Standardization of sonographic lung-to-head ratio measurements in isolated congenital diaphragmatic hernia: Impact on the reproducibility and efficacy to predict outcomes

Objectives - The purpose of this study was to evaluate the impact of standardization of the lung-to-head ratio measurements in isolated congenital diaphragmatic hernia on prediction of neonatal outcomes and reproducibility. Methods - We conducted a retrospective cohort study of 77 cases of isolated congenital diaphragmatic hernia managed in a single center between 2004 and 2012. We compared lung-to-head ratio measurements that were performed prospectively in our institution without standardization to standardized measurements performed according to a defined protocol. Results - The standardized lung-to-head ratio measurements were statistically more accurate than the nonstandardized measurements for predicting neonatal mortality (area under the receiver operating characteristic curve, 0.85 versus 0.732; P =.003). After standardization, there were no statistical differences in accuracy between measurements regardless of whether we considered observed-to-expected values (P >.05). Standardization of the lung-to-head ratio did not improve prediction of the need for extracorporeal membrane oxygenation (P >.05). Both intraoperator and interoperator reproducibility were good for the standardized lung-to-head ratio (intraclass correlation coefficient, 0.98 [95% confidence interval, 0.97-0.99]; bias, 0.02 [limits of agreement, -0.11 to +0.15], respectively). Conclusions - Standardization of lung-to-head ratio measurements improves prediction of neonatal outcomes. Further studies are needed to confirm these results and to assess the utility of standardization of other prognostic parameters.