Objective To evaluate the diagnostic value of endobronchial ultrasound-guided transbronchial needle aspiration(EBUS-TBNA) in mediastinal or lung malignant lesions and the prognosticate value of target therapy in lung cancer.
Methods Retrospective clinical trial of 91 consecutive patients with enhanced lymph node or mass in the mediastinum,with or without pulmonary lesions non-diagnosis by routine electronic bronchoscope were included in this study between May,2012 and December,2015.The samples were obtained from the lesions using EBUS-TBNA.EGFR mutation in 24 EBUS-TBNA samples was detected.The numbers of final diagnosis,the features of ultrasonic and complication were recorded.
Results Two hundred and twenty-seven lymph nodes and twenty-three masses were sampled by EBUS-TBNA in 91 malignant patients.The subcarinal nodes and right lower paratracheal lymph node accounted for 44.2%(304/692)and 31.5%(218/692);86 patients were confirmed by EUBS-TBNA.The overall sensitivity,specificity and accuracy of EBUS-TBNA were 94.5%,100.0% and 94.5%,respectively.The diagnosed cases included 48 cases lung adenocarcinoma(55.8%),23 cases of small lung cancer(26.7%),6 cases of squamous cell carcinoma(7.0%),5 NSCLC(5.8%),and 5 cases of non-small lung cancer unable to type;the metastases from extrapulmonary included 1 case of cervical cancer,1 case of colon cancer,1 case of esophageal cancer and 1 case of renal carcinoma.The round lymph nodes (L/S<1.5),heterogeneous,clear boundary,absence of the lymph door structure and irregular blood supply were the common phenomenon of malignant lesions for ultrasound image.EGFR mutation rate of EBUS-TBNA specimens was 70.8%,the mutation rate of tissue specimens was 63.6%(7/11),and the mutation rate of liquid cytology was 76.9%(10/13).The positive rate between two specimens showed no significant difference(χ
2=0.509 1,
P=0.47);Minor complications were found in 17 cases.
Conclusion EBUS-TBNA was not only a minimally invasive,safety and effective diagnosis technique for mediastinal or lung malignant lesions but also an ideal tool for prediction of target therapy in lung cancer.