| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Tristán Associates Harrisburg, PA 17111
In a recent technical innovation article, Sanders et al. [1] described their method for performing preoperative breast needle localization on a dedicated prone stereotactic table. Unfortunately, their method is neither technically elegant nor innovative. They seem to have taken the standard stereotactic needle localization technique widely taught in residency programs, continuing medical education courses, and manufacturers' training courses and to have simply subtracted an important portion of the procedure—the entire stereotactic calculation of lesion position and verification of needle placement—without adequately justifying their reasons for doing so. In addition, they made claims regarding their technique that are overreaching at best and unsubstantiated at worst.
Performing needle localization on a dedicated prone stereotactic biopsy table, rather than on a conventional mammography unit, has four advantages. First, the prone table makes fainting virtually impossible, and the entire procedure is conducted out of sight of the patient. Second, the digital acquisition and display of images speed the procedure considerably. These two advantages are shared by all procedures performed on a stereotactic table, whether with the technique described by Sanders et al. [1] or not. Third, the combination of the stereotactic calculations and the computer-controlled stage permits precise needle placement. Once the appropriate coordinates have been locked in, the front and back needle guides provide a foolproof guidance system for placing the needle, permitting skewering of even tiny targets. Fourth, stereotactic imaging provides nearly instant feedback, permitting verification of placement of the needle tip in three dimensions relative to the target before compression is released. By skipping the stereotactic portion of the needle localization procedure, Sanders et al. forego these last two advantages, in essence dumbing down a reliable and accurate procedure.
Furthermore, Sanders et al. [1] claim to have solved the problem of the accordion effect. I am not convinced they have. The accordion effect, which results in an error in the depth of placement of the localization needle, is an inevitable by-product of any needle localization procedure (whether performed on a conventional mammographic unit or on a dedicated prone stereotactic table) when compression, needle insertion, and the X-ray beam all travel in the same direction. The initial needle placement into the compressed breast may indeed be accurate, but when compression is released and applied in the orthogonal direction, the final position of the needle tip may be substantially different. Typically the needle withdraws to a variable degree. Kopans et al. [2] identified and addressed this problem in their seminal description of needle localization and placed the spring hookwire in the breast only after the appropriate depth had been adjusted and verified on an orthogonal view. Jackson [3] dispensed with any attempt to calculate the appropriate depth beforehand, instead advocating placing the needle as far as possible into the breast so that it can be withdrawn as needed on the basis of the orthogonal view; only then is the spring hookwire deployed.
Similarly, using the technique of Sanders et al. [1], one is forced to guess how deep to insert the needle. The single nonstereotactic image obtained after needle placement (figure 1B) does not confirm accurate needle placement. Only when the target, hub, and shaft of the needle are completely superimposed on one another does that single view confirm precise placement, and then only in two dimensions—depth cannot be depicted on a single mammogram. Anything else is largely guesswork. By contrast, using the full capabilities of the stereotactic table, one can choose to insert the needle so that the tip is, for example, exactly 1 cm beyond the lesion, and one can verify the position in three dimensions by stereotactic imaging before compression is released. Obviously, depth must still be verified by orthogonal imaging.
The only methods of which I am aware for avoiding the accordion effect during a mammographically guided needle localization are the freehand method (which has many problems of its own) or the lateral arm on a prone table (Mammotest; Fischer Imaging, Denver, CO). The stereotactic table (Stereoguide; Lorad/Trex, Danbury, CT), used by the authors has no such capability. But the authors report, "Ninety-degree orthogonal views are obtained, which confirm the relationship of the needle tip to the lesion without releasing compression" (my italics). They give no details clarifying how they accomplish such a feat, which is beyond the capabilities of the machine.
In summary, Sanders et al. [1] claim to have eliminated the accordion effect when in fact, they have not. Most of the other major advantages claimed by the authors relate not to their technique in particular but rather to all needle procedures performed on dedicated prone stereotactic tables. The authors would do well to scale back the overreaching claims for their technique. Better yet, they should take full advantage of the high-tech capabilities of their stereotactic table when performing needle localizations.
References
St. Barnabas Health Care System Livingston, NJ 07039
Note.—In the original article, the product Beekley O-Spots, Bristol,
CT, was incorrectly referred to as Beekley Mole Markers.
The stereotactic table (Stereoguide; Loard/Trex, Danbury, CT) is a versatile piece of equipment and may accommodate to its user preferences. Performing needle localizations with the traditional orthogonal technique on the Lorad table does not "dumb down" the procedure at all. Our technical note addresses the obvious advantages of using the Lorad table, which are reiterated by Dr. Guenin in his comments. We certainly agree that using the stereo pair to localize the lesion is quite precise; however, Dr. Guenin is not correct in stating that stereo imaging permits verification of the needle tip in three dimensions relative to the target. In fact, a true measure of distance is not possible unless the X-ray beam is 90° to the distance to be measured; otherwise, there is a fore-shortening effect. He is correct in his statement that the needle may be placed exactly 1 cm beyond the midpoint of the lesion with the breast in compression because he has dialed the needle to that position. The stereo pair subsequently obtained cannot "verify" any distance at all.
Guenin, however, misses the whole point of the technical note: the position of the needle tip with respect to the lesion with the breast in compression does not always reflect the final position of the needle or wire tip with respect to the lesion when the compression is released. This difference is due to the "accordion effect." The "accordion effect" refers to the migration of a needle or wire tip due to the change in its position in the breast and with respect to a mammographic lesion after the breast is released from compression. In any traditional orthogonal needle localization, the needle is placed while the breast is in compression. The breast is then released from compression, and the orthogonal view is obtained. Only then is the wire placed. The accordion effect, therefore, in the traditional biplane technique, only displaces the needle, which may be easily adjusted in the orthogonal position before wire placement. In the traditional orthogonal technique, the accordion effect, therefore, does not adversely affect the placement of the wire. In contrast, if the wire is placed stereotactically before the release of the breast from compression, its position is most certainly affected by this phenomenon, as we and others have experienced. Again, this problem is the major flaw in performing localizations stereotactically. That distinct disadvantage is inherent in the manufacturer's suggestion that during a stereotactically guided needle localization procedure, the needle be advanced in the range of 5-15 mm beyond the lesion. The manufacturer does not promise and cannot guarantee that using its guidelines, the stiffener will be centered on the lesion. This flaw of stereotactic needle placement is stated by Jackson and Bassett [2]—the stereotactic method does not allow biplane confirmation of needle placement, which is only critical because of the accordion effect. If there were no accordion effect, one could rely on the dial-in depth to assure a distance of 2 cm beyond the lesion.
The biplane technique requires a minimum of four images (1 straight-on view to localize, 1 straight-on view with the needle in place, 1 orthogonal view to document depth, and a final orthogonal view to document the wire in place). With the stereotactic technique, a minimum of six images are required (1 straight-on to localize, a pair to obtain coordinates, a second pair [purportedly] to show needle tip position, and then a final orthogonal view to document wire placement). In addition, only one needle guide is used during an orthogonal technique, whereas two guides are used during a localization performed stereotactically.
In our experience, using the traditional needle localization method, we have learned to gauge with accuracy how deep to place the needle (knowing in advance where the lesion is in the breast and having chosen the appropriate needle length), but in truth, we often adjust the depth of the needle, requiring a total of five images in most patients, still fewer than the minimum of six images required stereotactically.
After our technical note was published, our Lorad representative asked for a copy to distribute to clients who were concerned with the shortcomings of performing the needle localization procedure with the stereotactic method.
As the chief of breast intervention in my hospital, I (L.M.S.) have the responsibility of teaching residents breast intervention techniques. I routinely teach both needle localization techniques. The orthogonal approach works best for us, as mammographers, and it also work for our surgeons.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
