Background: The evidence of clinical value of positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose (FDG) in lymph node (LN) staging in non-small cell lung cancer (NSCLC) has been shown in numerous papers. However, few studies have assessed its limitations. The aim of the present study is to clarify clinico-pathologic factors responsible for false PET results.
Methods: From July 2000 through December 2001, 71 NSCLC patients underwent both FDG PET and surgical intervention at the National Cancer Center Hospital East, Chiba. Clinical records, computed tomographic (CT) scan findings, PET findings, and histologic findings were retrospectively reviewed.
Results: Sensitivity, specificity, accuracy in nodal staging for CT were 29, 83, and 65% and for PET were 39, 79, and 66%, respectively. There were 10 (14%) false-positive PET scans and 14 (20%) false-negative PET scans. The causative factors for false-positive PET scan were: (1) inflammatory conditions in seven patients; (2) PET mis-localization of an interlobar LN as a mediastinal LN in one patient; (3) inability to distinguish the endobronchial polypoid growth of a primary tumor from a lobar LN in one patient; (4) unknown in one patient. All false-positive LNs due to inflammatory conditions showed reactive lymphoid hyperplasia histologically. The causative factors for false-negative PET scan were: (1) limitation of spatial resolution of the PET scanner in 12 patients (maximum tumor focus dimensions in false-negative LNs ranging from 1 to 7.5 mm, with an average of 3.4 mm); (2) PET mis-localization of a mediastinal LN as a hilar LN in one patient; (3) weak FDG uptake by microscopic tumor foci due to necrosis with massive bleeding in a metastatic LN in one patient.
Conclusions: Inflammatory conditions were most responsible for false-positive PET scans, and spatial resolution limitation of FDG PET was the causative factor of false-negative PET scans. Recognizing these factors in advance would be clinically helpful in accurate nodal staging with FDG PET.