Analysis Of EBUS-TBNA Ultrasonographic And Chest CT Features In Benign And Malignantmediastinal Lymph Nodes

Objective: In particular,the diagnostic value of ultrasound bronchography and/or other imaging diagnostic methods for mediastinal lymph nodes is still unclear.We pectively analyzed the characteristics of ultrasonic imaging and CT of mediastinal lymph nodes to distinguish benign and malignant mediastinal lymph nodes,and to verify the diagnostic value of EBUS-TBNA in hilar and mediastinal lymph nodes.Methods: We reviewed patients with EBUS-TBNA admitted to the endoscopy center of the Affiliated tumor Hospital of Guangxi Medical University from February 2016 to June 2019,we analyzed each group of patients’ age,gender,ultrasonic images of EBUS-TBNA,CT images,pathology and cytology.Data analysis was performed using IBM SPSS Version 22.P < 0.05 was considered statistically significant.Results: We included 161 patients,with an average age of 57.57 years.A total of 265 lymph nodes were sampled,there were 102 malignant tumors and 33 benign cases,whilst 26 were unconfirmed.The diagnostic yield of EBUS-TBNA was 83.85%,the sensitivity was 79.69%,and the specificity was 100%.The positive predictive value was 100% and the negative predictive value was 55.93%.Malignant lymph nodes were less homogenous than benign lymph nodes(96.23%VS 80.40%,P = 0.001)and more prone to annular enhancement(10.06% VS 0%,P = 0.035).Small cell lung carcinomas were more likely to show lymph node fusion than adenocarcinoma(62.5% VS 23.81%,P = 0.005).The enhanced CT values of small-cell lung carcinoma lymph nodes were concentrated in the range< 15 HU.When the cut-off of the lymph node diameter was 1.87 cm,the accuracy of predicting pathologically positive lymph nodes was 65.7%.Upon comparison of the lymph nodes of lung adenocarcinoma patients and the lymph nodes of lung small-cell carcinoma patients,the accuracy of predicting lung adenocarcinoma was 69% when the degree of lymph node enhancement was-2.67 HU.Upon comparison of small-cell lung cancer and tuberculosis,the accuracy of predicting small-cell lung cancer was 74.4% at a cutoff of 2.04 cm.Conclusions: The uneve uniformity of lymph nodes in ultrasound and the annular enhancement of lymph nodes in CT are helpful for the identification of malignant lymph nodes.