Some infants of extremely low birth weight were noted to have respiratory deterioration (increased oxygen and ventilation requirements) around day 10 after initial improvement with surfactant (Survanta) replacement therapy. To characterize this deterioration ("slump"), we reviewed antenatal, neonatal, and short-term outcome variables in infants with a birth weight of 600 to 1,000 gm enrolled in our Survanta investigational new drug treatment protocol. Over a 20-month period 58 infants were enrolled. Of those, 51 (88%) survived. Survivors were grouped by number of doses given (in the first 48 hours): Group 1 (one, two, or three doses) (n = 29), Group 2 (four doses) (n = 22). Prevention and rescue treatments were included as per dosing protocol. On day 15, one (4%) in group 1 versus ten (45%) in group 2 required a fraction of inspired oxygen > or = 0.60, p < 0.001, and alveolar-arterial oxygen difference between groups < 200, p = 0.02. Differences between groups were not accounted for by antenatal steroid use, prevention versus rescue treatment, route of delivery, infant gender, patent ductus arteriosus, air leak, or documented infection. Bronchopulmonary dysplasia at 36 weeks' post-conceptional age was more common in group 2 (p < 0.05). We conclude that (1) the need for four doses of Survanta (administered as per protocol) was an early marker for this postsurfactant slump and (2) bronchopulmonary dysplasia at 36 weeks' postconceptional age was a better discriminator of chronic lung disease in this low-birth-weight group.