Dynamic MR perfusion imaging: Capability for quantitative assessment of disease extent and prediction of outcome for patients with acute pulmonary thromboembolism

Purpose: To compare directly the capabilities of multidetector-row computed tomography (MDCT) and MRI for disease severity assessment and outcome prediction for acute pulmonary thromboembolism (APTE) patients. Materials and Methods: Fifty consecutive APTE patients underwent MDCT, MR angiography, dynamic perfusion MRI, treatment and follow-up examination. Pulmonary blood flow (PBF), pulmonary blood volume, and mean transit time maps were generated from perfusion MRI, and all segmental parameters were determined by using region of interest measurements. Receiver operator curve analyses were used to determine the most accurate parameter for diagnosis of the APTE segment. Then, APTE index from perfusion MRI (PE_{perfusion MRI} index), right ventricle/left ventricle (RV/LV) diameter ratio and APTE indexes from embolic burdens observed on MDCT (PECT index) and MR angiography (PEMRA index) were calculated. Finally, ability to differentiate mortality (n = 8) from survival (n = 42) groups and to predict patient outcome were statistically assessed. Results: PBF was a significantly more accurate parameter than others (P < 0.05). When feasible threshold value was applied, specificity and accuracy of RV/LV diameter ratio and PE_{Perfusion MRI} index were significantly higher than those of PE_CT and PE_MRA indexes (P < 0.05). Logistic regression analysis demonstrated that each index was a significant predictor (P < 0.05). Conclusion: Dynamic perfusion MRI can be effective for disease extent assessment and outcome prediction for APTE patients.