Max Wintermark MD, Presenter: Nothing to Disclose

Salvador Pedraza MD, Abstract Co-Author: Nothing to Disclose

Dorith Goldsher MD, Abstract Co-Author: Nothing to Disclose

Birgitta Katinka Velthuis MD, Abstract Co-Author: Nothing to Disclose

James C. Anderson MD, Abstract Co-Author: Nothing to Disclose

Adam Eugene Flanders MD, Abstract Co-Author: Nothing to Disclose

Maarten van Leeuwen, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

Dynamic perfusion-CT (PCT) with deconvolution requires an arterial input function for data post-processing. The original model recommends selecting the arterial input as close as possible to the evaluated arterial territory. In the clinical settings, the anterior cerebral artery (ACA) is often chosen for this purpose. The goal of this study was to determine how the selection of the arterial input influences the PCT results in acute stroke patients.

125 patients with a suspected hemispheric stroke lasting less than 12 hours underwent admission PCT. Follow-up DWI was used to delineate the final infarct core. PCT datasets were post-processed using branches of the ACA, the right middle cerebral artery (MCA) and the left MCA as arterial input function. The regional cerebral blood volume (CBV), flow (rCBF) and mean transit time (MTT) were measured in the corresponding territories. The infarct core and tissue at risk were determined using MTT and CBV thresholds previously reported in the literature. PCT results were classified based on their calculation using an arterial input function supplying an ischemic territory or a normal territory. Results from the corresponding territories in the same patients were compared using paired t-tests. The infarct core and tissue at risk obtained with different arterial inputs were compared to the final infarct core inferred from the delayed DWI.

The MTT values obtained with the "ischemic" arterial input function tended to be shorter (p = 0.009 in "nonischemic" territories and p = 0.016 in "ischemic" territories), and the rCBF values lower (p = 0.024), compared to those obtained with a "nonischemic" arterial input function. The rCBV values were not influenced by the selection of the arterial input function. No statistically significant difference could be observed between the extent/location of the PCT infarct core and tissue at risk depending on which arterial input function was chosen.

In acute stroke patients, the selection of the arterial input function has no statistically significant impact of the PCT results; standardization of the PCT post-processing always using the ACA as the arterial input function is adequate.

M.W.,S.P.,D.G.,B.K.V.,J.C.A.,A.E.F.,M.v.: The present study was an international, multicenter trial supported by Philips Medical Systems

Wintermark, M, Pedraza, S, Goldsher, D, Velthuis, B, Anderson, J, Flanders, A, van Leeuwen, M, et al, , Which Arterial Input Function Is Correct for Perfusion-CT of Acute Stroke Patients: The One on the Ischemic Side or a "Nonischemic" One?.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4413433.html