Oesophageal atresia is correctable and survivable in infants less than 1 kg

Management of oesophageal atresia (OA) and trachea-oesophageal fistula (TOF) in babies of low birth weight is challenging especially when associated with other anomalies. Birth weight of <1500 g has previously formed part of a classification system designed to predict outcome, alongside the cardiac status of the patient. Improvements in neonatal care have led to increasing numbers of premature low birth weight infants surviving. The aim of this study was to look at the experience of our institution in the extremely low birth weight (ELBW) patients.

A retrospective review of our institutions OA database was performed from 1993 to June 2015. Patients of birth weight less than 1000 g were included. A review of our OA/TOF clinical database and notes review established the following; gestation, birth weight, associated anomalies, operative procedures, morbidity and mortality.

Of 349 patients with OA across the 22-year period, 9 ELBW patients were identified (<1000 g). Six males and three females. Gestational age ranged from 23 to 34 weeks and median birth weight was 815 g ranging from 630 to 950 g. Overall survival was 56 % (5/9). There were double the numbers of ELBW OA/TOF patients seen in the second half of the study period presumably the result of improving neonatal care. Seven patients had type C OA with TOF and underwent emergency TOF ligation, two had concomitant oesophageal repair. One of these patients died from NEC; the other survived. Of the five who had isolated TOF ligation three died—two from cardiac disease and one from prematurity. Both type A patients survived and after initial gastrostomy placement one had a primary delayed repair, the other a gastric transposition. All three babies under 800 g died—one from cardiac disease the others from conditions indicative of their prematurity—necrotising enterocolitis and intraventricular haemorrhage.

50 % survival is achievable in OA/TOF under 1 kg and the Spitz classification is still applicable in this group as a whole. However, none of the current classification systems are applicable in infants <800 g who in our study all had poor outcomes. We suggest these should be considered as separate group when predicting outcomes.