7/27/2019 1-s2.0-S0953620513000800-main

1/5

Original article

Serum magnesium in the metabolically-obese normal-weight andhealthy-obese subjects

Fernando Guerrero-Romero, Martha Rodriguez-Moran

Biomedical Research Unit, Mexican Social Security Institute, Predio Canoas # 100, Col. Los Angeles, ZC 34067, Durango, Mexico

a b s t r a c ta r t i c l e i n f o

Article history:

Received 20 December 2012

Received in revised form 21 February 2013Accepted 25 February 2013

Available online 20 March 2013

Keywords:

Metabolically obese normal weight

Metabolically healthy obese

Magnesium

Hypomagnesemia

Background: Given that hypomagnesemia is relatedwith hyperglycemia, hypertension, hypertriglyceridemia, and

insulin resistance, the objective of this study was to determine whether serum magnesium levels are associated

with the metabolically obese normal weight (MONW) and the metabolically healthy obese (MHO) phenotypes.

Methods: Population-based cross-sectional study that enrolled 427 subjects, men and non-pregnant women aged

20 to 65 years, to participate in the study. Subjects were allocated into groups with and without obesity; among

non-obese individuals, the subgroup of MONW subjects was compared with a control group of healthy

normal-weight individuals. Among obese individuals, the subgroup of MHO subjects was compared with a con-

trol group of obese subjects who exhibited at least one metabolic abnormality. In the absence of obesity, the pres-

ence of fasting hyperglycemia, insulin resistance, hypertriglyceridemia, and/or hypertension defined the presence

of MONW phenotype. In the absence of hypertension, insulin resistance and metabolic abnormalities of fasting

glucose and triglycerides levels, the phenotypically obese subjects were defined as MHO individuals.

Results: The sex-adjusted prevalence of MONW and MHO phenotypes was 40.8% and 27.9%. The multivariate lo-

gistic regression model adjusted by family history of diabetes, age, body mass index, and waist-circumference,

showed a positive association between hypomagnesemia and the MONW phenotype (OR 6.4; 95%CI 2.320.4)

and negative relationship between serum magnesium and the MHO phenotype (OR 0.32; 95%CI 0.170.61).

Conclusions: Our results show that hypomagnesemia is positively associated with the presence of MONW phe-

notype, and the normomagnesemia negatively with the MHO phenotype.

2013 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.

1. Introduction

The concept of metabolically obese but normal weight (MONW) in-

dividuals was originally developed to describe a subgroup of normal-

weight individuals displaying obesity-related phenotypic characteris-

tics [1]. It has been proposed that these individuals, despite having a

body mass index (BMI) b25 kg/m2, are characterized to have insulin

resistance, hyperglycemia, hypertriglyceridemia, and/or high blood

pressure [2,3]. It has been speculated that the originof increased cardio-

vascular risk among the normal-weight individuals maybe genetic and/

or related with diet composition [35].

The concept of metabolically healthy obese (MHO) individuals

was introduced to characterize the obese individuals who have no

metabolic disorders [6,7]. These individuals, despite the presence of

obesity, are free of components of the metabolic syndrome [8] and

are characterized by the early onset of obesity. The fat distribution,

insulin resistance, and physical activity levels account only partially

for differences between the obese and MHO individuals [9].

Among the components of metabolic syndrome, hypertension,

dislipidemia, and hyperglycemia are strongly related to low serum

magnesium levels [1012]; so, as magnesium intake is inadequate in

the western diet [13], hypomagnesemia could be a contributing factor

in the development of cardiovascular risk factors in the MONW pheno-

type. Given the scarce data about physiopathology of MONW and MHO

phenotypes and because magnesium depletion could be an important

contributing factor, we have hypothesized that hypomagnesemia

might be related with metabolic disturbances in the normal-weight

individuals and that normal magnesium levels might contribute to the

metabolically normal status in the obese individuals; thus, the objective

of this study was to determine whether serum magnesium levels are

associated with the MONW and MHO phenotypes.

2. Materials and methods

With the approval of the protocol by the Mexican Social Security

Institute Research Committee and after obtaining the written informed

consent, a population-based cross-sectional study was carried out.

The sampling strategy was based on advertising strategies to the

general population of Durango, city in northern Mexico, for inviting

European Journal of Internal Medicine 24 (2013) 639643

Grant support: This work was supported by grants from the Mexican Social Security

Institute Foundation, Civil Association.

Corresponding author at: Siqueiros 225, 34000, Durango, Dgo., Mexico. Tel.: +52

618 8120997; fax: +52 618 8132014.

E-mail address: rodriguez.moran.martha@gmail.com (M. Rodriguez-Moran).

0953-6205/$ see front matter 2013 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.ejim.2013.02.014

Contents lists available at ScienceDirect

European Journal of Internal Medicine

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / e j i m

http://dx.doi.org/10.1016/j.ejim.2013.02.014http://dx.doi.org/10.1016/j.ejim.2013.02.014http://dx.doi.org/10.1016/j.ejim.2013.02.014mailto:rodriguez.moran.martha@gmail.comhttp://dx.doi.org/10.1016/j.ejim.2013.02.014http://www.sciencedirect.com/science/journal/09536205http://www.sciencedirect.com/science/journal/09536205http://dx.doi.org/10.1016/j.ejim.2013.02.014mailto:rodriguez.moran.martha@gmail.comhttp://dx.doi.org/10.1016/j.ejim.2013.02.014

7/27/2019 1-s2.0-S0953620513000800-main

2/5

men and non-pregnant women aged 20 to 65 years, to participate in

the study.

A total of 427 individuals were enrolled. Subjects were allocated

into groups with and without obesity; among non-obese individuals,

the subgroup of MONW subjects was compared with a control group

of healthy normal-weight individuals. Among obese individuals, the

subgroup of MHO subjects was compared with a control group of

obese subjects who exhibited at least one metabolic abnormality.

A standardized interview, clinical examination, and laboratorytests were performed to determine the presence of diabetes, preg-

nancy, smoking, alcohol consumption, chronic diarrhea, renal disease,

malignancy, and the intake of magnesium supplements in the previ-

ous three months, which were the exclusion criteria.

2.1. Definitions

In the absence of obesity, the presence of fasting hyperglycemia, in-

sulin resistance, hypertriglyceridemia, and/or hypertension defined the

presence of the MONW phenotype. The phenotypically normal-weight

subjects, without metabolic abnormalities, defined the control group

of the MONW individuals.

In the absence of hypertension, insulin resistance, and metabolic

abnormalities of fasting glucose and triglycerides levels, the pheno-

typically obese subjects were defined as MHO individuals. The pheno-

typically obese subjects with at least one metabolic abnormality,

defined the control group of the MHO phenotype.

In order to minimize the rate of false positive test and consequent-

ly, to reduce the possibility of bias in the analysis, hypomagnesemia

was defined by serum magnesium concentration 0.7 mmol/dL

(1.7 mg/dL) and normomagnesemia by serum magnesium levels

>0.7 (1.7 mg/dL) and b1.15 mmol/L (2.8 mg/dL).

Normal-weight was defined by BMI b 25 kg/m2 and obesity by

BMI 25 kg/m2.

Fasting hyperglycemia was defined by fasting plasma glucose

5.6 and b7.0 mmol/L (100 and b126 mg/dL) [14].

Hypertriglyceridemia was defined by the presence of serum tri-

glyceride levels 1.7 mmol/L (150 mg/dL) [15].

Insulin resistance was defined by thepresence of HOMA-IR index3.

2.2. Measurements

In the standing position, weight and height were measured using a

fixed scale stadimeter with the subjects in light clothing and without

shoes. The BMI was calculated as weight (kilograms) divided by

height (meters) squared. The WC was measured to the nearest centi-

meter with a flexible steel tape measure with the subjects in standing

position. The anatomical landmarks were: laterally, midway between

the lowest portion of the rib cage and iliac crest, and the umbilicus

anteriorly.

Brachial artery blood pressure was measured in seated partici-

pants after they had rested for 5 min with the use of a baumanometer

(Microlife AG, Heerbrugg Switzerland) and stethoscope (3M LittmanClassic II, Neuss, Germany). The technique of blood pressure mea-

surement was according the criteria by the Seventh Report of the

Joint National Committee on Prevention, Detection, Evaluation, and

Treatment of High Blood Pressure [16].

The HOMA-IR index was calculated using the formula fasting insu-

lin (U/mL) fasting glucose (mmol/L) / 22.5 [17].

2.3. Assays

Whole blood sample was collected from antecubital venous after

810 h overnight fasting. Serum magnesium concentrations were

measured by colorimetric method, the intra- and interassay variations

were 1.0 and 1.5%, respectively. Serum glucose was measured using

the glucose-oxidase method; the intra- and inter-assay coefficients of

variation were 1.1% and 1.3%. Insulin levels were measured by micro-

particle enzyme immunoassay (Abbot Axsym System), with intra- and

inter-assay coefficients of variation 3.4% and 5.9%. Triglycerides were

measured enzymatically. HDL-cholesterol fraction was obtained after

precipitation by phosphotungstic reagent. The intra-and inter-assay co-

efficients of variation were 1.9% and 3.7% for triglycerides, and 1.5% and

3.1% for HDL-cholesterol.

Measurements were performed in a Data Pro Plus Clinical Analyzer

(Arlington, TX, USA).

2.4. Statistical analysis

Differences between the groups were assessed using unpaired

Student's ttest (MannWhitney Utest for skewed data) for numeric

variables, and the Chi-squared test for testing differences between

proportions.

The correlationship between serum magnesium levels with hy-

perglycemia, hypertriglyceridemia, and HOMA index in MONW sub-

jects, was estimated using Pearson correlation test.

Multiple logistic regression analysis adjusted by family history

of diabetes (FHD), age, BMI, and WC was used to compute the Odds

Ratio (OR) between serum magnesium levels (independent variable)

and the MONW and MHO phenotypes (independent variables). An ad-

ditional adjusted multiple logistic regression analysis was performed to

establish the relationship between serum magnesium and the cardio-

vascular risk factors (dependent variables), in the normal-weight and

obese individuals.

Data were analyzed using the statistical package SPSS 15.0 (SPSS

Inc., Chicago Il).

3. Results

A total of 427 individuals were enrolled; 274 (64.2%) women and

153 (35.8%) men, with average age and BMI of 41.5 13.7 years and

29.9 6.2 kg/m2.

Seventy-six (17.8%) individuals exhibited normal-weight; of

these, 31 (40.8%) had the MONW phenotype.

Table 1 shows the characteristics of the population with normal-weight according metabolically status. The percentage of FHD,

HOMA-IR value, and fasting glucose levels were higher, and serum

magnesium levels were lower, in the MONW individuals than in the

control group. There were no significant differences between the

groups for the BMI and WC.

Among the individuals with the MONW phenotype, the frequency

of hyperglycemia, hypertriglyceridemia, insulin resistance, and hy-

pertension was of 70.1%; 67.7%, 28.6%, and 3.2%, respectively; a total

of 2 (6.4%), 17 (54.8%), 12 (38.7%), and 0 (0.0%) of the MONW indi-

viduals exhibited 1, 2, and 3 cardiovascular risk factors.

There were a negative correlationship between serum magnesium

levels and hyperglycemia, hypertriglyceridemia, and HOMA index in

the MONW subjects, Fig. 1.

Hypomagnesemia was identified in 22 (70.1%) of the MONW indi-viduals and 8 (17.8%) of the normal-weight individuals without met-

abolic disturbances, p b 0.0005.

The crude OR that computes the relationship between hypomag-

nesemia and the MONW phenotype was 11.3 (95%CI 3.8 to 33.6,

p b 0.0005). In the adjusted multivariate logistic regression model,

the hypomagnesemia remained significantly associated with the

MONW phenotype (OR 6.4; 95%CI 2.320.4).

A total of 351 (82.2%) individuals exhibited obesity; of them, 98

(27.9%) subjects had the MHO phenotype.

Table 1 shows the characteristics of the obese population according

to metabolic status. The MHO individuals showed similar BMI and WC

than the obese individuals in the control group; however, the HOMA-IR

index, fasting glucose, insulin, and triglyceride levels were significantly

lower, whereas the serum magnesium levels were significantly higher,

640 F. Guerrero-Romero, M. Rodriguez-Moran / European Journal of Internal Medicine 24 (2013) 639643

7/27/2019 1-s2.0-S0953620513000800-main

3/5

in the MHO individuals as compared with the control group of obese

subjects.

Among the obese individuals, the frequency of hyperglycemia,

hypertriglyceridemia, hypertension, and insulin resistance was 48.2%;

54.1%, 38.7%, and 11.5%, respectively; a total of 100 (39.5%), 101

(39.9%), 48 (19.0%), and 4 (1.6%), of the obese individuals exhibited 1,

2, 3, and 4 cardiovascular risk factors. Hypomagnesemia was identified

in 10 (10.2%) of the MHO individuals and 121 (47.8%) of the obese

individuals with metabolic disturbances, p b 0.0005.

The crude OR that computes the relationship between

normomagnesemia and the MHO phenotype was 0.11 (95%CI 0.014

to 0.88, p = 0.03). In the adjusted multivariate logistic regression

model, the normomagnesemia remained negatively associated with

the MHO phenotype (OR 0.32; 95%CI 0.17 to 0.61).

Table 2 shows the adjusted ORs between hypomagnesemia and the

cardiovascular risk factors in the non-obese and obese groups. The

hypomagnesemia was strongly related with hypertriglyceridemia,

hyperglycemia, and insulin resistance in the non-obese group and, (in

a lesser degree) with insulin resistance and hypertriglyceridemia in

the obese group.

4. Discussion

Our results show that hypomagnesemia is positively associated

with the MONW phenotype, and the normomagnesemia negatively

with the MHO phenotype.

Furthermore, our results show that hypomagnesemia is strongly

related with hyperglycemia, hypertriglyceridemia, insulin resistance

in the non-obese individuals, and (in a lesser degree) with the

hypertriglyceridemia and insulin resistance in the obese individuals.

The underlying mechanisms and correlates of normal-weight indi-

viduals who display cardiovascular risk factor clustering and of obese

individuals who are resistant to the development of the adiposity-

associated cardio-metabolic abnormalities are not well known; ourresults support the hypothesis that serum magnesium levels might

play an important role in the development of these phenotypes.

Recently, Wildman et al. [18] reported the results of a cross-

sectional analysis performed on a sample of the National Health and

Nutrition Examination Surveys 19992004; they found a prevalence

of MONW and MHO, of 23.5% and 31.7%, respectively; findings that

emphasize the magnitude of this public health problem and encour-

age for further research into the underlying physiologic mechanisms

of these phenotypes.

On this regard, Hwang et al. [8] in a nation-wide population cohort,

in which the prevalence of the MHO phenotype was of 28.5%, report

that the MHO individuals are at higher risk to develop hypertension,

type 2 diabetes, and metabolicsyndrome than their non-obesecounter-

parts, providing evidence thatopposes the notion of MHO as a harmless

condition.

However, in both the National Health and Nutrition Examination

Surveys 19992004 [18] and the study by Hwang et al. [8] there are

no reports about the relationship between MONW and MHO with

magnesium status.

In our study, the prevalence of the MONW phenotype (40.8%) was

higher than that reported in the Asian (12.7% to 13.3%) and USA pop-

ulations (23.5%) [1820], whereas the prevalence of the MHO pheno-type (27.9%) was similar to previous reports in the Asian (31.7% to

47.9%) [19,20] and USA (23.5% to 31.7%) [8,18] subjects, findings

that support the possible role of ethnicity in the development of

both phenotypes and emphasizes, that regardless of weight status,

the health behaviors should be modified to prevent the development

of cardiovascular risk factors. On this matter, the main contribution of

our study suggests that serum magnesium status might play an im-

portant role in either the development of MONW or the prevention

of MHO phenotypes. Taking into account that serum magnesium

levels are easy to detect and magnesium deficiency is easy to modify,

through diet and/or oral magnesium supplementation, our finding

could be of interest for researchers and clinicians in the field. Further

research, based on double blind placebo controlled clinical trials is

mandatory.A growing body of evidence, derived from epidemiological studies

[2126] and clinical trials [2730], consistently shows an inverse

Table 1

Characteristics of the target population, according to the weight and metabolic status.

MONW Normal-weight MHO Obese

n = 31 n = 45 P value n = 98 n = 253 P value

Family history of type 2 diabetes, n (%) 11 (35.5) 6 (13.3) 0.04 17 (17.3) 44 (17.4) 0.88

Age, years 39.0 19.2 36.2 13.1 0.48 42.1 14.0 42.4 13.1 0.94

Body mass index 22.5 1.6 22.3 2.0 0.64 30.1 1.5 31.7 1.6 0.57

Waist circumference, cm 84.2 9.1 86.0 10.5 0.43 100.8 8.3 104.6 14.0 0.19

Systolic blood pressure, mm Hg 112.2 16.2 105.0 14.5 0.50 114.3 22.7 118.3 18.3 0.59

Diastolic blood pressure, mm Hg 87.0 8.0 86.1 9.2 0.67 77.6 11.6 76.2 10.9 0.38

Fasting glucose, mmol/L 5.8 1.0 4.8 0.5 0.0005 5.0 0.5 5.6 0.9 0.002

Fasting insulin, pmol/L 63.9 27.38 51.4 27.1 0.85 46.5 20.1 80.6 43.1 b0.0005

HOMA-IR 2.8 1.0 1.50 0.7 0.03 1.5 0.7 2.9 1.5 b0.0005

HDL-c, mmol/L 1.2 0.4 1.2 0.4 0.80 1.1 0.1 1.0 0.4 0.17

Triglycerides, mmol/L 3.7 2.8 1.3 0.2 b0.0005 1.2 0.4 2.4 2.1 b0.0005

Serum magnesium, mmol/L 0.66 0.21 0.78 0.12 0.04 0.78 0.12 0.66 0.21 0.01

MONW, metabolically obese normal-weight.

MHO, metabolically-healthy obese.

Fig. 1. Correlationship between serum magnesium and hyperglycemia, hypertriglyceridemia, and HOMA index in metabolically obese normal-weight subjects.

641F. Guerrero-Romero, M. Rodriguez-Moran / European Journal of Internal Medicine 24 (2013) 639643

7/27/2019 1-s2.0-S0953620513000800-main

4/5

relationship between dietary magnesium intake and serum magne-

sium levels with the risk of developing metabolic glucose disorders

and insulin resistance. In addition, also it has been emphasized the

strong relationship between magnesium depletion and hypertriglyc-

eridemia [11,20,29,3133]. Finally, it is well known that magnesium

affects blood pressure acting as a calcium channel antagonist, stimu-

lating production of prostacyclins and nitric oxide, and altering

vascular responses to vasoactive agonists [34]; on this regard, epide-

miological studies [3537] and clinical trials [3841] also have dem-

onstrated an inverse correlation between blood pressure and serum

magnesium levels.

However, to the best of our knowledge, the relationship between

serum magnesium levels with the MONW and MHO phenotypes has

not been previously evaluated; our results strongly suggest that, hy-

pomagnesemia plays an important role in the development of

MONW as well as that normomagnesemia is negatively associated

with MHO individuals.

We did not measure the customary diet, body fat composition,

physical activity, and gene expression, which is the main limitation

of this study. However, taking into account the aimed objective in

the study and the sampling strategy for including a representative

sample of the target population, this limitation does not exert an im-

portant role on our main conclusion.

Learning points

The frequency of the MONW and MHO phenotypes is common. Hypomagnesemia is positively associated to the MONW phenotype. Normomagnesemia is negatively associated to the MHO phenotype.

Hypomagnesemia is strongly related with hyperglycemia, hypertri-

glyceridemia and insulin resistance in the non-obese individuals

and (in a lesser degree) with the hypertriglyceridemia and insulin

resistance in the obese individuals.

Hypomagnesemia is strongly related with hypertriglyceridemia and

insulin resistance in the obese individuals.

Conflict of interests

The authors state that they have no conflicts of interest.

References

[1] Ruderman NB, Schneider SH, Berchtold P. The metabolically-obese, normal-weight individual. Am J Clin Nutr 1981;34:161721.

[2] Conus F, Rabasa-Lhoret R, Pronnet F. Characteristics of metabolically obesenormal-weight (MONW) subjects. Appl Physiol Nutr Metab 2007;32:412.

[3] St-Onge MP, Janssen I, Heymsfield SB. Metabolic syndrome in normal-weightAmericans: new definition of the metabolically obese, normal-weight individual.Diabetes Care 2004;27:22228.

[4] Tsai CH. Metabolic syndrome in non-obese Taiwanese: new definition of metabol-ically obese, normal-weight individual. Chin Med J (Engl) 2009;122:25349.

[5] Ruderman NB, Berchtold P, Schneider S. Obesity-associated disorders in normal-weight individuals: some speculations. Int J Obes 1982;6:17151.

[6] Karelis AD, St-Pierre DH, Conus F, Rabasa-Lhoret R, Poehlman ET. Metabolicand body composition factors in subgroups of obesity: what do we know? J ClinEndocrinol Metab 2004;89:256975.

[7] Primeau V, Coderre L, KarelisAD, Brochu M, Lavoie ME,Messier V, et al.Character-izing the profile of obese patients who are metabolically healthy. Int J Obes (Lond)2011;35:97181.

[8] Hwang LC, Bai CH, Sun CA, Chen CJ. Prevalence of metabolically healthy obesityand its impacts on incidences of hypertension, diabetes and the metabolic syn-drome in Taiwan. Asia Pac J Clin Nutr 2012;21:22733.

[9] Hayes L, Pearce MS, Firbank MJ, Walker M, Taylor R, Unwin NC. Do obese butmetabolically normal women differ in intra-abdominal fat and physical activitylevels from those with the expected metabolic abnormalities? BMC Public Health2010;10:723.

[10] Rasic-Milutinovic Z, Perunicic-Pekovic G, Jovanovic D, Gluvic Z, Cankovic-Kadijevic

M. Association of blood pressure and metabolic syndrome components withmagnesium levels in drinking water in some Serbian municipalities. J Water Health2012;10:1619.

[11] Guerrero-Romero F, Rodrguez-Morn M. Low serum magnesium levels and met-abolic syndrome. Acta Diabetol 2002;39:20913.

[12] Guerrero-Romero F, Rodrguez-Morn M. Hypomagnesemia, oxidative stress, in-flammation, and metabolic syndrome. Diabetes Metab Res Rev 2006;22:4716.

[13] Ford ES, Mokdad AH. Dietary magnesium intake in a national sample of US adults.J Nutr 2003;133:287982.

[14] The Expert Committee on the Diagnosis Classification of Diabetes Mellitus. Reportof the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus.Diabetes Care 1999;22:S5S19.

[15] Executive summary of the third report of the National Cholesterol EducationProgram expert panel on detection, evaluation, and treatment of high bloodcholesterol in adults. JAMA 2001;285:248697.

[16] Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo Jr JL, et al. JointNational Committee on Prevention, Detection, Evaluation, and Treatment ofHigh Blood Pressure. National Heart, Lung, and Blood Institute; National HighBlood Pressure Education Program Coordinating Committee. Seventh report of the

Joint National Committee on Prevention, Detection, Evaluation, and Treatment ofHigh Blood Pressure. Hypertension 2003;42:120652.

[17] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeo-stasis model assessment: insulin resistance and beta-cell function from fastingplasma glucose and insulin concentrations in man. Diabetologia 1985;28:4129.

[18] Wildman RP, Muntner P, Reynolds K, McGinn AP, Rajpathak S, Wylie-Rosett J,et al. The obese without cardiometabolic risk factor clustering and the normalweight with cardiometabolic risk factor clustering: prevalence and correlates of2 phenotypes among the US population (NHANES 19992004). Arch Intern Med2008;168:161724.

[19] Geetha L, Deepa M, Anjana RM, Mohan V. Prevalence and clinical profile ofmetabolic obesity and phenotypic obesity in Asian Indians. J Diabetes Sci Technol2011;5:43946.

[20] Lee K. Metabolically obese but normal weight (MONW) and metabolically healthybut obese (MHO) phenotypes in Koreans: characteristics and health behaviors.Asia Pac J Clin Nutr 2009;18:2804.

[21] Lopez-Ridaura R, Willett WC, Rimm EB, Liu S, Stampfer MJ, Manson JE, et al.Magnesium intake and risk of type 2 diabetes in men and women. DiabetesCare 2004;27:13440.

[22] He K, Liu K, Daviglus ML, Morris SJ, Loria CM, Van Horn L, et al. Magnesium intakeand incidence of metabolic syndrome among young adults. Circulation 2006;113:167582.

[23] Kirii K, Iso H, Date C, Fukui M, Tamakoshi A, JACC Study Group. Magnesium intakeand risk of self-reported type 2 diabetes among Japanese. J Am Coll Nutr 2010;29:99106.

[24] Ma J, Folsom AR, Melnick SL, Eckfeldt JH, Sharrett AR, Nabulsi AA, et al. Associa-tions of serum and dietary magnesium with cardiovascular disease, hypertension,diabetes, insulin, and carotid arterial wall thickness: the ARIC study. AtherosclerosisRisk in Communities study. J Clin Epidemiol 1995;48:92740.

[25] Kim DJ, Xun P, Liu K, Loria C, Yokota K, Jacobs Jr DR, et al. Magnesium intake in re-lation to systemic inflammation, insulin resistance, and the incidence of diabetes.Diabetes Care 2010;33:260410.

[26] Rumawas ME,McKeown NM,RogersG, Meigs JB,WilsonPW, JacquesPF. Magnesiumintake is related to improved insulin homeostasis in the Framingham offspringcohort. J Am Coll Nutr 2006;25:48692.

[27] Mooren FC,KrgerK, Vlker K, GolfSW, WadepuhlM, Kraus A. Oralmagnesiumsup-plementation reduces insulin resistance in non-diabetic subjects a double-blind,

placebo-controlled, randomized trial. Diabetes Obes Metab 2011;13:281

4.[28] Guerrero-Romero F, Tamez-Perez HE, Gonzlez-Gonzlez G, Salinas-Martnez AM,

Montes-Villarreal J, Trevio-Ortiz JH, et al. Oral magnesium supplementationimproves insulin sensitivity in non-diabetic subjects with insulin resistance. Adouble-blind placebo-controlled randomized trial. DiabetesMetab 2004;30:38253.

[29] Chacko SA, Sul J, Song Y, Li X, LeBlanc J, You Y, et al. Magnesium supplementation,metabolic and inflammatory markers, and global genomic and proteomic profiling:a randomized, double-blind, controlled, crossover trial in overweight individuals.Am J Clin Nutr 2011;93:46373.

[30] Rodrguez-Morn M, Guerrero-Romero F. Oral magnesium supplementation im-proves insulin sensitivity and metabolic control in type 2 diabetic subjects: a ran-domized double-blind controlled trial. Diabetes Care 2003;26:114752.

[31] Lima M de L, Cruz T, Rodrigues LE, Bomfim O, Melo J, Correia R, et al. Serum and in-tracellular magnesium deficiency in patients with metabolic syndrome evidencesfor its relation to insulin resistance. Diabetes Res Clin Pract 2009;83:25762.

[32] McKeown NM, Jacques PF, Zhang XL, Juan W, Sahyoun NR. Dietary magnesium in-take is relatedto metabolic syndrome in older Americans.Eur J Nutr 2008;47:2106.

[33] Belin RJ, He K. Magnesium physiology and pathogenic mechanisms that contrib-ute to the development of the metabolic syndrome. Magnes Res 2007;20:10729.

Table 2

Odds ratio (OR) that computes the relationship between low serum magnesium levels

and metabolic disorders in the non-obese (n = 76) and obese (n = 351) individuals.

Non-Obese Obese

OR 95% CI OR 95% CI

Hypertension 0.11 0.11 to 12.41 0.88 0.41 to 1.71

Hyperglycemia 4.15 1.4 to 11.8 1.48 0.97 to 2.26

Hypertriglyceridemia 6.67 2.1 to 20.4 1.61 1.5 to 2.46

HOMA-IR index 3 9.00 0.99 to 81.4 2.89 1.84 to 4.53

642 F. Guerrero-Romero, M. Rodriguez-Moran / European Journal of Internal Medicine 24 (2013) 639643

7/27/2019 1-s2.0-S0953620513000800-main

5/5

[34] Sontia B, Touyz RM. Role of magnesium in hypertension. Arch Biochem Biophys2007;458:339.

[35] Abbasi IU, Salim-ul-Haque, Kausar MW, Karira KA, Zubaris NA. Correlation ofdivalent cat ions (Ca++, Mg++) and serum renin in patients of essential hyper-tension. J Pak Med Assoc 2012;62:1348.

[36] Kesteloot H, Tzoulaki I, Brown IJ, Chan Q, Wijeyesekera A, Ueshima H, et al.Relation of urinary calcium and magnesium excretion to blood pressure: theInternational Study Of Macro- and Micro-nutrients and Blood Pressure and theInternational Cooperative Study on Salt, Other Factors, and Blood Pressure. Am JEpidemiol 2011;174:4451.

[37] Maier JA, Malpuech-Brugere C, Zimowska W, Rayssiguier Y, Mazur A. Low magne-

sium promotes endothelial cell dysfunction: implications for atherosclerosis,inflammation and thrombosis. Biochim Biophys Acta 2004;1689:1321.

[38] Guerrero-Romero F, Rodrguez-Morn M. The effect of lowering blood pressureby magnesium supplementation in diabetic hypertensive adults with low serummagnesium levels: a randomized, double-blind, placebo-controlled clinical trial.

J Hum Hypertens 2009;23:24551.[39] Motoyama T, Sano H, Fukuzaky H. Oral magnesium supplementation in patients

with essential hypertension. Hypertension 1989;13:22732.[40] WittemanJC, Grobbee DE, Derkx FH, BouillonR, de Bruijn AM, Hofman A. Reduction

of blood pressure with oral magnesium supplementation in women with mild tomoderate hypertension. Am J Clin Nutr 1994;60:12935.

[41] Wexler R, Aukerman G. Nonpharmacologic strategies for managing hypertension.Am Fam Physician 2006;73:19536.

643F. Guerrero-Romero, M. Rodriguez-Moran / European Journal of Internal Medicine 24 (2013) 639643