Relative Mediastinal Displacement Index (RMDI): A Prenatal MRI Indicator of Adverse Events in Fetuses With Isolated Left Congenital Diaphragmatic Hernia

Background: Extracorporeal membrane oxygenation (ECMO), has partly improved congenital diaphragmatic hernia (CDH) outcomes, yet the overall morbidity and mortality remain high. Existing prenatal indicators for CDH fetuses are operator-dependent, time-consuming, or less accurate, a new simple and accurate indicator to indicate adverse events in CDH patients is needed. Purpose: To propose and assess the association of a new MRI parameter, the relative mediastinal displacement index (RMDI), with adverse events including in-hospital deaths or the need for ECMO in fetuses with isolated left CDH (iLCDH). Study Type: Retrospective analysis. Subjects: One hundred thirty-nine fetuses were included in the iLCDH group (24 with adverse events and 115 without) and 257 fetuses were included in the control group from two centers in Guangzhou. Field Strength/Sequence: 3.0 T, T2WI-TRUFI; 1.5 T, T2WI-FIESTA. Assessment: Three operators independently measured the -> (DL),-> (DR) , and DH on the axial images. The calculation formula of the RMDI was -> (DL) + -> (DR) , and DH. Statistical Tests: The independent sample t test, Mann-Whitney U test, Chi-square test, Chi-square test continuity correction, Fisher's test, linear regression analysis, logistic regression analysis, intraclass correlation coefficient, receiver operating characteristic curve analysis, and Delong test. A P value <0.05 was considered statistically significant. Results: The RMDI did not change with gestational age in the iLCDH group (with [P = 0.189] and without [P = 0.567] adverse events) and the control group (P = 0.876). There were significant differences in RMDI between the iLCDH group (0.89 [0.65, 1.00]) and the control group (-0.23 [-0.34, -0.16]). In the iLCDH group, RMDI was the only indicator left for indicating adverse events, and the best cutoff value was 1.105. Moreover, there was a significant difference in diagnostic accuracy between the RMDI (AUC = 0.900) and MSA (AUC = 0.820), LHR (AUC = 0.753), o/e LHR (AUC = 0.709), and o/e TFLV (AUC = 0.728), respectively. Data Conclusion: The RMDI is expected to be a simple and accurate tool for indicating adverse events in fetuses with iLCDH.