RESUMEN
INTRODUCTION: Exercise have antidepressant effects in people with Major Depressive Disorder (MDD). However, about to half of patients do not respond to exercise. The identification of factors that moderates the antidepressant effects of exercise in people with MDD may help researchers and health professionals to identify sub-groups of patients that would benefit more from exercise. METHODS: A systematic review was carried out using Medline(PubMed), EMBASE and psycINFO up to April 2015. Individual and composite moderators were summarized and the strength of the evidence was assessed. RESULTS: Eleven studies were included for review resulting in the identification of potential individual (two biological, three clinical, two psychological and two social individual) and two potential composite moderators (the interaction between BDNF and Body Mass Index (BMI) and between family history of mental illness and gender). Only the two biological features and the BDNF x BMI interaction provided confirmatory evidence. LIMITATIONS: Due the different statistical approaches used in the studies, it was not possible to perform meta-analyses. The small number of studies and the exploratory nature of the evidence limits a wider generalization of the findings. CONCLUSION: Potential clinical, psychological, social or biological moderators were identified. However, the small number of studies and the limited strength of the evidence requires further studies before drawn definitive results. Further trials should consider the inclusion of moderators analysis using an a-priori, theoretical/evidence based hypothesis in order to provide high quality evidence for the use of personalized medicine in exercise for depression.