The clinical and predictive value of F-FDG PET/CT metabolic patterns in a clinical Chinese cohort with autoimmune encephalitis.

AIMS

To investigate the diagnostic and predictive role of F-FDG PET/CT in patients with autoimmune encephalitis (AE) as a whole group.

METHODS

Thrty-five patients (20 females and 15 males) with AE were recruited. A voxel-to-voxel semi-quantitative analysis based on SPM12 was used to analyze F-FDG PET/CT imaging data compared to healthy controls. Further comparison was made in different prognostic groups categorized by modified Rankin Scale (mRS).

RESULTS

In total, 24 patients (68.6%) were tested positive neuronal antibodies in serum and/or CSF. Psychiatric symptoms and seizure attacks were major clinical symptoms. In the acute stage, 13 patients (37.1%) demonstrated abnormal brain MRI results, while 33 (94.3%) presented abnormal metabolism patterns. F-FDG PET/CT was more sensitive than MRI (p < 0.05). Patients with AE mainly presented mixed metabolism patterns compared to the matched controls, demonstrating hypermetabolism mainly in the cerebellum, BG, MTL, brainstem, insula, middle frontal gyrus, and relatively hypometabolism in the frontal cortex, occipital cortex, temporal gyrus, right parietal gyrus, left cingulate gyrus (p < 0.05, FWE corrected). After a median follow-up of 26 months, the multivariable analysis identified a decreased level of consciousness as an independent risk factor associated with poor outcome of AE (HR = 3.591, p = 0.016). Meanwhile, decreased metabolism of right superior frontal gyrus along with increased metabolism of the middle and upper brainstem was more evident in patients with poor outcome (p < 0.001, uncorrected).

CONCLUSION

F-FDG PET/CT was more sensitive than MRI to detect neuroimaging abnormalities of AE. A mixed metabolic pattern, characterized by large areas of cortical hypometabolism with focal hypermetabolism was a general metabolic pattern. Decreased metabolism of right superior frontal gyrus with increased metabolism of the middle and upper brainstem may predict poor long-term prognosis of AE.