Ipsilateral Lung Volumes Assessed by Three-Dimensional Ultrasonography in Fetuses with Isolated Congenital Diaphragmatic Hernia

Objective: To evaluate the precision of three-dimensional ultrasonography (3DUS) in estimating the ipsilateral lung volume and the potential of this measurement to predict neonatal death in congenital diaphragmatic hernia (CDH). Methods: Between January 2002 and December 2004, the ipsilateral lung volumes were assessed by 3DUS using the technique of rotation of the multiplan imaging in 39 fetuses with CDH. The observed/expected ipsilateral lung volume ratios (o/e-IpsiFLVR) were compared to the lung/head ratios (LHR) and to the observed/expected total fetal lung volume ratios (o/e-TotFLVR) as well as to postnatal death. Results: Ipsilateral lung volumes (median 0.12, range 0.01–0.66) were more reduced than the total lung volumes (median 0.52, range 0.11–0.95, p < 0.001) in CDH. The bias and precision of 3DUS in estimating ipsilateral lung volumes were –0.61 and 0.99 cm 3 , respectively, with absolute limits of agreement from –2.56 to +1.33 cm 3 . The o/e-IpsiFLVR was lower in neonatal death cases (median 0.09, range 0.01–0.46) than in survivals (median 0.18, range 0.01–0.66), but this difference was not statistically significance (p > 0.05). The sensitivity, specificity, (positive and negative) predictive values and accuracy of o/e-IpsiFLVR in predicting neonatal death was 52.6% (10/19), 83.3% (10/12), 83.3% (10/12), 52.6% (10/19) and 64.5% (20/31), respectively. Conclusion: Although the ipsilateral lung volume can be measured by 3DUS, it cannot be used to predict neonatal death when considering it alone. However, it is important to measure it to calculate the total fetal lung volumes as the o/e-TotFLVR has the best efficacy in predicting neonatal death in isolated CDH.