Role of placenta in development of pre‐eclampsia: revisited

Abstract

Over the past two decades, the impact of pregnancy on the maternal cardiovascular system has been investigated widely, and these adaptations have been shown to persist over time1-4. In disorders of pregnancy, particularly pre-eclampsia, research has shown that maternal cardiovascular adaptation to pregnancy is disrupted, and that these dysfunctional changes may persist for months or years postpartum5-8. In pre-eclampsia, the physiological adaptation of the maternal cardiovascular system to pregnancy fails1, 9. Large epidemiological studies have consistently shown that women with a history of pre-eclampsia are at increased risk of cardiovascular morbidity and mortality in later life10, 11. Women with pre-eclampsia develop hypertension earlier and suffer higher mortality rates from congestive heart disease12, 13. However, analysis of a large population cohort showed that the risk of cardiovascular mortality is increased when pregnancy is complicated by preterm pre-eclampsia, while women with a diagnosis of pre-eclampsia with delivery at term exhibited no increased risk10. Cardiovascular mortality was also higher in women with preterm delivery for reasons other than pre-eclampsia10. Studies involving maternal echocardiography during and following a pre-eclamptic pregnancy have demonstrated severe structural and functional changes of the maternal heart14. These changes persisted postpartum, although only in women with early- rather than late-onset disease8, 15. These persistent structural changes of the maternal heart were perceived as potentially linking pre-eclampsia and subsequent maternal cardiovascular mortality16. In the past few years, the idea that the maternal cardiovascular system is the source and instigator of the development of pre-eclampsia, rather than the placenta, has gained considerable traction. The subject has been canvassed in Ultrasound in Obstetrics and Gynecology17 and elsewhere1, 18, 19, particularly by Thilaganathan, Lees and others. They have taken the notion a step further, saying that the cardiovascular system is the sole initiator of pre-eclampsia. However, the theory that the placenta is the source of pre-eclampsia is backed by decades of research. While the evidence shows that perturbation of maternal cardiovascular adaptation can lead to pre-eclampsia, the placenta remains the primary and essential initiator of pre-eclampsia in many instances. We will show that the far-reaching conclusion of these investigators that the maternal cardiovascular system is the ‘engine’ and the placenta is only the ‘radiator’ in the maternal–fetal unit, is unfounded. Echocardiographic studies of normal and pre-eclamptic pregnant women have shown characteristic changes in the cardiac index, total vascular resistance index (RI) and myocardial relaxation, altered heart geometry, such as ventricular wall thickness, and other changes1, 5-7. Indeed, the degree of these idiosyncratic changes in pre-eclamptic women appears to differentiate between early-onset and late-onset pre-eclampsia, as well as between those with and those without fetal growth restriction (FGR)8, 20. On the basis of these findings, proponents of the maternal cardiovascular theory of pre-eclampsia assert that the maternal heart is the most plausible source of pre-eclampsia. About 80% of pre-eclampsia occurs at term and most fetuses are not affected by FGR, which would be expected if the placenta was the source of the disease. Indeed, most fetuses delivered from pre-eclamptic pregnancies are appropriate- or even large-for-gestational age, and this has been used as evidence for the maternal cardiovascular theory17. Some researchers cite as support for the maternal cardiovascular theory that altered uterine artery blood flow is the cause of alterations in the placenta, not the result17. They point to evidence that high uterine artery Doppler RI, measured as early as the first trimester, indicates a high risk for the eventual development of pre-eclampsia. Leslie et al.21 showed that pregnancies with high first-trimester uterine artery RI also exhibited alterations in placental gene and protein expression in placental tissue collected during pregnancy terminations. Ophthalmic artery vascular indices have been studied in gravidae at various gestational ages, and the association between variations in these indices and the risk of pre-eclampsia22 has been promoted18, 19, 23 as supporting the idea that pre-eclampsia cannot be a consequence of impaired trophoblast invasion but rather it is related to maternal maladaptation to pregnancy23. Abdominal pregnancies may develop pre-eclampsia in the absence of invasion of the myometrium by the spiral arteries. Shallow spiral-artery invasion and failure of spiral-artery conversion are implicated in the pathophysiology of pre-eclampsia. In abdominal pregnancies, the placenta anchors to the peritoneum, maternal abdominal organs or the exterior aspects of the uterus. Despite this lack of direct connection between the trophoblasts and spiral-artery conversion, a characteristic decrease in uterine artery pulsatility index has been observed in abdominal pregnancies24. This finding is held up as an example of the ancillary nature of placental involvement. The placenta, it has been noted, cannot be considered ‘in isolation’. It depends on the maternal circulation for perfusion, and so placental dysfunction depends on maternal cardiovascular maladaptation17. This maladaptation clinically precedes pre-eclampsia in appearance, the pre-eclampsia cascade predominantly consists of cardiovascular signs, and cardiovascular disease risk persists many years postpartum. Taken together, these points are cited as evidence for the maternal cardiovascular origins of pre-eclampsia17, 18. In addition, pharmacological interventions to reduce the risk of pre-eclampsia in women who are screen-positive with a high risk of the disease include aspirin therapy, a drug with known vascular effects. This seems to support the notion that pre-eclampsia is preceded by cardiovascular maladaptation. Recent research25-27 has described the added value of repeat testing of soluble fms-like tyrosine kinase-1 (sFlt-1)/placental growth factor (PlGF) ratio, alone or with N-terminal pro B-type natriuretic peptide or C-terminal proendothelin-1 biomarkers, in the follow-up of pre-eclampsia. Both sFlt-1 and PlGF are produced by the placenta. Supporters of the maternal cardiovascular basis of pre-eclampsia reiterate that these biomarkers have known roles in the cardiovascular system, asserting, firstly, that their rise is a result of cardiac maladaptation and ischemia, and secondly, that the pre-eclampsia cascade worsens these cardiovascular effects. Parallels have been drawn between pre-eclampsia and gestational diabetes, in that both are conditions of pregnancy that are ‘cured’ by delivery. However, no one has shown diabetes to be caused by disordered placentation, and nor does gestational diabetes appear preponderantly in primigravidae. In the proposed maternal cardiovascular model of pre-eclampsia17, the placenta is presented as the ‘radiator’ while the maternal heart is the ‘engine’ on which the fetal–maternal unit depends. None of the facets comprising the model has their source in the placenta. This is an essential point. The placenta is far more than a mediator or bystander in pregnancy or in disorders of pregnancy. Decades of evidence corroborates the crucial role that the placenta plays. The maternal cardiovascular system plays a significant role that cannot be ignored in understanding the development of these disorders. However, by the same token, one cannot ignore the placenta, which in some instances is the main or sole instigator of the disease, and in others is a major player. Several pathologies of pregnancy exemplify this notion. The maternal–fetal–placental unit normally exists in harmony, and, while pre-eclampsia can develop in the absence of a fetus or in abnormal implantation of the pregnancy, without direct involvement of the uterus, it never develops without a placenta. Abdominal pregnancy is a rare complication but may progress to term with delivery of a live fetus28, 29, and may be further complicated by pre-eclampsia. This may support the proposition that it is the placenta which is necessary for pre-eclampsia, as the following cases of abdominal pregnancy have shown28, 29. In the first case, after delivery of a live fetus with the placenta left in-situ, pre-eclampsia persisted for 99 days until removal of the placenta, verified by the presence of clinical signs and kidney endotheliosis on biopsy. Kidney biopsy at 21 months postpartum revealed resolution of endotheliosis. The patient was alive and well 25 years postpartum with normal kidney function and blood pressure28. In another case, severe pre-eclampsia developed at term. A section of the placenta was left in situ after Cesarean delivery of a live term neonate. Eclampsia persisted for over 2 weeks in a ‘stormy postoperative course’29. We30 and others31 have described cases of twins discordant for FGR, in which selective feticide of the affected twin rapidly resolved pre-eclampsia. Despite the uterine and cardiovascular milieux being identical for both twins, elimination of the ailing placenta led to resolution of the clinical syndrome30, 31. A case of dichorionic twins discordant for severe FGR and severe brain anomaly was treated in our institution. Following selective feticide, the maternal symptoms of pre-eclampsia disappeared within 48 h. A randomized controlled trial was published recently in The Lancet32, that described the differential rates of pre-eclampsia in pregnancies conceived by assisted reproductive technology using frozen- vs fresh-embryo transfer (16/512 (3.1%) vs 4/401 (1.0%); relative risk, 3.13; 95% CI, 1.06–9.30; P = 0.029). The distinguishing factor between the groups was the preparation of the embryos32. In none of the above pathological cases can the associated pre-eclampsia be explained by maternal cardiovascular maladaptation. These cases demonstrate that pre-eclampsia may occur without direct involvement from the uterus, but not without a placenta. The physiology of placental implantation leads to anchoring via ectopic trophoblast invasion of other tissues, impacting vascular reactivity at other sites, even in the absence of spiral arteries. This alternative invasion provides pathways for nutrient and gas exchange to the developing fetus, as well as secretion of placental factors into the maternal circulation. Transformation of the spiral arteries and decreased resistance in the uterine artery are normally mediated by trophoblast invasion; however, this is not absolutely essential. Other circulating angiogenic and other factors can initiate a decrease in uterine artery resistance through physiological redundancy. Following the ‘radiator-vs-engine’ logic, pregnant women with congenital cardiovascular disease must be at the highest possible risk for pre-eclampsia. This, however, is not the case. In a large study, the Canadian CARPREG consortium investigated 599 pregnancies of 562 women with heart disease33. Severe maternal outcomes, such as pulmonary edema, arrhythmia, stroke or cardiac death, occurred in 13% of pregnancies. However, there was no increase in the incidence of pre-eclampsia in this group. Pregnancy-induced hypertension was diagnosed in 4% and pre-eclampsia in 2% of cases. The rate of premature delivery was 10%; however, 62 of the 105 preterm deliveries were due to premature labor. Proponents of the cardiovascular theory argue that the earlier median gestational age at delivery in these women might explain the identical or lack of increase in the incidence of pre-eclampsia. This is not backed by the data of this and other studies. Moreover, the most severe cardiovascular alterations are seen in early-onset pre-eclampsia. Such cases were not more frequent in this study. Therefore, when severe impairment of the maternal heart, i.e. ‘a broken engine’, does not lead to more pre-eclampsia, how is the maternal cardiovascular theory substantiated at all? Angiogenic factors, such as sFlt-1 and PlGF, have garnered attention over the past 15 years as a means to better diagnose and predict pre-eclampsia. Many studies have consistently shown their predictive value in various study designs, including large observational and interventional studies34, 35. However, the key aspect of these placental factors is the fact that they are central to the pathophysiology of the disease. In 2003, Maynard and colleagues at the Karumanchi lab showed that women with pre-eclampsia have altered placental expression and circulating serum levels of angiogenic and antiangiogenic factors such as sFlt-1 and PlGF36. In a small clinical study, they showed that levels of sFlt-1 were elevated and those of PlGF decreased in pre-eclamptic women. The degree of alteration had a dose–response-like relationship with the severity of the disease: the more dysregulated the placental expression and circulating levels in the peripheral blood, the more severe the disease. Additionally, and more importantly, the Karumanchi group presented the results of an animal study showing that pregnant rats overexpressing sFlt-1 developed hypertension and proteinuria, as well as glomerular endotheliosis, a histological hallmark of pre-eclampsia. Thus, inducing higher sFlt-1 levels led to the development of pre-eclampsia in these rodents, emphasizing its pathophysiological role in the disease process defined by hypertension and proteinuria36. At present, there is no treatment for pre-eclampsia. However, most current therapeutic approaches are focused on the placental sFlt-1/PlGF pathway. Extracorporeal apheresis, performed to remove sFlt-1 from the circulation of women with early-onset pre-eclampsia, has been shown to prolong pregnancy37, 38. Pravastatin, which is currently being tested as a prophylactic and therapeutic intervention for pre-eclampsia39, 40, inhibited sFlt-1-secretion in an ex-vivo model41. Recently, small interfering RNA (siRNA)-inhibition of placental sFlt-1 expression ameliorated symptoms of pre-eclampsia in a primate model42. These are preclinical results, however, they show the relevance of the placental sFlt-1 pathway as the most promising in the development of a therapy for the multisystem disorder, pre-eclampsia. Research therefore points to the placental angiogenic factor pathway, rather than the maternal cardiovascular system, as a promising target for therapy for the condition of interest. In summary, angiogenic factors are not specific to pre-eclampsia, as defined by hypertension and proteinuria, but rather to placental dysfunction. This should prompt a re-evaluation of the endpoint of interest. First pregnancy is an outlier in the risk factors for pre-eclampsia. Our collaborative investigations into the role of natural killer cells as builders49 of the maternal–fetal interface showed that the cytokine profiles of the decidua differ between nulligravidae and non-pregnant parous women50. In addition, differences between primigravid and parous women persist throughout pregnancy. These findings are present well before any cardiovascular adaptation could occur50, and it is therefore more plausible that placental and decidual pathologies contribute to the development of pre-eclampsia. Pre-eclampsia has not only been perceived as a disease of nulliparous women, but has also been shown to be a disease of primipaternity51. In women with hypertensive pregnancy disorders, compared to normotensive controls, the median duration of sexual cohabitation before conception was significantly shorter (2.4 months vs 17 months). Furthermore, the percentage of new paternity was significantly higher and the duration of unprotected intercourse eventually leading to a pregnancy was significantly shorter in these couples51. This clearly highlights the importance of the immune rather than the cardiovascular system for the etiology of the disease. The findings of investigation into the effect of smoking on the risk of pre-eclampsia52 provide further support for the role of PlGF in the development of pre-eclampsia from an unexpected quarter. The study revealed that pregnant smokers, even those with uterine artery Doppler indices placing them at high risk of pre-eclampsia, had higher levels of PlGF and a lower risk of pre-eclampsia. In the subgroup of women in the cohort who developed FGR, smokers had higher PlGF/sFlt-1 ratio than did non-smokers (P = 0.0311). The authors attributed the ‘protective’ effect of smoking to the observed increase in these placental factors, even in women with abnormal uterine artery Doppler52. The study of Leslie and colleagues21 of first-trimester uterine artery RI measured in pregnancies prior to termination, demonstrated that the placental samples of pregnancies with high uterine artery RI showed differential expression of 29 genes (six were upregulated and 23 downregulated). Interestingly, several of the downregulated genes have roles in immune and inflammatory responses. Evidence of oxidative stress was observed to be similar in all samples. Some apoptosis-pathway markers differed between normal and high-RI pregnancies while others did not21. Leslie et al.21 present results that support the notion of a placental basis of pre-eclampsia, since differential expression of placental factors (PlGF, sFlt-1 and others) broadly impacts the development of the fetal–placental–maternal interface, as well as the maternal cardiovascular system. The observed changes in ophthalmic artery Doppler measures22 are likewise a manifestation of the downstream effects of circulating placental factors. Some of the known risk factors for the development both of cardiovascular disease in women and of pre-eclampsia in pregnancy, overlap. These include age, ethnicity, obesity, diabetes, chronic hypertension and chronic kidney disease. Smoking, a known risk factor for heart disease, has been shown to somewhat reduce the risk of pre-eclampsia52. Some researchers have proposed that pre-eclampsia and cardiovascular disease share risk factors, although the development of cardiovascular disease later in life has not been shown to be mediated by pre-eclampsia, even in at-risk women11. Analysis of our delivery database (data not shown) revealed a cohort of parturient multiparae with no history of pre-eclampsia in their first pregnancy, who developed pre-eclampsia in their second or later pregnancy, followed by delivery in one of our wards of another healthy pregnancy in which pre-eclampsia did not develop (n = 527; 0.7% of multiparae). This subgroup of multiparous women who developed pre-eclampsia substantiates the concept that an inherent defect in the placenta of these pregnancies, as opposed to a deficiency in the maternal cardiovascular system or dysfunctional maternal cardiovascular adaptation to pregnancy, was the instigating factor in the development of the disorder. Pre-eclampsia is a multisystem disorder of pregnancy. The definitions have evolved over time, depending on advances in our ability to detect distinct features of the syndrome. Proteinuria was first described in association with eclampsia in 184353. While the triad of hypertension, proteinuria and edema was long regarded as the definitive standard of the disease, the symptom ‘edema’ was dropped, due to its lack of specificity, in the 1970s54, and proteinuria is now no longer mandatory for the diagnosis of the disease, according to the 2019 definition of the International Society for the Study of Hypertension in Pregnancy (ISSHP)55. Adoption of the definition ‘the new onset of hypertension and proteinuria after 20 weeks of gestation’ had long-lasting implications for research and clinical practice, as most studies applied this definition as an endpoint. ‘Hypertension and proteinuria’ still comprise the gold standard for disease diagnosis. However, this has been shown to have a positive predictive value of just 20% to detect pre-eclampsia-associated complications56. The new definition put forward by the ISSHP55 was shown to increase sensitivity but decrease diagnostic specificity, paralleled by decreased mean severity of outcomes such as gestational age at delivery and birth weight in screen-positive women57. Based on five decades of extensive basic and clinical research36, into the placental etiology of pre-eclampsia, and colleagues presented a of an definition of pre-eclampsia on model the main alternative pathways leading to the diagnosis of the of pre-eclampsia, with or without FGR, or leading to the development of FGR without pre-eclampsia. The authors between the of the according to maternal circulating PlGF as a of placental The maternal cardiovascular system plays a role in the development and clinical of pre-eclampsia. in cardiac function are in normal and pre-eclamptic pregnancies, and changes may persist for years postpartum. While the cardiovascular theory has not been shown to better of pre-eclampsia with to placental of maternal cardiac may be to the of women at high risk of or diagnosed with pre-eclampsia. of the unit and the between its is key to the understanding of pre-eclampsia. The clinical of pre-eclampsia disordered development in one or many of the While the maternal cascade and fetal effects may appear the may biomarkers, such as sFlt-1 and PlGF, a means to placental of maternal cardiac function may data for and of the placental and maternal of the disease may in both a means to develop an definition to predict placental maternal and fetal and therapeutic interventions to

Topics & Concepts