To evaluate and compare the clinical features, imaging, overlapping antibodies, and prognosis of pediatric and adult patients with anti-GFAP antibodies.
This study included 59 patients with anti-GFAP antibodies (28 females and 31 males) who were admitted between December 2019 and September 2022.
Out of 59 patients, 18 were children (under 18 years old), and 31 were adults. The overall cohort’s median age at onset was 32 years old, 7 for children, and 42 for adults. There were 23 (41.1%) patients with prodromic infection, 1 (1.7%) patient with a tumor, 29 (53.7%) patients with other non-neurological autoimmune diseases, and 17 (22.8%) patients with hyponatremia. Fourteen (23.7%) patients had multiple neural autoantibodies, with the AQP4 antibody being the most common. Encephalitis (30.5%) was the most common phenotypic syndrome. Common clinical symptoms included fever (59.3%), headache (47.5%), nausea and vomiting (35.6%), limb weakness (35.6%), and disturbance of consciousness (33.9%). Brain MRI lesions were primarily located in the cortex/subcortex (37.3%), brainstem (27.1%), thalamus (23.7%), and basal ganglia (22.0%). Spinal cord MRI lesions often involved the cervical and thoracic spinal cord. There was no statistically significant difference in the MRI lesion site between children and adults. Out of 58 patients, 47 (81.0%) had a monophasic course, and 4 died. The last follow-up showed that 41/58 (80.7%) patients had an improved functional outcome (mRS <3), and children were more likely than adults to have no residual disability symptoms (p = 0.001).
There was no statistically significant difference in clinical symptoms and imaging findings between children and adult patients with anti-GFAP antibodies; Patients with anti-GFAP antibodies may present with normal MRI findings or delayed MRI abnormalities, and patients with overlapping antibodies were common. Most patients had monophasic courses, and those with overlapping antibodies were more likely to relapse. Children were more likely than adults to have no disability. Finally, we hypothesize that the presence of anti-GFAP antibodies is a non-specific witness of inflammation.