| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Breast Health Center, California Pacific Medical Center, 3698 California St., San Francisco, CA 94118.
Received December 3, 2002;
accepted after revision July 3, 2003.
Address correspondence to F. R. Margolin.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. One hundred nineteen clips deployed through an 11-gauge vacuum-assisted biopsy probe at core biopsy sites were compared with 109 vascular ligating clips deployed at biopsy sites using an 18-gauge spinal needle. The distance of each clip from the position of the target calcification was assessed using stereotactic coordinates in 52 sequential cases and was measured on mammograms before and after biopsy in 108 clips deployed through an 11-gauge probe and 98 clips deployed using an 18-gauge needle. Variance in clip position between postbiopsy and follow-up mammograms was measured in 43 clips placed with an 11-gauge probe and in 44 clips placed with an 18-gauge needle. Comparable measurements of variance in position of fat necrosis calcifications between screening mammograms were used as controls.
RESULTS. Ninety-seven percent of the clips placed with an 11-gauge probe and 98% of the clips placed using an 18-gauge needle were within 1 cm of the target calcifications using stereotactic coordinates. On mammograms obtained after biopsy, 70% of the clips placed with an 11-gauge probe and 63% of the clips placed using an 18-gauge needle were within 1 cm of the target calcifications, and the position of 91% of the clips placed with an 11-gauge probe and 90% of the clips placed using an 18-gauge needle varied less than 15 mm on follow-up mammograms. Both clips provided accurate targets for wire-localized excisions. The cost of the 11-gauge needle and clip is $320. The 14-gauge probe, vascular clip, and 18-gauge spinal needle cost $191.58.
CONCLUSION. A vascular ligating clip delivered to a stereotactic core biopsy site by an 18-gauge spinal needle is comparable in apparent accuracy and stability to a clip deployed through an 11-gauge probe. This technique allows core biopsies to be performed with instruments smaller than 11-gauge and at a 40% savings in equipment cost.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Burbank and Forcier [1] in 1997 described a metallic clip designed to adhere to tissue at the biopsy site. This clip is deployed through an 11-gauge needle. Its use has increased fivefold from 1998 to 2001, according to a leading supplier of these instruments (Randall D, personal communication).
Three of four radiologists performing stereotactic core biopsies in our community hospital–based private practice obtained samples of all calcification cases with an 11-gauge vacuum-assisted biopsy probe (Mammotome, Biopsys/Ethicon-Endo Surgery, Cincinnati, OH), and when placement of a marker at the biopsy site was indicated, the tissue-marking clip described by Burbank and Forcier [1] (MicroMark and MicroMark II, Biopsys/Ethicon-Endo Surgery) and designed for use with this instrument was deployed through the 11-gauge biopsy probe.
In 1998, Curpen reported her experience with a simple, rapid, inexpensive, and accurate method for deployment of a small titanium clip through an 18-gauge spinal needle at stereotactic biopsy (Curpen NB et al., Radiologic Society of North America meeting, November 1998). This tissue-marking technique was used exclusively by one radiologist in our practice who sampled all calcification cases with a 14-gauge vacuum-assisted biopsy needle (Mammotome). In this report, we compare these two methods of clip deployment and tissue marking for initial placement accuracy, usefulness as a guide for wire localization, and stability over time. These two groups are identified by the gauge of instrument used for clip deployment and are referred to as "11-gauge clips" or "18-gauge clips" for simplicity.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
For biopsies performed with an 11-gauge probe, the marker clip designed for use with this instrument was deployed through the probe in accordance with the manufacturer's instructions and as described by Burbank and Forcier [1] and Liberman et al. [2].
For all cases in which a clip was used, stereotactic images were obtained to confirm deployment (Fig. 3B). Craniocaudal and mediolateral mammograms immediately after the procedure were compared with mammograms before biopsy to correlate the sites of tissue sampling and marker placement (Fig. 3C).
|
|
From March 11, 1999, to May 30, 2002, after informed consent for stereotactic core biopsy and clip placement was obtained, 551 lesions were sampled in 529 patients. Four hundred eighty-one procedures (87%) were done for calcifications. For all cases, an 11- or 14-gauge vacuum-assisted biopsy needle was used. Three radiologists used the 11-gauge probe exclusively in 241 consecutive calcification cases to place a clip through this needle at 119 biopsy sites. One radiologist sampled 240 calcifications in consecutive patients with a 14-gauge probe and deployed 109 clips with an 18-gauge spinal needle after removing the 14-gauge probe. For all cases, the size of the target calcifications, number of cores obtained, number of cores showing calcification on specimen radiographs, whether a clip was deployed, and final pathologic diagnosis were recorded.
After the institutional review board approved this investigation, retrospective data were analyzed from consecutive deployments of 110 11-gauge clips and 98 18-gauge clips. The mammograms for nine of the 11-gauge clips and 11 of the 18-gauge clips were unavailable for review.
Stereotactic Coordinates
For 52 consecutive clip deployments of each type, the x, y, and
z coordinates of the clip were determined and compared with those of
the target calcifications. The difference between these measurements was
recorded to the nearest half millimeter.
Mammograms After Biopsy
For 108 11-gauge clips and 98 18-gauge clips, craniocaudal, mediolateral,
and mediolateral oblique views obtained immediately after biopsy were compared
with prebiopsy mammograms. A horizontal line was drawn directly from the
nipple to the base of the breast on each view; this line was considered the
"central axis" of the breast. The distance of the prebiopsy target
calcifications and postbiopsy clip measured perpendicular to this line was
recorded to the nearest millimeter. The difference between these values on
each view was averaged for all views and was recorded as the lesion–clip
separation for each case.
Follow-Up Mammograms
In 43 11-gauge and 44 18-gauge cases with benign core biopsy results,
follow-up mediolateral oblique and craniocaudal mammograms were available at
least 6 months after the procedure. Measurements of clip position on these
examinations were made by the same method used on mammograms immediately after
biopsy, and variation in clip position between examinations was recorded to
the nearest millimeter.
Fat Necrosis Control Cases
In an effort to establish baseline variations for the measurement method
used in postbiopsy and follow-up examinations, we reviewed the mammograms of
123 consecutive screening patients with identical 3- to 8-mm calcific foci of
fat necrosis visible on examinations 12–61 months apart. Breast size in
these cases was recorded as small in 32 patients, medium in 62, and large in
29. In each case the distance of the calcific foci from the central axis of
the breast was determined on both craniocaudal and mediolateral oblique
mammograms. The measured difference between the location of these
calcifications on both examinations was recorded to the nearest millimeter.
This duplicated the measurement method used for postbiopsy and follow-up clip
placement cases. The effect of variations in breast position, compression, and
tube angulation on the location of fat necrosis calcifications on sequential
screening mammograms could then be used as a control or baseline variable and
applied to clip and lesion measurements in the study
groups.
|
Pathology in Cases with Atypical or Malignant Results
Surgical biopsy data were available for 46 lesions with atypical or
malignant core biopsy results. Twenty-six lesions were biopsied with an
11-gauge probe and 20 with a 14-gauge probe. All patients had clips deployed
and subsequent wire-localized surgical excisions. In each case, the
localization procedure and specimen radiographs were reviewed, and the
presence of the clip was recorded. The core biopsy and excisional procedure
diagnoses and the maximum size of the resected surgical specimen were
recorded.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
After the patient's breast was released from compression, standard mammographic views were obtained. The mean variance in millimeters of the deployed clip measured perpendicular to the central axis of the breast on craniocaudal, mediolateral oblique, and mediolateral views was compared with these same measurements applied to the center of the calcifications for biopsy. The difference, to the nearest millimeter, between the mean measured separation of the lesion and the clip is presented in Table 3. Seventy percent of 11-gauge clips and 63% of 18-gauge clips were within 1 cm of the prebiopsy calcifications. The results of this same measurement method applied to fat necrosis calcifications are presented in Table 4. This variance in the position of a fixed calcification on sequential mammograms (mean, 5.4 mm; median, 4 mm) applied to the measured separation of the clip and target lesion resulted in adjusted measurements more closely approximating those made using stereotactic coordinates in the compressed breast (Table 2). The data in Table 3, however, establish the comparability of placement accuracy for the 11- and 18-gauge clips.
|
|
The data in Table 4 allow the assessment of movement of each clip within the breast over time. If adjusted for the baseline variation in location of fat necrosis calcifications, these data suggest that little, if any, movement of either clip occurs within the breast over the intervals observed (6–34 months). The variance in measured position of fat necrosis calcifications on sequential mammograms is dependent on examination technique and breast size. We assessed and recorded breast size in each of the 123 screening patients with measured fat necrosis calcifications. Mean and median variances in position of these identical foci for small breasts were 1 mm for both values, whereas for breasts of medium size, these values were approximately 4 mm. However, for patients with breasts that were considered large, the mean variation in position of identical calcifications between examinations was 12 mm and the median variance was 11 mm. These results indicate that identical mammographic features are more likely to be consistently displayed in the same location in small breasts, whereas the effect of positional and technical differences between examinations in large breasts results in greater variation in the projected position of fat necrosis calcifications and, presumably, in the position of a deployed clip.
In 46 patients with atypical or malignant core biopsy results for whom surgical follow-up was available, the deployed clip was used as the target for wire localization in all cases. In 26 patients with 11-gauge clips and 20 with 18-gauge clips, the size of the target calcifications varied by less than 1 mm (mean, 5.7 mm; median, 5 mm; range, 2–21 mm). The mean time from core biopsy to surgery was similar (average, 2.4 weeks; range, 2–11 weeks). Specimen radiographs were available for review in 41 cases. The targeted clip was identified on the specimen radiograph in 23 of 24 of the 11-gauge cases and in 16 of 17 of the 18-gauge cases. Maximum size of the surgical specimen was 65 mm in the 11-gauge and 63 mm in the 14-gauge cases. Of 26 biopsies performed with the 11-gauge probe, ductal carcinoma in situ was diagnosed in 16 patients (62%), infiltrating ductal cancer in nine (35%), and atypical ductal hyperplasia in one (3%). In 20 biopsies performed with the 14-gauge probe, ductal carcinoma in situ was diagnosed in 13 patients (65%) and infiltrating ductal cancer in seven (35%).
Underestimation of disease (core biopsy diagnosis, ductal carcinoma in situ; surgical diagnosis, infiltrating ductal cancer) occurred in two 11-gauge and three 14-gauge biopsies. The atypical ductal hyperplasia case at core biopsy yielded only atypical ductal hyperplasia at surgery, and two 11-gauge and one 14-gauge core biopsies diagnosed as ductal carcinoma in situ showed no residual disease at surgery. The calcific clusters in these three cases were 2–4 mm, and no residual calcifications could be identified after core biopsy.
At our institution, the unit cost for an 11-gauge biopsy needle is $240 and that of the 14-gauge needle is $190. The 11-gauge deployed clip costs $80. Each 18-gauge deployed clip costs $0.41; an 18-gauge spinal needle costs $1.17. For each of the 11-gauge biopsies with clip deployment, the equipment cost for each procedure is $320. For each 14-gauge vacuum-assisted biopsy and 18-gauge clip deployment, the equipment cost is $191.58.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The most precise measurement of clip and lesion position can be achieved using stereotactic coordinates. This methodology was described by Liberman et al. [2] and was used in their study of 42 patients; this methodology was also used by Reynolds [3] in 94 patients and in 52 11-gauge clip deployments in this study. Using stereotactic coordinates, Tesolin and Curpen, reporting on 3,000 18-gauge clip deployments, found a mean deviation of 3.3 mm between the clip and target lesion (Tesolin M and Curpen NB, Radiologic Society of North America meeting, November 2001). This value is comparable to the 3.5-mm mean distance in our patients and is only slightly less than the 4.5- to 6-mm mean variance reported for 11-gauge deployed clips (Table 2).
These measurements indicate that the distance of either clip from the biopsy target is sufficiently small to allow accurate hookwire localization if further tissue sampling is needed. These stereotactic coordinates, however, are obtained with the breast in compression. As Liberman et al. [2] correctly describe, removing the breast from compression results in an "accordion effect," which tends to increase the clip-to-target distance in the direction of compression (z-axis). Assessing clip–target separation after release of compression, therefore, requires the use of a method to measure their relative positions on standard mammograms. Rosen and Vo [4] used superimposition of nipple, pectoral muscle, and other parenchymal landmarks on mammograms before and after biopsy to measure clip-to-lesion distances parallel to the plane of compression. Burbank and Forcier [1] and Tesolin and Curpen (Tesolin M and Curpen NB, Radiologic Society of North America meeting, November 2001) each used a mask technique in which the target lesion and other breast features were transferred from prebiopsy mammograms to clear film that was then superimposed on the image showing the clip. In an effort to establish a set of control measurements for this technique, Tesolin and Curpen identified 108 and Burbank and Forcier 22 benign breast lesions on sequential mammograms. They measured the variation in position of these lesions between studies to establish baseline variability. For the 22 cases described by Burbank and Forcier, the mean was 8 mm, and for the 108 cases described by Tesolin and Curpen, a mean of 7.7 mm was measured. All the benign lesions in the study by Burbank and Forcier were biopsy-proven and measured 1 cm or more. The size of the control lesions in the study by Tesolin and Curpen was not described.
Because all biopsies were performed using needle entry parallel to the chest wall and prior investigators have found the largest clip–lesion separation in the z-axis [2, 4], we elected to measure both target lesion and clip position in this plane, which is perpendicular to a horizontal line drawn from nipple to chest wall (the central axis or "equator" of the breast). We sought then to establish a baseline variability for this measurement method by applying it to identical fat necrosis calcifications of similar size to the core-biopsied lesions. These control values were related to breast size in our fat necrosis cases. Neither we nor prior investigators had, however, recorded breast size as a variable in clip placement cases. Our fat necrosis measurements suggest that as breast size increases, the position of any fixed point is subjected to greater variability on sequential mammograms.
We anticipate that our biopsy patients include relatively few women with small breasts because small breasts, in our practice, are frequently deemed unsuitable for stereotactic sampling for positional and technical reasons.
Rosen and Vo [4] assessed the position of 111 clips placed through 11-gauge probes using the superimposition-of-mammograms technique and found that 28% of the clips were located more than 1 cm from the target on at least one mammographic projection. Fourteen percent of the clips were 10–20 mm off-target, 6% were 20–30 mm from the target lesion, and 7% were located more than 3 cm from the biopsied lesion. The largest variances occurred parallel to the plane of compression (z-axis) and were attributed to the accordion effect, which occurs as the breast is released from compression. Their study recorded the greatest off-target distance in any plane. In our patients, as in the study by Burbank and Forcier [1], the average of off-target clip measurements on craniocaudal, mediolateral oblique, and mediolateral projections was used. Because we did not routinely record the plane of applied compression in our cases, our results are more comparable to those reported by Burbank and Forcier than to those reported by Rosen and Vo. Seven percent of deployed clips in our study were more than 24 mm from the target lesion. Three cases with the largest mean clip-to-target separation were recognized immediately on viewing the postbiopsy mammograms. In two of these cases, a clip had been deployed through an 18-gauge needle while the biopsy site was actively bleeding and the clip appeared to have been "washed out" along the blood-filled needle tract. In the other 18-gauge case with the widest separation between clip and lesion, the breast tissue was nearly entirely fat and separation occurred in the plane of compression (z-axis). One instance of a "disappearing clip" has been described [5]. This may have been related to bleeding at the biopsy site.
The apparent migration of an 11-gauge clip described in a single case report [6] is considered a relatively rare occurrence. In that case, the published mammograms suggest a breast of relatively fatty density. The 11-gauge clip is designed to adhere to tissue at the biopsy site, but whether it does so consistently has not been shown. The 18-gauge clip remains unattached to breast tissue after deployment. The lower tissue resistance of a fatty breast may allow either clip to migrate more readily.
A study recently reported in the surgical literature [7] describes 11-gauge deployed clips as having migrated (mean, 13.5 mm) in 93 patients, with migration being more than 20 mm in 21.5% of the cases. This investigation used a geometric measurement technique but failed to apply this to any control cases or to account for any technical differences between mammographic examinations. The "migration" described, therefore, may be more apparent than real. Development of a reproducible and accurate method for measurement of clip position is needed in order to establish the frequency and extent of clip migration. The measurements that we applied to fat necrosis calcifications on sequential screening mammograms suggest that a more extensive investigation of variability in apparent position of native calcifications in breast tissue as a function of technical and positional changes between examinations is indicated. These results should be applied to any future investigations of apparent clip migration.
The large size of breast tissue samples retrieved at stereotactic core needle biopsy with the 11-gauge probe has encouraged the wide use of the 11-gauge vacuum-assisted biopsy probe. Use of the 11-gauge probe versus the 14-gauge probe increased 190% from 1998 to 2001, according to a leading supplier of these instruments (Randall D, personal communication).
Although the 11-gauge extracted cores are larger, some evidence suggests that their diagnostic yield is not greater than that of 14-gauge vacuum-assisted cores. Jackman et al. [8] reported a ductal carcinoma in situ underestimation rate of 11% for both techniques (14-gauge biopsies, 38/348; 11-gauge biopsies, 69/605) in a large multiinstitution study. The important variable for diagnostic yield in that study was the number rather than the size of cores obtained. Brenner [9] suggests that these equal underestimation rates may be important for economic reasons because the 11-gauge probe is a more expensive instrument than the 14-gauge needle.
In our practice, equipment cost for a 14-gauge probe and 18-gauge clip deployment is 40% less than the 11-gauge biopsy clip system, but use of the 11-gauge biopsy clip system requires slightly more radiologist time. The one radiologist using the 18-gauge clip was able to prepare and deploy it in approximately 4 min, whereas the 11-gauge clip in most cases required half this time. These factors will vary among different operators and in different practice settings. That our experience with the 18-gauge clip deployment was restricted to only one radiologist who used the 14-gauge probe for all biopsies represents a limitation of our study.
We recognize the usefulness of the 18-gauge clip deployment technique for the marking of a sonographically guided breast biopsy site. We found this technique particularly helpful in correlating mammographic and sonographic findings [10] and in marking tumors to be initially treated with neoadjuvant chemotherapy.
Although approved for use in human subjects as a vascular ligating clip, deployment of the 18-gauge clip at a breast biopsy site represents an off-label use, so appropriate informed consent should be obtained before performing this tissue-marking technique. A substantially equivalent titanium clip more recently developed (Horizon ligating clip, Weck Closure Systems) has obtained Food and Drug Administration market clearance for both vascular ligation and tissue-marking applications (Young B, personal communication).
Acknowledgments
We thank Laura Liberman for manuscript review, Lori Greene and Bob Timlin
for data extraction and analysis, 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:
|
|
 |
|
  K. Calhoun, A. Giuliano, and R. J. Brenner Intraoperative Loss of Core Biopsy Clips: Clinical Implications Am. J. Roentgenol., March 1, 2008; 190(3): W196 - W200. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. R. Margolin, L. Kaufman, R. P. Jacobs, S. R. Denny, and J. D. Schrumpf Stereotactic Core Breast Biopsy of Malignant Calcifications: Diagnostic Yield of Cores with and Cores without Calcifications on Specimen Radiographs Radiology, October 1, 2004; 233(1): 251 - 254. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
