Outcome analysis for gastroschisis

Background/Purpose : Several factors are reportedly associated with an adverse outcome in gastroschisis, including mode of delivery, in utero diagnosis, type of closure, concurrent anomalies, intestinal atresia, and necrotizing enterocolitis (NEC). Since 1969, the authors have treated 185 patients who had gastroschisis. The authors analyzed their database to identify variables associated with increased morbidity and mortality. Methods : A retrospective study of all patients with gastroschisis treated at our institution in the last 30 years was performed. The characteristics of survivors and nonsurvivors were compared. A logistic regression analysis was performed, with survival as the dependent variable, and the following parameters as independent variables: in utero diagnosis, mode of delivery, gestational age and birth weight, era of repair, type of closure, presence of other associated anomalies, intestinal atresia, and development of necrotizing enterocolitis. Further logistic regression analysis was performed, with various indicators of morbidity as dependent variables. These included development of sepsis, bowel obstruction, and complications related to the closure or to the silo. No attempt at long-term follow-up was made. Results : A total of 185 infants with gastroschisis were treated at our institution from 1969 to 1999. Mean gestational age was 36.6 weeks, and the mean birth weight was 2,501 g. A total of 21 infants had intestinal atresia. NEC developed in 8 infants. Six infants had other serious anomalies. The overall survival rate was 91%. Survival improved in last 2 decades (94%). There were no differences in gestational age, birth weight and mode of delivery, method of closure, or presence of intestinal atresia between survivors and nonsurvivors. Only the era of repair ( P = .002), presence of necrotizing enterocolitis ( P = .044), and presence of other major anomalies ( P < .001) correlated with mortality in the logistic regression analysis. Sepsis, bowel obstruction, and closure complications accounted for most of the morbidity. Analysis of these three morbidity factors identified low gestational age ( P = .038) and development of necrotizing enterocolitis ( P = .020) as independent predictors of sepsis. Closure complications were only associated with lower birth weight ( P = .006). No predictors of bowel obstruction were identified. Conclusions : Mode of delivery, method of closure, birth weight and gestational age, and the presence of int