Usefulness of contrast-enhanced endoscopic ultrasonography in the differentiation between malignant and benign lymphadenopathy

BACKGROUND/AIMS: Endoscopic ultrasonography (EUS) is considered the most useful diagnostic modality for regional staging; however, it is still difficult to diagnose lymph node metastasis by EUS images only. In this study, we report the usefulness of contrast-enhanced EUS (CE-EUS) in the evaluation of benign lymph nodes (BLN) or malignant lymph nodes (MLN) based on blood flow patterns. SUBJECTS AND METHODS: In the retrospective study, CE-EUS was performed in 46 patients in whom EUS revealed lymph node in the mediastinum or abdominal cavity. The subjects consisted of 22 patients with BLN and 24 patients with MLN. The lesions were examined by EUS, and the maximal and minimal diameters of lymph nodes were measured. Thereafter, the shape and internal echoes were investigated, and the findings were morphologically classified based on Catalano's report. Enhancement effects and the diagnostic capability of CE-EUS were evaluated. In the prospective study, BLNs were differentiated from MLN using the enhancement patterns on CE-EUS based on the results of the retrospective study, and the diagnostic capability was evaluated. RESULTS: In the retrospective study, there were no significant differences in the maximal diameter and maximal/minimal diameter ratio between MLN and BLN. The morphology was classified into four types. Based on the morphological classification, the sensitivity, specificity, and accuracy rate were 88.2%, 77.3%, and 82.1%, respectively. On CE-EUS, the enhancement pattern was classified into three types. The BLN lesions showed uniform enhancement (19/22). In all patients with MLN, a defect of enhancement was observed (24/24). The sensitivity, specificity, and accuracy rate of CE-EUS were 100%, 86.4%, and 92.3%, respectively. In the prospective study, the sensitivity, specificity, and accuracy rate of CE-EUS were 100%, 81.8%, and 92.0%, respectively. CONCLUSIONS: CE-EUS is useful for differentiating BLN from MLN.