Early regional cerebral grey matter damage predicts long-term cognitive impairment phenotypes in multiple sclerosis: a 20-year study

Abstract

Despite grey matter atrophy in cortical and subcortical regions has been related to cognitive impairment in multiple sclerosis, only a few studies evaluated its predictive value for alterations in the long-term. We aimed to determine early predictors of cognitive status after 20 years of multiple sclerosis. In this longitudinal retrospective study, participants underwent a 1.5 T MRI scanning at diagnosis (T0) and after two years (T2), which included the evaluation of regional grey matter volume loss patterns. All individuals with multiple sclerosis underwent a comprehensive neuropsychological assessment at the end of the study and were classified considering their global and specific cognitive domains status (memory, attention/information processing speed, executive functioning). Clinical and MRI characteristics were assessed as predictors of long-term cognitive impairment. Analysis of covariance, t-test, unadjusted and adjusted (for age, sex, disease duration, volume of white matter lesions, volume of cortical lesions) logistic regression were conducted. One hundred seventy-five people with multiple sclerosis (118 females; mean ± SD age at the end of study = 47.7 ± 9.4 years) clinically followed for 20 years from onset (mean ± SD = 19.9 ± 5.1) were evaluated. At the end of the study, 81 (47%) were classified as cognitively impaired: 38 as mildly impaired (22%), and 43 as severely impaired (25%). In particular, 46 were impaired in memory (27%), 66 were impaired in attention/information processing speed (38%), and 71 were impaired in executive functioning (41%). Regression models identified precuneus (adjusted odds ratio = 3.37; P < 0.001), insula (adjusted odds ratio = 2.33; P = 0.036), parahippocampal gyrus (adjusted odds ratio = 2.07; P < 0.001) and cingulate (adjusted odds ratio = 1.81; P = 0.009) as the most associated regions with global cognitive impairment and domains-specific cognitive alterations after a mean of 20 years of multiple sclerosis, after adjusting for demographic and clinical variables as well as for focal white matter and grey matter damage. Early grey matter volume loss of specific cortical and deep grey matter regions predicts global and domain cognitive alterations after 20 years from multiple sclerosis diagnosis.

Keywords: cognitive impairment; multiple sclerosis; regional atrophy.

Graphical Abstract

Figure 1

Prevalence of global and domain cognitive status of people with multiple sclerosis. Global and domain-specific cognitive characterization of multiple sclerosis sample after 20 years from diagnosis (N = 175). MS, multiple sclerosis; CN, cognitive normal; mCI, mild cognitive impaired; sCI, severe cognitive impaired; MEM, memory; ATT/IPS, attention/information processing speed; EF, executive functions.

Figure 2

Significant regional predictors for global cognitive status. Illustration of the brain regions of interest (ROI) that showed statistical significance in the univariable logistic regression analyses (P < 0.05) with reference to global cognitive status.

Figure 3

Forest plot results for global cognitive status. The forest plot illustrates the odds ratios (95% confidence intervals, CIs) derived from univariable logistic regression performed considering all multiple sclerosis patients (N = 175) for each cortical area associated with global cognitive status.

Figure 4

Forest plot results for domain-specific cognitive status. The forest plot illustrates the odds ratios (95% confidence intervals, CIs) derived from univariable logistic regression performed considering all multiple sclerosis patients (N = 175) for each cortical area associated with domain cognitive status (A = MEM, B = ATT/IPS, C = EF). To increase graph readability, the upper limit for the CIs of the odds ratios has been capped at 16. MEM, memory; ATT/IPS, attention/information processing speed; EF, executive functions.