Nineteen studies were included. All were assessed as having a high risk of bias. In most studies follow-up was at end of intervention; one study had a three-month follow-up.
Aerobic exercise (four RCTs, two before and after studies and one experimental study design, n=113): Three studies reported statistically significant improvement in one or more measure of walking speed in the aerobic exercise group. The effect on fatigue varied between studies. Quality of life improved in three studies. There was improved fitness in all studies on a range of different measure.
Progressive resistance exercise (five before and after studies, n=59): The effect on activity level varied between studies; there was statistically significant improvement in some studies and not others following the intervention. Three studies that assessed fatigue reported statistically significant improvement and in all five studies at least one measure of strength significantly improved.
Combined exercise (four RCTs, n=301): One study that assessed balance reported no statistically significant improvement. None of the studies reported a statistically significant improvement in fitness. One study that assessed strength reported a significant improvement following the intervention.
Other interventions (two controlled studies and one case study, n=99): Fatigue significantly improved with participation in yoga. Breathing enhanced upper extremity exercises improved activity levels and pulmonary function. Aquatics exercise improved maximal oxygen consumption in two participants and improved fatigue in one person but not the other.