Many adults who survive community-acquired bacterial meningitis experience persistent neurologic or cognitive sequelae. A recent study from the Netherlands evaluated whether these long-term complications could be attributed to postinfectious autoimmune encephalitis.
In a cross-sectional observational study, researchers assessed the prevalence of neuronal autoantibodies in 118 adults who had recovered from community-acquired bacterial meningitis. The objective was to determine whether autoimmune mechanisms contributed to residual deficits.
Study Design and Methods
Patients were selected from the MeninGene cohort, a prospective nationwide study, and its follow-up arms: the MeninGene Serial Meningitis Sampling (SMS) and MeninGene Recall studies. Serum samples were collected during follow-up at 7 days, 3 months, or 1 to 5 years after hospital admission. A total of 42 patients participated in early follow-up, and 76 participated in late follow-up; 1 individual participated in both.
Follow-up testing was performed on serum only, as cerebrospinal fluid (CSF) was not collected at those time points. However, some patients had CSF tested at hospital admission, allowing retrospective comparison in a limited number of cases. Autoantibody detection followed a multi-step process beginning with immunohistochemistry (IHC) on sagittal rat brain sections. Samples showing reactivity underwent immunocytochemistry (ICC) on live hippocampal neurons, followed by cell-based assays (CBAs) or immunoblots, depending on the pattern and suspected target.
Patient Characteristics
The median age of participants was 59 years (interquartile range [IQR], 45–65 years), and 44% (52 of 118) were female. A causative pathogen was identified in 92% (108 of 118) of patients. The most common organism was Streptococcus pneumoniae (83%, 98 of 118), followed by Neisseria meningitidis (3%, 4 of 118), Haemophilus influenzae (2%, 2 of 118), Streptococcus agalactiae (2%, 2 of 118), Listeria monocytogenes (1%, 1 of 118), and Streptococcus salivarius (1%, 1 of 118). No pathogen was identified in 8.5% (10 of 118) of cases.
At hospital discharge, cognitive impairment was observed in 24% (24 of 100) and focal neurologic deficits in 13% (14 of 109). At long-term follow-up, 14% (11 of 76) had cognitive impairment and 32% (24 of 76) had neurologic deficits. Notably, the late follow-up group consisted exclusively of patients with pneumococcal meningitis, which may limit the generalizability of those specific findings to broader bacterial etiologies.
Autoantibody Testing Results
Two of the 118 patients (2%) had detectable neuronal autoantibodies in serum at follow-up.
One patient tested positive for leucine-rich glioma inactivated 1 (LGI1) antibodies via CBA. This individual was evaluated 4 years and 5 months after an episode of pneumococcal meningitis. Neurologic examination revealed left-sided sensory and motor deficits and gait disturbance. Neuropsychological assessment using the COGBAT computerized test battery revealed impairment in working memory, short-term memory, cognitive flexibility, and processing speed. No baseline serum or CSF was available for retrospective testing, and the patient declined further clinical follow-up. The possibility of preexisting autoimmunity or unrelated pathology cannot be excluded.
A second patient demonstrated antibody reactivity suggestive of an unknown extracellular target. No known antigen was identified through CBAs. The same staining pattern was observed in serum collected at both admission and follow-up. No neurologic or cognitive deficits were noted at follow-up, and the antibody was not considered clinically relevant.
None of the patients tested positive for antibodies to N-methyl-D-aspartate (NMDA) receptors.
Interpretation and Limitations
According to the authors, the low prevalence of neuronal autoantibodies in this cohort indicates that postinfectious autoimmune encephalitis is not a substantial contributing factor to long-term cognitive impairment or neurologic deficits following bacterial meningitis. The findings support the interpretation that acute brain injury related to the infection itself is a more likely mechanism underlying persistent symptoms.
Several limitations are noted. Follow-up testing was performed on serum only; CSF, which may be more sensitive for detecting certain antibodies, was not available at follow-up. Simultaneous serum and CSF testing is generally preferred to reduce false-negative results. Comprehensive cognitive testing using the COGBAT battery was only conducted in the late follow-up group, limiting interpretability for early-stage participants. Moreover, patients were not selected based on secondary deterioration, so autoantibody prevalence might be underestimated in more symptomatic populations.
The detection of LGI1 antibodies years after infection raises the possibility of preexisting autoimmunity or coincidental findings. In the case of unknown autoantibodies, the lack of associated clinical symptoms and the presence of antibodies at baseline suggest no causal link to the bacterial meningitis episode.
Conclusion
Neuronal autoantibodies were uncommon among adults recovering from community-acquired bacterial meningitis. These data do not support postinfectious autoimmune encephalitis as a frequent mechanism for long-term cognitive or neurologic sequelae in this population. Further studies are needed to explore targeted follow-up strategies for high-risk subgroups and determine the value of autoantibody screening in selected symptomatic cases.