High rate of unnecessary thymectomy and its cause. Can computed tomography distinguish thymoma, lymphoma, thymic hyperplasia, and thymic cysts?

Abstract

PurposeTo determine the non-therapeutic thymectomy rate in a recent six-year consecutive thymectomy cohort, the etiology of these unnecessary thymectomies, and the differentiating CT features of thymoma, lymphoma, thymic hyperplasia, and thymic cysts. Materials and methodsElectronic data base query of all thymectomies performed at the Massachusetts General Hospital from 2006 to 2012 yielded 160 thymectomy cases, 124 of which had available imaging. The non-therapeutic thymectomy rate (includes thymectomy for lymphoma and benign disease) was calculated. Preoperative clinical and CT imaging features were assessed by review of the in-house electronic medical record by 2 thoracic surgeons and 2 pathology-blinded radiologists, respectively. ResultsThe non-therapeutic thymectomy rate of 43.8% (70/160) was largely secondary to concern for thymoma and was comprised of lymphoma (54.3%, 38/70), thymic bed cysts (24.3%, 17/70), thymic hyperplasia (17.1%, 12/70), and reactive or atrophic tissue (4.3%, 3/70). Among these four lesions, there were significant differences in location with respect to midline, morphology, circumscription, homogeneity of attenuation, fatty intercalation, coexistent lymphadenopathy, overt pericardial invasion, and mass effect (p<0.001). True thymic cysts ranged in attenuation from −20 to 58Hounsfield units (HU), with a mean attenuation of 23HU. ConclusionThe high rate of unnecessary thymectomy was due to misinterpretation of thymic cysts, thymic hyperplasia, and lymphoma as thymoma on chest CT. This study demonstrates differentiating features between thymoma, lymphoma, thymic hyperplasia, and thymic cysts on chest CT which may help triage more patients away from thymectomy toward less invasive and non-invasive means of diagnosis and thereby lower the non-therapeutic thymectomy rate.