John Libbey Eurotext

Magnesium Research


Magnesium citrate supplementation decreased blood pressure and HbA1c in normomagnesemic subjects with metabolic syndrome: a 12-week, placebo-controlled, double-blinded pilot trial Volume 34, issue 3, July-August-September 2021


  • Figure 1


Metabolic syndrome (MetS) is a pathologic condition characterized by a set of metabolic impairments. To diagnose MetS, any three of following parameters should be met: insulin resistance, obesity, atherogenic dyslipidemia (high plasma triglyceride [TG] levels and/or low high-density lipoprotein cholesterol levels), and hypertension [1, 2]. These parameters are interrelated and share underlying mediators, mechanisms, and pathways, such as inflammation, oxidative stress, endothelial dysfunction, impaired function of β-cells, etc. [1, 3, 4]. The global prevalence of MetS can be estimated to be about one quarter of the world population [5]. This makes MetS a serious health problem, taking into account that MetS increases the risk of various cardiovascular diseases [6-8], of cancer [9], and of incident diabetes [10].

Changes in the concentration of cellular magnesium (Mg) might be an underlying common mechanism of the “insulin resistance” of the MetS, essential hypertension, altered glucose tolerance, and type 2 diabetes [11]. Mg plays an important role in insulin action as it is required for the activation of the tyrosine kinase domain of the insulin receptor, an important step in transmitting the insulin signal into cells [12]. Higher Mg intake decreases the risk of incident metabolic impairments (impaired glucose tolerance, insulin resistance, or hyperinsulinemia) [13]. Almost 80% of patients with MetS also suffer from hypertension, and the optimal antihypertensive treatment is being debated [2]. Recent large meta-analysis showed a correlation between Mg deficiency and hypertension and its successful therapy with Mg as well as a positive effect on increased systolic and diastolic blood pressure (BP) values [14].

Several lines of evidence confirm that Mg plays an important role in MetS. Patients with MetS have significantly lower serum Mg levels compared to healthy people [3, 15, 16]. Dietary Mg intake among nondiabetic individuals with MetS is way below recommended dietary allowance [17], and an inverse association was found between dietary Mg intake and the prevalence of the MetS [18-20]. Moreover, Mg intake was inversely associated with metabolic biomarkers of insulin resistance [17] and with systemic inflammation [18]. A meta-analysis of five studies has shown that for every 100 mg/day increment in Mg intake, the overall risk of having MetS was lowered by 17% [21]. Mg supplementation (382 mg/day for 4 months) significantly reduced systolic and diastolic BP values, insulin resistance, fasting glucose, and TG levels in metabolically obese, normal-weight individuals [22], and in patients with MetS [23]. Another study, however, showed that Mg supplementation (400 mg/day for 3 months) had no influence on BP, insulin resistance, levels of insulin, TGs, and fasting blood glucose in women with MetS without diabetes [24]. According to systematic review, approximately half of the clinical studies show no effect of Mg supplementation on certain components of MetS (high BP, hyperglycemia, hypertriglyceridemia, high-density lipoprotein cholesterol levels, and insulin sensitivity), although the majority of studies demonstrate improvement of at least one of these parameters upon Mg supplementation [25].

In hypomagnesemic patients, the effect of Mg supplementation is modular as it includes both the improvement of hypomagnesemia and the effect of additional Mg supplementation. We were interested in the latter one; therefore, in this study, we investigated the potential benefit of Mg supplementation in normomagnesemic individuals with MetS.

Materials and methods

This single-center, randomized, double-blinded, placebo-controlled pilot study was performed to assess the effect of 400 mg Mg in the form of Mg citrate on metabolic and physiologic parameters. All participants gave their written informed consent before starting the study. The study was conducted in accordance with the Declaration of Helsinki and the protocol was approved by the Ethical Committee at the Ärztekammer Westfalen-Lippe and Westfälischen Wilhelmsuniversität (2015-493-f-S). This trial was registered at (NCT03002545). Initially the study aimed to determine the effect of Mg supplementation on blood glucose control and insulin resistance within patients with type 2 diabetes (HbA1c > 6%). Due to low number of subjects with untreated diabetes within the prolonged recruiting phase (18 months), it was decided to change the focus and to proceed with available patients, majority of which (n = 42/50, figure 1) suffered from MetS.

Inclusion and exclusion criteria

Fifty patients were recruited among regular visits for different reasons in the St. Anna Hospital, Herne, Germany. Eligible individuals were men and women between 20 and 80 years of age with a MetS condition. MetS was determined by the presence of at least three of following symptoms: obesity determined as body mass index (BMI) > 25 kg/m2, BP ≥ 130/85 mmHg, hyperinsulinemia determined as HbA1c ≥ 5.7%, and dyslipidemia determined as TG > 150 mg/dL. Patients with two or less symptoms (n = 8) were not eligible for the study (figure 1).

Exclusion criteria included pregnancy and lactation, patients with severe kidney dysfunction (estimated glomerular filtration rate < 30 mL/min), anemia, the use of insulin, oral antidiabetics, diuretics, antacids, or proton pump inhibitors in the 4 week-period prior to study enrolment. Similarly, participants who had taken mineral or vitamin supplements, food supplements, or fortified foods that contain Mg in 4 weeks period prior to the study enrolment were also excluded.

Patients under hypertension treatment (n = 5 in the Mg group and n = 8 in the placebo group) were taking the following medication (single or multitherapy): Amlodipine, Aprovel Bipreterax, Beloc Zok, Candesartan, Carvedilol, Diovan, Lercanidipin, Nebilet, Ramipril, Valsartan, and Valsacor. During the study, the antihypertensive treatment was not changed.

Study design and procedures

Forty-two subjects (mean age 66 ± 9 years) were randomly assigned to receive either 400 mg Mg in the form of Mg citrate (Magnesium-Diasporal 400 EXTRA, Protina GmbH, Ismaning, Germany) or an identical looking and tasting placebo daily for 12 weeks. Mg citrate was chosen as it provides better bioavailability of Mg compared to inorganic salts such as Mg oxide [26]. Complete data set for analysis was obtained from n = 11 from the placebo group and n = 13 from the Mg group. Seventeen participants did not adhere to the treatment protocol (n = 7 in Mg group and n = 10 in placebo group), and one participant in placebo group dropped out for unknown reasons (figure 1).

Parameters were assessed at the beginning of the study (baseline) and after 12 weeks of supplementation. Body height (cm) was measured using a stadiometer (standard equipment) and body weight (kg) was assessed using regular scales to the nearest of 0.1 kg in nonfasting conditions with subjects wearing light clothing without shoes. BMI was calculated as body weight (kg) divided by height (m) squared. BP was determined according to recommendations proposed in the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure [27]. The auscultatory method of BP measurement was used for participants who had sat quietly for at least 5 minutes prior to the procedure. The average of two measurements of BP was recorded.Blood samples for serum and plasma were collected in tubes without anticoagulant or in tubes containing ethylenediaminetetraacetic acid, respectively, and processed according to standard procedures. Analyses were carried out by a Cobas analyzer (Roche, Mannheim, Germany). Blood samples for measurement of whole blood ionized Mg were collected in Eppendorf tubes that contained no anticoagulants, and right after the collection were subjected to measurement by a pHOx+ analyzer (NOVA, Rödermark, Germany).

Measured parameters included: systolic and diastolic BP, HbA1c, plasma glucose (nonfasting, so-called spontaneous or random), serum cholesterol, serum TGs, plasma Mg, plasma Ca, ionized Mg (from the whole blood), serum vitamin D, serum creatinine, serum interleukin-6 (IL-6), and serum C-reactive protein. Quality of life was assessed by SF-36 questionnaire.


Data are expressed as mean ± standard deviation. The normal distribution of our population (Gaussian distributions) was tested using the Kolmogorov and Smirnov method. Comparisons between groups were analyzed using the unpaired T test (equality of variances was assessed prior to the analysis). Paired T test was used to assess the difference between data in each group before and after intervention. For all statistical tests, P ≤ 0.05 was considered statistically significant.


Demographic data of the study population

All patients in this study were of Caucasian origin. Demographic and clinical characteristics were balanced between treatment groups (table 1). After several patients were removed from the analysis as they did not adhere to the treatment protocol, the difference in systolic BP between the groups became significant.

Effect of Mg supplementation on Mg levels

Subjects included in this study were normomagnesemic, with plasma Mg concentrations >0.75 mmol/L (mean concentration at baseline 0.79 ± 0.07 mmol/L) and blood-ionized Mg in the range of 0.55 to 0.75 mmol/L (mean concentration at baseline 0.62 ± 0.11 mmol/L). Plasma Mg concentrations in the Mg group significantly increased after 12 weeks (from 0.78 ± 0.07 to 0.83 ± 0.07 mmol/L, P = 0.01; table 1). There was neither a significant change in plasma Mg levels within the placebo group nor a significant difference between the groups at the baseline or at 12 weeks. In contrast to plasma Mg concentration, the change in blood-ionized Mg levels was not significant in both groups (–0.05 ± 0.08 and –0.01 ± 0.10 mmol/L in Mg and placebo groups, respectively).

Effect of Mg supplementation on glycemic control

To assess the effect of Mg supplementation on glycemic control, HbA1c and spontaneous plasma glucose levels were measured at baseline and after 12 weeks of intervention. HbA1c levels decreased significantly within the Mg group by –0.28 ± 0.27% (P < 0.01) but not within the placebo group (–0.04 ± 0.18%, P = 0.53). The difference between changes in both groups was also significant (P = 0.02).

Mg supplementation had no effect on the levels of plasma glucose, and there was no statistically significant difference between the groups at any time point.

Effect of Mgsupplementation on BP

We observed a difference at baseline with significantly higher values for systolic BP in the Mg group compared to the placebo group, which is likely due to BP medication. Five participants in the Mg group and eight in the placebo group were receiving an antihypertensive drug at baseline. During the study, the medication regimen for BP was not changed. After removing patients receiving antihypertensive medication from the analysis, we observed no difference between Mg and placebo groups at the baseline (data not shown), but the following effect of Mg treatment remained the same as described below (and with the similar level of significance).

After 12 weeks of intervention, systolic BP values were significantly lower in Mg group (120.8 ± 4.6 vs. 133.5 ± 7.6 mmHg, P < 0.001). Mg supplementation led to a pronounced decrease in systolic BP compared to placebo (–24.2 ± 11.5 vs. –1.0 ± 3.8 mmHg, respectively, P < 0.001).

Diastolic BP was significantly decreased within the Mg group after 12 weeks when compared with baseline (–6.8 ± 3.7 mmHg, P < 0.001). No significant difference was found in the placebo group (–1.0 ± 2.8 mmHg). After 12 weeks of supplementation, there was a significant difference in diastolic BP between Mg and placebo groups (P < 0.01).

Other parameters

During the study, no significant body weight changes were recorded.

After 12 weeks of supplementation with Mg citrate or placebo, vitamin D levels increased in both groups. However, the increase was significant only in Mg group (P < 0.01). This effect could be explained as study start was in springtime and ended in summer, which must have led to longer sun exposition periods and therefore increased vitamin D levels. Moreover, Mg plays an important role in vitamin D metabolism [28, 29] and Mg supplementation was shown to increase vitamin D concentration [30-32], which could lead to the greater difference in Mg group.

No significant changes occurred in both Mg and placebo groups in following parameters: cholesterol, creatinine, IL-6, and C-reactive protein. Level of TG significantly decreased in placebo group, but there was no significant difference between placebo and Mg group neither on baseline nor at 12 weeks. There were no changes in plasma Ca levels, but as plasma Mg levels increased in Mg group, Ca/Mg ratio showed significant decrease after 12 weeks of Mg supplementation (3.07 ± 0.33 to 2.88 ± 0.29, P = 0.03).

Neither of groups showed significant improvement in overall well-being (results quality of life questionnaire [SF-36], data not shown).

There were no adverse effects reported within the Mg or placebo group. One person in the Mg group reported about soft stool, but this effect was transient.


The aim of this study was to assess the effect of Mg supplementation in the form of Mg citrate on glycemic control and BP in normomagnesemic subjects with MetS.

To prove the efficacy of Mg supplementation in increasing Mg levels in organism, plasma and blood-ionized Mg concentrations were measured. An easy and simple test to measure body Mg status is still lacking; therefore, serum or plasma Mg concentration is commonly used to assess patients’ Mg status, although over 99% of total body Mg is extravascular with only 0.3% in serum [33]. According to the current reference values for plasma and ionized Mg concentrations, all study participants were normomagnesemic [33]. Twelve-week supplementation with 400 mg Mg in the form of Mg citrate led to a significant increase in plasma Mg concentrations after 12 weeks.

The Mg group also showed a significant increase in vitamin D concentration (14.7 ± 6.9 to 19.9 ± 8.3 ng/mL). This effect might be caused by the fact that enzymes involved in vitamin D metabolism are Mg dependent [28, 29]. Mg supplementation increased vitamin D concentration in the blood without any additional vitamin D supplementation [30-32].

Mg supplementation resulted in a significant reduction in HbA1c by 0.28%. The explanation is the importance of Mg for glycemic control. Serum Mg concentration negatively correlates with HbA1c as well as with serum glucose and insulin concentrations [34]. Mg supplementation significantly improved HbA1c, insulin levels, and HOMA-IR (homeostasic model assessment of insulin resistance in patients with type 2 diabetes [35]. In the review of 12 randomized controlled trials (RCTs), Morais et al. showed that the positive effect of Mg supplementation on several parameters, such as HbA1c, fasting insulin and glucose, and HOMA-IR, was more pronounced in hypomagnesemic patients [36]. Only a few studies assessed glycemic control in nondiabetic patients with normal Mg levels, with no clear consensus in results [37-39]. In a RCT by Mooren et al. [37], Mg supplementation improved fasting plasma glucose and insulin indices in normomagnesemic, overweight, and nondiabetic subjects. This result could not be replicated in Korean adults by Lee et al. [38]. Similarly, no effect of Mg supplementation on HbA1c and HOMA-IR was shown in healthy normomagnesemic young men with family history of MetS [39]. The differences in results probably arise from different study settings (duration of supplementation, dosage and formulation of Mg, ethnic background). A systematic review and meta-analysis of 21 RCTs by Simental-Mendia et al. [40] found that Mg supplementation for ≥4months improved HOMA-IR and fasting glucose in diabetic and nondiabetic subjects compared to a supplementation period et al. [25]. They identified 27 RCTs, focused on the improvement of at least one of the components of MetS. Six studies showed improvement of insulin sensitivity, and 4 out of 12 showed improvement of hyperglycemia upon Mg supplementation. Importantly, out of the nine studies that showed no positive outcome on the MetS, seven were conducted in healthy individuals, or in subjects with normal Mg levels at baseline.

Another parameter of MetS is dyslipidemia, characterized by high plasma TG levels and/or low high-density lipoprotein cholesterol levels. Patients in this study had TG levels of 209 ± 72 mg/dL, which is considered a high value (compared to normal value of et al., who showed no significant change in cholesterol and TG levels in women with MetS (76.8% normomagnesemic) upon 12-week Mg treatment [24]. In a review of Guerrero-Romero et al., 7 studies out of 11 showed no effect on TG levels and on high-density lipoprotein cholesterol levels [25].

Within the Mg group, there was a significant decrease in systolic and diastolic BP values. At baseline, subjects from the Mg group suffered from isolated systolic hypertension (grade 1 hypertension regarding systolic BP and high normal for diastolic BP values) according to the Guidelines of the European Society of Cardiology and European Society of Hypertension [41]. Systolic BP significantly decreased after 12 weeks of supplementation (from 145.0 ± 10.4 to 120.8 ± 4.6 mmHg, difference of –24.2 mmHg), reaching normal values according to the guidelines. Diastolic BP in Mg supplemented patients displayed decrease from 85.4 ± 3.1 to 78.5 ± 2.8 mmHg (difference of –6.8 mmHg), reaching optimal values. Neither systolic nor diastolic BP was affected in placebo group. We here acknowledge that, despite randomization, there was a significant difference in baseline systolic BP between the two patient groups, which arose after several patients were excluded from the study for not adhering to the treatment protocol. The main reason is probably higher proportion of patients receiving antihypertensive drugs in placebo group (8/11 compared to 5/13 in Mg group). Excluding these patients from analysis removed the difference in systolic BP at the baseline but had no influence on the results on Mg supplementation.

There is an inverse correlation between serum Mg concentration and hypertension [42, 43]. Mg-induced vasodilation through the reduction of intracellular Ca2+ concentration within vascular smooth muscle cells seems to be one of the mechanisms how Mg is involved in BP regulation. In addition, extracellular Mg contributes to vasodilation by reducing endothelin-1 expression, directly stimulating prostacyclin, and inhibiting nitric oxide formation [44].

Although not all trials with oral Mg supplementation in hypertension resulted in significant reduction of BP, several meta-analyses confirmed a significant decrease in systolic BP of 3 to 4 mmHg and diastolic BP of 2 to 3 mmHg [45-48]. Zhang et al. mentioned that in stratified analyses, a greater reduction in BP tended to be found in trials with high quality or low dropout rate [45]. Moreover, in the meta-analysis of Rosanoff and Plesset [49], studies on patients with high BP (systolic BP > 155 mmHg) reported much stronger effects of oral Mg treatment on systolic BP (–18.7 mmHg) and diastolic BP (–10.9 mmHg) similar to our results. Interestingly, comparable results have also been shown in borderline hypertension [50].

In general, within meta-analyses, the comparison of highly heterogeneous studies regarding Mg dosage, treatment duration, Mg status, sample sizes, trial quality, and participant characteristics may lead to an underestimated potential of Mg in hypertension in some but not all subjects.

It is worth mentioning that systolic and diastolic BPs influence the risk of adverse cardiovascular events independently from each other, although systolic hypertension had a greater effect [51]. Taking this in consideration, Mg supplementation could help reducing both systolic and diastolic BP in people at risk for cardiovascular events.

Several limitations of the study have to be mentioned:

  • (1)small sample size;
  • (2)HbA1c as a sole parameter has limited informative value on glycemic control, especially regarding insulin sensitivity/insulin resistance;
  • (3)and dietary Mg intake was not considered.


In conclusion, oral Mg citrate supplementation reduced HbA1c levels and reduced BP in normomagnesemic persons with MetS. Further interventional studies are required to determine the effect size of Mg supplementation on different parameters of MetS and thus provide optimal alternative strategies to medication.


Financial support: This study was funded by Protina Pharmazeutische GmbH. Conflict of interest: The authors Afitska K., Clavel J., and Werner T. are employed at Protina GmbH. The sponsor had no role in the study design and data collection. The authors had full access to all the data in the study.