| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Imaging, Brown Medical School, Rhode Island Hospital,
Providence, RI.
2 Present address: Department of Radiology, Beth Israel Deaconess Medical
Center, 330 Brookline Ave., Boston, MA 02215.
3 Department of Pathology, Brown Medical School, Rhode Island Hospital,
Providence, RI.
Received March 6, 2007;
accepted after revision May 27, 2007.
WEB
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. The reports of 1,223 consecutive stereotactic vacuum-assisted breast biopsies were retrospectively reviewed. An 11-gauge device was used to perform 828 and a 9-gauge device to perform 395 biopsies. The pathologic results were reviewed for all cases. Biopsy results of atypical ductal hyperplasia and ductal carcinoma in situ were compared with the pathologic results after surgical excision. Underestimation was defined as the need to upgrade atypical ductal hyperplasia to ductal carcinoma in situ or invasive carcinoma at surgery and to upgrade ductal carcinoma in situ to invasive carcinoma. Statistical significance was determined with the chi-square test and 95% CI.
RESULTS. In the 11-gauge group, 12 (26%) of 46 cases of atypical ductal hyperplasia were upgraded to ductal carcinoma in situ and one (2%) of the cases to invasive carcinoma. In the 9-gauge group, six (22%) of 27 cases of atypical ductal hyperplasia were upgraded to ductal carcinoma in situ and two (7%) of the cases to invasive carcinoma. In the 11-gauge group, 35 (28.7%) of 122 cases of ductal carcinoma in situ were upgraded to invasive carcinoma. In the 9-gauge group, 10 (23%) of 44 cases of ductal carcinoma in situ were upgraded to invasive carcinoma.
CONCLUSION. There was no statistically significant difference between 11-gauge biopsy and 9-gauge biopsy in underestimation of atypical ductal hyperplasia and ductal carcinoma in situ.
Keywords: atypical ductal hyperplasia • ductal carcinoma in situ • stereotactic breast biopsy • underestimation of disease • vacuum-assisted breast biopsy
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The amount of tissue obtained with each core biopsy sample increased from approximately 17 mg/core with automated 14-gauge biopsy to 34–40 mg/core with 14-gauge vacuum-assisted breast biopsy [17, 18]. Similarly, the average weight per core biopsy sample increased to 94–96 mg with 11-gauge vacuum-assisted breast biopsy [17, 19]. Although we found no published data on 9-gauge vacuum-assisted breast biopsy sample weights or volumes, one would expect a larger volume of tissue to be obtained with a 9-gauge needle. Brem and Gatewood [20] found a 39% increase in tissue acquisition when they compared 8- and 11-gauge vacuum-assisted breast biopsies.
The rate of histologic underestimation with 11-gauge vacuum-assisted breast biopsy is in the range of 10–27% for ADH and 5–18% for DCIS [14]. Although the use of 9-gauge vacuum-assisted biopsy has been reported in studies of MRI-guided breast biopsy [21, 22], to our knowledge there has been no comparison of histologic underestimation with 9- and 11-gauge stereotactic vacuum-assisted breast biopsies. Our hypothesis was that the rate of underestimation would be lower for 9-gauge vacuum-assisted breast biopsy than for 11-gauge vacuum-assisted breast biopsy. The purpose of our study was to compare the rate of histologic underestimation with 9-gauge probes with the rate for 11-gauge probes in stereotactic vacuum-assisted breast biopsy of lesions yielding ADH and DCIS.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Biopsy Procedure
From July 2000 to May 13, 2003, all stereotactic vacuum-assisted breast
biopsies were performed with an 11-gauge device (Mammotome, Biopsys
Medical/Ethicon Endo-Surgery). During this period, a total of 828 stereotactic
vacuum-assisted breast biopsies were performed. Beginning on May 14, 2003,
through December 2004, all stereotactic vacuum-assisted breast biopsies were
performed with a 9-gauge device (ATEC 9-G, Suros Surgical Systems). During
this period, a total of 395 stereotactic vacuum-assisted breast biopsies were
performed.
The biopsy protocol was identical for both groups of patients. Selection criteria for stereotactic biopsy also were identical for the 11- and 9-gauge groups. The same group of radiologists worked at the breast center throughout the study. Prebiopsy medication restriction guidelines were identical throughout the study period. Routine postbiopsy care (manual compression, dressing [Tegaderm, 3M Medical Specialties], and ice application), and postbiopsy instructions were identical throughout the study. In both the 11- and 9-gauge groups, an Ace bandage was applied in cases of acute postbiopsy hematoma.
Stereotactic biopsy was used for calcifications and noncalcified lesions (masses, architectural distortion, asymmetries) that could not be adequately visualized with breast sonography. All biopsy targets were nonpalpable BI-RADS category 4 or 5 lesions. Whenever technically feasible at our institution, BI-RADS category 4 and 5 lesions are percutaneously biopsied rather than localized for surgical biopsy. Lesion size and density of breast parenchyma were not used as criteria for exclusion from stereotactic biopsy. Informed consent for biopsy was obtained from each patient.
Stereotactic biopsy was performed with a dedicated prone unit (MammoTest, Fischer Imaging) and local anesthesia. Images were obtained before and after the biopsy device was activated to document accurate needle positioning within the targeted lesion. Retargeting was performed if necessary. Core biopsy specimens typically were obtained in a 360° rotation with the directional biopsy instrument, particularly when needle placement was within the center of the lesion. When radiographs obtained after the instrument was activated showed the needle to be immediately adjacent to the lesion, cores were obtained with the bowl of the needle directed toward the lesion. Standard practice at our institution at the time of the study was to obtain a minimum of six cores for both masses and calcifications and then to perform specimen radiography for calcifications. The radiologist assessed the specimen radiograph for adequacy of calcification retrieval and obtained additional cores as needed. Specimen radiography was not performed during biopsy of masses or in the presence of architectural distortion or asymmetries.
At the completion of the biopsy, a radiopaque biopsy marking clip (Gel Mark Ultra, SenoRx, or MicroMark, Biopsys/Ethicon–Endosurgery) was inserted into the biopsy site. Postbiopsy mammograms were obtained to confirm clip placement. Lesion type, number of cores obtained, and pathology results were recorded by the radiologist performing the procedure. Acute complications such as hematoma also were recorded. A spreadsheet program (Microsoft Office Excel 2003, Microsoft) was used to enter the data.
Data Collection and Analysis
The stereotactic breast biopsy database was reviewed to identify 302
patients in whom pathologic evaluation of stereotactic biopsy specimens
yielded ADH or DCIS during the study period. Medical records were then
reviewed to determine the pathology result at final surgical excision. In 63
cases, surgical pathology reports were not available, usually because the
patient had been referred for biopsy from an outside facility and there was no
record of the institution at which surgery was performed. These cases were
excluded from analysis. The final signed pathology reports of percutaneous
biopsy and surgical excision were accepted as determinants of the presence of
abnormalities in the breast in the remaining 239 cases. At our institution,
each breast pathology case is reviewed by two pathologists before the final
report is released. Individual slides were not reexamined for this study.
The percentage of lesions diagnosed as ADH or DCIS at vacuum-assisted breast biopsy and the pathology result at surgery were compared between the 9- and 11-gauge biopsy groups. Underestimations were defined as the need to upgrade ADH to DCIS or invasive carcinoma at surgery and as the need to upgrade DCIS to invasive carcinoma. Mammograms were available for 143 of 239 patients who underwent surgical follow-up. Mammographic size of the biopsy target was recorded and correlated with the underestimation rate.
Evaluation of 9-Gauge Sample Weight and Quality
Samples of 9-gauge specimen weights were obtained for comparison with
information on historic controls. Cores from 10 randomly selected patients
were weighed in the pathology laboratory before being processed in formalin.
Weights of each core specimen and a composite weight were obtained.
Statistical Analysis
Differences between pathology results for the two groups were analyzed for
statistical significance by use of the chi-square test and 95% CI. Statistical
significance was considered p < 0.05. The statistical calculations
were performed with a spreadsheet program (Microsoft Office Excel 2003). Data
on the number of specimens obtained per biopsy, mammographic size of the
biopsy target, and biopsy needle size were entered with statistical software
(SAS version 9.1.3, SAS Institute). Mean, median, and range of the
mammographic size of biopsy targets were calculated and compared by use of the
Wilcoxon's rank sum statistic. The relations of underestimation to biopsy
needle size, number of biopsy samples obtained, and mammographic size of the
biopsy target were evaluated by multiple logistic regression.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The average patient age was 57 years (range, 28–92 years) in the 11-gauge group and 56 years (range, 28–87 years) in the 9-gauge group. A mean of eight specimens per lesion (range, 6–24 specimens) were obtained with 11-gauge biopsy and a mean of seven specimens (range, 1–20 specimens) with 9-gauge biopsy. The difference in number of specimens obtained per lesion between the 11- and 9-gauge groups was analyzed for statistical significance with Wilcoxon's rank sum test and was found not statistically significant (p = 0.3). The mean mammographic size of the biopsy targets was 12.2 mm (range, 5–70 mm) for 11-gauge and 9.4 mm (range, 5–30 mm) for 9-gauge vacuum-assisted breast biopsy. Wilcoxon's rank sum test showed there was no statistically significant difference between the mammographic lesion sizes of the two groups (p = 0.4). Neither group had acute complications necessitating intervention. Twelve hematomas were reported in the 11-gauge group and one in the 9-gauge group, none of which required treatment. No infections were reported.
Biopsy Specimen
Nine-gauge biopsy sample weights averaged 124 mg/core in 10 randomly
selected patients; Berg et al.
[17] and Burbank
[19], however, had reported an
11-gauge specimen weight of approximately 94–96 mg/core. Each histologic
section obtained from a 9-gauge vacuum-assisted breast biopsy core sample was
approximately 2.5 mm wide, and each section obtained from an 11-gauge sample
was approximately 1.5 mm wide.
ADH Underestimation
Histologic examination yielded ADH in 62 (7.5%) of 828 lesions in which an
11-gauge device was used for stereotactic vacuum-assisted breast biopsy.
Sixteen lesions were excluded because a surgical pathology report was not
available. Surgical pathology results and underestimation rates for the other
46 cases are shown in Table 1.
Twelve cases were upgraded to DCIS and one case to invasive carcinoma.
Histologic examination yielded ADH in 33 (8.4%) of 395 cases in which a
9-gauge device was used. Six cases were excluded because a surgical pathology
report was not available. Surgical pathology results and underestimation rates
for the other 27 cases are shown in Table
1. Six cases were upgraded to DCIS and two to invasive
carcinoma.
|
DCIS Underestimation
Histologic examination yielded DCIS in 155 (18.7%) of 828 lesions in which
an 11-gauge device was used for stereotactic vacuum-assisted breast biopsy.
Thirty-three lesions were excluded because a surgical pathology report was not
available. Surgical pathology results and underestimation rates for the other
122 cases are shown in Table 1.
Among the 35 cases upgraded to invasive carcinoma at surgery, there was one
case of invasive lobular carcinoma. This instance was the only one in the
study in which the diagnosis of invasive carcinoma was not invasive ductal
carcinoma. Histologic examination yielded DCIS in 52 (13.2%) of 395 cases in
which a 9-gauge device was used for stereotactic vacuum-assisted breast
biopsy. Eight cases were excluded because a surgical pathology report was not
available. Surgical pathology results and underestimation rates for the other
44 cases are shown in Table
1.
Statistical Results
Differences in underestimation rates between the 11- and 9-gauge groups
were not statistically significant according to chisquare results (ADH, 28.3%
vs 29.6%, p = 0.9; DCIS, 28.7% vs 22.7%, p = 0.4) or results
of multiple logistic regression including needle gauge (ADH odds ratio [OR],
0.413, p = 0.7; DCIS OR, 2.660, p = 0.6), number of samples
(ADH OR, 2.303, p = 0.6; DCIS OR, 0.322, p = 0.5), and their
interaction (ADH OR, 0.928, p = 0.6; DCIS OR, 1.106, p =
0.5) in the model. Logistic regression (OR, 0.891, p = 0.1) also
showed no difference in rates of underestimation of disease in cases of ADH as
a function of the mammographic size of the biopsy target. In cases of DCIS,
however, there was a statistically significant increase in underestimation
rate as lesion size increased (OR, 1.039, p = 0.03).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Underestimation rates improved with the introduction of vacuum-assisted biopsy techniques and again with the use of larger-gauge biopsy needles. Several studies [24–26] have shown that underestimation is less common in cases in which all or nearly all of the targeted lesion is removed. The findings suggest that underestimation results from inadequate sampling. One study [27] showed a trend toward fewer DCIS underestimations in cases in which the mammographic target had been completely excised at 11-gauge vacuum-assisted breast biopsy. Our finding that underestimation of disease in cases diagnosed as DCIS at stereotactic biopsy increased with increasing mammographic size of the biopsy target is concordant with these results. Sampling errors should therefore be less likely in cases in which more tissue can be removed and analyzed.
In this study, the rates of underestimation of ADH and DCIS were slightly higher than those in a previous report [14]. In some studies [27–29], the number of biopsy samples obtained for each lesion has been more than 20 (mean, 13–15), whereas in our study means of seven (9-gauge) and eight (11-gauge) samples were obtained. The fewer biopsy samples obtained per lesion may account for the somewhat higher underestimation rates in our study than in other studies. An alternative explanation is that variation in pathologists' thresholds for diagnosing ADH and DCIS affect the histologic underestimation rate at stereotactic biopsy.
Philpotts et al. [29] reported a significantly higher underestimation rate for vacuum-assisted breast biopsy of calcifications than for vacuum-assisted breast biopsy of masses. This difference might have contributed to the slightly higher underestimation rate in our study, because most of the targeted lesions in both the 9- and 11-gauge vacuum-assisted breast biopsy groups were calcifications. Although Jackman and colleagues [30, 31] reported higher underestimation rates for mass lesions than for micro-calcifications, their study design differed substantially from ours in selection of cases for stereotactic biopsy. In one study [31], none of the mammographically evident lesions was biopsied with sonographic guidance during the study period. In the other study [30], the biopsy protocol varied among clinical sites and it is not clear whether sonographically guided biopsy was used on lesions evident on both mammography and sonography. In our study, all lesions reliably identified with sonography were biopsied under sonographic guidance.
Lomoschitz et al. [25] found no decrease in the underestimation rate using 20 rather than 12 samples per lesion for 11-gauge vacuum-assisted breast biopsy. Jackman et al. [30] reported that underestimation of DCIS and ADH was more frequent when 10 or fewer specimens were obtained. We expected that because the 9-gauge biopsy should obtain larger amounts of tissue, the diagnostic yield would improve. There was no significant difference, however, between 9- and 11-gauge biopsies in rate of underestimation of either ADH or DCIS. It may be that the difference in volumes of tissue harvested was too small to have a significant effect on diagnostic yield at pathologic examination. In addition, the mean number of samples in the 9-gauge group was slightly lower than the mean number of samples in the 11-gauge group (seven vs eight), although this difference was not statistically significant. The slight difference in number of samples likely occurred because in our practice adequate retrieval of calcifications was the deciding factor for how many cores to obtain. The larger, 9-gauge needles may have removed calcifications in the first set of six cores more adequately than did the 11-gauge needles. Therefore, it is possible that there would have been a difference in underestimation rate if the same number of cores had been obtained with each system. That there was no difference in underestimation rate as a function of number of cores is also likely due to our technique in which adequacy of specimen retrieval was the determining factor in the number of samples obtained. With this technique, we had few cases in which we obtained the large number of cores obtained in other studies [27–29].
Similarities in histologic underestimation for 9- and 11-gauge vacuum-assisted breast biopsies may also relate to how tissue is analyzed at pathologic examination. Although larger tissue specimens are obtained, the number of slides analyzed at pathologic examination at our institution generally depends on the number of core specimens obtained and thus may account in part for the similar underestimation rates for 9- and 11-gauge biopsies. Similarities in underestimation for the 11- and 9-gauge groups may also relate to the nature of the disease itself in that both ADH and DCIS frequently occur in scattered ducts rather than as a confluent area of abnormal cells. This factor makes sampling error nearly impossible to avoid in some cases and may explain why some investigators [30] have found that 25% of ADH underestimations are invasive carcinoma regardless of biopsy device. A study [32] comparing 8- and 11-gauge vacuum-assisted breast biopsies also showed no significant difference in breast cancer diagnosis between the two techniques.
In addition to differences in needle size, the 9- and 11-gauge devices were different biopsy systems, which might have affected the volume of tissue retrieved. The 11-gauge system delivered each core as it was obtained through an opening, providing vacuum assistance to deliver the specimens individually to the operator. The 9-gauge device was a closed system in which the cavity was lavaged with saline solution after each sample was obtained, and the samples were delivered together into a mesh basket. The saline lavage system offers the potential benefit of minimizing the number of samples left behind in the cavity or the biopsy unit.
The strengths of our study included identical inclusion criteria, biopsy methods, and histopathologic assessment for the two groups. The limitations included the retrospective design, which made it difficult to find surgical results in all cases. Follow-up was difficult in cases in which patients had been referred from outside hospitals for biopsy. In addition, fewer 9-gauge than 11-gauge biopsies were evaluated because 9-gauge vacuum-assisted breast biopsy is a newer technique. It is possible that the lack of a substantial difference in underestimation between the two groups was related to an insufficient number of patients in the sample. Further studies with larger numbers of cases and with larger number of cores obtained with 9-gauge vacuum-assisted breast biopsy would be helpful to further evaluate underestimation with this technique. Increasing the number of slides analyzed at pathologic examination also may improve the diagnostic accuracy of vacuum-assisted breast biopsy. In particular, obtaining additional slides through core biopsy specimens that contain identifiable parts of the targeted lesion on specimen radiographs may have a high yield. In addition, assessment of the complication rate in this study relied on the radiologist's subjective perception of hematoma and patients' reporting of infection and was therefore limited by the potential for delayed and unreported complications. In future studies, it would be helpful to establish objective criteria by which to compare complication rates for the two groups.
In conclusion, underestimation of disease remains a challenge in the diagnosis of ADH and DCIS with both 11- and 9-gauge vacuum-assisted breast biopsies. Surgical excision is necessary for further evaluation of ADH lesions with either type of biopsy. In lesions yielding DCIS at 11- or 9-gauge vacuum-assisted breast biopsy, surgery may reveal unsuspected areas of invasion. In our series, the underestimation rates for stereotactic vacuum-assisted breast biopsies performed with 9- and 11-gauge devices did not differ significantly. Further studies with larger series of patients are needed to further evaluate the effect of the tissue acquisition device on histologic underestimation at percutaneous breast biopsy.
Acknowledgments
We thank Jason T. Machan, PhD, for providing statistical analysis for this
study and the staff of the Anne C. Pappas Center for Breast Imaging for
assistance with data collection.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
