The effects of exercise on secondary prevention and health-related quality of life in people with existing vascular disease: systematic review and meta-analysis of randomised controlled trials

Background: Polyvascular disease (atherosclerosis across two or more vascular beds) is becoming increasingly common, yet systematic reviews of interventions such as exercise are traditionally targeted at people with a single disease. We aimed to determine the effect of exercise in the secondary prevention of major adverse cardiovascular events and health-related quality of life (HRQoL) in people with an existing vascular disease and to assess the impact of polyvascular disease. Methods: For this systematic review and meta-analysis, we searched databases (Cochrane Register of Studies Online, MEDLINE, Embase Ovid, CINAHL EBSCO, WHO-ICTRP and ClinicalTrials.gov) in January 2025 for randomised controlled trials (RCTs) of exercise in people with coronary artery disease, heart failure, stroke (including transient ischaemic attack (TIA)) and peripheral arterial disease (PAD). We excluded studies where exercise was delivered for <6 weeks. Two reviewers independently assessed articles for eligibility and extracted data. Disagreements were resolved through discussion. Critical outcomes were mortality (all-cause and cardiovascular-specific), vascular events (myocardial infarction, stroke, amputation, acute limb ischaemia (ALI)), vascular hospitalisations, and HRQoL (EQ-5D and SF-36). We extracted data at end of intervention, medium term (6–30 months follow-up), and long term (>30 months follow-up). We performed random-effects meta-analyses. Risk of bias was assessed using Cochrane's Risk of Bias 1 tool. The certainty of the evidence was assessed using GRADE. PROSPERO registration: CRD42024517019. Findings: We included 280 RCTs involving 23,419 participants. 114 (40·71%) studies did not report whether their populations had more than one vascular disease. Exercise may result in little to no difference in all-cause mortality compared to no exercise at end of intervention (risk ratio (RR) 0·92, 95% confidence interval (CI) 0·80–1·07; P = 0·30; 143 studies, 12,811 participants; low-certainty evidence). Similar effects were found at medium and long term. Exercise may result in little to no difference in cardiovascular mortality compared to no exercise at end of intervention (RR 0·92, 95% CI 0·75–1·12; P = 0·41; 77 studies, 7319 participants; low-certainty evidence). A similar effect was found at medium term. At long term there may be a difference favouring exercise on cardiovascular mortality (RR 0·81, 95% CI 0·64–1·01; P = 0·06; 10 studies, 3935 participants). Exercise probably reduces vascular hospitalisations compared to no exercise at end of intervention (RR 0·73, 95% CI 0·56–0·95; P = 0·02; 64 studies, 7101 participants; moderate-certainty evidence) and medium term (RR 0·83, 95% CI 0·70–0·99; P = 0·04; 49 studies, 7514 participants; low-certainty evidence), with little or no difference at long term. Exercise probably increases HRQoL as assessed by EQ-5D compared to no exercise at end of intervention (mean difference (MD), 6·20, 95% CI 2·21–10·20; P = 0·002; 8 studies, 805 participants; moderate-certainty evidence), with little or no difference at medium term (MD 2·23, 95% CI –3·19 to 7·66; P = 0·42; 7 studies, 707 participants; moderate-certainty evidence) and long term (MD 6·00, 95% CI –2·05 to 14·05; P = 0·14; 1 study, 73 participants). Exercise probably increases HRQoL as assessed by SF-36 compared to no exercise at end of intervention (MD 6·83, 95% CI 5·22–8·44; P < 0·0001; 50 studies, 3231 participants; moderate-certainty evidence) and medium term (MD 6·44, 95% CI 3·71–9·18; P < 0·0001; 15 studies, 1522 participants; moderate-certainty evidence). No studies reported SF-36 at long term. Data on vascular events were mixed and of low certainty. Evidence was limited, and therefore uncertain, for amputation and ALI. Limiting issues were poor descriptions of exercise, and poor, inconsistently reported study inclusion and exclusion criteria, therefore limiting our ability to categorise included populations as polyvascular/single. Interpretation: We believe this systematic review and meta-analysis to be the first to combine RCTs with vascular diseases and examine the effects of exercise in people with single conditions and polyvascular disease. We found consistent evidence that exercise improves HRQoL and reduces hospitalisations across vascular disease but does not appear to impact mortality. However, the vast majority of trials were designed to target people with a single vascular condition and did not report the presence of additional vascular diseases. Therefore, it was not possible to formally assess the impact of the addition of polyvascular disease on exercise outcomes or determine the applicability of our findings to a population with polyvascular disease. More trials are needed that include participants with polyvascular conditions to strengthen the evidence on safety of this intervention, in order to inform clinical guidelines. Funding: This study was funded by the NIHR Evidence Synthesis Programme (NIHR162044).