DIAGNOSTIC IMPACT OF COLOR DOPPLER ULTRASOUND-GUIDED CORE BIOPSY ON FINE-NEEDLE ASPIRATION OF ANTERIOR MEDIASTINAL MASSES

Although lymphoma and thymoma are common etiologies of anterior mediastinal masses (AMMs), smaller percentages and numbers of patients with these diseases have been enrolled in previous ultrasoundguided biopsy studies. To date, there has been no study of color Doppler sonographic features to support the differentiation of AMMs. For this retrospective cohort study, a search of the database of the China Medical University Hospital using the clinical coding "ultrasound-guided biopsy'' was conducted for the period December 2003 to February 2013. We selected patients diagnosed with AMMs (not cysts) using radiographic records. This search yielded a list of 80 cases. Real-time ultrasound-guided core needle biopsy (CNB) was performed in all but 5 patients without a sufficient safety range. In 89% (67/75) of these ultrasound-guided CNB cases, the diagnostic accuracy achieved subclassification. Fine-needle aspiration cytology achieved subclassification in only 10% of cases. On color Doppler sonography, 71% of lymphomas were characterized as "rich vascular with central/crisscross collocations'' and 29% as "avascular or localized/scattered peripheral vessels.'' However, decreased proportions of "rich vascular with central/crisscross collocations'' were found in lung cancer (4% [1/23], odds ratio = 0.018, 95% confidence interval: 0.002-0.154, p < 0.001) and thymoma/thymic carcinoma (25% [4/16]; odds ratio = 0.133, 95% confidence interval: 0.035-0.514, p = 0.003) compared with the lymphoma group. We conclude that the vessels in lymphoma AMMs have specific patterns on color Doppler sonography. Ultrasound-guided CNB of AMMs had an accuracy of <= 89% in diagnosis and subclassification. Fine-needle aspiration cytology itself cannot aid in the diagnosis. Color Doppler sonographic evaluation of AMMs followed by real-time CNB is a more efficient method. (C) 2014 World Federation for Ultrasound in Medicine & Biology.