Aim: This study aimed to examine the imaging characteristics and clinical and MRI correlates of brain 18F-fluorodeoxyglucose (18F-FDG)-PET imaging in patients with paraneoplatic neurological syndrome.
Materials and methods: Data of patients diagnosed with paraneoplastic neurological syndrome were retrospectively reviewed using the electronic medical records of the patients, looking specifically at records of hospital stays, laboratory findings and imaging reports. Both brain MRI and 18F-FDG-PET imaging characteristics were analyzed and compared.
Results: A total of 19 patients (ages 26–78; 13 female and six male patients) with clinical diagnoses of PNS were analyzed in this study. Limbic encephalitis (paraneoplastic limbic encephalitis) was found in 10 patients, seven of whom had a diagnosis of cancer. Brain 18F-FDG-PET showed bilaterally increased mesial temporal 18F-FDG uptake in eight of 10 patients with limbic encephalitis; seven of these eight patients exhibited memory loss. There was also a notable reduction in general cortical 18F-FDG uptake (including in the primary visual cortex) in six of the 10 patients with limbic encephalitis; three of the six patients had their primary motor cortices spared, two of them being spared bilaterally. Five of the seven limbic encephalitis patients with diagnosed cancer and two of the three without it had the aforementioned cortical and temporal lobe findings. Of the eight patients with onconeuronal antibodies, seven had temporal lobe enhancement and a total of six had diffuse cortical dysfunction. One patient with paraneoplastic limbic encephalitis without antibodies had demonstrated severely increased 18F-FDG uptake in both occipital lobes extending to the temporal lobes. The other patient without antibodies had a normal PET scan. Only one patient among four with paraneoplastic cerebellar degeneration had demonstrated decreased cerebellar uptake on 18F-FDG-PET that correlated with atrophy of the cerebellar vermis on MRI. Three patients had a clinical diagnosis of sensory neuropathy, of whom one demonstrated mild bilateral decrease in 18F-FDG uptake in the mesial temporal lobes, one showed notable increase in 18F-FDG uptake in the right mesial temporal lobe (with normal MRI) and the other had a normal brain 18F-FDG-PET. One patient was found to have cerebellar findings paired with oculomotor findings and showed increased 18F-FDG uptake in the cerebellum. One patient with stiff-person syndrome had normal brain 18F-FDG-PET.
Conclusion: The pattern of abnormalities in the brain 18F-FDG-PET images usually correlates well with the corresponding clinical settings and presentations. Although quite frequently findings correlate with those of MRI, at times 18F-FDG-PET can demonstrate functional abnormality in the absence of any MRI finding, which could give a therapeutic window before anatomical changes set in.