The placenta becomes the primary source of estrogen after week 9 of human pregnancy, when levels of estradiol increase by 50-fold during pregnancy (4). In vitro studies have demonstrated that estradiol-17β induces proliferation, migration and adhesion of endothelial cells (ECs) (5–8), critical processes underlying uterine angiogenesis and facilitating the increase in UPBF for a healthy pregnancy establishment. These effects have been observed in both human umbilical vein endothelial cells (HUVECs) and uterine arterial endothelial cells (UAECs) (5–8). In addition, estrogens contribute to uterine vascular remodeling by enhancing the transcriptional activity of serum and glucocorticoid-inducible kinase-1, which in turn modulates matrix metalloproteinases-2 (MMP-2) and E-cadherin expression in human villous samples (9).

Animal studies further demonstrate the role of estrogens in promoting vascular remodeling and hemodynamic adaptation. During early primate pregnancy, estrogens promote trophoblast cell proliferation, differentiation, viability, and invasion of the spiral UA (10). In pregnant rats, estradiol was shown to activate extracellular MMPs inducer, which in turn increased the expression and activity of MMP-2 and MMP-9 in uterus and aorta of late-pregnant rats vs virgin and mid-pregnant rats, supporting its role in vascular remodeling (11). However, in baboons, as gestation advances, the rise of estrogenssuppresses trophoblast uterine invasion (10) by inhibiting vascular endothelial growth factor (VEGF) expression and increasing soluble fms-like tyrosine kinase-1(sFlt-1) levels (12).

Furthermore, several studies have shown that estrogens vasodilate UA (13–16), favoring the augmentation in UPBF during pregnancy. In pregnant ewes, systemic infusion of estradiol-17β induced a 5-fold increase in UPBF, mainly mediated by large-conductance Ca²⁺-activated K⁺ channel (BK_ca) and NO (17). In rats, estrogens were shown to act through the G protein-coupled estrogen receptor (GPER) (18, 19), promoting endothelium-dependent vasodilation via the NO–cyclic guanosine monophosphate (NO–cGMP) signaling pathway (20). GPER-mediated vasodilation also involves L-type Ca²⁺ channels and ERK1/2 activation in VSMCs (16).

In human myometrial arteries, vasodilation in response to 17β-estradiol is mediated by the greater expression of estrogen receptors-α, and -β (ER-α, -β) (15).

The effects of estrogens on uterine circulation during pregnancy have been extensively reviewed elsewhere (18, 21).

Progesterone is a steroid hormone whose levels during pregnancy are almost 10 times higher than during the luteal phase of the menstrual cycle (22). Around the 10^th week of gestation, the placenta takes over progesterone production from the corpus luteum, maintaining serum concentrations of at least 10 ng/ml, which are essential for sustaining a healthy pregnancy (23). Progesterone contributes to thickening and expanding the endometrial lining, thereby enlarging the surface area available for implantation of the fertilized egg (24). It also facilitates trophoblastic invasion of the spiral UA, thereby promoting growth and remodeling of the uterine vasculature and increasing UPBF. Moreover, progesterone induces uterine vessel vasodilation by acting on both ECs and VSMCs. In ECs, it upregulates the expression and activity of eNOS (25) as well as cyclooxygenase-1 and -2 (COX-1, COX-2) (26), increasing the production of NO and PGI₂ respectively (27). In VSMCs, progesterone downregulates the protein kinase C pathway (PKC), further contributing to vessel vasodilation (28).

Furthermore, progesterone, along with estrogen, promotes vascular remodeling by enhancing MMPs expression and activity (11). However, the effect of progesterone in regulating UPBF during pregnancy appears controversial (25); therefore, further studies are needed to elucidate its effects and the molecular mechanisms through which progesterone modulates the uterine vascular function during pregnancy.

Relaxin is produced mainly by the corpus luteum and, during pregnancy, also by the placenta. Relaxin reaches a peak of about 1 ng/ml by the end of the first trimester, then its levels gradually decline and plateau at ~0.6 ng/ml (29). Relaxin contributes to the growth and softening of the cervix, facilitating rapid labor, and it also contributes to uterine vasodilation and increases compliance, favoring the augmentation in UPBF (29). In vitro studies have demonstrated that relaxin stimulates angiogenesis. Experiments on human endometrial stromal cells showed that relaxin treatment induces a variety of angiogenesis-related genes, including VEGF-A (30). Moreover, relaxin has been shown to induce VEGF and basic fibroblast growth factor (31) expression in ischemic wound sites, supporting its pro-angiogenic activity.

Animal studies have provided strong evidence for relaxin role in vascular remodeling and uterine compliance. Kaftanovskaya et al., have demonstrated that relaxin-receptor-deficient mice had reduced lengths of the pubic symphysis and increased collagen density in reproductive tissues (32). Similarly, in pregnant relaxin-mutant mice, UA exhibited increased elastin expression and decreased levels of MMPs and adhesion molecules. Interestingly, 5 days of exogenous relaxin treatment reversed arterial stiffness and improved fetal weight in relaxin-deficient mice (33). Consistent with these findings, neutralization of relaxin in late pregnancy using the monoclonal antibody-1 (MCA-1) has been shown to increase uterine artery stiffness (34). Moreover, relaxin induced vasodilation in UA isolated from pregnant rats, with a greater effect during mid- compared to late pregnancy through the NO-sGC pathway (35). Recent evidence suggests that the vasodilatory responses of relaxin are mediated by its major receptor, the relaxin/insulin-like family peptide 1 receptor (RXFP1), which is most highly expressed in the UA during early pregnancy (34). Newly emerging data support that relaxin binds to RXFP1 activates Gαi/o protein, coupling to phosphatidylinositol-3 kinase/Akt (protein kinase B)-dependent phosphorylation and activation of eNOS (36). Nonetheless, relaxin inhibits spontaneous myometrial contractile activity in mid-gestation, while it has no effect at term (35).

Human studies suggest a potential role of relaxin in regulating uteroplacental vascular function. A positive correlation between serum relaxin levels and UA resistance index at 10–12 weeks of gestation has been observed, suggesting that relaxin contributes to the regulation of the uteroplacental vasculature. Moreover, suppression of circulating relaxin throughout mid-pregnancy abolished the cardiovascular adaptations required to sustain a healthy pregnancy (37). Thus, the sensitivity of the uterus to relaxin is subjected to modulation through pregnancy, suggesting that relaxin sustains an adequate UPBF during mid-gestation, while facilitating labor at term, allowing for vasoconstriction.

In summary, while animal studies show promising roles in vascular remodeling and compliance, human studies are limited and mostly observational, lacking mechanistic insights into RXFP1 signaling in the uterine vasculature during normal and pathological pregnancies

In addition to steroid hormones, the placenta secretes a wide range of angiogenic factors, including VEGF, placental growth factor (PlGF) and sFlt-1, which collectively contribute to the regulation of the uterine circulation. In healthy pregnancies, maternal plasma VEGF concentrations are markedly elevated compared to non-pregnant baselines. It rises during the first trimester, peaks at 10–14 weeks, remains elevated up to the 20^th week of pregnancy, then declines in the third trimester (38, 39). During the first trimester of pregnancy, VEGF levels increase 4–5 fold (163.2 ± 81.6 pg/mL) compared to the non-pregnant state (18.5 ± 16.8 pg/mL) (40). VEGF contributes to maternal adaptations to pregnancy by increasing vascular permeability, stimulating angiogenesis and inducing vasodilation of the uterine circulation.

The expression of VEGF mRNA in both the uterine endometrium (41) and placenta (42) indicates its active role at the feto-maternal interface, where tightly regulated modulation of vascular permeability and angiogenesis is crucial for the establishment of a successful uteroplacental circulation. VEGF-mediated vascular permeability has been demonstrated in the uterine vasculature and is further enhanced by pregnancy (43, 44). The increase in permeability facilitates the extravasation of plasma proteins into surrounding tissue, facilitating ECs migration and thus the angiogenic process (45). In addition, VEGF induces the expression of serine proteases urokinase, tissue-type plasminogen activators, PA inhibitor 1 (46) and MMPs interstitial collagenase (47), consistent with a pro-degradative environment, which facilitates migration and sprouting of ECs during angiogenesis. Notably, VEGF angiogenic properties are enhanced during pregnancy (48). VEGF role in mediating vasodilation has also been shown in vitro. Indeed, VEGF promotes NO production in cultured UAECs (48) and stimulates PGI₂ synthesis in a time- and concentration-dependent manner in UVECs via activation of cytosolic phospholipase A₂/p42/p44 MAP kinases pathways (49).

The critical role of VEGF in angiogenesis is supported by evidence from genetically modified pregnant mice lacking VEGF gene or deficient in VEGF receptors, both of which result in impaired vascular development and pregnancy loss (50, 51). VEGF promotes NO production in vivo, in the UA of pregnant ovine (52), while reducing UA contractility and increasing UPBF (52–54) short- and long-term. Consistent with these findings, VEGF induced vasodilation of uterine arcuate artery (UAA) of both non-pregnant and pregnant rats, acting through an endothelium-dependent mechanism (55).

PlGF belongs to the VEGF growth factor family, and it binds specifically to VEGFR-1 (56). Plasma PlGF concentrations increase from week 11 to 12 onward to peak at week 30 in healthy pregnancy (57), coinciding with implantation and early vascular development. PlGF is expressed in villous trophoblasts and vascular endothelium in human placenta at term (58), and the correlation between PlGF expression and placental perfusion suggests that PlGF may contribute to ensuring adequate vascular development and function of the placenta early in gestation (59). PlGF contributes to uteroplacental circulation establishment by supporting angiogenesis, immune modulation and trophoblast invasion. Its role in vascular remodeling is supported by a study demonstrating a reduced uterine natural killer population in PlGF null mice (60). This sub-population of natural killer cells is the major participant in the early vascular changes in the pregnant endometrium. Thus, demonstrating that PlGF-mediated remodeling of the UA at the feto-maternal interface is necessary to sustain an adequate UPBF. Supporting this finding, an increased level of PlGF correlates with improved placental perfusion (59). In addition, PlGF potentiates the angiogenic response to VEGF on microvascular ECs (61), thereby promoting the angiogenic process at the feto-maternal interface. PlGF regulates the UPBF, inducing vasodilation, as demonstrated in placental arteries (62) and uterine circulation of rats and humans (63). The mechanism of PlGF-mediated vasodilation has been associated with signaling through VEGFR-1 and NO involvement, according to the vascular bed (63).

In addition to the well-known sex hormones and growth factors discussed above, whose role in pregnancy-associated uterine vascular remodeling has been widely demonstrated and established, several studies have suggested other placental hormones, proteins and growth factors as critical contributors to a successful uterine vascular remodeling.

PE is defined as new-onset hypertension after 20 weeks of pregnancy, with or without proteinuria and evidence of end-organ damage. Clinically, PE is associated with several maternal and fetal complications (133), and is generally classified in early-onset (<34 weeks of gestation), which tends to be more severe, and late-onset (>34 weeks of gestation), which is usually milder. The placenta has a crucial role in the pathophysiology of both early- and late-onset PE, though the specific mechanisms differ. Early-onset PE is more strongly associated with abnormal placentation, poor uterine perfusion, and IUGR (134), whereas late-onset is primarily attributed to predisposing maternal factors and placental senescence (134).

It is generally accepted that PE originates from the abnormal transformation of the spiral arteries underlying the placenta, due to impaired invasion of the uterine wall by migrating trophoblast cells (135, 136). In addition to incomplete or absent vascular remodeling, signs of vascular damage, resembling atherosclerosis-like lesions, have been described in placental bed samples from women with PE compared to women with healthy pregnancies (135). This maladaptive phenotype leads to narrower maternal uterine vessels and relative placental ischemia, ultimately reducing the UPBF as demonstrated by Doppler ultrasound (137–139).

Numerous hormones and proteins of placental origin have been investigated to elucidate the molecular mechanisms underlying the aberrant uterine vascular phenotype observed in PE. It is well-established that PE is associated with a decrease in pro-angiogenic factors such as PlGF and VEGF, coupled with an increase in anti-angiogenic factors such as sFlt-1 and sEng.

Supporting this notion, a reduction in circulating PlGF has been associated with placental hypoperfusion in both human (140) and animal models of PE (141). Placental hypoperfusion, indicative of defective adaptations of the uterine circulation, may be partially explained by a substantial elevation of sFlt-1 observed in PlGf^-/- mice (142). Notably, PlGF administration dampened sFlt-1 increase, restoring angiogenic signaling in a PE-like model (141). The anti-angiogenic effects of sFlt-1 are further exacerbated by sEng, which has been shown to induce a severe state of PE in pregnant rats (143). The exacerbation of the PE-like symptoms is further supported by in vitro evidence showing that, in UVEC, sEng treatment contributes to endothelial dysfunction by inhibiting NO production, reducing cell viability, impairing trophoblast invasiveness, as well as suppressing MMPs expression (144).

Thus, PE is characterized by a shift toward an antiangiogenic profile, which disrupts the normal vascular homeostasis, resulting in endothelial cell dysfunction, including decreased NO production, and release of procoagulant proteins. Hence, contributing to hypertension and reduced UPBF.

IUGR is defined as fetal weight estimated to be below the 10^th percentile for its gestational age. The most common cause of IUGR is poor placental function and placental ischemia due to deteriorated uteroplacental perfusion (145, 146). A strong association between IUGR and PE has been consistently observed, with IUGR frequently arising in pregnancies complicated by severe PE (147). Histopathological observations revealed a significant decrease in villous vascular density (148) and a maldevelopment of the placental terminal villous tree in placentae from IUGR pregnancies compared to those from normal pregnancies (149). These findings suggest that the aberrant vascular formation is a leading determinant in IUGR, explaining the sub-optimal UPBF. From a molecular standpoint, the angiogenic profile of IUGR placentas presents inconsistencies that may reflect compensatory and pathogenic mechanisms. A study from Barut et al., reported increased placental expression of VEGF-A, basic fibroblast growth factor and eNOS, in placental samples from IUGR mothers compared to control samples collected during the third trimester (150). This result suggests a possibly compensatory upregulation aimed at resolving placental hypoxia. However, Lyall et al. (151), found a reduction in VEGF expression in placental villous tissue from pregnancies complicated by IUGR and PE, indicating that angiogenic insufficiency may dominate in such pathological contexts. Similarly, the expression of PlGF mRNA in placental samples from IUGR pregnancies has been inconsistent, with evidence demonstrating a decrease in PlGF expression only in severe cases of IUGR (152). Study using an ovine model of IUGR has demonstrated an increase in VEGF mRNA levels in the IUGR group compared to the control in early gestation, possibly reflecting early compensatory mechanisms. However, during mid-gestation, the expression of VEGF receptors in fetal tissue was significantly reduced, potentially limiting VEGF and PlGF signaling, thereby impairing angiogenesis and placental vascular development (153).

This finding may reflect a progressive failure of the placenta to sustain proangiogenic signaling under chronic stress, hence affecting UPBF.

GDM, defined as hyperglycemia first diagnosed during pregnancy (154), affects up to 20% of pregnancies worldwide (155), and it is frequently associated with uteroplacental insufficiency, as well as increased risk of IUGR and PE (156, 157).

Morphological and structural abnormalities have been documented in human GDM placentas (158, 159) as well as in placentas from hyperglycemic animal model, resulting in IUGR (160). Placentas from hyperglycemic-induced animal model exhibited disrupted trophoblast invasion, inadequate spiral UA remodeling (160), as well as dysregulation in PGI₂ levels (161, 162) and decreased levels of VEGF and PlGF (163, 164).

Collectively, the molecular, structural and functional abnormalities observed in GDM pregnancy suggest a defective vascularization due to impaired placental development in GDM pregnancy, affecting the uteroplacental circulation and thus compromising the UPBF.

PAS encompasses a spectrum of conditions characterized by abnormal placental adhesion to the uterine wall, which fails to detach at birth (165). The vast majority of PAS cases are attributed to scarring from cesarean births, which disrupts the endometrial-myometrial interface and promotes aberrant implantation (166–168). In PAS, EVTs excessively invade the myometrium (169), often within a rigid, collagen-rich extracellular matrix (170), accompanied by inflammation and hypoxia in uterus (171, 172). The absence of proper decidualization and EVT invasiveness prevents the remodeling of the uterine spiral artery in PAS (173), resulting in an abnormal pulsatile and high-velocity blood flow at the feto-maternal interface (174). Histopathologic analysis has revealed excessive vascularity in scarred uterine tissue in PAS (173) (175) and a lack of structural integrity of the vessel, potentially due to Von Willebrand factor suppression as demonstrated in ECs of PAS epiplacental artery (176).

The excessive vascularization is supported by upregulation of VEGF, angiopoietin-2 (177) and PlGF (178) in PAS lysates, and reduced expression of antiangiogenic factors such as vascular endothelial growth factor receptor-2 (VEGFR-2), endothelial cell tyrosine kinase receptor (Tie-2), and sflt-1 in syncytiotrophoblast cells from PAS placenta specimens (177). While evidence on VEGF levels in PAS remains controversial (178), hormonal regulation of angiogenic signaling may contribute to this imbalance. Indeed, the relaxin gene is upregulated in the basal plate of PAS placentas, together with RFXP1 in both the basal plate and villous trophoblast (179).

Animal models have recapitulated this condition, including placental dysplasia, incomplete remodeling of the spiral arteries, deep trophoblast invasion at the feto-maternal interface and reduced placental perfusion, as well as imbalances in angiogenic and anti-angiogenic factors within the placenta and in peripheral blood (180).

The ability of most drugs to reach therapeutic concentrations is limited by the placenta, which acts as a selective barrier between fetal and maternal circulation. To face these challenges, a range of methods for placental-targeted drug delivery have been developed. These include stem cell-based therapies, lipid nanoparticles therapies, and gene therapies.

Hormonal and protein biomarkers - Low levels of PAPP-A during the first trimester have been consistently associated with IUGR (248, 249). The predictive value improves when PAPP-A is combined with UA Doppler pulsatility index (249).

While some studies found no significant association between β-hCG and the occurrence of IUGR (249), a meta-analysis reported increased levels of β-hCG during the second trimester in pregnancy complicated by IUGR (250).

PP13 has shown potential as a predictive biomarker for IUGR, particularly when assessed during the second trimester, where PP13 levels are significantly. However, its predictive accuracy is higher for PE than for IUGR (251). Nonetheless, maternal serum PP13 in the first trimester may predict the risk of developing IUGR, PE, or PE complicated by IUGR (252). However, PP13 role in predicting IUGR remains contradictory (253).

A positive correlation between maternal PL levels and estimated fetal weight has been reported in a cross-sectional study, suggesting its potential utility in monitoring fetal growth (235, 254). Importantly, PL reduction has been significantly associated not only with the onset of PE, but also with asymmetrical IUGR, supporting its role as a marker of placental dysfunction (235).

Elevated maternal serum leptin concentrations at delivery have been documented in pregnancies complicated by IUGR compared to normal pregnancies (255). Similarly, elevated CRH levels in the third trimester of pregnancy have been associated with a 3.6-fold increase in IUGR risk (256). Molecular studies using in silico approaches further support these findings, showing increased RNA expression of PAPP-A2 and CRH in both maternal blood and placental tissue from IUGR pregnancies (257).

Additionally, higher levels of inhibin A and B have been observed in fetuses affected by IUGR compared to controls (258). Lastly, lower levels of serum PSG-1 have been reported in the late second trimester of IUGR pregnancies (259).

Growth factors biomarkers – During the third trimester, pregnancies with IUGR showed significantly lower levels of free and total PlGF (260). A meta-analysis revealed that pregnant women with IUGR and PE may have higher sFlt-1/PlGF ratio. A sFlt-1/PlGF ratio >33 showed strong predictive capacity in FGR (261). Nonetheless, similar to Flt-1/PlGF, sEng has high predictive value for IUGR along with PE and HELLP (262). IUGR pregnancies, along with PE, showed significantly elevated sEng concentrations and a strong positive correlation with sFlt-1 (263). Remarkably, placental transcript abundance for VEGF was significantly lower in IUGR placentae compared to healthy placentae (264). Additionally, increased sEng levels have been consistently associated with GDM, PE, and IUGR across different stages of gestation (265, 266). Despite these associations, a meta-analysis by Conde-Agudelo et al., including 53 studies and 39,974 women, concluded a minimal predictive accuracy of angiogenic factors for IUGR (267).

Hormonal and protein biomarkers - Increased levels of PAPP-A during the first trimester of pregnancy have been observed in women at higher risk of PAS (284, 285). In contrast, another study reported that PAPP-A levels were lower in PAS cases, while β-hCG levels were increased compared to the control group. Interestingly, combining these biomarkers in predictive models substantially improved diagnostic accuracy over single-marker use, with a sensitivity of 100% and specificity of 72% (286). Studies have also examined the diagnostic value of hyperglycosylated human CG. One study found that levels of this hormone were lower in PAS cases compared to controls in both the second and third trimesters of pregnancy. However, its diagnostic performance was limited, showing only a modest sensitivity and specificity (287).

Another hormone of interest is PL. Research has shown that during the third trimester, PL mRNA levels are increased in the plasma of women with invasive placentation (288). These results are in agreement with the study from Jing et al., which confirmed that maternal plasma PL mRNA concentrations were significantly increased in PAS cases compared to controls (289).

In addition to protein hormones, alpha-fetoprotein (AFP), a glycoprotein produced by the fetal liver and transported to maternal serum through the placenta or by diffusion across fetal membranes, has also emerged as a potential biomarker for PAS. A retrospective case-control study found increased levels of AFP in the PAS group during the second trimester, with a sensitivity of 71% and a specificity of 46% (290). Similar results were obtained in a subsequent study, which demonstrated a positive association between elevated second-trimester serum AFP and placenta accreta, suggesting its potential use as a screening tool for pregnancy at high risk of PAS (291).

Growth factors biomarkers - PAS is characterized by an excessive neovascularization, thus angiogenic factors have also been studied as potential biomarkers. A study found significantly increased maternal serum PlGF levels in PAS cases compared to the healthy group during the third trimester, while no differences were found in serum levels of VEGF (178). In the same study, PlGF is elevated in maternal serum and the placental bed of patients with PAS disorders compared to those without PAS (178).

In contrast, studies investigating sFlt-1 levels have yielded less conclusive results. One study reported no association between third-trimester sFlt-1 levels and PAS disorders (292). Another study suggested that although sFlt-1 and PlGF expression slightly differ depending on the depth of placental invasion, there is no direct correlation between serum levels and PlGF and sFlt-1 expression in the placenta (293).

Although these biomarkers show potential for detecting adverse pregnancy outcomes, large-scale studies are needed to conclude whether they could be used in early screening for pregnancy complications such as IUGR, PE, GDM, and PAS.