| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 All authors: Department of Radiology, Mayo Clinic, 201 1st St. SW, Rochester, MN 55905.
Received October 17, 2003; accepted after revision February 16, 2004.
Address correspondence to D. M. Lanners.
OBJECTIVE. The purpose of our study was to assess the MRI risk factors for the patient after an unusual case of inadvertent deployment of the mounting sheath of a metallic clip used at stereotactic breast biopsy.
MATERIALS AND METHODS. We evaluated the materials for ferromagnetic properties, heating, and artifacts.
RESULTS. Our analysis showed significant deflection and prominent susceptibility artifacts of the sheath at 1.5 T, although the clip itself showed no deflection and only minimal artifact. Our study shows that a device and its delivery apparatus may have significantly different ferromagnetic properties.
CONCLUSION. In case of inadvertent deployment of the sheath, a delay of 6–8 weeks before MRI is recommended as a conservative approach to ensure tissue ingrowth and to minimize the chance of harm to the patient. In the case of metallic clips used for breast biopsy, caution is warranted when a portion of the device unintended for placement is introduced into the breast.
The vacuum-assisted technique for percutaneous breast biopsy is widely used, often to evaluate small (< 1 cm) mammographic abnormalities such as microcalcifications [1]. Increasingly, metallic clips are being deployed at the conclusion of these procedures to mark the location of lesions that may not be visible after the biopsy. Clips are necessary to mark the site should future surgery be required. Several retrospective studies have evaluated the accuracy of placement of these markers and show that the clips are generally accurately placed but may vary from the biopsy site immediately after placement [1–3]. Migration of these metallic clips within the breast has also been reported [3–5]. If no further surgery is required, the marker may be left in the breast indefinitely. The clips, many of which are made of stainless steel, have been shown to be safe for MRI, with only minimal artifacts occurring at field strengths up to 1.5 T [6].
During deployment of a commonly used clip, the Biopsys 1 Micromark II (Ethicon Endo-Surgery), the clip is released with vacuum assistance from its mounting, which consists of plastic tubing with a metallic sheath mounted on a wire (Figs. 1A and 1B). We report an unusual case in which the metallic sheath was deployed inadvertently along with the marker at the end of a routine stereotactic biopsy procedure and our subsequent MRI safety analysis.
|
|
A 69-year-old woman presented at our institution for annual screening mammography. The examination showed a new area of microcalcifications in the left breast. The patient underwent stereotactic biopsy in the usual fashion. Because most of the microcalcifications had been removed during the biopsy procedure, the decision was made to deploy a metallic clip (Biopsys 1 Micromark II) at the biopsy site should future surgery be required. A mammogram obtained in the craniocaudad and mediolateral projections after the procedure showed the device to be in good position but with the metallic sheath still attached to the clip (Figs. 2A and 2B). Biopsy results showed a fibroadenoma and proliferative fibrocystic change, and no further surgery was required. The patient was offered the option of having the clip and sheath surgically removed but declined. Screening mammography performed 1 year later showed the clip and sheath still attached with no migration or change in position within the patient's breast.
|
|
Correspondence with the manufacturer (Ethicon Endo-Surgery) regarding the mounting device indicated that the device was composed of the same stainless steel as the clip itself, but no data regarding MRI compatibility of the mount (sheath and wire) were available (Itani T, personal communication). For this reason, testing of the sheath, mounting wire, and clip for ferromagnetic properties, temperature, and metallic artifacts was performed.
Materials and Methods
A new, undeployed clip was obtained for the purpose of testing (Figs. 1A and 1B). The metallic clip, sheath, and wire were separated from each other and the plastic portion of the mount. Each component was individually tested for ferromagnetic properties (deflection), temperature, and image artifacts under differing scanning conditions.
Ferromagnetic Properties
The three parts of the device were tested for attraction by first using a
handheld magnet (
180 G). Qualitative testing was conducted by placing the
individual items in an enclosed plastic case and placing them in the bore of a
1.5-T MRI system to visually observe any deflection in air in the magnet. The
items were also tested by filling the plastic case with water and then with a
material of increased viscosity (mineral oil) to evaluate deflection in the
MRI environment. All components were quantitatively evaluated using a
string-based test to quantitatively assess the magnitude of deflection and
force [7,
8].
Temperature Measurements
Using a fiberoptic thermometer that is unaffected by radiofrequency and
magnetic fields (3100 Fluoroptic Thermometer, Luxon), we measured temperatures
of the clip, sheath, and wire before and immediately after MRI. A 2D fast
spin-echo sequence (TR/TE, 2,000/17; echo-train length, 12; number of
excitations, 2) was performed with a 512 x 256 acquisition matrix, 24-cm
field of view, and 3-mm slice thickness with a 1.5-mm interslice gap.
Radiofrequency power level was indicated by a transmit gain of 190 (0.1 dB;
these are relative gain units), which we estimate to yield approximately 7 kW
(16 kW maximum, –1 dB, 20% line loss, 31% coil loss). Each component was
placed at the center of an 80-mL phantom composed of 0.05 g/L of MnCl and 2.5
g/L of NaCl with temperature probes located within 1 mm of the object.
Artifact Analysis
Three imaging techniques were used for scanning the clip, sheath, and wire,
as follows: spin-echo (500/20; slice thickness, 1 mm; scanning time, 2 min 16
sec), fast spin-echo (2,000/17; echo-train length, 12; slice thickness, 1.5
mm; scanning time, 48 sec), and gradient-refocused echo (500/13.4; flip angle,
30°; slice thickness, 1.5 mm; scanning time, 2 min 12 sec) sequences. All
acquisitions were 2D imaging sequences in the sagittal plane using a bandwidth
of 32 kHz; field of view, 10 cm; skip, 0.5 mm; slices, 7; matrix, 512 x
256; excitations, 1; and frequency encoding along the anterior to posterior
direction.
The metal components of the device were taped to the inside of an 80-mL plastic container filled with the previously described solution and imaged. The resulting artifacts varied in severity at the location of the clip, sheath, and wire, as shown in Figures 3A, 3B, 3C, and 3D. Quantitatively, the artifact was measured as a cylindric region of signal loss, as shown in Figures 4A, 4B, and 4C.
|
|
|
|
|
|
|
Results
The ferromagnetic property analysis (Table 1) showed no deflection of the clip itself. In contrast, the sheath and wire showed significant movement in all the media tested. Quantitative string testing of the sheath and wire showed no deflection of the clip itself but marked deflection of both the sheath (62° ± 2°) and mounting wire (89° ± 2°) (Table 1). Temperature analysis (Table 2) showed no change in temperature from the prescanning baseline for any of the three components. Significant differences were seen in artifacts caused by the different components (Figs. 3A, 3B, 3C, and 3D). The clip itself caused little distortion, but the sheath and wire caused much more local signal loss. Quantitative measurements of artifact radii from axial images (Table 3) were consistent with the observed distortion (Figs. 4A, 4B, and 4C).
|
|
|
Discussion
Previous investigators have performed ex vivo testing of the metallic
marking clips used for stereotactic biopsy, and these clips have been deemed
to be no risk for patients undergoing MRI examinations at 1.5 T or less
[6,
9]. Our results regarding the
clip itself are consistent with these findings, showing no heating or
deflection at 1.5 T and minimal local artifacts. With the inadvertent
deployment of the metallic sheath of the delivery apparatus along with the
clip, further evaluation of the magnetic properties of this object was
warranted. Although the manufacturer had indicated that the material used for
the delivery device was the same as the clip itself, our tests show the
delivery apparatus to be markedly ferromagnetic, whereas the clip is not. This
difference may be due to manufacturing processes that can affect the
ferromagnetic properties of stainless steel, and it prompted further analysis.
Although no heating was found, significant artifacts were noted at the sheath
location independent of acquisition type. The string test yields a measured
angle indicating the force-to-weight ratio (tan 
= force / weight).
Because the measured angle of the sheath (with error bars) is in a range that
the force-to-weight ratio changes dramatically (1.73 
60° to 2.05
64°), it is important to treat this conservatively. The
force-to-weight ratio may be 2.0 or above, which has been reported to be a
critical threshold [10].
In general, the risk of minimal movement of ferromagnetic materials in the body is critically dependent on the location of the material or device. Shrapnel in the leg is bothersome because it causes local artifacts that may limit the diagnostic value of an MRI study of that area, but MRI of the same material located in the orbit or lung is of considerably more risk. In our patient, the material is located in the breast. Although rare reports exist of extramammary migration of localization wires used for surgical biopsy [11, 12], to our knowledge migration of percutaneously placed localization clips outside the breast has not been reported, suggesting that the degree and extent of movement of these objects are limited [2, 3, 5, 13]. In general, such material is considered safe for MRI after the process of tissue ingrowth has occurred [9, 14, 15]. Tissue ingrowth has been variously said to occur from 6 to 8 weeks after the implantation of biomedical implants such as inferior vena cava filters or coronary artery stents [15]. In our patient, the clip with the sheath remained in place 1 year after the original biopsy, and we can assume that the patient will not come to harm if MRI is required. In the rare instance such as ours in which the metallic portion of the delivery device is deployed along with the clip, a conservative approach would be warranted, with no MRI performed in the first 2 months after placement because of the strong deflection of the metallic sheath in the static magnetic field. Because of the ferromagnetic character of the material used for the sheath and wire, metal artifacts are much more significant and potentially obscure pathology in the image. The clip itself causes little artifact, with only minimal local distortion. The metallic sheath showed much more severe local signal loss that could potentially confound interpretation of this region on MRI. These artifacts are most exaggerated on gradient-echo imaging, a mainstay of contrast-enhanced breast MRI [16, 17]. Alternative imaging sequences, such as fast and conventional spin echo with or without metallic artifact reduction, may be used to improve quality in the event that this patient might require diagnostic MRI of the affected breast [18, 19].
In conclusion, the clip intended for marking after the stereotactic biopsy is MRI-safe at 1.5 T, with no deflection and no measurable change in temperature. Artifacts related to the clip, even on gradient-echo imaging, are limited. In contrast, the sheath and wire of the mounting apparatus both showed ferromagnetic properties differing from those of the clip, with significant deflection at 1.5 T and prominent susceptibility artifacts. With this in mind, it remains important to assess objects for ferromagnetic properties. Because of the potential for movement of the mounting device in the breast, a conservative approach to MRI after the procedure is recommended in cases such as ours. A minimum delay of 6–8 weeks should ensure tissue in-growth and minimize the chance of any potential harm to the patient in cases of inadvertent deployment of the mounting device with the clip after percutaneous breast biopsy.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
