The association between dietary mineral intake and the risk of preeclampsia in Chinese pregnant women: a matched case–control study

Previous studies on the relationship between dietary minerals and preeclampsia (PE) have given inconsistent results. The aim of this study was to further clarify the relationship between dietary minerals intake and PE in Chinese pregnant women. In this study, 440 pairs of hospital–based preeclamptic and healthy women were matched 1:1. Dietary intake was obtained through a 78–item semi–quantitative food frequency questionnaire. Multivariate conditional logistic regression was used to estimate the odds ratios (ORs) and 95% confidence intervals (CIs). Restricted cubic splines were plotted to evaluate the dose–response relationship between dietary minerals intake and PE. This study found significant inverse associations for dietary intake of calcium, magnesium, phosphorus, iron, copper, manganese and zinc and the risk of PE in both univariate and multivariate models (all P- trend < 0.05). After adjusting for possible confounders, compared with the lowest quartile, the odds ratio of the highest quartile was 0.74 (95% CI 0.56–0.98) for calcium, 0.63 (95% CI 0.42–0.93) for magnesium, 0.45 (95% CI 0.31–0.65) for phosphorus, 0.44 (95% CI 0.30–0.65) for iron, 0.72 (95% CI 0.53–0.97) for copper, 0.66 (95% CI 0.48–0.91) for manganese and 0.38 (95% CI 0.25–0.57) for zinc. In addition, a reverse J–shaped relationship between dietary minerals intake and PE risk was observed (P–overall association < 0.05). In Chinese pregnant women, a higher intake of dietary minerals, including calcium, magnesium, phosphorus, copper, iron, manganese, and zinc was associated with a lower odds of PE.

In addition, many minerals act as antioxidants to maintain the normal development of the placenta 13 .A lack of these elements in pregnant women increase the risk of complications, such as high blood pressure and PE 14 .In addition, inflammation has deleterious effects on placental function in humans and rodents, providing a potential mechanism underlying the development of PE 15,16 .Therefore, moderate mineral intake during pregnancy may play an important role in preventing PE.
There have been many studies on the relationship between mineral intake during pregnancy and the risk of PE.An unmatched case-control study in Ethiopia 17 showed a significant association between low dietary calcium intake and low serum calcium concentrations and PE.A meta-analysis by Ma et al. on the relationship between zinc and PE showed that moderate zinc supplementation during pregnancy reduced the incidence of PE 18 .A cross-sectional study in India found that iron supplementation during pregnancy reduced the incidence of PE 19 .However, another studies found no association between dietary calcium and magnesium intake and PE 20 .In a randomized controlled study in China 21 , women were randomly assigned to one of the three treatment groups: folic acid, folic acid and iron, or folic acid, iron and 13 additional vitamins and minerals.No differences were found in the incidence of hypertension between the groups.Given these conflicting results, differences in dietary habits across countries and the severe consequences of PE, we conducted a matched case-control study to clarify the relationship between dietary mineral intake and PE in Chinese women.

Participants
This 1:1 matched case-control study was conducted from March 2016 to June 2019 in the First Affiliated Hospital of Zhengzhou University, China.This 1:1 matched case-control study was conducted at the First Affiliated Hospital of Zhengzhou University in China from March 2016 to June 2019.The inclusion and exclusion criteria were the same as those used in previous studies 22 .According to the Guidelines for the Diagnosis and Treatment of Hypertensive Diseases during Pregnancy (2015) 23 , the case group were singleton pregnant woman of childbearing age (≥ 18 years) after 28 weeks of gestation diagnosed with PE.Control subjects were women preparing for delivery at the same hospital who did not have hypertension or albuminuria and were matched for age (± 3 years), gestation week (± 1 week) and gestational diabetes mellitus (GDM).The exclusion criteria for subjects in both groups were as follows: (1) patients with heart disease, malignancy, hyperthyroidism, immune system diseases, chronic renal insufciency, and other endocrine system diseases; and (2) patients with epilepsy, depression and other mental or cognitive dysfunction.
The study was conducted in accordance with the Declaration of Helsinki and the research protocols were approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University (Scientific Research No. 2016-LW-34).All participants signed an informed consent form before the collection of epidemiological data and biological specimens.All data used for analysis were anonymized.

Sample size calculation
The sample size calculation was based on previous studies of the association between dietary minerals and PE (odds ratio [OR] = 0.42) 17 .Based on cross-sectional, cohort, and intervention studies on dietary mineral intake during pregnancy, we estimated that 15% of the general pregnant population has adequate dietary mineral intake 24 .The statistical power (β) was set at 80% and the significance level was set at 0.05.The sample size required for each group was calculated as 132.

Data collection
The structured questionnaire was filled out by trained researchers during a face-to-face session.The questionnaire mainly interrogated the following: (1) food intake, (2) demographic characteristics, (3) anthropometric data and (4) nutritional supplement intake.A 78-item, semi-quantitative food frequency questionnaire (FFQ) was used to assess the dietary status of pregnant women in the 3 months before delivery.The following four levels of food-intake frequency were recorded for each item: daily, weekly, monthly or never.The reliability and validity of the FFQ have been demonstrated in previous studies 25 .The assessments of dietary minerals intake (mg/d) and energy intake (kcal/d) were calculated using the Chinese Food Composition Tables (2004) 26 and the China Food Composition (2nd Edition) 27 .Only mineral intake from food was considered.Mineral intake from nutritional supplements was excluded.Demographic characteristics, such as age, occupation and family history of hypertension, diseases, menstruation, pregnancy and childbirth, were collected.Anthropometric data included height, weight, blood pressure.

Statistical analysis
A paired Student's t-test or Wilcoxon signed rank-sum test was used to compare the quantitative data between the two groups.McNemar's test was used to compare qualitative data.Dietary mineral intake was adjusted for energy using the residual method 28 .
Based on the distribution in the control group, dietary mineral intake levels were divided into four equal parts (Q1-Q4), with Q1 as the control group.The association of dietary mineral intake with PE was analyzed using univariate and multivariate conditional logistic regression analysis.Confounding factors adjusted in the multivariable models were as follows: age, gestational week, pre-pregnancy body mass index (BMI), weight gain during pregnancy, educational level, income, parity, history of adverse pregnancy, family history of hypertension, use of multivitamin and mineral supplements, physical activity, energy intake and energy-adjusted vegetable and fruit intake.The median of each metric was entered into the model as a continuous variable for trend testing.A sensitivity analysis was also performed in which we analyzed the association of dietary mineral intake with PE after excluding pregnant women with GDM.A restricted cubic spline (RCS) plot was used to analyze

Association between dietary mineral intake and the risk of developing PE
The relationship between dietary mineral intake and the risk of developing PE is shown in Table 2. Significant inverse and dose-response associations for the dietary intake of calcium, magnesium, phosphorus, iron, copper, manganese and zinc were found in the both univariate and multivariate models.After adjusting for possible confounders, compared with the lowest quartile, the OR of the highest quartile was 0.74 (95% confidence interval [CI] 0.56-0.98)for calcium, 0.63 (95% CI 0.42-0.93)for magnesium, 0.45 (95% CI 0.31-0.65)for phosphorus, 0.44 (95% CI 0.30-0.65)for iron, 0.72 (95% CI 0.53-0.97)for copper, 0.66 (95% CI 0.48-0.91)for manganese and 0.38 (95% CI 0.25-0.57)for zinc.

Ethics approval
The study was conducted in accordance with the Declaration of Helsinki and the research protocols were approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University (Scientific Research No. 2016-LW-34).All participants signed an informed consent form before the collection of epidemiological data and biological specimens.All data used for analysis were anonymized.

Discussion
This 1:1 matched case-control study found an inverse association between the intake of dietary minerals, i.e., calcium, magnesium, phosphorus, copper, iron, manganese, and zinc, and the development of PE.This associations persisted after the exclusion of participants with GDM.This finding has important public health implications for the prevention of PE in pregnant women in China.
The question of whether low calcium intake is related to PE has been controversial.There are many epidemiological studies that have observed an association between calcium and the development of PE 17,29 , consistent with the results of this study (calcium intake of control group = 555.31mg/d, case group = 615.73mg/d, P < 0.001).Clinical trials have also been conducted to determine the potential benefits of preventive calcium supplementation in pregnant women 30,31 .In their meta-analysis, Patrelli et al. showed that calcium supplementation during pregnancy reduced the risk of PE more clearly in people with a low calcium intake and with a higher risk of PE than the general population 32 .However, some studies have not shown an effect of calcium on the development of PE [33][34][35] .Gupta et al. conducted a community-based cross-sectional study, and found no association between low calcium intake or hypocalcemia and PE (P = 0.57 and P = 0.74, respectively) 33 .In a case-control study conducted in Washington, D.C., no statistical difference in dietary calcium intake was found between a group of women with PE and a group of women without PE (P = 0.59) 34 .In general, the protective effect of calcium depends on the region, maternal calcium levels and health status 36,37 .More studies on the relationship between dietary calcium intake and PE are needed.
Our results show a significant negative association between dietary copper intake and PE, with RCS curves suggesting a reverse J-shaped relationship.This is consistent with the findings of Kim et al. 38 and Bo et al. 39 .In the Fig. 1e at certain level of copper 15 mg/d on the 95% CI for the risk of PE is increased.Through reactive oxygenmediated reactions, excess copper intake can cause DNA damage, lipid peroxidation and protein modification, thereby affecting blood pressure levels 40 .In addition, excess copper causes an increase in the amount of mature collagen and increases resistance to blood vessels 41 .However, further mechanistic studies are needed to uncover the pathways involved in this association.
Many previous studies have evaluated the associations between magnesium, phosphorus, iron, manganese and zinc and the risk of PE.Our findings are consistent with those of some previous epidemiological studies that found dietary magnesium intake to be inversely associated with the development of PE [42][43][44] .Jain et al. examined blood samples from pregnant women with and without PE and found that a decrease in serum magnesium concentration during pregnancy may be a cause of PE 42 .A meta-analysis of magnesium intake and PE also showed that magnesium supplementation during pregnancy reduces the risk of PE (risk ratio [RR] = 0.54, 95%CI 0.59-0.98,P = 0.04) 43 .Our results for phosphorus 45 , iron 46 , manganese 47 and zinc 48 were similar to those obtained in previous studies.Hajianfar et al. conducted a prospective cohort study to investigate the association between dietary iron intake during the first trimester of pregnancy and pregnancy outcomes 46 .The results of www.nature.com/scientificreports/this study showed that higher haeme, non-haeme, and total iron intakes were associated with a lower risk of PE (haeme: crude P = 0.05; non-haeme iron: adjusted P = 0.02; total iron: adjusted P = 0.05).Similarly, in a study of Nigerian women, serum concentrations of zinc were significantly lower in women with PE than in women without (P < 0.001) 48 .These results suggest that the intake of magnesium, phosphorus, iron, manganese and zinc during pregnancy may reduce the incidence of PE.
Previous studies have suggested several mechanisms by which minerals may affect the pathogenesis of PE.Oxidative stress and inflammation have been shown to be important in the pathogenesis of adverse pregnancy outcomes 13,49 .PE is characterized by impaired placental function and may be caused by abnormal remodelling of spiral arteries.Oxidative stress due to inadequate spiral artery remodelling is an important factor associated with PE 50 .In the second trimester of pregnancy, the placenta gradually secretes a large number of anti-angiogenic factors, causing vascular inflammation, endothelial dysfunction and maternal vascular injury.The end result of altered angiogenesis is hypertension and multiple organ damage 2 .Magnesium, copper and zinc are all required for the proper functioning of enzymes like superoxide dismutase, which are needed to scavenge free radicals.A lack of these elements during pregnancy may impair the antioxidant potential of cells by reducing superoxide Table 2. Risk of PE during pregnancy according to quartiles of dietary minerals intake.OR odds ratios, CI confidence interval, Q quartile.*P < 0.05; **P < 0.01.# Determined by entering the median intake from each quartile as a continuous variable in the regression models.a Adjusted OR from the conditional logistic model adjusted for age, gestational week, pre-pregnancy body mass index, weight gain during pregnancy, educational level, income, parity, adverse pregnancy history, family history of hypertension, physical activity, the use of multivitamin and mineral supplements, daily energy intake and energy-adjusted daily vegetable and fruit intake using the Enter Method.dismutase activity and increasing lipid peroxidation, leading to elevated blood pressure [51][52][53] .In addition, low levels of manganese may reduce the activity of manganese superoxide dismutase, an antioxidant located in the mitochondria, eventually leading to the accumulation of reactive oxygen species that promote the development of PE 54 .Mineral deficiencies may increase blood pressure by promoting the production of certain hormones.Iron deficiency may lead to hypoxia, which stimulates the secretion of stress hormones, e.g., norepinephrine and cortisol, which increase the risk of placental oxidative stress 55 .Low calcium intake stimulates the release of parathyroid hormone or renin, which increases intracellular calcium concentration in vascular smooth muscle cells, leading to vasoconstriction, which may cause hypertension 56 .Moderate mineral intake during pregnancy may, therefore effectively reduce the risk of PE.This study has some limitations.First, as it is a case-control study, it was not able to discern causality.High-quality, larger randomized controlled trials are needed to evaluate the effects of multiple-mineral intakes, from both the diet and supplements, on the development of PE.Second, the dietary data collected from participants, 3 months before delivery may be subject to recall bias.To minimize the impact of recall bias on the results, face-to-face interviews were conducted by trained researchers and food photographs were used to help Table 3. Risk of PE during pregnancy according to quartiles of dietary minerals intake by excluding participants with gestational diabetes mellitus.OR odds ratios, CI confidence interval, Q quartile.*P < 0.05; **P < 0.01.# Determined by entering the median intake from each quartile as a continuous variable in the regression models.a Adjusted OR from the conditional logistic model adjusted for age, gestational week, pre-pregnancy body mass index, weight gain during pregnancy, educational level, income, parity, adverse pregnancy history, family history of hypertension, physical activity, the use of multivitamin and mineral supplements, daily energy intake and energy-adjusted daily vegetable and fruit intake using the Enter Method.

Figure 1 .
Figure 1.Multivariable-adjusted ORs (solid lines) and 95% CIs (dashed lines) for risk of PE according to dietary intake of calcium (a), magnesium (b), phosphorus (c), iron (d), copper (e), manganese (f) and zinc (g).The model was adjusted for age, gestational week, prepregnancy BMI, weight gain during pregnancy, educational level, income, parity, adverse pregnancy history, family history of hypertension, physical activity, the use of multivitamin and mineral nutritional supplements, daily energy intake and energy-adjusted daily vegetables and fruits intake.OR odds ratios, CI confidence interval.

Table 1 .
Characteristics and energy-adjusted dietary minerals intake of the participants.SD standard deviation, IQR interquartile ranges.*Continuous variables were evaluated using paired Student's t-tests or Wilcoxon rank-sum tests.Categorical variables were evaluated using paired chi-square tests.a P value < 0.05 as significance level.b Intake adjusted for energy.