Feasibility of aerobic exercise training in acute stroke (version 3)
Research type
Research Study
Full title
The feasibility of aerobic exercise training in the acute phase of stroke; a single-centre randomised, controlled, feasibility trial.
IRAS ID
280320
Contact name
Tom Maden-Wilkinson
Contact email
Sponsor organisation
Sheffield Hallam University
Clinicaltrials.gov Identifier
Duration of Study in the UK
1 years, 9 months, 0 days
Research summary
Research Summary:
In the acute phase of ischaemic stroke (1-7 days after stroke) the body increases the production of proteins called neurotrophins which help the nervous system recover from injury. Aerobic exercise (e.g. walking or cycling) stimulates nerves and blood vessels to produce and release a greater amount of neurotrophins than what would occur at rest. During aerobic exercise, active nerves in the brain release neurotrophins into the surrounding brain tissue. In addition, aerobic exercise increases blood flow to the muscles and brain. This causes blood vessels throughout the body to stretch, triggering the release of neurotrophins into the blood which are then transported to injured brain tissue. Animal studies show that starting aerobic exercise training in the acute phase of stroke improves brain recovery and reduces disability, which is mediated in part by neurotrophin concentration. However, no studies have attempted to see if this works in humans.
The purpose of this study is to see if it is feasible to do a larger trial which will assess if aerobic exercise training improves stroke survivors' recovery better than standard care. Many people have a lower-limb disability after stroke which makes it difficult to exercise. We will use a power-assisted bike which allows people to cycle (whilst in their own bed) even if they have reduced leg function. Participants will exercise once per day at a moderate intensity for 10-30 minutes over five days. The research will be completed on the Royal Hallamshire Hospital stroke wards. We will assess feasibility outcomes including exercise safety, intervention fidelity and attrition. We will also investigate how aerobic exercise affects the amount of neurotrophins in the blood and how brain blood flow changes during exercise. Lastly, we will interview participants and staff regarding the acceptability of the intervention and study procedures.
Summary of Results:
Introduction
Numerous animal studies have shown that aerobic exercise training (e.g., use a running wheel) implemented early after a stroke, reduces brain damage and neurological disability. However, there is uncertainty about the safety and effectiveness of aerobic exercise training early after stroke in human stroke survivors. Before attempting to investigate if aerobic exercise improves recovery from stroke, it is important to understand if it is safe and feasible to do in a clinical setting. It is also necessary to understand if it is feasible to conduct a research trial. The results of this study will help researchers decide whether it is worthwhile conducting a more expensive, large-scale, multi-site trial, and will help improve the quality of future research by understanding what works well and what could be improved.
The main aim of this feasibility study was to understand if stroke survivors could take part in a 5-session cycling exercise programme, starting between 24 hours to 7 days after stroke onset. We wanted to understand how participants tolerated cycling exercise, if it was safe, and if people liked or disliked it. In addition, our aim was to find out if it is feasible to conduct a research study, so we assessed how quickly participants could be recruited, if and why people dropped out, and if it was possible to do follow-up assessments after people were discharged from hospital.
Recruitment
Based on previous hospital-based rehabilitation research, we set a realistic target of recruiting 1-2 participants per month. We recruited 30 participants in 13 months, and our recruitment rate was 2.3 participants per month. Participants were randomly allocated to receive usual care physiotherapy (usual care group) or 5 additional cycling exercise sessions (intervention group). The groups were split randomly so that the average level of disability was similar between groups. The aim of this was to make sure that one group did not have an unfair advantage over the other. This is process is called ‘stratification’. For small studies, it is recommended not to use several stratification factors because this can lead to unbalanced group sizes. Therefore, we decided only to use ‘disability’ as a stratification factor. Unfortunately, this resulted in the usual care group being, on average, 12 years older than the intervention group. We will adjust for this age difference in future data analyses. In hindsight, it would have been more suitable to use a technique called ‘minimisation’ which makes it possible to randomly balance groups based on multiple characteristics.
Intervention delivery
The intervention consisted of 5 cycling exercise sessions over 5 days. The target exercise intensity was moderate (i.e., feeling slightly out of breath). The first exercise session involved 15 minutes of moderate-intensity exercise. The exercise duration increased by 5 minutes per session, reaching 35 minutes in the final session. A total of 75 exercise sessions were planned (5 sessions per participant). We conducted 55 exercise sessions in total, which was 73% of the planned amount.
The main reasons for not conducting exercise sessions were:
- 4 people were discharged from hospital earlier than anticipated (12 sessions missed).
- 1 person was too fatigued (1 session missed).
- 1 person developed an illness (3 sessions missed).
- 1 person dropped out because they disliked using the bike and heart rate monitor (4 sessions missed).Intervention safety
There were no adverse events related to the intervention. One participant developed a urinary tract infection and a raised temperature, which led to their exclusion from the intervention. The study doctor determined that this adverse event was not related to the intervention.
Intervention acceptability
13 out of 15 participants (87%) enjoyed the intervention. In general, cycling exercise gave people a sense of achievement and hope, improved their mood, and broke up the boredom of being on a hospital ward. One person was disappointed that they could not continue using the bike after completing the exercise programme. One participant dropped out of the study because they thought using a heart rate monitor (elasticated fabric band) around their chest was too invasive, and they did not like having their legs strapped into a bike. One participant thought that bike was not relevant to his stroke recovery.
Therapists supported the idea of using of exercise bikes as part of in-hospital stroke rehabilitation programmes, but most thought it would not be feasible for this type of intervention to be delivered by NHS staff due to a lack of time and staff availability. Therapists suggested that therapy assistants would be suitable and cost-effective staff members to supervise cycling exercise sessions if implemented in the NHS.
Other findings
We conducted some additional investigations to understand what effects cycling exercise had on participants. We assessed short-term changes (i.e., the response to a single exercise session) and longer-term changes (measured 3 months after stroke). Feasibility studies have small sample sizes, therefore they cannot prove if novel interventions are better or worse than usual care. However, preliminary data from feasibility studies can help identify a ‘signal’ which may reveal trends toward benefit or harm which can inform the decision to stop or continue research.
Short-term effects
Using ultrasound probes attached to a headset, we found that blood flow to the brain increased by 10-15% during exercise. This is a promising finding because previous research has shown that aerobic exercise has beneficial effects on post-stroke brain health, and increased blood flow to the brain is likely to play an important role in this process. We also investigated the effects of aerobic exercise on a blood protein called brain-derived neurotrophic factor (BDNF). This protein has an important role in the maintaining the health of the neurons and repairing damaged neurons. Previous research in animals has found that BDNF levels increase in the brain after exercise. We did not see any changes in the amount of BDNF in participants’ blood after exercise. However, it is important to note that BDNF found in the blood may not represent changes in the brain.
Longer-term effects
We conducted an ‘as-treated’ analysis which includes data from participants who were admitted to hospital for at least 6 days. This timeframe was chosen because it is enough time for people to complete the 5-session intervention (beginning the day after recruitment). This left 12 participants in the usual care group (2 removed due to short length of stay, 1 dropped out) and 11 participants in the intervention group (2 removed due to short length of stay, 1 dropped out, 1 developed an illness).
On average, the intervention group had a better recovery when assessed 3 months after their stroke. In the usual care group, one participant died whilst in hospital, and 7 out of 12 participants (58%) were independent and able to walk 3 months after their stroke. In the intervention group, 11 out of 11 participants (100%) were able to walk and were independent 3 months after stroke. This suggests that the intervention may have a beneficial effect on post-stroke recovery which justifies further research. However, it is important to note that the intervention group was 12 years younger (67.5 years vs. 79.5 years; mean average). As older people tend to have poorer post-stroke outcomes, our preliminary data should be interpreted with this limitation in mind, and future research should control for age as well as disability.
We measured numerous other outcome measures in this study, including: quadricep muscle size, physical and cognitive function, mood and quality of life. Data analyses for these outcome measures are ongoing.
REC name
Yorkshire & The Humber - Bradford Leeds Research Ethics Committee
REC reference
20/YH/0285
Date of REC Opinion
26 Nov 2020
REC opinion
Further Information Favourable Opinion