Abstract: Objective To investigate the diagnosis effect of transbronchial needle aspiration under CT-guided for mediastinal mass and pulmonary hilar mass.Methods One hundred and forty patients underwent mediastinal mass and pulmonary hilar mass were randomly divided into either observation group (n=70) or control group (n=70) by random number table method after signing informed consent. In the observation group, transbronchial needle aspiration was carried out under real-time CT guidance. In the control group, transbronchial needle aspiration was carried out without CT guidance. The number of positive patients, false positive patients, negative patients and false negative patients were recorded in both groups. The sensitivity, specificity, positive predictive value and negative predictive value were calculated in both groups. The incidence of complication was also recorded in both groups. For enumeration data, χ² test was applied to test significance. A P-value less than 0.05 was deemed significant.Results Forty patients with malignant tumor, 1 benign tumor, 13 tuberculosis, 2 sarcoidosis and 14 normal tissues were identified in the observation group. Thirty-eight patients with malignant tumor, 2 benign tumor, 11 tuberculosis, 3 sarcoidosis and 16 normal patients were found in the control group. There was no difference between two groups (The χ² values were respectively 0.029, 0.000, 0.050, 0.000 and 0.042. The P-values were respectively 0.734, 1.000, 0.823 1.000 and 0.837). The false negative patients (1 case) in the observation group were less than those (7 case) in the control group (χ²=4.275, P=0.039). The number of the positive patients (65 case) in the observation group were similar to those in (59 case) the control group (χ²=1.764, P=0.184). The number of the false positive patients (1 case) in the observation group were similar to those in (2 case) the control group (χ²=0.000, P=1.000). The number of the negative patients (3 case) in the observation group were similar to those in (2 case) the control group (χ²=0.000, P=1.000). The sensitivity, specificity, positive predictive value and negative predictive value in the observation group were 98.48%, 75.00%, 98.48%, and 75.00%, respectively. The sensitivity, specificity, positive predictive value and negative predictive value in the control group were 89.39%, 50.00%, 96.72%, and 22.22%, respectively. There was not mediastinal emphysema or pneumothorax patient in the observation group, but six mediastinal emphysema or pneumothorax patients in the control group. There was significant difference between two groups (χ²=4.353, P=0.037). There was one hemorrhage patient in the observation group, but eight hemorrhage patients in the control group. There was significant difference between two groups (χ²=4.275, P=0.039).Conclusion Transbronchial needle aspiration under CT is more effective for diagnosing mediastinal mass and pulmonary hilar mass and its complication is less.