Objective:
The goal of our study was to investigate the relationship between autoimmune encephalitis (AE) phenotype domains and in-hospital and discharge outcomes.
Background:
AE is a rare cause of subacute encephalopathy, presenting with amnesia, behavioral changes, and psychiatric manifestations. Despite advancements in neuronal autoantibody identification, diagnosis often relies on phenotypic presentation due to delays in antibody confirmation. Phenotypic features present early and may help determine urgency of treatment; however, there is insufficient literature defining the relationship between predominant phenotypes and inpatient outcomes. We hypothesized that phenotype-based presentation in AE would be associated with distinct patterns of inpatient severity, resource utilization, and discharge outcomes.
Design/Methods:
Hospitalizations for AE were abstracted from the 2016-2022 National Inpatient Sample and categorized based on non-mutually exclusive phenotype groups of seizure, cognitive change/confusion, movement, or new-onset psychosis. Specific phenotype combinations were also examined. Outcomes investigated included in-hospital mortality, hospital length of stay (LOS) and cost, discharge disposition, and in-hospital severe events (shock, mechanical ventilation).
Results:
There were an estimated 19,175 hospitalizations for AE; of those, seizure was the most frequent phenotype (Weighted N: 8,405), followed by cognitive/confusion (Weighted N: 6,495). The cognitive/confusion phenotype had the worst clinical outcomes, with an inpatient mortality rate of 3.85%, mechanical ventilation was performed in 19.5% of hospitalizations, and the average LOS and hospital cost were 20.26 days (SD: 0.73 days), and $77,735 (SD $3,647), respectively. Rare phenotype combinations demonstrated greatest hospital burden, with movement and new psychosis combination group having the longest average LOS (47.64 ± 6.27 days) and highest average cost ($161,324 ± 8,501).
Conclusions:
Amongst phenotypic groups, seizure and cognitive/confusion traits were most prevalent, whereas cognitive/confusion phenotypes were associated with the worst outcomes. Phenotype combinations were linked to increased hospital burden. These findings suggest that clinical phenotypes may serve as important prognostic factors in AE.
Generative AI Usage
No, did not use generative AI in the drafting or editing in this abstract.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.