Cognitive performances in people with neuromyelitis optica spectrum disorder: a systematic review and meta-analysis

Abstract

BACKGROUND: Recent literature indicates that neuromyelitis optica spectrum disorder (NMOSD) patients commonly experience cognitive impairments (CI). This review aimed to compare cognitive functions between NMOSD patients and healthy controls (HCs) and synthesize the overall prevalence of CI in NMOSD patients. METHOD: PubMed/MEDLINE, Embase, Scopus, and Web of Science were searched systematically up to March 24, 2025, to identify relevant articles evaluating the cognitive functions of NMOSD patients. Meta-analysis was conducted using the random-effect model to determine a) the pooled prevalence of CI in NMOSD patients and b) the pooled standard mean difference (SMD) of cognitive performance between NMOSD patients and HC, with their 95 confidence interval by R software version 4.4.0. RESULTS: Sixty-three studies, encompassing 3370 NMOSD patients and 4641 HCs, met the inclusion criteria. Meta-analysis revealed that the pooled prevalence of CI was 45.4 (95 CI; 39.6 to 51.2, I(2)=88 ) among NMOSD patients. The analysis demonstrated that NMOSD patients had significant poor performances on Montreal Cognitive Assessment (MoCA) (SMD=-1.53, 95 CI: -2.03 to -1.04, p < 0.001, I(2)=93 ), Brief Visuospatial Memory Test-Revised (BVMT-R) (SMD=-1.14, 95 CI: -1.69 to -0.59, p < 0.001, I(2)=86 ), Mini Mental State Examination (MMSE) (SMD=-1.1, 95 CI: -1.65 to -0.55, p < 0.001, I(2)=85 ), Symbol Digit Modalities Test (SDMT) (SMD=-1.03, 95 CI: -1.37 to -0.69, p < 0.001, I(2)=92 ) than HCs. CONCLUSION: This comprehensive review indicated that NMOSD patients exhibited worse cognitive performance compared to healthy individuals. Additionally, CI is often a complication of NMOSD. These findings accentuate the imperative of integrating cognitive assessments into the comprehensive evaluation of these patients.