| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Breast Health Center, Department of Radiology, California Pacific Medical
Center, 3698 California St., 2/F, San Francisco, CA 94118.
2
Department of Radiology, Brigham and Women's Hospital, Harvard Medical School,
75 Francis St., Boston, MA 02115.
Received December 5, 2000;
accepted after revision March 12, 2001.
Address correspondence to F. R. Margolin.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We reviewed the prospectively collected results of our imaging—guided core biopsy program during its first 5 years (1994-1998). A total of 1333 lesions (94% of which were Breast Imaging Reporting and Data System (BI-RADS) assessment category 4) were sampled in 1183 patients. Patients with BI-RADS assessment category 5 lesions were referred to surgeons. Stereotactic guidance was used for the core biopsy of 506 lesions, and sonography was used to guide the predominantly 16-gauge needle core biopsy of 827 solid masses.
RESULTS. One hundred forty-seven cancers were diagnosed in 1333 biopsies, resulting in a positive yield of 11%. Of 1020 patients with benign, concordant core biopsy results, 981 (96%) had at least one follow-up imaging examination within 36 months of the biopsy. Nineteen (2%) of these 1020 patients had a suspicious change at follow-up; 18 of these patients underwent surgical excision with benign findings. No cancers were found at imaging follow-up or by tumor registry linkage. All malignant core biopsy results were confirmed as malignant at surgical excision (positive predictive value 100%). Twenty-two patients with atypical ductal hyperplasia at core biopsy had subsequent surgery, and 12 (55%) of them were found to have cancer at surgery.
CONCLUSION. An imaging-guided core biopsy program, developed and implemented by a small group of radiologists in a community hospital, can achieve successful results and provide an important service to patients and a cost-effective alternative to surgical biopsy. Our program emphasized sonographic guidance and achieved high follow-up compliance.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In 439 of the 506 stereotactic core biopsies, a 14-gauge biopsy needle and an automated gun with a 22-mm excursion (Pro-Mag; Manan Medical Products, Northbrook, IL) were used. A directional vacuum-assisted biopsy device (Mammotome; Ethicon Endo-Surgery, Cincinnati, OH) was acquired in August 1997 and was used in the remaining 67 stereotactic core biopsy cases (14-gauge needle in 45 patients and 11-gauge needle in 22 patients). Stereotactic core samples from calcified lesions were radiographed at x1.5 magnification in a dedicated specimen X-ray unit (Faxitron; Faxitron X-Ray, Buffalo Grove, IL). The number of core samples containing calcification was recorded. Samples with and without calcification were separately submitted for pathologic examination, which included polarized light microscopy for identification of calcium oxalate crystals. For masses sampled under stereotactic guidance, the scout digital image was repeated after the biopsy to reveal needle tracts through the target.
Eight hundred twenty-seven solid masses were biopsied using sonographic guidance (5200-S Envision; Acoustic Imaging, Denver, CO) and a free-hand technique. A nonsterile 7.5-MHz linear transducer and povidone-iodine ointment (Betadine; Purdue Frederick, Norwalk, CT) as a coupling agent were used in all patients. Of the 827 sonographically biopsied masses, 758 (92%) were sampled using an automated biopsy gun with a 22-mm throw and a 16-gauge needle (Pro-Mag; Manan Medical Products). Twenty-two masses were sampled with a 14-gauge needle and the remaining 47 masses with an 18-gauge needle. Postfire orthogonal images were obtained after each needle pass to document needle position in two planes.
For core biopsies performed using either stereotactic or sonographic guidance, maximum lesion diameter in millimeters and number of cores obtained were recorded. All procedures were performed by one of three radiologists. The total time spent by the radiologist in the procedure room was recorded for the 596 procedures performed by one radiologist. The specialty of the referring physician was recorded for 554 patients.
Selection of Cases and Method of Imaging Guidance
A dedicated prone stereotactic biopsy table with digital imaging was
acquired in March 1993. During the following 9 months, under an institutional
review board—approved protocol, 22 patients with lesions scheduled for a
needle-localized surgical biopsy underwent a preoperative stereotactic core
biopsy. Core and surgical histology correlated in 21 patients. To secure the
cooperation of our breast surgeons, it was agreed that all patients with
lesions that were highly suggestive of malignancy (Breast Imaging Reporting
and Data System [BI-RADS] [5]
assessment category 5) would be referred for surgical consultation. Probably
benign (BI-RADS assessment category 3) lesions were assigned to short-term
follow-up unless the patient or her referring physician insisted on a
biopsy.
On January 1, 1994, our imaging-guided core biopsy program was implemented in clinical practice, primarily for use in patients with BI-RADS 4 lesions. Table 1 summarizes management recommendations and interventional procedures in our practice from January 1, 1994, through December 31, 1998. During these 5 years, 130,189 screening and 50,629 diagnostic mammographic examinations were performed, and 705 cancers were diagnosed at surgical or imaging-guided core biopsies.
|
Patients with lesions assessed as suspicious (BI-RADS 4) and suitable for imaging-guided core biopsy were offered this option at the time of diagnostic imaging. Patients agreeing to have the biopsy were then scheduled for the procedure. Referring physicians were notified by telephone or fax of the scheduled biopsy.
Stereotaxis was used to biopsy all cases of calcifications and focal asymmetric densities and any mass that was best visualized mammographically. Solid masses presenting with discrete sonographic findings were sampled using sonographic guidance whether or not they were palpable or could be identified on mammograms.
A customized computerized databse was created to store data on each patient. Specific histology was recorded for all core biopsies. The frequency of fibroadenoma diagnoses prompted identifying this entity separately from other benign histologies.
Management and Follow-Up
When atypia or malignancy was identified at core biopsy, the referring
physician was notified, and a coordinated plan for patient notification and
surgical referral was developed. Patients with concordant benign diagnoses
were telephoned by the radiologist with their biopsy results and
recommendations for imaging follow-up. All patients with benign diagnoses were
placed in a 3-year follow-up protocol with the first imaging examination
scheduled at 6 months. Calcifications were followed mammographically. Masses
were reevaluated by either mammography or sonography, depending upon which
technique would provide the most accurate assessment when compared with
prebiopsy images. Patients with lesions exhibiting any suspicious change at
follow-up were referred for surgical excision.
In March 2000, our database for patients with benign, concordant core biopsy results who were assigned to follow-up only was linked with our regional Surveillance, Epidemiology, and End Results (SEER) tumor registry. This linkage provided us with follow-up data ranging from 15 months for patients biopsied at the end of 1998 to 75 months for those biopsied early in 1994. Included were those patients who did not return to our institution for follow-up but not those who may have moved from our geographic area.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Stereotactically Guided Biopsy Results
The results of stereotactically guided core biopsy of 506 lesions in 470
patients are presented in Table
3. Of the 506 lesions, there were 344 (68%) clusters of
microcalcifications and 162 (32%) noncalcified lesions. Among the results of
506 stereotactic core biopsies, 59 cancers were diagnosed, including 18
invasive carcinomas and 41 cases of ductal carcinoma in situ. Hence, the
positive yield of stereotactic core biopsy was 12%. In 35 (85%) of the 41
patients with ductal carcinoma in situ, the disease presented as
calcifications. All lesions sampled using stereotactic guidance were
nonpalpable.
|
In the subgroup of 344 patients with calcifications, mean patient age was 55 years (range, 31-92 years). An average of six core samples was obtained (range, 1-19 samples). Fewer than five samples were obtained only when the procedure was discontinued because of the patient's inability or unwillingness to cooperate or tolerate further sampling. Of 344 targeted calcification clusters, 161 (47%) were 5 mm or less in greatest dimension (range, 3-70 mm).
Calcification was present in at least one core on specimen radiography in 314 (91%) of 344 lesions. In seven patients, calcifications not seen in core radiographs were identified in microscopic sections, five with benign and two with malignant histology. In seven patients with no calcium seen in core samples at either specimen radiography or histology, postbiopsy images revealed either decreased or no calcification at the biopsy site. Twenty-five patients with no calcification visible in core radiographs and with benign histology received follow-up mammograms from 6 to 54 months after biopsy. Twenty-four showed no change. The mammogram of one patient showed increased calcifications at 7 months; surgical excision revealed benign fibrocystic change. No patient whose core histology had been benign and in whom stereotactic biopsy had failed to extract calcifications was diagnosed subsequently with breast cancer in the biopsied quadrant, according to SEER tumor registry data.
One hundred sixty-two noncalcified lesions (136 masses, 23 asymmetric densities, and three areas of architectural distortion) in 149 patients were biopsied using stereotactic guidance. Mean patient age was 52 years (age range, 33-85 years). An average of six core samples (range, 1-9 samples) was obtained from each patient. Mean lesion size in this subgroup was 13.3 mm, and median size was 6 mm.
Sonographically Guided Biopsy Results
Among the 827 sonographically guided biopsies in 713 patients, 88 cancers
were diagnosed, including 80 invasive carcinomas and eight cases of ductal
carcinoma in situ. Hence, the positive yield of sonographically guided core
biopsy was 11%. Palpability was recorded for 698 of 827 solid masses biopsied
with sonographic guidance. Of the 698 masses, 514 (74%) were palpable, whereas
184 (26%) were not. Table 4
shows the histologic results of the 827 sonographically guided core biopsies
as a function of palpability. Median patient age was 40 years (age range,
17-93 years). Mean number of core samples obtained was four (range, 1-8
samples). Mean diameter of palpable masses was 16.3 mm (range, 4-56 mm), and
mean diameter of nonpalpable masses was 12.2 mm (range, 5-35 mm).
|
Complications and Discordance
No serious complication occurred in any patient as a result of
imaging-guided core biopsy. Two patients received oral antibiotics for
clinical findings of cellulitis at the entrance site of a 14-gauge
stereotactic biopsy needle.
Excluding those cancers underestimated in biopsies yielding atypical ductal hyperplasia (ADH), only one of 148 cancers in this series was not initially diagnosed by core biopsy. In this patient, a sonographically guided core biopsy using a 16-gauge needle was performed on a palpable 3-cm solid mass. Histologic examination revealed the samples to be nondiagnostic. This pathologic result was considered discordant with the radiologic finding, and the lesion was surgically excised 2 weeks later. This mass was a 22-mm infiltrating ductal cancer.
Atypical Hyperplasia
ADH was diagnosed in 24 patients, including 18 who underwent stereotactic
core biopsies for calcifications and six who underwent sonographically guided
biopsies of solid masses. Twenty-three patients had prompt surgical excision
of the biopsied area. One patient was lost to follow-up. Pathology reports
were available for review for all but one patient who had surgery in
Europe.
Cancer was found in 12 (55%) of 22 patients in whom a pathologic diagnosis was available, including ductal carcinoma in situ in eight patients and invasive ductal cancer in four. ADH was the final diagnosis in one patient, whereas eight surgical biopsies yielded only benign histology. In one other patient, ADH was described in the epithelial component of a fibroadenoma in both core and surgical specimens.
Atypical lobular hyperplasia was diagnosed in one patient who had stereotactically guided biopsy of calcifications and in two who had sonographically guided biopsies of solid masses. Both masses were surgically excised and were found to be benign. The calcifications were followed mammographically at 3, 9, 12, and 24 months with no suspicious change detected.
Follow-Up of Benign Biopsies
Of 1183 patients in our study, 1020 had benign, concordant biopsy results
and were entered in a follow-up protocol with imaging recommended at 6, 12,
24, and 36 months. Of the 1020 eligible patients, 981 (96%) had at least one
follow-up examination within this 3-year period. In 19 patients (14 with
masses, five with clustered calcifications), follow-up imaging at 6-36 months
after biopsy (mean time since biopsy, 14 months) revealed a suspicious
change—either an increase in size of the mass or in the number of
calcifications. Subsequently, 18 of the 19 lesions were excised, and none of
these 18 lesions was malignant. One of the 19 patients (who was herself a
referring physician) declined surgical excision. In her case, after the
initial increase in calcifications, multiple subsequent mammograms obtained up
to 47 months after the core biopsy showed these calcifications to be
stable.
In 17 (94%) of the 18 lesions excised, surgery confirmed the initial benign core biopsy diagnosis. Eleven patients with an initial sonographically guided core biopsy diagnosis of fibroadenoma had surgical excision when, at follow-up (mean follow-up time, 14 months; range, 6-24 months), the lesion appeared to have enlarged. Ten were found to have benign fibroadenomas at subsequent surgery. In one patient with an enlarging fibroadenoma, a benign phyllodes tumor was surgically removed. The other three enlarging masses were diagnosed as tubular adenoma, papilloma, and nodular adenosis at surgical excision. Increasing calcifications prompted open biopsy for three of four patients with fibrocystic change at core biopsy. All were found to have benign fibrocystic change at surgical excision.
The records of the remaining 962 patients with benign core biopsy results and with stability established by follow-up mammograms were linked with our regional SEER tumor registry data. No cancer was recorded in the area biopsied in any of these patients. The follow-up interval at the time of linkage ranged from 15 to 75 months.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cancer Yield, Lesion Selection, and Referral Sources
In a survey of 458 breast-imaging practices nationally, surgeons were the
most commonly cited referral source
[1]. In contrast, surgeons were
the referring physicians for only 10% of our patients. We had adopted a
protocol under which patients with BI-RADS 5 lesions would be referred for
surgical consultation. Such patients were seldom then referred for
imaging-guided core biopsy. Our overall core biopsy yield of 147 (11%) cancers
in 1333 biopsies is thus lower than that reported in other series. Malignancy
was found in 17% of 6152 core biopsies at 20 community institutions
[6] and in 24% of a total of
3276 core biopsies reported from two large referral hospitals
[7,
8].
When patients from our practice with BI-RADS 5 lesions were referred for core biopsy, 92% of the lesions were found to be malignant. More recently, we have performed core biopsies in more patients with suspicious (BI-RADS 5) lesions who would previously have been referred directly to surgeons. The yield of malignancy for core biopsy for 29 months (January 1, 1999, to June 1, 2001) has thus increased to 87 (19%) of 459 stereotactic core biopsies and 153 (17%) of 898 solid masses sampled with sonographic guidance. In the first five months of 2001, 51 (25%) of 203 solid masses biopsied with sonographic guidance were malignant.
Of the 1027 core biopsy procedures with a BI-RADS assessment category recorded, 967 (94%) had results assessed as BI-RADS category 4. Among these 967 cases, there were 100 cancers (10%, Table 2). Therefore, imaging-guided core biopsy in our practice was most effective in obviating the need for surgical biopsy in the 90% of patients with BI-RADS 4 lesions but benign core biopsy results. We recognize that the broad spectrum of lesions encompassed by the BI-RADS 4 category allowed the inclusion of many findings of low suspicion for which follow-up may now be elected. However, at the inception of our program in 1994, core biopsy was recommended for such lesions, which, in our practice, would have otherwise been surgically sampled.
From January 1, 1997, to March 1, 1998, 262 needle-localized surgical biopsies were performed at our institution, with 105 cancers found (positive yield of 40%). This result compared with a cancer yield of 29% for needle-localized surgical biopsy at our hospital before 1994 [9]. Similar increases in cancers found at surgical biopsy after implementing an imaging-guided core biopsy program have been reported by others [8, 10].
Diagnostic Yield and Follow-Up of Patients with Benign Results
Jackman et al. [11]
reviewed eight studies in which stereotactically guided core biopsy was used
to diagnose 1276 cancers. The percentage of missed lesions ranged from
0.3-8.2%. In another series [7]
that described the highest sensitivity (99.7%), all procedures were performed
by one of three radiologists. Our decision to also limit performance of
imaging-guided core biopsy to three radiologists took advantage of the short
but steep learning curve that characterizes these procedures and the improved
accuracy documented with increased operator experience
[12]. Institutions with active
training programs may, therefore, reasonably be expected to report somewhat
lower sensitivity when the procedure is performed by multiple operators with
varying degrees of experience. Liberman et al.
[13] have reported that early
in an operator's stereotactic biopsy experience, a disproportionate number of
lesions are missed because of technical errors.
Among the 1020 patients with a benign diagnosis at imaging-guided core biopsy in our series, 979 (96%) had at least one visit for follow-up imaging recorded within 6 to 36 months after biopsy. Jackman et al. [11] instituted a program at a multispecialty clinic to vigorously pursue follow-up imaging in patients with benign imaging-guided core biopsy results. This approach was successful in the follow-up of 259 of 262 such patients. In the study by Jackman et al., as well as in our own, the specific times of postbiopsy return visits were not recorded. The tumor registry linkage we performed, however, should have captured patients with false-negative core biopsy results who failed to fully comply with our follow-up protocol. The same tumor registry has been successfully used to identify breast cancer cases in our catchment area [14].
We, too, have engaged dedicated personnel to monitor follow-up, mail reminders to patients, and notify referring physicians. These efforts have resulted in fewer patients with no recorded follow-up visits in our series compared with others [7, 8, 15]. A stable patient population in our geographic area, relatively few outside referrals, and a substantial managed care population also contributed to successful follow-up compliance.
To our knowledge, only one imaging-guided biopsy series reported a regional tumor registry interrogated to discover any cases of cancer missed by core biopsy [16]. This search did not identify any cancer diagnosis among 64 women for whom no mammographic follow-up or for whom less than 12 months of follow-up data were available. In our study of 1333 biopsies, one cancer was diagnosed at surgical excision after core biopsy yielded only nondiagnostic results. This cancer was diagnosed without delay because we excised all lesions when the pathology results were discordant with the imaging findings. No patient in our study with a benign core biopsy result was later diagnosed with breast cancer in the quadrant biopsied, through either imaging follow-up or tumor registry linkage.
All patients with a malignant core biopsy diagnosis had a confirmed surgical diagnosis of malignancy in the quadrant biopsied. No false-positive core biopsies were identified, permitting a positive predictive value calculation of 100%.
Core specimen radiographs did not convincingly reveal calcium in 25 patients who had undergone stereotactic biopsy for calcifications. These patients represented 7% of our stereotactic calcification cases. Twenty (80%) of these biopsies were done early in our experience. We elected follow-up in lieu of repeat biopsy in these patients for four reasons: First, nine (36%) of these 25 patients had been assigned a BI-RADS assessment category 3 with a biopsy performed to assuage the anxiety of the patient and referring physician. Second, a review of images indicated that the region of the calcifications had been adequately sampled (mean number of cores per lesion, 8; range of number of cores, 4-16). This finding was most frequent when nonclustered calcifications appeared dispersed on stereotactic digital images and would have been difficult to target with a 14-gauge automated biopsy gun. Third, in five patients, calcifications were identified in microscopic sections, and benign histology was judged to be concordant with mammographic findings. Finally, in seven other patients, postbiopsy images showed no or diminished calcifications. These were patients with atypical microcystic calcifications and have been previously reported [17].
These selection criteria may account for the variance between our results and those described by Liberman et al. [13], in whose series four patients without calcifications in cores had surgical biopsies with malignant results. In those cases, some initial biopsies were judged to be technically inadequate because of inaccurate positioning or use of a short-excursion gun. In another series of 236 calcification clusters sampled with 14-gauge stereotactic cores, seven cases failed to reveal calcium in core radiographs, and the patients underwent surgical biopsy with one malignancy and six benign results [18]. Although we did not miss any cancers when calcium was not revealed in core radiographs, a relatively low prior probability of breast cancer was present in our core biopsy patients because of selection criteria. At that time, other studies had not yet been published to guide management. Because of the experience that now has been reported by others, we recommend repeat biopsy for calcifications in patients with BI-RADS 4 or 5 assessments if no calcifications are identified at core specimen radiography.
Nineteen (2%) of 981 patients with benign core biopsy results who had complied with follow-up imaging showed suspicious changes. Images of 11 patients with an initial core biopsy diagnosis of fibroadenoma showed enlargement of the lesion at a follow-up examination. All underwent surgical excision, which revealed 10 fibroadenomas and one benign phyllodes tumor. Similar benign surgical results have been reported by Lee et al. [16] for seven fibroadenomas that enlarged after core biopsy and by Leung et al. [19] for 17 fibroadenomas that enlarged after core biopsy. Nevertheless, we continue to recommend surgical excision for enlarging masses, because malignancy has been reported in a small proportion of such lesions [11, 16].
Twelve (55%) of our 22 patients with ADH for whom surgical follow-up data were available were found to have a malignancy at surgery. This rate is somewhat higher than the 37% incidence of malignancy described by Reynolds [20] in an extensive literature review of 579 patients with ADH diagnosed at imaging-guided core biopsy. Three of the histologic underestimates occurred in samples obtained using sonographic guidance with 16-gauge needle, whereas eight underestimates occurred in samples obtained using stereotactic sampling with a 14-gauge automated gun. Only one underestimate of malignancy in a core biopsy yielding ADH occurred in samples obtained using an 11-gauge vacuum-assisted device.
Sonographic Guidance and Palpable Lesions
Sonography was selected as the guidance technique for 827 (86%) of the 961
masses sampled by core biopsy. This usage differs from that described in many
reports of imaging-guided core biopsy in which stereotactic guidance was
either the exclusive or predominant technique used
[2,
4,
6,
8,
18,
21]. Sonographic guidance
provided several advantages over the stereotactic technique. Use of the
sonographically guided procedure allowed patients to avoid exposure to
radiation, the discomfort of the stereotactic biopsy table, and problems
related to motion. In addition, it allowed real-time and orthogonal imaging
and could be performed more rapidly and at lower cost ($472 vs. $1048 at our
institution) than the stereotactic procedure. We elected to perform most of
the sonographically guided core biopsies with a 16-gauge rather than a
14-gauge needle, which is the needle size most commonly used, according to
reports in the stereotactic biopsy literature
[22]. Our pathologists found
that 16-gauge cores provided a sufficient sample for confident histologic
diagnosis. Technically, the smaller needle could be more readily maneuvered in
resistant, fibrous breast tissue. Others have described a similar preference
for smaller gauge needles for these same reasons
[23].
Other studies describing sonographically guided core biopsy have either largely excluded palpable lesions [2] or failed to indicate whether the masses sampled were or were not palpable. Information on whether a mass was palpable or not was recorded for 698 masses sampled by sonographically guided core biopsy in our series. Five hundred fourteen (74%) of these were palpable. A recent report by Liberman et al. [24] describes 115 palpable masses in patients referred (85% by surgeons) for core biopsy; further diagnostic tissue sampling was avoided in 73% of these patients. Of the 514 palpable masses in our series, 72 (14%) were malignant or showed ADH and required further surgery. Hence, 86% of our patients with palpable masses were spared further evaluation and diagnostic work-up by sonographically guided core biopsy. In our practice, core biopsy could be scheduled more promptly than surgical consultation for fine-needle aspiration or open biopsy after the patient's initial visit for diagnostic imaging. This advantage was an important consideration for anxious patients with recently discovered breast lumps. Both the series of Liberman et al. [24] and our experience suggest that not all palpable lesions need to be excised surgically.
In our series, the frequency of fibroadenomas (72%) found among palpable lesions suggests that imaging follow-up rather than biopsy may have been elected for some of these masses. Although clinical and sonographic features were strongly suggestive of fibroadenoma in many patients, we chose to recommend core biopsy for those patients with masses that had been discovered recently or were a source of concern or anxiety to patients or their referring physicians.
We hope that our experience with imaging-guided core biopsy in a community hospital will encourage other private practice radiologists to offer these procedures to their patients and to collect and tabulate the data that will further support the contribution of these procedures to cost-effective patient care.
Acknowledgments
We thank R. James Brenner, W. Phil Evans III, and Laura Liberman for
manuscript review and helpful suggestions; John Barclay for Northern
California Cancer Center Registry data linkage; Eric Wilcox for patient data
retrieval; and Estella Liu for manuscript preparation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. W. T. Leung, F. R. Margolin, K. E. Dee, R. P. Jacobs, S. R. Denny, and John. D. Schrumpf Performance Parameters for Screening and Diagnostic Mammography in a Community Practice: Are There Differences Between Specialists and General Radiologists? Am. J. Roentgenol., January 1, 2007; 188(1): 236 - 241. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Mendez, F. Cabanillas, M. Echenique, K. Malekshamran, I. Perez, and E. Ramos Mammographic features and correlation with biopsy findings using 11-gauge stereotactic vacuum-assisted breast biopsy (SVABB) Ann. Onc., March 1, 2004; 15(3): 450 - 454. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Jackman and F. A. Marzoni Jr. Stereotactic Histologic Biopsy with Patients Prone: Technical Feasibility in 98% of Mammographically Detected Lesions Am. J. Roentgenol., March 1, 2003; 180(3): 785 - 794. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Guenin and F. R. Margolin Full Disclosure of Breast Biopsy Options Am. J. Roentgenol., April 1, 2002; 178(4): 1029 - 1030. [Full Text] [PDF]
|
|
|
|
 |
|
 L. Esserman, H. Cowley, C. Eberle, A. Kirkpatrick, S. Chang, K. Berbaum, and A. Gale Improving the Accuracy of Mammography: Volume and Outcome Relationships J Natl Cancer Inst, March 6, 2002; 94(5): 369 - 375. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
