ABSTRACT
<p><b>OBJECTIVE</b>To evaluate the accuracy of [18F] fluorodeoxyglucose (FDG) triple-head coincidence imaging in the setting of suspected lung cancer.</p><p><b>METHODS</b>109 patients with suspected lung cancer were enrolled in the present study. According to the diameter of the lesion (> 1.5 cm, </= 1.5 cm), patients were divided into two groups. The sensitivity, specificity and accuracy of the results were calculated and the results were compared with pathological results.</p><p><b>RESULTS</b>Of the total of 109 patients, 86 cases were confirmed as lung cancer whereas 23 cases were benign by pathological study. The sensitivity, specificity and accuracy of FDG imaging were 95% (82/86), 74% (17/23) and 91% (99/109), respectively. For the group of lesion diameter > 1.5 cm, the sensitivity, specificity and accuracy were 98% (81/83)), 63% (10/16) and 92% (91/99), respectively. While for the group of lesion diameter </= 1.5 cm, the results were 1/3, 7/7, and 8/10, respectively. The difference between the two groups was significant (P = 0.005).</p><p><b>CONCLUSIONS</b>The technique of [18F] FDG imaging is an useful method for routine evaluation of patients with suspected lung cancer. The sensitivity of detection is related to lesion size, if lesions </= 1.5 cm in diameter the results of FDG imaging should be carefully analyzed.</p>