The association between MRI brain volumes and computerized cognitive scores of people with multiple sclerosis

Abstract

BackgroundComputerized cognitive assessment facilitates the incorporation of multi-domain cognitive monitoring into routine clinical care. The predictive validity of computerized cognitive assessment among people with multiple sclerosis (PwMS) has scarcely been investigated. ObjectiveTo explore the associations between brain volumes and cognitive scores from a computerized cognitive assessment battery (CAB, NeuroTrax) among PwMS. MethodsPwMS were evaluated with the CAB and underwent brain MRI within 40 days. Cognitive assessment yielded age- and education-adjusted scores in 9 cognitive domains: memory, executive function, attention, information processing speed, visual spatial, verbal function, motor skills, problem solving, and working memory. The global cognitive score (GCS) is the average of all domain scores. MRI brain and lesion volumes were assessed with icobrain ms, a fully automated tissue and lesion segmentation and quantification software. Results91 PwMS were included [Age: 52.1 ± 11.7 years, 64 (70%) female, EDSS: 3.4 ± 2.0, 79 (87%) with a relapsing remitting course]. Significant correlations were found between the GCS and whole brain, white matter, grey matter, thalamic, lateral ventricles, hippocampal and lesion volumes (Correlation coefficients: 0.46, 0.40, 0.25, 0.42, −0.36, 0.21, −0.3, respectively). Regression analysis revealed that lateral ventricles and thalamic volumes were the most consistent predictors of all cognitive domain scores. ConclusionComputerized cognitive scores were significantly associated with quantified MRI. These findings support the predictive validity of multi-domain computerized cognitive assessment for people with multiple sclerosis.