Anti-N-methyl-D-aspartate receptor encephalitis in adults: a systematic review and analysis.

Purpose: To analyze predictors of treatment outcome for anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis in adults.

Methods: We performed a comprehensive literature search of PubMed, PsycInfo, and OVID. We included 424 patients from case reports and case series. Demographics, anti-NMDAR antibodies, prodromal and presenting symptoms, diagnostic workup, and treatment variables were recorded. Inferential analyses were performed in the subset (n = 299) of those with known treatment outcomes. Multivariate multinomial logistic regression analysis for treatment outcome compared full recovery versus partial recovery and full recovery versus death.

Results: Treatment outcomes consisted of 34.67% full recovery (n = 147), 30.90% partial recovery (n = 131), 4.95% death (n = 21), and 29.48% unknown (n = 125). Speech/language abnormality and abnormal electroencephalogram (EEG) were each significantly associated with a higher relative risk for a full recovery. Treatment with intravenous immunoglobulin and plasmapheresis were each significantly associated with a higher relative risk for partial recovery. The analysis comparing death to full recovery found that catatonia was significantly associated with a lower relative risk for death. Increased age, orofacial dyskinesia, and no tumor removal were each significantly associated with a higher relative risk for death.

Conclusion: Increased age, orofacial dyskinesia, and no tumor removal were associated with a higher relative risk for death in anti-NMDAR encephalitis in adults. Clinicians should monitor and appropriately treat anti-NMDAR encephalitis with these findings to minimize the risk of death.