Aim: The objective of this study was to evaluate the clinicoradiological findings of thoracic Castleman disease.
Methods: The study included 34 patients (22 male and 12 female patients; mean age, 32 [SD, 18.1] years) with thoracic Castleman disease. Clinicoradiological findings of the 34 patients were analyzed. Regarding computed tomography findings, lesion number, location, degree of enhancement (moderate, >20 Hounsfield units than back muscle enhancement; high, >40 Hounsfield units), and associated findings were recorded.
Results: Of 34 patients, hyaline-vascular type (HVT) was found in 27 patients (79%), plasma cell type (PCT) in 5 patients (15%), and mixed type (6%) in 2 patients. In HVTs (n = 27), lesions were found, in decreasing order, in the lower neck (n = 9, 33%), pulmonary hilum (n = 6, 22%), and the upper paratracheal area (n = 4, 15%). Ten (37%) of 27 HVT patients had symptoms, whereas all (100%) with PCT had generalized symptoms. In 26 (96%) of 27 HVT patients, disease was unicentric, whereas it was multicentric in all PCT patients. Moderate to high degree of lesion enhancement was seen in 22 (92%) of 24 HVT patients and 4 (80%) of 5 PCT patients. Feeding vessels or draining veins were identified in 12 (44%) of 27 HVT patients and 2 (40%) of 5 PCT patients. The diseases were cured with surgical removal in HVT, whereas they showed variable prognosis in PCT.
Conclusions: Irrespective of subtypes, Castleman disease is characterized radiologically by unicentric or multicentric enhancing lymph node enlargement; in HVT, they show good prognosis after surgical treatment, but in PCT, they show variable prognosis.
From the Departments of *Radiology and Center for Imaging Science and †Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea.
Received for publication June 28, 2012; accepted August 22, 2012.
Reprints: Kyung Soo Lee, MD, PhD, Department of Radiology Samsung Medical Center, Sungkyunkwan University School of Medicine 50 Ilwon-dong, Gangnam-gu, Seoul 135-710, Korea (e-mail: email@example.com).
The authors have no conflicts of interest.