| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Developments |
1 From the Department of Radiology, Duke University Medical Center, Erwin Rd, Box 3808, Durham, NC 27710. From the 1998 RSNA scientific assembly. Received February 17, 2000; revision requested April 3; revision received June 2; accepted June 5. Address correspondence to R.C.N. (e-mail: nelso017@mc.duke.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
Index terms: Interventional procedures, technology • Phantoms • Ultrasound (US), guidance, **.129852, **.12986 • Ultrasound (US), technology
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
Although sonography is widely used in Europe and Asia for guiding percutaneous intervention, radiologists in the United States have not as readily adopted this technique. This is probably because of the wide availability of computed tomographic (CT) scanners in this country and because many residency and fellowship training programs have emphasized CT rather than sonographic guidance. However, the recent emphasis on cost containment and patient throughput, especially with high-end CT scanners, has tipped the scales toward the use of sonography, particularly for percutaneous biopsy (1,2). Furthermore, application of this technique has expanded from use in superficial and/or large lesions to use in deep and/or small lesions (3,4).
Despite the advances in sonographic technology, including the use of narrowly collimated ultrasound beams and attached needle guides, limitations continue to exist. One such limitation that both the novice and experienced interventionalist encounter is the indistinct acoustic interface between the needle tip and surrounding soft tissue. This limitation results in difficulty in finding and following the needle tip, particularly with the narrow-caliber needles (ie, 20 and 22 gauge) used for diagnostic tissue aspiration (5). A second limitation is that most attached needle guides require the needle to be passed at a specific or fixed angle relative to the transducer, as well as in the plane of imaging. The freehand technique may not have this particular problem, but it does require considerable practice and experience to master the two-handed coordination necessary for needle visualization and targeting. Furthermore, as with the use of attached needle guides, the freehand technique is essentially limited to an in-plane approach.
Several years ago, our abdominal interventional group decided to alter our practice of using primarily CT for guidance to using primarily sonography. As a result, we now perform the majority of our percutaneous biopsies in the abdomen and pelvis with sonographic guidance. Recently, the UltraGuide 1000 (UltraGuide, Tirat Hacarmel, Israel) was introduced to complement the currently used guidance techniques for interventional sonography and, especially, to enhance the freehand technique. This device provides graphically displayed guidance information obtained from small position sensors located within a weak magnetic field generated by the base unit. This design allows the operator to approach the lesion from any angle relative to the transducer, whether in plane or out of plane.
The purpose of this study was to objectively compare the results of using the device in a phantom with those of other commonly used techniques for sonographically guided biopsy and to subjectively evaluate use of the device in patients undergoing percutaneous biopsy and drainage.
|
Materials and Methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
|
|
The information is displayed as a graphic overlay on a flat-screen liquid crystal monitor, which carries the real-time sonographic image through a coaxial cable from the output port available on most sonographic systems (Fig 3). The flat screen can be easily tilted or rotated for viewing from the operator’s position and offers an additional viewing monitor to that of the sonographic unit for any secondary personnel assisting in the procedure. The graphic display (ie, color, thickness, and pattern of lines showing the actual and predicted needle path) can also be adjusted for individual operator preference prior to or during the intervention. Note that the device cannot be used to predict the needle path accurately when there is bending along the shaft. That is, narrow-caliber needles of 22 gauge or smaller do not work well with the device since they are flexible and are subject to considerable bending.
|
|
A commercially available breast gel phantom (Computerized Imaging Reference Systems, Norfolk, Va) 13 cm long x 10 cm wide x 8 cm deep was used; it contained several implanted lesions of various sizes at varying depths. The lesions ranged up to 11 mm in diameter and 7.5 cm in depth. All passes were performed in random order by the participants by using a computed sonographic unit (model 128; Acuson, Mountain View, Calif) with 20-gauge Crown needles (Medi-Tech/Boston Scientific, Natick, Mass).
The different techniques were randomized to reduce the possibility of an experience bias, whereby the last technique used has an advantage over the first technique due to experience with the device. Each participant performed four biopsies by using the following guidance methods: (a) a plastic needle guide attached to the side of the transducer alone, (b) the needle guide with benefit of the device, (c) the freehand technique alone, and (d) the freehand technique with benefit of the device. Targeting was considered successful when the needle tip was positioned well within the lesion, a placement that would likely yield adequate tissue if biopsy had actually been performed. For the phantom study, only the in-plane approach was used (Fig 5) to keep the experiment simple for the 18 operators being introduced to the device.
|
Clinical Study
After reviewing the results of the phantom study, we began to use the device in patients. For the clinical portion of the examination, only the freehand technique with the device was evaluated. Initiation of data collection for the clinical portion of the study was delayed for approximately 3 months so that all of the radiologists participating in the study could become familiar with the device. For a variety of reasons, the device was not used sequentially in every patient who underwent a sonographically guided procedure. Reasons for not using the device included the fact that for some procedures it was unnecessary (eg, a superficial or large lesion was involved), that there was an occasional equipment malfunction, that some faculty chose not to take the additional time (estimated time, 2 minutes) to set up the device, and that some faculty chose not to participate in the clinical portion of the study.
During 12 months (July 18, 1998, through July 26, 1999), 43 patients underwent a total of 83 needle placements during sonographically guided biopsy (n = 35) or drainage (n = 9) with use of this device. A total of five radiologists (including R.C.N., M.A.K., E.K.P., D.H.S.), all of whom were abdominal imaging faculty, participated in this portion of the study. All procedures were performed in a dedicated sonographic interventional suite by using a Logiq 700 system (GE Medical Systems, Milwaukee, Wis).
A total of 83 needle passes were performed during 44 procedures. This total included passes in 50 fine-needle aspirations in which 47 Franseen needles (Baxter Healthcare, Deerfield, Ill; 42 20-gauge and five 22-gauge) were used and in which three 18-gauge trocar needles (Cook, Bloomington, Ind) were used as the introducer needle in a coaxial approach. There were also 24 core-needle biopsies, all performed with Achieve needles (Allegiance Healthcare, McGaw Park, Ill), of which five were 14 gauge, four were 18 gauge, and 15 were 20 gauge. All nine fluid collections were initially accessed with an 18-gauge trocar needle, which was subsequently used for guide wire and catheter placement by using a modified Seldinger technique.
Both objective and subjective data were collected by having the operator fill out a data sheet immediately after the procedure. The objective data included that related to the lesion and biopsy hardware used during each procedure and included lesion location, needle type and gauge, and whether the biopsy was performed in plane or out of plane. The subjective data included the operator’s assessment of visualization of the needle tip, divergence of the needle tip, overall contribution of the UltraGuide 1000, and degree of difficulty. The first two parameters were chosen because they are key factors in determining the success of needle placement. The latter two factors were selected as key contributors toward technology assessment.
Needle tip visualization was based on a scoring system from 1 to 4 as follows: 1, visualization throughout the procedure; 2, visualization with minimal adjustment or manipulation; 3, visualization with major adjustment or manipulation; or 4, no visualization.
The degree of needle tip divergence, defined as deviation from the path predicted by the device, was scored on a scale from 1 to 3 as follows: 1, no divergence; 2, slight divergence; or 3, substantial divergence.
A subjective impression of the overall contribution of the device to the procedure was recorded as follows: 1, the device worked with a decrease in time, effort, or both; 2, the device worked, but there was no decrease in time, effort, or both; 3, the device worked with an increase in time, effort, or both; or 4, the device did not work. The reference for this subjective impression was based on the faculty’s experience with performing sonographically guided intervention by using an attached needle guide.
The degree of difficulty was also recorded for each procedure as follows: (a) straightforward; (b) difficult due to the nature (ie, difficult for the operator to visualize due to the echo features) or location (ie, deep, behind a rib, surrounded by blood vessels) of the lesion itself; or (c) difficult due to technical factors not related to the lesion (ie, patients who were uncooperative, patients who could not hold their breath, or patients in whom it was difficult to achieve pain control).
Statistical Analysis
The four guidance techniques used in the phantom study were compared by using a repeated-measures analysis of variance, or ANOVA, model. For the clinical study, the relationship between the technical difficulty of the procedure and the contribution of the device was analyzed by using the Cochran-Mantel-Haenszel test. A P value of less than .05 was considered to indicate a statistically significant difference.
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
|
Individual preference for each technique was ranked from 1 to 4, with 1 being the most favored technique and 4 being the least favored technique. The mean preference ranking was calculated and is shown in Table 1. The freehand technique with the device was the most highly preferred technique, followed closely by the use of the attached needle guide either with or without the device. Although there was no statistically significant difference among these three approaches, all were ranked significantly higher than the freehand technique alone (P < .001). While there was a general correlation between the operator’s preference ranking and the standardized times and number of passes, the flexibility of the freehand technique with the device was considered to be advantageous by many of the participants, even if it required slightly more effort to target a lesion.
Clinical Study
Data were collected for a total of 44 procedures in 43 patients (23 men, 20 women; age range, 29–81 years; mean age, 59 years). The types of procedures included biopsy of a focal mass in a solid organ in 24 patients, biopsy of a lymph node in eight patients, core-needle biopsy of hepatic parenchyma in three patients, and drainage of a fluid collection in nine patients. Biopsy of a focal lesion in a solid organ included the liver in 14 patients; the pancreas in four patients; the thyroid in three patients; and the kidney, spleen, and adnexa in one patient each. Sites for lymph node biopsy included a portacaval node in two patients, a mesenteric node in two patients, a retroperitoneal node in two patients, a groin node in one patient, and a lymph node for which the site was not specified in one patient.
The participants were asked to score the degree of needle tip visualization for every pass in each procedure. Data were recorded for a total of 83 needle passes (Table 2). The mean tip visualization score was 2.1 ± 0.9. Needle tip visualization was given a score of 1 (visualization throughout the procedure) in 22 (27%) passes, a score of 2 (visualization with minimal adjustment or manipulation) in 43 (52%), a score of 3 (visualization with major adjustment or manipulation) in 11 (13%), and a score of 4 (no visualization) in seven (8%).
|
A subjective impression of the overall contribution of the device was recorded for 43 procedures, again relative to the faculty’s experience with the use of an attached needle guide. For these procedures, a score of 1 (the device worked with a decrease in time, effort, or both) was assigned to 14 (33%); a score of 2 (the device worked, but there was no decrease in time, effort, or both) was assigned to 20 (47%); a score of 3 (the device worked with an increase in time, effort, or both) was assigned to 8 (19%); and a score of 4 (the device did not work) was assigned to one (2%).
A description of the overall degree of difficulty of the procedure was recorded for 44 procedures. Twenty-nine (66%) were considered to be straightforward, nine (20%) were considered difficult for technical reasons, and six (14%) were difficult due to the location of the lesion. To determine whether there was a correlation between the operator’s subjective impression of the overall contribution of the device to the procedure and the degree of difficulty, the mean subjective impression score, with the use of the four-point scale noted in the previous paragraph, was then determined according to the assessment of procedural difficulty. In those 29 procedures considered to be straightforward, the mean subjective impression score was 1.8 ± 0.7. In those 15 procedures considered difficult for whatever reason, the mean score was 2.1 ± 1.0. This difference, however, was not statistically significant (P = .245).
|
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
The opportunity cost for sonography is not nearly as substantial. As a result, CT is reserved at our institution for the minority of cases in which the lesion cannot be visualized at sonography or those in which there is a concern about transgressing the bowel. The ability of sonography to depict lesions is substantially enhanced by the application of transducer compression, which can considerably reduce the distance between the skin and the lesion; this technique not only improves depiction but also shortens the needle path. Fisher et al (4) reported a mean reduction in the distance from skin to retroperitoneal lymph nodes of 49% with the use of transducer compression.
The key to sonographically guided intervention is real-time needle tip visualization. At biopsy, tip visualization is critical, not only during needle placement but also during tissue acquisition. Furthermore, the best guidance techniques are those that have a relatively flat learning curve and the flexibility of allowing both in-plane and out-of-plane approaches.
There are three main techniques for performing needle placement with sonographic guidance. These include use of an attached needle guide, use of the freehand technique, and use of electronic guidance, such as that with the UltraGuide 1000. Each of these techniques has advantages and disadvantages, which will be detailed next.
The attached needle guide is a small plastic or metallic device that is attached to the transducer parallel to the plane of the ultrasound beam. This device is coupled with software provided by the manufacturer of the sonographic unit, which depicts a pair of parallel or diverging lines on the screen along the proposed needle tract. The software also provides depth measurements, from either the skin or the insertion point on the device. The advantages of this attachable device are that it is easy to learn to use, it is precise, it provides reasonable depiction of the tip, and it reduces needle bending. The disadvantages of this device include a slight increase in setup time and cost, fewer degrees of freedom in needle placement, limitations in the performance of steep angulation, and exclusion of out-of-plane needle placement as an option.
The freehand technique is the oldest sonographic guidance option, as it has no special hardware or software requirements. This technique is primarily used with in-plane imaging whereby the plane of the transducer and needle path are aligned manually by using eye-hand coordination. This technique has the advantages of unlimited degrees of freedom for an in-plane approach and the lack of additional setup time and cost. The disadvantages, however, include a steep learning curve, particularly for deep and sharply angulated targeting, exclusion of out-of-plane imaging as an option, and problems with needle bending. This technique is well suited to experienced interventionalists, but in our opinion it can be challenging for new interventionalists or for those individuals attempting to alter their practice from use of primarily CT guidance to use of primarily sonographic guidance.
Electronic guidance is the newest and most provocative technique for assistance with needle placement. With the UltraGuide 1000, a low–strength magnetic field is established in the region where biopsy is being performed, and position sensors are attached to the side of the transducer and to the needle shaft. With the use of operator-defined needle lengths, the proposed path and tip of the needle are demonstrated on a liquid crystal screen projected over the real-time image. The overlay is depicted in two planes, the imaging plane and a bird’s-eye view, to provide the operator with an overall three-dimensional concept of both lesion and needle tip orientation. The advantages of this technique include the ability to use innumerable needle approaches, both in plane and out of plane (up to 90°), and the ability to use either the freehand technique or an attached needle guide. The disadvantages include additional setup time and equipment cost, problems encountered with needle bending (especially with 22-gauge needles), and a moderately steep learning curve. Although the device used in this study did not depict color Doppler sonograms, the new version of the UltraGuide 1000 has red-green-blue capability as an option. In addition, placement of a position sensor on the shaft of a large Tru-Cut needle (Baxter Healthcare) (eg, 14-gauge) may at times hinder the spring-loaded mechanism of the needle.
In our practice, it is estimated that the device would be beneficial in approximately half of the sonographically guided biopsies of the abdomen and pelvis. In the remaining half, biopsy is performed in either superficial or large lesions in which needle placement is relatively easy.
The results of our phantom study suggest that the device performs in a fashion similar to that of the attached needle guide, although both techniques were substantially superior to the freehand technique. This was evident both in terms of the number of needle passes and in the overall time required for targeting.
When the clinical study of patients commenced, however, the faculty found targeting to be more challenging than in the phantom study. There were several reasons that the learning curve proved to be steeper in patients. The lesions tended to be deeper; therefore, needle bending caused more problems, particularly for longer and narrow-gauge (ie, 22-gauge) aspiration needles. Divergence was typically encountered not only when an attempt was made to puncture the skin but also when internal fascial planes were traversed. Furthermore, the sonographic machine used in the clinical portion of the study was newer and more technologically advanced (Logiq 700; GE Medical Systems) than that used in the phantom portion (model 128; Acuson). Perhaps the section thickness on the Logiq 700 machine (GE Medical Systems) was more thinly collimated, making it more difficult to align with the needle shaft.
With experience, it was soon learned that several minor alterations could reduce bending substantially. These included use of a 20-gauge needle rather than of a 22-gauge needle, use of a scalpel to make a tiny skin incision, and placement of the electronic sensor along the shaft of the needle at the maximum required depth of needle placement rather than at the hub.
Furthermore, with added experience, it became apparent that the out-of-plane capability had substantial advantages in needle placement. That is, an understanding of the three-dimensional relationship between the lesion and the needle markedly improved both visualization and targeting. This three-dimensional concept, however, is not entirely intuitive and requires some time and experience to master prior to achievement of consistently successful needle visualization and placement. Although the equipment was somewhat calibration-sensitive at the outset, several software upgrades in the interim have improved this situation.
There are a few shortcomings of this study. First, the clinical phase was not a randomized comparison of lesion targeting with and without the device. Since a comparison of both approaches in the same patient would have resulted in an inordinate increase in the number of needle passes, the faculty were instead asked to describe their experience with the device in the context of having previously performed almost all biopsies with an attached needle guide. Therefore, there is an admitted bias toward the use of an attached guide rather than toward the use of a freehand approach.
Other shortcomings include the nonconsecutive nature of the patients, the relatively small number of patients, limited use of the out-of-plane approach, and unequal use of equipment by all faculty participants. These factors were primarily related to the wide variety of procedures performed in our dedicated sonographic interventional suite, faculty preference for different techniques, and overall timing considerations. That is, there were instances in a busy clinical practice when the faculty elected not to take the time (approximately 2 minutes) to set up the equipment prior to a procedure. Furthermore, the faculty began to take advantage of the out-of-plane approach only in the latter portion of the study after they became comfortable with the in-plane approach.
In conclusion, we believe that the UltraGuide 1000 is promising technology that can be helpful for those individuals who wish to make the transition from the use of primarily CT to the use of primarily sonographic guidance for percutaneous intervention. Although the experience and comfort level for the use of CT is substantial, the cost considerations in a busy clinical practice are also substantial and are beginning to force interventionalists to find less expensive and less time-consuming methods of guidance.
Furthermore, for those individuals with considerable experience in the performance of sonographically guided intervention, this device will allow them to make the transition from primarily the use of the attached needle guide to the use of the freehand technique not only in plane but also up to 90° out of plane. The ultimate result with this device is the added flexibility in approaching a large number of patients with a wide variety of lesions. Future research and development with this device should include incorporation of the electronic sensors into the transducer hardware, the development of a wireless attachment to the needle shaft, and the development of specialized needles.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, M.H.H., R.C.N.; study concepts and design, R.C.N., E.K.P., M.H.H., D.H.S.; definition of intellectual content, M.H.H., R.C.N.; literature research, M.H.H., R.C.N.; clinical studies, M.H.H., R.C.N.; data acquisition, M.H.H., R.C.N.; data analysis, M.H.H., R.C.N., M.A.K.; statistical analysis, M.A.K.; manuscript preparation, M.H.H., R.C.N.; manuscript editing and review, all authors; manuscript final version approval, R.C.N., E.K.P., M.A.K., D.H.S.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. Bergin, J. N. Pappas, J. J. Hwang, D. H. Sheafor, and E. K. Paulson Echogenic Polymer Coating: Does It Improve Needle Visualization in Sonographically Guided Biopsy? Am. J. Roentgenol., May 1, 2002; 178(5): 1188 - 1190. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
