Patient population
Ten patients (5 men and 5 women; mean age, 82.9 years) who had undergone non-contrast CT, CT perfusion, and CT angiography as part of the standard acute stroke assessment from April 2016 to February 2019 were enrolled in this retrospective study. The occlusion regions were as follows: internal carotid artery, 5 patients; and proximal segment (M1) of the middle cerebral artery, 5 patients. All patients had successful recanalization defined by the thrombolysis in cerebral infarction (TICI) scale 3 via mechanical thrombectomy.
This study was approved by the Hospital Ethics Committee. It granted permission to use pre-acquired anonymized patient data, and requirement for individual informed consent was waived.
Image acquisition
CT perfusion was performed using an 80-detector row CT scanner (Aquilion Prime; Canon Medical Systems, Ohtawara, Japan). For CT perfusion imaging, 40 mL of iohexol (Iopaque, 300 mg/mL of iodine; Fuji Pharma, Tokyo, Japan) was administered intravenously at a flow rate of 5 mL/s, followed by 10 mL of saline solution at the same flow rate using a power injector (Dual Shot; Nemoto Kyorindo, Tokyo, Japan). The following scanning parameters were used: tube voltage, 80 kVp; tube current, 50 mA; rotation, 1.5 s/rotation; beam width, 10 mm × 4 rows; slice thickness, 10 mm; kernel, FC43; and scan duration, 40 to 60 s.
Data analysis
Simulated low-dose CT perfusion data sets (60, 45, 30, and 15 mAs) were generated by adding Gaussian noise to original CT perfusion data (75 mAs) according to the following well-known relationship:
SD ∝ 1 ⁄ √mAs
Here, SD indicates the standard deviation of CT values at the level of basal ganglia and mAs indicates tube current–time product ( Fig. 1 ).
CT perfusion analysis was performed for all data sets using a Bayesian estimation algorithm implemented in a commercially available software package (Vitrea, Version 7.9.0; Canon Medical Systems). A single radiological technologist measured the quantitative perfusion values (cerebral blood volume [CBV], cerebral blood flow [CBF], mean transit time [MTT], and time to peak [TTP]) in affected and contralateral normal middle cerebral artery regions at the level of basal ganglia with a pre-defined region of interest template ( Fig. 2 ).
Statistical analysis
For statistical analysis, a free statistical software (R, version 3.5.1; The R Foundation for Statistical Computing, Vienna, Austria) was used. The quantitative values of low-dose CT perfusion data sets were compared with those of original CT perfusion data as reference standard using a Wilcoxon signed-rank test. A p-value of<0.05 was considered to indicate statistically significant difference.