The aerobic capacity - fatigue relationship in persons with Multiple Sclerosis is not reproducible in a pooled analysis of two randomized controlled trials

Background: Fatigue is one of the most frequent symptoms of persons with Multiple Sclerosis (pwMS) but has limited treatment options. Aerobic capacity and endurance training have been discussed as relevant factors to improve fatigue. However, over the last decades, results have been equivocal. This secondary analysis of two pooled parallel group RCTs of three weeks of endurance training (high intensity interval training (HIIT) and moderate continuous training (MCT)) for pwMS aimed to (I) reproduce reported associations between aerobic capacity and fatigue on a cross-sectional and interventional level. The analysis further aimed to (II) investigate intervention effects on fatigue in a severely fatigued subgroup and (III) analyze differences in changes of fatigue between peak oxygen uptake (VO2peak) responders and non-responders. Methods: Both RCTs were conducted in the same inpatient rehabilitation clinic in Valens, Switzerland. Original primary outcomes were cognitive function (RCT1) and change in proportion of circulating regulatory T-cells (RCT2). PwMS (n = 131) with a relapsing-remitting or secondary progressive MS phenotype and Expanded Disability Status Scale score between 1 6.5 were eligible. Over the two studies participants exercised 3 5 times per week on cycle ergometers at intensities of 65 70% of maximum heart rate (HRmax) for 30 min (MCT groups) or three times per week with five 90 - 180 s intervals at intensities of 85% - 100% of HRmax and 90 s rest intervals (HIIT groups). Main outcome measures for the present secondary analysis were VO2peak measured during a cardiopulmonary exercise test and the Fatigue Scale for Motor and Cognitive Functions (FSMC), both assessed at the start and end of inpatient rehabilitation. Results: Baseline correlations did not reveal a significant association between VO2peak and FSMC. There were no significant improvements in fatigue after the HIIT and MCT in the overall sample or the subsample of severely fatigued pwMS and no significant differences in fatigue changes between VO2peak-responders and non responders. Conclusions: Our analysis did not confirm the aerobic capacity - fatigue relationship on a cross-sectional and experimental level, even when analyzing subgroups that should benefit the most according to proposed hypotheses.