I. Palm et al., Brain metastases of lung cancer: Diagnostic accuracy of positron emission tomography with fluoro-deoxyglucose (FDG-PET), MED KLIN, 94(4), 1999, pp. 224-227
The value of FDG-PET in oncology is currently investigated in clinical stud
ies. There is only limited information on the usefulness of FDG-PET in the
evaluation of distant metastases of lung cancer. The purpose of the present
prospective investigation was to determine the diagnostic accuracy of FDG-
PET in the detection of brain metastases of lung cancer.
After intravenous injection of 220 +/- 50 MBq F-18-deoxyglucose PET acquisi
tion was carried out using an ECAT ART scanner (CTI Siemens). Images were r
econstructed using a filtered backprojection with a Hanning filter. PET dat
a were analyzed by visual interpretation of coronal, sagittal and transvers
al slices. PET scans were interpreted by two experienced nuclear medicine p
hysicians without prior knowledge of the results of other imaging studies o
r clinical data.
Between March 1997 and July 1998 whole-body PET was performed in 417 patien
ts with suspected lung cancer. 402 patients were used for statistical analy
sis. Based on conventional brain imaging with CT (occasionally MRI), brain
metastases were suspected in 17 patients (prevalence 4.2%). For FDG-PET alo
ne, sensitivity was 82% (14/17) and specificity 38% (14/37). Therefore, dia
gnostic accuracy of FDG-PET in detection of brain metastases was 93.5%. The
low specificity of FDG-PET can be explained by reduced tracer uptake mainl
y due to brain infarction or vascular encephalopathy in this group of elder
ly patients.
Our results indicate that due to its low specificity FDG-PET is not useful
for the evaluation of brain metastases in the primary staging of patients w
ith lung cancer.