Introduction

Preeclampsia (PE), a multisystem vascular syndrome, is characterized by the gestational onset of hypertension and proteinuria and generally occurred after 20 weeks’ of gestation [1]. PE is one of the major causes of maternal and perinatal morbidity and mortality, especially in low- and middle-income countries [2, 3]. It affects approximately 5–8% of all pregnancies around the world [4], and is responsible for almost 350,000 maternal deaths [3, 5], six million perinatal deaths [6], eight million preterm births [7], and approximately 20 million low-birth-weight newborns in developing countries [8]. Moreover, it is demonstrated that PE is associated with higher risks of chronic noncommunicable diseases in later life of affected women [9]. Despite several researches to identification of the major risk factors and potential mechanisms and much advance in our knowledge, the PE’s etiology remains elusive. However, it is hypothesized that PE’s etiology is multifactorial, involving both maternal and placental contributions [10].

The maternal infections, especially those that are transmissible in utero, are responsible for several incidences of morbidity and mortality during pregnancy [11]. TORCH complex, comprise toxoplasmosis, other (syphilis, varicella-zoster, parvovirus B19, Hepatitis B), Rubella, Cytomegalovirus (CMV), and Herpes infections, hepatitis infections, human immunodeficiency virus (HIV) are the most common transplacentally acquired infections by the fetus [11, 12]. These infections are responsible for several congenital anomalies like still birth, abortion, intrauterine fetal deaths, congenital malformations, and other congenital failures [11, 13,14,15]. Zika virus infection is considered as the newest TORCH complex that is associated with several complications such as intrauterine fetal infection, microcephaly, neurological abnormalities, and Guillain–Barré syndrome [16,17,18]. Moreover, coinfection with the above mentioned microorganisms can lead to more adverse effect on mother and fetus and also can lead to difficulties in actual diagnosis [19, 20].

In recent years, a growing body of literatures suggests that infections by bacteria, viruses, and parasites and their related inflammations play an important role in the pathogenesis of PE [21, 22]. Several maternal infectious agents including HIV, malaria, different bacteria, and periodontal disease or urinary tract infections (UTIs) have been suggested to increase the risk of PE [21, 22]. This narrative review will discuss the role the infectious agents in development of PE and possible mechanisms related to these infections that are involved in PE. For this purpose, we searched PubMed, Google scholar, and Cochrane databases using the following search words: “infection and preeclampsia,” “bacterial infection and preeclampsia,” “viral infection and preeclampsia,” and “parasitic infection and preeclampsia.” Publications in English language were considered, but we did not impose any study design or geographic limitations. Review was conducted on more than 38 potentially relevant articles published between 2009 and 2017.

Bacterial infection

In the last three decades, several epidemiological and casual studies have evaluated the possible relationship between maternal bacterial and viral infections and PE. Many of these studies indicated a positive association between bacterial and viral infections and PE [23,24,25,26,27,28]. Moreover, intrauterine infections and their resulting inflammatory responses are responsible for early preterm births and labor in pregnant women [29,30,31]. Rustveld et al. [21] in a meta-analysis study have demonstrated that women with either a bacterial or viral infection had twofold higher risk to develop PE compared with women without infection (OR 2.1; 95% CI 1.6–2.7).

Several retrospective and prospective studies demonstrated potential risk of UTIs by bacteria to induce PE [32,33,34,35]. Several pathogenic bacteria are responsible for UTIs including Escherichia coli, responsible for 70–80% of all UTIs in pregnancy; Gram-negative bacteria include Klebsiella, Enterobacter, Proteus, Pseudomonas, Citrobacter, and Gram-positive bacteria, for example, group B Streptococci [36]. Moreover, other bacteria, including Ureaplasma parvum, Gardnerella vaginalis, Mycoplasma hominis, lactobacilli, and Chlamydia trachomatis, have also been reported to induce UTIs [33, 37,38,39]. It is demonstrated that the early diagnosis and treatment of UTI decreased the incidence of PE by 64% [40]. Hsu et al. in a retrospective study on 13,852 pregnant women reported that risk of PE was significantly more (OR 4.2; 95% CI 1.1–5.1) among women who developed UTIs during pregnancy [32]. Conde-Agudelo et al. [22] in a comprehensive meta-analysis study have shown that that risk of PE was increased in pregnant women with UTIs (OR 1.57; 95% CI 1.45–1.70).

Periodontal disease (PD) is another most prevalent human infection disease that has been reported as potential risk factor for the development of PE. Several genera of bacteria include Treponema, Bacteroides, Porphyromonas, Prevotella, Capnocytophaga, Peptostreptococcus, Fusobacterium, Actinobacillus, Tannerella, and Eikenella, and their related species are involved in PD [41]. In a case–control study, Contreras et al. have found that the presence of microorganisms related in PDs such as Porphyromonas gingivalis (OR 1.8; 95% CI 1.1–2.8), Tannerella forsythia (OR 1.8; 95% CI 1.1–3.0), and Eikenella corrodens (OR 1.8; 95% CI 1.1–2.8) were significantly associated with the development of PE in pregnant women [42]. Conde-Agudelo et al. in their meta-analysis reported significant relationship between the presence of PD and the induction of PE (OR 1.76; 95% CI 1.43–2.18) [22]. Similar finding (OR 2.79; 95% CI 2.01–3.01) was achieved in a recent meta-analysis study by Wei et al. [43]. Two comprehensive studies by Rustveld et al. and Conde-Agudelo et al. have been published in 2008; we summarized recently conducted studies (2009–2016) regarding the association between UTIs and PD [2, 34, 35, 44,45,46,47,48,49,50,51] with development of PE in Table 1.

Table 1 Studies regarding the associations between increased risk of preeclampsia and urinary tract infections (UTIs) and periodontal disease (PD)
Full size table

Chlamydia pneumonia, Ureaplasma urealyticum, and Helicobacter pylori are other bacterial organisms that were reported as potential risk factors for the development of PE [52, 53]. Heine et al. using a case–control study demonstrated that women with elevated titers of IgG to C. pneumoniae have a threefold increased risk of PE (OR 3.1, 95% CI 1.2–7.9) compared with healthy controls [23]. Although in other studies, no significant correlation was observed between C. pneumoniae and PE [54,55,56], Dadelszen et al. in a prospective cohort study reported that women with early-onset PE had significantly higher levels of IgG to Chlamydophila pneumonia compared with normotensive pregnant women [24].

The significant relationship between H. pylori infection and PE was described for the first time by Ponzetto et al. [57]. They reported that seropositive women for H. pylori had almost threefold higher risk to develop PE [57]. Subsequent studies reported also a significantly higher H. pylori seropositivity rate in preeclamptic women compared with controls [25, 26, 58,59,60,61, 65, 66]. Moreover, recent studies described that infection with Cytotoxin-associated antigen A (CagA)- and Vacuolating cytotoxin A (VacA)-positive H. pylori-strains is significantly related with PE and, especially with “placental PE” [58]. This could be explained by the fact that CagA- and VacA-positive H. pylori strains are generally associated to higher levels of inflammatory mediators compared with negative strains [62]. Recently conducted studies (2009-2017) regarding the associations between C. pneumonia, C. trachomatis, and H. pylori infections and PE [25, 26, 39, 52, 58,59,60,61, 63,64,65,66] are summarized in Table 2.

Table 2 Studies regarding the associations between increased risk of preeclampsia and bacteria
Full size table

Viral infection

Among the viral pathogens, Cytomegalovirus (CMV), Adeno-Associated Virus-2 (AAV-2), herpes simplex virus type-2 (HSV2), Epstein-Barr virus (EBV), and human immunodeficiency virus (HIV) have been more studied and found to have more likelihood to be effective in PE. Dadelszen et al. through a nested case–control study indicated that women with early-onset PE had higher anti-CMV levels than women with late-onset PE and normotensive women (P < 0.05) [24]. Similar results were obtained by Xie et al. in two subsequent studies [28, 67]. Moreover, their results showed that women with PE have upregulated TLR-2/-4 mRNA expression, increased levels of serum IL-6 and TNF-α, and reduced IL-10 compared with matched normal and nonpregnancy controls [67]. Arechavaleta-Velasco et al. [68] in a molecular study reported that rates of AAV-2 placental infection were significantly higher among women with severe PE compared with women having normotensive placentas (P = 0.002). In a subsequent study, same team reported that the first-trimester maternal IgM seropositivity for AAV-2 was 5.6 times more prevalent among PE (P = 0.0004) than in healthy controls [69]. Rustveld et al. have found that seroconversion for HSV 1/2 or CMV was associated with a fivefold increased risk for developing PE (OR 5.4; 95% CI 1.0–29.0) after adjusting for education, income, smoking, years of cohabitation, medical insurance, and type of birth control [70]. In another study, Trogstad et al. reported an increased risk of PE among women who were seronegative for, and therefore at the risk of acquiring EBV (OR 3.5; 95% CI 1.1–10.6), CMV (OR 1.6; 95% CI 0.8–3.2), and HSV-2 (OR 1.7; 95% CI 0.7–4.2) infections [71]. Moreover, there are several studies regarding the effect of HIV on hypertensive disorders of pregnancy [72,73,74,75,76]. Based on data from a comprehensive meta-analysis study [77], however, no evidence was found regarding the relationship between HIV infection and PE (OR 1.04; 95% CI 0.60–1.79), but significant associations were observed regarding HIV infection with hypertension (OR 1.46; 95% CI 1.03–2.05) and eclampsia (OR 2.56; 95% CI 0.15–44.11) among pregnant women. Recent studies (2009–2017) regarding the associations between viral infections and PE [28, 39, 64, 67, 78,79,80,81,82,83,84] are summarized in Table 3.

Table 3 Studies regarding the associations between the increased risk of preeclampsia and viral infections
Full size table

Mechanisms for bacterial and viral infections

Inflammation responses against infections play important roles in the initiation and enhancement of acute uteroplacental atherosis or destruction of trophoblast cells, major risks known to induce PE [68]. Moreover, clinical and epidemiologic data indicated that acute atherosis is directly associated with PE [70, 85]. The increase of monocytes circulation resulting from infections and establishment of macrophage foam cells in the arterial intima could be the key factors to induce early lesion in atherosclerosis [70, 86,87,88]. Moreover, several studies have demonstrated that inflammatory responses are excessive in preeclamptic pregnancies compared with normal pregnancies [89,90,91,92,93]. Bacterial and viral infections during pregnancy could stimulate release of high level of pro-inflammatory cytokines (TNF-ɑ, IL-12, IFN-γ, etc.) and also increase of oxidative stress and endothelial cell dysfunction, all of which could lead to initiation of hypertension disorders including PE [70]. Increased levels of oxidative stress induced by Chronic or acute infections could impair the production and bioactivity of nitric oxide (NO) that can lead to endothelial dysfunction, a crucial event to induce the PE. In agreement with this statement, PE-like manifestations were observed in experimental models by blocking endothelial production of NO [94,95,96,97].

In addition to the above mentioned mechanisms, some antigenic factors of microorganisms like Cytotoxin-associated antigen A (CagA) may be directly related PE [98]. Recent studies showed that anti-CagA antibodies are able to cross-react with antigens (β-actin proteins) of endothelial cells and cytotrophoblast cells of placenta that can lead to negative effects on its invasiveness ability [99, 100]. Moreover, it is reported that anti-CagA antibodies are able to inhibit the activation of mediator factors that are important during trophoblast proliferation, such as ERK and Nuclear Factor-kB [99].

Parasitic infections

Among the parasite diseases, infections with protozoa such as Plasmodium spp., Toxoplasma gondii, Trichomonas vaginalis, and Trypanosoma cruzi could be potential risk factors for PE, mainly regarding vertical transmission, placental infection, and the host immune response to them. Infection with these protozoa during the pregnancy could result to low birth weight, still birth, spontaneous abortion, growth restriction, intrauterine fetal death, and fetal abnormalities [101]. There are no study exploring role of the trypanosomiasis in development of PE, although some evidences are available regarding parasite invasion to the placenta and inhibit the implantation and cell division [101]. Toxoplasmosis is one of the most prevalent infection disease with worldwide distribution [102]. It is the cause of many adverse complications in immunocompromised patients and pregnant women [103, 104]. Todros et al. [105] in a cohort study indicated that pregnant women treated with spiramycin have shown lower pregnancy-induced hypertension (OR 0.092; 95% CI 0.021–0.399) compared with women who did not take any antibiotic during pregnancy. Although in a recently conducted study in Mexico, it was reported that chronic toxoplasmosis is not associated with hypertensive disorders in pregnant women [106].

Although, there is any study indicating the role of T. vaginalis in PE, but secretion of galectin family (galectin-1 and galectin-3) by cervical and vaginal epithelial cells upon T. vaginalis infection could be possible mechanism to induce PE, as these galectins modulate the inflammatory responses [107]. Than et al. [108] have demonstrated that placental expression of galectin-1 was significantly higher in patients with severe PE than in normal controls, and these increases may represent a fetal response to an exaggerated systemic maternal inflammation. Moreover, T. vaginalis lipophosphoglycan motivates species-specific inflammatory response and selective chemokine (IL-8 and macrophage inflammatory protein-3ɑ) upregulation by human cervical and vaginal epithelial cells [107, 109].

Malaria especially induced by P. falciparum has been the most-described parasitic mechanism to cause PE. Annually, 125 million pregnant women are at risk of malarial infection in malarious areas, and the excess risk of infection varies with gravidity [110, 111]. There are several epidemiological overlaps between malaria infection and PE [112]. Both malaria and PE have markedly higher risk in young and primigravidae women and also similar seasonal distribution [112,113,114,115,116]. In addition, some observational studies reported placental malaria as a potential risk factor for developing increased maternal hypertension and related disorders, including PE [117,118,119,120,121]. Sartelet et al. [117] in 1996 reported that placental malaria was significantly related with PE in Senegalian pregnant women (OR 3.0; 95% CI 1.3–6.9). Muehlenbachs et al. in Tanzania indicated that malaria was associated with the increased risk of hypertension in young (18–20 years old) first-time mothers (OR 3.1; 95% CI 1.1–9.0) [115]. In a recent case–control study in Sudan, placental malaria (OR 2.3; 95% CI 1.0–5.2) was significantly associated with PE [120].

There are several evidences to multiple mechanisms contributing to PE associated with malaria in pregnant women including anemia, alternation in immunological milieu, and increase of anti-angiogenic proteins that contribute to the pathogenesis of PE such as sFlt-1 and endoglin [115, 116, 122, 123]. Moreover, Muehlenbachs et al. [115] reported that soluble vascular endothelial growth factor receptor 1 (sVEGFR1), a preeclampsia biomarker, was significantly increased in first-time mothers with both malaria infection and hypertension. Complement C5a, which plays an important role in induction of inflammation and initiation of acquired immune response, is another factor that is elevated in malarial infection and has significant role in the induction of PE [122, 124,125,126]. The shift of immunological cytokine profile of Th2, typical immune profile associated with pregnancy, to Th1 includes TNF-ɑ, IL-12, and IFN-γ. immune profile more suited to parasite killing is another important possible mechanism by malarial infection to induce preeclampsia [15, 127] [110, 111].

Conclusion

PE is the major public health problem in both developed and developing countries. However, etiology of PE is multifactorial, but infections and their related inflammations have important roles in the development of PE. Pro-inflammatory cytokines and other cellular mediators induced by periodontal, vaginal, or urinary infections cause endothelial dysfunction that is essential alteration in the pathophysiology of PE. Early diagnosis and treatment of bacterial, viral, and parasitic infections could be an effective strategy to reduce the incidence of PE. In order to achieve significant decreases in maternal and perinatal mortality rates resulting from PE, it is crucial that the screening and treatment of common maternal infections be incorporated into the prenatal care programs.