In this population-based study of singleton VLBW infants born before 30 weeks of gestation, infants with PIH mothers had significantly higher odds of RDS and BPD than infants with non-PIH mothers. After adjusting for potential confounders, there were no significant differences in severe IVH, PVL, ROP, or death during NICU admission between infants with PIH mothers and infants with non-PIH mothers.
Although the pathophysiology of PIH remains uncertain, inadequate placental implantation and abnormal vascularization may play certain roles in the development of PIH24. Inadequate placental implantation leads to uteroplacental ischemia. As a consequence of uteroplacental ischemia, angiogenic or antiangiogenic factors, reactive oxygen species and inflammatory cytokines in the bloodstream of mothers with PIH can cross the placenta, reach the fetus and possibly affect the developing lungs25, 26.
The association between maternal PIH and RDS in preterm infants remains controversial10, 11, 27–29. A cohort study in the Netherlands reported the protective effect of preeclampsia on RDS28. Another single-center cohort study including preterm infants born at 23–28 gestational weeks demonstrated that preeclampsia increased the risk of RDS29. Similar to several studies examining the association between maternal PIH and RDS in preterm infants, we also showed a twofold increased risk of RDS in infants born to mothers with PIH, after adjustment for confounding factors10, 11, 29. Vascular endothelial growth factor (VEGF), which binds to the soluble VEGF receptor-1, is important for pulmonary vascular development and surfactant protein production30, 31. Soluble VEGF receptor-1 antagonizes VEGF by binding its molecules in the circulation, preventing them from interacting with their endogenous receptors and affecting their downstream signaling32. The elevated soluble VEGF receptor-1 concentrations in the amniotic fluid of mothers with PIH may inhibit VEGF signaling, leading to surfactant insufficiency10. RDS is secondary to surfactant insufficiency, and the association between PIH and RDS is biologically plausible as both are characterized by relative VEGF deficiency.
After adjustment for confounding factors, we found that the increased risk of BPD in infants with PIH mothers is in accordance with previous studies8, 9. Because angiogenesis and alveolar development are interactive in the fetal lung, BPD has been recently recognized as a manifestation of vascular disease of the lungs33. Impaired pulmonary vascular growth by altered signaling of angiogenic or antiangiogenic factors derived from mothers with PIH may play a role in the pathogenesis of BPD34. Tang et al. suggested that in infants born to mothers with PIH, impaired VEGF signaling in utero, due to high amniotic fluid levels of soluble VEGF receptor-1, disrupted lung growth and contributed to the increased risk of BPD35. Another study has revealed a relationship with endoglin, which is an antiangiogenic factor that acts by inhibiting TGF-β signaling36, 37. The study demonstrated that increased soluble endoglin in cord blood was associated with the development of BPD in preterm infants born to mothers with PIH. However, conflicting results on the effect of maternal PIH on BPD in preterm infants have been reported8, 9, 38, 39. An international cohort study specifically examining preterm infants born at 24 to 28 weeks of gestation reported that the risk for BPD increased in infants born to mothers with PIH8. A meta-analysis demonstrated that PIH was associated with BPD in preterm infants born at < 29 weeks of gestation9. By contrast, a meta-analysis of Australian cohorts showed that maternal preeclampsia did not influence the risk of BPD in extremely low-birth-weight infants38. Another large population-based study also reported that maternal preeclampsia was associated with a decreased risk of BPD in VLBW infants39. In that study, a negative association between maternal preeclampsia and BPD was shown only in a subgroup with a GA greater than 31 weeks. Considering that the study populations in these reports comprised data from birth weight-based registries, not GA-based registries, these discrepancies might be attributable to different study populations. A high number of SGA infants born at a relatively later GA would be preferentially included in birth weight-based registries. To minimize this bias in the KNN cohort, we only included VLBW infants born earlier than 30 weeks of GA in this study.
No difference was found in mortality before NICU discharge between infants born to PIH mothers and infants born to non-PIH mothers. The association of maternal PIH and mortality in preterm infants has not been consistent7, 14–17, 40. Some studies have demonstrated that PIH is associated with an increased or decreased risk of mortality7, 14–17. Another study has reported no association between PIH and neonatal mortality in preterm infants, similar to our study40. Conflicting results can be accounted for in part by differences in statistical approaches, differences in GA and birth weight ranges, and differences in sample sizes.
Several reports have noted a lower incidence of severe brain injuries, such as IVH or PVL, among preterm infants born to mothers with PIH8, 12, 13. They suggested that PIH may serve some adaptive role for the fetus in the face of uteroplacental dysfunction41. However, despite the suggested protective mechanisms, no significant differences were observed after adjusting for confounding factors in the rate of severe IVH and PVL between infants born to PIH mothers and infants born to non-PIH mothers in our study. This finding is in accordance with a previous meta-analysis showing that the rate of IVH and PVL in infants with PIH mothers was comparable to that in infants with non-PIH mothers9.
In this study, the overall rate of PIH was 17.1%. The maternal PIH rates in our study were comparable to those in previous studies14, 42. Considering that the worldwide incidence of PIH is increasing with the rising prevalence of predisposing factors, such as increasing maternal age, obesity, assisted reproductive technologies and diabetes, PIH deserves significant attention5, 7.
There are several strengths to our study. Our data were derived from a large national cohort prospectively collected in Korea. The KNN database is reliable and has been validated with a high degree of precision. Considering the bias that may exist in birth weight-based registries, as mentioned above, we excluded infants with a GA of 30 weeks or more from the study population. Additionally, we only included infants born as singleton in this study, which may have influenced the results by plurality14. Our study has some limitations. One limitation of the current study lies in the KNN definition of PIH. The KNN definition of PIH does not discriminate between gestational hypertension, preeclampsia, and eclampsia. It is important to know that a subgroup of PIH, such as gestational hypertension, preeclampsia, and eclampsia, may result in different adverse outcomes for infants. Moreover, the KNN database does not include information regarding aspects of the management of maternal PIH, including fetal monitoring and usage of medications such as magnesium sulfate, that may have affected outcomes for the infants43.