Despite one randomized trial suggesting a beneficial effect of steroids on the duration of neonatal ECMO (Westfall et al., Surgery, 1991), there is no consensus on the role of steroid therapy in neonatal ECMO. We surveyed the medical directors of 105 ECMO centers in May 1996 on their use of steroids in neonatal ECMO. We obtained the names of all medical directors of ECMO centers on the Extracorporeal Life Support Organization (ELSO) mailing list and sent them a brief multiple-choice questionnaire. This questionnaire requested information on the use of steroids in neonatal ECMO, the indications for such use, and the type and duration of steroid administered. Of the 69 respondents, 36 (52%) reported never using steroids for neonates at the initiation of ECMO, but 17 of those 36 had used steroids before the conclusion of ECMO. Thus 50 of 69 respondents (72%) used steroids in the course of neonatal ECMO; 14% reported using steroids ”frequently“ or”always“. The most frequently cited reason for using steroids was improvement in pulmonary function or reduced capillary leak. Of 56 respondents based in the USA, 37 indicated interest in participating in a multicenter randomized controlled trial of steroid use in neonates undergoing ECMO. We conclude that while there is no general consensus about the use of steroids in neonates undergoing ECMO, steroids are frequently administered to infants who receive ECMO. When steroids are employed, the time for initiation of therapy and the duration of therapy are variable. Before steroid therapy during ECMO becomes routine, a prospective randomized controlled trial on the efficacy of steroids in neonatal ECMO is urgently required.