DOZ047.24: Predictors of early and late mortality in esophageal atresia patients

Abstract

Abstract Background Esophageal atresia (EA) is associated with mortality, despite significant advances in neonatal care. Predictors of early mortality (EM) are well described in the literature. However, little is known about late mortality (LM). This study aimed to describe EM and LM rates and their associations in EA patients. Methods Patients with EA, managed at The Royal Children's Hospital, Melbourne (1980–2017), who subsequently died, were identified from the Nate Myers Esophageal Atresia database. Data collected included patient demographics, VACTERL associations, mortality risk factors, and preoperative, operative, and postoperative findings. Early mortality was defined as death during initial admission. Results A total of 90/628 (14.3%) patients died during the study period—EM occurred in 68/90 (75.6%), whilst LM occurred in 22/90 (24.4%). Causes of EM were most often palliation for respiratory anomalies (17/68, 25.0%), followed by associated syndromes (11/68. 16.2%) and neurological anomalies (10/68, 14.7%). The most commonly associated syndrome leading to palliation was trisomy 18 (7/68, 10.3%). Causes of LM were most often respiratory compromise (6/22, 27.3%), sudden unexplained deaths (6/22, 27.3%), and Fanconi's anemia (2/22, 9.1%). Of patients discharged from hospital, 22/560 (3.9%) subsequently died. There was no difference in VACTERL association between EM (32/63, 50.8%) and LM (11/20, 55%) (P = 0.80), and no difference in the incidence of twins between EM (8/58, 13.8%) and LM (2/18, 11.1%) (P = 1). Conclusion We identified predictors of early and late mortality in esophageal atresia in a large cohort. Parents of children with esophageal atresia must be counseled appropriately as to the likelihood of death following discharge from hospital.