Placental dysfunction and impaired fetal growth: a relationship with bronchopulmonary dysplasia and pulmonary hypertension

More than five decades ago, Northway et al described the clinical, radiological and pathological features of a chronic lung disease in preterm infants characterised by lung emphysema, fibrosis, pulmonary vascular remodelling and cardiac hypertrophy. At that time, this chronic lung disease of prematurity, which they named bronchopulmonary dysplasia (BPD), was thought to occur as a consequence of postnatal events, including severe respiratory distress syndrome, oxygen toxicity and volutrauma. Despite improvements in neonatal care, postnatal adverse exposures continue to be important factors in BPD pathogenesis. Emerging evidence also suggests that pulmonary outcomes after preterm birth are likely determined not only by postnatal factors but also by the in utero environment. Male gender, white race, smoking, hypertension and infection have all been linked to increased BPD susceptibility. In addition, growing evidence suggests that fetal growth restriction (FGR) is a major risk factor for BPD. Abnormal placentation resulting in impaired placental perfusion is one of the most frequent causes of early FGR (<32 weeks gestational age) and preeclampsia. In the French EPIPAGE-2 study of preterm infants <32 weeks gestation, the risk of moderate/severe BPD was more than sixfold higher in FGR infants. Importantly, FGR has also been linked to a specific vascular phenotype of BPD in which pulmonary hypertension (PH) contributes to increased morbidity and mortality. In this issue of the Journal, Pierro et al present findings of their comprehensive metaanalysis of more than 200 studies demonstrating that preterm infants with FGR have roughly twofold higher risk of BPD and PH. These important findings highlight the critical influence of placental health in preterm pulmonary outcomes. The placenta is an essential conduit for oxygen and nutrient delivery from the mother to the developing fetus. Histological evidence of placental vascular malperfusion (fetal or maternal) is present in more than 50% of preterm placentas. When placental insufficiency is severe enough, the fetus becomes hypoxic and redistributes blood flow away from the periphery to the brain and heart. Not only does somatic growth become compromised but also human and animal studies show that fetal hypoxia adversely affects multiple organs and programmes chronic diseases into adulthood. Prospective cohort studies demonstrate that histological evidence of placental vascular maldevelopment increases the risk of BPD and PH. Sheep FGR models show profound impairment in alveolar and lung vascular growth, even in the absence of postnatal adverse exposures. Still unanswered though is how FGR alters pulmonary vascular development, leading to BPD and PH. Emerging evidence suggests that a complex interplay of angiogenic imbalance, inflammation, oxidative stress, epigenetic alterations and elevated endoplasmic reticulum stress may be involved. Interestingly, decreased cord blood angiogenic biomarkers are associated with BPD and PH. Preclinical studies reveal that pulmonary vascular endothelial cells isolated from FGR lambs have impaired angiogenic potential. Furthermore, endothelial cells obtained from human FGR placentas have defective function, suggesting that chronic intrauterine hypoxia dysregulates common pathways in placental and fetal lung endothelial cells. The systematic review by Pierro et al in the current issue of the Journal suggests that the degree of placental vascular dysfunction, as indicated by FGR, positively correlates with BPD and PH risk. The multifactorial prenatal phenotypes associated with BPD and PH are well illustrated in these comprehensive metaanalyses that dissect the contribution of various hypertensive conditions as well as FGR. These data could further strengthen BPD and PH prediction models. However, interpretation of the metaanalysis of observational studies is not straightforward, should be done with caution and has potential shortcomings. Metaanalyses benefit from the assumed balance between populations achieved in individual trials via randomisation. In observational studies, such as those included in these metaanalyses, the exposed and nonexposed populations can differ in terms of confounding factors, for example, gestational age. These individual studies use multivariate statistical methods to ascertain the adjusted risk, which can differ considerably from the nonadjusted risk. The issue related to risk factors that act as confounders or mediators in the pathway toward poor outcome needs to be considered. Conditions such as preeclampsia can lead to birth at earlier gestational ages. The impact of these conditions on the outcome of the general neonatal population will be in part mediated by the lower gestational age. However, prenatal exposure to these conditions can also increase the risk for poor outcome in infants of comparable gestational age. As the authors report, hypertensive conditions appeared not to be significantly associated with moderate/severe BPD when examining all studies but were shown to be significantly associated with BPD in the subgroup of studies where gestational age did not differ between exposed and nonexposed groups. Regardless of how a factor is deemed a confounder or mediator, one cannot discount their individual contribution to the outcomes. In the future, it will be essential to have more uniform FGR definitions as not all small for gestational age infants have FGR and vice versa. Moreover, intrauterine hypoxia programmes multiorgan dysfunction, some of which are potential contributors to the development of BPD and PH, and these confounders must be accounted for. Maternal illnesses and exposures leading to abnormal uteroplacental perfusion may also alter overlapping lung developmental signalling pathways, making risk assessment even more complex. The systematic review by Pierro et al highlights the necessity for antenatal approaches targeting placental health to reduce BPD and PH. It will be vital to better understand the mechanistic role of ischaemic placental disease leading to FGR and preeclampsia in the development of BPD and PH. Not only will such studies have far reaching implications for shortterm and longterm preterm pulmonary outcomes but they may also contribute University of Miami Miller School of Medicine, Miami, Florida, USA