| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, Advocate Illinois Masonic Medical Center,
836 Wellington St., Chicago, IL 22005.
2 Department of Surgery, Advocate Illinois Masonic Medical Center, Chicago, IL
22005.
3 Department of Pathology, Advocate Illinois Masonic Medical Center, Chicago, IL
22005.
Received June 5, 2002;
accepted after revision February 26, 2003.
Address correspondence to K. J. Kirshenbaum.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Between September 1994 and February 2001, 506 stereotactic core needle biopsies of mammographic lesions in 492 patients were performed in our center on a mammography unit with add-on stereotactic equipment. Of the initial 92 patients, 80 underwent stereotactic core needle biopsy and surgical excision simultaneously. In subsequent cases, surgical biopsy was performed after core biopsy in patients who had malignant or atypical histologic results or discordance between mammographic and pathologic findings. Follow-up mammography was advised for all patients whose core biopsy results were diagnosed as benign lesions.
RESULTS. Histologic results for 506 lesions undergoing stereotactic core needle biopsy were as follows: 113 (22.3%) were malignant; 369 (72.9%), benign; and 24 (4.7%), atypical. Of 113 malignant lesions identified at stereotactic core needle biopsy, 111 were confirmed as malignant, whereas two showed no evidence of malignancy at surgical excision. Of 369 lesions diagnosed as benign at stereotactic core needle biopsy, 172 (46.6%) showed no change on follow-up mammography, 114 (30.9%) were lost to follow-up, and 83 (22%) underwent surgical excision. Of 24 lesions with atypical histology, 23 had surgical follow-up, six were malignant, nine were benign, and eight were confirmed as showing atypical histology. Stereotactic core needle biopsy of the 506 lesions was complicated by five (1.0%) cases of vasovagal attack and four (0.8%) cases of bleeding. The resulting sensitivity, specificity, and positive and negative predictive values were 98.3%, 93.0%, 86.0%, and 99.2% respectively.
CONCLUSION. Biopsy with an add-on unit is safe, reliable, accurate, and cost-effective with results comparable to those reported for dedicated prone biopsy devices.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Most published studies confirming the efficacy of core needle biopsy have focused on the use of dedicated prone units. Caines et al. [11] were the first to describe stereotactic core needle biopsy using a conventional mammography unit with an add-on stereotactic device. Only three other articles have highlighted the use of add-on units, two of which described the authors' limited experience with a reclinable chair [12, 13]; the remaining article described the use of biopsy in patients with microcalcifications alone [14].
Since 1994, our center has performed stereotactic core needle biopsy exclusively on a conventional mammography machine with an add-on unit. We undertook this prospective study to assess the accuracy of this technique.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In an attempt to establish the accuracy of the stereotactic core needle biopsy technique, the departments of surgery and radiology established a protocol in which the first 80 core needle biopsy procedures performed would have surgical confirmation immediately after the core needle biopsy procedure. Of the first 92 patients who underwent a core needle biopsy procedure at our institution, 12 elected to have the core biopsy procedure alone, foregoing surgical confirmation.
Subsequently, a protocol was established that called for all nonpalpable mammographic lesions referred for biopsy to undergo stereotactic core needle biopsy as the initial biopsy technique. Surgical biopsy alone was reserved for lesions deemed technically difficult to biopsy using the stereotactic core needle technique as well as for lesions with malignant or atypical histology, microcalcifications that were absent on both core specimen radiography and at histologic evaluation in patients whose indication for stereotactic core needle biopsy was the presence of microcalcifications, cases of discordant mammographic and pathologic findings (i.e., the imaging features of a lesion were suspicious enough to prompt excision in spite of a nonatypical or nonmalignant result at stereotactic core needle biopsy), or cases in which the patient or surgeon preferred that surgery be performed. All core biopsy specimens were interpreted by one of six pathologists, with results subsequently reviewed by the director of surgical pathology.
All biopsies were performed by one of three radiologists who specialize in breast imaging. A reusable biopsy gun (Bard, Murray Hill, NJ) with a disposable 14-gauge needle (Bard, Covington, CA) was used with a 22-mm needle throw for all automated core biopsy procedures. An 11-gauge vacuum-assisted biopsy (Mammotome, Biopsys Medical–Ethicon Endo-Surgery, Cincinnati, OH) was used for 23 of the final 32 biopsies performed for microcalcifications because this product only recently became available for use with add-on units. An automated 14-gauge core needle biopsy procedure was used for microcalcification clusters smaller than 5 mm in diameter, and an 11-gauge vacuum-assisted biopsy procedure was used for microcalcification clusters equal to or larger than 5 mm in diameter. Stereotactic core needle biopsies were performed using a stereotactic add-on device (Siemens, Erlangen, Germany) (Fig. 1A) attached to either a Mammomat 3 or 3000 conventional mammography unit (Siemens).
|
After initial diagnostic mammograms were reviewed, the target area was selected using a method that has been previously described [11]. The craniocaudal approach with the patient in an upright seated position was used whenever possible; however, patients with inferior lesions that were too closely positioned near the Bucky table required alternate positioning. We placed these patients in the recumbent position using a convertible recumbent chair (Bennet Medical X-ray Technologies, Copiague, NY) with subsequent needle entry from the medial or lateral aspect of the breast (Fig. 1B). After thorough skin cleansing and raising of a skin wheal using a 2% solution of lidocaine (Xylocaine, Astra-Zeneca Pharmaceuticals, Wilmington, DE), we punctured the patient's skin with the automated core needle, or, in the case of a patient imaged on a Mammotome unit, made a small incision with a scalpel. Subsequent biopsies were all performed through this initial puncture site.
|
Needle position was documented on a postfire stereotactic image-pair with additional routine automated core needle biopsies obtained 2–5 mm from the initial central biopsy site at 12-, 3-, 6-, and 9-o'clock positions. Vacuum-assisted Mammotome biopsies were routinely obtained by sampling tissue at the 12-, 3-, 6-, and 9-o'clock positions as well. Additional biopsies were then taken if deemed necessary on the basis of initial postfire stereotactic image-pairs, gross motion by the patient, or absence of calcifications on core specimen radiography (performed for all patients with microcalcifications to document their presence at the time of biopsy).
Histopathologic diagnoses of the results of automated 14-gauge core needle biopsy and vacuum-assisted 11-gauge biopsies were classified according to a modified classification system described by Brenner et al. [15]: nonspecific benign finding (e.g., fibrocystic change, ductal hyperplasia, or stromal fibrosis), specific benign finding (e.g., fibroadenoma), atypical finding (e.g., atypical lobular hyperplasia, atypical ductal hyperplasia, or atypical fibroepithelial lesion), atypical–high-risk finding (e.g., benign phyllodes tumor or radial scar), ductal carcinoma in situ, and invasive carcinoma (including invasive ductal carcinoma, invasive lobular carcinoma, and malignant phyllodes tumor).
After undergoing stereotactic core needle biopsy, all patients with a histologic diagnosis of atypical or atypical–high-risk finding, ductal carcinoma in situ, or invasive carcinoma were referred for surgical excision. Those with concordant nonspecific or specific benign stereotactic core needle biopsy and mammographic diagnoses were advised to have follow-up ipsilateral mammography at 6-month intervals for 2 years. Attempts to contact these patients were routinely made to obtain follow-up mammographic studies. Lesion characteristics were retrospectively reviewed and categorized on the basis of findings for the original mammographic images. Lesion type (mass, calcification, or asymmetric density), lesion size, patient position, number of cores per biopsy, and pathologic results were recorded.
Using a statistical analysis previously described by Brenner et al. [15], we calculated the overall sensitivity and specificity on the basis of whether the core biopsy yielded histologic results that prompted the appropriate management [9, 14, 16]. A stereotactic core needle biopsy diagnosis of a nonspecific or specific benign diagnosis in a lesion that subsequently proved to be malignant was considered to be a false-negative result. Atypical lesions that subsequently proved to be malignant were considered true-positive results, whereas those with benign or atypical excisional results at pathology were classified as false-positive results. Lesions with discordant results between mammography and pathology that were subsequently shown to be benign at excisional biopsy were considered false-positive results as well.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Histologic results for 506 lesions undergoing stereotactic core needle biopsy are presented in Table 1. Included are the histologic results of the first 80 lesions that automatically underwent immediate surgical biopsy: 19 (24%) were malignant, 59 (74%) benign, and 2 (2%) atypical findings.
|
Among our study population, 121 malignancies were diagnosed at surgery, stereotactic core needle biopsy, or both (a 23.9% malignancy rate). We found 41 cases of ductal carcinoma in situ, 75 infiltrating ductal carcinomas, two malignant phyllodes tumors, and three infiltrating lobular carcinomas. Two cases of ductal carcinoma in situ diagnosed at stereotactic core needle biopsy did not show malignancy at surgery and were considered to have been entirely removed at core biopsy [16]. Core needle biopsy results for the remaining 119 surgically confirmed malignancies were 111 (93.3%) malignant findings, two (1.7%) benign findings, and six (5%) atypical findings. Both of the lesions with benign core needle biopsy results that were subsequently proven to be malignant at surgical excision (performed because of discordant mammographic and pathologic findings) were cases of ductal carcinoma in situ.
Of 369 lesions diagnosed at stereotactic core needle biopsy as benign, nonspecific benign, or both, 172 (46.6%) showed no change on follow-up mammography (range, from 3 months to 5 years; mean, 2.1 years), 114 (30.9%) were lost to follow-up, and 83 (22.5%) underwent surgical excision. Of the 83 cases that underwent surgical excision, 59 (71.1%) were among the initial cases that automatically underwent surgery (all were found to be benign at surgery), 21 (25.3%) were removed because of either patient or surgeon preference (all were subsequently found to be benign at surgery), and three (3.6%) were removed because of mammographic–pathologic discordance. In two of these three discordant cases—one case of ductal carcinoma in situ and one atypical ductal hyperplasia—microcalcifications were absent on specimen radiography or at histologic evaluation. In the third discordant case of ductal carcinoma in situ, only a single microcalcification was removed. In all other cases, successful retrieval of microcalcifications was documented on specimen radiography, histologic evaluation, or both.
Of 24 lesions diagnosed at stereotactic core needle biopsy as atypical, surgical follow-up showed six were ductal carcinomas in situ, seven were cases of atypical ductal hyperplasia, one was a case of atypical lobular hyperplasia, three were benign phyllodes tumors, and six were benign or nonspecific benign legions. One lesion with an atypical result at stereotactic core needle biopsy was not surgically excised and was lost to follow-up.
Among our study population, 219 lesions underwent surgical excision (Table 2). If we excluded the 113 cases of malignancy diagnosed at stereotactic core needle biopsy, 106 lesions with nonmalignant histologic findings at core needle biopsy were surgically excised. If we excluded those lesions that were automatically surgically excised in the first 80 patients (61 nonmalignant findings: 59 benign and two atypical lesions), 45 lesions diagnosed as nonmalignant at stereotactic core needle biopsy were surgically excised either at the request of the patient or discretion of the surgeon (21 lesions) or because of discordant mammographic–pathologic findings (three lesions) or atypical results (21 lesions). A total of 393 lesions were diagnosed as being benign or atypical histologically at stereotactic core needle biopsy, 115 of which were lost to follow-up (114 benign lesions and one atypical lesion) and 278 of which had either surgical or mammographic follow-up. Therefore, these 45 lesions represent a rebiopsy rate of 16.2% (45/278). If those cases in which patients underwent surgical excision because of patient or surgeon preference were excluded, the rebiopsy rate was actually 8.6% (24/278).
|
Stereotactic core needle biopsy was complicated by five vasovagal episodes (1.0%, 5/506), with the procedure being terminated early (three cases with one core retrieval and two cases with two core retrievals). All three participating radiologists had at least one procedure complicated by a vasovagal reaction. The primary investigator responsible for performing most of the core biopsy procedures (80.8%, 409/506) had a vasovagal complication rate of three in 409 (0.7%), whereas the two other investigators had complication rates of 2.1% (1/47), and 1.9% (1/53), respectively. All vasovagal attacks involved patients who underwent biopsy while in the upright position; however, all the specimens obtained during the biopsies were diagnostic, with no false-negative results noted (three invasive carcinomas and two benign lesions with no change at the 6-month and 8-month followup, respectively). Minor bleeding was encountered in three (0.6%, 3/506) cases. There was one case (0.2%) of major bleeding that required surgical control (11-gauge vacuum-assisted biopsy needle was used). No complications were noted in any of the 13 patients who underwent multibiopsy procedures.
Several patients who were referred for stereotactic core needle biopsy cancelled the procedure because mammographic abnormalities resolved or prebiopsy stereotactic mammography allowed us to determine that the lesion was not persistant. Although a specific tally of cancelled procedures was not recorded, none of the cancellations had to do with factors related to the add-on unit.
We retrospectively reviewed both core and excisional biopsy histologic findings, with 19 discrepant results between surgical and stereotactic core needle biopsy findings (Table 3). No changes in final histologic results were noted on this review. Excluding the 115 patients lost to follow-up resulted in sensitivity, specificity, and positive and negative predictive values of 98.3%, 93.0%, 86.0%, and 99.2%, respectively.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Although three radiologists participated in our series, one was responsible for approximately 80% of the lesions biopsied, and this fact may have skewed our results toward greater accuracy rates than those previously reported. In addition, our study was limited in that follow-up data were unavailable in 30% of our cases with benign findings at stereotactic core needle biopsy.
The problem of patients dropping out of follow-up is common to most biopsy studies, with rates ranging from 25–39% [20–24]. According to Kopans [25], 24–77% of cases with benign stereotactic core needle biopsy results in multiple prone biopsy studies lack longterm follow-up data. In two large multiinstitutional trials of the prone biopsy technique [9, 15], 20–35% of women did not return for a 6-month follow-up study. The lack of follow-up data does make it difficult to assess the true accuracy of stereotactic core needle biopsy with an add-on unit, but our data are no more limited than the prone biopsy data that has preceded it. Although most patients underwent biopsy in an upright seated position and were approached craniocaudally, 27% of patients were recumbent in a reclinable chair. Positioning of patients was chosen based on shortest distance from the skin to the lesion and on which position the lesion was better seen to enable the most accurate targeting. In addition, decubitus positioning was used in patients if craniocaudal positioning placed the lesion too close to the Bucky table (< 20 mm from the tabletop).
Our overall sensitivity (98.3%), specificity (93.0%), positive predictive value (86.0%), and negative predictive value (99.2%) compared favorably with those reported by the previously cited investigators for the prone biopsy system. Our rebiopsy rate of 16.5% included those cases in which the patient or surgeon requested excision. If we excluded this subset of patients, our rebiopsy rate of 8.6% compared favorably with the rates of most prone biopsy series, which have ranged from 2% to 17% [20].
The primary purported disadvantage of an add-on stereotactic core needle biopsy system has been its higher incidence of vasovagal complications with resultant patient motion and potential for reduced diagnostic accuracy. In our series, we encountered five vasovagal episodes (incidence, 1%) necessitating early termination of the procedure. The primary investigator performed most of the stereotactic core needle biopsy procedures (80.8%, 409/506) and had the lowest rate of vasovagal complications (0.7%, 3/409). Caines et al. [11] reported a similar low incidence of vasovagal reactions that compared favorably with prior prone studies. Doyle et al. [12] and Welle et al. [13] described vasovagal rates of 29% and 38%, respectively. Operator experience, difficult lesion location, increased procedure time, and increased number of core samples obtained are all possible factors that could play a role in this significant discrepancy in vasovagal rates reported. Thorough patient preparation—with the physician and support personnel explaining the procedure to the patient—establishes a rapport with prospective biopsy patients and is critical to ensuring the success of stereotactic core needle biopsy using an add-on unit.
One major bleeding episode that required surgical control occurred during one of our first 11-gauge vacuum-assisted biopsies. Three minor bleeding cases (hematomas < 2 cm) resolved without intervention. In most cases, very little bleeding was noted.
The cost differential between add-on and dedicated prone biopsy units is significant. A breast imaging center need spend only approximately $90,000 (average list price of add-on device) to make an existing mammography unit biopsy-ready. For a dedicated prone biopsy table, a center would need to spend $226,000 (average list price). If one includes the additional cost of purchasing a mammography machine (average list price, $80,000) that might be required because the add-on biopsy unit is incompatible with an existing mammography machine, the cost differential is substantially reduced. However, when not being used for biopsies (which is most of the time at a small to mid-sized imaging center), add-on units can be used for general screening and diagnostic work, whereas prone units can only be used for biopsies. In addition, the space requirements for a prone system are significantly greater than those required for an add-on system. Together, the cost and space requirements for a prone system become prohibitive for small and mid-sized centers.
Vacuum-assisted 11-gauge biopsy did not become available for add-on units until late in our study series; we biopsied only 23 cases of microcalcifications using this system. Although accuracy should increase with the use of the Mammotome system [26, 27], we did experience our only significant bleeding episode with the Mammotome.
In conclusion, stereotactic core needle biopsy performed on a conventional mammography unit with an add-on stereotactic device is a safe, reliable, and accurate method in the diagnosis of nonpalpable breast lesions seen on mammography. The add-on system offers sensitivity, specificity, and accuracy for stereotactic core needle biopsies comparable to those of a dedicated prone biopsy system in addition to significant cost and space-saving advantages.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Georgian-Smith, C. D'Orsi, K. J. Kirshenbaum, T. Voruganti, and C. Overbeeke Beware of False Claims Am. J. Roentgenol., April 1, 2004; 182(4): 1084 - 1084. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
