Fatigue is a common problem experienced by people with multiple sclerosis (PwMS) and can impact physical, cognitive, and psychosocial aspects of daily living and quality of life. The tracking of meaningful longitudinal change in subjective fatigue that occurs as a result of MS activity may be enhanced by incorporating objective neurophysiological measures into longitudinal assessment. To examine this possibility, we examined the longitudinal relationship between visual evoked potential (VEP) measures and a variety of fatigue measures over an approximately two-year period in PwMS.

VEP measures were obtained using a checkerboard pattern-reversal paradigm. Fatigue was assessed with the Modified Fatigue Impact Scale (MFIS Global, Physical, Cognitive, and Psychosocial subscales) and the Fatigue Severity Scale (FSS) questionnaires. Multiple linear regression analyses were conducted in which the change in each fatigue scale score from baseline to follow-up (T1-to-T2) served as the outcome variables for separate models. Predictor variables included the peak latency of the P100 component of the VEP (maximum peak among the two eyes) and the inter-ocular latency (IOL) at T1, the T1-to-T2 change score for maximum VEP latency and IOL, and the fatigue score at T1 that corresponded to each outcome measure.

Prolonged baseline VEP latency was a significant predictor of the T1-to-T2 increase in MFIS Global score, and increased VEP latency from baseline to follow-up was significantly associated with MFIS Cognitive score over the same time period. Furthermore, VEP latency measures in these two models were better predictors of changes in fatigue than baseline fatigue scores were, based on the magnitude of the standardized beta coefficients. Subsequent post-hoc analyses revealed that the relationship between change in VEP latency and change in MFIS Cognitive score was evident primarily for PwMS that had elevated MFIS Cognitive score at baseline.

The present study provides novel evidence that prolonged VEP latency is predictive of worsening of global and cognitive fatigue in PwMS. VEP latency measures may therefore provide clinical utility for monitoring changes in fatigue in PwMS, when used in conjunction with other clinical tools.