| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Department of Radiology (D.E.M., B.F.C., S.V.K.), Baystate Comprehensive Breast Center/Regional Cancer Program (R.B.B.), Department of Pathology (R.A.G., M.E.B.), Department of Surgery (J.L.F.), and Division of Hematology and Oncology (G.M.J.), Baystate Medical Center, 759 Chestnut St, Springfield, MA 01199. From the 2001 RSNA scientific assembly. Received April 25, 2002; revision requested June 21; revision received August 16; accepted September 30. Supported by an RSNA Seed Grant and by the Rays of Hope charitable fund. Address correspondence to D.E.M.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Thirty-four women with breast masses underwent US-guided biopsy with an 11-gauge VAB device, with which removal of all evidence of the lesion was attempted. Histologic findings were compared with results of surgery and follow-up imaging. Patient tolerance and perceptions of the procedure and the ability of the procedure to eliminate a palpable finding were evaluated with questionnaires and findings at follow-up physical examination.
RESULTS: The biopsy protocol was completed in all cases. Twenty-six benign lesions (76%) and eight malignancies (24%) were diagnosed. After VAB, 10 patients (29%) underwent surgery on the basis of histologic findings of invasive carcinoma (n = 7), ductal carcinoma in situ (n = 1), lobular neoplasia (n = 1), or atypical ductal hyperplasia (n = 1). VAB resulted in complete excision of four of 10 lesions: two of eight malignancies and two of two benign lesions. Among 21 patients with benign lesions who underwent 6-month follow-up imaging, eight (38%) had a definite residual mass. At 6-month follow-up examination, VAB was seen to have eliminated the palpable abnormality in seven (88%) of eight patients with initially palpable benign masses. Thirty-two patients (94%) described no or mild pain during biopsy, and 33 patients (97%) rated care as excellent.
CONCLUSION: After removal of all US evidence of breast masses with a VAB device, there was a substantial probability that residual lesion that was not visualized during the procedure would later be found at surgery or follow-up imaging. A palpable mass (
1.2 cm in mean diameter) was eliminated in 88% of cases, and patient tolerance and perceptions of the procedure were favorable.
© RSNA, 2003
Index terms: Breast, biopsy, 00.1261, 00.1262 • Breast neoplasms, localization, 00.12985 • Breast neoplasms, therapy, 00.1262 • Breast neoplasms, US, 00.1298, 00.12985 • Ultrasound (US), guidance, 00.12985
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
During this prospective study, 348 patients were referred for possible US-guided biopsy, and 346 were screened by our research associate (R.B.B.) for possible enrollment. The research associate, who consulted with a participating radiologist (D.E.M., B.F.C., or S.V.K.) when necessary, reviewed US images obtained in potential participants and performed measurements (length, width, height) of the lesion and of its distance from the skin surface and chest wall to determine patient eligibility for the study.
Two patients were not screened because they were scheduled for and underwent biopsy before the research associate could review the cases and contact the patients about possible participation. Fifty-seven screened patients (16%) met the study requirements and were offered inclusion; 38 (67%) of the eligible patients chose to participate. Four patients who agreed to participate were excluded because their lesions represented complex cysts that were aspirated; this resulted in a final study population of 34 women.
Descriptions of the shape and margin characteristics of masses on the basis of their combined mammographic and sonographic features were recorded before biopsy by one of three radiologist authors (D.E.M., B.F.C., or S.V.K.), who used mammography terminology previously described by the American College of Radiology (7). The mean age of participating patients was 50 years (range, 27–84 years). The institutional review board of Baystate Medical Center reviewed and approved this study, and participating patients read and signed a separate informed consent form.
Biopsy Procedure and Review of Histologic Findings
The biopsy protocol used in this study involves removal of a substantially larger tissue volume than is our usual practice. Biopsy procedures were performed by one of the three participating radiologists (D.E.M., B.F.C., or S.V.K.) with an 11-gauge biopsy device (Ethicon Endo-Surgery, Cincinnati, Ohio; U.S. Food and Drug Administration approval for use in diagnosis, 1999). This handheld vacuum-assisted device is used for freehand biopsy with US guidance. With the device, one can apply suction through the biopsy probe to draw tissue into a sample notch, separate a sample with a rotating cutter, and transport the specimen to a collection chamber for removal. Directional sampling is achieved by manually rotating the probe at the biopsy site. Identical US units (Ultramark 9 HDI; Philips Medical Systems, Bothell, Wash) equipped with 5–10-MHz transducers were used to guide the procedures.
US was used to localize the mass and to enable measurement of the distance from the skin surface to the anterior margin of the lesion and the distance from the deep margin of the lesion to the anterior pectoralis margin. Anesthetization of the skin and deeper tissues leading to and around the lesion was achieved with administration of a mean of 8 mL of buffered 1% lidocaine and a mean of 9 mL of 1% lidocaine and epinephrine in a 1:100,000 mixture, respectively. With US guidance, the probe was positioned deep to the lesion through a small skin incision. A distance of at least 0.2 cm was documented at US between the open sampling notch and the deep margin of the lesion to allow for inclusion of the deep margin during biopsy. Samples were obtained between the 8-o’clock and 4-o’clock locations, rotating through the 12-o’clock position. Sampling of tissue deep to the probe (from the 5-o’clock to the 7-o’clock positions), the process of which would have been obscured by acoustic shadowing from the probe and required that the sampling notch be directed toward the chest wall, was thereby avoided. An attempt was made to remove all evidence of the target lesion with continuous US guidance.
After biopsy, US was performed in multiple planes, with and without application of suction to collapse the biopsy cavity, so that we could assess for the presence of residual mass. Any visible remaining lesion was removed at that time. A clear-probe mode on the VAB unit permits the application of pulsed suction through the biopsy probe for the removal of residual tissue fragments. This technique was used, and any removed tissue was placed in formalin with the lesion samples. With the same biopsy probe, a second series of core samples was then obtained from the margins of the biopsy cavity. These samples were removed to enable determination of whether margin sampling is an accurate means of predicting the presence or absence of residual target lesion. To accomplish this, the biopsy probe was rotated within the cavity; the directional capability of the device enabled sampling of tissue from different portions of the cavity wall. Samples were obtained from between the 8-o’clock and 4-o’clock locations, rotating through the 12-o’clock position. The cavity was again suctioned by using the clear-probe mode, and any retrieved fragments were included with the margin samples. The procedure time (from injection of local anesthetic to removal of the biopsy probe) and number of biopsy passes were recorded by an assisting technologist.
After biopsy, a metallic tissue marker (MicroMark Clip; Ethicon Endo-Surgery) was deployed through the biopsy probe, the probe was removed, and hemostasis was achieved by means of manual compression. The incision was closed with adhesive strips, and a sterile pressure dressing was applied. Patients then underwent unilateral mammography, which is routine in our practice after clip placement. Specimen radiography of biopsy samples was performed when the target lesion contained calcifications. Patients were instructed to apply an ice pack for the remainder of the day, to avoid strenuous activities, and to take acetaminophen if they experienced discomfort.
Histologic slides of tissue taken from the lesion and from the margins were interpreted by one of two experienced breast pathologists (R.A.G., M.E.B.). These pathologists also reviewed histologic material from all patients who underwent subsequent surgical procedures and recorded the size and distribution of any residual target lesion and measured its distance from the edge of the VAB cavity.
Data Analysis
The VAB procedure was considered diagnostically accurate when (a) a specific histologic diagnosis was established that was concordant with the imaging findings, (b) there was no substantial change in diagnosis (ie, no benign or atypical histologic findings at VAB but malignancy diagnosed at excision, and no in situ carcinoma diagnosed at VAB but invasive carcinoma revealed at excision) of lesions that were subsequently excised, and (c) there was no evidence of misdiagnosis at the time of follow-up imaging.
Masses were considered initially palpable if they could be felt by the referring provider, patient, or radiologist before biopsy and were considered impalpable at 6-month follow-up if neither the patient nor the radiologist could palpate the mass. Lesions were considered completely excised at VAB only in patients who underwent subsequent surgical removal of tissue from the biopsy site with no evidence of residual target lesion in the surgical specimen; lesions were considered not completely excised if residual lesion was present in the surgical specimen or at follow-up imaging. The accuracy of margin sampling as a predictor of the presence or absence of residual lesion was evaluated on the basis of comparison with histologic material obtained during subsequent surgery in 10 patients.
Patient Tolerance, Perceptions, Cosmetic Outcome, and Follow-up Imaging
Assessment of patient tolerance and perceptions of the procedure was based on patients’ ability to tolerate completion of the study protocol and on their responses to self-administered questionnaires, which were distributed by the research associate while the radiologist was not present. The first questionnaire was completed after the procedure on the day of biopsy. Patients were asked to rate pain experienced during biopsy (as none, mild, moderate, or severe) and the quality of care received (as excellent, good, fair, or poor). During a follow-up appointment 2–5 days after biopsy, all patients completed a second questionnaire and responded to questions about whether they had taken medication for pain or experienced any bleeding (none, mild, moderate, or severe) and whether the procedure had interfered with their usual daily life (not at all, a little, somewhat, or a lot). One of the three radiologist authors examined the patient. Any palpable abnormality at the biopsy site was measured, and any ecchymosis or change in the breast contour was noted.
Patients who did not undergo surgery and who returned for follow-up approximately 6 months after biopsy were administered a third questionnaire to determine their level of satisfaction (very satisfied, satisfied, or not at all satisfied) regarding how the biopsy site appeared and felt, whether they would recommend the procedure to someone in a similar situation, and whether they could feel any abnormality at the biopsy site. Patients underwent physical examination of the biopsy site performed by one of two radiologist authors (D.E.M., S.V.K.) (17 of 21 patients) or by one of two different radiologists (four of 21 patients; two patients each). The examining radiologist inspected the biopsy site for visible scar, which was measured, and for any skin retraction or other contour abnormality. Physical examination of the area was performed to assess for palpable abnormalities.
Patients who returned at 6 months also underwent imaging to evaluate the area of biopsy. All patients underwent unilateral mammography except one—a 28-year-old woman who had undergone VAB of a palpable mass that was mammographically occult—and all patients underwent US examination of the biopsy region. These studies were interpreted by the same radiologist who performed the physical examinations. The radiologist scanned the area of biopsy in each case and compared the mammographic and real-time US findings with the appearance and location of the mass on mammograms and US images obtained before biopsy.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Diagnostic Accuracy
For each lesion, a specific histologic diagnosis concordant with the imaging findings was established. Twenty-four patients had benign lesions (Table 1); 21 of these patients underwent 6-month follow-up imaging. Surgery was performed in 10 patients for eight malignancies and two atypical lesions that were diagnosed at VAB (Table 2). Among those patients who underwent subsequent surgical excision, there were no substantial changes in the histologic diagnosis established at VAB (Table 2). One patient with fibroadenoma and atypical ductal hyperplasia at VAB had no malignancy or atypical ductal hyperplasia at excision. One patient with pseudoangiomatous stromal hyperplasia and lobular neoplasia at VAB had no DCIS or invasive carcinoma at excision. One patient with DCIS at VAB had no invasive carcinoma at excision. One patient with invasive carcinoma only at VAB had a focus of DCIS at excision, in addition to invasive carcinoma. The patient with invasive carcinoma with ductal and lobular features and DCIS at VAB had only residual invasive tumor at mastectomy. Six-month follow-up imaging was performed in 21 patients with benign lesions and revealed no findings that indicated potential misdiagnosis and no indications for further intervention.
|
|
Among 21 patients with benign lesions who returned for 6-month follow-up, eight (38%) had lesions that were not fully excised at VAB and definite residual mass at US (Fig 1), eight (38%) had no evidence of residual lesion at follow-up imaging, and five (24%) had equivocal findings at follow-up imaging (ie, it was uncertain if residual target lesion—versus postbiopsy changes or a different lesion—was present). For the 10 patients who underwent surgery after the VAB procedure, the histologic presence or absence of target lesion in the VAB margin samples correctly predicted the presence or absence of residual target lesion at surgery in four of eight malignancies and two of two benign lesions (Fig 2).
|
|
|
|
|
|
All 34 patients returned 2–5 days after their procedure to undergo follow-up physical examination and to complete a second questionnaire. Twenty patients (59%) took acetaminophen for discomfort and 14 (41%) took no analgesics. No patients reported bleeding. When asked how much the procedure interfered with their usual daily activities, 16 patients (47%) indicated "not at all;" 14 (41%), "a little;" four (12%), "somewhat;" and none, "a lot." At examination, 24 patients (71%) had ecchymosis, nine (26%) had no visible abnormality aside from the skin incision, and one (3%) had slight skin convexity without ecchymosis at the biopsy site. Twenty-five patients (74%) had no palpable findings in the biopsy region, and nine (26%) had focal palpable hematomas or edematous tissue measuring 2.8 cm in median (mean, 4.0 cm) diameter. These values include outlying data from a patient who developed a large hematoma (discussed below).
Twenty-four patients with benign lesions did not undergo surgery, and 21 (88%) of these 24 patients underwent follow-up imaging and physical examination and completed a third questionnaire approximately 6 months after VAB. Nineteen (90%) of these 21 patients were very satisfied with the appearance of the biopsy area, two (10%) were satisfied, and none was dissatisfied; 16 patients (76%) were very satisfied with how the biopsy area felt, five (24%) were satisfied, and none was dissatisfied. All 21 patients stated that they would recommend this biopsy technique to someone in a similar situation. At clinical examination, four patients (19%) had no visible scarring, and 17 (81%) had minimal scarring (scar length range, 2–9 mm). No patients had skin retraction, concavity, convexity, or other change in breast contour. Among the 13 patients whose lesions were initially impalpable and who underwent 6-month follow-up examination, none described a new palpable finding at the biopsy site. In one of these patients, the examining radiologist noted palpable scar tissue in the region of biopsy. Mammography (n = 20) and US (n = 21) showed no evidence of chronic fluid collection or other complications in any of the 21 patients who underwent follow-up imaging and physical examination.
Ability to Eliminate a Palpable Finding
The lesion was initially palpable in 10 (29%) of 34 patients, including two patients with malignant and eight with benign pathologic findings. All eight patients with benign lesions returned at approximately 6 months, and there was no residual palpable mass felt by the patient or radiologist in seven (88%) of these patients (Fig 3). One patient, who had no residual palpable mass, did describe a change in the texture of the breast at the biopsy site.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
However, the removal of all imaging evidence of a lesion by using available percutaneous biopsy methods with US or stereotactic guidance has not been shown to consistently indicate complete histologic excision. Parker et al (4) reported removal of all imaging evidence of 88% of lesions during 11-gauge US-guided biopsy, with one (5.5%) of 18 infiltrating carcinomas revealed at subsequent surgery to have been completely excised. Perez-Fuentes et al (5) removed all US evidence of 88.6% of lesions, including 10 (71%) of 14 cancers. In their study, subsequent surgical findings confirmed histologic excision of five (50%) of 10 cancers in which no residual abnormality was revealed at US after VAB. Liberman et al found that removal of all mammographic evidence of malignant lesions during stereotactic biopsy resulted in complete histologic excision of eight (35%) of 23 infiltrating carcinomas (14) and four (27%) of 15 infiltrating and in situ carcinomas (15). Use of a large-bore cannula for biopsy with stereotactic guidance has also been found to be unreliable as an excisional technique, with 23 (85.2%) of 27 malignancies in one study (16) having positive margins.
In our series, all US evidence of the target lesions, as visualized on the day of biopsy, was removed in all 34 study patients. However, incomplete excision (revealed at histologic examination of the final surgical specimen) was proved in six (60%) of 10 patients who underwent surgery, and a smaller residual mass was visualized in eight (38%) of 21 patients with benign lesions at 6-month follow-up imaging. It is therefore apparent that the complete excision of some masses was hindered by the loss of visualization of the remaining portion of the lesion during and immediately after VAB owing to the presence of local anesthetic, edema, blood, and/or air, which made the true extent of the lesion difficult to determine. Similar experiences during US-guided VAB have been described (6,17), and we believe that obscuration of the residual portion of lesions was a major obstacle to our ability to more consistently excise entire masses.
In five of the six patients in this study with malignancies that were incompletely excised at VAB, the remaining tumor extended 0.1–1.2 cm (mean, 0.6 cm) from the cavity edge. The thickness range of the residual tumors in our study is similar to thickness ranges of tumors that have been successfully necrosed in early studies of radio-frequency and laser ablation of breast cancers (18,19). The inability of mammography and US to reveal the true extent of a lesion was a limiting factor in at least one patient in our series; this patient had infiltrating lobular carcinoma, which was visualized as a 0.4-cm mass at mammography and US but was revealed at surgery to be multifocal, extending approximately 4.0 cm from the edge of the VAB cavity. Magnetic resonance (MR) imaging can help define the extent of breast carcinoma more accurately than can mammography and US in some patients (20,21), and its potential applications in breast cancer treatment have been described (22). If percutaneous treatment methods for breast cancer are to be developed, tumor ablation and MR imaging before and after treatment would be potential areas for future study.
Results of our technique of sampling the cavity margins after VAB accurately reflected completeness of excision in six of 10 patients. The inaccurate results may be due to noncontiguous sampling, presence of lesion proximal or distal to the sampling notch, protrusions of tumor into the tissues adjacent to the biopsy cavity, presence of discrete islands of tumor distant from the target lesion, or to a combination of these factors.
In our small series, we found that a palpable finding was removed in seven of eight patients and that VAB offers an option for percutaneous diagnosis with potential elimination of a palpable mass, which may be a desirable outcome for some women (23).
Two patients (6%) developed complications from VAB in this study. This rate is in the range of 1.1%–10% that has been reported by other investigators for 11-gauge US-guided VAB (4–6,24) and is higher than complication rates of less than 1% that have been reported for US-guided breast biopsy with a 14-gauge automated device (25,26) and complication rates of less than 2% that have been reported for stereotactic VAB (27,28). Reports of complications from breast biopsy performed with a large-bore cannula device have ranged from 0% to 20.8% (12,29,30).
In summary, US-guided VAB with removal of all evidence of breast lesions is technically feasible and diagnostically accurate but does not reliably result in the excision of benign or malignant masses. The procedure eliminates a small palpable finding in the large majority of cases. Although two complications occurred in this small series, overall patient tolerance and perceptions of the procedure were favorable.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, D.E.M., R.B.B.; study concepts and design, D.E.M., B.F.C.; literature research, R.B.B., D.E.M.; clinical studies, D.E.M., B.F.C., S.V.K.; data acquisition, R.B.B., R.A.G., M.E.B.; data analysis/interpretation, D.E.M.; manuscript preparation, D.E.M., R.B.B.; manuscript definition of intellectual content, D.E.M.; manuscript editing and revision/review, all authors; manuscript final version approval, D.E.M.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Y. Ko, Y.-A Bae, M.-J. Kim, K. S. Lee, Y. Lee, and L. S. Kim Factors Affecting the Efficacy of Ultrasound-Guided Vacuum-Assisted Percutaneous Excision for Removal of Benign Breast Lesions J. Ultrasound Med., January 1, 2008; 27(1): 65 - 73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Littrup, L. Freeman-Gibb, A. Andea, M. White, K. C. Amerikia, D. Bouwman, T. Harb, and W. Sakr Cryotherapy for Breast Fibroadenomas Radiology, January 1, 2005; 234(1): 63 - 72. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
