AJCN. First published ahead of print May 27, 2015 as doi: 10.3945/ajcn.114.098616.

Magnesium supplementation affects metabolic status and pregnancy outcomes in gestational diabetes: a randomized, double-blind, placebo-controlled trial1 Zatollah Asemi,2 Maryam Karamali,3 Mehri Jamilian,3 Fatemeh Foroozanfard,4 Fereshteh Bahmani,2 Zahra Heidarzadeh,2 Sanaz Benisi-Kohansal,6 Pamela J Surkan,5 and Ahmad Esmaillzadeh6* 2

Research Center for Biochemistry and Nutrition in Metabolic Diseases, Kashan University of Medical Sciences, Kashan, Iran; 3Department of Gynecology and Obstetrics, School of Medicine, Arak University of Medical Sciences, Arak, Iran; 4Department of Gynecology and Obstetrics, School of Medicine, Kashan University of Medical Sciences, Kashan, Iran; 5Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; and 6Food Security Research Center and Department of Community Nutrition, School of Nutrition and Food Science, Isfahan University of Medical Sciences, Isfahan, Iran

ABSTRACT Background: To our knowledge, prior research has not examined the effects of magnesium supplementation on metabolic status and pregnancy outcomes in maternal-child dyads affected by gestational diabetes (GDM). Objective: This study was designed to assess the effects of magnesium supplementation on metabolic status and pregnancy outcomes of magnesium-deficient pregnant women with GDM. Design: A randomized double-blind placebo-controlled clinical trial was performed among 70 women with GDM. Patients were randomly assigned to receive either 250 mg magnesium oxide (n = 35) or a placebo (n = 35) for 6 wk. Fasting blood samples were taken at baseline and after a 6-wk intervention. Results: The change in serum magnesium concentration was greater in women consuming magnesium than in the placebo group (+0.06 6 0.3 vs. 20.1 6 0.3 mg/dL, P = 0.02). However, after controlling for baseline magnesium concentrations, the changes in serum magnesium concentrations were not significantly different between the groups. Changes in fasting plasma glucose (29.7 6 10.1 vs. +1.8 6 8.1 mg/dL, P , 0.001), serum insulin concentration (22.1 6 6.5 vs. +5.7 6 10.7 mIU/mL, P = 0.001), homeostasis model of assessment-estimated insulin resistance (20.5 6 1.3 vs. +1.4 6 2.3, P , 0.001), homeostasis model of assessment– estimated b-cell function (24.0 6 28.7 vs. +22.0 6 43.8, P = 0.006), and the quantitative insulin sensitivity check index (+0.004 6 0.021 vs. 20.012 6 0.015, P = 0.005) in supplemented women were significantly different from those of women in the placebo group. Changes in serum triglycerides (+2.1 6 63.0 vs. +38.9 6 37.5 mg/dL, P = 0.005), high sensitivity C-reactive protein (2432.8 6 2521.0 vs. +783.2 6 2470.1 ng/mL, P = 0.03), and plasma malondialdehyde concentrations (20.5 6 1.6 vs. +0.3 6 1.2 mmol/L, P = 0.01) were significantly different between the supplemented women and placebo group. Magnesium supplementation resulted in a lower incidence of newborn hyperbilirubinemia (8.8% vs. 29.4%, P = 0.03) and newborn hospitalization (5.9% vs. 26.5%, P = 0.02). Conclusions: Magnesium supplementation among women with GDM had beneficial effects on metabolic status and pregnancy outcomes. This trial was registered at www.irct.ir as IRCT201503055623N39. Am J Clin Nutr doi: 10.3945/ajcn.114.098616.

Keywords: magnesium, supplementation, gestational diabetes, pregnant women, pregnancy outcomes INTRODUCTION

Gestational diabetes mellitus (GDM),7 defined as impaired carbohydrate metabolism during pregnancy (1), affects 3–10% of pregnant women (2). Increased hormones, including estrogen and progesterone produced during pregnancy, can lead to elevated inflammatory factors and biomarkers of oxidative stress (3, 4), resulting in glucose intolerance and impaired insulin metabolism (5). GDM has important health consequences for the mother and the fetus (6). It is associated with increased subsequent risk of developing diabetes, metabolic syndrome, and cardiovascular events (7). In addition to strategies focused on diet (8, 9), vitamin supplementation (10–12), and pharmaceutical interventions (13), some studies have shown a significant inverse relation between dietary magnesium intake and glucose homeostasis in patients with GDM (2, 14). Bardicef et al. (15) showed that intracellular magnesium depletion can occur in pregnancy, especially in pregnant women affected by GDM. Current data have suggested a beneficial effect of magnesium supplementation on the metabolic status of pregnant women (16) as well as diabetic and hypertensive subjects (17). Rodríguez-Moran and 1 Supported by a grant from the Kashan University of Medical Sciences. The financial support for conception, design, data analysis, and development of the manuscript comes from the Research Center for Biochemistry and Nutrition in Metabolic Diseases, Kashan University of Medical Sciences, Kashan, Iran. 7 Abbreviations used: FPG, fasting plasma glucose; GDM, gestational diabetes mellitus; GSH, total glutathione; HOMA-B, homeostasis model of assessment–estimated b-cell function; hs-CRP, high-sensitivity C-reactive protein; NO, nitric oxide; QUICKI, quantitative insulin sensitivity check index; TAC, total antioxidant capacity; T2DM, type 2 diabetes mellitus. *To whom correspondence should be addressed. E-mail: Esmaillzadeh@ hlth.mui.ac.ir. Received August 28, 2014. Accepted for publication April 16, 2015. doi: 10.3945/ajcn.114.098616.

Am J Clin Nutr doi: 10.3945/ajcn.114.098616. Printed in USA. Ó 2015 American Society for Nutrition

Copyright (C) 2015 by the American Society for Nutrition

1 of 8

2 of 8

ASEMI ET AL.

Guerrero-Romero (18) reported that consumption of 382 mg magnesium per day resulted in a significant decrease in fasting glucose, serum triglyceride concentrations, and HOMA-IR in normal-weight subjects; however, no significant effect was observed on HDL cholesterol concentrations. In another study among healthy middle-aged overweight women with adequate magnesium concentrations, administration of 250 mg magnesium as magnesium oxide per day did not influence inflammatory factors (19). Metabolic abnormalities found in gestational diabetes are similar to those of metabolic syndrome in nonpregnant individuals. Magnesium is an essential cofactor in the enzymatic process of high-energy phosphate (20), acts as a calcium channel antagonist, and stimulates production of prostacyclins and nitric oxide (NO) (21). Magnesium seems to have anti-inflammatory effects due to its antagonism to calcium that play an important role in inflammation as well as in protein synthesis and transmembrane ion transport (22). Considering inadequate dietary magnesium intake among pregnant women (23) as well as a relatively high prevalence of hypomagnesemia in the urban Iranian population (24), we hypothesized that magnesium supplementation might help patients with GDM to control their metabolic profile and pregnancy outcomes. The objective of this study was to examine the effects of magnesium supplementation on metabolic status and pregnancy outcomes of pregnant women with GDM.

METHODS

Participants This randomized, double-blind, placebo-controlled, parallel clinical trial was done in Kashan, Iran, from February 2014 to July 2014 in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. The Ethics Committee of Arak University of Medical Sciences reviewed and approved the study protocol (registered in the Iranian registry of clinical trials as IRCT201503055623N39). Written informed consent was obtained from all participants before the intervention. To estimate the sample size, we used a randomized clinical trial sample size formula in which type I (a) and type II errors (b) were 0.05 and 0.20 (power = 80%), respectively. On the basis of a previous study (10), we used an SD of 2.04 and a difference in mean (d) of 1.50, considering HOMA-IR as the key variable. The calculation indicated 30 subjects were needed in each group. Assuming a dropout of 5 participants per group, the final sample size was determined to be 35 patients per group. We chose HOMA-IR to estimate sample size because it was the most important variable under consideration in patients with GDM. Furthermore, the largest sample size was obtained when we used this variable. Eligible participants were aged 18–40 y (at weeks 24–28 of gestation) who were diagnosed with GDM by a “onestep” 2-h 75-g oral glucose tolerance test. Gestational age was assessed from the date of last menstrual period and concurrent clinical assessment (25). A diagnosis of GDM was based on the American Diabetes Association criteria (26). Women whose plasma glucose met one of the following criteria were diagnosed with GDM: fasting, $92 mg/dL; 1 h, $180 mg/dL; or 2 h, $153 mg/dL. After screening 950 pregnant women in the Naghavi maternity clinic affiliated with Kashan University of Medical Sciences, Kashan, Iran, 70 subjects were eligible for enrollment

[865 women were excluded for not having GDM and 15 women were excluded after being diagnosed with GDM class A2 and needed insulin therapy: fasting plasma glucose (FPG) .105 mg/ dL and blood sugar 2 h postprandial .120 mg/dL). These 70 pregnant women were randomly assigned to either magnesium supplementation (n = 35) or a placebo (n = 35) for 6 wk. Before random assignment, patients were stratified based on BMI (in kg/ m2; ,30 and $30) and weeks of gestation (,26 or $26 wk). Study design Participants were randomly assigned to consume either 250 mg magnesium supplements/d as magnesium oxide or a placebo for 6 wk. Although the duration of the intervention was 6 wk, we followed patients until delivery. However, we did not continue magnesium supplementation until delivery because assessment of compliance to supplements through blood samples was very difficult in the final weeks of pregnancy. Furthermore, because we wanted to examine the effects of supplementation on metabolic profiles, we were concerned that a large number of women would drop out if they were asked to provide blood near the end of their pregnancies. Magnesium supplements and the placebo were manufactured by 21st Century Pharmaceutical Company and Barij Essence Pharmaceutical Company, respectively. All magnesium and placebo tablets were provided by Barij Essence Pharmaceutical Company in prepackaged bottles that were numbered for each patient according to a randomization sequence. Each patient was assigned to a number to establish participant order and received the magnesium or placebo in the corresponding prepackaged bottles. The placebo and magnesium tablets were the same in size, weight, color, and taste. All patients, clinical investigators (including the person administering demographic and food recall questionnaires), and other health care personnel were blinded to treatment assignment. Participants were asked not to alter their routine physical activity or usual dietary intakes during the study and not to consume any supplements other than those provided to them by the investigators. All pregnant women also consumed 400 mg folic acid/d starting at the beginning of pregnancy and 60 mg ferrous sulfate/d as of the second trimester. Compliance to the magnesium supplementation was assessed through quantification of serum magnesium concentrations. The use of magnesium supplementation and the placebo during the study was checked by asking participants to return the medication containers. To increase compliance, all patients received brief daily cell phone reminders to take the supplements. All participants provided 3 dietary recalls (once during the weekend and on 2 weekdays) and 3 physical activity records to verify that they maintained their usual diet and physical activity during the intervention. Both dietary recalls and physical activity records were taken at weeks 2, 4, and 6 of the intervention. To obtain information on participant nutrient intake based on these 3-d food diaries, we used Nutritionist IV software (First Databank) modified for Iranian foods. Assessment of anthropometric variables Data on prepregnancy weight and height (measured values) were obtained from the women’s clinical records. A trained midwife at the maternity clinic took anthropometric measurements at baseline and 6 wk after the intervention. Height and weight (Seca) were measured while the participants wore light

MAGNESIUM SUPPLEMENTATION AND GDM

clothing and no shoes. BMI was calculated as weight (in kilograms) divided by height (in meters) squared. Infant length and weight were measured by using standard methods (Seca 155 Scale) during the first 24 h after birth and were recorded to the nearest 1 mm and 10 g, respectively. Infant head circumference was measured to the nearest 1 mm with a Seca girth measuring tape. We also collected data on infants’ 1- and 5-min Apgar scores. Macrosomic babies were defined as those whose birth weight was .4000 g (27). Biochemical and polyhydramnios assessment Before the onset and after the end of intervention, 10-mL blood samples were taken from each patient at the Kashan reference laboratory. Blood was collected in 2 separate tubes: 1) one without EDTA to separate the serum, to quantify serum magnesium, insulin, lipid profiles, and high-sensitivity C-reactive protein (hs-CRP) concentrations and 2) another one containing EDTA to examine plasma NO and biomarkers of oxidative stress. FPG concentrations were measured on the day blood was collected. Blood samples were immediately centrifuged (D78532; Hettich) at 1465 3 g for 10 min to separate the serum. Serum lipid profiles were also quantified on the day of blood collection. The samples were then stored at 2708C until analyzed at the Kashan University of Medical Sciences reference laboratory. Commercial kits were used to measure serum magnesium, FPG, serum cholesterol, triglycerides, and VLDL, LDL, and HDL cholesterol concentrations (Pars Azmun). All interand intra-assay CVs for magnesium, FPG, and lipid profile measurements were ,5%. Serum insulin concentrations were assayed by ELISA (Monobind). The intra- and interassay CVs for serum insulin were 2.9% and 5.8%, respectively. HOMA-IR and homeostasis model of assessment–estimated b-cell function (HOMA-B) and quantitative insulin sensitivity check index

FIGURE 1

3 of 8

(QUICKI) were calculated based on suggested formulas (28). Serum hs-CRP was quantified by using an ELISA kit (LDN) with intra- and interassay CVs of 2.8% and 4.4%, respectively. The plasma NO concentration was determined by the Griess method (29). Plasma total antioxidant capacity (TAC) was assessed by the use of the ferric reducing antioxidant power method developed by Benzie and Strain (30). Plasma total glutathione (GSH) was examined by using the method of Beutler and Gelbart (31). The plasma malondialdehyde concentrations were determined by the thiobarbituric acid reactive substance spectrophotometric test (32). CVs for plasma TAC, GSH, and malondialdehyde were 0.8%, 2.6%, and 3.4%, respectively. Hyperbilirubinemia was defined as total serum bilirubin concentrations $15 mg/dL (257 mol/L) in infants 25–48 h old, 18 mg/dL (308 mol/L) in infants 49–72 h old, and 20 mg/dL (342 mol/L) in infants older than 72 h (33). Polyhydramnios was diagnosed by using sonographic estimation at postintervention. On the basis of this measurement, polyhydramnios was defined as an amniotic fluid index in excess of 25 cm (34). Statistical analysis We used the Kolmogorov-Smirnov test to examine if variables were normally distributed. Intention-to-treat analysis of the primary study endpoint was performed for all the randomly assigned participants. To determine the effects of magnesium supplementation on insulin metabolism, lipid profiles, inflammatory factors, and biomarkers of oxidative stress, we used mixed-model repeated-measures ANOVA. Because the mixedmodel analysis without any ad hoc imputation has been shown to provide equal or more power than does analysis using mixed models with missing values imputed by ad hoc imputation methods, we did not impute missing values by considering the assumption that these values are missing at random (35). To

Summary of patient flow diagram.

4 of 8

ASEMI ET AL.

assess if the magnitude of the change in dependent variables depended on the baseline maternal age and weight, we controlled all analyses for baseline values of maternal age and weight to avoid potential bias. To identify the effect of magnesium supplementation on pregnancy outcomes of maternal-child dyads, we applied a x2 test for categorical variables and 1-factor ANOVA for continuous variables (newborns’ weight, length, and head circumference and Apgar score). To examine if prepregnancy BMI, maternal FPG at baseline, and maternal age influenced these findings, we applied ANCOVA controlling for these variables. P , 0.05 was considered statistically significant. All statistical analyses were done by using the Statistical Package for Social Sciences version 17 (SPSS, Inc.).

RESULTS

As demonstrated in the study flow diagram (Figure 1), during the intervention phase of the study, 3 patients were excluded from the magnesium group [withdrawn due to personal reasons (n = 2) and hospitalization (n = 1)] and 3 [withdrawn due to personal reasons (n = 2) and insulin therapy (n = 1)] from the placebo group. Finally, 64 participants [magnesium (n = 32) and placebo (n = 32)] completed the trial. However, because the analysis was based on the intention-to-treat principle, all 70 women (35 in each group) were included in the final analysis. On average, the compliance rate in our study was high, such that 100% of capsules were taken during the course of the study in both groups. No side effects were reported after consumption of magnesium supplements in patients with GDM or in their infants. Participants’ mean age, prepregnancy weight, and BMI were 29.3 6 3.9 y, 70.1 6 11.5 kg, and 27.2 6 4.0, respectively. Baseline demographic characteristics of patients in the 2 groups were not clinically and statistically different (Table 1). Based on the 3-d dietary recalls obtained throughout the intervention, no significant differences were observed between the 2 groups in terms of dietary intakes of energy, carbohydrates, proteins, fats, SFAs, PUFAs, MUFAs, cholesterol, crude fiber, total dietary fiber, magnesium, and manganese (Table 2). The change in serum magnesium concentrations after 6 wk of supplementation was greater in magnesium-supplemented women compared with women in the placebo group (+0.06 6 TABLE 1 General characteristics of pregnant women with GDM who received either magnesium supplements or a placebo1 Placebo group (n = 35) Maternal age, y Height, cm Prepregnancy weight,2 kg Weight at study baseline, kg Weight at end of trial, kg Weight change, kg Prepregnancy BMI,2 kg/m2 BMI at study baseline, kg/m2 BMI at end of trial, kg/m2 BMI change, kg/m2

29.4 160.6 70.3 73.1 75.1 2.1 27.2 28.3 29.1 0.8

6 6 6 6 6 6 6 6 6 6

3.1 3.9 8.1 9.5 9.4 1.1 2.9 3.5 3.5 0.4

Magnesium group (n = 35) 29.1 160.1 70.0 73.6 76.0 2.4 27.3 28.7 29.6 0.9

6 6 6 6 6 6 6 6 6 6

4.6 7.1 14.2 14.8 12.4 2.3 4.9 5.3 5.4 0.9

Values are means 6 SDs. GDM, gestational diabetes mellitus. Based on participants’ measured weight and height from maternity clinic records. 1 2

TABLE 2 Dietary intake of pregnant women with GDM who received either magnesium supplements or a placebo1 Placebo group (n = 35) Energy, kcal/d Carbohydrates, g/d Protein, g/d Fat, g/d SFAs, g/d PUFAs, g/d MUFAs, g/d Cholesterol, mg/d Crude fiber, g/d TDF, g/d Magnesium, mg/d Manganese, mg/d

2381 321.6 86.3 86.8 25.4 27.5 24.3 237.3 5.5 18.0 282.1 2.3

6 6 6 6 6 6 6 6 6 6 6 6

194 45.2 19.8 17.8 7.0 6.1 7.7 152.7 1.7 4.7 72.3 0.7

Magnesium group (n = 35) 2451 335.7 86.6 88.9 25.3 29.7 23.3 208.3 5.6 18.8 301.5 2.3

6 6 6 6 6 6 6 6 6 6 6 6

186 42.2 9.2 13.8 5.3 8.1 4.8 116.0 1.6 4.8 76.0 0.8

1 Values are means 6 SDs. There were no significant differences between the groups. GDM, gestational diabetes mellitus; TDF, total dietary fiber.

0.31 vs. 20.11 6 0.29 mg/dL, P = 0.02) (Table 3). In addition, magnesium-supplemented women had a significant reduction in FPG (29.7 6 10.1 vs. +1.8 6 8.1 mg/dL, P , 0.001), serum insulin concentrations (22.1 6 6.5 vs. +5.7 6 10.7 mIU/mL, P = 0.001), HOMA-IR (20.5 6 1.3 vs. +1.4 6 2.3, P , 0.001), and HOMA-B (24.0 6 28.7 vs. +22.0 6 43.8, P = 0.006) and an increase in QUICKI (+0.004 6 0.021 vs. 20.012 6 0.015, P = 0.005) compared with women in the placebo group. Changes in serum triglycerides (+2.1 6 63.0 vs. +38.9 6 37.5 mg/dL, P = 0.005), serum hs-CRP (2432.8 6 2521.0 vs. +783.2 6 2470.1 ng/ mL, P = 0.03) and plasma malondialdehyde concentrations (20.5 6 1.6 vs. +0.3 6 1.2 mmol/L, P = 0.01) differed significantly between the 2 groups. We also observed a trend toward a significant effect of magnesium supplementation on increasing plasma NO concentrations (+15.9 6 37.0 vs. +1.1 6 39.0 mmol/L, P = 0.05). There were no significant differences between the magnesium and placebo groups in terms of changes in other lipid profiles and plasma TAC and GSH concentrations. Baseline concentrations of magnesium differed significantly between the 2 groups. Therefore, baseline concentrations were controlled for in the analyses. However, after this adjustment, no substantial changes in our findings occurred, except for magnesium (P = 0.89) and total cholesterol concentrations (P = 0.01). Additional adjustments for age and baseline weight did not affect our findings, except for serum magnesium (P = 0.89) and total cholesterol concentrations (P = 0.01) (Table 4). Taking magnesium supplements resulted in a lower incidence of newborn hyperbilirubinemia (8.8% vs. 29.4%, P = 0.03) and lower newborn hospitalization rate (5.9% vs. 26.5%, P = 0.02) (Table 5). We did not observe significant differences in the cesarean delivery rate, need for insulin therapy after the intervention, polyhydramnios, maternal hospitalization, preterm delivery, gestational age, newborn birth size, Apgar score, and newborn hypoglycemia between the 2 groups. Adjustment for prepregnancy BMI, baseline maternal FPG, and age did not alter the findings.

DISCUSSION

On the basis of serum magnesium concentrations at baseline, all patients with GDM in our study were magnesium deficient.

5 of 8

MAGNESIUM SUPPLEMENTATION AND GDM

TABLE 3 Metabolic profiles, inflammatory factors, and biomarkers of oxidative stress at baseline and after a 6-wk intervention in pregnant women with GDM who received either magnesium supplements or a placebo1 Placebo group (n = 35) Week 0

Week 6

Magnesium group (n = 35) Change

Week 0

Week 6

Change

Magnesium, mg/dL 1.62 6 0.33 1.51 6 0.33* 20.11 6 0.29 1.32 6 0.34** 1.35 6 0.31 0.06 6 0.31 FPG, mg/dL 91.4 6 9.0 93.7 6 11.0 1.8 6 8.1 95.1 6 12.6 85.8 6 6.1* 29.7 6 10.1 Insulin, mIU/mL 14.1 6 5.7 19.9 6 12.6* 5.7 6 10.7 13.1 6 7.1 10.7 6 3.1 22.1 6 6.5 HOMA-IR 3.2 6 1.6 4.7 6 3.1* 1.4 6 2.3 3.1 6 1.8 2.7 6 0.6* 20.5 6 1.3 HOMA-B 52.0 6 20.8 74.1 6 50.8* 22.0 6 43.8 46.5 6 31.3 42.4 6 14.1 24.0 6 28.7 QUICKI 0.351 6 0.022 0.338 6 0.028* 20.012 6 0.015 0.343 6 0.023 0.349 6 0.019 0.004 6 0.021 6.3 6 20.6 188.1 6 42.8 182.2 6 31.8 23.8 6 33.0 Total cholesterol, mg/dL 198.8 6 51.4 204.1 6 44.8 Triglycerides, mg/dL 166.5 6 73.7 201.5 6 91.2* 38.9 6 37.5 173.1 6 97.9 171.0 6 96.7 2.1 6 63.0 VLDL cholesterol, mg/dL 33.1 6 14.5 39.9 6 18.3* 6.3 6 7.0 34.1 6 19.4 33.9 6 18.8 0.51 6 11.7 LDL cholesterol, mg/dL 108.1 6 37.4 107.2 6 31.6 21.5 6 15.0 98.8 6 31.1 96.9 6 25.3 23.2 6 22.3 HDL cholesterol, mg/dL 58.1 6 15.7 59.5 6 11.7 0.8 6 6.0 53.1 6 12.0 54.7 6 9.2 1.7 6 6.0 Total/HDL cholesterol ratio 3.4 6 0.8 3.8 6 0.8 0.3 6 0.5 3.7 6 0.8 3.4 6 0.6 20.2 6 0.6 hs-CRP, ng/mL 6101.1 6 3715.0 6881.1 6 4007.0 783.2 6 2470.1 5731.3 6 3932.2 5305.5 6 4091.0 2432.8 6 2521.0 NO, mmol/L 101.1 6 36.4 1.1 6 39.0 100.2 6 27.7 116.1 6 36.8* 15.9 6 37.0 101.7 6 54.6 TAC, mmol/L 715.1 6 155.8 722.2 6 134.5 5.4 6 82.8 710.4 6 151.2 777.5 6 466.4 67.3 6 456.8 GSH, mmol/L 470.1 6 238.7 469.4 6 173.5 22.1 6 118.0 511.3 6 231.5 431.5 6 71.9 280.2 6 233.7 MDA, mmol/L 3.4 6 1.4 3.7 6 1.5 0.3 6 1.2 4.0 6 1.1 3.5 6 1.5 20.5 6 1.6

P value2 0.02 ,0.001 0.001 ,0.001 0.006 0.005 0.10 0.005 0.005 0.41 0.61 0.09 0.03 0.05 0.39 0.07 0.01

Values are means 6 SDs. Values in this table were not adjusted for baseline concentrations, maternal age, and baseline weight. Baseline values of magnesium differed between the 2 groups. *Significantly different from baseline (P , 0.05). **Significantly different from placebo group at week 0 (P , 0.05). FPG, fasting plasma glucose; GDM, gestational diabetes mellitus; GSH, total glutathione; HOMA-B, homeostasis model of assessment–estimated b-cell function; hs-CRP, high-sensitivity Creactive protein; MDA, malondialdehyde; NO, nitric oxide; QUICKI, quantitative insulin sensitivity check index; TAC, total antioxidant capacity. 2 P values represent the time 3 group interaction (computed by mixed-model repeated-measures ANOVA). 1

Magnesium supplementation for 6 wk significantly decreased FPG, insulin, HOMA-IR, HOMA-B, hs-CRP, and malondialdehyde and increased QUICKI compared with the placebo. In interpreting the results, it is important to keep in mind that women participating in the study were magnesium deficient, explaining why we found an increase in serum magnesium concentrations in the supplementation compared with the placebo group. However, after controlling for baseline magnesium concentrations, the changes in serum magnesium concentrations were not significantly different. This implies that baseline serum magnesium concentrations were involved in the response to magnesium supplementation. It is also important to note that serum magnesium concentrations do not thoroughly reflect dietary or supplemental magnesium intake. However, we were not able to assess intracellular magnesium concentrations in the current study. Some investigators have also recommended the use of erythrocyte magnesium content to assess dietary intake. Others have shown that the magnesium content of white blood cells is a better index of intracellular magnesium in skeletal and cardiac muscle (36). Patients with GDM are susceptible to metabolic abnormalities, inflammation, and oxidative stress (11). Our findings showed that the administration of magnesium supplements for 6 wk in women with GDM resulted in a significant decrease in FPG, serum insulin concentrations, HOMA-IR, and HOMA-B and a significant rise in QUICKI compared with placebo. A growing body of evidence has suggested an inverse association between dietary magnesium intake (37) and serum magnesium concentrations (38) and the risk of developing insulin resistance and type 2 diabetes mellitus (T2DM). Consistent with our study, reduced HOMA-IR has been observed after intake of 2.5 g magnesium chloride/d (39). It has been reported that the acetyl-CoA carboxylase enzyme that catalyzes the formation

of malonyl-CoA, which is implicated in physiologic insulin secretion, is stimulated via magnesium in a concentration-dependent TABLE 4 Adjusted changes in metabolic variables of pregnant women with GDM who received either magnesium supplements or a placebo1 Placebo group (n = 35)

Magnesium group (n = 35) P value2

Magnesium, mg/dL 20.03 6 0.04 20.03 6 0.04 FPG, mg/dL 0.4 6 1.2 27.8 6 1.4 Insulin, mIU/mL 6.3 6 1.4 22.5 6 1.4 HOMA-IR 1.7 6 0.4 20.9 6 0.4 HOMA-B 21.0 6 5.9 24.9 6 5.9 QUICKI 20.018 6 0.003 0.008 6 0.003 Total cholesterol, mg/dL 9.1 6 4.3 24.9 6 4.2 Triglycerides, mg/dL 34.8 6 8.0 2.0 6 8.1 VLDL cholesterol, mg/dL 7.7 6 1.4 0.4 6 1.4 LDL cholesterol, mg/dL 1.6 6 3.1 26.1 6 3.1 HDL cholesterol, mg/dL 0.6 6 0.7 21.0 6 0.4 Total/HDL cholesterol ratio 0.1 6 0.1 20.1 6 0.1 hs-CRP, ng/mL 834.8 6 411.0 2482.2 6 414.7 NO, mmol/L 21.9 6 6.0 16.9 6 6.1 TAC, mmol/L 22.1 6 53.1 67.0 6 52.1 GSH, mmol/L 217.5 6 17.7 266.0 6 17.7 MDA, mmol/L 0.5 6 0.3 20.2 6 0.2

0.89 ,0.001 ,0.001 ,0.001 0.001 0.001 0.01 0.005 0.005 0.10 0.55 0.06 0.02 0.05 0.28 0.05 0.04

1 Values are means 6 SEs. Values are adjusted for baseline values, maternal age, and baseline weight. FPG, fasting plasma glucose; GDM, gestational diabetes mellitus; GSH, total glutathione; HOMA-B, homeostasis model of assessment–estimated b-cell function; hs-CRP, high-sensitivity Creactive protein; MDA, malondialdehyde; NO, nitric oxide; QUICKI, quantitative insulin sensitivity check index; TAC, total antioxidant capacity. 2 Obtained from mixed-model repeated-measure ANOVA (time 3 group interaction).

6 of 8

ASEMI ET AL. TABLE 5 Associations between magnesium supplementation and pregnancy outcomes1 Placebo group (n = 34) Cesarean delivery, n (%) Need to insulin therapy after intervention, n (%) Preeclampsia, n (%) Polyhydramnios, n (%) Maternal hospitalization, n (%) Preterm delivery, n (%) Macrosomia .4000 g, n (%) Gestational age, wk Newborn weight, g Crude Model 14 Model 25 Newborn length, cm Crude Model 14 Model 25 Newborn head circumference, cm Crude Model 14 Model 25 1-min Apgar score Crude Model 14 Model 25 5-min Apgar score Crude Model 14 Model 25 Newborn hyperbilirubinemia, n (%) Newborn hospitalization, n (%) Newborn hypoglycemia, n (%)

Magnesium group (n = 34)

P value

11 (32.4) 1 (2.9) 1 (2.9) 2 (5.9) 1 (2.9) 2 (5.9) 2 (5.9) 39.1 6 0.2

0.212 0.552 0.312 0.552 0.552 0.552 0.232 0.252

3417.0 6 91.2 3421.2 6 91.0 3412.6 6 91.2

3208.1 6 91.2 3209.5 6 90.8 3214.5 6 90.9

0.07 0.08 0.13

50.6 6 0.4 50.6 6 0.4 50.6 6 0.4

50.9 6 0.4 50.9 6 0.4 51.0 6 0.4

0.61 0.60 0.51

35.1 6 0.3 35.1 6 0.3 35.0 6 0.3

35.3 6 0.3 35.3 6 0.3 35.3 6 0.3

0.61 0.62 0.56

16 (47.1) 2 (5.9) 0 (0) 1 (2.9) 2 (5.9) 1 (2.9) 5 (14.7) 38.7 6 0.23

8.8 6 0.03 8.9 6 0.03 9.0 6 0.03

8.8 6 0.03 9.0 6 0.03 9.0 6 0.03

0.99 0.95 0.91

9.9 6 0.03 9.9 6 0.03 9.9 6 0.03 10 (29.4) 9 (26.5) 1 (2.9)

9.9 6 0.03 9.9 6 0.03 9.9 6 0.03 3 (8.8) 2 (5.9) 2 (5.9)

1 0.99 0.96 0.032 0.022 0.552

1

We excluded one patient and her newborn in each group due to intrauterine fetal death. Obtained from x2 test. 3 Mean 6 SE (all such values). 4 Obtained from ANCOVA adjusted for prepregnancy BMI. 5 Obtained from ANCOVA adjusted for prepregnancy BMI, maternal fasting plasma glucose at baseline, and maternal 2

age.

manner (40). In addition, magnesium may competitively inhibit the voltage-dependent calcium channel, which is known to play a role in insulin secretion (41). Our study showed that in patients with GDM, magnesium supplement intake led to a significant difference in changes in serum triglycerides and VLDL cholesterol concentrations but did not influence other serum lipid profiles. In line with our findings, a significant reduction in serum triglyceride concentrations was observed after intake of magnesium supplements (18, 42). In addition, in a crossover study in healthy participants, taking magnesium oxide reduced total and LDL cholesterol concentrations (43). However, magnesium supplementation has been found to not affect lipid profiles among patients with T2DM (44). Furthermore, no significant change in lipid concentrations was seen after intake of 384 mg magnesium chloride/d among patients with T2DM (45). This might be explained by the near-normal concentrations of serum magnesium in the participants of that study (44). Magnesium might suppress postprandial hyperlipidemia through promoting the formation of insoluble compounds and the excretion of fat (46). Furthermore, VLDL cholesterol lipolysis is facilitated by magnesium because this nutrient is known to be a cofactor for lipoprotein lipase (47).

Findings from the current study showed that magnesium supplementation in patients with GDM led to greater decreases in serum hs-CRP concentrations compared with the placebo. Although magnesium supplementation resulted in a significant increase in plasma NO concentrations, the changes were not significantly different from those of the placebo group. In agreement with our results, Nielsen et al. (48) found that taking magnesium supplements improved indicators of inflammation in elderly people. However, cyclosporine and magnesium supplementation did not change NO concentrations in rats (49). In addition, research has shown that magnesium supplementation does not significantly attenuate inflammatory markers (19, 50). Anti-inflammatory effects of magnesium may be mediated via its antagonism to calcium, the ion playing an important role in inflammation (51). Inactivation of N-methyl-D-aspartate receptors and the inhibition of nuclear transcription factor kB by magnesium have been considered potential mechanisms leading to a decreased inflammatory response (52). We found that magnesium supplementation was associated with a significant reduction in plasma malondialdehyde concentrations but did not affect plasma TAC and GSH concentrations. Supporting our results, supplementation with vitamin

MAGNESIUM SUPPLEMENTATION AND GDM

combinations (vitamins C and E) and minerals (magnesium and zinc) has led to decreased malondialdehyde concentrations among patients with T2DM (53). These findings were also reported by other investigators (54). Magnesium supplementation increased GSH concentrations among atopic asthmatic children (55). It has been suggested that when the concentrations of free radicals are high, including GDM, magnesium supplementation might reduce free radical production and lipid hydroperoxides through decreasing reactive oxygen species production (56) and increasing glutathione peroxidase activity (57). Our study demonstrated that magnesium supplementation in patients with GDM was associated with a decreased incidence of newborn hyperbilirubinemia and hospitalizations but did not affect other pregnancy outcomes. In line with our study, Misra et al. (58) indicated that serum total magnesium concentrations in 30 newborns with hemolytic hyperbilirubinemia were significantly lower than those of healthy subjects. However, others observed no significant association of low dietary intake of magnesium with preeclampsia, small-for-gestational age infants, or preterm labor (59). However, Pintov et al. (60) showed no significant association between umbilical cord magnesium concentrations and newborns’ bilirubin concentrations in the 48th hour of life. It is speculated that gestational malnutrition may cause maternal and neonatal hypomagnesemia by negatively affecting enzymes in bilirubin metabolism and antioxidant enzymes in erythrocytes, thus leading to newborns’ indirect hyperbilirubinemia (61). To interpret our findings, some limitations need to be taken into account. Due to limited funding, we did not examine the effects of magnesium supplementation on other biomarkers of systemic inflammation or oxidative stress. The use of serum magnesium concentrations to measure compliance, instead of intracellular or urinary concentrations, may have led to less accurate assessment (62). The dosage of magnesium supplements that can influence serum or intracellular magnesium concentrations must also be examined in future studies. Some investigators have used higher dosages of supplemental magnesium than those used in this study to increase plasma/serum magnesium concentrations (63). However, supplements of #450 mg/d could result in increased serum magnesium concentrations in magnesium-deficient subjects (44). Furthermore, we demonstrated that 250 mg magnesium oxide/d for 6 wk resulted in increased serum magnesium concentrations by 0.06 mg/dL. Future studies are needed to further confirm these findings. Although we controlled for several confounders in the analysis, it must be kept in mind that even without any adjustment, the assessment of numerous outcomes in the current study inflates the type I error. In addition, multiple-comparison issues associated with comparing groups on a large number of outcomes also must be taken into account. In conclusion, magnesium supplementation in pregnant women with GDM was associated with decreased insulin, hsCRP, malondialdehyde concentrations, newborn hyperbilirubinemia, and hospitalization but did not influence other lipid profiles, NO, TAC, GSH, or other pregnancy outcomes. The authors’ responsibilities were as follows—ZA and AE: contributed in conception, design, statistical analysis, and drafting of the manuscript; MK, MJ, FF, FB, ZH, and SB-K: contributed in data collection and manuscript drafting; PJS: contributed to editing of the manuscript; AE: supervised the study. All authors read and approved the final version of the paper. None of the authors declared any personal or financial conflict of interest.

7 of 8

REFERENCES 1. Nolan CJ. Controversies in gestational diabetes. Best Pract Res Clin Obstet Gynaecol 2011;25:37–49. 2. Nair AV, Hocher B, Verkaart S, van Zeeland F, Pfab T, Slowinski T, Chen YP, Schlingmann KP, Schaller A, Gallati S, et al. Loss of insulin-induced activation of TRPM6 magnesium channels results in impaired glucose tolerance during pregnancy. Proc Natl Acad Sci USA 2012;109:11324–9. 3. Volpe L, Di Cianni G, Lencioni C, Cuccuru I, Benzi L, Del Prato S. Gestational diabetes, inflammation, and late vascular disease. J Endocrinol Invest 2007;30:873–9. 4. Herrera E, Ortega-Senovilla H. Disturbances in lipid metabolism in diabetic pregnancy—are these the cause of the problem? Best Pract Res Clin Endocrinol Metab 2010;24:515–25. 5. Reece EA, Leguizamon G, Wiznitzer A. Gestational diabetes: the need for a common ground. Lancet 2009;373:1789–97. 6. Nielsen KK, Kapur A, Damm P, de Courten M, Bygbjerg IC. From screening to postpartum follow-up—the determinants and barriers for gestational diabetes mellitus (GDM) services, a systematic review. BMC Pregnancy Childbirth 2014;14:41. 7. Brewster S, Zinman B, Retnakaran R, Floras JS. Cardiometabolic consequences of gestational dysglycemia. J Am Coll Cardiol 2013;62: 677–84. 8. Louie JC, Markovic TP, Perera N, Foote D, Petocz P, Ross GP, BrandMiller JC. A randomized controlled trial investigating the effects of a low-glycemic index diet on pregnancy outcomes in gestational diabetes mellitus. Diabetes Care 2011;34:2341–6. 9. Asemi Z, Samimi M, Tabassi Z, Sabihi SS, Esmaillzadeh A. A randomized controlled clinical trial investigating the effect of DASH diet on insulin resistance, inflammation, and oxidative stress in gestational diabetes. Nutrition 2013;29:619–24. 10. Asemi Z, Hashemi T, Karamali M, Samimi M, Esmaillzadeh A. Effects of vitamin D supplementation on glucose metabolism, lipid concentrations, inflammation, and oxidative stress in gestational diabetes: a double-blind randomized controlled clinical trial. Am J Clin Nutr 2013;98:1425–32. 11. Asemi Z, Karamali M, Esmaillzadeh A. Effects of calcium–vitamin D co-supplementation on glycaemic control, inflammation and oxidative stress in gestational diabetes: a randomised placebo-controlled trial. Diabetologia 2014;57:1798–806. 12. Samimi M, Jamilian M, Asemi Z, Esmaillzadeh A. Effects of omega-3 fatty acid supplementation on insulin metabolism and lipid profiles in gestational diabetes: randomized, double-blind, placebo-controlled trial. Clin Nutr 2014 Jun 17 (Epub ahead of print; DOI: 10.1016/ j.clnu.2014.06.005). 13. Gui J, Liu Q, Feng L. Metformin vs insulin in the management of gestational diabetes: a meta-analysis. PLoS ONE 2013;8:e64585. 14. Del Gobbo LC, Song Y, Elin RJ, Meltzer SJ, Egeland GM. Gestational glucose intolerance modifies the association between magnesium and glycemic variables in mothers and daughters 15 years post-partum. Magnes Res 2012;25:54–63. 15. Bardicef M, Bardicef O, Sorokin Y, Altura BM, Altura BT, Resnick LM. Perinatal cellular ion metabolism: 31P-nuclear magnetic resonance spectroscopic analysis of intracellular free magnesium and pH in maternal and cord blood erythrocytes. J Soc Gynecol Investig 1996;3: 66–70. 16. Bullarbo M, Odman N, Nestler A, Nielsen T, Kolisek M, Vormann J, Rylander R. Magnesium supplementation to prevent high blood pressure in pregnancy: a randomised placebo control trial. Arch Gynecol Obstet 2013;288:1269–74. 17. Barbagallo M, Dominguez LJ, Galioto A, Pineo A, Belvedere M. Oral magnesium supplementation improves vascular function in elderly diabetic patients. Magnes Res 2010;23:131–7. 18. Rodríguez-Moran M, Guerrero-Romero F. Oral magnesium supplementation improves the metabolic profile of metabolically obese, normal-weight individuals: a randomized double-blind placebocontrolled trial. Arch Med Res 2014;45:388–93. 19. Moslehi N, Vafa M, Rahimi-Foroushani A, Golestan B. Effects of oral magnesium supplementation on inflammatory markers in middle-aged overweight women. J Res Med Sci 2012;17:607–14. 20. Barbagallo M, Dominguez LJ, Galioto A, Ferlisi A, Cani C, Malfa L, Pineo A, Busardo A, Paolisso G. Role of magnesium in insulin action, diabetes and cardio-metabolic syndrome X. Mol Aspects Med 2003;24:39–52. 21. Sontia B, Touyz RM. Role of magnesium in hypertension. Arch Biochem Biophys 2007;458:33–9.

8 of 8

ASEMI ET AL.

22. Almoznino-Sarafian D, Berman S, Mor A, Shteinshnaider M, Gorelik O, Tzur I, Alon I, Modai D, Cohen N. Magnesium and C-reactive protein in heart failure: an anti-inflammatory effect of magnesium administration? Eur J Nutr 2007;46:230–7. 23. Esmaillzadeh A, Samareh S, Azadbakht L. Dietary patterns among pregnant women in the west-north of Iran. Pak J Biol Sci 2008;11:793–6. 24. Syedmoradi L, Ghasemi A, Zahediasl S, Azizi F. Prevalence of hypoand hypermagnesemia in an Iranian urban population. Ann Hum Biol 2011;38:150–5. 25. Gupta P, Narang M, Banerjee BD, Basu S. Oxidative stress in term small for gestational age neonates born to undernourished mothers: a case control study. BMC Pediatr 2004;4:14. 26. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2014;37(Suppl 1):S81–90. 27. Boulet SL, Alexander GR, Salihu HM, Pass M. Macrosomic births in the United States: determinants, outcomes, and proposed grades of risk. Am J Obstet Gynecol 2003;188:1372–8. 28. Pisprasert V, Ingram KH, Lopez-Davila MF, Munoz AJ, Garvey WT. Limitations in the use of indices using glucose and insulin levels to predict insulin sensitivity: impact of race and gender and superiority of the indices derived from oral glucose tolerance test in African Americans. Diabetes Care 2013;36:845–53. 29. Tatsch E, Bochi GV, Pereira Rda S, Kober H, Agertt VA, de Campos MM, Gomes P, Duarte MM, Moresco RN. A simple and inexpensive automated technique for measurement of serum nitrite/nitrate. Clin Biochem 2011;44:348–50. 30. Benzie IF, Strain JJ. The ferric reducing ability of plasma (FRAP) as a measure of “antioxidant power”: the FRAP assay. Anal Biochem 1996;239:70–6. 31. Beutler E, Gelbart T. Plasma glutathione in health and in patients with malignant disease. J Lab Clin Med 1985;105:581–4. 32. Janero DR. Malondialdehyde and thiobarbituric acid–reactivity as diagnostic indices of lipid peroxidation and peroxidative tissue injury. Free Radic Biol Med 1990;9:515–40. 33. Porter ML, Dennis BL. Hyperbilirubinemia in the term newborn. Am Fam Physician 2002;65:599–606. 34. Nobile de Santis MS, Radaelli T, Taricco E, Bertini S, Cetin I. Excess of amniotic fluid: pathophysiology, correlated diseases and clinical management. Acta Biomed. 2004;75(Suppl 1):53–5. 35. RTI International. Triangle Park (NC): RTI International. 2007 [cited 2015 Feb 22]. Available from: http://www.rti.org/rtipress 36. Ryan MP, Ryan MF, Counihan TB. The effect of diuretics on lymphocyte magnesium and potassium. Acta Med Scand Suppl 1981;647:153–61. 37. Rumawas ME, McKeown NM, Rogers G, Meigs JB, Wilson PW, Jacques PF. Magnesium intake is related to improved insulin homeostasis in the Framingham offspring cohort. J Am Coll Nutr 2006;25:486–92. 38. Guerrero-Romero F, Rodriguez-Moran M. Hypomagnesemia, oxidative stress, inflammation, and metabolic syndrome. Diabetes Metab Res Rev 2006;22:471–6. 39. Guerrero-Romero F, Tamez-Perez HE, Gonzalez-Gonzalez G, SalinasMartinez AM, Montes-Villarreal J, Trevino-Ortiz JH, Rodriguez-Moran M. Oral magnesium supplementation improves insulin sensitivity in non-diabetic subjects with insulin resistance: a double-blind placebocontrolled randomized trial. Diabetes Metab 2004;30:253–8. 40. Kowluru A, Chen HQ, Modrick LM, Stefanelli C. Activation of acetylCoA carboxylase by a glutamate- and magnesium-sensitive protein phosphatase in the islet beta-cell. Diabetes 2001;50:1580–7. 41. Murakami M, Ishizuka J, Sumi S, Nickols GA, Cooper CW, Townsend CM Jr., Thompson JC. Role of extracellular magnesium in insulin secretion from rat insulinoma cells. Proc Soc Exp Biol Med 1992;200:490–4. 42. Lal J, Vasudev K, Kela AK, Jain SK. Effect of oral magnesium supplementation on the lipid profile and blood glucose of patients with type 2 diabetes mellitus. J Assoc Physicians India 2003;51:37–42. 43. Shechter M, Saad T, Shechter A, Koren-Morag N, Silver BB, Matetzky S. Comparison of magnesium status using X-ray dispersion analysis following magnesium oxide and magnesium citrate treatment of healthy subjects. Magnes Res 2012;25:28–39.

44. Rodríguez-Moran M, Guerrero-Romero F. Oral magnesium supplementation improves insulin sensitivity and metabolic control in type 2 diabetic subjects: a randomized double-blind controlled trial. Diabetes Care 2003;26:1147–52. 45. Purvis JR, Cummings DM, Landsman P, Carroll R, Barakat H, Bray J, Whitley C, Horner RD. Effect of oral magnesium supplementation on selected cardiovascular risk factors in non-insulin-dependent diabetics. Arch Fam Med 1994;3:503–8. 46. Kishimoto Y, Tani M, Uto-Kondo H, Saita E, Iizuka M, Sone H, Yokota K, Kondo K. Effects of magnesium on postprandial serum lipid responses in healthy human subjects. Br J Nutr 2010;103:469–72. 47. Rayssiguier Y, Gueux E. Magnesium and lipids in cardiovascular disease. J Am Coll Nutr 1986;5:507–19. 48. Nielsen FH, Johnson LK, Zeng H. Magnesium supplementation improves indicators of low magnesium status and inflammatory stress in adults older than 51 years with poor quality sleep. Magnes Res 2010; 23:158–68. 49. Yuan J, Zhou J, Chen BC, Zhang X, Zhou HM, Du DF, Chang S, Chen ZK. Magnesium supplementation prevents chronic cyclosporine nephrotoxicity via adjusting nitric oxide synthase activity. Transplant Proc 2005;37:1892–5. 50. Mortazavi M, Moeinzadeh F, Saadatnia M, Shahidi S, McGee JC, Minagar A. Effect of magnesium supplementation on carotid intimamedia thickness and flow-mediated dilatation among hemodialysis patients: a double-blind, randomized, placebo-controlled trial. Eur Neurol 2013;69:309–16. 51. Aneiros E, Philipp S, Lis A, Freichel M, Cavalie A. Modulation of Ca2 + signaling by Na+/Ca2+ exchangers in mast cells. J Immunol 2005; 174:119–30. 52. Mazur A, Maier JA, Rock E, Gueux E, Nowacki W, Rayssiguier Y. Magnesium and the inflammatory response: potential physiopathological implications. Arch Biochem Biophys 2007;458:48–56. 53. Farvid MS, Jalali M, Siassi F, Hosseini M. Comparison of the effects of vitamins and/or mineral supplementation on glomerular and tubular dysfunction in type 2 diabetes. Diabetes Care 2005;28:2458–64. 54. Farvid MS, Jalali M, Siassi F, Saadat N, Hosseini M. The impact of vitamins and/or mineral supplementation on blood pressure in type 2 diabetes. J Am Coll Nutr 2004;23:272–9. 55. Bede O, Nagy D, Suranyi A, Horvath I, Szlavik M, Gyurkovits K. Effects of magnesium supplementation on the glutathione redox system in atopic asthmatic children. Inflamm Res 2008;57:279–86. 56. Liu YX, Guo YM, Wang Z. Effect of magnesium on reactive oxygen species production in the thigh muscles of broiler chickens. Br Poult Sci 2007;48:84–9. 57. Boujelben M, Ghorbel F, Vincent C, Makni-Ayadi F, Guermazi F, Croute F, El-Feki A. Lipid peroxidation and HSP72/73 expression in rat following cadmium chloride administration: interactions of magnesium supplementation. Exp Toxicol Pathol 2006;57:437–43. 58. Misra PK, Kapoor RK, Dixit S, Seth TD. Trace metals in neonatal hyperbilirubinemia. Indian Pediatr 1988;25:761–4. 59. Skajaa K, Dorup I, Sandstrom BM. Magnesium intake and status and pregnancy outcome in a Danish population. Br J Obstet Gynaecol 1991;98:919–28. 60. Pintov S, Kohelet D, Arbel E, Goldberg M. Predictive inability of cord zinc, magnesium and copper levels on the development of benign hyperbilirubinemia in the newborn. Acta Paediatr 1992;81:868–9. 61. Tunc¸er M, Demirsoy S, Ozsoylu S, Erdem G. The significance of zinc, copper and magnesium levels of maternal, cord and newborns’ sera in hyperbilirubinemia of unknown etiology. Turk J Pediatr 1982;24: 227–33. 62. Arnaud MJ. Update on the assessment of magnesium status. Br J Nutr 2008;99(Suppl 3):S24–36. 63. Carpenter TO, DeLucia MC, Zhang JH, Bejnerowicz G, Tartamella L, Dziura J, Petersen KF, Befroy D, Cohen D. A randomized controlled study of effects of dietary magnesium oxide supplementation on bone mineral content in healthy girls. J Clin Endocrinol Metab 2006;91: 4866–72.