PURPOSE: Early intervention is needed to treat patent ductus arteriosus (PDA) as it is a major cause of increased mortality in preterm infants. However, it is uncertain which is better, medical versus surgical management. We reviewed medical records to compare the treatment course and outcome between medically and surgically treated preterm PDA infants. METHODS: Thirth-two Mechanically ventilated pretem infants (gestational age<34 wks, birth weight<2,000gm) who survived beyond 30 days were studied. Treatment course and outcome were compared between indomethacin-treated (INDO, n=15) and surgically treated who have not responded to indomethacin (Surg, n=17). RESULTS: Volume of administered fluid and urine output during the first five days of life were similar, however, initial weight loss were lower in the SURG group than INDO group (p=0.031). Size of PDA on the echocardiogram were larger in SURG group (mean 3.4 mm) than INDO group (mean 2.5 mm) (p=0.046). Duration of hospitalization was longer in the SURG group (mean 46 days) than INDO group (mean 72 days) (p=0.033), however, time to start feeding, ventilator duration and weaning time were similar in both groups. Incidence of intraventricular hemorrhage was lower in the SURG group (47%) than INDO group (6%) (p=0.009). CONCLUSION: Preterm infants with poor initial weight loss and large size of PDA were likely to become surgical candidates and required longer periods of hospitalization and showed increased incidence of IVH. Although surgical treatment of PDA in preterm infants is definitive, fluid restriction and medical management at early postnatal period is recommended.