Profound improvements in perfusion and diffusion tensor imaging over the past few decades are changing the ways in which radiologists understand disease processes, especially those involving small blood vessels in the brain.
Some advancements are now current practice, Dr. Denis Le Bihan from NeuroSpin, CEA Saclay Center in Paris, told delegates at the 2007 European Congress of Radiology. Diffusion MRI, for instance, is commonly used both for assessing the extent of acute ischemic disease to speed the treatment of stroke patients and for mapping white matter fibers to identify for neurosurgeons which parts of the brain are functional and should not be touched during surgery.
Future directions for diffusion MRI are not yet clear, but there are two potential applications, Le Bihan said. The first relates to problems in communication between various parts of the brain when, for example, the frontal lobe is not talking to the hippocampus, or not talking in the right way.
"There are some papers showing that in schizophrenia, depression, or autism, the brain anatomy looks almost normal. The issue is that communication is not well established, and communication goes through the white matter fibers. So if we have a way to map out white matter fibers, we may be able to show that something is abnormal; there are not as many fibers as we expect, or the fibers are not well organized. These are things that could be detected with diffusion tensor imaging," Le Bihan said.
Diffusion MRI also may be used to detect activation of the brain. During activation, the diffusion of water molecules slows down. The reason is not yet known, but it may be because neurons swell when they are activated. If diffusion MRI proves to be accurate in visualizing changes in the microscopic structure of tissues that occur during activation of the neurons, it could reveal a process that is intrinsically linked to brain activation.
"Blood oxygen level-dependent MRI is an indirect way to detect brain activation. It detects an increase in the cerebral blood flow in the regions that have been activated. With diffusion MRI, we could have a way to detect directly these regions," he said.
These potential uses of MRI will require further technological developments to become reality. For one thing, diffusion MRI is not available on all MR systems. In addition, the acquisition of data requires considerable postprocessing. Technical improvements are needed so that one can just press a button and get the results, Le Bihan said.
A couple of MRI sequences can be done routinely today to reveal pathologies at the level of the blood vessel, according to Prof. Anne Osborn, executive vice president of Amirisys in Salt Lake City.
"Everyone can do gradient-refocused T2* scans; every manufacturer has them. The scans are just not used as much as they should be. And they can be very revealing by providing a look at the microscopic disease processes," she said.
Radiologists commonly use a single "wastebasket" term, such as "small vessel vascular disease or microangiopathy of the brain," without recognizing that a wide variety of pathologies can account for different imaging appearances, she said. Yet with a constellation of imaging findings, radiologists can distinguish problems such as amyloid angiopathy from arteriolosclerosis with a high degree of certainty.
Imaging studies can reveal pathologies on all three sides of the microcirculation. In arterioles, MRI can reveal arteriolosclerosis as well as lipohyalinosis, which are manifestations of atherosclerosis, and amyloid angiopathy, which has different disease underpinnings and imaging findings. At the capillary level, imaging can show what happens when capillary junctions do not function properly, and on the venous side, it can provide useful information on autoimmune diseases.
"There certainly is a new frontier for radiologists, and, in a way, it has been there all along. It is to look at microscopic disease processes and attempt to translate them into why we see what we see on imaging studies," Osborn said.
