‘Exercise’ (or ‘exercise training’) is defined as a subset of physical activity that is planned, structured, repetitive and performed deliberately to improve or maintain one or more components of physical fitness, which was defined in chapter 4 (p. 77). Most exercise training interventions are classified as either cardiorespiratory (to increase cardiorespiratory fitness), resistance (to improve muscle strength and/or power) or mixed (a combination of both). A defining feature of any exercise training intervention is some form of progression in the training stimulus throughout the programme; for example, where the frequency, duration and/or the intensity of exercise sessions increases week by week. In the context of therapeutic interventions, exercise training is somewhat unusual in that it can confer functional and health benefits for people of any age even if already healthy, and can continue to do so in those who are already trained.

Theoretical Benefit of Exercise

Why might exercise training be important for people post stroke? What benefits can they expect in terms of functional and psychosocial outcomes?

Rationale for Fitness Training in People with Stroke

The evidence presented in chapter 4 suggests that physical fitness is low after stroke and that there is scope for improving physical fitness. There is no biological reason why physical fitness after stroke cannot be increased by physical fitness training, providing the training is safe and feasible. If fitness is improved there are a range of potential benefits including increased participation in activities of daily living and improved muscle strength and power, all of which are of value to people with stroke. Exercise also provides a range of physical, social and psychological benefits that are not always dependent on fitness improvement.

Physical Benefits

The strength and nature of the associations between low fitness and post-stroke problems (chapter 4, p. 85) suggests that improving fitness may improve some common post-stroke problems.

Post-stroke physical activity and fitness are low, and these low levels are associated with common post-stroke functional limitations. Increased fitness and physical function could, in theory, also help with other post-stroke problems, such as reducing fatigue and the risk of falls and fractures, compensating for the increased energetic cost of the hemiparetic gait, reducing disability, and improving independence, quality of life and mood.

There are other potential benefits that may not be dependent on improvements in physical fitness. Physical therapies are known to promote structural brain remodelling (Gauthier et al. 2008) and this may improve post-stroke motor deficits. There is systematic review evidence that repetitive practice of some common day-to-day activities produces some modest improvements in mobility and activities of daily living in stroke patients (French et al. 2010). Therefore, participation in repetitive, task-related training may have functional benefits even if fitness is not improved.

Psychosocial Benefits

Exercising in a group with other people has been found to have psychosocial benefits in people with stroke (Carin-Levy et al. 2009, Patterson and Ross-Edwards 2009, Mead 2005). Self-reported benefits include social and motivational support gained by sharing the experience with others. Therefore, simply participating in exercise may be beneficial particularly where group activities are involved.

Mortality and Morbidity

Exercise is known to be beneficial for people with a number of conditions which are risk factors for stroke or common co-morbid conditions associated with stroke. For example, systematic review evidence (p. 96) shows that exercise can lower blood pressure (Dickinson et al. 2006), improve vascular risk factors in people with obesity (Shaw et al. 2006) and type II diabetes (Thomas et al. 2006), reduce mortality in coronary heart disease patients (Jolliffe et al. 2000) and could benefit people with depression (Mead et al. 2008). Therefore, exercise and fitness training after stroke, particularly cardiorespiratory training, could reduce morbidity and mortality through secondary prevention of stroke and comorbid diseases.

These studies suggest that exercise training is not simply an intervention that is provided with the primary aim of improving fitness, but that it might have multiple other benefits after stroke.

Risks and Hazards

However, there may also be risks associated with exercise such as training-induced soft-tissue injuries, altered muscle tone, falls and vascular events. These risks need to be considered and managed in planning exercise programmes and are discussed in chapters 9 and 10.

Evidence of Benefits

There are different types of scientific study based on observation and experiment that provide empirical evidence. There is a hierarchy within these study designs that provides an increasing ‘level of evidence’; that is, more trustworthy data which allows stronger recommendations to be made.

Levels of Evidence

Observational Studies

Much of the data linking fitness and function in chapter 4 is from observational studies. These designs reveal associations between low fitness and post-stroke problems, but they do not prove that low fitness causes problems or that increasing fitness will alleviate problems. They do, however, allow one to hypothesise that increasing fitness may have benefits. However, to demonstrate that physical fitness training improves post-stroke problems, one needs to perform a randomised controlled trial.

Randomised Controlled Trials

These controlled study designs enable one to attribute any observed benefits of an intervention to the intervention itself, rather than other factors, such as type of patient recruited. Study participants, after assessment of eligibility and recruitment, but before the intervention to be studied begins, are randomly allocated to receive one or other of the alternative treatments. In the case of trials of fitness training, this might be either an exercise intervention (in addition to usual stroke care) or usual stroke care alone. After randomisation, the two (or more) groups of patients are followed up in exactly the same way. Thus any differences in outcomes at the end of the trial can be attributed to the intervention itself, in this case exercise.

Thus, randomised controlled trials (RCTs), particularly if they are large, provide a superior level of evidence about the effects to that of observational, uncontrolled studies. Generally, if clinical practice is to be changed, one needs robust evidence from RCTs. However, large trials of exercise – particularly after stroke – are difficult to carry out and, therefore, pooling the data from a number of studies is an effective way of taking stock of as much of the available evidence as possible; the mechanism to do this is provided by systematic reviews and meta-analyses.

Systematic Reviews

The best level of evidence about the effects of an intervention such as exercise is provided by systematic review of RCT evidence. Systematic reviews are based on a specific question about a health-care intervention. Their rigorous methodology identifies and reduces bias (systematic error). Pooling data from a number of RCTs by meta-analysis increases statistical power (reducing random error) and can reveal consistent effects. Systematic reviews and meta-analyses produced by the Cochrane Collaboration (http://www.cochrane.org/) are subject to a process of continual updating and considered the least biased type of systematic review. The level of evidence provided by these reviews currently means they often make a substantial contribution to guiding health-care decisions and policy (see p. 104).

Randomised Trial Evidence

In this section, we introduce the STroke: A Randomised Trial of Exercise or Relaxation (STARTER) trial (Mead et al. 2007). This is an example of how to design and perform a randomised trial of an exercise intervention. This trial controlled for the effects of social interaction by providing relaxation to the patients not allocated exercise, thereby allowing researchers to determine whether any beneficial effects of exercise were due to the exercise itself rather than the effects of social interaction. The intervention provided in STARTER was designed so that it could be delivered to groups of patients in the community setting.

STARTER was an exploratory trial which aimed to determine the feasibility of exercise training after stroke and to estimate how effective fitness training might be. It randomly allocated 66 participants, mean age 72 years, who had completed their stroke rehabilitation and who were able to walk without assistance from another person, to either fitness training or relaxation (control group). Both groups were held three times a week for 12 weeks. Both the exercise and relaxation (attention control) interventions were delivered in the same rehabilitation hospital by an advanced exercise professional three times a week (Monday, Wednesday, Friday) for groups of up to seven participants. A summary of the trial including the outcome measures is presented in Box 5.1. Details of the actual exercises are further described in chapter 10.

Box 5.1 Randomised controlled trial – summary

Participants	66 stroke patients
Intervention	Mixed training (group circuit training)
Comparison	Non-exercise attention control (relaxation)
Outcomes	Physical fitnessActivity limitationsQuality of lifeMood

Content of the Programme

During week 1, the instructor familiarised participants with techniques and equipment. At the start of each of the sessions the instructor measured blood pressure (as an additional safety check) and enquired whether the participants had fallen since the last session. Each session lasted 1 hour and 15 minutes (including ’tea-and-chat’ after the interventions). Transport (minibus or taxi) was provided. Participants unable to attend every session of the 12-week programme were offered up to three additional ‘catch-up’ sessions.

The mode of exercise, initial exercise intensity and rate of progression were based on an exercise intervention designed to reduce falls in older frail people (Skelton et al. 2005) many of whom had had a previous stroke and community exercise sessions designed for the UK charity ‘Different Strokes’ (http://www.differentstrokes.co.uk/). Modifications to the intervention were made to meet the therapeutic aims of normalising body symmetry, anatomical alignment and muscle tone as much as possible during the training process. Further adaptations, e.g. inclusion of the stair climbing/descending exercise, were made by the study physiotherapist. Participants unable to perform or complete a particular exercise were given a shortened, modified or alternative task, i.e. the exercise was ‘tailored’ to the needs of individual patients (Chapter 10). Although we aimed for progression every 2–4 weeks, individuals not ready to progress (e.g. insufficient strength, endurance or technique) remained at their current prescription and only progressed when assessed by the instructor to be ready. The length of our exercise programme (12 weeks) was similar to previous studies in stroke participants at the time (Saunders et al. 2004).

Each session started with a warm-up to enhance circulation and mobility (15 to 20 minutes). The total duration of the exercise training increased from 15 minutes (week 1) to 40 minutes by week 12. Each session comprised a cardiorespiratory training component followed by a resistance training component.