| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
EDUCATION EXHIBIT |
Diagnostic Imaging Centers, Kansas City, Mo
I appreciate the opportunity to comment on the excellent articles by Barth et al (1) and Akisik et al (2) in this issue of RadioGraphics. The authors present two different perspectives on the advantages and disadvantages of body imaging with 3.0-T magnetic resonance (MR) imaging. At 3.0-T body imaging, as in all MR imaging applications, trade-offs constantly must be made to optimize the image quality within the specific constraints of the individual examination. The adage "There is no such thing as a free lunch" is often invoked to convey the need to strike a balance among the goals of maximizing the signal-to-noise ratio (SNR), optimizing the spatial resolution, and minimizing the acquisition time for a particular MR imaging examination. This adage has held true for body imaging at 3.0 T, in which there is the increased cost of the magnet as well as the additional challenge of managing artifacts that are more prominent at 3.0 T.
Given an assumed direct correlation between the SNR and the field strength, the obvious advantage of 3.0-T imaging is that the SNR at 3.0 T is double that at 1.5 T. However, we now know that the SNR at 3.0 T usually is not doubled, mainly because of changes in imaging parameters related to the increase in T1 and in the specific absorption rate (SAR), a measure of energy deposition in the patient’s tissues (1). Parallel imaging techniques play an important role in maintaining the increased SNR and an acceptable SAR while achieving decreased acquisition time and, possibly, improved spatial resolution at 3.0-T body imaging (3). Hyperechoes and transition between pseudo steady states (TRAPS), two recently developed modifications of decreased-flip-angle fast spin-echo sequences, are useful for reducing the SAR without significantly decreasing the SNR (4,5).
Both articles describe many of the obstacles that are most commonly faced in performing high-quality body imaging at 3.0 T. The increased SNR at 3.0 T must be balanced against increased RF magnetic field inhomogeneity and resultant standing wave and conductivity effects, increased chemical shift effects, and susceptibility artifacts. Imaging parameters have been altered over the past few years to improve contrast resolution at 3.0 T. The modifications in imaging parameters were necessary because of the change in tissue-specific relaxation kinetics compared with those at 1.5 T (6). Coil design also has been improved to help minimize artifacts that occur at 3.0-T imaging (7,8). Parallel imaging techniques have been applied at 3.0-T body imaging to take advantage of the increased SNR while maximizing the benefits of reduced acquisition time and allowing reduced echo train lengths, which may help decrease the SAR (9,10).
Yet, the question that may be most important for most private practice radiologists remains unanswered: In these days of declining reimbursement for MR imaging examinations, are the extra cost and effort required for the siting and optimization of a 3.0-T system justified in a practice that already has optimal 1.5-T imaging capabilities (11)? Body imaging applications in which 3.0 T is probably superior to 1.5 T include three-dimensional high-resolution MR cholangiopancreatography, staging of cervical carcinoma and evaluation of extension beyond the cervical stroma, and high-resolution MR angiography for evaluation of renal and mesenteric arteries (12–16). In addition, 3.0-T MR imaging is probably better than 1.5-T imaging for the staging of rectal carcinoma but less accurate than endorectal ultrasonography for the assessment of perirectal soft-tissue invasion (17). For MR spectroscopy in the abdomen and pelvis, 3.0-T imaging with parallel acquisition has an inherent advantage over 1.5 T in that the chemical shift effect is doubled, producing improved spectral resolution of metabolites that are obscure at 1.5 T (18,19). However, this advantage is somewhat eroded by the increased intravoxel field inhomogeneities related to the increased susceptibility effects at 3.0-T MR imaging (2). The higher SNR at 3.0-T imaging also may eliminate the need to use endocavitary coils for applications such as prostate imaging (19). As Barth et al indicate, the improved spatial resolution and the finer detail on reformatted images from 3.0-T MR imaging may result in clearer visibility and better characterization of lesions. In addition, because less time is needed for body imaging with 3.0 T than with 1.5 T, while the SNR and spatial resolution are comparable or slightly better, patient throughput can be maximized. However, to my knowledge, no large systematic studies have yet been performed that confirm the clinical benefits of 3.0-T MR imaging compared with 1.5-T imaging.
There is no doubt that the previously described advantages of 3.0-T MR imaging are worth some of the additional cost of a 3.0-T system compared with that of a 1.5-T system. The vendors are determined to expand the 3.0-T market, and sales of 3.0-T magnets dramatically increased in recent years: The number of units sold in 2006 was more than twice that sold in 2004. In fact, information from the vendors indicates that 20% of all dollars spent to purchase MR imaging equipment in 2006 went to purchase 3.0-T units, compared with approximately 10% just a couple of years earlier. Certainly, a large portion of the demand for 3.0-T magnets is generated by the demand for MR applications other than body imaging. It is likely that this trend will continue as we gain more knowledge and as the quality of 3.0-T MR imaging improves while the difference in cost between 3.0-T MR imaging systems and 1.5-T systems narrows.
Another question that remains to be answered is related to MR safety. At this time, it appears that most patients who are eligible for 1.5-T imaging will be able to undergo 3.0-T imaging. However, for some patients (eg, pregnant women), the increased magnetic field strength may represent an additional risk (7). In addition, we do not yet have enough experience with 3.0-T magnets to reassure all patients who have various implants, devices, or other accessories that 3.0-T imaging is as safe as 1.5-T imaging. Patients with a surgical implant or device that may lead to increased susceptibility artifacts probably are better evaluated with a lower-field-strength magnet. Pregnant patients and patients with ascites also may be better evaluated with a 1.5-T MR imaging system than with a 3.0-T system because of RF inhomogeneity artifacts, which may lead to non-diagnostic images (20).
The opportunities provided by 3.0-T technology to improve present imaging capabilities are exciting. The authors of both articles demonstrate that we have come a long way toward understanding how to use this relatively new technology to maximize its potential for body imaging. For radiology practices that have already purchased a 3.0-T MR imaging system, there are benefits for musculoskeletal and neurologic MR imaging, as well as for body imaging, that help rationalize the purchase (3). For those that are still weighing the cost of a 3.0-T system against the potential benefits for routine body imaging, more studies are needed to determine whether the additional investment in field strength is justified by a consistent improvement in image quality (2).
 |
References |
|---|
 Top References |
|---|
Related Articles
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOGRAPHICS | RADIOLOGY | RSNA JOURNALS ONLINE |
