Misdiagnoses of 11C-choline combined with 18F-FDG PET imaging in brain tumours

Abstract

The widely applied F-FDG is known for its disadvantage in brain tumour PET imaging because of its high background uptake. C-choline can achieve high contrast of brain tumour imaging and was expected to have higher sensitivity and specificity. We analysed the misdiagnoses in C-choline PET imaging in brain tumours with the aim of improving the accuracy of diagnosis with C-choline PET imaging. We selected 10 patients proven to have been misdiagnosed on the basis of histopathological correlation and clinical follow-up among 94 patients (110 studies) who underwent C-choline PET/CT for diagnosed or suspected brain tumour between 23 March 2005 and 8 February 2007. C-choline PET imaging were performed on a Biograph Sensation 16 PET/CT scanner (Siemens Medical Systems), F-FDG imaging was also performed as reference. Of all 10 misdiagnosed patients, five were false positive (one abscess, one tuberculosis, one benign gliocyte proliferation, one inflammatory granuloma and one demyelination), four were false negative (two metastases from lung cancer, one lymphoma, one grade II glioma) and one was misdiagnosed by wrong interpretation due to lack of experience. The rate of false positives was (5/110) 4.55%; the rate of false negatives was (4/110) 3.64%; the accuracy of C-choline alone was (93/110) 84.5%; by comparison, the accuracy of F-FDG alone was (78/110) 70.9%. C-choline imaging has a certain rate of false positivity and false negativity. With proper application, C-choline might have greater potential than F-FDG for brain tumour PET imaging.