C. C. Style1,5, T. C. Lee1,5, A. R. Mehollin-Ray4, P. E. Lau1, M. A. Verla1,5, S. M. Cruz1, J. Espinoza2, C. J. Fernandes3, S. G. Keswani1,5, D. L. Cass5, O. O. Olutoye1,2,5 1Baylor College Of Medicine,Micheal E. DeBakey Department Of Surgery,Houston, TEXAS, USA 2Texas Children’s Hospital,Department Of Obstetrics And Gynecology,Houston, TEXAS, USA 3Texas Children’s Hospital,Department Of Pediatrics,Houston, TEXAS, USA 4Texas Children’s Hospital,Department Of Radiology,Houston, TEXAS, USA 5Texas Children’s Fetal Center,Division Of Pediatric Surgery,Houston, TEXAS, USA 6Baylor College Of Medicine,Micheal E. DeBakey Department Of Surgery,Houston, TEXAS, USA
Introduction: Congenital diaphragmatic hernia (CDH) is a congenital anomaly that causes significant morbidity and mortality. Fetal MRI has been used to characterize and stratify the severity of CDH after diagnosis. The purpose of this study was to determine the utility of serial MRI in predicting severity of outcomes in CDH.
Methods: A single institution retrospective review was performed of fetuses referred to our institution from April 2004 to March 2017 with a diagnosis of CDH. Prenatal MRI indices including total fetal lung volume (TFLV), observed to expected total lung volume (O/E TLV), and presence of intrathoracic liver herniation were evaluated to predict outcomes. Primary outcomes included survival, need for extracorporeal membrane oxygenation (ECMO) and pulmonary hypertension (PH). Patients were included if they underwent fetal MRI at least twice prior to delivery. Fetuses that underwent endotracheal balloon occlusion were excluded. Data were analyzed using student’s t-test and logistic regression, a p value <0.05 was considered significant.
Results:Of 250 fetuses evaluated for CDH, 193 were further characterized by MRI; a total of 73 had serial MRIs, with 53 having at least one occurring between 20 – 28 weeks (2nd trimester) and one occurring at >28 weeks gestation (3rd trimester). Mean gestational age at the 2nd trimester and 3rd trimester MRI were 23.5 ± 2.1 and 33.1 ± 1.70 weeks, respectively. Of the 53 patients, 67% (n=36) were male and 85% (n=45) had a left-sided CDH. There was an average of 10.1 weeks (± 4.4) between the 1st and 2nd MRI. Of these patients, 98% (n=52) had an increase of TFLV with a mean growth of 1.45 ± 0.85 mm/week between the 2nd and 3rd trimester. There was a significant correlation of interval TFLV growth and PH postnatally (p=0.009). O/E TLV either increased or decreased by >10% in 74% with no statistical correlation to postnatal outcome. When comparing all three MRI parameters in the overall cohort (n=193), O/E TLV, TFLV, and liver herniation were predictive of infant survival (p<0.05). TFLV in the 3rd trimester was an independent predictor of PH, however O/E TLV at initial presentation was a strong predictor for need for ECMO and survival when comparing serial MRIs (Table 1). Overall cohort survival was 79%.
Conclusion: TFLV in the 3rd trimester is a strong predictor of pulmonary hypertension, but MRI assessment of lung volumes in the second trimester is more predictive of survival than 3rd trimester studies. Our results suggest that the timing of MRI assessment of lung volumes in CDH patients who have not undergone fetal intervention is important in predicting outcomes, and that serial MRI assessments may be of value in selected cases.
