Background: Valvular heart disease (VHD) is a type of cardiovascular disease with abnormal heart valve structure and/or function and a rapidly growing cause of global cardiovascular morbidity and mortality. Physical inactivity is a problem for patients with VHD, especially after surgery. However, there is no data on the effects of exercise on VHD from large multicentre randomised controlled trials (RCTs). Therefore, we conducted a systematic review and meta-analysis to provide a comprehensive analysis of small RCTs to evaluate the effects of exercise on cardiopulmonary function in patients with VHD and provide an evidence-based medicine basis for developing and guiding the clinical application of exercise in patients with VHD. Methods: We conducted a systematic review and meta-analysis of RCTs. We systematically searched electronic databases (PubMed, Web of Science, Embase, Cochrane Central Register of Controlled Trials, China National Knowledge Infrastructure [CNKI], China Science and Technology Journal Database [VIP], WanFang Database, and SinoMed [CBM]) for all studies on exercise and VHD from their inception to January 2023. The quality of included studies was assessed using the Cochrane risk-of-bias tool. The primary outcomes were the six-minute walk test distance (6MWD), left ventricular ejection fraction (LVEF), and short-form 36-item health survey (SF-36). Results: This systematic review included 22 RCTs with 1520 subjects (869 men and 651 women). The meta-analysis results showed that exercise significantly improved exercise capacity measured by the 6MWD (mean difference [MD] = 25.54, 95% confidence interval [CI] = 19.98–31.11, I{}^{\mathrm{2}} = 0%, p 0.00001), LVEF (MD = 6.20, 95% CI = 4.76–7.65, I{}^{\mathrm{2}} = 66%, p 0.00001), and quality of life measured by the SF-36 (physical function: MD = 3.42, 95% CI = 2.12–4.72, I{}^{\mathrm{2}} = 12%, p 0.00001; mental health: MD = 3.86, 95% CI = 0.52–7.20, I{}^{\mathrm{2}} = 68%, p = 0.020; social function: MD = 2.30, 95% CI = 0.64–3.97, I{}^{\mathrm{2}} = 45%, p = 0.007; bodily pain: MD = 2.60, 95% CI = 0.83–4.37, I{}^{\mathrm{2}} = 22%, p = 0.004) in patients with VHD compared to healthy controls. Conclusions: This study suggests that exercise can significantly improve cardiopulmonary function, enhance physical and social function, reduce bodily pain, and potentially improve mental health in patients with VHD, providing an evidence-based basis for better recovery in patients with VHD.