Advertisement

Environmental Science and Pollution Research

, Volume 26, Issue 3, pp 2238–2252 | Cite as

Chronic mercury exposure and blood pressure in children and adolescents: a systematic review

  • Gema Gallego-Viñas
  • Ferran Ballester
  • Sabrina LlopEmail author
Review Article
  • 259 Downloads

Abstract

The aim of this paper is to systematically review the scientific literature on the possible relation of chronic mercury exposure and blood pressure among children and adolescents. We searched for observational studies in 6 electronic databases and grey literature for English, French or Spanish language studies published up to 30th November 2017. We performed a quality assessment of primary studies. We identified 8 articles involving 5 cohorts, 1 cross-sectional study and 1 case-control study. The participants had mean ages of between 3 and 17 years. Mercury was analysed in different matrices and periods of exposure. Four articles evaluated prenatal exposure, 2 evaluated both prenatal and postnatal exposures and 2 postnatal exposure. Blood pressure was measured according to different protocols. The association between mercury and blood pressure was adjusted by different covariates in each study. Four articles found a positive significant association between chronic mercury exposure and blood pressure in children or adolescents. Among these 4 articles, three of them evaluated prenatal exposure. There are still few studies assessing chronic mercury exposure and blood pressure in children and adolescents with inconsistency in results. Designs are very heterogeneous, which hampers their comparability. Evidence of this association is scarce and further research is needed.

Keywords

Environmental exposure Mercury Cardiovascular Blood pressure Paediatrics Public health Epidemiology 

Abbreviations:

BP

Blood pressure

DBP

Diastolic blood pressure

DHA

Docosahexaenoic acid

EPA

Eicosapentaenoic acid

MeHg

Methylmercury

PCB

Polychlorinated biphenyls

PUFAs

Polyunsaturated fatty acids

SBP

Systolic blood pressure

WOS

Web of Science

Introduction

Mercury is a ubiquitous pollutant considered as highly toxic to human health (WHO 2007; Sheehan et al. 2014). Both natural and anthropogenic sources are possible, although the human activities are currently the most important (UNEP 2013; Streets et al. 2017). Trends in environmental mercury levels in the future are uncertain. Future mercury emissions depend upon many variables as development of national and regional economies, implantation of technologies, and regulatory changes (Sundseth et al. 2017). In this context, the recent Minamata Convention on Mercury aims to control anthropogenic emissions (Kessler 2013; Lancet 2017).

Mercury metabolism and toxicity depends on its chemical form: elemental (metallic), inorganic, or organic. Elemental mercury exposure originates from mercury spills (broken thermostats, e.g.), gold mining, dental fillings made with mercury amalgam, skin-lightening creams and soaps, traditional medicines, and religious, ethnic or cultural practices (Voodoo and Santeria). Inorganic mercury has found in some antiseptics, laxatives, cosmetic products, bactericides, fungicides, and insecticides. Organic mercury is frequently detected as methylmercury (MeHg) and ethylmercury. It has been found in fishing products, thimerosal-containing vaccines, and pharmaceutical products (WHO-UNEP 2008; WHO 2010; Bose-O’Reilly et al. 2010; Park and Zheng 2012).

Around the world, mercury exposure mainly occurs through consumption of sea products contaminated with MeHg (WHO 2013a; O’Brien 2017). Both blood and hair can be used to assess MeHg exposure. The presence of mercury in blood or red blood cells indicates recent or current exposure to mercury. Once incorporated in hair, mercury does not return to blood; therefore, it provides a good long-term marker of exposure to MeHg. Most mercury in hair is in the form of MeHg, especially among populations that consume fish. Cord blood concentrations characterise children’s prenatal MeHg exposure. On the other hand, urine mercury levels are considered the best measure of recent exposures to inorganic mercury or elemental mercury vapour, although serum or plasma is also known as a good biomarker of these exposures (WHO 1991; WHO-UNEP 2008).

The nervous system is considered to be the most sensitive target organ to mercury toxicity (WHO 2007), although negative effects have been also reported in other systems including the cardiovascular, reproductive, and respiratory systems, renal function, skin, and liver (Holmes et al. 2009; Bernhoft 2012; Syversen and Kaur 2012; Rice et al. 2014).

Furthermore, high systolic blood pressure (BP) is considered one of the leading risk factors for global disease burden (GBD 2016 Disease and Injury Incidence and Prevalence Collaborators 2017). High BP shows an increasing prevalence attributed to population growth, ageing, and behavioural risk factors. The prevalence of paediatric hypertension worldwide is difficult to determine because of many factors, as different definition (Lurbe et al. 2016). A recent systematic review (Roulet et al. 2017) found a secular decrease in BP in the paediatric population, although it is also recognised in many studies, a significant degree of underdiagnosis of paediatric elevated BP and hypertension (Hansen et al. 2007; Brady et al. 2010; Dionne 2017; Rinke et al. 2018). Current estimates suggest that prevalence of paediatric hypertension is 3.5% in American children, 2.2–4.9% in central European countries, and 9–13% in Southern and Western Europe (Brady et al. 2018). In any case, high BP is globally the strongest modifiable risk factor for cardiovascular disease and related disability (WHO 2013b; O’Brien 2017).

Children may be particularly vulnerable to the effects that mercury exposure can have on health because of their immature systems and their rapid growth and development. Furthermore, children can be more prone to higher levels of exposure than adults (due to behavioural patterns, bigger intakes in relation to their corporal weight, etc.). A growing amount of evidence points to developmental exposure to environmental contaminants (such as metals) as a possible cause of epigenetic changes, thus suggesting mechanisms for subsequent diseases in adulthood (Gluckman et al. 2005; Balbus et al. 2013). Early childhood exposure to mercury may have a persistent impact on the quality of life in the adolescent and adult years (WHO 2010).

Regarding the cardiovascular system, mercury exposure has been related in adults to hypertension, coronary heart disease, myocardial infarction, stroke, oxidation, atherosclerosis, and changes in heart rate variability (Virtanen et al. 2007; Roman et al. 2011; Houston 2011; Gribble et al. 2015; Genchi et al. 2017). Moreover, experimental studies conducted in rats and mice found evidences of increasing blood pressure with MeHg exposure (Wakita 1987; Grotto et al. 2009; Islam et al. 2016).

Prospective studies with a long enough follow-up period to establish relationships between BP in childhood and cardiovascular events or mortality in adulthood are not yet available. Nevertheless, it has been observed that children with high BP are more likely to have hypertension in adulthood, as well as the presence of intermediate markers of target organ damage in hypertensive children and adolescents, which marks the importance of identifying casual factors and controlling BP in childhood (Hardy et al. 2015; Lurbe et al. 2016).

The aim of this paper is to systematically review the scientific literature on the possible relation of chronic mercury exposure and BP among children and adolescents (0–18 years old).

Methods

Search strategy

We followed the recommendations of Meta-Analysis of Observational Studies in Epidemiology (MOOSE) guidelines (Stroup et al. 2000). The search strategy was developed based on the knowledge of the authors. Key words and controlled vocabulary were chosen based on previous findings (“mercury”, “mercury compounds”, “organomercury compounds”, “blood pressure”, “hypertension”, “vascular diseases”, “autonomic nervous system”). We searched for published evidence from observational studies analysing the effect of chronic mercury exposure on blood pressure in children and/or adolescents (birth to age 18). We searched relevant studies on 7 December 2017 using the electronic bibliographic databases PubMed, Embase, Scopus, Web of Science (WOS), Lilacs, and Índice Médico Español (IME) (http://bddoc.csic.es:8080/). If supported by the electronic database, we limited the search to studies conducted in humans from birth to 18 years old, published up to 30 November 2017 and written in English, Spanish, or French. Otherwise, we manually applied these filters. The search strings are provided in the Supplementary File.

For the articles considered relevant for inclusion in this systematic review, their reference lists and the articles resulting of the “Times cited” function in WOS were hand-searched for additional pertinent publications. Reference lists of review articles were also hand-searched.

We also searched for open access graduate theses and dissertations on the website www.oatd.org on 7 December 2017, using the following search string: (mercury OR methylmercury) AND ("blood pressure" OR hypertension OR cardiovascular).

Inclusion criteria

Articles that met all these inclusion criteria were included in this systematic review:
  1. 1)

    Original researches from observational studies conducted in human populations.

     
  2. 2)

    Written in English, French, or Spanish.

     
  3. 3)

    Published up to 30 November 2017.

     
  4. 4)

    Exposure: chronic mercury exposure. According to the Agency for Toxic Substances and Disease Registry, we considered “chronic exposure” the contact with mercury that occurred over a long time (over than 1 year), in contrast to “acute exposure” (once or for only a short time, up to 14 days) and “intermediate duration exposure” (more than 14 days and less than 1 year) (Agency for Toxic Substances and Disease Registry 2018).

     
  5. 5)

    Outcome: blood pressure (systolic and/or diastolic) levels or hypertension.

     
  6. 6)

    Population: children and/or adolescents (from birth to age 18).

     
  7. 7)

    Eligible studies had to quantify the association between mercury and blood pressure and to show separate results for the adult and paediatric populations if populations were mixed.

     

Selection of studies

We performed the study selection through the following exclusion process: the first author (GGV) and the corresponding author reviewed the title and abstract of the studies and excluded those that did not meet the inclusion criteria mentioned above. Then, both authors read the full text of the remaining studies and excluded those that did not meet the inclusion criteria.

Similarly, GGV read the title and abstract of the theses and dissertations identified. The first author also excluded those references that did not meet the inclusion criteria mentioned above. Finally, she searched for the inclusion criteria in the text of two doctoral theses whose inclusion we had doubts about.

Data extraction

The first author contacted one author for a manuscript we were unable to access. The full text of the remaining articles was systematically reviewed by three authors. We followed the criteria suggested by the STROBE Statement (Vandenbroucke et al. 2007) for the data extraction. For each relevant study identified, we extracted and tabulated the following data: study design, location, participants, relevant periods (recruitment, exposure, follow-up, data collection), inclusion criteria, study size, child’s age at data collection, variables (exposure, outcomes, and other covariates), data sources, exposure and outcome assessment, statistical methods, main results (confounded-adjusted estimates and their precision), strengths, and limitations. If studies reported more than one multivariable model, we recorded the model adjusted for the most covariates. Conflicts were resolved with the agreement of all authors.

Study quality

Different bias assessment tools are used for assessing quality and susceptibility to bias in observational studies in epidemiology (Sanderson et al. 2007; Cascaes da Silva et al. 2013). We used a bias assessment tool designed for single use in this systematic review to assess the risk of bias of the articles. Quality evaluation criteria included the evaluation of six types of biases: sample selection, exposure assessment, outcome assessment, confounder adjustment, analytical issues, and attrition. Each criterion, except attrition, was classified according to risk of bias: minimal, low, moderate, or high. The bias assessment tool is provided in Supplementary File Table 1.

Statistical analysis

We had initially considered of summarising the results as a meta-analysis, but heterogeneity in the study design, exposure assessment, and outcome assessment, as well as methodological limitations across the studies, prevented us from combining these results. Thus, the results are qualitatively summarised between similar exposure groups.

Results

Search results

A total of 1033 studies were identified from all databases and 66 theses and dissertations from the website mentioned above (Fig. 1). Of the 1033 retrieved studies, only 17 studies met our inclusion criteria based on a review of the title and abstract. After reading the full text of those 17 studies, 2 studies were excluded because their outcome was not BP. Seven papers studied effects of chronic mercury exposure on BP, including children and adults in their study population, but since results were not shown separately, they were also excluded. A total of 8 articles were considered relevant for inclusion in our systematic review. We reviewed the reference lists of relevant studies and reviews, as well as their citing articles from WOS, and we did not identify additional studies that met the inclusion criteria.
Fig. 1

Search and selection of articles for systematic review on chronic mercury exposure and blood pressure in children and adolescents

Additionally, we read the title and abstract of the 66 theses and dissertations identified. Six of them were not conducted in humans, 2 were written in a language other than English, French, or Spanish, and 56 did not assess the possible association of mercury and BP, so they were excluded. After reading the sections “Study population” and “Results” from the dissertation written by Mozhgon Rajaee in 2015, this dissertation was excluded as the study population was adult. Likewise, according to “Study design and sample” section in the master’s thesis written by Trine Louise Jul Larsen in 2017, population included was only adult. In consequence, none of those theses met our inclusion criteria.

Quality assessment

The quality assessment for the studies included in this systematic review is summarised in Table 1. In most studies, risk of bias was considered minimal or moderate, except in one study in which exposure assessment and analysis had some limitations (Poursafa et al. 2014). In this last study, exposure to mercury was assessed after blood pressure measurement. Moreover, the biomarker used (mercury in serum) was not the most appropriate to evaluate a chronic exposure to mercury (Joint FAO/WHO Expert Committee on Food Additives (JECFA) 2011; Ruggieri et al. 2017). In addition, some inaccuracies in the statistical analyses, such as mistakes in mercury categorisation and risk factor values, were identified.
Table 1

Quality of the articles based on the evaluation of biases

Reference

Study design

Risk of bias

Selection

Exposure

Outcome

Confounder

Analytical

Sørensen et al. (1999)

Cohort

Minimal

Minimal

Low

Low

Minimal

Grandjean et al. (2004)

Cohort

Minimal

Minimal

Minimal

Low

Minimal

Thurston et al. (2007)

Cohort

Minimal

Minimal

Minimal

Moderate

Minimal

Valera et al. (2011)

Cross-sectional

Minimal

Minimal

Minimal

Minimal

Minimal

Valera et al. (2012)

Cohort

Minimal

Minimal

Minimal

Minimal

Minimal

Kalish et al. (2014)

Cohort

Minimal

Minimal

Moderate

Minimal

Minimal

Poursafa et al. (2014)

Case-control

Minimal

High

Moderate

Moderate

High

Gregory et al. (2016)

Cohort

Minimal

Minimal

Minimal

High

Low

Attrition in the 5 reviewed birth cohort studies was different. In the Faroese cohort, Sørensen et al. (1999) examined 89.7% of initial participants at 7 years and Grandjean et al. (2004) examined 85.9% of initial participants at age 14. In the Seychelles cohort, 83.4% of initial participants were examined at age 12 and 72.9% at age 15 (Thurston et al. 2007). In the Inuit cohort, Valera et al. (2012) examined 49.5% of initial participants. Kalish et al. (2014), in the US cohort, examined 48.4% of live births at age 3 and 40.6% of live births at age 8. In ALSPAC cohort, Gregory et al. examined 49.3% of the offspring for whom there was a prenatal blood mercury measurement at age 7, 47.4% at age 9, 43.5% at age 11, 34.3% at age 13, 33.3% at age 15, and 28.3% at age 17 (Gregory et al. 2016).

Descriptive analysis of the included studies

Date, location, design, and study population

We identified 8 relevant articles, published from 1999 to 2016, from the following observational studies: The Faroese cohort (Sørensen et al. 1999; Grandjean et al. 2004), the cohort of the Seychelles Child Development Study (Thurston et al. 2007), a cross-sectional study in 4 islands in French Polynesia (Valera et al. 2011), the cohort of Inuit children from Nunavik (Arctic Québec, Canada) (Valera et al. 2012), the Project Viva, a cohort study in Massachussets (USA) (Kalish et al. 2014), a case-control study in 27 provincial counties in Iran (Poursafa et al. 2014), and the ALSPAC cohort in UK (Gregory et al. 2016).

All cohort studies were birth cohorts and they covered mean ages from 3 to 17. The cross-sectional study (Valera et al. 2011) and the case-control study (Poursafa et al. 2014) only included teenagers (mean ages 14–15). The characteristics of these studies are summarised in Table 2.
Table 2

Descriptive analysis of the studies on mercury exposure and blood pressure (BP) at paediatric age

Reference

Location

Study period

Study design

Inclusion criteria

Sample size

Age, in years, at BP measurementa

Covariates consideredb

Sørensen et al. (1999)

Faroe Islands

1986–1994

Cohort

Singleton births

917

6.9 (0.3)

2, 4, 5, 6, 8, 11, 12, 13, 14, 15, 16

20, 23, 24, 25, 26

30, 31, 32, 33, 34, 36

Grandjean et al. (2004)

Faroe Islands

1986–2001

Cohort

Singleton births

878

13.8 (0.3)

1, 2, 4, 5, 6

20, 23, 25, 26

31, 32, 33, 34, 37, 38, 39, 41

Thurston et al. (2007)

Seychelles Islands

1989–2004

Cohort

Mother-child pairs without medical problems that might seriously affect development

644

Boys: 12.7 (range 11.7–13.4)

Girls: 12.6 (range 11.6–13.4)

2

23, 26

31, 32, 33, 34

559

Boys: 15.4 (range 14.8–16.2)

Girls: 15.4 (range 14.7–16.3)

Valera et al. (2011)

French Polynesia

2007

Cross-sectional

Teenagers born on the Austral Islands and living in Papeete (Tahiti) or on 3 austral islands. Included students from 2 colleges or from secondary school

101

14.2 (1.5)

26

31, 32, 33, 34, 35, 38, 39, 40, 43, 44, 45, 47, 49, 51, 53

Valera et al. (2012)

Nunavik, Canada

1993–2010

Cohort

Pregnant women who arrived for delivery and all newborns born at 2 health centers

226

11.3 (0.6)

1, 2, 4, 5

20, 22, 23, 26

31, 32, 33, 34, 47, 48, 49, 50, 51, 52, 53, 54, 55

Kalish et al. (2014)

Massachussets, USA

1999–2008

Cohort

Pregnant women with initial prenatal visit at < 22 weeks of gestation, singleton pregnancy, did not plan to move away from the study area prior to delivery and could complete study forms in English

1031

3.3 (0.3)

2, 3, 4, 5, 6, 7, 9, 10, 11, 12

20, 21, 22, 23, 25, 26

30, 31, 32, 33, 34, 42, 46, 47

865

7.9 (0.8)

Poursafa et al. (2014)

Iran

2009–2010

Case-control

Students without any history of acute or chronic diseases and any medication use

320

15.0 (2.6)

26

31, 34, 35, 36

Gregory et al. (2016)

UK

1991–2008

Cohort

Mother-child pairs residing in Avon

1268–2207

7, 9, 11, 13, 15, 17

4, 5, 6, 9, 17, 18, 19

25, 26

30, 46

BP blood pressure

aMean (standard deviation) or as indicated

bCovariates considered: Parental: 1, prenatal methylmercury exposure; 2, maternal hypertension or family history of hypertension; 3, maternal third trimester SBP; 4, alcohol during pregnancy; 5, smoking during pregnancy; 6, maternal age; 7, maternal race/ethnicity; 8, parental race/ethnicity; 9, maternal education level; 10, marital status; 11, pre-pregnancy weight; 12, pre-pregnancy body mass index; 13, mother unskilled; 14, father unskilled; 15, father unemployed; 16, day-care; 17, social factors (family adversity score, housing tenure, household crowding, stressful life events in the 1st half of pregnancy); 18, selenium in mother blood during pregnancy; 19, fish intake during pregnancy; Child at birth: 20, gestational age; 21, foetal growth z score; 22, birth length; 23, birth weight; 24, placenta weight; 25, parity; 26, sex; Child at BP measurement: 30, breastfeeding duration; 31, age at testing blood pressure; 32, child weight; 33, child height; 34, child body mass index; 35, child waist circumference; 36, child residence; 37, physical activity; 38, teenager smoking habits; 39, teenager alcohol consumption; 40, child anti-hypertensive treatment; 41, examiner (BP measurement); 42, BP measurement conditions (child state and position, arm used, cuff size, measurement sequence number); 43, fasting glucose; 44, fasting insulin; 45, triglycerides; 46, maternal fish intake during pregnancy; 47, DHA + EPA intake; 48, total n-3 PUFAs (cord blood); 49, total n-3 PUFAs (child blood); 50, selenium (cord blood); 51, selenium (child blood); 52, lead (cord blood); 53, lead (child blood); 54, PCB 153 (cord blood); 55, PCB 153 (child blood)

Blood pressure measurement

All studies showed results of BP as a quantitative variable (BP levels), although 1 of them (Kalish et al. 2014) only showed results for SBP.

Blood pressure was measured according to different guidelines and protocols. In addition, conditions of measurement were not totally reported (Table 3). Thurston et al. (2007) extracted children’s BP data from a public database, while, in the rest of identified studies, BP was measured in the office. Three studies used auscultatory technique (Sørensen et al. 1999; Grandjean et al. 2004; Valera et al. 2011, 2012; Gregory et al. 2016), 3 studies used oscillometric technique (Thurston et al. 2007; Kalish et al. 2014; Gregory et al. 2016), and a further study (Poursafa et al. 2014) did not report the device and measurement technique. Regarding cuff location, 2 studies used the left arm (Sørensen et al. 1999; Grandjean et al. 2004), 1 used the right arm (Thurston et al. 2007), 1 used an unidentified arm (Kalish et al. 2014), and 4 did not report it. The number of BP readings ranged from 1 to 5. The use of data to calculate mean BP levels was also heterogeneous.
Table 3

Blood pressure measurement conditions reported in identified studies assessing mercury exposure and blood pressure at paediatric age

Reference

Subjects (prior to measurement)

Device

Conditions of measurement

Measurement result

Physical and mental relaxation

Avoiding circumstances

Method (operation)

Type/model

Hour of measurement

Activity of child

Position

Cuff (location)

Environment (place, temperature,…)

Observers

Number of readings (interval between them)

Use of data to calculate SBP/DBP means

Sørensen et al. (1999)

Standardised conditions (relaxing in a chair)

NR

Auscultation (manual)

Sphygmomanometer

Each day: 4 in the morning, 4 in the afternoon

NR

Sitting

Covered ½–2/3 of upper arm (left arm)

NR

Health service professionals

1

--

Grandjean et al. (2004)

Child relaxing in a chair

NR

Auscultation (manual)

NR

NR

NR

Sitting

Covered ½-2/3 of upper arm (left arm)

NR

2 paediatricians

3 (NR)

Mean of 3

Thurston et al. (2007)

Rest for several minutes seated with their arm comfortably resting on a table

NR

Oscilometry (automatic)

Omron, HEM 711 AC

NR

NR

Sitting

Paediatric, standard or large (right arm)

Routine school health examinations

Trained school nurses

2 (≥ 1 min)

Mean of 2

Valera et al. (2011)

Rest for 5 min

Not eaten or smoked for at least 30 min

Auscultation

Mercury sphygmomanometers, 15-in. stethoscopes

NR

NR

NR

Cuffs sized to the subjects’ arms (NR)

NR

NR

3 (NR)

Mean of last 2

Valera et al. (2012)

Seated position for 5 min

Not eaten or smoked for at least 30 min

Auscultation

Mercury sphygmomanometer

NR

NR

Sitting

Appropriate cuff size (NR)

NR

NR

3 (NR)

Mean of last 2

Kalish et al. (2014)

NR

NR

Oscilometry (automatic)

Dinamap Pro 100 or Pro 200 (Critikon Inc.)

NR

3 years: sleeping, quiet awake, active awake, crying

8 years: quiet, still, talking, moving

Sitting, semi-reclining or standing

Child, small adult, adult, large adult (arm)

NR

Trained research assistants

Up to 5 times (1 min)

Mean of the (up to) 5

Poursafa et al. (2014)

NR

NR

NR

Calibrated instruments

NR

NR

NR

NR (NR)

NR

A trained team of health professionals

NR

NR

Gregory et al. (2016)

Relaxing atmosphere (13 years), fasting conditions (15 years), NR (7, 9, 11, 17 years)

NR

Oscilometry (automatic)

Dinamap 9301 Vital Signs Monitor

Morning or afternoon

Silent, talking or fidgeting

NR

NR

A specially designed clinic (temperature recoded but NR)

NR

2 (NR)

Mean of 2

DBP diastolic blood pressure, NR not reported, SBP systolic blood pressure

Mercury exposure assessment

Four studies (Sørensen et al. 1999; Thurston et al. 2007; Kalish et al. 2014; Gregory et al. 2016) assessed prenatal exposure, 2 evaluated both pre and postnatal exposures (Grandjean et al. 2004; Valera et al. 2012), and 2 evaluated postnatal exposure (Valera et al. 2011; Poursafa et al. 2014).

Matrices used for assessing mercury exposure are shown in Table 4. All the studies determined total mercury in their respective matrices by different analytical methods. Cord blood was used in the cohorts of Faroe Islands (Sørensen et al. 1999; Grandjean et al. 2004) and Nunavik (Valera et al. 2012). Maternal hair was used in 2 studies (Sørensen et al. 1999; Thurston et al. 2007). Erythrocyte mercury in maternal blood during pregnancy was used in 1 study (Kalish et al. 2014). Two studies used child hair (Grandjean et al. 2004; Valera et al. 2012), 3 studies used child whole blood (Valera et al. 2011, 2012; Gregory et al. 2016), and 1 study used child blood serum (Poursafa et al. 2014) to assess postnatal mercury exposure. Two of these 7 studies used 2 biomarkers of pre or postnatal exposure (Sørensen et al. 1999; Valera et al. 2012).
Table 4

Association between chronic mercury exposure and blood pressure at paediatric age

Reference/study

N

Age/s (years)

Mercury exposure

Effect on blood pressure

Model adjustmentsd

Matrix

Levela

Change unit

Measure of effect

SBP

DBP

Prenatal exposure

 Sørensen et al. (1999)

894

7

Cord blood

31.77 μg/L

A tenfold increase in Hg exposure

Change in BP level (mmHg) (95% CI)

14.6 (8.3–20.8)

13.9 (7.4–20.4)

SBP: 32

DBP: 2, 32

914

7

Maternal hair at parturition

5.65 μg/g

8.6 (0.9–16.2)

3.6 (− 4.3–11.5)

 Grandjean et al. (2004)

837

14

Cord blood

22.6 μg/L

Per doubling of Hg exposure

Adjusted β (p value)

0.045 (0.84)

0.121 (0.60)

2, 5

23, 26

31, 32, 33, 37, 38, 41

 Thurston et al. (2007)

644

12

Maternal hair during pregnancy

Boys: 6.6 μg/g

Girls: 7.0 μg/g

Per 1 unit increase in Hg (μg/g)

Adjusted β (95% CI)c (p value)

0.07 (− 0.11 to 0.25) (0.46)

− 0.03 (− 0.17 to 0.11) (0.62)

2

23, 26

31, 32, 33, 34

559

15

Maternal hair during pregnancy

Boys: 6.5 μg/g

Girls: 7.0 μg/g

Per 1 unit increase in Hg (μg/g)

Adjusted β (95% CI)c (p value)

0.10 (− 0.10 to 0.30) (0.32)

0.17 (0.03 to 0.31) (0.02)

Boys: 0.36 (0.12 to 0.60) (0.003)

Girls: 0.05 (− 0.15 to 0.25) (0.59)

2

23

31, 33, 34

 Valera et al. (2012)

226

11

Cord blood

107.3 nmol/Lb

Per 1 unit increase in Hg (nmol/L)

Adjusted β (p value)

0.12 (0.13)

− 0.15 (0.09)

5

23, 26

31, 33, 34, 48, 50, 52, 54

 Kalish et al. (2014)

1103

3, 8

Maternal blood at 2nd trimester of pregnancy

4.0 ng/g erythrocytes

Hg quartiles:

Q1 1.0 ng/g

Q2 2.2 ng/g

Q3 3.8 ng/g

Q4 7.0 ng/g

Adjusted β (95% CI) across quartiles

Q1: ref

Q2: − 0.1 (− 1.5 to 1.3)

Q3: 0.8 (− 0.6 to 2.2)

Q4: 0.0 (− 1.5 to 1.5)

--

3, 5, 6, 7, 9, 10, 12

21, 26

31, 34, 42, 46, 47

 Gregory et al. (2016)

2207 (7 years)

2125 (9 years)

1950 (11 years)

1540 (13 years)

1494 (15 years)

1268 (17 years)

7, 9, 11, 13, 15, 17

Maternal blood at 1st trimester of pregnancy

Median: 1.86 μg/L

Per 1 SD increase in Hg exposure

Adjusted β (95% CI)

7 years: 0.10 (− 0.31 to 0.51)

9 years: 0.07 (− 0.34 to 0.49)

11 years: 0.21 (− 0.26 to 0.69)

13 years: − 0.03 (− 0.56 to 0.51)

15 years: 0.19 (− 0.48 to 0.85)

17 years: 0.15 (− 0.44 to 0.73)

7 years: 0.24 (− 0.04 to 0.52)

9 years: 0.27 (− 0.00 to 0.55)

11 years: 0.09 (− 0.21 to 0.39)

13 years: 0.27 (− 0.05 to 0.59)

15 years: − 0.10 (− 0.59 to 0.38)

17 years: − 0.01 (− 0.38 to 0.36)

SBP: 4,5,6, 9, 17, 18, 25, 30

DBP: 4,5,6, 9, 17, 25, 30

Postnatal exposure

 Grandjean et al. (2004)

837

14

Child hair

0.96 μg/g

Per doubling of Hg exposure

Adjusted β (p value)

− 0.017 (0.91)

− 0.086 (0.61)

1, 2, 5

23, 26

31, 32, 33, 37, 38, 41

 Valera et al. (2011)

101

14

Child blood

GM: 8.1 μg/L

Hg tertiles:

T1: 1.2–7.3

T2: 7.4–10.0

T3: 11.0–26.0

Adjusted mean ± SE across tertiles (p value)

T1: 109 ± 2.9

T2: 112 ± 2.8

T3: 113 ± 2.9

(0.38)

T1: 71 ± 2.4

T2: 72 ± 2.3

T3: 74 ± 2.3

(0.49)

26

31, 35, 43, 45, 49, 51

 Valera et al. (2012)

226

11

Child blood

22.4 nmol/Lb

Per 1 unit increase in Hg (nmol/L)

Adjusted β (p value)

0.15 (0.11)

0.15 (0.15)

1, 5

23, 26

31, 33, 34, 48, 50, 52, 54

Child hair

6.61 nmol/gb

Per 1 unit increase in Hg (nmol/g)

Adjusted β (p value)

0.09 (0.31)

− 0.04 (0.63)

 Poursafa et al. (2014)

320

15

Child blood serum

0.17 μg/L in cases and 0.10 μg/L in controls

Hg quartiles:

Q1 ≤ 0.5

Q2 0.6–0.7

Q3 0.8–0.9

Q4 ≥ 1.0

Adjusted mean (SD) across quartiles (p value)

Girls:

Q1: 110.91 (15.03)

Q2: 109.59 (16.15)

Q3: 109.35 (17.29)

Q4: 125.73 (9.77)

(0.007)

Girls:

Q1: 69.09 (15.71)

Q2: 67.73 (13.39)

Q3: 69.94 (10.44)

Q4: 81.96 (5.76)

(0.003)

26

31, 34, 35

Boys:

Q1: 108.42 (16.25)

Q2: 107.49 (15.03)

Q3: 109.79 (14.42)

Q4: 122.77 (16.45)

(0.008)

Boys:

Q1: 68.89 (13.16)

Q2: 71.17 (10.85)

Q3: 72.10 (11.39)

Q4: 79.30 (6.79)

(0.002)

β regression coefficient, N sample size, SE standard error, SD standard deviation, GM geometric mean, CI confidence interval, SBP systolic blood pressure, DBP diastolic blood pressure

aArithmetic mean of total mercury, or as indicated

bResults given in nanomol may be converted to micrograms by dividing by 5

cResults given in “Adjusted β (SE)” have been converted to “Adjusted β (CI 95%)” by these formulas: Lower Limit CI 95% = mean − 1.96 × SE and Upper Limit CI 95% = mean + 1.96 × SE.

dCovariates considered: Parental: 1, prenatal methylmercury exposure; 2, maternal hypertension or family history of hypertension; 3, maternal third trimester SBP; 4, alcohol during pregnancy; 5, smoking during pregnancy; 6, maternal age; 7, maternal race/ethnicity; 8, parental race/ethnicity; 9, maternal education level; 10, marital status; 11, pre-pregnancy weight; 12, pre-pregnancy body mass index; 13, mother unskilled; 14, father unskilled; 15 father unemployed; 16, day-care; 17, social factors (family adversity score, housing tenure, household crowding, stressful life events in the 1st half of pregnancy); 18, selenium in mother blood during pregnancy; 19, fish intake during pregnancy; Child at birth: 20, gestational age; 21, foetal growth z score; 22, birth length; 23, birth weight; 24, placenta weight; 25, parity; 26, sex; Child at BP measurement: 30, breastfeeding duration; 31, age at testing blood pressure; 32, child weight; 33, child height; 34, child body mass index; 35, child waist circumference; 36, child residence; 37, physical activity; 38, teenager smoking habits; 39, teenager alcohol consumption; 40, child anti-hypertensive treatment; 41, examiner (BP measurement); 42, BP measurement conditions (child state and position, arm used, cuff size, measurement sequence number); 43, fasting glucose; 44, fasting insulin; 45, triglycerides; 46, maternal fish intake during pregnancy; 47, DHA + EPA intake; 48, total n-3 PUFAs (cord blood); 49, total n-3 PUFAs (child blood); 50, selenium (cord blood); 51, selenium (child blood); 52, lead (cord blood); 53, lead (child blood); 54, PCB 153 (cord blood); 55, PCB 153 (child blood).

In bold, p values < 0.05 considered as statistically significant

Means of mercury levels in different studies ranged from 21.5 to 31.8 μg/L in cord blood, from 0.17 to 12.76 μg/L in maternal blood during pregnancy, from 5.6 to 7.0 μg/g in maternal hair, from 1.0 to 1.3 μg/g in child hair, from 4.5 to 8.1 μg/L in child blood, and from 0.1 to 0.2 μg/L in child serum. Mean mercury level was 4.0 ng/g erythrocytes in maternal blood during pregnancy (Table 4).

Adjustment for covariates

Covariates considered in each article are shown in Table 2. All studies, except for Gregory et al. (2016), adjusted by child anthropometric covariates at BP measurement (sex, age, weight or BMI, height). All the cohort studies, except for Gregory et al. (2016), adjusted by family history of hypertension (Grandjean et al. 2004; Kalish et al. 2014) or maternal hypertension during pregnancy (Sørensen et al. 1999; Thurston et al. 2007; Valera et al. 2012). The cross-sectional study (Valera et al. 2011) and the case-control study (Poursafa et al. 2014) did not adjust by this factor.

Sørensen et al. (1999), Grandjean et al. (2004), Valera et al. (2012), Kalish et al. (2014), and Gregory et al. (2016) adjusted by parental sociodemographic covariates (prenatal exposures like alcohol consumption or smoking during pregnancy, maternal age, race/ethnicity, educational level,…). Grandjean et al. (2004) and Valera et al. (2011) also adjusted by adolescent consumption of alcohol and tobacco.

Both studies conducted by Valera et al. (2011, 2012) considered metabolic and dietary factors, such as polyunsaturated fatty acids (PUFAs) and selenium levels (measured in blood), as potential confounders, along with other environmental pollutants such as polychlorinated biphenyls (PCBs) and lead. Kalish et al. (2014) and Gregory et al. (2016) used fish consumption during pregnancy as a potential confounder. Polyinsaturated fatty acids intake was also used as a confounder by Kalish et al. (2014).

Grandjean et al. (2004) adjusted for examiner and Kalish et al. (2014) adjusted for several conditions of measurement (activity of the child, cuff size, arm used, position, and measurement sequence number). The rest of the studies did not reported if they had adjusted their analysis for any condition of measurement.

Mercury exposure and blood pressure

The main results of identified studies on mercury exposure and BP in children and adolescents are reported in Table 4.

Five studies (6 articles) assessed the relationship between prenatal mercury exposure and BP in children or adolescents. The Faroese study found a positive significant association between prenatal exposure to mercury and BP at age 7, with a stronger effect in low birth weight children (Sørensen et al. 1999), but this association disappeared at age 14 (Grandjean et al. 2004). In the Seychelles Islands, Thurston et al. (2007) found a positive and significant association between prenatal exposure and DBP in boys (not in girls) at age 15, but this association was not significant at age 12. Valera et al. (2012) found a positive association between prenatal mercury exposure and SBP and a negative association with DBP in Inuit children at age 11, but none were statistically significant. Kalish et al. (2014) did not find any significant association between prenatal mercury exposure and BP in American children. Finally, in the UK, Gregory et al. (2016) found a positive significant association between prenatal exposure and SBP at age 11 whose mothers were non-fish eaters, but this association was not significant in other age groups not even in the offspring of mothers who ate fish.

Two studies assessed postnatal exposure taking into account prenatal exposure as a potential confounder. Grandjean et al. (2004) re-examined the Faroese cohort at age 14 and observed no significant association between postnatal mercury exposure and BP. Mercury exposure was not significantly associated with both SBP and DBP in the study conducted by Valera et al. (2012) in Inuit children at age 11.

No significant association was found between postnatal exposure and BP among French Polynesian teenagers at age 14 (Valera et al. 2011). However, SBP and DPB had a significant increase across mercury quartiles among Iranian adolescents at age 15 (Poursafa et al. 2014).

Discussion

In this systematic review, we summarised the state-of-the-art in the investigation of the relation of mercury exposure on BP among children and adolescents. The studies identified showed heterogeneous results, which could in part be explained by large differences between them in terms of study design, study population, assessment of mercury exposure, characteristics of BP measurement, and covariates considered.

Mercury exposure and blood pressure

Population at 4 studies carried out until 2012 (Sørensen et al. 1999; Grandjean et al. 2004; Thurston et al. 2007; Valera et al. 2011, 2012) presented high mercury levels. Subsequent 3 studies included in this systematic review (Kalish et al. 2014; Poursafa et al. 2014; Gregory et al. 2016) have been carried out with populations with lower mercury levels than previous studies mentioned above, although these levels were similar to those assessed around the world (Sheehan et al. 2014).

Four of the 7 studies found a positive significant association between chronic mercury exposure and blood pressure in children or adolescents (Sørensen et al. 1999; Thurston et al. 2007; Poursafa et al. 2014; Gregory et al. 2016). Among these 4 studies, 3 of them evaluated prenatal exposure (Sørensen et al. 1999; Thurston et al. 2007; Gregory et al. 2016) and 1 evaluated postnatal exposure to mercury (Poursafa et al. 2014). This could suggest that the foetal period is the most vulnerable, although more studies are required to confirm this.

Two of the 3 studies that found significant associations between prenatal exposure to mercury and offspring BP included highly exposed populations (Sørensen et al. 1999; Thurston et al. 2007). They are high fish-consuming populations, so the predominant toxic form of mercury may be MeHg (National Research Council (NRC) 2000). It is worth highlighting that Thurston et al. (2007) found a significant association in boys, but not in girls. There is another study of prenatal exposure to mercury with similar levels but it did not find any significant association (Valera et al. 2012). The third study that evaluated prenatal exposure to mercury with significant associations with BP in children was conducted in a less-exposed population (Gregory et al. 2016), although their authors concluded that it could be a spurious association. No significant association was found in another study assessing prenatal exposure in which mercury levels were also lower (Kalish et al. 2014). This might indicate the existence of a threshold value of mercury exposure below which no BP effects are expected.

In addition, Sørensen et al. (1999) found that the greatest mercury-associated changes in BP occurred in children with cord blood mercury levels below 10 μg/L, showing a linear association. However, over 10 μg/L, BP was not associated with mercury exposure. Thurston et al. (2007) also found this pattern in adolescents. This could suggest the existence of a ceiling effect in the toxic effects of mercury on BP. Further studies are needed to confirm the possible existence of a threshold value and/or a ceiling effect in prenatal mercury exposure and its effects on BP.

Regarding the 4 studies that evaluated postnatal exposure to mercury, 3 of them with the most exposed populations did not find any significant association. These three populations were highly exposed to MeHg due to seafood consumption (Grandjean et al. 2004; Valera et al. 2011, 2012). One study found a significant association between postnatal mercury exposure and BP, but the risk of bias of this study was considered to be high in exposure assessment and analysis of data (Poursafa et al. 2014). Therefore, more studies are needed to clarify the possible relationship between postnatal exposure and BP at paediatric age.

In reference to the biological plausibility of the association between mercury and blood pressure, Sørensen et al. (1999) suggested that MeHg could affect targets involved in BP regulation, as cation regulation (Calcium channels) and originate parasympathetic dysfunction. Nowadays, it is known that mercury has a high affinity for the sulfhydryl group (-SH), inactivating numerous enzymes, amino acids, and sulphur-containing antioxidants. Mercury also induces mitochondrial dysfunction and lipid peroxidation and promotes platelet aggregation and blood coagulation (Houston 2014). The decreased oxidant defence and increased oxidative stress is an early biological response that can produce vascular endothelial cell damage by promoting inflammation and vasoconstriction (Roman et al. 2011).

Quality assessment

The exposure assessment is a key factor in epidemiological studies. Thus, the use of inappropriate indicators of mercury exposure leads to the danger of getting biased results. It has been reported that the best matrices to measure total mercury, as a biomarker of methylmercury, are hair, blood, and erythrocytes (WHO 2010). Therefore, the measurement of mercury in other matrices such as serum (Poursafa et al. 2014) could lead to misclassification (Berglund et al. 2005; Grandjean et al. 2005). Furthermore, in this study (Poursafa et al. 2014), exposure was measured after health effect, so risk of exposure bias was considered high.

The measurement of BP requires standardised conditions and accurate instruments (Stergiou et al. 2012). None of the 3 studies that assessed BP levels with oscillometric technique used a device validated for paediatric population (Thurston et al. 2007; Kalish et al. 2014; Gregory et al. 2016) (for an updated list of validated devices, see www.dableducational.org, accessed April 21 2018). European guidelines recommend taking 3 BP measurements by manual auscultation, with an interval of 3 min between measurements and use the average of the last two (Lurbe et al. 2016). According to American guidelines, when the initial BP reading at an office visit is elevated (≥ 90th percentile), it is necessary to obtain 2 additional readings at the same visit and average them and, if it is used an oscillometric technique, discard the first one and average the subsequent readings (Flynn et al. 2017). Only Sørensen et al. (1999) took 1 reading of BP, so we expected a biased measurement of BP in this article, specifically, increased levels of SBP and DBP (Lurbe et al. 2016; Flynn et al. 2017).

At paediatric age, it is known that BP is a very variable parameter that presents normal values according to sex and it also increases progressively with body growth and development. Therefore, it was considered that the study that did not adjusted by the anthropometric characteristics of children had a high risk of confounding bias (Gregory et al. 2016). Furthermore, it is recognised that some nutrients present in fish (such as selenium and PUFAs) may confound the relationship between mercury and cardiovascular diseases, including high BP (Mozaffarian and Rimm 2006; Choi et al. 2008; WHO-UNEP 2008; Mozaffarian and Wu 2011). Although essential fatty acids from fish may reduce the risk of acute coronary events, mercury in fish could attenuate this beneficial effect (Mozaffarian and Rimm 2006). Consequently, risk of confounding bias was considered minimal in studies including these confounders (Valera et al. 2011, 2012; Kalish et al. 2014).

Strengths and limitations

The principal strength of this systematic review is the exhaustiveness of the search strategy. The selection of databases was done in order to maximise geographic, temporal, and thematic coverage (Abad García et al. 2015). Lilacs database was included to cover Latin-American and Caribbean literature and IME database in order to extend temporal coverage of Spanish journals. Furthermore, the search was extended to give access to theses and dissertations. Four of the 6 databases used (PubMed, Embase, Scopus, WOS) recovered all articles included in this review. Lilacs recovered 3 articles (43%) and IME did not recover any article. Other strength of this study was the performance of a quality assessment of primary studies.

This systematic review has several limitations. First, we restricted the literature search to English, French, and Spanish languages. Secondly, we could not perform a meta-analysis, due to the high heterogeneity in the studies, so results of this review are qualitative.

Recommendations for future research

The evidence on early exposure to mercury and BP in children and adolescents is still limited and heterogeneous. The number of epidemiological studies addressing this issue is scarce and we have not identified a clear pattern of mercury exposure and BP effects among the reviewed studies. Additional research is warranted and some considerations should be taken into account.

Recommendations for future research include cohort-based studies assessing the relationship between prenatal exposure to mercury and BP, in order to clarify if the foetal period is the most vulnerable window of exposure.

The possible association between postnatal mercury exposure and BP is not yet elucidated. Future cohort studies should consider both pre and postnatal exposures.

The potential for nonlinear dose-response relationships (e.g. a threshold dose response or ceiling effect) should be considered, as well as the possibility of long-term effects of mercury exposure.

Regarding the assessment of mercury exposure, future studies could improve the precision of prenatal exposure biomarkers by adjusting maternal blood or cord blood concentrations by haemoglobin (Kim et al. 2014). Furthermore, biomarker imprecision should be considered (1) assessing biomarker imprecision with three independent exposure indicators or (2) incorporating an imprecision of 25–50% in sensitive analyses (Grandjean and Budtz-Jørgensen 2010).

Differences in individual susceptibilities to mercury exposure with respect to genetics could be also considered in future research. Some genetic polymorphisms with influence on mercury metabolism have been identified (Llop et al. 2015) especially those linked to glutathione. Some of these polymorphisms have been also linked to an increasing risk of myocardial infarction (Nakamura et al. 2002) and hypertension (Hao et al. 2011) in adults. However, very few studies have assessed the influence of the interaction between mercury and these polymorphisms on cardiovascular outcomes, especially among the younger population. More research on this topic is warranted in order to elucidate the genetic susceptibility for mercury at early ages.

Another topic that should be explored in further studies is the effect modification caused by the children’s gender. Despite evidence of possible differences in mercury neurotoxicity between the genders (Llop et al. 2013), only Thurston et al. (2007) among the reviewed studies considered this factor. Boys were more susceptible than girls to the cardiovascular toxicity of mercury. However, this pattern should be confirmed in further studies.

The standardisation of BP measurement protocols in childhood epidemiological studies, or at least the publication of complete measurement protocols, could improve the comparability of studies.

Finally, the inclusion of fish consumption as a confounder and other covariates previously not considered but currently underresearch (such as sodium intake, postnatal weight gain, or air pollution) could help to clarify the relationship between mercury exposure and BP at paediatric age (Geleijnse et al. 2002; Simkhovich et al. 2008; Foraster et al. 2014; Lurbe et al. 2014).

Conclusions

There are still relatively few studies on mercury exposure and BP in children and adolescents. The results of the identified studies were inconsistent, so, with the current available scientific evidence, it cannot be concluded that such an association exists.

Notes

Acknowledgments

We thank Beatriz Valera, Ph.D. for providing her manuscript and M. Francisca Abad, Ph.D. for her recommendations on databases and documentation.

Contributors

All the authors participated in the conception of the article. Gema Gallego-Viñas designed the search strategy. Gema Gallego-Viñas and Sabrina Llop searched for and selected relevant articles. All the authors participated in data extraction, analysis and interpretation of data, drafting the article, and final approval of the version to be published.

Funding

This work was supported by Miguel Servet-FEDER (MS 15/0025) and FIS-FEDER (16_1288).

Compliance with ethical standards

Competing interests

The authors declare that they have no conflict of interest.

Supplementary material

11356_2018_3796_MOESM1_ESM.docx (20 kb)
ESM 1 (DOCX 20 kb)
11356_2018_3796_MOESM2_ESM.docx (20 kb)
ESM 2 (DOCX 19 kb)

References

  1. Abad García MF, González Teruel A, Argento J, Rodríguez Gairín JM (2015) Characteristics and visibility of Spanish journals in Health Sciences and databases. Abad García María Francisca González Teruel Aurora M Argento Javier Rodríguez Gairín Josep Man 2015. Prof Inf 24(5):537–550.  https://doi.org/10.3145/epi.2015.sep.04 (in Spanish)Google Scholar
  2. Agency for Toxic Substances and Disease Registry (2018) Glossary of terms | ATSDR. https://www.atsdr.cdc.gov/glossary.html#G-A-. Accessed 28 Apr 2018
  3. Balbus JM, Barouki R, Birnbaum LS et al (2013) Early-life prevention of non-communicable diseases. Lancet Lond Engl 381:3–4.  https://doi.org/10.1016/S0140-6736(12)61609-2 CrossRefGoogle Scholar
  4. Berglund M, Lind B, Björnberg KA et al (2005) Inter-individual variations of human mercury exposure biomarkers: a cross-sectional assessment. Environ Health Glob Access Sci Source 4:20.  https://doi.org/10.1186/1476-069X-4-20 Google Scholar
  5. Bernhoft RA (2012) Mercury toxicity and treatment: a review of the literature. J Environ Public Health 2012:460508.  https://doi.org/10.1155/2012/460508 CrossRefGoogle Scholar
  6. Bose-O’Reilly S, McCarty KM, Steckling N, Lettmeier B (2010) Mercury exposure and children’s health. Curr Probl Pediatr Adolesc Health Care 40:186–215.  https://doi.org/10.1016/j.cppeds.2010.07.002 CrossRefGoogle Scholar
  7. Brady TM, Solomon BS, Neu AM et al (2010) Patient-, provider-, and clinic-level predictors of unrecognized elevated blood pressure in children. Pediatrics 125:e1286–e1293.  https://doi.org/10.1542/peds.2009-0555 CrossRefGoogle Scholar
  8. Brady TM, Stefani-Glücksberg A, Simonetti GD (2018) Management of high blood pressure in children: similarities and differences between US and European guidelines. Pediatr Nephrol Berl Ger.  https://doi.org/10.1007/s00467-018-3946-y
  9. Cascaes da Silva F, Arancibia V, Angélica B et al (2013) Evaluation lists and scales for the quality of scientific studies. Rev Cuba Inf En Cienc Salud 24:295–312 (in Spanish)Google Scholar
  10. Choi AL, Cordier S, Weihe P, Grandjean P (2008) Negative confounding in the evaluation of toxicity: the case of methylmercury in fish and seafood. Crit Rev Toxicol 38:877–893.  https://doi.org/10.1080/10408440802273164 CrossRefGoogle Scholar
  11. Dionne JM (2017) Updated guideline may improve the recognition and diagnosis of hypertension in children and adolescents; review of the 2017 AAP blood pressure clinical practice guideline. Curr Hypertens Rep 19:84.  https://doi.org/10.1007/s11906-017-0780-8 CrossRefGoogle Scholar
  12. Flynn JT, Kaelber DC, Baker-Smith CM et al (2017) Clinical practice guideline for screening and management of high blood pressure in children and adolescents. Pediatrics 140.  https://doi.org/10.1542/peds.2017-1904
  13. Foraster M, Basagaña X, Aguilera I et al (2014) Association of long-term exposure to traffic-related air pollution with blood pressure and hypertension in an adult population-based cohort in Spain (the REGICOR study). Environ Health Perspect 122:404–411.  https://doi.org/10.1289/ehp.1306497 CrossRefGoogle Scholar
  14. GBD 2016 Disease and Injury Incidence and Prevalence Collaborators (2017) Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Lond Engl 390:1211–1259.  https://doi.org/10.1016/S0140-6736(17)32154-2 Google Scholar
  15. Geleijnse JM, Giltay EJ, Grobbee DE et al (2002) Blood pressure response to fish oil supplementation: metaregression analysis of randomized trials. J Hypertens 20:1493–1499CrossRefGoogle Scholar
  16. Genchi G, Sinicropi MS, Carocci A et al (2017) Mercury exposure and heart diseases. Int J Environ Res Public Health 14.  https://doi.org/10.3390/ijerph14010074
  17. Gluckman PD, Cutfield W, Hofman P, Hanson MA (2005) The fetal, neonatal, and infant environments-the long-term consequences for disease risk. Early Hum Dev 81:51–59.  https://doi.org/10.1016/j.earlhumdev.2004.10.003 CrossRefGoogle Scholar
  18. Grandjean P, Budtz-Jørgensen E (2010) An ignored risk factor in toxicology: the total imprecision of exposure assessment. Pure Appl Chem Chim Pure Appl 82:383–391.  https://doi.org/10.1351/PAC-CON-09-05-04 CrossRefGoogle Scholar
  19. Grandjean P, Murata K, Budtz-Jørgensen E, Weihe P (2004) Cardiac autonomic activity in methylmercury neurotoxicity: 14-year follow-up of a Faroese birth cohort. J Pediatr 144:169–176.  https://doi.org/10.1016/j.jpeds.2003.10.058 CrossRefGoogle Scholar
  20. Grandjean P, Budtz-Jørgensen E, Jørgensen PJ, Weihe P (2005) Umbilical cord mercury concentration as biomarker of prenatal exposure to methylmercury. Environ Health Perspect 113:905–908.  https://doi.org/10.1289/ehp.7842 CrossRefGoogle Scholar
  21. Gregory S, Iles-Caven Y, Hibbeln JR et al (2016) Are prenatal mercury levels associated with subsequent blood pressure in childhood and adolescence? The Avon prebirth cohort study. BMJ Open 6:e012425.  https://doi.org/10.1136/bmjopen-2016-012425 CrossRefGoogle Scholar
  22. Gribble MO, Cheng A, Berger RD et al (2015) Mercury exposure and heart rate variability: a systematic review. Curr Environ Health Rep 2:304–314.  https://doi.org/10.1007/s40572-015-0053-0 CrossRefGoogle Scholar
  23. Grotto D, de Castro MM, Barcelos GRM et al (2009) Low level and sub-chronic exposure to methylmercury induces hypertension in rats: nitric oxide depletion and oxidative damage as possible mechanisms. Arch Toxicol 83:653–662.  https://doi.org/10.1007/s00204-009-0437-8 CrossRefGoogle Scholar
  24. Hansen ML, Gunn PW, Kaelber DC (2007) Underdiagnosis of hypertension in children and adolescents. JAMA J Am Med Assoc 298:874–879.  https://doi.org/10.1001/jama.298.8.874 CrossRefGoogle Scholar
  25. Hao Y, Wu B, Shi J et al (2011) Association of tag SNPs of GPx-3 with essential hypertension in rural Han Chinese in Fuxin, Liaoning, China. Chin Med J 124:2113–2116Google Scholar
  26. Hardy R, Lawlor DA, Kuh D (2015) A life course approach to cardiovascular aging. Futur Cardiol 11:101–113.  https://doi.org/10.2217/fca.14.67 CrossRefGoogle Scholar
  27. Holmes P, James KAF, Levy LS (2009) Is low-level environmental mercury exposure of concern to human health? Sci Total Environ 408:171–182.  https://doi.org/10.1016/j.scitotenv.2009.09.043 CrossRefGoogle Scholar
  28. Houston MC (2011) Role of mercury toxicity in hypertension, cardiovascular disease, and stroke. J Clin Hypertens Greenwich Conn 13:621–627.  https://doi.org/10.1111/j.1751-7176.2011.00489.x CrossRefGoogle Scholar
  29. Houston MC (2014) The role of mercury in cardiovascular disease. J Cardiovasc Dis Diagn 2014.  https://doi.org/10.4172/2329-9517.1000170
  30. Islam MZ, Van Dao C, Shiraishi M, Miyamoto A (2016) Methylmercury affects cerebrovascular reactivity to angiotensin II and acetylcholine via Rho-kinase and nitric oxide pathways in mice. Life Sci 147:30–38.  https://doi.org/10.1016/j.lfs.2016.01.033 CrossRefGoogle Scholar
  31. Joint FAO/WHO Expert Committee on Food Additives (JECFA) (2011) Safety evaluation of certain food additives and contaminants. Geneva, Switzerland. http://whqlibdoc.who.int/publications/2011/9789241660631_eng.pdf. Accessed 26 July 2016
  32. Kalish BT, Rifas-Shiman SL, Wright RO et al (2014) Associations of prenatal maternal blood mercury concentrations with early and mid-childhood blood pressure: a prospective study. Environ Res 133:327–333.  https://doi.org/10.1016/j.envres.2014.06.004 CrossRefGoogle Scholar
  33. Kessler R (2013) The Minamata Convention on Mercury: a first step toward protecting future generations. Environ Health Perspect 121:A304–A309Google Scholar
  34. Kim B-M, Choi AL, Ha E-H et al (2014) Effect of hemoglobin adjustment on the precision of mercury concentrations in maternal and cord blood. Environ Res 132:407–412.  https://doi.org/10.1016/j.envres.2014.04.030 CrossRefGoogle Scholar
  35. Lancet T (2017) Minamata Convention on mercury: a contemporary reminder. Lancet 390:822.  https://doi.org/10.1016/S0140-6736(17)32287-0 Google Scholar
  36. Llop S, Lopez-Espinosa M-J, Rebagliato M, Ballester F (2013) Gender differences in the neurotoxicity of metals in children. Toxicology 311:3–12.  https://doi.org/10.1016/j.tox.2013.04.015 CrossRefGoogle Scholar
  37. Llop S, Ballester F, Broberg K (2015) Effect of gene-mercury interactions on mercury toxicokinetics and neurotoxicity. Curr Environ Health Rep 2:179–194.  https://doi.org/10.1007/s40572-015-0047-y CrossRefGoogle Scholar
  38. Lurbe E, Garcia-Vicent C, Torro MI et al (2014) Associations of birth weight and postnatal weight gain with cardiometabolic risk parameters at 5 years of age. Hypertension 63:1326–1332.  https://doi.org/10.1161/HYPERTENSIONAHA.114.03137 CrossRefGoogle Scholar
  39. Lurbe E, Agabiti-Rosei E, Cruickshank JK et al (2016) 2016 European Society of Hypertension guidelines for the management of high blood pressure in children and adolescents. J Hypertens.  https://doi.org/10.1097/HJH.0000000000001039
  40. Mozaffarian D, Rimm EB (2006) Fish intake, contaminants, and human health: evaluating the risks and the benefits. JAMA J Am Med Assoc 296:1885–1899.  https://doi.org/10.1001/jama.296.15.1885 CrossRefGoogle Scholar
  41. Mozaffarian D, Wu JHY (2011) Omega-3 fatty acids and cardiovascular disease: effects on risk factors, molecular pathways, and clinical events. J Am Coll Cardiol 58:2047–2067.  https://doi.org/10.1016/j.jacc.2011.06.063 CrossRefGoogle Scholar
  42. Nakamura S, Kugiyama K, Sugiyama S et al (2002) Polymorphism in the 5’-flanking region of human glutamate-cysteine ligase modifier subunit gene is associated with myocardial infarction. Circulation 105:2968–2973CrossRefGoogle Scholar
  43. National Research Council (NRC) (2000) Toxicological effects of methylmercury. Academic Press, Washington, DCGoogle Scholar
  44. O’Brien E (2017) The Lancet Commission on hypertension: addressing the global burden of raised blood pressure on current and future generations. J Clin Hypertens Greenwich Conn 19:564–568.  https://doi.org/10.1111/jch.12998 CrossRefGoogle Scholar
  45. Park J-D, Zheng W (2012) Human exposure and health effects of inorganic and elemental mercury. J Prev Med Public Health Yebang Ŭihakhoe Chi 45:344–352.  https://doi.org/10.3961/jpmph.2012.45.6.344 CrossRefGoogle Scholar
  46. Poursafa P, Ataee E, Motlagh ME et al (2014) Association of serum lead and mercury level with cardiometabolic risk factors and liver enzymes in a nationally representative sample of adolescents: the CASPIAN-III study. Environ Sci Pollut Res Int 21:13496–13502.  https://doi.org/10.1007/s11356-014-3238-4 CrossRefGoogle Scholar
  47. Rice KM, Walker EM, Wu M et al (2014) Environmental mercury and its toxic effects. J Prev Med Public Health Yebang Ŭihakhoe Chi 47:74–83.  https://doi.org/10.3961/jpmph.2014.47.2.74 CrossRefGoogle Scholar
  48. Rinke ML, Singh H, Heo M et al (2018) Diagnostic errors in primary care pediatrics: project RedDE. Acad Pediatr 18:220–227.  https://doi.org/10.1016/j.acap.2017.08.005 CrossRefGoogle Scholar
  49. Roman HA, Walsh TL, Coull BA et al (2011) Evaluation of the cardiovascular effects of methylmercury exposures: current evidence supports development of a dose-response function for regulatory benefits analysis. Environ Health Perspect 119:607–614.  https://doi.org/10.1289/ehp.1003012 CrossRefGoogle Scholar
  50. Roulet C, Bovet P, Brauchli T et al (2017) Secular trends in blood pressure in children: a systematic review. J Clin Hypertens Greenwich Conn 19:488–497.  https://doi.org/10.1111/jch.12955 CrossRefGoogle Scholar
  51. Ruggieri F, Majorani C, Domanico F, Alimonti A (2017) Mercury in children: current state on exposure through human biomonitoring studies. Int J Environ Res Public Health 14.  https://doi.org/10.3390/ijerph14050519
  52. Sanderson S, Tatt ID, Higgins JP (2007) Tools for assessing quality and susceptibility to bias in observational studies in epidemiology: a systematic review and annotated bibliography. Int J Epidemiol 36:666–676.  https://doi.org/10.1093/ije/dym018 CrossRefGoogle Scholar
  53. Sheehan MC, Burke TA, Navas-Acien A et al (2014) Global methylmercury exposure from seafood consumption and risk of developmental neurotoxicity: a systematic review. Bull World Health Organ 92:254–269F.  https://doi.org/10.2471/BLT.12.116152 CrossRefGoogle Scholar
  54. Simkhovich BZ, Kleinman MT, Kloner RA (2008) Air pollution and cardiovascular injury epidemiology, toxicology, and mechanisms. J Am Coll Cardiol 52:719–726.  https://doi.org/10.1016/j.jacc.2008.05.029 CrossRefGoogle Scholar
  55. Sørensen N, Murata K, Budtz-Jørgensen E et al (1999) Prenatal methylmercury exposure as a cardiovascular risk factor at seven years of age. Epidemiol Camb Mass 10:370–375CrossRefGoogle Scholar
  56. Stergiou GS, Parati G, Asmar R et al (2012) Requirements for professional office blood pressure monitors. J Hypertens 30:537–542.  https://doi.org/10.1097/HJH.0b013e32834fcfa5 CrossRefGoogle Scholar
  57. Streets DG, Horowitz HM, Jacob DJ et al (2017) Total mercury released to the environment by human activities. Environ Sci Technol 51:5969–5977.  https://doi.org/10.1021/acs.est.7b00451 CrossRefGoogle Scholar
  58. Stroup DF, Berlin JA, Morton SC et al (2000) Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 283:2008–2012CrossRefGoogle Scholar
  59. Sundseth K, Pacyna JM, Pacyna EG et al (2017) Global sources and pathways of mercury in the context of human health. Int J Environ Res Public Health 14.  https://doi.org/10.3390/ijerph14010105
  60. Syversen T, Kaur P (2012) The toxicology of mercury and its compounds. J Trace Elem Med Biol 26:215–226.  https://doi.org/10.1016/j.jtemb.2012.02.004 CrossRefGoogle Scholar
  61. Thurston SW, Bovet P, Myers GJ et al (2007) Does prenatal methylmercury exposure from fish consumption affect blood pressure in childhood? Neurotoxicology 28:924–930.  https://doi.org/10.1016/j.neuro.2007.06.002 CrossRefGoogle Scholar
  62. United Nations Environment Programme. (2013) Mercury: Time to act. Geneva, Switzerland: UNEP Chemicals Branch. http://cwm.unitar.org/cwmplatformscms/site/assets/files/1254/mercury_timetoact.pdf. Accessed 26 July 2016
  63. Valera B, Dewailly E, Poirier P et al (2011) Influence of mercury exposure on blood pressure, resting heart rate and heart rate variability in French Polynesians: a cross-sectional study. Environ Health Glob Access Sci Source 10:99.  https://doi.org/10.1186/1476-069X-10-99 Google Scholar
  64. Valera B, Muckle G, Poirier P et al (2012) Cardiac autonomic activity and blood pressure among Inuit children exposed to mercury. Neurotoxicology 33:1067–1074CrossRefGoogle Scholar
  65. Vandenbroucke JP, von Elm E, Altman DG et al (2007) Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med 4:e297.  https://doi.org/10.1371/journal.pmed.0040297 CrossRefGoogle Scholar
  66. Virtanen JK, Rissanen TH, Voutilainen S, Tuomainen T-P (2007) Mercury as a risk factor for cardiovascular diseases. J Nutr Biochem 18:75–85.  https://doi.org/10.1016/j.jnutbio.2006.05.001 CrossRefGoogle Scholar
  67. Wakita Y (1987) Hypertension induced by methyl mercury in rats. Toxicol Appl Pharmacol 89:144–147.  https://doi.org/10.1016/0041-008X(87)90185-2 CrossRefGoogle Scholar
  68. WHO (1991) Environmental Health Criteria 118. Inorganic Mercury. Switzerland, Geneva http://www.inchem.org/documents/ehc/ehc/ehc118.htm. Accessed 24 Apr 2018Google Scholar
  69. WHO (2007) Exposure to mercury: a major public health concern. Switzerland, Geneva http://www.who.int/phe/news/Mercury-flyer.pdf. Accessed 26 July 2016Google Scholar
  70. WHO (2010) Children’s exposure to mercury compounds. Switzerland, Geneva http://apps.who.int/iris/bitstream/handle/10665/44445/9789241500456_eng.pdf;jsessionid=06ED69CBBAF04FFD9FB778C8B308C141?sequence=1. Accessed 01 July 2017Google Scholar
  71. WHO (2013a) A global brief on hypertension. Silent killer, global public health crisis. Switzerland, Geneva http://apps.who.int/iris/bitstream/handle/10665/79059/WHO_DCO_WHD_2013.2_eng.pdf?sequence=1. Accessed 28 Apr 2018Google Scholar
  72. WHO (2013b) Mercury and health. Switzerland, Geneva http://www.who.int/en/news-room/fact-sheets/detail/mercury-and-health. Accessed 24 Apr 2018Google Scholar
  73. WHO-UNEP (2008) Guidance for identifying populations at risk from mercury exposure. Switzerland, Geneva https://wedocs.unep.org/bitstream/handle/20.500.11822/11786/IdentifyingPopnatRiskExposuretoMercury_2008Web.pdf?sequence=1&isAllowed=y. Accessed 25 June 2017Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Conselleria de Sanitat Universal i Salut PúblicaGeneralitat ValencianaValenciaSpain
  2. 2.FISABIO-UJI-Universitat de València Joint Research Unit in Epidemiology and Environmental HealthValenciaSpain
  3. 3.Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP)MadridSpain
  4. 4.Nursing DepartmentUniversity of ValenciaValenciaSpain
  5. 5.Foundation for the Promotion of Health and Biomedical Research in the Valencian RegionFISABIO-Public HealthValenciaSpain

Personalised recommendations