Characterization of Suboptimal Responses to Fetoscopic Endoluminal Tracheal Occlusion in Congenital Diaphragmatic Hernia.

INTRODUCTION

The aim of the study was to characterize the changes in fetal lung volume following fetoscopic endoluminal tracheal occlusion (FETO) that are associated with infant survival and need for extracorporeal membrane oxygenation (ECMO) in congenital diaphragmatic hernia (CDH).

METHODS

Fetuses with CDH who underwent FETO at a single institution were included. CDH cases were reclassified by MRI metrics [observed-to-expected total lung volume (O/E TLV) and percent liver herniation]. The percent changes of MRI metrics after FETO were calculated. ROC-derived cutoffs of these changes were derived to predict infant survival to discharge. Regression analyses were done to determine the association between these cutoffs with infant survival and ECMO need, adjusted for site of CDH, gestational age at delivery, fetal sex, and CDH severity.

RESULTS

Thirty CDH cases were included. ROC analysis demonstrated that post-FETO increases in O/E TLV had an area under the curve of 0.74 (p = 0.035) for the prediction of survival to hospital discharge; a cutoff of less than 10% was selected. Fetuses with a <10% post-FETO increase in O/E TLV had lower survival to hospital discharge [44.8% vs. 91.7%; p = 0.018] and higher ECMO use [61.1% vs. 16.7%; p = 0.026] compared to those with an O/E TLV increase ≥10%. Similar results were observed when the analyses were restricted to left-sided CDH cases. A post-FETO <10% increase in O/E TLV was independently associated with lower survival at hospital discharge (aOR: 0.073, 95% CI: 0.008-0.689; p = 0.022) and at 12 months of age (aOR: 0.091, 95% CI: 0.01-0.825; p = 0.036) as well as with higher ECMO use (aOR: 7.88, 95% CI: 1.31-47.04; p = 0.024).

CONCLUSION

Fetuses with less than 10% increase in O/E TLV following the FETO procedure are at increased risk for requiring ECMO and for death in the postnatal period when adjusted for gestational age at delivery, CDH severity, and other confounders.