Congenital diaphragmatic hernia (CDH) is considered by most researchers to be a surgical emergency. However, early repair does not necessarily improve respiratory function or reverse fetal circulation, and many patients deteriorate postoperatively. As a result, in 1985, we began to employ a protocol in which surgery was delayed until the PCO2 was maintained below 40 and the child was hemodynamically stable; children in whom these criteria could not be achieved died without surgical repair. Sixty-one consecutive infants with CDH were managed over 4 years; 31 from 1983 to 1984 (group 1) and 30 from 1985 to 1986 (group 2). The groups were similar with respect to sex, side of the defect, birth weight, gestational age, incidence of pneumothorax, and blood gases. High frequency oscillation was used with increasing frequency during the study period, for patients with refractory hypercarbia (13% in group 1, 30% in group 2). All patients were initially paralyzed and ventilated. Mean time from admission to surgery was 4.1 hours in group 1 and 24.4 hours in group 2 (P less than .05). In group 1, 87% of patients had surgical repair (77% within eight hours of admission, 10% after eight hours), and in group 2 only 70% of patients had surgery (10% within eight hours, 60% after eight hours). All patients who were not operated on died. Overall mortality was 58% in group 1 and 50% in group 2; this difference was not statistically significant. These data indicate that our current approach has not increased overall mortality.(ABSTRACT TRUNCATED AT 250 WORDS)