Aim of the study: Severe lung hypoplasia and persistent pulmonary hypertension are the main determining factors of survival in infants with congenital diaphragmatic hernia (CDH). The oxygenation index (ratio of delivered oxygen and its arterial level) closely reflects lung function. Single-institution studies have reported that best oxygenation index on day 1 of life (BOI-d1) is the most reliable postnatal predictor of survival in CDH. The aim of this study was to evaluate the predictive value of BOI-d1 in four disparate high volume centers in Europe.
Methods: A retrospective, multicenter study of infants with CDH born between 2000 and 2009 in four European tertiary institutions was conducted. Ethical approval was obtained from institutional review boards. Centers no. 1 and. 4 offered extracorporeal membrane oxygenation (ECMO), whereas center no. 3 offered fetal endoluminal tracheal occlusion (FETO) in fetuses defined as poor prognosis (lung-to-head ratio [LHR]≤ 1.0 and "liver-up" position). Prenatal LHR and perinatal variables, including gestational age, birth weight, defect side, liver position, BOI-d1, and patch requirement, were analyzed. Receiver operating characteristic curves were used to determine cutoff values for continuous variables. Comparison was made between survivors and nonsurvivors using univariate analysis and logistic regression analysis, p<0.05 was considered significant.
Results: A total of 235 infants (center no. 1, n=29; no. 2, n=64; no. 3, n=113; and no. 4, n=29) were included. One infant required (2%) ECMO and 66 (28%) had FETO. LHR was available in 83 patients (36%). Overall survival (discharge from hospital) and 28-day survival were 67.6% (n=159) and 72.3% (n=170), respectively. Univariate analysis showed that significant categorical predictors of 28-day survival were liver-down position (p<0.0001), LHR >1 (p=0.003), and primary repair (p=0.02) but not defect side (p=0.83). Area under the receiver operating characteristic (AUROC) curve for continuous variables; gestational age, birth weight, and BOI-d1 were 0.70, 0.68, and 0.88, respectively. AUROC for BOI-d1 (28-day survival) was 0.91 and had sensitivities (73 and 91%) and specificities (92 and 80%) for cutoffs of 40 and 82, respectively.
Conclusion: This multicenter study showed, that except from the defect side, all the prenatal variables studied have predictive value but the most useful is BOI-d1. This is simple to calculate and represents an excellent marker for lung function and a reliable early postnatal predictor of survival.
Georg Thieme Verlag KG Stuttgart · New York.