The purpose of this study was to examine the effects of therapeutic exercise on pain, stiffness, quality of life, physical function, disease activity, health-related fitness and cardiovascular risk factors in adults with spondyloarthritis (SpA). Electronic databases (Cochrane Central Register of Controlled Trials, EMBASE, MEDLINE/PubMed, PEDro, AMED, CINAHL) were systematically searched from inception to October 2013 using medical subject headings and keywords. This was augmented by searching conference abstracts and a hand search of reference lists of included studies. Randomised and quasi-randomised studies of adults with SpA in which at least one of the comparison groups received an exercise intervention were included. The primary outcomes were pain, stiffness, quality of life, physical function and disease activity. Secondary outcomes were health-related fitness and cardiovascular risk factors. Two reviewers independently screened studies for inclusion. Methodological quality was assessed by two reviewers using the Cochrane risk of bias tool and the PEDro scale. Twenty-four studies, involving 1,498 participants, met the inclusion criteria. Meta-analyses were not undertaken due to clinical heterogeneity, and this review focused on qualitative synthesis. Moderate evidence supports exercise interventions in improving physical function, disease activity and chest expansion in comparison to controls; there is low-level evidence of improved pain, stiffness, spinal mobility and cardiorespiratory function. Supervised group exercise yields better outcomes than unsupervised home exercise. The addition of aerobic components to flexibility programmes improves cardiorespiratory outcomes, but not cardiovascular risk factors. The most effective exercise protocol is still unclear. Current evidence indicates that therapeutic exercises provide benefits for adults with ankylosing spondylitis; effects on other SpA subtypes are unknown.