Purpose: This study was undertaken in order to analyze clinical data conceming cases of esophageal atresia cases we encountered to elucidate the risk factors that may more accurately predict a prognosis. Methods: We retrospec5vely reviewed the medical charts of 19 infants with esophageal atresia diagnosed at Sanggye Paik Hospital from Juoe 1991 to M8y 2000. Resulte: The mean birth ht of the infants was 2.46 kg (0.97-3.99 kg). Associated anomalies occurred in t2 infants (63.2%), including cardiovascular anomalies in 10 (52.6%), anorectal anomaly in 1 (5.3%), renal anomaly in 1 (5.3%), skeletal anomaly in 1 (5.3%) and chromosomal anomaly in 1 (5.3%). We performed primary end-to-end anastomosis with one layer of interrupted suture in 12 infants. Primary repair was carried out in 7 infants, simultaneously with gastrostomy in 1, and gastrostomy & delayed end-to-end anastomosis was performed in 4. Postoperative complications included pneumonia in 8(66.7%), leakage in 4 (33.3%), stricture in 4 (33.3%), sepsis in 2 (16,7%), wound infection in 1 (8.3%) and gastroesophageal reflux in 1 (8.3%). The postoperative mortality r8te was 25.0%(3/12). Causes of death included sepsis (n=2) & heart failure (n=1). According to the Waterston criteria, 4 infants (21.1%) were classified as group A, 6 (31.6%) as group 8, and 9 (47.3%) as group C. The postoperative survival rates of group A, B, and C were 100% (3/3), 80% (4/5), 8Ad 50% (2/4) respectively. The postoperative survival rates of class l and class II by Montreal classification were 88.9%(8/9) and 33.3% (1/3), respectively. Conclusion: Our 4ata suggests that associated anomalies and general conditions are more important prognostic factors than birth weights in patients with esophageal atresia. (J Korean Surg Soc 2001;61:277-281)