Efficacy of Iodine Perfusion Maps from Dual-energy Computed Tomography of the Pulmonary Arteries in Pulmonary Embolism Assessment

Introduction: There has been increasing availability and use of dual-energy computed tomography (DECT) over recent years. The aim of this study was to evaluate the sensitivity, specificity, and accuracy of iodine perfusion maps in diagnosing pulmonary embolus (PE) using a DECT scanner. Methods: A retrospective study was performed comparing the detection of PE and correlation of PE types (central vs. segmental and occlusive vs. non-occlusive) on computed tomographic pulmonary angiography (CTPA) and iodine perfusion maps. Diagnostic performance parameters were calculated for each reader and for the different types of PE. Interobserver variability was measured for the two techniques. Results: Both radiologists demonstrated consistent diagnostic performance in detecting PEs on CTPA. However, diagnostic performance varied widely in the detection of a defect on iodine perfusion maps. Iodine perfusion scans demonstrated a high sensitivity for occlusive and central PE but a low sensitivity for non-occlusive and segmental PE, whereas their specificity in the detection of central PE was similar to that of segmental PE. Positive predictive value (PPV) of iodine perfusion scans was higher for central PE than for segmental PE, whereas negative predictive value (NPV) was lower for segmental PE than for central PE. Occlusive PE had a low PPV and a high NPV, whereas non-occlusive PE had a lower PPV and NPV. Interobserver agreement was high in the interpretation of CTPA and fair in the interpretation of iodine perfusion maps. Conclusion: Iodine perfusion maps should be read in conjunction with CTPA. A normal iodine perfusion map excludes a central or occlusive PE.