Summary of background: ECMO has been used in neonates and children for some time. However, its use on adults remains controversial, as studies reporting both positive and negative outcomes have been published. Aim: To determine if the application of ECMO in the support of critically ill adult patients with cardiac and/or respiratory failure is better, in detriment of the usual standard care. Methods: A systematic review of observational studies and clinical trials was performed by searching PubMed/Medline, SCOPUS and ISI Web of Knowledge for studies comparing the use of ECMO with conventional therapies. Database search retrieved 2164. Ten met the inclusion criteria and proceeded for data extraction, which was pooled for meta-analysis. Mortality rate was defined as the primary outcome. Sub-group analysis by respiratory or cardiac indication was performed. Results: Ten studies were included, six related to respiratory insufficiency and four to cardiac failure. No statistical differences (p<0,05) between ECMO technic and conventional therapies were observed. ECMO showed a non-significant tendency to have a lower mortality rate in cases of respiratory illnesses, and a higher in cardiac indications. The main complications reported for ECMO were haemorrhagic, followed by neurological and mechanical. Conclusion: The use of ECMO in critically ill adult patients does not offer an improvement over conventional techniques. Therefore, its use should be carefully evaluated, before instituting this technique in detriment of some of the conventional therapies.