CT still preferred for stroke, but pending MRI improvements could change that
Relative location of scanners and motion artifacts have trumped MR’s ability to image whole brain and clearly indicate diffusion
BY JOHN C. HAYES
Diagnostic Imaging asked Dr. William Bradley Jr., chair of radiology at the University of California, San Diego and a longtime expert on MR, what it will take to move MR to a more prominent position in emergency stroke imaging.
Diagnostic Imaging: MRI has been around for over two decades.1 Why is most acute stroke imaging today performed using CT?
Dr. Bradley: Because CT is usually sited closer to the emergency department than MRI. We installed an MRI close to the ED several years ago, primarily to manage acute strokes.
DI: So are you using MRI now for all your acute strokes?
Bradley: We have done a number of acute strokes. However, some of these patients have had motion artifacts. This required repeating the sequences, which extended the scan time, delaying treatment. So we have gone back to CT and CT perfusion (CTP) for now, but this has its problems as well.
DI: Can you use PROPELLER or BLADE to get rid of motion artifact on MR?
Bradley: Yes, we can, but they increase the scan time. We have a new motion artifact reduction technique, developed by Anders Dale here at UCSD, called PROMO.2 It adds only a few milliseconds to the scan time. When it is commercialized enough to go on our clinical magnets, we will transfer the acute strokes back to MRI.
DI: What other problems have you had with CT and CTP?
Bradley: We cover only 4 cm of brain with our 64-slice CT scanners. While we can toggle back and forth to cover
8 cm of brain, we are concerned that the low temporal resolution (3+ sec) in toggle mode will reduce the quality of the CTP scan. As you know, MRI diffusion and perfusion cover the entire brain.
DI: Are there any other problems with CTP?
Bradley: Frankly our biggest problem is getting the CTP study processed and read in the middle of the night. This is done automatically in MR perfusion. Hopefully it will be done automatically with CTP in the future. But there will still be problems relative to MRI.
DI: Such as?
Bradley: Well, 20% of the time MR diffusion imaging is done in an acute stroke, there are multiple diffusion abnormalities on whole-brain MRI. Some of these will be older than the middle cerebral artery stroke that brought [the patient] in—and some will be older than three hours, precluding intravenous tPA. You don’t get this information with CT or CTP and these patients are at greater risk for hemorrhage if they get tPA.
DI: Isn’t reduced cerebral blood volume (CBV) on CTP equivalent to positive diffusion on MRI in indicating the core of the infarct?
Bradley: Very good question. I suspect positive diffusion on MRI is a whole lot more obvious than reduced CBV on CTP. Several years ago we frequently couldn’t even define a CBV abnormality on MRI when the diffusion was clearly positive. I’m not sure if we would have the same problem with CTP since it is so hard to compare CT and MRI in the same patient with an acute stroke.
DI: Are you concerned about the radiation dose in CT?
Bradley: Like everyone else, we were concerned about the radiation dose. But our service engineers tell us we are OK. Still, there is more radiation for CT and CTP compared with MRI. That message seems to have gotten through to the public, since our CT volume has decreased over the last year. The Europeans are much more radiation-phobic than we have been. If we now develop that fear in the U.S., that, too, will be a reason to migrate acute stroke evaluation to MRI.
1. Bradley WG. MR evaluation of acute stroke. Applied Radiology 1999;28(5):22-26.
2. Brown TT, Kuperman JM, Erhart M, et al. Prospective motion correction of high-resolution magnetic resonance imaging data in children. Neuroimage 2010;53(1):139-145. Epub 2010 Jun 11.