Mortality risk indicators may be useful adjuncts to clinical judgment in considering the use of extraordinary and relatively high-risk treatments such as high-frequency ventilation, extracorporeal membrane oxygenation, and nitric oxide therapy. We retrospectively evaluated the reliability of published indications of high mortality, developed additional high-risk indicators, and determined the interval between the time when high-risk indicators were met and when death occurred in near-term and term neonates with respiratory failure. Patients were included in the analysis if they met the following criteria: > or = 35 weeks gestation, > or = 2 kg birth weight, < or = 7 days of age, and receiving a fraction of inspired oxygen of > or = 0.8 and mechanical ventilation. Patients were excluded if they had congenital heart disease, intracranial hemorrhage, untreatable bleeding diathesis, or lethal congenital anomaly. Fifteen patients in the 1980 to 1981 group and 41 patients in the 1985 to 1987 group met these criteria. We observed 100% mortality in our 1980 to 1981 patients who met previously published criteria predictive for > or = 80% mortality in near-term and term neonates with respiratory failure; however, mortality risk was 60% to 80% in the 1985 to 1987 group.(ABSTRACT TRUNCATED AT 250 WORDS)