Highlighting the Mechanistic Relationship Between Perinatal Depression and Preeclampsia: A Scoping Review

Background: Although there is scientific literature supporting an association between depression and preeclampsia (PE), little is known about the underlying mechanistic pathways that may explain these observed associations. Thus, this study aimed to outline the relationship between depression and PE, and to highlight the underlying cardiovascular and metabolic risk factors that are common to both. Methods: A scoping review of the literature was conducted in Medline, Scopus, and Web of Science. Results: From 706 articles initially identified, 23 articles met the inclusion criteria and were included in this review. Although some studies reported a positive association between PE and postpartum depressive symptoms, challenges comparing different methodologies, measurement instruments and when measurements were administered, and patient populations do not permit a decisive conclusion. In addition, very few studies addressed potential underlying mechanisms that may be contributing to observed associations; thus, a secondary search was conducted to identify cardiovascular and metabolic risk factors that are common to both depression and PE. Conclusion: The cardiovascular and metabolic risk factors (i.e., increased inflammation and oxidative stress and decreased vascular and endothelial function) common to both depression and PE suggest that these factors may contribute as underlying mechanisms in both conditions. These similarities underscore the importance to better understand these mechanisms so preventative and therapeutic strategies could be developed to improve maternal health.


Background
Hypertensive disorders of pregnancy are a broad spectrum of conditions ranging from chronic hypertension to preeclampsia (PE) and eclampsia. In 2018, *7% of all pregnancies in Canada were complicated by hypertensive disorders of pregnancy. 1 PE is defined as having hypertension (systolic blood pressure ‡140 mmHg and/or diastolic blood pressure ‡90 mmHg), certain end-organ dysfunction, and with or without newonset proteinuria. 1,2 PE can lead to health consequences such as long-term end-organ dysfunction, fetal developmental issues, and even death for both mother and fetus. 2,3 Known risk factors for PE include age, nulliparity, chronic hypertension, body mass index (BMI) >30 kg/m 2 , and preexisting diabetes. 4 Of concern is evidence from epidemiologic studies demonstrating the increased prevalence of these risk factors in women of childbearing age. 5,6 In Ontario, between 2012 and 2016, 17.8% of mothers with live singleton or twin births had a prepregnancy BMI >30 kg/m 2 , and 1.1% had established chronic hypertension. 7 Average maternal age has also steadily risen since the mid-1960s, with the average maternal age at first pregnancy now at 30.8 years. 8 Additional factors presenting challenges to pregnant women are mental health concerns, particularly depression and depressive symptoms. In Canada, *5% of women have major depressive disorder, 9 in Ontario, 7.7% of mothers were affected with depression during pregnancy, 10 8.6% of women experience postpartum depressive symptoms, 11 and 54.2% of women who experienced postpartum depression had depression previously. 10,12 Increasing attention is being paid to the interrelationships between preexisting mental health concerns, particularly depression, obesity, and PE. For example, one study reported that, in pregnant women, a one unit increase in prepregnancy BMI was associated with *3% higher odds of antenatal depression. 13 These findings suggest an interplay between physical and mental health, or risk factors that are similar for both entities.
Despite known associations between depression and PE, much less is understood about the possible mechanistic pathways connecting these two conditions. Thus, the first objective for this study was to conduct a scoping review to summarize research findings regarding the association between depression and PE. The secondary objective was to extract information from previously reported studies, particularly related to underlying cardiovascular and metabolic risk factors, and integrate this information with the goal to provide evidence for and insights into the possible mechanistic pathways connecting depression and PE.

Methods
Identifying the research question This scoping review was conducted using the methodological framework described by Arksey and O'Malley. 14 The primary research question for this scoping review was, ''what is the current evidence for an association between depression and preeclampsia''? As most of the screening tools used to assess depression during pregnancy within research are solely to identify those in need of further clinical assessment, the distinction between depressive symptoms and clinical diagnosis of depression is not often made, thus for this scoping review, the term ''depression'' includes the entire spectrum of the disease. The secondary research question for this scoping review was, ''what information has been reported that provides insight into pathophysiologic mechanisms that might link depression and preeclampsia''?
Identifying relevant studies This scoping review identified, retrieved, and evaluated information from peer-reviewed scientific articles that examined the association between depression and PE in pregnant women. The focus was on studies published between 2000 and 2020, as it reflected the most recently published literature. Two authors performed the search using a combination of search terms related to our target group, depression, and PE (Table 1) within the following international databases: Medline (2000-April 2020), Scopus (2000-2019), and Web of Science (2000-2020). The search strategies were limited to scientific journal articles and English language only. Selecting the literature Inclusion and exclusion criteria. Inclusion criteria for studies included in this scoping review were as follows: (1) an examination of the associations between depression and PE during pregnancy and/or postpartum, and (2) measurement of maternal depression and PE during pregnancy and/or postpartum. Studies were excluded if: (1) depression was pooled with other variables, such as anxiety, stress, and mood disorder, or the PE measurement was pooled with perinatal outcomes, and (2) if the study did not separately assess depression without antidepressant medication. The exclusion of these articles allowed for a more focused examination of the relationships between depression and PE, as covariates such as antidepressant medication have been reported to affect either depression or PE. The specific inclusion and exclusion criteria are given in Table 2.
As we included different types of studies in this scoping review (primary studies and review studies), and there is not a single approach to compare the quality of these different types of studies, we did not include a formal assessment of the methodological quality of the included studies. However, we include in the summary table multiple characteristics of the study, including sample size and design, for readers to review and consider. Duplicate articles were identified by the search tool and removed from the database before screening. A two-step process was performed by two authors (M.Y. and S.F.) (Fig. 1). During the first stage, the titles and abstracts of the identified articles were screened according to the exclusion and inclusion criteria described previously. Uncertainty during the initial title and abstract screening was resolved by consulting the fulltext articles.
Abstracting information and summarizing data from included studies. Studies that met the inclusion criteria were screened further as full articles. Full-text screening was assessed for final inclusion into the study, with the selected full-text articles passing an additional bibliographic reference list check to identify any additional eligible publications. The articles reviewed were categorized and summarized depending upon whether the study assessed the outcomes of interest during prenatal period (before the birth of the newborn) or postpartum period (after delivery of the newborn). Information such as the author, year of publication, study design, study population, outcomes, relevant findings, and information about underlying cardiovascular or metabolic risk factors examined was also abstracted.

Supplemental literature search
To better interpret and supplement the information abstracted from studies included in this scoping review related to the secondary research question (pathophysiologic mechanisms that might link depression and PE), a supplemental literature search was conducted. This general search, conducted in PubMed, identified research studies with information that could better explain information regarding the underlying mechanisms of depression and PE obtained from studies included in this scoping review.

Results
Twenty-three articles were included in this scoping review. Nine studies were conducted in North America, six in Europe, five in Asia, two in South America, and one in Africa. Four meta-analyses were included as they fit the inclusion criteria in addressing the research question, despite having a few studies within the metaanalyses that overlapped with those already included within the review. Twelve studies were published in the most recent 5 years.

Associations between prenatal depression and PE during pregnancy
Of all 23 articles reviewed, 13 explored the associations between prenatal depression and PE during pregnancy (Table 3). Some studies reported a significant positive association between the two factors, 15-20 but other studies did not observe such associations or reported associations that were not statistically significant. [21][22][23] In a prospective case-control study carried out in Iran, Kharaghani et al observed that women with mild depression and severe to moderate depression had a 1.81-fold and 2.52-fold increased risk of developing PE, respectively. 17 Conversely, Ghaffar et al    Studies included in the meta-analyses are given in Table 4.
observed no significant association in hospitalized women between depression and PE. 21 Vollebregt et al also reported no significant associations between depression and PE even after adjusting for medical and socioenvironmental factors. 23 Pooled analyses from two meta-analyses, Hu et al (pooled odds ratio [OR] = 1.48; 95% confidence interval [CI], 1.04-2.01) and Zhang et al (pooled OR = 1.5; 95% CI, 1.10-20.5) reported statistically significant positive associations between prenatal depressive symptoms and PE. 24,25 Conversely, Grigoriadis et al did not observe any statistically significant associations between exposure to antenatal depressive symptoms and an increased risk of PE (OR = 1.35, 95% CI, 0.95-1.92). 12 Summaries and details of these three meta-analyses are given in Table 4.

Associations between PE and postpartum depression
Ten of the 23 articles included in this scoping review investigated associations between diagnosis of PE and development of postpartum depression (Table 5). Seven studies reported significant associations between increased prevalence of postpartum depression with the diagnosis of PE. [26][27][28][29][30][31][32] Auger et al observed that women with PE had a higher incidence of hospitalization for depression 28 years after indexed delivery and that late-onset PE was more consistently associated with hospitalization for depression within 1-4 years after pregnancy. 30 Similarly, Bergink et al reported an increased incidence risk ratio of 2.85 (95% CI, 1.84-4.42) for unipolar depression within 0-3 months postpartum in women who developed PE. 31 Caropreso et al was the only meta-analysis that explored the association between PE and postpartum depression in our search. The analysis reported the standard mean difference between groups of women with and without PE as 1.04 (95% CI, 0.22-1.86), which suggested high severity of depressive symptoms in women with a history of PE. 33 Study details of this meta-analysis are given in Table 6.

Examination of common underlying cardiovascular and metabolic risk factors
The majority of the articles included in this scoping review accounted for factors associated with cardiovascular risk, such as prepregnancy BMI, smoking, preexisting hypertension, and diabetes. In the studies investigating associations between prenatal depression and PE, the associations observed between depression and PE remained after adjusting for these cardiovascular risk covariates. 18,19,23,34 In studies assessing the association between PE and postpartum depression, three of the nine studies adjusted for the confounding variable of prepregnancy BMI. As given in Table 5, all three articles, Auger et al, Mommersteeg et al, and Chen et al, reported that positive associations between PE and postpartum depression remained significant after adjustment for prepregnancy BMI. 26,30,35 Integration of information from supplemental literature search Although all included studies were examined for common underlying cardiovascular and metabolic risk factors, it is important to note that none of the studies reported the magnitude of an independent association between the covariates and the association between depression and PE, or more advanced analyses such as stratification or mediation analysis that might provide insight into the characteristics underlying the observed association between PE and depression. Similarly, none of the studies specifically assessed potential common cardiovascular and metabolic risk factors such as arterial stiffness, endothelial dysfunction, or obesity between depressive symptoms and PE despite previous studies suggesting that these factors are associated with both PE and depression individually. [36][37][38][39] To provide better insight into possible pathophysiologic mechanisms that might link depression and PE, integration of the results from this scoping review and information from a supplemental scoping review focused on BMI, vascular stiffness, endothelial dysfunction, inflammation, and oxidative stress are now presented.
Body mass index. A high prepregnancy BMI is associated with an increased risk of antenatal depression. 40 This is unsurprising as obesity and depression outside of pregnancy are often linked and are thought to share similar biological mechanisms. 41 In addition, studies have shown that an elevated BMI increases the risk for developing PE 42,43 ; although obesity alone is not a direct indicator of cardiometabolic health and recent studies have begun to consider the differences between ''healthy'' obesity and obesity with associated metabolic abnormalities. 44 Arterial stiffness. Arterial stiffness is a good indicator of cardiovascular health and common measurements of arterial stiffness include augmentation index, which  17 Kim et al, 20 Kurki et al, 15 Qiu et al, 18 Vollebregt et al 23 17 Kurki et al, 15 Qiu et al, 18 Vollebregt

Not examined in this study
Summary and synthesis 9 out of the 23 studies met the inclusion criteria for this review. Eight of the nine studies reported statistically significant associations between PE (PE) and postpartum depressive symptoms.
a Studies included in the meta-analysis in Table 6. ACOG, American College of Obstetrics and Gynecology; AOR, adjusted OR; HDP, hypertensive disorders of pregnancy, HR, hazards ratio; ISSHP, International Society for the Study of Hypertension in Pregnancy; MNH, Muhimbili National Hospital; NICU, neonatal intensive care unit; PPD, postpartum depression.
represents arterial adaptability, and pulse-wave velocity. In the nonpregnant literature, depression has been associated with a higher augmentation index and pulse wave velocity measurements, 37,[45][46][47] suggesting that individuals with depression have increased arterial stiffness that may contribute to increased cardiovascular risk. Similarly, multiple studies have reported that individuals with a current and/or previous history of PE have significantly higher pulse wave velocity and augmentation indices compared with those who had healthy, uncomplicated pregnancies. 46,[48][49][50][51][52][53][54][55][56] Endothelial dysfunction. Impaired measurements of flow-mediated dilation, an ultrasound-based imaging assessment of dilation in the brachial artery that is used to quantify the health or dysfunction of the endothelium, have been associated with depression in the nonpregnant population. 57,58 Prenatal measurements of flow-mediated dilation have been reported to be impaired in pregnant women with established PE or those who develop PE later in pregnancy when compared with women with normotensive pregnancies. 49,50,[59][60][61] Inflammation. Inflammation has been implicated in both depression and PE. Studies have found increased levels of inflammatory markers to be associated with both depression and PE. For example, outside of pregnancy, commonly studied inflammatory markers, such as interleukin-1, interleukin-6, and C-reactive protein, have been shown to be positively associated with depression. 62 In pregnancy specifically, individuals with a history of depression have increased plasma C-reactive protein and glycoprotein acetyls. 63 Similarly, increased levels of these markers have been shown to be associated with PE. 64,65 Oxidative stress. Markers of oxidative stress have also been studied in both depression and PE. In the general population, depression has been shown to be associated with increased oxidative stress and reduced antioxidant activity. 66 Furthermore, one study found that antenatal depression was significantly associated with increased levels of 8-isoprostane, a specific biomarker of oxidative stress. 67 Likewise, individuals with PE have elevated markers of oxidative stress, including 8isoprostane as well as malondialdehyde and prostaglandin F2a. 64,68 The well-established relationships between inflammation, oxidative stress, and vascular function and the risk factors associated with both depression and PE (i.e., obesity and hypertension) 55,69,70 suggest that these factors may contribute as underlying risk factors between depression and PE. These similarities underscore the importance to better understand these mechanisms, so preventative and therapeutic strategies could be developed to improve maternal health.

Discussion
The scoping review addressing the first objective of this study summarized research findings regarding the association between depression and PE and found conflicting results, although these inconsistent results may reflect the challenges of ''standardizing'' evaluations during a complex time period. Contextually, pregnancy can span from conception until the delivery of the newborn, suggesting various time points where depressive symptoms can be assessed (i.e., 1st, 2nd, and 3rd trimester). Fluctuating moods and emotions between different stages of the pregnancy may lead to varying levels of negative affect and thus rendering some associations significant and others insignificant. For this reason, the prevalence of depression in pregnancy may be over-or underestimated, depending on the timing of assessment.
The definition of the postpartum period varied from immediately postdelivery, 1-6 months after delivery, and up to several years after the indexed delivery. Despite the varying time of assessment, the majority of the studies reported that there was an increased risk of experiencing postpartum depression following PE. These findings aligned with the meta-analysis performed by Caropreso et al, 33 which reported that 8 of 13 studies observed associations between PE and postpartum depression or depression outside the perinatal period.
As most of the studies included in our scoping review were epidemiological and thus only assessed the prevalence and association of these disorders in pregnant women, following a secondary literature search, we outlined some of the cardiovascular and metabolic factors associated with both pathologies (e.g., preexisting obesity, increased systemic inflammation, oxidative stress, vascular changes, and endothelial dysfunction) that may contribute to the common underlying etiologies between depression and PE. The complex relationship between these pathologies and risk factors are given in Figure 2.
Many of these risk factors interact with one another and may be linked in a vicious cycle affecting overall FIG. 2. Concept map of associations between risk factors and conditions. Solid lines represent explored and known associations. Dash lines represent relationships of interest for this scoping review: (1) primary interest (orange and red) and (2) potential underlying mechanism (blue and green). cardiometabolic health. For example, both PE and depression are associated with increased rates of obesity and cardiovascular disease, which too share many of these underlying risk factors. 41,44,[71][72][73] There is substantial evidence from both animal and human studies that depression and cardiovascular disease are linked through changes in oxidative stress, hyperactivation of the hypothalamic-pituitary-adrenal axis and the immune system. 71,[74][75][76] These changes are also considered mechanistic links between depression and obesity. 41 PE has been associated with an imbalance between the antioxidant defense system and the production of reactive oxygen species, which then leads to an increase in generalized inflammatory processes and endothelial dysfunction. 77,78 Endothelial dysfunction in women with PE has also been associated with significantly higher levels of oxidative stress, markers of systemic inflammation, and overactivation of the hypothalamic-pituitary-adrenal axis. 79,80 Risk factors such as vascular endothelial dysfunction, as evidenced by reductions to the flow-mediated dilation response, are tightly linked with prooxidant and proinflammatory environments across a wide of array of conditions. The ''positive feedback loop'' that many investigators have identified in linking elevated oxidant stress with chronic inflammation can have the impact of reducing mediators of vascular function such as nitric oxide and prostacyclin bioavailability and increase vascular production of constrictor metabolites such as thromboxane A 2 among other metabolites of arachidonic acid. Each of these potential contributors may act to increase vascular resistance within human subjects and alter patterns of cerebral perfusion-which has been identified as a likely contributor to cognitive impairments including depressive symptoms.
In addition, reductions to arterial and vascular wall distensibility will have significant impacts on the penetrance of pulse wave hemodynamics deeper into the cerebrovascular circulation and has been implicated in cerebrovascular network remodeling impairing mass transport and exchange within the microcirculation. Taken together, this constellation of vascular impairments and elevated risk factor profiles may represent a significant contributor to the emergence of depressive symptoms, potentially of vascular origin. This is an area of investigation that will require significantly more study in the coming years to better understand how to identify and support women who may be at higher risk for these conditions and help alleviate subsequent complications.
Limitations of this scoping review include restricting to English-only articles from three databases and including all eligible studies without evaluation of the quality or size of the study. In addition, to specifically assess the association between depression during pregnancy and PE, this review excluded any studies that evaluated the possible mediating effects of antidepressant medications. Strengths of this study include the differentiation between prenatal and postpartum depression and the abstraction and synthesis of information related to potential underlying causal mechanisms. Previous reviews on this topic have only assessed the relationship between either prenatal depressive symptoms or postpartum depressive symptoms and PE. Furthermore, this scoping review focused on identifying common underlying cardiovascular risk factors that may contribute to the relationship between the two conditions. This is the only scoping review thus far that investigates the available literature on depression and PE throughout the perinatal period with consideration for common underlying risk factors.

Conclusion
In conclusion, pregnant women with prenatal depressive symptoms may be at increased risk of developing PE and women who developed PE are at increased risk of developing postpartum depressive symptoms. Moreover, this review highlighted the cardiovascular and metabolic changes that may be contributing to the relationship between the two pathologies and may be useful as early pregnancy indicators of those who are at a higher risk for adverse pregnancy outcomes. Although universal screening for prenatal depression and PE is beneficial and are goals for many obstetrical settings, an overall assessment of cardiometabolic health in early pregnancy may be more useful to detect those who are most at risk for depression and/or PE and their associated complications.
Author Disclosure Statement All the authors are members of the University of Western Ontario and have no competing interests to declare.
Funding Information M.Y. received funding from the Canada Graduate Scholarship-Master's Program from the National Science Engineering Council.