When exploring the capabilities of 3T MR, most imagers proceed directly to the brain. The higher field strength scanning has proven especially illuminating in functional imaging and for providing greater resolution than 1.5T for a range of neurological applications. For body imaging, however, 3T is just hitting its stride.
With its higher signal-to-noise ratio, 3T produces more detailed scans than 1.5T, providing better resolution and reduced acquisition times for many body imaging sequences. Field strength has improved spectroscopy as well, with greater chemical shift dispersion. With more space between spectral peaks, chemical compounds are easier to identify. As the chemical shift dispersion improves, so does the fat suppression.
"Still, this is not just a case of more horsepower makes a better car; it's not that straightforward," said Dr. Robert Herfkens, director of MRI at Stanford University. "It's a complex issue. There are some very clear advantages in the brain, and there is not much doubt that overall image quality has improved, and that translates into better diagnostic methods. I don't think you can make a statement that everything is better at 3T."
Increased SNR has provided its own unique challenges, raising issues about contrast, homogeneity, radio-frequency deposition, and motion artifact, to name a few. Availability of coils for 3T has also been a problem in getting it up to speed in clinical practice. Over the past two to three years, however, manufacturers and radiologists have worked together to refine protocols that offset imaging limitations and have produced an array of surface coils to meet most body applications.
Increased-channel phased-array surface coils have made it easier to get clearer images when a radiologist needs an isolated view of, for example, the knee, shoulder, or back. Dr. Lawrence Tanenbaum, section chief of MR and CT at Edison Imaging/JFK Medical Center in Edison, NJ, relies on state-of-the-art, four to eight-channel phased-array coils to obtain very precise body images.
"Depending on what system you're using and how sophisticated it is, there can be discrepancies in performance," Tanenbaum said.
Dr. Joel Felmlee, a medical physicist and biomedical engineer at the Mayo Clinic's Magnetic Resonance Research Laboratory in Rochester, MN, has pioneered improvements in coil performance. Along with Dr. Richard Ehman, a Mayo Clinic radiologist, Felmlee developed the technique of spatial presaturation, which decreases artifact caused by the motion of the blood in the vessels, and navigator-guided motion correction.
"The idea has always been that if you track motion using these techniques you could correct the artifacts," Felmlee said. "Now if you're trying to do a 3T high-resolution wrist image, your pixels can be anywhere from 0.4 to 0.2 to even 0.1 mm in size. Subtle motion can corrupt that image."
Correcting motion artifact is an issue when surface coils are used, particularly endorectal coils that are used to detect prostate cancer. The standard endorectal coil is placed as close to the prostate as possible, and its use for regular clinical exams is a subject of debate. Patient comfort is a primary consideration, especially since discomfort causes patients to move, and that can decrease the quality of the images. But if the exam is done correctly, the signal from the prostate and the image can be very clear because of the endorectal coil's proximity to the gland.
"In terms of signal to noise, this is probably the best you can do," Herfkens said. "When you're routinely imaging the prostate, you don't see any of the lymph nodes, for example. But with the endorectal coil you can."
Although it is still in research stages at Emory University School of Medicine, prostate imaging at 3T has proved successful in detecting early stages of cancer.
"The 3T is giving modulations or alternations in signal that are easier to discern, and that makes it good for spotting tumors," said Dr. Hu Xiaoping, director of the Biomedical Imaging Technology Center at Emory. "Plus, the contrast of tumors is not that high, and you need high resolution to see them."
Endorectal coils are not widely available commercially, but vendors expect to introduce them within the next year.
ADJUSTMENTS IN PROTOCOL
Depending on the application, radiologists decide how to "spend" the added field strength of 3T: by decreasing the image acquisition time or improving the image resolution. Dr. Kimberly Amrami, a musculoskeletal radiologist at the Mayo Clinic, specializes in hand and wrist imaging, and she most frequently adjusts her protocol to obtain sharper images.
In typical T1-weighted wrist imaging at 1.5T, Amrami would use a 10-cm field-of-view with a wrist coil. The imaging matrix would be 256 x 256 pixels, the slice thickness might be 4 mm, and the TR time 500 msec. Acquisition may require the average of two or three signals.
In a T1-weighted protocol at 3T field strength, Amrami can increase TR to 700 (using a longer acquisition time to gain an improved image) and choose a 6- or 8-cm field-of-view and a thinner slice thickness. The number of excitations can be reduced to one, and she can select a higher resolution matrix such as 512 x 384. Acquisition would require averaging more signals to achieve the same quality available at 1.5T, which would make the process longer.
Amrami and other radiologists have found that because tissue relaxation times are longer, contrast for T1-weighted imaging can be poorer with traditional 1.5T protocols. To improve contrast at 3T, new protocols incorporate gradient-echo acquisition, fast spin-echo, inversion recovery, or rapid acquisition with relaxation enhancement (RARE).
"I adjust the imaging parameters to improve contrast, but sometimes I accept that the image is different and requires a slightly different way of looking at things," Amrami said.
Tanenbaum confirmed that there are challenges with certain types of contrast.
"But with creative approaches, we're getting spectacular T1 contrast that leaves nothing to be desired," he said.
HEAT AND NOISE LIMITATIONS
Challenges pertaining to heat and acoustic noise have, for most practical purposes, been addressed by the latest scanning technology, but they remain a concern for radiologists. As field strength increases, so does the amount of radio-frequency deposited in a patient's body. A high specific absorption rate (SAR), the measurement of RF energy actually absorbed by the human body, has the potential to limit the number of slices per scan. The U.S. Food and Drug Administration has set limits for the amount of heat deposition, and today's 3T scanners work within those guidelines.
"You cannot do any body imaging at higher field strengths, say at 7T, as of yet. That is pushing the limit," Herfkens said. "Seven-T is being used for research purposes on the head only, and that is pushing a whole host of problems to the limit."
Acoustic noise increases in the 3T gradients as well. It's a louder system that approaches the limits established by the Occupational Safety and Health Administration, so hearing protection is necessary.
OTHER APPLICATIONS AND CHALLENGES
Across the board, 3T with the proper protocols is producing extremely detailed images of structures that were not visible at 1.5T because of background noise. Using a finger coil, Amrami can study the 0.2 to 0.3-mm-thick extensor tendon and the nerves going to the fingertips.
"We had a case where a person had surgery for a little cyst," she said. "They had some loss of sensation along one side of the finger. The hand surgeon wanted to know if the nerve had been accidentally cut. We were able to tell that the nerve was continuous. It was just temporary damage to the nerve, so surgery was avoided."
Other areas where 3T shows promise include cardiovascular, breast, and pediatric joint imaging.
- Cardiovascular imaging. Hu at Emory University has acquired encouraging results using 3T for cardiovascular imaging. Functional information collected in a series of cines allows him to study heart motion, looking for abnormal anatomy and function. His primary interest is in detecting an ischemic event, which can typically lead to a stroke.
"Three-T has refined our imaging techniques as we examine morphology functions," Hu said. "Because of the limitations of signal to noise, such events were not noticed at 1.5T."
Hu pays careful attention to B1 and B0 homogeneity, as some sequences are more sensitive to homogeneity. Increased signal to noise and inhomogeneity can cause more shading of the images. Uniformity overall is better at 1.5, while 3T requires more shimming to produce a uniform image, he said.
"It's fair to say that at 3T that you need good or even better shim coils than at 1.5. High-order shim coils are necessary," Felmlee said.
- Breast imaging. Dr. Mitch Schnall, chief of MRI at the University of Pennsylvania Medical Center, uses 3T extensively for breast cancer detection.
"Breast MRI has the capability of finding small cancerous lesions often missed by regular mammograms and self-examinations," he said.
Because a breast typically contains a lot of fat, the exceptional fat suppression capabilities of 3T make a tremendous difference in imaging the breast. Eliminating the competing fat signal increases contrast.
"Knowing the extent of cancer within a breast can impact a patient's decision on the type of breast therapy chosen, and that can mean the difference between choosing a mastectomy or a lumpectomy," Schnall said.
- Pediatric joint imaging. Cincinnati Children's Hospital is the first pediatric institution in the U.S. to install a 3T scanner that enables whole-body imaging for clinical applications. The higher resolution images allow doctors to depict smaller anatomy more clearly.
"The 3T is especially good for scanning small joints such as a child's ankle or wrist," said Dr. Blaise Jones, chief of neuroradiology.
The scanner has also aided studies of juvenile rheumatoid arthritis and other musculoskeletal disorders.
"The most pleasant surprise about the 3T is that within two weeks we were running a full schedule on it. It's not like a sports car that's taken out one weekend a month," Jones said.
MR. RAUF is a freelance writer in Seattle, Washington.