In our study, we analyzed the placental morphology and expression of angioactive factors at two different time points during the early stages of SARS-CoV-19 infection: first, during viremia (RT-qPCR-positive pregnant women, early infection) and late infections (RT-qPCR-negative but IgG-positive) at the time of birth. We did not observe any degenerative changes in placental structure in patients who tested positive for viral infection (RT-qPCR-positive). However, vascular dilation and congestion were observed in the capillary villous vasculature. This has been interpreted as a compensatory hemodynamic process or, in some cases, a sign of vascular dysfunction [16]. COVID-19 is linked to blood clotting issues and venous thromboembolism [17, 18], which could potentially affect the maternal–fetal hemodynamic system. Research involving image-based modeling of blood flow has indicated that optimal vascular dilation can enhance oxygen transfer [19]. Therefore, the observed vascular structure may suggest an early adaptive response of the placenta to maternal vascular changes, possibly aimed at increasing the oxygen and nutrient absorption capabilities.

However, later infections (IgG positive patients) completely changed this situation. Infected patients show increased degenerative changes in their placentas, including syncytial knots, intervillous fibrin deposition, central villous infarction, and chorangiosis, suggesting that these changes develop during infection due to chronicity or an organic response. This aligns with other studies that identified increased central and perivillous fibrin deposition, syncytial knotting as placental markers of maternal vascular malperfusion in COVID-19 patients [20,21,22,23].

Several studies have identified changes in the placenta due to SARS-CoV-19 infection; the exact cause of these changes remains unclear. Evidence suggests that there may be partial or intermittent obstruction of blood flow to the fetus, such as blockage of the chorionic plate and villous vessels, which can lead to blood clotting and fibrin deposition in the placenta. Perna et al. [24] suggested that the development of these lesions depends on the duration of infection and the pregnancy stage. The authors showed that the villous fibrin deposits and other changes are more pronounced when infection occurs during the third trimester, which supports our results.

Later infections also present with chorangiosis (increased capillaries in the chorionic villi). Although chorangiosis is normal at term, it is exacerbated in pregnancies associated with maternal complications such as chronic prenatal hypoxia [25, 26]. The chorangiosis here is seen as a secondary placental response to SARS-CoV-19 infection, suggesting a need for increased blood and oxygen flow in the chorionic villi, as evidenced by premature increases in vessel dilation and congestion in early infected placentas.

SARS-CoV-2 infection triggers the release of pro-inflammatory cytokines, which play a crucial role in activating immune and blood vessel cells (leukocytes and endothelial cells). This activation is strongly associated with the prothrombotic state during SARS-CoV-2 infection [9]. This condition is also thought to be closely linked to angiogenesis, which promotes the formation of new blood vessels to ensure sufficient oxygen and nutrients during stressful situations [27]. In line with our morphological findings, we examined the protein expression of angioactive factors through immunohistochemical staining for VEGF, PlGF, iNOS, eNOS, and COX2.

In our study, we found high levels of VEGF protein at both stages of infection, indicating continued pro-angiogenic, pro-permeable, and vasodilatory placental activity since the early stages. Several studies suggest that serum levels of angiogenic factors, such as VEGF and PlGF, change in pregnant women with COVID-19, particularly when the infection occurs during the third trimester [28,29,30,31]. The placenta actively produces these factors, suggesting active participation in regulating serum levels [13, 14, 32, 34], which aligns with our results and increases VEGF placental expression.

Although PlGF also exhibits angiogenic activity, we observed reduced levels after the acute phase of infection. Previous reports have suggested that in COVID-19, endothelial damage may reduce maternal PlGF levels, similar to preeclampsia. This decrease is linked to pathological interference with the angiotensin receptor ACE2 [34]. Although existing research has focused on maternal serum levels, it is essential to recognize that the placenta is the primary source of PlGF, and any changes in maternal levels may also reflect alterations in placental production [33]. The decrease in placental PlGF protein levels in our study supports this possibility.

The development of blood vessels in the placenta, and their expansion and relaxation, mainly depend on VEGF expression. However, other signaling molecules are involved in or associated with their action. While the exact pathways that control this process are not fully understood, it is well established that VEGF-induced permeability involves the production of nitric oxide (NO) by endothelial nitric oxide synthase (eNOS) [12, 35]. Krause and co-workers demonstrated that nitric oxide, generated by endothelial and inducible nitric oxide synthases (eNOS and iNOS, respectively), plays a vital role in placental physiology and functions as the primary vasodilator [36]. Two primary mechanisms have been suggested for this activity: opening the pores between endothelial cells and forming junctions between them [37].

To address whether nitric oxide could contribute to placental angioactivity in infected mothers, we also analyzed the expression of eNOS and iNOS proteins. Our results showed sustained expression of NOS isoforms, with eNOS predominating in the early stages of maternal infection, followed by an increase in iNOS in the later stages. Although placental NOS isoforms are temporally expressed during gestation [36], an increase in these isoforms, along with an increase in VEGF, may indicate a possible need for increased vasodilation at the maternal–fetal interface in response to the thrombogenic activity of SARS-CoV-2. Interestingly, the later expression of iNOS occurs in parallel with that of COX2, both of which are implicated in the pathophysiology of inflammation and angiogenesis [38, 39]. Similar to eNOS, iNOS is also involved in the production of nitric oxide (NO), but only after induction by an inflammatory condition [40]. COX2 is an essential enzyme in the synthesis of lipid mediators, particularly prostaglandin E2. Enhanced COX2-induced prostaglandin synthesis stimulates angiogenesis, among many other associated processes, including modulation of inflammatory responses, which are also observed following microbial infections [41, 42].

Robust scientific knowledge indicates that immunological dysregulation and inflammation contribute to the development, progression, and severity of COVID-19 [43]. In this context, the presence of factors such as iNOS and COX2, which are not only associated with angiogenesis but are also linked to an inflammatory response, aligns with these findings. It is important to note that the increased expression of these factors occurs in the later stages of infection, after the inflammatory process has already been established.

This study outlined several limiting factors, including sample size, gestational comorbidities during SARS-CoV-2 infection, and the precise timing of the infection. Our samples were collected at a municipal public care hospital in the initial months following the announcement of the COVID-19 pandemic. Our study design aimed to exclude pregnant women diagnosed solely based on symptoms that were not consistently confirmed by molecular tests. Thus, the samples were limited to patients who underwent at least two tests during the peripartum phase (RT-qPCR and IgG), and whose neonates and placentas were also analyzed, which significantly reduced our sampling. Another limiting factor in our study was the inclusion of pregnant women with comorbidities such as diabetes and hypertension. To address this potential issue, we excluded these samples from an additional analysis, as detailed in Supplementary Tables 1 and 3. The results continued to demonstrate statistical differences between the control and infectious groups, similar to those observed in the full sample, suggesting that the changes were associated with the infectious condition. Another critical point in our study was the exact moment of infection, which is usually unclear. Instead of determining a single moment of infection, we defined two periods: early infection for patients who were RT-qPCR-positive and IgG-negative, under the assumption that older infections should also test positive for IgG; and late infection for patients who were IgG-reactive, indicating an infection that had been contracted previously and was only detectable through IgG levels. Our data indicates significant morphological and functional placental adaptations during COVID-19 infection. However, these limitations highlight the need for future studies with larger, more homogeneous samples to validate our findings.