Long-Term Cognitive, Functional, and Patient-Reported Outcomes in Patients With Anti-NMDAR Encephalitis

Background and Objectives:

Anti-NMDA receptor (anti-NMDAR) encephalitis generally manifests in young adults. Although 80%–90% returns to independence, the majority experience persistent cognitive and psychosocial difficulties. Studies have demonstrated that cognitive recovery may continue for years; the temporal trajectory is largely unknown, as are factors influencing cognitive/ psychosocial recovery. Objectives were to (1) describe the cognitive recovery trajectory, (2) assess self-reported outcomes, (3) identify factors relating to outcome, and (4) explore the relation between cognitive and self-reported outcomes, and participation. 

Methods: 

We performed a large-scale cross-sectional and prospective cohort study. We addressed our nationwide cohort, provided they were (1) older than 16 years, (2) independent preillness, and (3) able to perform cognitive tests and/or self-report. Patients completed Patient-Reported Outcome Measures and neuropsychological assessments (memory, language, perception and construction, and attention and executive functions), and functional outcomes were established (modified Rankin Scale [mRS] score and return-to-work/-education). Outcomes were compared with references and between groups based on clinical characteristics and functional outcomes (T-tests for normalized data and nonparametric tests for patient-reported data). Recovery was visualized by plotting outcomes against time-of-assessment. 

Results: 

We included 92 patients (age 29 ± 2 years; 77% female). Cognitive scores improved with time-of-assessment, up to 36 months after diagnosis (R = 0.35, p = 0.022), with the most enhanced improvement in the first 6 months. This result could be reproduced in prospective patients (n = 12). Beyond 36 months (n = 44), 34% of patients had a persistent impairment (z-score <−1.5 SD) and 65% scored below-average (<−1 SD) in 1 or more cognitive domains, despite a “favorable” outcome measured by mRS (≤2) in the majority (91%). Most affected were memory (mean −0.67 ± 0.89 SD, p = 0.25) and language (−0.75 ± 1.06 SD, p = 0.23). Self-reported complaints remained in emotional well-being (mean 72 ± 25 SD vs norm 82 ± 33 SD, p < 0.001), social functioning (73 ± 26 SD vs 84 ± 22 SD, p < 0.001), energy levels (57 ± 19 SD vs 69 ± 19 SD, p < 0.001), and quality of life (0.85 ± 0.14 SD vs 0.93 ± 0.11 SD, p < 0.001). Many patients did not resume school/work (30%) or needed adjustments (18%). Resuming school/work related to processing speed (−0.14 ± 0.78 SD vs −0.84 ± 1.05 SD, p = 0.039) and well-being (EuroQol 5 Dimensions 5 Levels median 0.90 vs 0.81, p = 0.016). 

Discussion: 

Recovery from anti-NMDAR encephalitis may continue for 3 years, with risk of persisting cognitive deficits, notably in memory and language, and sequelae in social functioning, energy levels, and well-being. The frequently applied outcome measure mRS does not fully capture outcomes. Almost half of patients struggled resuming school/work, associated with cognitive deficits and well-being.