People who have had a stroke or transient ischaemic attack (TIA) have an increased risk of a second cardiovascular event (eg, stroke, heart attack) and commonly exhibit cardiovascular risk factors (eg, hypertension, diabetes, dyslipidemia). To-date, research on prevention of cardiovascular events and risk factors has focused on drug interventions. The evidence-base of non-drug treatments, particularly exercise, is less clear. The aim of this systematic review was to estimate the average effect of exercise interventions on reducing secondary cardiovascular events or risk factors in people with stroke or TIA.

The inclusion criteria were: recruiting adults who had sustained a stroke or TIA; evaluating exercise interventions; compared to usual care or control conditions; reporting the outcomes of cardiovascular events (including cardiovascular death) and/or cardiovascular risk factors; and, a randomised controlled trial design. Exercise interventions included aerobic, resistance, or a mixture of both aerobic and resistance. Motor training interventions aimed at improving function were only included if they had a substantive aerobic or resistance component. The cardiovascular risk factors evaluated were blood pressure (systolic and diastolic), lipid profile (total, low-density, and high-density cholesterol), fasting plasma glucose, and body mass index. Sensitive searches were performed in 11 databases (including PubMed and China National Knowledge Infrastructure). Two reviewers independently selected the included trials and extracted data, with any disagreements resolved by consensus discussions or arbitration by a third reviewer. The Consensus on Exercise Reporting Template (CERT) was used to evaluate completeness of reporting of the interventions. The Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach (that included trial methodological quality evaluated using the Cochrane risk of bias tool) was used to categorise the overall quality of the evidence for each outcome. Meta-analysis was used to calculate the mean effect and 95% confidence interval (CI) for each outcome. Pre-specified subgroup analyses were used to estimate the effects of early (< 6 months) versus late (> 6 months) exercise initiation; incorporation of health education in the intervention; and, participant type (TIA or non-disabling stroke versus severe stroke).

Twenty trials (n = 1,031 participants) were included in the review. Most trials only recruited people with stroke (16 trials; three exclusively recruited people with disabling stroke), initiated the intervention within 6 months of the primary stroke or TIA (11 trials), and compared exercise to usual care (12 trials). Ten trials used aerobic exercise, two used resistance exercise, and eight used a combination of aerobic and resistance exercise. Four trials incorporated health education with the exercise intervention.

Only one small trial (70 participants) reported the cardiovascular events outcome. In this trial exercise reduced the risk of hospitalisation or death due to stroke, myocardial infarction or peripheral arterial disease (adjusted hazard ratio 0.194, 95% CI 0.121 to 0.737) compared to the control group. There was moderate quality evidence that exercise reduced systolic blood pressure (-4 mmHg, 95% CI -7 to -2, 12 trials, 606 participants) but had no effect on fasting glucose (-0.14 mmol/L, 95% CI -0.29 to 0.01, 7 trials, 364 participants) and body mass index (0.00 kg/m2, 95% CI -0.26 to 0.25, 8 trials, 446 participants) compared to control interventions. There was low quality evidence that exercise reduced diastolic blood pressure (-3 mmHg, 95% CI -5 to -1, 12 trials, 606 participants) and total cholesterol (-0.27 mmol/L, 95% CI -0.54 to 0.00, 9 trials, 370 participants) compared to control interventions. There was very low quality evidence that exercise had no effect on low-density (-0.28 mmol/L, 95% CI -0.63 to 0.07, 7 trials, 303 participants) and high-density (0.08 mmol/L, 95% CI -0.02 to 0.17, 9 trials, 394 participants) cholesterol compared to control interventions.

Data from the subgroup analyses were reported for the outcomes of systolic and diastolic blood pressure only. The reduction in systolic blood pressure was most pronounced in trials that initiated the exercise intervention within 6 months of stroke or TIA (< 6 months -8 mmHg, 95% CI -12 to -5, 6 trials, 298 participants; > 6 months -2 mmHg, 95% CI -4 to -1, 6 trials, 308 participants), incorporated an educational component (exercise + education -8 mmHg, 95% CI -14 to -1, 4 trials, 200 participants; exercise only -3 mmHg, 95% CI -4 to -1, 8 trials, 406 participants), and recruited participants with TIA or non-disabling stroke (TIA or non-disabling stroke -5 mmHg, 95% CI -9 to -2, 9 trials 428 participants; disabling stroke -3 mmHg, 95% CI -4 to -1, 3 trials, 178 participants). The reduction in diastolic blood pressure was also more pronounced in trials that initiated the exercise intervention within 6 months of stroke or TIA (< 6 months -3 mmHg, 95% CI -6 to -1, 6 trials, 297 participants; > 6 months -2 mmHg, 95% CI -5 to 1, 6 trials, 309 participants) and recruited participants with TIA or non-disabling stroke (TIA or non-disabling stroke -4 mmHg, 95% CI -5 to -2, 9 trials 428 participants; disabling stroke 1 mmHg, 95% CI 0 to 2, 3 trials, 178 participants). In contrast, the effects on diastolic blood pressure were less pronounced in trials that incorporated an education component (exercise + education -2 mmHg, 95% CI -5 to 1, 4 trials, 200 participants; exercise only -3 mmHg, 95% CI -6 to 0, 8 trials, 406 participants).

Aerobic and/or exercise can result in reductions in some cardiovascular risk factors (systolic and diastolic blood pressure, total cholesterol) in people with stroke or TIA, particularly if initiated early, provided alongside education, and targeting people with TIA and non-disabling stroke. Investigating the effects of exercise on preventing cardiovascular events is a future research priority.

Wang C, et al. Aerobic exercise interventions reduce blood pressure in patients after stroke or transient ischaemic attack: a systematic review and meta-analysis. Br J Sports Med 2019;53(24):1515-25

Read more on PEDro.