Radiology: CTA protocol may overestimate stroke infarct
 |
The discrepancies could have inappropriately excluded up to 90 percent of eligible stroke patients from reperfusion therapy if they had been used to inform treatment decision making. Although diffusion-weighted (DW) MR imaging provides the highest accuracy for depicting acute ischemic stroke, its use has been primarily restricted to major medical centers. CT scanners are more widely available and hypoattentuation in the unenhanced images may predict final infarct volume, albeit without the sensitivity of DW MRI.
With newer CT systems, hospitals have adopted revised CTA protocols and increased speed to maximize arterial-phase contrast enhancement. These changes may prevent the time delay required for a steady state between arterial and tissue contrast to be reached, and consequently overestimate infarct size.
Researchers from Massachusetts General Hospital (MGH) in Boston designed a retrospective study to determine the influence of CTA acquisition protocols on infarct size estimates and evaluate the potential effect on treatment decision making.
Benjamin Pulli, MD, from the division of neuroradiology at MGH, and colleagues examined 100 consecutive stroke patients who underwent CT and DW MRI exams between January 2000 and February 2010.
A total of 35 patients were categorized as protocol 1 and 65 as protocol 2, with the latter imaged via a newer, faster CTA protocol. There were no differences in clinical and demographic data between the two groups.
Two neuroradiologists blinded to patient information except the side of stroke involvement reviewed the imaging exams and estimated infarct volumes in millimeters.
“[We] demonstrated hypoattentuating volumes on CT angiography source images obtained with our newer, rapid CT angiography protocol represented significantly overestimated volumes of restricted diffusion at concurrent DW imaging, while analysis of images obtained with our older, slower protocol revealed reasonable agreement and no consistent overestimated volumes compared with DW imaging infarct volumes,” wrote Pulli et al. In addition, inter-rater agreement was lower with the new protocol.
Among protocol 2 patients, volume on CTA source images was significantly larger than DW images. In this group, volume at DW imaging was overestimated by at least 20 percent in 90.3 percent of patients vs. 9.7 percent in protocol 1, according to Pulli and colleagues.
The findings could have major implications for organizations that rely on CTA source images to evaluate brain parenchyma and estimate infarct size and for clinical trials that rely on mismatch data to select patients for thrombolysis, according to the researchers. They estimated that the rapid protocol results would have inappropriately excluded approximately 45 to 60 percent of patients from treatment. Pulli and colleagues emphasized that neither the mismatch concept nor absolute infarct size at DW imaging is recommended outside of clinical trials for treatment decision making.
“We suggest that when CT angiography source images are used to evaluate the parenchyma during ischemic stroke, a protocol that is empirically validated to provide a good estimate of the infarct core be established at each institution,” the researchers wrote. They noted that a delay of 40 seconds between the protocol to visualize the intracranial arteries and the protocol to evaluate the parenchyma could suffice on the basis of their data.
Shortcoming of the study included its retrospective design, use of DW imaging as the reference standard to estimate the infarct core and the possibility of reperfusion between CTA and MR imaging.