BACKGROUND: Survivors of stroke have long-term physical and psychological consequences that impact their quality of life. Few interventions are available in the community to address these problems. Yoga, a type of mindfulness-based intervention, is shown to be effective in people with other chronic illnesses and may have the potential to address many of the problems reported by survivors of stroke. OBJECTIVES: To date only narrative reviews have been published. We sought to perform, the first systematic review with meta-analyses of randomized controlled trials (RCTs) that investigated yoga for its potential benefit for chronic survivors of stroke. METHODS: Ovid Medline, CINHAL plus, AMED, PubMed, PsychINFO, PeDro, Cochrane database, Sport Discuss, and Google Scholar were searched for papers published between January 1950 and August 2016. Reference lists of included papers, review articles and OpenGrey for Grey literature were also searched. We used a modified Cochrane tool to evaluate risk of bias. The methodological quality of RCTs was assessed using the GRADE approach, results were collated, and random effects meta-analyses performed where appropriate. RESULTS: The search yielded five eligible papers from four RCTs with small sample sizes (n = 17-47). Quality of RCTs was rated as low to moderate. Yoga is beneficial in reducing state anxiety symptoms and depression in the intervention group compared to the control group (mean differences for state anxiety 6.05, 95% CI:-0.02 to 12.12; p = 0.05 and standardized mean differences for depression: 0.50, 95% CI:-0.01 to 1.02; p = 0.05). Consistent but nonsignificant improvements were demonstrated for balance, trait anxiety, and overall quality of life. CONCLUSIONS: Yoga may be effective for ameliorating some of the long-term consequences of stroke. Large well-designed RCTs are needed to confirm these findings.