Iodine Distribution Map in Dual-Energy Computed Tomography Pulmonary Artery Imaging with Rapid kVp Switching for the Diagnostic Analysis and Quantitative Evaluation of Acute Pulmonary Embolism

Rationale and Objectives To assess the diagnostic value of dual-energy (DE) computed tomography pulmonary angiography (CTPA) for acute pulmonary embolism (PE) using a helical DE scan mode with rapid kVp switching. Materials and Methods Seventy-six patients with suspected acute PE underwent DE CTPA. Two readers independently assessed and measured the iodine maps. CTPA images were assessed for the presence, location, and degree of PE as the standard of reference. Iodine maps were used to identify the perfusion defect (PD), and the diagnostic accuracy of iodine maps was calculated. The iodine concentrations of PDs and normal lung parenchyma were also measured and compared. Results A per-patient analysis showed the 84.6% sensitivity and 96.0% specificity of iodine map for PE, and on per-segment analysis, the sensitivity and specificity for PE were 82.9% and 99.6%, respectively. Intraobserver and interobserver variability correlations were excellent, with k values from 0.806 to 1.000. Quantitative analysis showed there was a significant difference for iodine concentration between circumscribed/patchy PDs or wedge-shaped PDs consistent with PE and normal lung parenchyma (P < .05). The intraobserver reliability of reader 1 was from 0.928 to 0.997, and reader 2 was from 0.912 to 0.995. And, the interobserver reliability between two readers was from 0.967 to 0.999. Conclusions CTPA based on DE scanning with rapid kVp switching can provide both morphologic analysis and quantitative evaluation of PD related to acute PE in addition to standard CTPA data. Quantification of iodine concentration may be helpful for identifying the presence or absence of PE. To assess the diagnostic value of dual-energy (DE) computed tomography pulmonary angiography (CTPA) for acute pulmonary embolism (PE) using a helical DE scan mode with rapid kVp switching. Seventy-six patients with suspected acute PE underwent DE CTPA. Two readers independently assessed and measured the iodine maps. CTPA images were assessed for the presence, location, and degree of PE as the standard of reference. Iodine maps were used to identify the perfusion defect (PD), and the diagnostic accuracy of iodine maps was calculated. The iodine concentrations of PDs and normal lung parenchyma were also measured and compared. A per-patient analysis showed the 84.6% sensitivity and 96.0% specificity of iodine map for PE, and on per-segment analysis, the sensitivity and specificity for PE were 82.9% and 99.6%, respectively. Intraobserver and interobserver variability correlations were excellent, with k values from 0.806 to 1.000. Quantitative analysis showed there was a significant difference for iodine concentration between circumscribed/patchy PDs or wedge-shaped PDs consistent with PE and normal lung parenchyma (P < .05). The intraobserver reliability of reader 1 was from 0.928 to 0.997, and reader 2 was from 0.912 to 0.995. And, the interobserver reliability between two readers was from 0.967 to 0.999. CTPA based on DE scanning with rapid kVp switching can provide both morphologic analysis and quantitative evaluation of PD related to acute PE in addition to standard CTPA data. Quantification of iodine concentration may be helpful for identifying the presence or absence of PE.