Neonatal duodenal obstruction: a review of 30 consecutive cases

Delayed return of normal duodenal function necessitating a prolonged hospital stay may often follow operative treatment of neonatal duodenal obstruction (NDO). Previously suggested ways to improve the operative result include variations in the anastomic technique, duodenal tapering, and the omission of gastrostomy. We have analysed the experience at the Prince of Wales Children's Hospital of 30 consecutive cases of NDO treated over a 7-year period (1984–1990) in order to define factors in the treatment that might influence the establishment of oral feeding and duration of hospital stay. There were 19 males and 11 females, with duodenal atresia occurring in 10 and extrinsic or intrinsic causes of duodenal stenosis in the remaining 20. Two patients died without operation. Corrective procedures included 17 duodenoduodenostomies, 9 duodenoplasties, and 2 patients had lysis of Ladd's bands that produced duodenal obstruction from the time of birth. Only 1 patient underwent duodenal tapering. There were 20 gastrostomies performed according to the preferences of the primary-care surgeons. It was shown that the duration of establishing oral feeding and hospital stay were not influenced by the type of corrective procedure, although Ladd's bands duodenal obstruction was associated with the shortest hospital stay. The use of a gastrostomy was related to higher morbidity, a longer period to establish oral feeding, and a prolonged hospital stay. Prenatal diagnosis has also emerged as an important influencing factor. It is concluded that gastrostomy should be omitted in the treatment of NDO. The various types of corrective procedures were similar in final outcome in terms of morbidity and total hospitalisation time.Presented at the 7th Asian Congress of Paediatrics, Perth, 1991.Correspondence to: E. Shi