Increased determinism in brain electrical activity occurs in association with multiple sclerosis

Abstract

Objective: Increased determinism (decreased complexity) of brain electrical activity has been associated with some brain diseases. Our objective was to determine whether a similar association occurred for multiple sclerosis (MS).

Methods: Ten subjects with a relapsing–remitting course of MS who were in remission were studied; the controls were age- and gender-matched clinically normal subjects. Recurrence plots were calculated using representative electroencephalogram (EEG) epochs (1–7 seconds) from six derivations; the plots were quantified using the nonlinear variables percent recurrence (%R) and percent determinism (%D). The results were averaged over all derivations for each participant, and the means were compared between the groups. As a linear control procedure the groups were also compared using spectral analysis.

Results: The mean±SD of %R for the MS subjects was 6·6±1·3%, compared with 5·1±1·3% in the normal group (P = 0·017), indicating that brain activity in the subjects with MS was less complex, as hypothesized. The groups were not distinguishable using %D or spectral analysis.

Discussion: Taken together with our earlier report that %R could be used to discriminate between MS and normal subjects based on the ability to exhibit evoked potentials, the evidence suggests that complexity analysis of the EEG has potential for development as a diagnostic test for MS.

Keywords:

• Complexity
• Multiple sclerosis
• Nonlinear modeling
• Phase space
• Recurrence plot
• Recurrence quantification analysis