The use of the alveolar-arterial oxygen difference P(A-a)O2 and the oxygenation index (mean airway pressure [Paw] FIO2 x 100/PaO2) have been proposed for selecting infants who will require extracorporeal membrane oxygenation (ECMO) therapy. However, the use of the oxygenation index (OI) in conjunction with Paw in an exclusive population of patients with meconium aspiration syndrome (MAS) has not been reported. Fourteen patients born in our facility and managed with conventional therapy and five infants treated with ECMO were enrolled in the study. All patients had clinical and x-ray evidence of MAS. Infants who received conventional treatment required mechanical ventilation greater than 48 h, FIO2 1.0, and were under the care and supervision of one neonatologist. Management was directed to minimize barotrauma by avoidance of routine hyperventilation, use of lower Paw, and sufficient expiratory time. One patient died before ECMO and 13 infants survived. Six survivors had an OI greater than 25 (three had an OI greater than 40), six had a Paw greater than or equal to 12 cm H2O (12 to 15 cm H2O in five infants) and six patients had a P(A-a)O2 greater than or equal to 610 torr. One surviving infant was transferred for ECMO therapy (OI 67, Paw 20 cm H2O). The five patients treated with ECMO survived (OI 48 to 92, Paw 20 to 29.5 cm H2O P(A-a)O2 627 to 650 torr). One patient in each group developed chronic lung disease with evidence of resting tachypnea. Our findings indicate that an OI greater than 40 in association with a Paw greater than or equal to 20 cm H2O may be helpful in predicting which infants with MAS need ECMO, whereas patients requiring a Paw less than or equal to 15 cm H2O can be managed with conventional therapy. An OI greater than or equal to 25 but less than 40 is not associated with high mortality in these patients. The predictive value of Paw of 16 to 20 cm H2O and the duration of an OI greater than 40 in patients with MAS need further investigation.