Click here to


Are you sure ?

Yes, do it No, cancel

Impact of Scan Duration On CT Perfusion-Derived Tissue Volumes in Acute Ischemic Stroke Patients

E Wright1*, L Hur1 , M Goyal2,3 , A Demchuk2 , B Menon2,3 , T Lee1 , (1) Medical Biophysics, University of Western Ontario, London, Ontario, Canada. (2) Radiology, Foothills Medical Center, Calgary, Alberta, Canada. (3) Clinical Neuroscience, Foothills Medical Center, Calgary, Alberta, Canada.


(Tuesday, 7/31/2018) 3:45 PM - 4:15 PM

Room: Exhibit Hall | Forum 8

Purpose: Applying thresholds to CBF, Tmax and other parameters from CT Perfusion (CTP) allows volume of irreversibly damaged infarct and salvageable penumbra, which are important for determining eligibility for endovascular therapy, to be measured in ischemic stroke patients. CTP scan durations range from 40-120s despite the possibility of truncation artifacts from delayed wash-in/wash-out of contrast affecting the accuracy of parameter calculations, and by extension, infarct/penumbra volumes used for triaging. The purpose of this study was to determine the minimum scan duration that produced reliable tissue volume estimates.

Methods: 65 patients underwent admission CTP of 150s duration. Images were removed from the end of the acquisition to simulate scan durations of 120, 90, 60, and 40s, maps were computed for each duration. Tissue volumes derived by the following thresholds were recorded at each scan duration; CBF<7 and 13mL∙min-1∙100g-1, and Tmax>9, 16, and 21s. Two-one-sided-tests procedures were used for each threshold to find the shortest scan duration where tissue volume was equivalent to the 150s scan within ±2mL.

Results: Volumes derived by both CBF thresholds were equivalent to their volumes from the 150s scan for all shorter scan durations (p<0.025). Decreasing scan duration from 150 to 40s caused volumes with CBF<7mL∙min-1∙100g-1 and 13mL∙min-1∙100g-1 to decrease by 0.7±0.2mL (p<0.001) and 1.4±0.3mL (p<0.025) respectively. Volumes derived by Tmax>21s in the 120s scans decreased by only 1.27±0.3mL (p<0.025) relative to the 150s scans, all shorter scan durations were not equivalent. Volumes derived by Tmax>9s and 16s were not equivalent to the 150s duration for any shorter scan durations. Volumes measured by Tmax>9, 16, and 21s were underestimated by 13.3±0.8mL, 14.3±1.3mL, and 11.5±1.3mL respectively, relative to their corresponding values from the 150s duration.

Conclusion: CBF-derived infarct/penumbra volumes remain equivalent across scan durations ranging from 40-150s. Tmax-derived infarct/penumbra volumes are underestimated at scan durations shorter than 120-150s.

Funding Support, Disclosures, and Conflict of Interest: Ting-Yim Lee has a licensing agreement with GE Healthcare for the Perfusion software used in this work.


Stroke, Perfusion Imaging, Image Analysis


IM- CT: Quantitative imaging/analysis

Contact Email