Evidence for physical activity in post-stroke rehabilitation
First, animal studies have demonstrated that physical activity enhances neural plasticity and learning, processes central to functional recovery after stroke, by increasing neurotrophic substances like brain-derived neurotrophic factor, neurotransmitters such as dopamine, long-term potentiation, and possibly even neurogenesis .
Second, physical therapies are known to promote structural brain remodelling in humans , and this can influence post-stroke motor deficits. A recent systematic review indicated that repetitive practice of some common day-to-day activities lead to modest improvements in mobility and ADL in stroke patients .
However, conclusive evidence for beneficial effects of physical activity after stroke on ADL and walking speed is still missing. Most of the studies included in the 2011 meta-analysis by Brazzelli and colleagues  on physical activity-based interventions featured limited sample sizes and had a relatively short-term follow-up. Moreover, a variety of interventions were tested in diverse stroke populations with a range of outcome measures. The trials varied substantially in training intensity and frequency, total number of sessions, timing of training (acute, subacute or chronic), and locomotor impairment severity (non-ambulators to community ambulators), rendering it difficult to draw definite conclusions. Results indicate modest treatment effects associated with physical fitness training for stroke patients, particularly around interventions targeting “fitness” approaches rather than “strengthening”. Significant effects were demonstrated for an increase in maximal oxygen uptake and walking speed, with promising results for ADL (after intervention and on follow-up; both trends), and quality of life (directly after intervention). However, a subsequently published large trial on 408 stroke patients , comparing treadmill-based locomotor training either 2 or 6 months post-stroke to progressive exercise at home (non-specific, low intensity exercise intervention), did not find a significant benefit of locomotor training with regard to increasing the proportion of study participants who had higher functional walking levels (comfortable walking speed) at 1 year post-stroke. Of note, participants entered the trial only 2 months post-stroke in an ambulatory setting, and treadmill training was not aimed at enhancing cardiovascular fitness.
Third, in healthy humans undergoing physical fitness training, an increase not only in physical fitness, gait speed, and motor function but also in cognitive functions such as memory , executive functions , and language learning  have been demonstrated. Therefore, patients with not only motor deficits but also language and other cognitive dysfunctions may benefit from physical fitness training, a hypothesis that has not been addressed in most studies.
Fourth, profound effects on mood have been demonstrated by physical activity immediately after intervention in stroke . Therefore, mood and quality of life measures post-stroke might be expected to improve at long-term follow-up, possibly depending on continuation of exercise.
Fifth, post-stroke physical activity and fitness levels are low, and these low levels are associated with common post-stroke functional limitations [14, 15]. Increased fitness and physical function, shown to result from post-stroke physical activity training [8, 14], could benefit a range of other common post-stroke problems by reducing fatigue, reducing the incidence of falls and fractures, compensating for the increased energy cost of a hemiparetic gait, reducing disability and improving independence [14, 16].
Sixth, physical fitness training is beneficial for people with other comorbidities or risk factors for stroke. Systematic review evidence shows that interventions involving physical fitness training reduce blood pressure , improve vascular risk factors such as obesity  and type II diabetes , and reduce mortality in coronary heart disease. Therefore, post-stroke cardiorespiratory training, in particular, could reduce morbidity and mortality through secondary prevention of stroke and comorbid conditions.
In summary, physical fitness training, delivered as cardiovascular active, supported or unsupported ambulation training, may benefit a range of common post-stroke problems. It is likely that it does not simply provide a mechanism to increase fitness and improve walking ability, but has multiple mechanisms of action such as enhancing neurotrophins and overall plasticity and thus possibly exerts neuroplastic effects on motor, language and cognitive systems . Therefore, it provides a spectrum of plausible benefits that are relevant to many people with stroke. However, there may also be risks, such as training-induced vascular events, soft tissue injuries, fatigue, altered muscle tone, and falls , although there is good evidence that cardiovascular active training can be administered successfully even in severely affected stroke patients [8, 21, 22].