Extracorporeal membrane oxygenation (ECMO) improves survival in mature neonates with reversible lung disease. However, ECMO could result in survival of infants with severe respiratory dysfunction who would otherwise have died. Alternatively, infants receiving ECMO might be spared prolonged ventilation and consequent barotrauma, resulting in improved respiratory function. Our aim was to compare respiratory function at 1 yr of age in infants assigned to receive either ECMO or conventional management (CM). Seventy-eight surviving infants of the United Kingdom (UK) ECMO trial (51 in the ECMO group) were studied at 1 yr of age. Questionnaires provided details of respiratory symptoms, and laboratory measurements of respiratory function were made for respiratory rate, tidal volume, lung volume, airway conductance, specific airway conductance, and maximal expiratory flow at FRC (Vmax (FRC)). Data were exchanged on floppy disk for cross-analysis and to ensure that investigators were blinded to the status of the infants. There was a wide spectrum of respiratory function, from normal to markedly abnormal. There were few differences between the groups, but in the CM group lung volume was increased (95% confidence intervals [CIs] of the difference in ECMO versus CM subjects: -67; -4 ml), and inspiratory specific conductance was lower (95% CI: 0.03; 0.98 s(-)(1). kPa(-)(1)). There was a trend toward a lower V max(FRC) (95% CI: -2; 67 ml/s(-)(1) in the CM group. In addition to providing a survival advantage, ECMO did not worsen lung function in infants assigned to receive it. Indeed, their lung function appeared slightly better than that of infants treated conventionally.