Abstract Purpose The purpose of this study was to determine whether prenatal imaging parameters are predictive of postnatal CDH-associated pulmonary morbidity. Methods The records of all neonates with CDH treated from 2004 to 2012 were reviewed. Patients requiring supplemental oxygen at 30 days of life (DOL) were classified as having chronic lung disease (CLD). Fetal MRI-measured observed/expected total fetal lung volume (O/E-TFLV) and percent liver herniation (%LH) were recorded. Receiver operating characteristic (ROC) curves and multivariate regression were applied to assess the prognostic value of O/E-TFLV and %LH for development of CLD. Results Of 172 neonates with CDH, 108 had fetal MRIs, and survival was 76%. 82% (89/108) were alive at DOL 30, 46 (52%) of whom had CLD. Neonates with CLD had lower mean O/E-TFLV (30vs.42%; p = 0.001) and higher %LH (21.3 ± 2.8 vs.7.1 ± 1.8%; p < 0.001) compared to neonates without CLD. Using ROC analysis, the best cutoffs in predicting CLD were an O/E-TFLV < 35% (AUC = 0.74; p < 0.001) and %LH > 20% (AUC =0.78; p < 0.001). On logistic regression, O/E-TFLV < 35% and a %LH > 20% were highly associated with indicators of long-term pulmonary sequelae. On multivariate analysis, %LH was the strongest predictor of CLD in patients with CDH (OR: 10.96, 95%CI: 2.5–48.9, p = 0.002). Conclusion Prenatal measurement of O/E-TFLV and %LH is predictive of CDH pulmonary morbidity and can aid in establishing parental expectations of postnatal outcomes.