Lymphoscintigraphy for Lymphatic Mapping in Breast Carcinoma Radiology, August 1, 2003; 228(2): 313 - 315. [Full Text] [PDF]
|
|
|
|
|
|
P. J. Tanis, R. A. Valdes Olmos, S. H. Muller, and O. E. Nieweg Lymphatic Mapping in Patients with Breast Carcinoma: Reproducibility of Lymphoscintigraphic Results Radiology, August 1, 2003; 228(2): 546 - 551. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. March, B. F. Coughlin, R. B. Barham, R. A. Goulart, S. V. Klein, M. E. Bur, J. L. Frank, and G. Makari-Judson Breast Masses: Removal of All US Evidence during Biopsy by Using a Handheld Vacuum-assisted Device--Initial Experience Radiology, May 1, 2003; 227(2): 549 - 555. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Bonnema and C. J. H. van de Velde Sentinel lymph node biopsy in breast cancer Ann. Onc., October 1, 2002; 13(10): 1531 - 1537. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
