CT perfusion on par with MRI in assessing stroke risk

Quantitative CT perfusion (CTP) mismatch classification provides similar results to perfusion-diffusion MRI in identifying tissue at risk for infarction, according to a study published in the October issue of Stroke.

The results suggest CTP could provide a more accessible alternative to MRI, according to authors Bruce C.V. Campbell, PhD, of Royal Melbourne Hospital, University of Melbourne, Australia, and colleagues.

“A mismatch between the irreversibly damaged infarct core and the extent of hypoperfused tissue at risk for infarction is an attractive paradigm with which to select ischemic stroke patients for reperfusion therapies,” they wrote.

The authors explained that diffusion imaging has been established as an accurate representation of irreversible infarction, with a time-to-peak of the deconvolved tissue residue function (Tmax) of five to six seconds shown to be the best predictor of tissue at risk for infarction without reperfusion. While the MRI perfusion-diffusion mismatch paradigm has been extensively studied, Campbell et al wrote that there has been limited validation of CTP parameters.

To test correspondence of CTP-derived mismatch with MRI, the authors analyzed the results from 49 acute ischemic stroke patients who underwent CTP within three to six hours after stroke onset and perfusion-diffusion MRI within one hour after CTP.

Since relative cerebral blood flow (relCBF) has recently been shown to perform better than other parameters in predicting the infarct core, relCBF and Tmax were calculated. Campbell and colleagues restricted the core to voxels with low relCBF and delayed time-to-peak in order to reduce the number of white matter false-positive lesions.

Results showed that constraining relCBF less than 35 percent within the automated relative time-to-peak (relTTP) perfusion lesion region of interest reduced the median magnitude of volumetric error compared with diffusion-weighted imaging from 47.5 mL to 15.8 mL. The optimal CT-Tmax threshold to match the six second threshold of MRI was 6.2 seconds, and this threshold, combined with the constrained relCBF core, resulted in concordant status between CT-based and MRI-based mismatch in 90 percent of cases, according to the authors.

“The precision of MRI perfusion-diffusion imaging, particularly in accurately identifying infarct core, may remain superior to CTP. However, there is often restricted access to urgent MRI and contraindications (eg, uncharacterized metallic foreign bodies) create difficulties in the emergency setting,” wrote Campbell and colleagues. “We have demonstrated that CTP can provide information similar to MRI at the level of treatment decision-making.”