Adiponectin as a Biomarker of Preeclampsia: A Systematic Review

Abstract

Background/Objectives: Classified as a hypertensive disorder of pregnancy, preeclampsia is one of the leading causes of maternal and fetal morbidity and mortality. The abnormal trophoblast invasion that leads to a failed transformation of the uterine spiral arteries during placentation remains the most probable cause for preeclampsia. It is known that adiponectin acts on the placenta, playing a regulatory role in placentation processes. Therefore, the aim of this systematic review is to compile scientific evidence to evaluate the role of adiponectin as a biomarker for preeclampsia. Methods: The protocol for this systematic review was registered on the PROSPERO database (ID CRD42024542403) and follows the PRISMA 2020 guidelines. Overall, twenty-nine studies were selected from the PubMed and Scopus databases, including case–control, prospective and retrospective cohort, cross-sectional, and bidirectional Mendelian randomization studies. Results: From the articles analyzed, nine studies indicated an increase in adiponectin levels in preeclampsia, eleven reported a decrease, eight detected no significant changes, and in two studies, it was not possible to determine the glycoprotein levels. Analysis of the evidence quality revealed that moderate and low evidence levels predominate, with stronger evidence for decreased adiponectin levels. Conclusions: Promoting the advancement of scientific research is crucial, particularly exploring the association between adiponectin and other biomarkers. This approach could facilitate the development of screening and diagnostic methods, enabling the implementation of specific preventive and therapeutic strategies.

1. Introduction

Preeclampsia is a hypertensive disorder affecting approximately 2–8% of all pregnancies, representing one of the leading causes of maternal and fetal morbidity and mortality [1]. It is associated with proteinuria which develops after 20 weeks of gestation and pathological edema can also occur [2]. One of the risk factors for developing preeclampsia is maternal obesity, a condition characterized by adipose tissue dysfunction and, consequently, alterations in hormonal secretion patterns [3,4,5]. Although the pathophysiology of preeclampsia is not fully understood, most authors state that a failure in the transformation of uterine spiral arteries, resulting from an abnormal trophoblastic invasion during placentation, explains the development of this pathology [3,6,7]. As adiponectin is influenced by variations in body adipose tissue percentage and can act on the placenta, playing a significant role in regulating placental processes, a potential link is hypothesized between altered adiponectin secretion patterns and the development of preeclampsia [8,9].

1.1. Preeclampsia

Preeclampsia is characterized by the onset of hypertension, edema, and proteinuria after 20 weeks of gestation in a previously normotensive pregnant woman. However, this condition can also be associated with the development of systemic manifestations [2]. Therefore, additional criteria, when combined with the development of hypertension after 20 weeks of gestation, can aid in the diagnosis of preeclampsia (

Table 1. Diagnostic criteria for preeclampsia. Adapted from [3,10].

Diagnostic Criteria for Preeclampsia
Blood Pressure Elevation	- SBP ≥ 140 mmHg or DBP ≥ 90 mmHg on two occasions at least 4 h apart, after 20 weeks of gestation in a previously normotensive woman.
	- SBP ≥ 160 mmHg or DBP ≥ 110 mmHg, on one occasion.
and
Proteinuria	- ≥ 300 mg in a 24 h urine collection (or extrapolated from a timed collection). - Protein/creatinine ratio ≥ 0.3.
	- Urine dipstick test 2+ (if other methods are unavailable).
or (in the absence of proteinuria at least one of the following criteria)
Thrombocytopenia	Platelet count < 100,000/mm3 or other hematological complications such as hemolysis or disseminated intravascular coagulation.
Renal Insufficiency	Serum creatinine concentration > 1.1 mg/dL or doubling of serum creatinine levels in the absence of other renal diseases.
Impaired Liver Function	Elevated liver transaminases to twice the normal level, with or without upper right quadrant or epigastric abdominal pain.
Pulmonary Edema	-
Neurological Complications	New-onset headache unresponsive to medication and not attributable to another diagnosis; visual disturbances; altered mental status; eclampsia; stroke; or clonus.
Uteroplacental Dysfunction	Stillbirth, fetal growth restriction, or abnormal umbilical artery Doppler waveforms.

) [3].

For risk stratification purposes, preeclampsia was initially classified into two categories: mild and severe, based on quantitative values for blood pressure and proteinuria. Mild preeclampsia is characterized by a systolic blood pressure (SBP) of 140 mmHg or higher or a diastolic blood pressure (DBP) of 90 mmHg or higher, along with proteinuria exceeding 300 mg/24 h, a urine dipstick reading of 1+, or a protein-to-creatinine ratio of 0.3 mg/dL or higher. In contrast, severe preeclampsia is defined by an SBP exceeding 160 mmHg or a DBP exceeding 110 mmHg, proteinuria greater than 2 g/24 h, a urine dipstick reading of 3+, or the presence of symptoms involving multiple maternal organ systems [10].

Subsequently, a different classification was proposed based on gestational age [6,11]. Early-onset preeclampsia is defined as occurring before 34 weeks of gestation, while late-onset preeclampsia occurs after 34 weeks. Notably, early-onset preeclampsia carries higher risks and more severe consequences for both the mother and the fetus [6]. Additionally, some authors also consider preterm preeclampsia, when the delivery occurs between 34 and 37 weeks of gestation, and late-onset preeclampsia for term deliveries (after 37 weeks) [12].

1.1.1. Pathophysiology of Preeclampsia

Despite extensive research on preeclampsia, the disease’s pathophysiology remains highly heterogeneous and still uncertain [3,10]. It has been stated that it occurs in two steps: the first where abnormal placentation occurs in an early stage leading to the second step where maternal clinical manifestations of preeclampsia arise [13]. Although there is no supporting evidence obtained from pregnant women, several hypotheses have emerged for the placental dysfunction, including abnormal trophoblast invasion, oxidative stress, dysregulated natural killer (NK) cells at the maternal–fetal interface, as well as genetic and environmental influences [2,6,14].

In a healthy pregnancy, the trophoblast invades the decidua basalis of the endometrium and part of the myometrium to transform the spiral arteries into high-capacitance and low-resistance vessels, ensuring adequate blood flow to the placenta and fetus [2,6,14]. In preeclampsia, this process is defective due to abnormal trophoblast invasion, leading to placental ischemia and hypoxia [7,15]. In addition, in a healthy pregnancy, both systolic and diastolic blood flow in the uterine artery are within the normal values while, in a preeclamptic pregnancy, diastolic blood flow is compromised, indicated by a characteristic notch in the waveform that precedes the clinical manifestations of the disease [2].

It is well known that the trophoblast invasion theory has been the most cited and scientifically accepted, however, several data clearly refute this hypothesis. Specifically, trophoblast invasion only occurs in approximately 20% of preeclampsia cases, usually in those related to fetal growth restriction, which is mainly related to early-onset preeclampsia [12]. Furthermore, although this mechanism is thought to promote placental hypoxia—triggering the release of pro-inflammatory cytokines (e.g., IFN-γ, TNF-α, ILs) and anti-angiogenic factors (e.g., sFlt-1, endoglin)—which, together with oxidative stress, contribute to systemic inflammation, endothelial dysfunction, and hypertension [2,7,14,15,16,17], it has not yet been demonstrated in cases of preeclampsia. On the other hand, three different studies have already shown opposite results. When compared to normal pregnancies, in early-onset cases of fetal growth restriction with preeclampsia a hyperoxic placenta and a hypoxic fetus were found, due to reduced oxygen extraction by the placenta, despite the same placental blood flow [18], and through assessment of the oxygenated hemoglobin concentration, there was an increase in the oxygen partial pressure [12,19,20]. Thus, new hypotheses have emerged that take into account three main processes: (1) increased maternal susceptibility to factors released by the placenta, especially in women with previous risk factors; (2) abnormal development of the villous trophoblast, leading to the release of apoptotic/necrotic particles that activate and damage the maternal vascular system; and (3) failure of the extravillous trophoblast to adequately invade the spiral arteries, compromising placental perfusion [12].

In addition to an increased sensitivity to vasoconstrictors (angiotensin II and endothelin 1) [4,21] and an imbalance between thromboxane A2 and prostacyclin [22], adiponectin has emerged as a potential biomarker of preeclampsia [23]. Evidence suggests that altered circulating levels of adiponectin may reflect placental and endothelial dysfunction, highlighting its relevance for both understanding disease mechanisms and potential clinical applications in prediction, diagnosis, and management [8,23].

1.1.2. Risk Factors

Several factors can promote the development of preeclampsia in pregnant women [3,5,6,24]. Previous or chronic hypertension, history of preeclampsia in the pregnant woman or in the family, obesity, and diabetes mellitus, as well as autoimmune diseases and chronic kidney disease, are some of the main risk factors for preeclampsia. In addition, the pregnancy history also influences the development of this pathology, namely being the first pregnancy or multiple pregnancies, an long interval between pregnancies (>5 years), and conditions during gestation (intrauterine growth restriction and premature placental abruption). Maternal age is also presented as a risk factor, however, the exact age is still uncertain. Some authors consider it to be a risk factor when the mother is over 35 years old, while others refer to it as such when she is beyond 40 years old [3,5,6,24]. These are all summarized in

Table 2. Risk factors for developing preeclampsia [3,5,6].

Moderate Risk Factors	High Risk Factors
Nulliparity	Chronic hypertension
BMI > 30 kg/m2, obesity	Chronic kidney disease
Maternal age	Pregestational and gestational diabetes
Medically assisted reproduction	Autoimmune diseases (anti-phospholipid antibody syndrome; systemic lupus erythematosus)
History of stillbirth	Thrombophilia
History of intrauterine growth restriction	Premature placental abruption
Interval between pregnancies > 5 years	Preeclampsia in previous pregnancy
Low schooling	Family history of preeclampsia
Obstructive sleep apnea syndrome	Multiple pregnancy

.

1.2. Adiponectin

Adiponectin is one of several hormones affected by variations in body fat percentage. However, in contrast to the secretion profiles of other adipocytokines produced by adipocytes, obesity and increased fat mass are associated with a reduction in the concentration of this hormone [25,26]. This effect is observed primarily with increased visceral adipose tissue since, when there is an increase in subcutaneous adipose tissue, adiponectin levels rise, as it is positively correlated with the concentration of this glycoprotein [27,28].

Adiponectin circulates in the plasma at high concentrations, constituting approximately 0.01% to 0.05% of total serum proteins [27]. It plays a crucial role in regulating energy homeostasis due to its anti-diabetic, anti-atherogenic, anti-inflammatory, and angiogenic properties, as summarized in Figure 1 [8]. Given its functions, a reduction in adiponectin concentration is expected to be associated with the development of metabolic syndrome, increased cardiovascular risk, and conditions such as hypertension, atherogenic dyslipidemia, hyperinsulinemia, and insulin resistance [29,30]. Additionally, adiponectin might play a role in regulating reproductive function, as it has potential effects on the hypothalamus, ovaries, placenta, endometrium, and even the embryo itself [8,9].

Adiponectin, along with its receptors (AdipoR1 and AdipoR2), is expressed in the cytotrophoblast and syncytiotrophoblast of the placenta. Together with extravillous trophoblasts, these constitute the three main subtypes of trophoblasts found in the human body [23,31]. By acting on the placenta, adiponectin significantly inhibits trophoblast proliferation, playing a regulatory role in processes associated with placentation (Figure 1) [23]. Therefore, certain pathologies that develop during pregnancy may correlate with changes in the concentration of this glycoprotein, such as hypertensive disorders of pregnancy [8,23].

Adiponectin Levels in Healthy Pregnancies

During the first trimester of pregnancy, adiponectin levels in women with normal weight (body mass index—BMI < 25 kg/m²) are similar to those in non-pregnant women, except at 11–14 and 15–18 weeks of gestation, when an increase in the average plasma concentration of adiponectin is observed [23]. Subsequently, there is a slight decrease in its concentration throughout pregnancy, demonstrating an inverse correlation between adiponectin levels and gestational age in healthy, normal-weight pregnant women [32]. This inverse correlation may be explained by increased visceral fat deposition during pregnancy [23,32]. Overweight pregnant women (BMI ≥ 25 kg/m²) exhibit lower mean plasma adiponectin concentrations compared to normal-weight women. However, unlike normal-weight pregnancies, in overweight pregnancies, there is no correlation between adiponectin concentrations and gestational age [32]. In this sense, the aim of this systematic review is to verify the role of adiponectin as a biomarker in pre-eclampsia.

2. Methods

The protocol for this systematic review, with ID number CRD42024542403, was submitted to the registry of protocols and systematic reviews PROSPERO (International Prospective Register of Systematic Reviews).

2.1. Search Strategy

For this systematic review, the following databases were consulted: PubMed and Scopus. The literature search was conducted between March and 30 July 2023. Original articles were identified using the following MeSH keywords: “Preeclampsia” AND “Adiponectin” in PubMed, and “Preeclampsia” [Abstract] AND “Adiponectin” [Abstract] in Scopus.

2.2. Inclusion Criteria

This systematic review includes original research articles published in English or Portuguese since 2013, focusing on human studies involving pregnant women with and without preeclampsia. The studies must quantify adiponectin concentrations in pregnant women. Case–control, prospective and retrospective cohort studies, cross-sectional studies, and bidirectional Mendelian randomization studies are included.

2.3. Exclusion Criteria

Exclusion criteria include inaccessible articles, those published prior to 2013, meta-analyses, literature reviews, animal studies, studies conducted solely on pregnant women without preeclampsia, studies without measurements of adiponectin concentrations, and original articles published in languages other than English or Portuguese.

2.4. Study Selection

This systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 flow diagram [33]. As presented in Figure 2, a total of 231 articles were retrieved from the database search, 133 from PubMed and 98 from Scopus. At this stage, the articles that did not fall within the stipulated time frame or were not in the selected languages and duplicates were identified and excluded. Subsequently, the 74 eligible articles were selected by evaluating the title and abstract. This selection resulted in the exclusion of 43 articles due to irrelevance, inaccessibility, and study designs not within the inclusion criteria. Reviews, systematic reviews, and meta-analyses were also excluded. The full text was then analyzed and the articles included according to eligibility. According to all these criteria, 29 studies were included in this review.

3. Results

Among the 29 included studies, findings regarding adiponectin levels in preeclampsia were heterogeneous. Several studies (Tendean et al. [34], Thagaard et al. [35], de Knegt et al. [36], Bawah et al. [37], Zhang et al. [38]) consistently reported reduced adiponectin levels in women with preeclampsia, establishing a negative correlation between this adipokine and disease occurrence. Additionally, one study by Chandrasekaran et al. [39] showed significantly lower adiponectin levels in mild preeclampsia compared to severe preeclampsia and controls. On the other hand, Tendean and co-workers found a negative correlation, however, in their study the authors only analyzed severe preeclampsia cases and compared with normotensive controls [34]. When analyzing gene expression, Dong et al. [40] found significantly reduced placental mRNA expression of adiponectin in preeclamptic patients compared to controls. In line with these results, Vieira et al. [41] reported a positive correlation between higher adiponectin levels and uncomplicated pregnancies, that is, higher adiponectin concentrations may be protective against preeclampsia and related complications. Lomakova et al. [42] further highlighted ethnic differences, with African American women presenting lower adiponectin levels and a higher incidence of preeclampsia.

Conversely, three studies by the same research group [43,44,45] observed increased adiponectin levels in preeclamptic pregnancies, similar to the findings by Abraham et al. [46], where adiponectin concentrations were higher in preeclamptic women compared to controls and those with gestational hypertension. Similarly, Noureldeen [47] reported increased adiponectin levels in obese women (BMI ≥ 30 kg/m²) with preeclampsia. Salimi et al. [48] and Weedon-Fekjaer et al. [49] found increased maternal serum adiponectin in both early- and late-onset preeclampsia, although no differences were observed between mild and severe cases. Zhou et al. [50] showed higher adiponectin levels in umbilical cord blood from severe preeclampsia cases, while Khosrowbeygi and Ahmadvand [51] also reported increases in both mild and severe preeclampsia. However, in a different study, the same authors [52] found decreased adiponectin levels after adjusting for BMI, suggesting that the relationship may be partly independent of adiposity.

Other investigations, including Song et al. [53], Chen et al. [54], Tobinaga et al. [55], Güngör et al. [56], Cetinkaya Demir et al. [57], Chandrasekaran et al. [58], Eleuterio et al. [59], and Nevalainen et al. [60], reported no significant differences in adiponectin levels between groups. Similarly, Martinez-Fierro et al. [61] found no relevant expression of adiponectin in peripheral blood mononuclear cells, failing to establish a relationship between adiponectin and preeclampsia. Rao et al. [62] also observed no significant overall difference, although stratification by severity revealed higher adiponectin concentrations in severe compared with mild preeclampsia.

Importantly, of the studies that included BMI in the analysis, most showed no significant differences in adiponectin levels between pregnant women with and without preeclampsia ([34,36,44,48,49,51,53,55,56,58,62]). When adjusted for BMI, Khosrowbeygi [52] and Lomakova [42] reported that adiponectin remained independently associated with preeclampsia, indicating that adiposity alone does not fully explain the observed associations.

Regarding biological sampling, most studies analyzed venous blood; however, timing varied considerably, with some collecting samples during the first, second, or third trimesters and others only after preeclampsia diagnosis. A few studies assessed placental tissue (Dong et al. [40], Weedon-Fekjaer et al. [49]) or umbilical cord blood (Zhou et al. [50]). These methodological differences may partly account for the conflicting findings across studies.

The results from the 30 selected studies are all summarized in

Table 3. Characteristics of the studies included in the systematic review.

Authors and Year	Participants	Markers Analyzed	BMI Handling	Main Results
Tendean et al. [34] 2021	52 participants: - 26 pregnant women with severe preeclampsia - 26 pregnant women without preeclampsia	- Adiponectin	BMI reported but not adjusted	- Adiponectin levels are correlated with the development of severe preeclampsia, since pregnant women with preeclampsia showed significantly lower levels of adiponectin when compared to normotensive pregnant women. - No relationship between BMI and the incidence of preeclampsia.
Thagaard et al. [35] 2019	2503 participants stratified according to BMI (normal, moderate or severe obesity): - 93 pregnant women with hypertensive disorders of pregnancy (29 with hypertension and 64 with preeclampsia).	- Adiponectin - Leptin	Matched and adjusted for BMI	- Obese women have a lower concentration of adiponectin in preeclampsia, with a lower concentration in severe obesity. No association was found in normal-weight women. Leptin concentration had no association with the disease in normal-weight and moderately obese women; however, in women with severe obesity, a lower level of leptin was found - The adiponectin/leptin ratio was not associated with the development of preeclampsia in any of the groups.
de Knegt et al. [36] 2023	423 participants: - 126 pregnant women with preeclampsia (98- mild preeclampsia, 21- severe preeclampsia, 7- HELLP syndrome) - 297 pregnant women without preeclampsia	- Adiponectin - Leptin - Adiponectin/leptin ratio as a surrogate marker of insulin sensitivity - PAPP-A - β-hCG	Adjusted for BMI	- The adiponectin/leptin ratio was significantly lower in pregnancies with preeclampsia compared to controls. - Adiponectin and PAPP-A were negatively associated with preeclampsia, while leptin was positively associated, however the adiponectin/leptin ratio was a better predictor of the disease, although not clinically relevant as a single marker. - There was no association between the adiponectin/leptin ratio and clinical severity or time of onset of preeclampsia. - β-hCG concentrations were not significantly different between the two groups.
Bawah et al. [37] 2020	190 participants: - 90 pregnant women with preeclampsia - 100 pregnant women without preeclampsia	- Adiponectin - Leptin - Resistin - Visfatin - Lipids	Adjusted for BMI	- There was no significant difference in the lipid profile, with the exception of HDL-C, which was significantly lower in the preeclampsia group. BMI was significantly higher in pregnant women who developed preeclampsia. - Leptin, resistin and visfatin levels were significantly higher in pregnant women who developed preeclampsia. Adiponectin showed significantly lower levels in pregnant women who developed preeclampsia. - Resistin was considered the best predictor of the disease after controlling for BMI. However, adiponectin was considered the best predictor after controlling for BMI, age, parity and family history of diabetes and preeclampsia.
Zhang et al. [38] 2023	208 participants: - 118 pregnant women with severe preeclampsia - 90 pregnant women without preeclampsia	- Adiponectin	Restricted to normal BMI range (18.5–24.9 kg/m2)	- Adiponectin levels in patients with preeclampsia were significantly lower than in the healthy control group. - In the ultrasound evaluation of the umbilical artery, the pulsatility and resistance indices increased in the preeclampsia group compared to the control group. Furthermore, when correlated with serum adiponectin levels, it was shown that adiponectin levels were negatively related to pulsatility and resistance indices.
Chandrasekaran et al. [39] 2020	64 participants: - 16 pregnant women with non-severe preeclampsia - 30 pregnant women with severe preeclampsia - 18 pregnant women without preeclampsia	- Adiponectin - Leptin - Resistin	BMI reported but not adjusted or matched	- Plasma concentrations of adiponectin were significantly lower in patients with preeclampsia without severity characteristics compared to the other two groups (on average 1 to 2 years after delivery). Leptin and resistin concentrations did not differ between the groups. - After separating the participants according to body fat percentage, in the high-fat group (≥38%), leptin remained positively associated with visceral and subcutaneous fat areas, adiponectin correlated negatively with visceral fat area and resistin was positively associated with subcutaneous fat area. Among low-fat women (<38%), leptin was positively associated with the subcutaneous fat area, adiponectin showed a negative relationship for both areas, while resistin was not significantly associated with either fat compartment.
Dong et al. [40] 2018	82 participants: - 52 pregnant women with preeclampsia - 30 pregnant women without preeclampsia	- Adiponectin - Pathway P38 MAPK-STAT5	Not reported	- p-38 with high expression and p-STAT5 with lower expression in placental trophoblasts of pregnant women with preeclampsia, when compared to pregnant women without the condition. - Adiponectin mRNA expression was found in both groups, but the level of expression was significantly lower in patients with preeclampsia. - The level of p-p38 expression was negatively correlated and the level of p-STAT5 expression was positively correlated with adiponectin expression levels.
Vieira et al. [41] 2017	1409 participants with a BMI ≥ 30 kg/m2: - 904 complicated pregnancies - 505 uncomplicated pregnancies	- Adiponectin; leptin - Haemoglobin A1c (HbA1c); fructosamine; insulin and C-peptide - IL-6; high-sensitivity C-reactive protein; t-PA antigen - Lipids (triglycerides (TG), total cholesterol, LDL-C (low-density lipoprotein cholesterol), HDL-C (high-density lipoprotein cholesterol)) - Liver-associated markers (aspartate aminotransferase, alanine aminotransferase, gamma-glutamyl transferase (γGT), SHBG, and ferritin) - Vitamin D	Restricted to obese women (BMI ≥ 30 kg/m2)	- Higher concentrations of adiponectin were associated with a greater likelihood of uncomplicated pregnancy and labour. - Higher concentrations of HbA1c, insulin, SHBG and γGT were associated with a lower probability of pregnancy and uncomplicated labour.
Lomakova et al. [42] 2022	1776 participants: Pregnant, young and generally healthy - n = 646 African American women - n = 853 Hispanic women - n = 277 Caucasians	- Adiponectin - IL-6, IL-8, IL-10, TNF-α and GMCSF (granulocyte-macrophage colony-stimulating factor) - Resistin	Adjusted for BMI	- African American women had lower levels of adiponectin and higher levels of resistin and GMCSF compared to Hispanic or Caucasian women. Although African American pregnant women had higher levels of IL-8 and TNF-α, the trends for all cytokines (IL-8, TNF-α, IL-6, IL-10) were not statistically significant. - A higher number of African American women developed preeclampsia compared to the other ethnic groups. However, fewer African American women developed gestational diabetes. - Women with lower levels of adiponectin had a higher risk of developing preeclampsia when analysed with and without adjustment for pre-pregnancy BMI.
Eleuterio et al. [43] 2013	117 participants: - 47 pregnant women with preeclampsia - 70 pregnant women without preeclampsia Exclusion of pregnant women with BMI ≥ 30 kg/m2	- Adiponectin - Nitrite - Endogenous NO inhibitor - Asymmetric dimethylarginine (ADMA)	Restricted to non-obese women (BMI ≤ 30 kg/m2)	- Increased concentration of adiponectin in preeclampsia. - Decreased nitrite concentration in preeclampsia. - Negative correlation between adiponectin and BMI in a healthy pregnancy; no correlation in preeclampsia. - Positive correlation between adiponectin and nitrite in a healthy pregnancy; no correlation in preeclampsia. - Strong positive correlation between adiponectin and nitrite levels when ADMA levels are reduced in pregnant women with preeclampsia, suggesting that high concentrations of this inhibitor may interfere with the physiological activation of endothelial nitric oxide synthase by adiponectin in preeclampsia.
Eleuterio et al. [44] 2014	63 participants: - 27 pregnant women with preeclampsia - 36 pregnant women without preeclampsia Exclusion of pregnant women with BMI ≥ 30 kg/m2	- Adiponectin - Leptin - sFlt-1 - sEng	Restricted to non-obese women (BMI ≤ 30 kg/m2)	- Increased concentrations of adiponectin, leptin, sFlt-1 and sEng in pregnant women with preeclampsia compared to pregnant women without the condition. - Strong positive correlation between adiponectin and sFlt-1 and sEng in preeclampsia, but not in healthy pregnancies. - Significant positive correlation between leptin and sFlt-1 and sEng in preeclampsia, but not in healthy pregnancies. - Negative correlation between adiponectin and BMI and positive correlation between leptin and BMI in the group of pregnant women without preeclampsia. However, both correlations were not found in the group of pregnant women with preeclampsia. - Negative correlation between sFlt-1 and BMI in pregnant women with preeclampsia.
Eleuterio et al. [45] 2015	105 participants: - 59 pregnant women with preeclampsia - 46 pregnant women without preeclampsia Exclusion of pregnant women with BMI ≥ 30 kg/m2	- Adiponectin - Leptin - Matrix metalloproteinase 2 and 9 (MMP 2 and 9) and tissue inhibitor of metalloproteinase 1 and 2 (TIMP 1 and 2)	Restricted to non-obese women (BMI ≤ 30 kg/m2)	- Levels of adiponectin, leptin, MMP2, TIMP1 and TIMP2 increased in pregnant women with preeclampsia. - There was a positive correlation between adiponectin and MMP2 and between adiponectin and TIMP2 in the group of pregnant women with preeclampsia.
Abraham et al. [46] 2020	248 participants: - 66 pregnant women with early preeclampsia - 68 pregnant women with late preeclampsia - 55 pregnant women with gestational hypertension - 59 healthy pregnant women	- Adiponectin - Leptin - Resistin - Redox markers malondialdehyde (MDA) and total antioxidant status (TAS)	Not reported	- Adiponectin levels were significantly higher in the preeclampsia group compared to the control and gestational hypertension groups. - Adiponectin levels in the gestational hypertension group were comparable to the control group. - Increased serum adiponectin, MDA and TAS were associated with adverse maternal outcomes in hypertensive diseases of pregnancy. - Leptin levels did not differ significantly between the groups. The leptin/adiponectin ratio and resistin levels were not significantly different between the groups. The correlation between MDA, leptin and adiponectin was not statistically significant.
Noureldeen [47] 2014	62 participants: 25 pregnant women with preeclampsia - 9 with BMI < 30 kg/m2 - 16 with BMI ≥ 30 kg/m2 37 normotensive pregnant women - 15 with BMI < 30 kg/m2 - 22 with BMI ≥ 30 kg/m2	- Leptin - Adiponectin - Resistin - Visfatin - TNF-α - IL-6	Matched by BMI	- Changes in adipocytokines were more pronounced in obese pregnant women with the disease than in normal-weight pregnant women complicated by preeclampsia. - When pregnant women with a BMI ≥ 30 kg/m2 the adipocytokines that showed significant changes included: - Leptin: decreased concentration in pregnant women with preeclampsia - Adiponectin: increased concentration in pregnant women with preeclampsia - Resistin: increased concentration in pregnant women with preeclampsia - TNF-α: increased concentration in pregnant women with preeclampsia
Salimi et al. [48] 2014	90 participants: - 45 pregnant women with preeclampsia - 45 pregnant women without preeclampsia	- Adiponectin - Leptin	Matched by BMI	- Maternal serum leptin and adiponectin showed significantly higher levels in pregnant women with preeclampsia. - In contrast to leptin, serum adiponectin did not differ between women with early onset and late onset preeclampsia. - Serum adiponectin was significantly higher in severe preeclampsia compared to pregnant women without preeclampsia. However, the difference between adiponectin levels in severe and mild preeclampsia was not significant. - A significant positive correlation was observed in the control group between: leptin and adiponectin; leptin and BMI. - A significant positive correlation was observed in the control group and in patients with preeclampsia between: adiponectin and BMI. - There were no significant differences in the leptin/adiponectin ratio between pregnant women with and without preeclampsia.
Weedon- Fekjaer et al. [49] 2014	72 participants: - 23 pregnant women with early onset preeclampsia - 26 pregnant women with late onset preeclampsia - 23 pregnant women without preeclampsia	- Adiponectin - Leptin - Resistin	Adjusted for BMI	- The concentration of adiponectin was significantly higher in the maternal circulation in both early and late onset preeclampsia compared to the control group. Placental expression of the adiponectin gene showed no significant difference between the groups. - Circulating leptin levels were significantly higher in both preeclampsia groups compared to the control group. Placental expression of the leptin gene was significantly higher in the early onset preeclampsia group compared to the late onset group. - There were no significant differences between the groups in terms of the plasma concentration of resistin and the expression of its gene in the placenta.
Zhou et al. [50] 2020	120 participants: - 60 with severe preeclampsia - 60 without preeclampsia	- Aquaporin 3 (APQ3) - Adiponectin - TG - LDL-C - HDL-C - Apolipoprotein (Apo) A and Apo B	Not reported	- APQ3 expression was lower in placental tissue and higher in foetal membranes in the case group compared to the control group. - Adiponectin levels in the umbilical cord blood of newborns in the case group were higher than in the control group. - TG and C- LDL levels were significantly higher in the case group. However, no significant differences were found for other parameters, including cholesterol, C- HDL, ApoA, ApoB and ApoB/ApoA.
Khosrowbeygi and Ahmadvand [51] 2013	60 participants: - 30 pregnant women with preeclampsia - 30 pregnant women without preeclampsia	- Adiponectin - Homocysteine	Matched by BMI	- When compared to pregnant women without preeclampsia: - Increased concentrations of total adiponectin and total homocysteine in mild and severe preeclampsia. Increased homocysteine-adiponectin ratio in severe preeclampsia. - No significant differences between mild and severe preeclampsia.
Khosrowbeygi and Ahmadvand [52] 2013	60 participants: - 30 pregnant women with preeclampsia - 30 pregnant women without preeclampsia	- Adiponectin - Leptin	Matched and adjusted for BMI	- Significant increase in adiponectin and leptin levels in pregnant women with mild and severe preeclampsia, compared to pregnant women without preeclampsia, before adjusting for BMI. No significant difference was found between mild and severe preeclampsia. - Significant increase in the leptin/adiponectin ratio in pregnant women with severe preeclampsia, compared to pregnant women without preeclampsia and with mild preeclampsia, before adjusting for BMI. No change when comparing pregnant women without preeclampsia and those with mild preeclampsia. - After adjustment for BMI: - Leptin values were slightly higher in the preeclampsia group - Adiponectin values were significantly lower in the preeclampsia group - The adjusted leptin/adiponectin ratio was significantly higher in the preeclampsia group
Song et al. [53] 2016	153 participants: - 74 pregnant women with preeclampsia - 79 pregnant women without preeclampsia	- Leptin - Adiponectin - Resistin	Adjusted for BMI	- There were no significant differences in serum adiponectin levels between the groups, even when stratifying between mild and severe preeclampsia. - BMI, serum leptin and resistin levels and the resistin/creatinine ratio were significantly higher in the preeclampsia group. - No correlation between serum adiponectin levels and BMI. No correlation between serum resistin levels and BMI. - Positive correlation between serum leptin levels and BMI, only in pregnant women without preeclampsia.
Chen et al. [54] 2022	141,068 participants of European descent: - 4743 cases of preeclampsia - 136,325 in the control group	- Adiponectin - Leptin - Resistin -sOB-R (soluble leptin receptor) - PAI-1	Adjusted for BMI	- No correlation between levels of adipokines, particularly adiponectin, and preeclampsia. - PAI-1 could be a useful biomarker in the diagnosis and monitoring of preeclampsia therapy.
Tobinaga et al. [55] 2014	108 participants: - 54 pregnant women with preeclampsia - 54 pregnant women without pr-eclampsia	- Soluble fms-like tyrosine kinase 1 (sFlt-1) - Soluble Endoglin (sEng) - Adiponectin - Plasminogen activator inhibitor-1 (PAI-1)	Adjusted for BMI	- Adiponectin levels were similar in both groups, regardless of BMI and unrelated to uterine artery resistance. - PAI-1 levels were significantly higher in preeclampsia, but unrelated to uterine artery resistance. - Mean serum levels of sFlt-1, sEng and PAI-1 were significantly higher in patients with preeclampsia. - Serum levels of sFlt-1 and sEng were significantly higher in women with early onset preeclampsia compared to late onset preeclampsia, while concentrations of adiponectin and PAI-1 were similar between these subgroups. - Patients with preeclampsia had significantly higher mean uterine artery resistance. - Patients with preeclampsia and abnormal uterine artery doppler had higher serum sEng levels than those with normal doppler.
Güngör et al. [56] 2017	79 participants: - 30 pregnant women with early onset preeclampsia - 22 pregnant women with late onset preeclampsia - 27 pregnant women without preeclampsia Considering the division between early and late at 32 weeks’ gestation	- Plasma angiogenic factors (PlGF, VEGF) - Plasma anti-angiogenic factors (sFlt-1, endoglin) - Leptin, adiponectin and ghrelin - Endothelial dysfunction markers (vWF, NO) - Platelet function markers (ADP and collagen-induced platelet aggregation, P-selectin)	Matched by BMI	- Endoglin, leptin and vWF levels are increased in preeclampsia. - Levels of PIGF, collagen-induced platelet aggregation and P-selectin are decreased in preeclampsia. - No significant differences were found in adiponectin and ghrelin levels when comparing both groups.
Demir et al. [57] 2013	80 participants: - 52 pregnant women with preeclampsia - 28 pregnant women without preeclampsia	- Adiponectin - Visfatin	Not reported	- No significant changes in plasma concentrations of adiponectin and visfatin in pregnant women with preeclampsia compared to healthy pregnant women. - Preeclampsia severity was not shown to affect plasma values of adiponectin and visfatin.
Chandrasekaran et al. [58] 2020	117 participants stratified into subgroups: - Women with obesity - Normal weight women - 61 pregnant women with preeclampsia (n = 36 obese; n = 25 normal weight) - 56 pregnant women without preeclampsia (n = 29 obese; n = 27 normal weight)	- Visfatin and resistin (related to visceral fat) - Leptin and adiponectin (related to general adiposity) - Inflammatory cytokines: IFN-γ, IL-1β, IL-6, IL-2	Adjusted for BMI	- Association between preeclampsia and significantly higher maternal concentrations of visfatin, resistin and inflammatory cytokines. - Preeclampsia was associated with increased levels of leptin, but not adiponectin. - There were no interactions between preeclampsia and obesity, suggesting that the increase in resistin, visfatin, inflammatory cytokines and leptin present in the condition did not differ according to the degree of obesity. - The concentrations of visfatin, resistin, adiponectin and leptin did not differ when pregnant women with preeclampsia were stratified according to degree of severity. Inflammatory cytokines, on the other hand, showed significantly higher concentrations in pregnant women with characteristics of disease severity.
Eleuterio et al. [59] 2016	108 participants: - 31 pregnant women with preeclampsia - 27 pregnant women with gestational hypertension - 50 healthy pregnant women with uncomplicated pregnancies	- Adiponectin - Leptin - Oxidative stress markers: malondialdehyde-thiobarbituric acid reactive substances (MDA-TBARS) - Plasma antioxidant activity [Ferric Reducing Ability of Plasma (FRAP)]	Not reported	- Plasma levels of adiponectin and leptin were similar in all the study groups. - FRAP showed gradual increases between the three groups, with a significant difference between the gestational hypertension and preeclampsia groups, the latter being higher. - MDA-TBARS plasma levels were similar between the groups. - There was a significant negative correlation between MDA-TBARS and adiponectin, suggesting a relationship between antioxidant levels and the glycoprotein in healthy pregnancies, which is altered in patients with gestational hypertension or preeclampsia.
Nevalainen et al. [60] 2017	864 participants distributed into training and test groups: - 71 pregnant women with early onset preeclampsia - 793 pregnant women without preeclampsia Training groups: - Controls (n = 652) - Early onset preeclampsia (n = 29) Test groups: - Controls (n = 141) - Early onset preeclampsia (n = 42)	- Alpha-fetoprotein (AFP); Placental growth factor (PIGF) - Soluble tumour necrosis factor receptor 1 (sTNFR1); Retinol binding protein 4 (RBP4); Disintegrin and metalloprotease-12 (ADAM12) - Soluble P-selectin - Follistatin like-3 (FSTL3) - Adiponectin; Angiopoietin-2 - Sex hormone binding globulin (SHBG) - PAPP-A (Pregnancy-associated plasma protein A); β-hCG (human chorionic gonadotrophic hormone)	Not reported	- The best individual maternal serum biomarkers in the first trimester of pregnancy for predicting the development of early preeclampsia were: AFP, PIGF, RBP4 and sTNFR1. - The best screening combination in the test set was achieved by adding sTNFR1, mean arterial pressure and AFP, PlGF or RBP4 to the combined first trimester screening information of maternal characteristics, PAPP-A and β-hCG. - When assessing adiponectin, SHBG and FSTL3 levels, the results were contradictory: adiponectin and SHBG values increased in the test set and decreased in the training sample set; FSTL3 values decreased in the test set and increased in the training sample set.
Martinez- Fierro et al. [61] 2014	177 participants: - 108 pregnant women with preeclampsia - 60 with mild preeclampsia - 48 with severe preeclampsia - 69 normotensive pregnant women	Gene expression: -Hemeoxygenase 1 - Superoxide dismutase - Vascular endothelial growth factor A - Transforming growth factor beta 1 (TGF-β1) - IL-6, IL-15 - Adiponectin	Not reported	- The onset and severity of preeclampsia was reflected in the imbalance in the expression of the Vascular Endothelial Growth Factor A and TGF-β1 genes, with their under-expression being a constant in most cases. - IL-6, IL-15 and adiponectin showed low or no expression in the peripheral blood mononuclear cell samples evaluated. Therefore, the association between IL6, IL-15 and adiponectin expression and preeclampsia could not be assessed.
Rao et al. [62] 2021	120 participants: - 60 pregnant women with preeclampsia - 60 pregnant women without preeclampsia	- Adiponectin - Leptin	BMI reported but not adjusted	- Adiponectin levels were higher in the study group than in the control group. However, the difference was not statistically significant. - Leptin levels in the study group were significantly higher than in the control group. - The adiponectin/leptin ratio was significantly lower in the study group. However, it was significantly higher in pregnant women with severe preeclampsia compared to pregnant women with mild preeclampsia. - Severe preeclampsia is associated with significantly higher adiponectin levels compared to mild preeclampsia.

.

Analysis of Evidence Levels in Included Studies

Based on the analysis of the studies included in this systematic review, a table of the levels of evidence was made according to the Oxford Centre for Evidence-Based Medicine (CEBM) system [63]. This is a pyramid-shaped hierarchy that ranks the strength of evidence according to the type of study [63]. Five levels of evidence are defined according to the study design: level 1 corresponds to randomized clinical trials; level 2 to prospective cohort studies and bidirectional Mendelian randomization studies; level 3 to retrospective cohort and case–control studies; level 4 to cross-sectional studies and case series; and level 5 to experimental and mechanism-based reasoning studies. Accordingly, the studies were classified as: high evidence for levels 1 and 2 (highlighted in green), moderate evidence for level 3 (highlighted in yellow), and low evidence for levels 4 and 5 (highlighted in orange). The numbers in

Table 4. Level of evidence of the relationship between adiponectin levels and the development of preeclampsia taking into consideration the BMI. Version adapted from OCEBM and Marx et al. [64,65].

	BMI Considered
Adiponectin in PE	Inclusion of Only Obese Pregnant Women			Exclusion of Underweight and/or Overweight and/or Obese Pregnant Women			No BMI Restriction in the Exclusion Criteria			No BMI Restriction in the Exclusion Criteria, but Result Observed Only in Obese Pregnant Women
Increase					1	2		3	2			1
No change				1	1		1	3	2
Decreased	1				1		2	3	3		1
It was not possible to determine								1

Legend: green: High; yellow: Moderate; orange: low.

represent the number of studies categorized under each corresponding result [63,64].

The levels of evidence were analyzed for the difference in adiponectin levels in pregnant women with preeclampsia, taking into account BMI, as illustrated in Table 4. In detail, adiponectin levels in cases of preeclampsia were classified as increased, decreased, or unchanged compared to controls and were then related to BMI, divided into four types of studies: (1) those that only included obese women, (2) those that excluded underweight and/or overweight and/or obese pregnant women, (3) those with no BMI restriction in the exclusion criteria, and (4) those with no BMI restriction in the exclusion criteria, but results occurred only in obese pregnant women.

The interpretation of Table 4 shows that, in general, the majority of studies are concentrated in the low and moderate levels of evidence categories and that there is a high degree of heterogeneity in the results obtained. This means that there are few randomized clinical trials and prospective cohort studies performed on this topic, hence the lower levels of evidence. Specifically, Table 4 shows that a greater number of studies report a decrease in adiponectin levels in women with preeclampsia, as do the levels of evidence, with high-level-evidence studies. Furthermore, most of the studies analyzed focus on those with no BMI restriction in the exclusion criteria, that means that the BMI was not used as an exclusion criterion in the sample selection. Although including this factor increases the population’s heterogeneity, it may also introduce a confounding factor, since BMI is strongly associated with adipokine levels, such as adiponectin, and the risk of preeclampsia.

4. Discussion

Of the studies analyzed, most suggest a decrease in adiponectin levels in pregnant women with preeclampsia. However, the results are very heterogeneous, since of the thirty studies, eleven reported a decrease in adiponectin, nine indicated an increase, eight detected no significant changes, and in two studies it was not possible to perform quantification.

Overall, variations in reported adiponectin levels across the studies analyzed may result from a combination of biological factors. Specifically, maternal metabolic status, including obesity, insulin resistance, and inflammatory profiles, can influence circulating levels of adiponectin, while hormonal fluctuations across the trimesters may further modulate its levels. Nonetheless, these divergent results are in accordance with previous studies. Daskalaskis et al. conducted a systematic review aiming to analyze the role of serum levels of various adipocytokines in preeclampsia, including adiponectin. The conclusions indicate an association between preeclampsia and increased levels of leptin, chemerin, and fatty acid-binding protein 4. However, data on the remaining adipocytokines remain inconsistent [65]. Furthermore, considering the influence of BMI and adiponectin levels, a different systematic review carried out by Salihu et al. aimed to establish the relationship between obesity and preeclampsia. In this study, obesity was considered an exposure variable, and the findings consistently demonstrated an association with preeclampsia. However, potentially relevant biomarkers were identified, such as leptin, adiponectin, matrix metalloproteinases, C-reactive protein, and sex-hormone-binding globulin. Nonetheless, establishing a causal relationship was not possible [66]. In a different study, focusing on biochemical and molecular predictive tests for preeclampsia restricted to the first trimester of pregnancy, several biomarkers were assessed, including uterine artery pulsatility index, pregnancy-associated plasma protein A, human chorionic gonadotropin, inhibin-A, and adiponectin, the focus of the present systematic review [67]. According to Abdi et al., there is inconsistency in adiponectin levels. Even so, despite focusing solely on first-trimester samples, their review corroborated the findings of the present study, thus suggesting that, given the complexity of preeclampsia involving multiple biological pathways and various risk factors, integrating multiple markers is essential for predicting this condition [67].

The conduction of this systematic review adhered to rigorous methodological criteria to minimize bias and ensure replicability. The inclusion of populations from diverse geographical regions (notably Indonesia, Egypt, Denmark, Brazil, China, India, Africa, Iran, Turkey, the United States, Norway, Finland, Mexico, and England) provides a comprehensive perspective on adiponectin levels in varying contexts. However, sociodemographic variables such as ethnicity, lifestyle, and access to healthcare may influence adiponectin levels and the development of preeclampsia, contributing to the heterogeneity observed in the interpretation of the results.

In addition, the number of participants varied significantly across the studies considered in this review, ranging from 52 to 141,068 participants. This variability allows for specific insights from studies with smaller sample sizes and broader conclusions from those with larger populations. However, it may also contribute to the discrepancies observed in the results. While the diversity of study designs included in the sample allows for a comprehensive and multidimensional analysis, differences in adiponectin measurement methodologies, study designs, and inclusion/exclusion criteria, particularly regarding the BMI of pregnant women, contribute to the heterogeneity observed when comparing different studies. Additionally, adiponectin is a glycoprotein with several isoforms, and the relative expression of these isoforms may vary, further influencing the divergence in results.

The timing of sample collection and the types of samples analyzed represent another limitation of this systematic review. Samples were collected at various time points, including the first, second, and/or third trimesters, as well as during delivery and/or postpartum, while other studies did not specify the timing of sample collection. Furthermore, while most studies analyzed blood samples, others included placental tissue samples collected during delivery, and one study relied on genetic data analysis. Thus, discrepancies in the results obtained may also arise from the gestational timing of sample collection, the type of sample analyzed (serum, plasma, or placental tissue), and the assay employed (e.g., ELISA vs. radioimmunoassay).

Another limiting factor is the quality of evidence in the studies included in this systematic review. The results on the levels of evidence indicate that most studies were rated as providing moderate- or low-quality evidence and that there was some variability in BMI within the inclusion and exclusion criteria. Both the quality of evidence and the heterogeneity in BMI across study populations are critical factors to consider when interpreting the findings. Few studies exclusively included pregnant women with obesity, while some explicitly excluded them. Additionally, there were fewer studies classified as high-quality evidence compared to other categories.

Altogether, these factors may explain why some studies report heterogenous results, with some showing increases, decreases, or no changes in adiponectin levels in women with preeclampsia, highlighting the need for careful interpretation and standardization across research designs.

5. Conclusions and Future Perspectives

The abnormal trophoblast invasion during placentation is considered the most probable cause of the pathophysiology of preeclampsia. And, given the role of adiponectin in regulating placental processes, there is a possibility that adiponectin could be used as a biomarker for preeclampsia. This systematic review was performed to answer this question. However, inconsistent results were found throughout the studies analyzed, where a positive, negative, or no association was found between levels of adiponectin and the development of preeclampsia. In addition, the levels of evidence showed stronger evidence for decreased adiponectin levels, mainly in studies where the BMI was not used as an exclusion criterion in the sample selection, which may introduce a confounding effect. Thus, to enhance accuracy and ensure comparability of the results, it is crucial that future studies initially focus on pregnant women with a normal BMI (between 18.5 kg/m² and 24.9 kg/m²), thereby establishing a reliable baseline. This focus is essential because, beyond the variation in adiponectin level associated with pregnancy, it also varies with BMI. The addition of overweight and obese pregnant women can introduce numerous confounding variables, complicating the analysis of the results.

Regarding the types of samples collected, standardizing maternal blood sampling is recommended, as it is a practical method used in most of the analyzed studies and facilitates the potential future introduction of a screening test. Additionally, the timing of sample collection should ideally occur during the first trimester or at least before the 20th week of gestation, as this is the period before preeclampsia typically develops. Early identification would significantly improve outcomes. Additionally, early detection of preeclampsia is critical, however, no current screening methods effectively promote early detection of this condition. This systematic review concludes that further research is essential to understand and assess adiponectin’s potential as a biomarker for preeclampsia. Future studies should consider combining adiponectin with other biomarkers, as using it in isolation has shown considerable variability in results.