Comparison of CT perfusion summary maps to early diffusion-weighted images in suspected acute middle cerebral artery stroke. Issue 4 (April 2015)

• Record Type: Journal Article
• Authors: Benson, John; Payabvash, Seyedmehdi; Salazar, Pascal; Jagadeesan, Bharathi; Palmer, Christopher S.; Truwit, Charles L.; McKinney, Alexander M.
• Is Part Of: European journal of radiology. Volume 84:Issue 4(2015)
• Page Start: 682
• Page End: 689
• Publication Date: 2015-04
• Subjects: ACA anterior cerebral artery -- ASPECTS Alberta Stroke Program Early CT score -- CBV cerebral blood volume -- CBF cerebral blood flow -- CTP CT perfusion -- DC-SVD delay-corrected SVD -- DS-SVD delay-sensitive SVD -- DWI diffusion-weighted imaging -- MCA middle cerebral artery -- MTT mean transit time -- PCA posterior cerebral artery -- SVD singular value decomposition -- TTP time to peak; Stroke -- CT perfusion -- MRI; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2014.12.026
• Languages: English
• ISSNs: 0720-048X

Abstract

Abstract: Objectives: To assess the accuracy and reliability of one vendor's ( Vital Images, Toshiba Medical, Minnetonka, MN ) automated CT perfusion (CTP) summary maps in identification and volume estimation of infarcted tissue in patients with acute middle cerebral artery (MCA) distribution infarcts. Subjects and methods: From 1085 CTP examinations over 5.5 years, 43 diffusion-weighted imaging (DWI)-positive patients were included who underwent both CTP and DWI <12 h after symptom onset, with another 43 age-matched patients as controls (DWI-negative). Automated delay-corrected postprocessing software (DC-SVD) generated both infarct "core only" and "core + penumbra" CTP summary maps. Three reviewers independently tabulated Alberta Stroke Program Early CT scores (ASPECTS) of both CTP summary maps and coregistered DWI. Results: Of 86 included patients, 36 had DWI infarct volumes ≤70 ml, 7 had volumes >70 ml, and 43 were negative; the automated CTP "core only" map correctly classified each as >70 ml or ≤70 ml, while the "core + penumbra" map misclassified 4 as >70 ml. There were strong correlations between DWI volume with both summary map-based volumes: "core only" ( r = 0.93), and "core + penumbra" ( r = 0.77) ( both p < 0.0001). Agreement between ASPECTS scores of infarct core on DWI with summary maps was 0.65–0.74 for "core only" map, and 0.61–0.65 for "core + penumbra" ( both p < 0.0001). Using DWI-based ASPECTS scores as the standard, the accuracy of the CTP-based maps were 79.1–86.0% for the "core only" map, and 83.7–88.4% for "core + penumbra." Conclusion: Automated CTP summary maps appear to be relatively accurate in both the detection of acute MCA distribution infarcts, and the discrimination of volumes using a 70 ml threshold.