Report from ASNR: CE-MRI identifies susceptibility to hemorrhage after stroke

May 2, 2006

Contrast-enhanced MRI reliably predicts whether acute stroke patients receiving intravenous tPA treatment are particularly susceptible to intracranial hemorrhage, according to a study presented Monday at the American Society of Neuroradiology meeting in San Diego.

Contrast-enhanced MRI reliably predicts whether acute stroke patients receiving intravenous tPA treatment are particularly susceptible to intracranial hemorrhage, according to a study presented Monday at the American Society of Neuroradiology meeting in San Diego.

Dr. David J. Mikulis, an associate professor of neuroradiology at Toronto Western Hospital, reported that the presence of blood-brain barrier breakdown identified with postgadolinium T1-weighted spin-echo imaging produced 100% positive and negative predictive values for 10 of 24 patients who developed hemorrhagic transformation.

Hyperintensities were absent in all 14 patients without intracranial bleeding but were seen in five of 10 patients who developed hemorrhage in the week following treatment.

"Basically, if you don't bleed, we don't see these signs," Mikulis said.

The imaging protocol involved three gadolinium injections to measure permeability, perfusion, and vascular occlusion with CE-MR angiography. A T1-weighted axial sequence determined the degree of enhancement in the middle cerebral artery region.

As a screening tool, the technique promises to reduce the risk of intracranial bleeding among stroke patients who receive tPA thrombolysis. Although tPA can reverse the neurological deficits of stroke, it also increases risk of hemorrhagic transformation 10-fold, Mikulis said.

Studies presented by Mikulis and other researchers at the meeting added weight to growing evidence demonstrating the power of MRI and CT to determine when tPA thrombolysis should be attempted beyond the traditional three-hour treatment window.

Previous research suggested that microbleeds could serve as markers of risk for hemorrhagic transformation, but evidence accumulated by Dr. Eliana E. Bonfante and Dr. Karen C. Albright at the University of Texas in Houston found no significant correlation between microbleeding and the incidence of hemorrhagic conversion after thrombolytic therapy.

Dr. Max Wintermark, an assistant professor of radiology at the University of California, San Francisco, was one of several neuroradiologists who presented evidence showing equivalence between perfusion CT and diffusion-perfusion MRI for identifying salvageable brain tissue to help select patients for intra-arterial thrombolysis three to nine hours after the onset of stroke symptoms.

Excellent agreement between the two techniques was seen among 45 patients for determining the extent of cortical involvement, calculating penumbra size, and identifying 14 patients for intra-arterial tPA treatment. Only one patient who would have been treated on the basis of the MRI evaluation would not have received thrombolysis on the basis of the functional CT and CTA studies.

Noncontrast CT is routinely used to discriminate between hemorrhagic and ischemic stroke, so add-on perfusion CT could offer a convenient way to measure the presence and extent of salvageable brain tissue.

Dr. Pamela Schaefer, an associate professor of neuroradiology at Massachusetts General Hospital, presented research showing the diagnostic power of a faster and easier MR method for isolating ischemic penumbra. The MGH study of 44 patients found the five-minute enhancement mapping technique performs as well as methods using time-consuming deconvolution-based algorithms.

Based on DWI-PWI mismatch criteria, readers came up with the same decision whether to administer intra-arterial tPA in 36 of 44 patients (images were uninterpretable in two cases, and in two other cases, the perfusion deficits were almost exactly 20% larger than the diffusion deficits, making decision making difficult). Among these 36 studies, the sensitivity and specificity of MTE maps for detecting mismatch were 87% and 85%, respectively, for reader one and 93% and 95%, respectively, for reader two.

For more information from the Diagnostic Imaging archives:

The window expands for more effective stroke treatment

Aggressive tPA stroke treatment promises to lower long-term costs

Seventh measurement can increase six-stroke parameter sensitivity