Recently PET/CT systems have become popular and been introduced at many institutions throughout Japan and the clinical importance should increase for managing many kinds of malignant diseases including lung cancers. A 2-[fluorine-18] fluoro-2-deoxy-D-glucose (FDG) has been the most important tracer in clinical usage and seems to play a major role during these several years. Since March 2004 we began operating the first PET/CT scanner in Japan with a 4-slice multi-detector CT (MDCT) and we have now introduced a second PET/CT scanner with 8-slice MDCT and have been performing 25 to 30 PET studies per day. Although some problems have been reported based on extensive experience, the ability of FDG-PET for staging lung cancers is still indispensable, like other conventional modalities such as CT, MRI and SPECT. It is usually hard to differentiate malignant from benign lesions except for lesions with markedly elevated FDG accumulation. FDG accumulation in macrophages and sarcoid reaction may lead to false positive results. It is essential to interpret FDG-PET findings with the morphological characteristics of CT and/or MRI. FDG-PET has been considered superior to CT for evaluating N factor. However recent studies mention that FDG-PET might show considerable both false-positive and false-negative results particularly in patients with small primary nodules. FDG-PET is useful for integrating and simplifying many kinds of examination methods to determine the existence of distant metastases. Physiological FDG accumulation in brown fat and skeletal muscles, misregistration artifacts between CT and PET and metal devices such as pace-makers might cause incorrect results. In this paper we demonstrate the pitfalls of FDG-PET in managing patients with lung cancer. © 2007 The Japan Lung Cancer Society.