The value of pulmonary embolism detection and double energy perfusion imaging of dual source CT in diagnosing peripheral pulmonary embolism / 实用放射学杂志

ABSTRACT

Objective To explore the value of post processing technique of dual source CT in diagnosing peripheral PE.Methods One hundred and fifty patients with suspected PE were underwent dual source CT scan,and 20 of the all patients met the criterion. The raw CT data were reconstructed by the pulmonary embolism detection (PED)software,double energy perfusion imaging (DEPI)and CT pulmonary artery angiography (CTPA).The PED,DEPI and CTPA map were analyzed by 2 senior doctors in cardiovascular diagnosis profession.According to the diagnosis standard of PE,the location,number of the emboli in segmental and sub-segmental pulmonary arteries were respectively recorded on the PED and DEPI map.We calculated the detection rate,evaluated the significant difference and evaluated the diagnosis consistency between the DEPI and CTPA map.Results Emboli were found in 30 segmental and 40 sub-segmental arteries on the CTPA map with the detection rate of 7.50% and 5.00%,respectively.Emboli were found in 48 segmental and 62 sub-segmental arteries on PED map with the detection rate of 12.00% and 7.75%,respectively.There was significant difference between the CTPA and PED map (χ2 =4.60,5.06,P<0.05).The DEPI and PED map had higher consistency in diagnosing PE.The Kappa coefficient was 0.94 if the PED map was regarded as the reference standard,Among 48 cases with segmental PE detected by the PED map,there were 13 cases of complete filling defects and 35 cases of partial filling defects.Among 13 cases of complete filling defects,there were perfusion defect in 10, and perfusion thin in 3 on the DEPI map.Among 35 cases of partial filling defects,there were perfusion defect in 2,perfusion thin in 29,and perfusion normal in 4 on the DEPI map.Conclusion The PED map combined the DEPI map of dual source CT can significantly improve the diagnosis rate of peripheral PE and has a high clinical value.