Congenital heart defects (CHDs), defined as gross structural malformation of the heart or large blood vessels during the fetal period, are the most frequent congenital anomalies in newborns, affecting 8.6–10.3 per 1,000 live births (1–3). Heart defects are a major cause of infant non-infectious morbidity and mortality, which can lead to about 6.0% neonatal and 46% congenital deaths (4, 5). Evidence showed that more than one million children with CHDs are born every year in the world, and more than half of them need surgical intervention within 1-year-old (6, 7). Although the surgical repair of CHD has achieved remarkable success, many interventions are palliative rather than curative (8). Survivors with CHDs are usually at significant risk of cardiovascular complications and neurodevelopmental disorders in the long term (8, 9). At present, the cause of CHDs is largely unknown.

Hypertensive disorders in pregnancy (HDP) are a group of maternal disorders characterized by increased blood pressure during pregnancy, which include gestational hypertension, pre-eclampsia or eclampsia, chronic hypertension, and preeclampsia superimposed on chronic hypertension (10). It was estimated that approximately 5–8% of all pregnant women worldwide were affected by HDP (11). Previous studies suggested that HDP produces an adverse in utero environment, which may increase the risk of adverse pregnancy outcomes (for example, low birth weight, neonatal death, intrauterine growth restriction, congenital malformations, and so on) (12, 13). These indicated that mothers with HDP may increase the risk for heart defects in offspring. Despite some published original studies that have evaluated the relationship between HDP and CHDs risk, the findings were still inconsistent (14–18).

Until now, one meta-analysis has been published to discuss this question (19). This study reported a significant association between maternal hypertension and the risk of CHDs in offspring (19). However, this review did insufficient consider the heterogeneity of results when exploring the association of maternal hypertension with CHDs (19), so the risk estimates are not entirely precise and robust. Meanwhile, the association between hypertension and specific CHD phenotypes was limited in terms of sample size, which may cause the bias of results due to the lack of statistical power. Additionally, of note, the previous review did not focus on the associated different types of HDP with the risk of CHDs. Furthermore, given the increasing prevalence of HDP, and partially after the publication of that meta-analysis, several large-scale cohort studies with inconsistent outcomes have been reported, collating the existing evidence of the association of HDP with CHDs is required (14–18).

Therefore, based on the above-mentioned situation, we synthesized the available published studies on the association between maternal HDP and the risk of CHDs in offspring in a systematic review and meta-analysis. The objective of this study is as follows: (i) to review and summarize the epidemiologic evidence on the association of maternal HDP on total CHDs and specific CHD phenotypes in offspring; (ii) to estimate the association of different types of HDP with total CHDs in offspring; and (iii) to identify potential sources of heterogeneity by subgroup, sensitivity analyses, and Galbraith plot.

Our meta-analysis aimed to synthesize the published literature on the association of maternal HDP with the risk of CHDs in offspring. Four principal findings were obtained. First, our study suggested that mothers who had HDP exposure were associated with a 70% increased risk of CHDs in offspring compared with non-exposure. Second, our study indicated that maternal HDP exposure could significantly increase the risk of most CHD phenotypes in offspring including CTD (OR = 1.55), TOF (OR = 1.35), ASD (OR = 1.78), HLHS (OR = 2.10), COA (OR = 1.73), LVOTO (OR = 1.33), and RVOTO (OR = 1.69). Third, the results of subgroup analyses indicated that the development risk of total CHDs in offspring had significantly increased when data were confined only to specific HDP types, such as pre-eclampsia (OR = 1.79), gestational hypertension (OR = 1.16), and chronic hypertension (OR = 1.68). Fourth, our results showed that the risk of total CHDs was also significantly increased when restricted to specific preeclampsia types, e.g., mild preeclampsia (OR = 1.33), severe preeclampsia (OR = 2.31), early onset preeclampsia (OR = 4.03), and late-onset preeclampsia (OR = 1.27). These findings indicated that early onset and severe preeclampsia had a higher risk of CHDs than late-onset and mild preeclampsia, respectively, and the increased exposure to the hypertensive disorder could generate more important complications. Besides, our results found that the association of maternal HDP with CHDs still exists after subgroup analyses, sensitivity analyses, and Galbraith plot, which indicated that our findings were credible.

To our knowledge, the potential mechanisms involved in the association of maternal HDP exposure with CHDs risk in offspring remain uncertain. However, several underlying mechanisms have been put forward to try to explain the association of maternal HDP with CHDs risk. The most potential explanation is that maternal HDP can occur in early damage of the vasculogenesis, leading to abnormal development of the placenta (18). Previous studies showed that placental dysplasia may cause alteration in placental genes, oxygen concentration, and protein expression (18). These alterations can lead to altered antioxidant enzyme activity, increased apoptosis, reduction of angiogenic factors, the abnormal balance of growth factors, hypoxic stress, and ultimately to lead to the abnormal development of the fetal cardiovascular system (15, 42, 47). For example, animal studies demonstrated that the blockage of vascular endothelial growth factor (VEGF) receptors resulted in a functional and structural defect in heart valve development, indicating that these receptors are involved in heart valve formation (48, 49). Moreover, animal studies also corroborated that hypoxic stress is associated with abnormal proteins activation and signaling, which may increase the risk of CHD (50). Meanwhile, persisting elevated blood pressure levels may affect cardiovascular health through a variety of mechanisms, including increased afterload, myocardial ischemia, myocardial fibrosis, and mechanical damage on the aortic valve. These alterations may induce abnormally high stress on aortic leaflets, turbulent flow, endothelial injury, and subsequent progression toward alteration in aortic valve morphology, finally causing aortic regurgitation, stenosis, and aortic dilatation, suggesting that persisting hypertension are closely related to cardiac structural abnormalities (51).

Additionally, there are a few possible explanations for the association of maternal HDP with CHDs risk in offspring. On the one hand, evidence indicates that maternal HDP exposure can lead to compromised blood flow to the developing fetus. Decreasing blood flow to the uterus during pregnancy could cause fetal intracardiac blood flow alterations and cell death, thus affecting normal heart development in the fetus (40, 52). On the other hand, although the heart develops were earlier than the period of pre-eclampsia from the theory of primitive embryo development [in embryology, the period of organogenesis starts 3 weeks and ends 8 weeks (42)], researchers have proposed that pathophysiologic changes in preeclampsia begin well before 20 weeks. Recently study has detected imbalances in angiogenic biomarkers as early as the first trimester in women who later developed preeclampsia (38). Among them, overexpression of the antiangiogenic biomarkers soluble endoglin (sEng) and soluble fms-like tyrosine kinase 1 (sFlt-1) associated with angiogenic placental growth factor and vascular endothelial growth factor begins at the start of pregnancy (38, 53). Meanwhile, a researcher found that a fetal imbalance toward anti-angiogenesis at the trophoblast stage could result in abnormal heart formation (38). Therefore, we can speculate that pre-eclampsia may already have clinical changes in early pregnancy, which may affect the development of the heart and lead to an increased risk of CHDs. Additionally, research showed that gestational hypertension may be in the spectrum of pre-eclampsia and was included in the definition of mild preeclampsia (54). Thus, the effect of pre-eclampsia on CHD may be consistent with that of gestational hypertension on CHD. Furthermore, findings regarding the association of maternal HDP with CHDs subtypes are varied in our study. Despite our study suggesting that maternal HDP may increase the risk for most CHDs, the mechanisms are almost unknown. We think that the differences between HDP and specific CHDs phenotypes could be explained by the fact that the pathophysiology of CHDs is multifactorial, involving genetic and environmental factors.

This systematic review had several strengths compared to the existing meta-analysis. First, as far as we know, our study was the first meta-analysis to independently estimate the association of pre-eclampsia, gestational hypertension, and chronic hypertension with the risk of CHDs. Our results indicated that the risk of total CHDs in offspring had significantly increased when data were restricted to pre-eclampsia (OR = 1.79), gestational hypertension (OR = 1.16), and chronic hypertension (OR = 1.68). Second, compared with the previous review (19), our study supplied up-to-date evidence on this topic and involved a large number of studies that not only included America and Europe populations, but also Asia, Oceania, and Africa populations. With the accumulation of evidence and the expansion of sample size and population, we have enhanced our statistical power to provide generalizable risk estimates. Third, although the previous meta-analysis also estimates the association of maternal hypertension with specific CHDs phenotypes (19), the number of studies included in the analysis is limited, including only 2–3 studies. Our study further enlarged the sample size in our analysis, which may contribute to improving the accuracy and reliability of risk estimates. Therefore, different from the results of the previous study (19), our results suggested a significant association of maternal HDP with most CHD phenotypes, which not only included CTD and ASD, but also TOF, HLHS, COA, LVOTO, and RVOTO. Besides, our study also fully considered the effect of heterogeneity and the most relevant heterogeneity moderators have been identified by subgroup analysis.

However, several limitations should be noted in our study. Firstly, in our review, results were limited to English-language literature, which may lead to possible language bias. Moreover, potential publication bias with under-publication of negative results may influence the findings, and it cannot be ignored. Secondly, substantial heterogeneity was found in the our study. This is not surprising, given the different study populations and methodologies. Based on the results of subgroup analyses, we found that types of HDP and geographic region were the potential source of heterogeneity. The possible explanation was that the risk of CHDs and HDP were not consistent in different geographical regions due to differences in race, medical level, and health awareness. For example, in some developing countries (e.g., Africa and Asia), there are still many pregnant women who do not receive adequate antenatal and pregnancy care because of the limited medical service, and the incidence and mortality of HDP were relatively high in these countries. Moreover, the health awareness of pregnant women in these developing countries is relatively weak, which may lead to the diagnosis delay of HDP, and result in a lack of effective early clinical interventions for HDP. Besides, since the definition of HDP may change over time, bias on outcome detection is difficult to avoid, which may increase clinical heterogeneity between studies. Nevertheless, we should still view the results with caution as the heterogeneity cannot be ignored. Thirdly, both case-control and cohort studies were selected in our study. Case-control studies are widely acknowledged to be prone to recall and selection biases, which restrict the strength and quality of evidence. Although the exposure history of maternal HDP was confirmed through medical records and/or physician-diagnosed self-reporting, the potential bias was unavoidable.

Fourth, considering that other subtypes of HDP were mostly diagnosed after organogenesis in addition to chronic hypertension, it is difficult to establish a causal link between pre-eclampsia or gestational hypertension and CHDs based on existing studies. Therefore, our study was mainly used to evaluate the association of maternal pre-eclampsia or gestational hypertension with CHDs in offspring, and could not explain the causal relationship. Fifth, of note, although our results found that early onset and severe preeclampsia had a higher risk of CHDs than late-onset and mild preeclampsia, respectively, the number of studies included in the analysis is limited, only three and four studies. In the future, more studies was needed to further explore the potential association between different types of preeclampsia and the risk of CHDs, which may contribute to improving the accuracy and reliability of risk estimates. Sixth, it is difficult to accurately evaluate the effect of the management of these disorders or the use of antihypertensive treatment on the occurrence of complications because of the limited information from original studies. Last but not least, the potential limitations including these studies (studies that were old or poor consideration of confounding factors) are inevitable. Although most of the included studies have tried to control for some potential confounding factors and restricting the analysis to studies that have controlled confounding factors did not substantially change the pooled risk estimate, the possible influences of confounding factors might not be completely omitted. Therefore, further studies with adequate adjustment of potential confounding factors are still warranted to clarify the association of HDP with CHDs in offspring.

In conclusion, our findings have two implications significant for the public health of pregnant women. First, our study indicated maternal HDP exposure including pre-eclampsia, gestational hypertension, and chronic hypertension may be associated with an increase in the risk of CHDs. Second, our findings showed that maternal HDP exposure was significantly associated with the risk of most CHD phenotypes in offspring, such as CTD, ASD, and so on. These findings emphasize that maternal HDP exposure could seriously endanger the health of their babies, and lead to an increased risk of death in offspring. Therefore, for pregnancies, especially pregnant women at high risk of HDP, clinical workers should form a coalition of obstetricians, pediatricians, sonographers, and specialists in pregnancy-related hypertensive disease to conduct dynamic monitoring of these pregnant women, and increase cardiac developmental early screening of HDP-exposed infants in order to early intervention, so as to avoid missing the best treatment time and reduce the risk of death in children with CHDs. However, considering that the mechanisms involved in the association between maternal HDP and CHDs in offspring are unclear, further studies are essential to explore the potential mechanisms and find probable interventions.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

SZ, LZC, and YJ made substantial contributions concerning: conception and design, acquisition of data, analysis and interpretation of data, drafting the article and revising it critically for important intellectual content, and final approval of the version to be published. XQ, TW, JL, JD, YL, LTC, and JQ made substantial contributions concerning: analysis and interpretation of data, drafting the article and revising it critically for important intellectual content, and final approval of the version to be published. LZC and YJ was the guarantor for the study. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.