Abstract
Objective To explore the diagnostic value of maximum standard uptake value (SUVmax) from 18F-FDG PET/CT images in enlarged mediastinal lymph nodes of unknown etiology. Methods We performed a retrospective study of patients with enlarged mediastinal lymph nodes on 18F-FDG PET/CT scans. SUVmax and the short axis and long axis of lymph nodes were recorded. These parameters were compared among the five commonest causes of mediastinal lymphadenopathy: lymphoma, metastatic disease, sarcoidosis, tuberculosis, and lymphadenitis. Histopathologic diagnosis was recorded as the final golden standard. Results A total of 94 patients (62 men and 32 women; age range 7–85 y) were included with final diagnoses of 42 patients with benign pathology and 52 patients with malignancies. The sensitivity, specificity, and the accuracy of PET/CT in diagnosis of the benign and malignant mediastinal lymph nodes were 94.2%, 73.8%, and 85.1%, respectively. The SUVmax of benign and malignant groups were 13.10 ± 5.21 and 12.59 ± 5.50, respectively, which had no statistical difference (P > 0.05). However, the long axis and the short axis of lymph nodes in the benign and malignant groups were 2.86 ± 1.02 cm, 1.77 ± 0.60 cm and 6.04 ± 3.83 cm, 3.95 ± 2.08 cm, respectively (P < 0.05). The diagnostic values of PET/CT were higher than those of the long or short axis. However, the specificity of PET/CT was lower (73.8%) than that from the long or short axis (90.5% and 92.9%, respectively), although no statistical difference existed. Among the five common causes of mediastinal lymphadenopathy, significant differences could be seen in SUVmax and in the long axis and the short axis of lymph nodes (P < 0.05). Conclusions SUVmax, a commonly used semiquantitative measurement, was not helpful for differentiation between benign and malignant lesions in patients with enlarged mediastinal lymph nodes in this study. Many benign lesions, such as sarcoidosis and tuberculosis, had high FDG uptake, possibly a trend that the size of the lymph nodes seems to have some diagnostic value.
Highlights
Unexplained mediastinal lymphadenopathy is not uncommon in clinical
Some studies reported the sensitivity for Med, transbronchial needleContrast Media & Molecular Imaging aspiration (TBNA), EBUS-TBNA, and EUS-FNA in detecting malignancy were 80%, 78%, 89%, and 91%, respectively, and the specificity were 100%, 100%, 100%, and 100%, respectively [1,2,3,4]. e reason for the difference of the sensitivity may be related to the biopsy methods which could not access all the lymph nodes in mediastinum
Patients with enlarged mediastinal lymph nodes of unknown etiology and 18F-FDG PET/computed tomography (CT) scans were included in this retrospective study. e following inclusion criteria were used to select patients: (1) the enlarged mediastinal lymph nodes were defined as the long axis >1 cm or generalized pulmonary hilar enlargement on CT images; (2) the enlarged mediastinal lymph nodes had higher FDG uptake than that of the adjacent blood pool; (3) the patients had not undergone treatment; (4) clinical data were complete, and formal follow-up was recorded; (5) histopathologic diagnosis was recorded as the final golden standard
Summary
Unexplained mediastinal lymphadenopathy is not uncommon in clinical. Some patients visit a doctor due to dysphagia, hoarseness, or enlarged lymph nodes occasionally found in the physical examination. e symptoms may be caused by enlarged lymph nodes that compress the esophagus and recurrent laryngeal nerves. Ere are invasive methods for evaluation of abnormal mediastinal lymph nodes, including mediastinoscopy (Med) [1], thoracoscopy [1], transbronchial needle. Some studies reported the sensitivity for Med, TBNA, EBUS-TBNA, and EUS-FNA in detecting malignancy were 80%, 78%, 89%, and 91%, respectively, and the specificity were 100%, 100%, 100%, and 100%, respectively [1,2,3,4]. E reason for the difference of the sensitivity may be related to the biopsy methods which could not access all the lymph nodes in mediastinum. Med and EBUSTBNA could not reach prevascular, subaortic, paraaortic, paraesophageal, and pulmonary ligament nodes [5]. These methods can obtain pathological results and have high specificity, they are invasive and may lead to complications. TBNA can lead to mediastinal gas, bleeding, infection, and so on, while these incidence rates are low in EBUS-TBNA
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
