Clinical Procedures for the Prevention of Preeclampsia in Pregnant Women: A Systematic Review

Abstract Objective To identify the most effective procedures recommended for the prevention of preeclampsia. Data Sources A systematic review was performed in the following databases: Pubmed/MEDLINE, CINAHL, Web of Science, Cochrane and LILACS via the Virtual Health Library (VHL). A manual search was also performed to find additional references. The risk of bias, the quality of the evidence, and the classification of the strength of the recommendations were evaluated using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach. Selection of Studies In the initial search in the databases, the total number of articles retrieved was 351, and 2 were retrieved through the manual search; after duplicate articles were removed, 333 citations remained. After a thorough review of the titles and abstracts, 315 references were excluded. Accordingly, 18 articles were maintained for selection of the complete text (phase 2). This process led to the exclusion of 6 studies. In total, 12 articles were selected for data extraction and qualitative synthesis. Data Collection The articles selected for the study were analyzed, and we inserted the synthesis of the evidence in the online software GRADEpro Guideline Development Tool (GDT) (McMaster University and Evidence Prime Inc. All right reserved. McMaster University, Hamilton, Ontário, Canada); thus, it was possible to develop a table of evidence, with the quality of the evidence and the classification of the strength of the recommendations. Data Synthesis In total, seven studies recommended the individual use of aspirin, or aspirin combined with calcium, heparin or dipyridamole. The use of calcium alone or in combination with phytonutrients was also highlighted. All of the studies were with women at a high risk of developing preeclampsia. Conclusion According to the studies evaluated, the administration of aspirin is still the best procedure to be used in the clinical practice to prevent preeclampsia.


Introduction
Preeclampsia (PE) is a systemic disease, and much of its etiology is unknown. Evidence suggests that it is characterized by angiogenic imbalance, exaggerated inflammation and endothelial dysfunction, leading to clinical manifestations of pregnancy-specific hypertension, which usually occurs in the second half of pregnancy. 1,2 It is one of the main causes of maternal and perinatal death, affecting 3% to 5% of pregnancies worldwide, and accounting for 25.7% of all maternal deaths in Latin America and the Caribbean. 3,4 The risk of these complications is greater when the disease has an early onset (before 34 weeks of gestation), and it is serious, as it causes restricted intrauterine growth (RIG) or fetuses that are small for the gestational age (SGA), severe maternal morbidities, and it also causes women to become more prone to develop cardiovascular risk factors. The costs for each birth with the presence of preeclampsia have been shown to increase by US$ 6,583. [5][6][7][8] One of the major challenges for modern obstetrics is the early identification of women at a high risk of developing preterm preeclampsia (before 37 weeks of gestation). Thus, the prediction of preeclampsia is paramount in performing appropriate procedures to reduce the prevalence of the disease by modifying antenatal care and pharmacological interventions in a timely manner. 9,10 Several clinical, biochemical and biophysical tests have been proposed to predict PE; however, none of them have proved to be useful in isolation to classify the preexisting maternal risk for PE with sufficient specificity and sensitivity for clinical use. Ideally, the predictive model should be simple, inexpensive, and easy to apply, so that it can be widely used in the clinical practice. 11 Researchers developed a predictive model that is able to predict the risk of developing PE in the first trimester (11 weeks, 0 day -13 weeks, 6 days), based on a combination of maternal factors that includes the mean arterial pressure (MAP), the pulsatility index (PI) of the uterine artery, and the serum placental growth factor (PlGF). This model can predict Conclusion According to the studies evaluated, the administration of aspirin is still the best procedure to be used in the clinical practice to prevent preeclampsia.

Resumo
Objetivo Identificar quais são as condutas recomendadas para a prevenção de préeclâmpsia em gestantes. Fontes de Dados Foi feita uma revisão sistemática da literatura, e foram desenvolvidas estratégias detalhadas de busca individual nas bases de dados PubMed/MEDLINE, CINAHL, Web of Science, Cochrane e LILACS pela Biblioteca Virtual em Saúde (BVS). Uma pesquisa manual também foi realizada para encontrar referências adicionais. O risco de viés, a qualidade da evidência, e a classificação da força das recomendações foram avaliadas usando a abordagem Classificação 12 Accordingly, strategies to reduce the risk of pregnancyrelated hypertension disorders are a global priority, and screening only makes sense if there is an effective intervention available to prevent the disease and control the unfavorable outcomes. 13,14 Therefore, the present study sought to answer the following question: what are the most up-todate and effective procedures to prevent PE? The aim of the present review was to identify the most effective procedures recommended for the prevention of PE.

Type of Study
A systematic review based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 15

Eligibility Criteria
The inclusion criteria were 1) randomized controlled trials (RCTs), systematic reviews/meta-analyses, and prospective or retrospective cohort studies 2) that included pregnant women of all ages, with a single pregnancy and without preeclampsia in the current gestation, 3) which reported the outcome of interest (prevention and control of PE). There were no language restrictions. Articles published between the years of 2013 and 2018 were included. Studies were excluded due to the following reasons: they did not cover the topic; and studies that were letters, abstracts of congresses, personal opinions, chapter of books, descriptive studies, case reports, or case series, study protocols and qualitative studies.

Information Sources and Search Strategy
The studies were identified through a search strategy adapted for each electronic database: PubMed/MEDLINE, CINAHL, Web of Science, Cochrane and LILACS via the Virtual Health Library (VHL). All of the electronic database searches were performed on December 14, 2018. In a complementary way, a manual search was performed on the reference lists of the selected articles, to ensure that any additional references that might not have been retrieved in the electronic search were duly found. To perform the searches in the electronic databases, the following medical subject headings (MESH) were used: pregnant women, preeclampsia, prevention and control; we also used the AND Boolean operator to search for the combination (pregnant women AND preeclampsia AND prevention and control).

Selection of the Studies
After collecting all of the references, duplicate articles were deleted using a free online bibliographic reference manager (EndNote Web). The selection of the studies was performed in phases. In phase 1, two investigators independently examined the titles and abstracts of potentially relevant studies and selected articles that appeared to meet the inclusion criteria based on their abstracts. In phase 2, the same reviewers read the full text of all of the independently-selected articles and excluded studies that did not meet the inclusion criteria. Any disagreements, whether in the first or second phases, were solved through discussions between the two investigators. In the event of failure to reach a consensus, a third reviewer was called to make a final decision.
The studies that were excluded and the reasons for exclusion are presented in ►Fig. 1.

Data Collection Process
Two investigators independently collected data from the articles, including: study characteristics (references, year of publication, objective), population characteristics (sample size, gestational age), procedure characteristics (study design, follow-up period) and main results.

Summary of the Measures
The primary outcome of interest in the study was the prevention of PE. In addition, fetal outcomes (preterm delivery, premature rupture of membranes, restricted intrauterine growth, Apgar score, and fetal/neonatal death) were also considered as secondary outcomes.

Risk of Bias in the Studies and Classification of the Strength of the Recommendations
The quality of the evidence and the classification of the strength of the recommendations were evaluated through the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach. 16 The synthesis of the evidence was digitalized using the online software GRADEpro Guideline Development Tool (GDT) (McMaster University and Evidence Prime Inc. All right reserved. McMaster University, Hamilton, Ontário, Canada), which made it possible for us to develop the evidence table. The criteria for this evaluation were study design, risk of bias, inconsistency, indirect evidence, inaccuracy, possible publication bias, whether the study had a large effect size, the potential confounding factors, and the dose-response gradients of the procedures. With this, the quality of the evidence was characterized as high, moderate, low or very low.

Selection of the Studies
In phase 1 of the study selection, 351 citations were identified in the 5 databases consulted. The manual search from the reference lists of the identified studies provided two additional studies. After duplicate articles were removed, 333 citations remained. After a thorough review of the titles and abstracts, 315 references were excluded. Accordingly, 18 articles were maintained for selection of the complete text (phase 2). This process led to the exclusion of six studies (►Fig. 1). In total, 12 articles were selected for data extraction and qualitative synthesis.

Characteristics of the Studies
Of the twelve studies selected for the present review, three of them used only aspirin, four used aspirin combined with calcium, heparin or dipyridamole, while the others used only calcium or calcium combined with phytonutrients, L-arginine, heparin or pravastatin as procedures for the prevention of PE. Most studies included women at a high risk of developing PE in their samples; for this, those who had at least one risk factor for the disease were considered at a high risk of developing PE; one of the risk factors analyzed was pregnant women with systemic hypertension before pregnancy, aiming to evaluate superimposed PE.
The descriptive characteristics of the studies are summarized in Chart 1.

Risk of Bias in the Studies and Classification of the Strength of the Recommendations
The quality of the evidence of the results evaluated through the GRADE system was classified as high certainty, moderate certainty, low certainty and very low certainty (Chart 2).
Most of the studies classified as high-certainty were those that used the administration of aspirin individually as the procedure. The other high-certainty studies used L-arginine, phytonutrients and aspirin with calcium, and they demonstrate that further studies are unlikely to change the confidence in the estimate of the effect of these findings, which allows us to have confidence in these results. Among the high-certainty studies, the use of phytonutrients was not statistically significant in the reduction of the risk of developing PE, a fact that discourages their isolated use as a preventive measure.
One study 20 involving the addition of low-molecularweight heparin (LMWH) or unfractionated heparin to a low dose of aspirin and one 21 that used a low dose of aspirin (75 mg) were moderately certain, indicating that additional studies should be performed, and are likely to have an impact on the confidence in the estimate of the effect, and may  To estimate the effectiveness and safety of L-arginine to prevent PE in high-risk patients.
L-arginine 3 g once daily orally in 600 mg capsules.
Beginning at the 20th week of gestation with follow-up every 3 weeks during pregnancy and 2 weeks after birth.
The incidence of severe PE was higher in the placebo group (n ¼ 7) than in the Larginine group (n ¼ ;,

Randomized controlled trial
To evaluate the maternal-fetal safety and pharmacokinetic parameters of pravastatin when used in pregnant women at high risk of developing PE.
Pravastatin 10 mg, 1 capsule orally per day until birth. Beginning in the 1st trimester (weeks 12 to 16) and followup until birth.
There were 4 cases of PE in the placebo group and none in the pravastatin group, with 5 preterm births before 37 weeks in the placebo group compared with 1 in the pravastatin group.
Talari et al (2014) 9 Randomized controlled trial To determine whether the treatment with acetylsalicylic acid (ASA) reduces the incidence of PE among pregnant women with abnormal uterine artery flow.
Aspirin group (n ¼ 40); placebo group (n ¼ 40). change the estimate. The study 13 that used calcium for the prevention of PE presented low certainty, even though this drug is already used for this purpose, and is well accepted in populations with calcium deficiency in the diet. This indicates that further studies are likely to have a significant impact on the confidence in the estimate of the effect, and may change the estimate.
Finally, only two studies 3,23 were classified as very-lowcertainty, suggesting that the estimation of the effect of the administration of pravastatin and heparin as the procedure for the prevention of PE is very uncertain. Perhaps this was due to the fact that these studies had small samples, but statistically significant results, and due to the fact that these procedures are still not used much for this purpose. Preterm preeclampsia occurred in 13 out of the 798 participants (1.6%) in the aspirin group, compared with 35 out of 822 (4.3%) in the placebo group, when aspirin was started between the 11th and 14th gestational weeks and continued until the 36th week. There was a significant difference between the aspirin and placebo groups in the incidence of preeclampsia (2.5% versus 22.5% respectively). The aspirin group presented an $ 11-fold reduced risk of developing preeclampsia. No difference was observed in the incidence of preeclampsia between the phytonutrient and placebo groups (15.9% versus 16.3% respectively). Follow-up: variation from the 12th week to birth. Aspirin was associated with a significant reduction in the prevalence of preeclampsia, severe preeclampsia and fetal growth restriction, with a significant dose-response relationship when initiated before the 16th week of gestation. The primary outcome, being a composite of preeclampsia, birth of a newborn small for gestational age (percentile < 10), placental abruption, or termination of pregnancy > 20 weeks, were significantly reduced by low-molecular-weight heparin, with an RR reduction of 0.52 (95%CI: 0.32-0.86; p <.01) The four trials with low risk of bias, all in women at a high risk of developing preeclampsia, showed a consistent reduction in the incidence of preeclampsia, even with calcium started at the end of the pregnancy. RCT, randomized controlled trial; RR, risk ratio. Notes: Ã The risk in the intervention group (and its 95% confidence interval) is based on the risk assumed for the comparison group and the relative effect of the intervention (and its 95% confidence interval). a Inaccurate due to not presenting confidence interval. b Highly suspicious due to sample size, form of sample selection, and for presenting statistically significant results. c Due to the heterogeneity of the selected studies.

Discussion
The present review included 12 studies that addressed different procedures for the prevention of PE, and they were published in the previous 5 years. Most studies used the administration of aspirin as the procedure to prevent PE. Three of them used only aspirin 5,9,21 , and another four 14,19,20,25 used aspirin combined with calcium, heparin or dipyridamole. These findings are in line with the international recommendations for women at a high risk of developing PE. 17,18 In all of the studies, the expected outcome, the prevention of PE, was achieved. This fact confirms the efficacy of aspirin when administered to pregnant women at a high risk of developing PE when it is started prior to the 16th week of pregnancy, preferably between weeks 11 and 14. Regarding the ideal dose, the studies have administered doses between 50 mg and 150 mg, and most suggest that the dose should be higher than 100 mg in order to achieve satisfactory effects. 5,9,14,[19][20][21] The recommendation for the use of aspirin is associated with uteroplacental underperfusion due to impaired placentation, placental vascular lesions, and incomplete transformation of the spiral uterine arteries. 14 Approximately 30 thousand women have already participated in RCTs with aspirin, thus making this one of the most intensely studied drugs in obstetrics. 22 Physiologically, reduced placental blood flow is thought to lead to the activation of the coagulation cascade, platelet aggregation and endothelial dysfunction, which may be partially mediated by prostacyclin and thromboxane A2, so that aspirin selectively reduces thromboxane A2 without significantly affecting prostacyclin levels and also inhibiting placental aggregation. 21 The low dose of aspirin inhibits the biosynthesis of platelet thromboxane A2 by altering the imbalance between thromboxane A2 and prostacyclin, and is therefore thought to favor vasodilation, which assists in the reduction of blood pressure and the prevention of PE. 23 This mechanism has achieved good results in practice, more in reducing the incidence of PE in high-risk women than in those of low and moderate risk. 14 It is recommended that antiplatelet agents be administered to pregnant women at a high risk of developing PE or to those with chronic hypertension, as they lead to a wide reduction in adverse pregnancy outcomes for both the mother and the newborn. 24 The results of the systematic reviews showed that antiplatelet agents are associated with a reduction in the risk of developing preterm PE (relative risk [RR]: 0.62; 95%CI: 0.45-0.87), when initiated before the 16th gestational week, and they are also associated with a significant reduction in the prevalence of PE, severe PE, and restricted intrauterine growth. 19,25 For LMWH, a meta-analysis showed that heparin alone reduced the risk of: severe or early onset of PE ( 26 The second most frequent procedure in the selected studies was the prescription of calcium, although administered alone it obtained low certainty. 13 In the other studies 14,27 , calcium was combined with other medications, such as phytonutrients and aspirin, both of which obtained high certainty. Calcium is also used for the prevention of PE because there is epidemiological, clinical and laboratory evidence that indicates the association of dietary calcium deficiency with hypertensive disorders of pregnancy. 13 Like aspirin, calcium is recommended by international societies, and the doses range from 1.5 g to 2.0 g of elemental calcium daily for pregnant women residing in areas with low dietary intake of calcium. 17,18 One systematic review 28 highlighted calcium supplementation (1 g to 2 g/day) for the prevention of PE, and found that calcium supplementation significantly reduces the risk (between 55% and 64%) in women with low dietary calcium intake. In addition, calcium supplements have a significant protective effect on maternal mortality and severe morbidity and premature delivery, with a tendency to reduce the risks of low birth weight and stillbirth. 24 Another procedure found in one study 27 was the use of phytonutrients for the prevention of PE. The contents of the antioxidant supplement capsule consisted of a blend of fruits and vegetable juice powder concentrate derived from acerola, cherry, apple, beetroot, broccoli, cabbage, carrot, blackberry, kale, orange, peach, papaya, parsley, pineapple, spinach and tomato, with each capsule providing 7.5 mg of β-carotene, 234 mg of vitamin C, 30 mg of vitamin E in the form of RRR-αtocopherol, 420 mg of folate, and 60mg of calcium. It was initiated in the 12th week of gestation, and the use was maintained until birth. 27 This study, 27 even with high certainty, found no difference between the phytonutrient and placebo groups in the prevention of PE, discouraging its use in the clinical practice.
In relation to the administration of pravastatin and Larginine, both are innovative procedures that are little used in the prevention of PE. However, their real benefits are still unclear, as studies are still incipient. In the present review, the studies 3,10 regarding their administration were classified as very-low-certainty and high-certainty respectively. Pravastatin (10 mg) is a hydrophilic statin used to lower serum cholesterol and prevent cardiovascular disease. Some of the main characteristics of PE are angiogenic imbalance, exaggerated inflammation, and endothelial dysfunction, which lead to clinical manifestations of hypertension, proteinuria and target-organ damage. Furthermore, although PE is a disease exclusive to pregnancy, it has pathogenic similarities and is associated with many risk factors for cardiovascular disease in adults. 3 The use of statins in the treatment of PE has taken a leading role in recent years. In particular that of pravastatin, with many studies and strong scientific evidence supporting its use as the statin of first therapeutic option for the treatment of early and severe forms, which is in disagreement with the findings of the present review. 29 The ability of statins to reverse the angiogenic imbalance that is characteristic of PE, and to restore endothelial dysfunction in animal models, demonstrates the need for clinical trials that further assess the efficacy of pravastatin. 29,30 Unlike pravastatin, L-arginine (3 g) was classified as highcertainty. L-arginine is a basic amino acid nitric oxide precursor that has the ability to dilate vessels and improve blood flow and circulation. 10 In a study 10 included in the present review, the number of pregnant women with severe PE was higher in the placebo group; in addition, the systolic blood pressure, the diastolic blood pressure and the mean arterial pressure were significantly lower in the L-arginine group compared with the placebo group. This demonstrates that the oral treatment with 3 g per day of L-arginine has a significant effect on the prevention of PE in high-risk patients, as well as on the ability to avoid more severe outcomes.
The cost of L-arginine is high compared with that of aspirin; however, the benefits of L-arginine in the prevention of PE exceed the cost of treating PE. Furthermore, Larginine showed good results when administered to patients with 19 to 20 weeks of gestation. L-arginine therapy may be important because many patients with PE have their first consultation with a physician when it is too late for other treatments, such as the one with aspirin, to be effective.
Although there was no limit to the type of conduct for the prevention of PE, to obtain more publications on the subject, none of the studies verified the relationship between nonmedicated treatment and the prevention of PE. Moreover, we found no studies that evaluated women classified as low risk for PE, and the conduct for this population.

Conclusion
Based on the studies included in the present review, the administration of aspirin is considered the best procedure for the prevention of PE in high-risk women, which corroborates the results of several studies published in recent years. Further studies are needed to identify and test other medications and their benefits during pregnancy in the prevention of adverse maternal and perinatal outcomes.