Abstract
BACKGROUND: Non-contrast computed tomography (NCCT) is a common imaging technique for suspected acute ischemic stroke (AIS); however, early ischemic changes (EICs) in AIS patients often present with subtle differences on NCCT. Dual-layer spectral computed tomography (DLCT) can reduce artifacts and provide more accurate tissue characterization. This study aimed to assess the effectiveness of DLCT in detecting EICs in AIS patients. METHODS: This study included 49 AIS patients who underwent DLCT and magnetic resonance imaging (MRI) within 24 hours. The imaging data were retrospectively analyzed, focusing on quantitative measurements from regions identified as acute infarctions on diffusion-weighted MRI. Various parameters, including conventional computed tomography (CT), virtual monoenergetic (MonoE) CT, effective atomic number (Zeff), and electron density relative to water (EDW) values, were compared between the ischemic and normal brain regions. RESULTS: The area under the curve (AUC) for EDW was the highest [0.859; 95% confidence interval (CI): 0.787-0.914], followed by MonoE 100 keV (0.830; 95% CI: 0.754-0.890). Both of which were higher than that of conventional CT (0.757; 95% CI: 0.674-0.828; P=0.01); however, there was no statistically significant difference between the AUC values of EDW and MonoE 100 keV themselves (P=0.22). EDW had the highest sensitivity (0.831) with a threshold of 102.33%, while MonoE 100 keV had a high specificity (0.846) with a threshold set at 25.83 HU. The AUC of conventional CT was comparable to that of MonoE 70 keV (0.791; 95% CI: 0.711-0.858; P=0.10), while MonoE 40 keV had the lowest AUC (0.577; 95% CI: 0.478-0.675). CONCLUSIONS: Overall, the findings suggest that EDW and MonoE 100 keV CT images from DLCT significantly enhance the detection of AIS compared to conventional NCCT imaging.