18F-FDG PET/CT metabolic patterns and clinical parameters correlate with disease severity in autoimmune encephalitis

Purpose To investigate the correlation of 18 F-FDG PET/CT metabolic patterns and clinical parameters with disease severity in patients with autoimmune encephalitis (AE). Methods We retrospectively analyzed 35 AE patients who underwent 18 F-FDG PET/CT at the Second Affiliated Hospital of Soochow University between March 2021 and December 2023. Clinical and auxiliary examination data were collected and analyzed using the t -test, Mann–Whitney U test, Fisher’s exact test, and Spearman correlation. Metabolic patterns on PET/CT were assessed both visually and using Cortex ID software across severity groups. Results Among the 35 AE patients (21 males, 14 females), 16 were classified as mild and 19 as moderate-to-severe. Compared to the mild group, the moderate-to-severe group exhibited significantly elevated neutrophil percentages and neutrophil-to-lymphocyte ratio (NLR), alongside reduced lymphocyte percentages, lymphocyte counts, and serum calcium ion levels (Z = −2.75 to 2.30; all p < 0.05). 18 F-FDG PET/CT demonstrated a higher abnormality detection rate than brain MRI ( p < 0.05). The most common imaging finding was hypometabolism in the brain parenchyma, predominantly involving the parietal and occipital lobes. Furthermore, CASE scores demonstrated a negative correlation with Z-scores in the right medial frontal lobe, right superior parietal lobe, and bilateral lateral occipital lobes ( r = −0.456 to −0.567; all p < 0.05). Conclusion In addition to clinical symptoms, specific cerebrospinal fluid and serum biomarkers may help evaluate AE severity. 18 F-FDG PET/CT demonstrates a higher abnormality detection rate than brain MRI, and we preliminarily observed a correlation between regional hypometabolism and clinical severity scores.