ABSTRACT. The turnover and pool size of surfactant has Abbreviations been studied in animals, but there is little similar information in humans. In the present investigation lung effluent phospholipids were studied in 29 small preterm infants with severe RDS. Thirteen were treated with mechanical ventilation, and 16 additionally received natural human surfactant. The first dose (60 mg surfactantlkg body wt) was given between 2 and 10 h of age, and the surfactant was given again if there was an insufficient response. Together 260 aspirates, recovered during routine suctioning of the airways, were analyzed for phospholipids. Phosphatidylglycerol, present only in exogenous surfactant, was used as a specific marker to estimate the apparent pool size and the half-life of surfactant phospholipid. In addi-.tion, the saturated phosphatidylcholine/sphingomyelin ratios were correlated with the ventilatorv index (mean airway pressure X fractional inspiratory oxygen/arterial oxygen tension). There was a linear correlation between the ventilatory index and the saturated phosphatidylcholine/ sphingomyelin ( r --0.70) but no consistent correlation between the ventilatory index and the amount of phospholipids in the aspirate. The saturated phosphatidylcholine/ sphingomyelin ratio increased during the surfactant-induced remission of respiratory failure, decreased during the relapse of respiratory failure (present among 50% of the surfactant-treated infants), and increased again during the recovery. The control infants tended to have lower saturated phosphatidylcholine/sphingomyelin ratios during the first week than the surfactant-treated infants. Therecipients of surfactant had slightly more severe lung disease than the controls, when the results were adjusted by covariance to remove the differences in the saturated phosphatidylcholine/sphingomyelin ratio. Exogenous surfactant increased the amarent endogenous ~o o l size at least fivefold. The apparent half-life of ~hosph~tidylglycerol was 30 h (20-36 h). The half-life was independent of the amount of exogenous surfactant (60 versus 120 n~g/kg). Therefore, the apparent turnover rate after 120 was higher than after 60 mg/kg surfactant (p < 0.01).