Death postponement and increased chronic lung disease: the hidden costs of mortality reduction in the post-surfactant era.

To determine changes in the morbidity and mortality of ventilated, very low-birth-weight infants before and after the introduction of surfactant therapy in 1990, we retrospectively studied the infants in a tertiary neonatal intensive care unit. All ventilated patients admitted from January 1, 1984, to December 31, 1997, with birth weight of 1500 g or less, excluding 76 transferred to other hospitals, were included in the analysis (N = 1336). The primary outcomes studied included mortality, length of stay, and survival without bronchopulmonary dysplasia. Although post-surfactant mortality and pulmonary air leak for infants with birth weight greater than 750 g decreased more than 50% (P < or = .026), the proportion of infants surviving without bronchopulmonary dysplasia also decreased (P < or = .034). For surviving infants with birth weight of 751 to 1000 g, mean post-surfactant length of stay increased 16 days (P = .008). Postnatal age at death also increased in the post-surfactant period for infants with birth weight of 750 g or less (P = .002). For infants with birth weight of 1000 g or less, post-surfactant increases were seen in the mean duration of assisted ventilation (+22% to 32%, P < or = .005) and the incidence of bronchopulmonary dysplasia at 36 weeks' postmenstrual age (+62% to 162%, P < .001). For all infants, the incidence of bacteremia, duration of supplemental oxygen therapy, and likelihood of discharge on home oxygen were increased in the post-surfactant period (P < or = .011). The implicit benefits of mortality reduction in the post-surfactant era may be offset by increasing respiratory morbidity in some survivors and by an unwelcome prolongation of death for some nonsurvivors. We speculate that the ultimate costs of these undesirable outcomes may greatly surpass the ostensible cost benefits of neonatal surfactant therapy.