Anti-N-methyl-D-aspartate receptor (anti-NMDAR) encephalitis is the most frequent autoimmune encephalitis in children and may cause long-term sequelae. Although MRI plays a key role in diagnosis and follow-up, its correlation with cognitive outcome remains unclear. We retrospectively reviewed pediatric cases of anti-NMDAR encephalitis diagnosed from 2007 to 2024, including only antibody-positive patients with at least one MRI. Clinical data encompassed presenting symptoms, EEG, therapies, relapses, and the latest neuropsychological assessment. MRI scans were evaluated by three neuroradiologists using a structured atrophy scale, and volumetric T1-weighted sequences, when available, were analyzed with Volbrain. Twenty-two patients (9 males, 13 females; median onset age 11.9 years) were included. The most frequent presenting symptoms were seizures (82%) and behavioral disturbances (59%), while EEG abnormalities were detected in 95%. At onset, 9 patients (41%) showed MRI abnormalities, most commonly fronto-parieto-occipital atrophy. Long-term cognitive follow-up (median 4.8 years), available for all patients, demonstrated that 80% of patients had normal or only mild deficits, but those with abnormal baseline MRI were more likely to develop long-term cognitive impairment, although the limited sample size precluded robust statistical analysis. Long-term MRI was available in 8 patients (median follow-up 10 years); volumetric analysis revealed parietal lobe volume loss in 6, including some with visually normal scans. Overall, baseline MRI alterations appeared associated with higher risk of subsequent cognitive deficits, while volumetric data highlighted subtle parietal atrophy even in patients with normal-appearing scans and, in some cases, without neuropsychological impairment. These findings support the complementary role of quantitative MRI and systematic neuropsychological monitoring in pediatric anti-NMDAR encephalitis.
Building similarity graph...
Analyzing shared references across papers
Loading...
Fabio Martino Doniselli
Anna Paola Savoldi
João Nuno Ramos
NeuroImage Clinical
Fondazione IRCCS Istituto Neurologico Carlo Besta
Centro Hospitalar de Vila Nova de Gaia
Building similarity graph...
Analyzing shared references across papers
Loading...
Doniselli et al. (Sun,) studied this question.
www.synapsesocial.com/papers/69abc1765af8044f7a4ea2eb — DOI: https://doi.org/10.1016/j.nicl.2026.103986