Remodeling of pulmonary arteries in human congenital diaphragmatic hernia with or without extracorporeal membrane oxygenation

PURPOSE: The aim of this study was to describe in detail the perinatal developmental profile of the pulmonary vasculature in congenital diaphragmatic hernia (CDH) and to examine the potential beneficial effects of extracorporeal membrane oxygenation (ECMO) on the vascular morphology. Additionally the authors aimed to identify the differences in pulmonary vascular morphology among CDH cases according to the primary cause of death: either extreme lung hypoplasia (LH) or persistent pulmonary hypertension (PPH). METHODS: The authors studied autopsy sections from 30 high-risk CDH cases with respect to the pulmonary arteries in relation to gestational age (GA) and ECMO treatment. They were grouped into CDH-I: 20 cases with GA greater than 34 weeks who were not subjected to ECMO and CDH-II: 10 cases with GA greater than 34 weeks, who were subjected to ECMO for an average time of 237 hours. Five age-matched neonates who died from placental insufficiency or birth asphyxia without evidence of lung hypoplasia served as controls (CON). Medial and adventitial thicknesses of pulmonary arteries were measured in lung sections stained with Elastic van Gieson by 2 investigators blinded for the clinical data. Immunohistological staining with anti-alpha-smooth muscle actin (alpha-SMA) was performed to confirm the precise location of the arterial media before morphometry. CDH cases were subgrouped and compared according to the primary cause of death. Unpaired Student t test was used for statistics, with significant P value < or =.05. RESULTS: In CDH newborns, a significant increase in medial, adventitial, and total wall thickness was found in pulmonary arteries with an external diameter of less than 200 microm as compared with age-matched controls (P<.004, .0001, and .0009, respectively). ECMO-treated CDH newborns showed a significantly thinner arterial adventitia than CDH patients who did not receive this treatment (P<.0001), approaching normal values. However, the medial thickness remained increased. Morphometrically, no significant differences in CDH cases between patients dying of PPH or severe LH could be determined. CONCLUSIONS: (1) In CDH, there is failure of the normal arterial remodeling processes occurring in the perinatal period. (2) Pulmonary vascular morphology in CDH does not differ between the groups with lung hypoplasia or persistent pulmonary hypertension as primary cause of death. (3) Adventitial thinning of these arteries might be one of the mechanisms by which ECMO alters PPH in CDH cases.