In coronary patients, emotional disturbances in the sense of increased anxiety and depression have often been documented. Over the last years, there has been a growing interest in the effects of exercise on emotion. This meta-analysis, based on 13 to 15 studies on psychological effects of exercise programmes in coronary patients, showed a positive effect size both for anxiety (dmean = .3137) and depression (dmean = .4569). Nevertheless, these effect sizes have to be evaluated as less than medium by convention and in comparison to the effect size of psychotherapeutic interventions. The effect sizes on anxiety or depression were not related to methodological characteristics of the included studies, like duration of the exercise programme, time chosen for follow-up, publication year or use of control groups. But there was a significantly negative correlation between sample size and effect size. This might indicate a potential publication bias in the sense that significant rather than non-significant results may be published on small subject samples. We concluded that exercise programmes should not be considered the only treatment for emotional disturbances in coronary patients, but that they can be psychologically beneficial as an additional treatment component.