Severe MOGAD in Children: ICU Study Reveals Seizures and Brain Pressure Risks

Myelin oligodendrocyte glycoprotein antibody disease, known as MOGAD, has long been viewed as a demyelinating condition of the central nervous system that most children recover from fully, particularly when treated promptly with high-dose intravenous steroids. A new retrospective study from Texas Children’s Hospital, published in the journal Neurocritical Care, is now drawing attention to a subgroup of pediatric patients whose disease is severe enough to require intensive care, and whose clinical courses are complicated by seizures, status epilepticus, and dangerously elevated pressure inside the skull. The findings suggest that recognizing and treating these secondary complications early may be as important as controlling the underlying autoimmune inflammation itself.

The research team, led by investigators at Baylor College of Medicine and Texas Children’s Hospital, reviewed the medical records of all pediatric patients diagnosed with MOGAD at their institution between January 2017 and December 2024. To be included, patients had to experience a core clinical demyelinating event and test positive for MOG-IgG antibodies on clinical cell-based assays, with either a clear positive titer of at least 1:100 or a low-positive titer supported by consistent clinical or magnetic resonance imaging features, in line with the international consensus diagnostic criteria established in 2023. Patients with isolated optic neuritis were excluded from the comparison group because their disease course differs fundamentally from other phenotypes and rarely involves seizures, hydrocephalus, or intracranial hypertension.

Of the 124 pediatric patients with MOGAD identified during the eight-year window, 37 required admission to the pediatric intensive care unit, driven by altered mental status, seizures, or the rapid onset of weakness. Sixteen of these critically ill children had to be intubated because of altered consciousness or seizures, with a median duration of mechanical ventilation of three days. Fever was strikingly common in the ICU group, present in 24 of the 37 patients, and cerebrospinal fluid white blood cell counts were significantly higher among those admitted to intensive care compared with those managed on general wards. The ICU cohort also skewed male, and children with the cerebral cortical encephalitis phenotype were dramatically overrepresented, accounting for nearly a third of ICU admissions compared with under five percent of non-ICU cases.

Perhaps the most consequential finding concerns intracranial pressure. Twelve patients showed neuroimaging evidence of increased intracranial pressure, including four with signs of herniation or cerebral shift, five with optic nerve head protrusion and posterior globe flattening in the absence of clinical optic neuritis, and others with sulcal effacement. When the investigators combined imaging evidence, elevated opening pressures on lumbar puncture, and the use of hyperosmolar therapy, they identified 16 unique patients, more than 40 percent of the ICU cohort, with evidence of elevated intracranial pressure. Seven children received hyperosmolar therapy with hypertonic saline, and two underwent placement of invasive intracranial pressure monitoring devices, one of which recorded pressures as high as 50 millimeters of mercury and remained in place for 14 days to guide pharmacologic treatment.

Seizures were another dominant feature of critical MOGAD. Eighteen of the 37 ICU patients experienced seizures, and eleven developed status epilepticus, including one child who met criteria for new-onset refractory status epilepticus, a life-threatening form that resists standard antiseizure medications. Notably, five patients had electrographic seizures visible only on continuous electroencephalography, with no clinical correlate, and all five carried the cerebral cortical encephalitis phenotype. Twenty-seven patients underwent continuous EEG monitoring, revealing a heterogeneous mix of background abnormalities, focal slowing, interictal epileptiform discharges, and rhythmic and periodic patterns on the ictal-interictal continuum, including lateralized and generalized periodic discharges and rhythmic delta activity with and without superimposed spikes. Importantly, the authors found no specific EEG biomarker that reliably distinguishes MOGAD from other autoimmune encephalopathies, contradicting earlier suggestions that the anterior temporal lobe is spared from ictal activity in this disease.

Treatment patterns in the cohort reflected both the promise and the uncertainty of current immunotherapy practice. Most children, 32 of the 37, received immunotherapy during their ICU stay, with intravenous methylprednisolone as the standard first-line agent, alone in 26 cases and combined with intravenous immunoglobulin or plasmapheresis in others. Thirteen patients were adequately treated with first-line therapy, but the remaining nineteen required escalation, most commonly to plasmapheresis or intravenous immunoglobulin, with a median time to escalation of three days. One severely affected child with markedly elevated interleukin-6 levels received the IL-6 receptor antagonist tocilizumab after failing steroids, plasmapheresis, and immunoglobulin, and another received intrathecal dexamethasone. The authors argue that in life-threatening presentations, clinicians should consider early dual immunotherapy, and that emerging data on IL-6 receptor blockade may justify earlier use of tocilizumab in refractory cases.

The study also exposes a diagnostic trap that carries real consequences. Five ICU patients initially received no immunotherapy because their fever and elevated cerebrospinal fluid white cell counts led clinicians to suspect viral encephalitis. Although all five appeared to return to their neurologic baseline without intervention, three later deteriorated and required readmission with new neurologic symptoms before finally receiving immunotherapy. The authors emphasize that fever and cerebrospinal fluid pleocytosis, typically associated with infection, frequently accompany MOGAD attacks and may simply reflect the intensity of neuroinflammation. Their recommendation is that when MOGAD is suspected in a critically ill child, high-dose intravenous steroids should not be delayed while infectious workups proceed in parallel, with infectious disease specialists involved to help weigh the risks.

Outcomes data underscore the stakes of getting this right. Children admitted to the ICU had significantly worse scores on the Functional Status Scale and the Pediatric Cerebral Performance Category scale at follow-up than those managed outside the ICU, were far more likely to require inpatient rehabilitation, and were more than six times as likely to be diagnosed with epilepsy later in life. Intriguingly, the risk of future epilepsy did not correlate with whether a child had seizures or even status epilepticus during the acute illness. Instead, it tracked with evidence of elevated intracranial pressure: 71 percent of children who went on to develop epilepsy had clinical or imaging concern for raised intracranial pressure, compared with just 15.5 percent of those who did not. The authors speculate that secondary cerebral injury, including encephalomalacia and cortical laminar necrosis visible on follow-up MRI, may create the substrate for epileptogenesis, making intracranial hypertension a modifiable target for preventing long-term harm.

The study is not without limitations. Its retrospective design means the clinical approach to assessing and treating intracranial pressure was not standardized, and the authors acknowledge that many patients with evidence of raised pressure did not receive pressure-directed therapy, possibly because the complication was underrecognized in the setting of autoimmune cerebral inflammation. The single-center, tertiary-quaternary referral setting also introduces referral bias. Still, the investigators argue that the findings make a compelling case for systematic neuromonitoring in critically ill children with MOGAD, including routine measurement of opening pressure during lumbar puncture, cautious use of fundoscopy, which proved insensitive in this cohort, and greater adoption of transcranial doppler ultrasonography, which showed elevated middle cerebral artery velocities suggestive of early hyperemia in the four patients tested. They call for early multidisciplinary involvement spanning critical care, neurocritical care, neurosurgery, and infectious disease, and for prospective studies to determine whether aggressive, standardized management of seizures and intracranial pressure can improve the long-term outlook for these children.

Subject of Research: Critical illness, seizures, and intracranial hypertension in pediatric MOG antibody disease requiring intensive care

Article Title: Pediatric MOGAD in the Intensive Care Unit: A Case Series

Article References: Runco, A. D., Davila-Williams, D., Moeller, K. K., Sandweiss, A. J., Ballinger, E., Shukla, N., Riviello, J. J., Erklauer, J. C., & Fisher, K. S. (2026). Pediatric MOGAD in the Intensive Care Unit: A Case Series. Neurocritical Care. https://doi.org/10.1007/s12028-026-02653-6

Image Credits: AI Generated

DOI: 10.1007/s12028-026-02653-6

Keywords: MOGAD, pediatric neurology, neurocritical care, intracranial pressure, status epilepticus, demyelinating disease, continuous EEG, immunotherapy, cerebral cortical encephalitis, ADEM, epilepsy risk, intensive care