Abstract: Objective:To explore the characteristics of the CT Hemoperfusion for lung masses and its clinical value.Methods:35 consecutive patients(21 men and 14 women;mean age,64.7 years;age range,24-83 years) underwent perfusion CT.With a spiral plus 64 scanner,an plain scan was performed through the anatomical region in question.Then sections through the masses were selected.Using an injector pump,50 mL of 300 mg/100 mL ultravist contrast agent was injected via cubital vein at a rate 4 mL/s,and 5 s delay,then a multisection perfusion procedure was performed with a cine mode up to 35-40 s.The parameters of hemoperfusion such as blood volume and permeability surface products in masses,were acquired by a software package perfusion 3.These values of hemoperfusion(BV and PS)of three independent samples(lung malignant lesion group,lung tubercular lesion group and lung inflammatory lesion group)were analyzed statistically.Results:The mean of BV and PS in lung malignant lesion,in lung tubercular lesions and lung inflammation were(8.43±2.44)mL·100 g^-1, (24.30±5.80)mL·min^-1·100 g^-1,(1.90±0.60)mL·min^-1·100 g^-1,(1.19±0.37)mL·min^-1·100 g^-1,(19.73±6.65) mL·100 g^-1,(30.01±9.06)mL·min^-1·100 g^-1,respectively.The mean BV and PS between lung malignant tumor and lung tubercular lesion,between inflammatory lesions and tubercular lesions,and between lung malignant tumor and inflammatory lesions were statistically significant(p<0.05).On the other side,the mean of BV between lung malignant lesion and inflammatory lesions were statistically significant(p=0.014,t=2.639),but not of PS(p=0.506,t=0.674).Conclusions:CT Hemoperfusion parameters BV and PS for lung showed a significant difference between lung malignant lesion and lung tubercular lesion,so did between inflammatory lesions and tubercular lesions.Though it is difficulty to differentiate lung malignant lesion from lung inflammatory lesions solely depending on the CT Hemoperfusion parameters,hemoperfusion is an indirect reflection of biological behavior of lung masses and has important value for the diagnosis and prognosis of lung cancer.