To date, uniform standards for congenital diaphragmatic hernia (CDH) management have not existed. The purpose of this study was to evaluate the evolving clinical outcome of the patients with CDH and to present our recent management protocol using echocardiography. Sixty patients treated for CDH at our hospital from 1978 through 2001 were reviewed. Periods of treatments were divided arbitrarily into three periods;1978-1991 (period I, n=26), 1992-1994 (period II, n=6), 1995-2001 (period III, n=28). Immediate repair was performed during period I. We performed preoperative stabilization and delayed repair since the start of period II, and nitric oxide (NO) was introduced in period III. In period III, our management strategy was the use of fentanyl for sedation and analgesia; vasoactive agents such as dopamine, dobutamine, and prostaglandin E1 in selected cases; the use of high-frequency oscillating ventilation (HFOV), inhaled NO; and venovenous extracorporeal membrane oxygenation (ECMO) if indicated. The details of stabilization management and the timing of surgery were determined using echocardiography to evaluate pulmonary hypertension (PH) by measuring dimension and shunt patterns through the ductus arteriosus (DA), right pulmonary artery (rPA) and left pulmonary artery (lPA). Overall, 42 of 60 patients survived (70%). The number of patients surviving in each period was 14 of 26 (54%) in period I, 4 of 6 (67%) in period II, and 24 of 28 (86%) in period III. Seventeen of 28 patients in period III required inhaled NO (group A). Of these 17 patients, 5 required ECMO; of these 5, 3 were long-term survivors. The remaining 11 patients from period III who were managed without NO (group B) survived. In left-sided CDH cases, the dimension of DA at admission in group A (5.07+/-1.79 mm) was significantly larger than in group B (2.99+/-1.68 mm) (P<0.01). The dimension of rPA in group A (3.37+/-0.80 mm) was significantly smaller as compared with group B (4.28+/-0.72 mm) (P<0.01). Although the dimension of lPA was not significantly different between group A (3.03+/-0.74 mm) and group B (3.46+/-0.48 mm), lPA blood flow was noticeably stronger in group B. DA shunt patterns were bi-directional (53%), right-to-left (40%) and left-to-right (7%) in group A, whereas no patients in group B showed a right-to-left shunt pattern. After confirmation of closure of DA or dominant left-to-right shunt, and marked increase of pulmonary arterial blood flow, patients in both group A and B underwent surgery successfully. In four non-survivors, findings of improving PH were not observed. We conclude that echocardiographic examination is useful to manage persistent pulmonary hypertension with recent treatment modalities including NO and HFOV and to determine the proper timing of surgery, which contributes to an improved outcome of CDH.