DOI: 10.1148/radiol.2323031128
(Radiology 2004;232:911-914.)
© RSNA, 2004
MR-guided Vacuum-assisted Breast Biopsy: Accuracy of Targeting and Success in Sampling in a Phantom Model1
Constance D. Lehman, MD, PhD and
Taro Aikawa, MD
1 From the Department of Radiology, University of Washington Medical Center, Seattle, Wash (C.D.L.); Seattle Cancer Care Alliance, 825 Eastlake Ave E, G4–830, Seattle, WA 98109-1023 (C.D.L.); and Department of Radiology, Columbia University, New York, NY (T.A.). From the 2002 RSNA scientific assembly. Received July 21, 2003; revision requested October 3; revision received November 10; accepted January 12, 2004. Supported by an unrestricted grant from Ethicon Endo-Surgery, Cincinnati, Ohio, with full control by the authors for experimentation and data reporting. Address correspondence to C.D.L. (e-mail: lehman@seattlecca.org).
 |
ABSTRACT
|
|---|
An 11-gauge MR-compatible system was designed for use in magnetic resonance vacuum-assisted breast biopsy. The system uses a detachable needle with minimal artifact to allow imaging after placement and before biopsy to confirm lesion location. A phantom study involving 16 biopsies of lesions smaller than 10 mm was conducted to assess the performance of this system. Fifteen (94%) of 16 biopsies resulted in successful lesion sampling. Eighty-five percent of all core samples contained specimen materials targeted lesion material. Further research to evaluate the efficiency, accuracy, and safety of this system in a patient population is recommended.
© RSNA, 2004
Index terms: Breast, biopsy, 00.1261 • Breast, MR, 00.1214 • Magnetic resonance (MR), experimental studies • Magnetic resonance (MR), guidance, 00.1214, 00.1261 • Phantoms
|
INTRODUCTION
|
|---|
Vacuum-assisted breast biopsy has proved to be more effective than 14-gauge automated gun biopsy (1,2), but it has been limited to lesions that are visible with mammography or ultrasonography (US). With the increasing use of magnetic resonance (MR) in breast imaging, however, a need exists for accurate and successful tissue sampling of lesions seen only on MR images. The advantages of vacuum-assisted breast biopsy include single probe insertion, directional sampling, larger tissue samples, and the ability to obtain more tissue in less time. In addition, when a lesion that is seen on an MR image is sampled for biopsy, clip placement enables future surgical interventions to be performed with mammographic or US guidance.
Preliminary reports indicate that MR-guided vacuum-assisted breast biopsy is technically feasible and clinically applicable to lesions discovered with MR imaging (3). Studies conducted at five European sites demonstrated successful biopsy in 98% of 341 lesions (4). The methods describe MR-guided vacuum-assisted breast biopsy with a coaxial or substitute needle configuration performed outside a closed magnet with or without breast compression. An 11-gauge MR-compatible system with an aiming device mounted to the patient table was used (5). Needle artifacts, tissue shift during probe insertion, and washout of contrast material during the procedure limited applicability to lesions larger than 10 mm in diameter (3).
To address these concerns, an experimental device has been developed that facilitates vacuum-assisted breast biopsy performed with MR guidance by using a closed MR system, without the need for a coaxial or substitute needle. The system uses a detachable needle that produces few artifacts, thus allowing imaging after clip placement and before biopsy to confirm location prior to washout of contrast material. The purpose of our study was to evaluate the accuracy of targeting, success in sampling, and feasibility of fiducial marker placement by using this method of MR-guided vacuum-assisted breast biopsy in a phantom model.
|
Materials and Methods
|
|---|
Vacuum-assisted Breast Biopsy Unit
The vacuum-assisted breast biopsy unit used was an 11-gauge MR-compatible Mammotome system (Ethicon Endo-Surgery, Cincinnati, Ohio) that was available for experimental use only. System components include (a) a tissue immobilization and/or localization device that attaches to the breast coil and (b) a vacuum-assisted biopsy system that consists of a reusable control module and holster and a disposable probe, stylet, and fiducial marker. The holster is of the same design as that used for US-guided vacuum-assisted breast biopsy with the 11-gauge MR-compatible system. This holster can be inserted in a sterile sleeve or cleaned with an enzymatic detergent solution after each use. The probe has a detachable distal needle assembly that consists of a dual lumen plastic needle with a titanium metallic blade tip. This tip provides a sharper cutting edge than the previously available pyramidal tip. Near the tip of the device is a sample notch that can be rotated with a thumbwheel on the probe shaft. The plastic needle and thumbwheel are attached to a plastic housing that joins the probe to the remaining holster and the localization mechanism on the breast coil. This localization mechanism can be adapted to a variety of currently available open coils. Vacuum tubing and connectors provide a means to aspirate and inject fluids through the lower lumen of the probe. To prevent fluids from flowing back through the detached needle, a flexible stylet is inserted in the upper needle lumen.
The proximal portion of the probe consists of a nonferrous metallic rotating cutter tube, plastic gearing that is used to advance and rotate the cutter, and a knockout pin that is used to eject the tissue from the cutter after sampling. The reusable biopsy components consist of standard unmodified components, such as a control module, handheld holster assembly, remote keypad, and handheld software. An aluminum stand supports the cables (Fig 1).
Breast immobilization is provided by a lateral grid plate attached to the breast coil. A cartesian coordinate system is used to localize lesions in the breast phantom that are registered to fiducial markers on the lateral grid plate. A pedestal provides a means to insert the needle into the tissue, a slide base provides lateral motion, and a cross arm allows vertical motion and a means for probe attachment (Fig 2). All of these components are made of plastic.
Generic Description of How the Method Works
During the prebiopsy examination, the disposable probe is calibrated with the entire 11-gauge MR-compatible system and then removed. The distal needle assembly is detached from the holster and fixed to the localization mechanism on the open breast coil (MRI Devices, Waukesha, Wis). The stylet is inserted into the cutter lumen to prevent backflow of fluids through the needle. During this time, the 11-gauge MR-compatible system remains outside the MR imaging suite.
A vitamin E capsule serves as a fiducial marker and is placed near the proposed biopsy site. Three imaging sequences of the breast are performed: (a) a fast spin-echo T1-weighted examination of the entire breast in the sagittal plane (repetition time msec/echo time msec, 500/12), with a 4-mm section thickness and 1-mm skip before biopsy; (b) a fast spin-echo T1-weighted limited region of interest examination in the transverse plane (500/12), with five sections obtained at 4-mm intervals with the probe in position; and (c) a limited region of interest examination in the transverse plane after biopsy and clip placement.
The x, y, and z coordinates of the lesion and fiducial marker are recorded from the time of the first sagittal examination. The x coordinate refers to the superior-inferior position, the y coordinate refers to the anterior-posterior position, and the z coordinate refers to the medial-lateral value or the depth of needle penetration into the breast. The differences between the x, y, and z coordinates of the fiducial marker and lesion are calculated, and the guidance system is adjusted so that the needle guide is in line with the lesion in the superior-inferior and anterior-posterior planes.
When the suspicious lesion is localized, the needle is positioned at the calculated superior-inferior (x coordinate) and anterior-posterior (y coordinate) positions. A stop is provided to a calculated medial-lateral (z coordinate) position to enable the correct depth of needle penetration into the breast. The probe is then inserted so that the center of the sampling chamber is in line with the lesion. With the needle in the breast phantom, the patient is then shuttled into a closed-bore magnet to confirm the location of the needle. A limited transverse examination is performed through the region of the lesion prior to biopsy.
Once the needle location is confirmed and the patient is shuttled out of the imager, the 11-gauge MR-compatible system control module and the handheld holster assembly are positioned in the MR imaging suite outside the 5-G line. The stylet is removed, and the holster is attached. The control module and cabling provide suction to pull tissue into the sample notch and advance and rotate the cutter to obtain tissue samples. Control of the cutter position and suction is provided with buttons on the remote keypad.
Sampling is performed preferentially in the region of the lesion. For example, if the lesion is noted to be inferior to the probe on images obtained with the transverse sequence after probe insertion, samples are obtained from positions between the 4 and 8 o’clock positions. When tissue sampling is complete, the probe is retracted 5 mm, and a MicroMark tissue marker (Ethicon Endo-Surgery) is inserted through the needle assembly and deployed into the biopsy cavity. The holster is detached, which allows the needle assembly to remain in the tissue. The stylet is inserted into the needle, and the patient is shuttled into the imager to undergo postbiopsy confirmation imaging.
Phantom Study
In this study, 16 MR-guided vacuum-assisted biopsies were performed in turkey breast models with embedded phantom lesions. The lesions were 3.2–4.7 mm in diameter (mean, 4.0 mm ± 0.5 [standard deviation]) and were solid synthetic nodules made of gelatin mixed with vitamin E and blue food coloring. The nodule material consisted of 1 tablespoon of Knox gelatin (Kraft Foods, Northfield Ill), 20 cc of water, four drops of blue dye, and one large vitamin E capsule. This mixture was heated until boiling. It was then poured into a mold to create spheres with a 5-mm diameter. Imaging was performed with a 1.5-T Signa MR imager (GE Medical Systems, Milwaukee, Wis).
For each biopsy, x, y, and z coordinates of the lesion and fiducial marker were recorded from the initial sagittal sequence (Fig 3a, 3b). The probe was inserted, and the x, y, and z coordinates of the chamber center were recorded from the transverse sequence (Fig 3c). Six samples were obtained in the region of each lesion. Tweezers were used to extract each tissue sample from the device. The presence or absence of blue dye within each lesion was recorded. At the completion of the biopsy, the probe was pulled back 5 mm, and a tissue marker was placed. The probe was then removed, and a final postbiopsy limited transverse sequence was obtained through the lesion, as described previously (Fig 3d). The x, y, and z coordinates of the clip were recorded. All biopsies were performed by a radiologist (C.D.L.).
View larger version (91K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Biopsy and clip placement in phantom breast. (a) Sagittal MR image of phantom with compression grid and fiducial marker (arrow). (b) Sagittal MR image with a targeted lesion (arrow). (c) Transverse MR image with probe placement adjacent to lesion (arrow). (d) Transverse MR image confirms clip placement (arrow).
|
|
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Biopsy and clip placement in phantom breast. (a) Sagittal MR image of phantom with compression grid and fiducial marker (arrow). (b) Sagittal MR image with a targeted lesion (arrow). (c) Transverse MR image with probe placement adjacent to lesion (arrow). (d) Transverse MR image confirms clip placement (arrow).
|
|
View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. Biopsy and clip placement in phantom breast. (a) Sagittal MR image of phantom with compression grid and fiducial marker (arrow). (b) Sagittal MR image with a targeted lesion (arrow). (c) Transverse MR image with probe placement adjacent to lesion (arrow). (d) Transverse MR image confirms clip placement (arrow).
|
|
View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3d. Biopsy and clip placement in phantom breast. (a) Sagittal MR image of phantom with compression grid and fiducial marker (arrow). (b) Sagittal MR image with a targeted lesion (arrow). (c) Transverse MR image with probe placement adjacent to lesion (arrow). (d) Transverse MR image confirms clip placement (arrow).
|
|
Successful biopsy was defined as presence of blue lesion material within one of the core samples. Accuracy of targeting and clip placement was determined by the differences between x, y, and z coordinates of the lesion, chamber center, and clip. Individual core sample weights and the time to perform a biopsy (defined as the time from acquisition of the first image to the time of probe removal) were recorded.
Data Analysis
The percentage of successful biopsies was documented. Means and standard deviations of the differences in coordinates between targeted lesions and chamber centers and between targeted lesions and marker clips were calculated. The amount of time required to perform each examination was recorded, and the overall average time and range of times required to perform each biopsy were calculated.
|
Results
|
|---|
Fifteen of 16 biopsies resulted in successful samples. One biopsy was unsuccessful and was aborted because of probe failure. In this case, the needle was replaced, and subsequent biopsy resulted in a successful sample. In another case, six additional core biopsies were needed to acquire portions of the lesion. In this case, imaging revealed a large air gap around the inserted lesion. Eighty-five percent of all core samples contained specimen material. The means and standard deviations of the x, y, and z coordinate differences between targeted lesions and chamber centers were 2.9 mm ± 1.7, 3.7 mm ± 2.4, and 2.4 mm ± 3.8, respectively. The means and standard deviations of the x, y, and z coordinate differences between the targeted lesions and the clips were 3.6 mm ± 3.4, 2.5 mm ± 1.9, and 0.9 mm ± 2.1, respectively. The mean individual sample weight was 64 mg ± 22 (standard deviation). The average time required to perform a biopsy was 16 minutes 41 seconds; times ranged from 11 minutes 44 seconds to 22 minutes 38 seconds.
|
Discussion
|
|---|
Results from the MR-guided vacuum-assisted breast biopsy of phantom models without a coaxial or substitute needle suggest that the Mammotome system shows promise in the accurate targeting and sampling of lesions. In addition, it is feasible to place a marker at the biopsy site by using this technique. We used the MR-guided vacuum-assisted breast biopsy system described previously and were able to sample 15 of 16 targeted lesions within the phantoms. The mean sample weight was 64 mg. After successful sampling, we were able to place a site marker precisely at the biopsy location (mean, <4 mm from biopsy location). The procedure was efficient, with a mean time of less than 20 minutes.
The results cited above are comparable with those achieved by others. For example, Heywang-Kobrunner et al (7) reported that 54 of 55 MR-guided vacuum-assisted breast biopsy procedures (including 15 in lesions 
5 mm and 26 in lesions between 5 and 10 mm) were successful.
The mean weight of our core samples was 64 mg ± 22 (standard deviation). Berg et al (6) reported that they obtained average sample weights of 94.4 mg when they tested an 11-gauge vacuum-assisted breast biopsy system with turkey breast phantoms. The weights we measured may be less because of differences in phantom lesion materials.
As an added benefit, the Mammotome device uses a detachable needle with minimal artifact that allows imaging after needle placement and before biopsy. This allows for sampling of lesions smaller than 10 mm. The mean lesion size sampled was 4 mm (range, 3.2–4.7 mm).
There are limitations of this phantom study. While turkey breasts are commonly used as phantoms in breast biopsy work, a turkey breast does not have the consistency of muscle to fat composition that a human breast has. Tissue difference may affect the accuracy of clip marker placement and the success of biopsy in human tissue. Finally, the challenge of contrast material washout in lesions in humans was not evaluated in this phantom study. Results from this phantom study need to be replicated in clinical trials.
The experimental biopsy device used in this study has not been approved by the Food and Drug Administration. There is currently a vacuum-assisted breast biopsy device that is commercially available for use in the MR suite (ATEC System; Suros Surgical Systems, Indianapolis, Ind). The ATEC System is currently undergoing clinical trials for further evaluation of the accuracy of MR-guided vacuum-assisted breast biopsy in a clinical population.
|
ACKNOWLEDGMENTS
|
|---|
We thank Bonnie Thursten, CT, and John McCloskey, CT, for their contributions to the success of this project. We thank John Hibner and David Beck of Ethicon Endo-Surgery, Cincinnati, Ohio, for their expertise and assistance in producing this work. We thank Diane Guay for her assistance in project development and administration. We thank Sue Peacock, MSc, and Tamara Fernando for their assistance in manuscript preparation.
|
FOOTNOTES
|
|---|
Author contributions: Guarantor of integrity of entire study, C.D.L.; study concepts and design, C.D.L.; literature research, C.D.L., T.A.; experimental studies, C.D.L., T.A.; data acquisition and analysis/interpretation, C.D.L., T.A.; manuscript preparation, C.D.L., T.A.; manuscript definition of intellectual content, C.D.L.; manuscript editing, revision/review, and final version approval, C.D.L., T.A.
|
REFERENCES
|
|---|
- Burbank F. Stereotactic breast biopsy of atypical ductal hyperplasia and ductal carcinoma in situ lesions: improved accuracy with directional, vacuum-assisted biopsy. Radiology 1997; 202:843-847.[Abstract/Free Full Text]
- Philpotts LE, Shaheen NA, Carter D, Lange RC, Lee CH. Comparison of rebiopsy rates after stereotactic core needle biopsy of the breast with 11-gauge vacuum suction probe versus 14-gauge needle and automatic gun. AJR Am J Roentgenol 1999; 172:683-687.[Abstract/Free Full Text]
- Heywang-Kobrunner SH, Heinig A, Pickuth D, Alberich T, Spielmann RP. Interventional MRI of the breast: lesion localisation and biopsy. Eur Radiol 2000; 10:36-45.[CrossRef][Medline]
- Perlet C, Heinig A, Prat X, et al. Multicenter study for the evaluation of a dedicated biopsy device for MR-guided vacuum biopsy of the breast. Eur Radiol 2002; 12:1463-1470.[CrossRef][Medline]
- Viehweg P, Heinig A, Amaya B, Alberich T, Laniado M, Heywang-Kobrunner SH. MR-guided interventional breast procedures considering vacuum biopsy in particular. Eur J Radiol 2002; 42:32-39.[CrossRef][Medline]
- Berg WA, Krebs TL, Campassi C, Magder LS, Sun CJ. Evaluation of 14- and 11-gauge directional vacuum-assisted biopsy probes and 14-gauge biopsy guns in a breast parenchymal model. Radiology 1997; 205:203-208.[Abstract/Free Full Text]
- Heywang-Kobrunner SH, Heinig A, Schaumloffel U, et al. MR-guided percutaneous excisional and incisional biopsy of breast lesions. Eur Radiol 1999; 9:1656-1665.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
C. D. Lehman, E. R. DePeri, S. Peacock, M. D. McDonough, W. B. DeMartini, and J. Shook
Clinical Experience with MRI-Guided Vacuum-Assisted Breast Biopsy
Am. J. Roentgenol.,
June 1, 2005;
184(6):
1782 - 1787.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
