Correlation sometimes implies causation: Possible roles of correlation analysis between 18fluorine-fluorodeoxyglucose positron emission tomography/computed tomography and thymic epithelial neoplasms

摘要

We read with interest the paper of Fukumoto et a. [1] on the assessment of the usefulness of ~(18)fluorine-fluorodeoxyglucose positron emission tomography/computed tomography (~(18)F-FDG PET/CT) in distinguishing the histo-logical subtypes of thymic epithelial neoplasms (TENs). The authors divided the early-stage TENs patient cohort according to the WHO histological classification into two groups (low- and high-risk tumors), and the focal FDG accumulation was evaluated merely through the determination of the maximum standardized uptake value (SUV_(max)). The authors found that the SUV_(max) values of the low-risk and high-risk TENs were significantly different. However, the two figures reported in the paper regarding SUV_(max) in subgroups according to a simplified WHO histological classification and the SUV_(max) of the Stages III and IV TENs show large confidence intervals, and no statistical correlation of data was performed.