Table 1.

Summary of the included studies of individual metal exposures or co-exposure models

Author, year of publication; title	Location	Sample Size	Study population characteristics	Year at baseline, study name if available	Study design	Exposure(s)	Metal concentrations	Outcome	Covariates	Key findings
Arsenic
Farzan et al., 2017; Maternal and infant inflammatory markers in relation to prenatal arsenic exposure in a US pregnancy cohort	New Hampshire, USA	563 mothers; 500 paired infants	Mean maternal age at enrollment: 31.2 years	New Hampshire BIrth Cohort Study; ongoing recruitment since January 2009	Prospective cohort	Arsenic [As] levels measured in maternal urine samples at gestational weeks (GW) 24–28	Mean (SD) Maternal urinary As (μg/L): 5.78 (13.34)	Maternal and infant Inflammation levels measured by plasma and cord blood samples markers: MCP1, TNFalpha, ICAM1, VCAM1	Enrollment age, education, maternal smoking and/or second hand smoke exposure during pregnancy, urinary creatinine, and batch assignment	Maternal arsenic exposure during pregnancy was positively associated with infant cord blood levels of ICAM and VCAM. A dose response relationship was found between maternal urinary As and maternal VCAM. Maternal VCAM and ICAM mediated the relationship between arsenic and infant VCAM and ICAM
Chen et al., 2019; Early life and adolescent arsenic exposure from drinking water and blood pressure in adolescence	Araihazar, Bangladesh	726 adolescents	Mean age at child assessment: 14.75 years	Health Effects of Arsenic Longitudinal Study (HEALS)	Prospective cohort	Arsenic [As] levels repeatedly measured in maternal urine from 1 year prior to birth (in utero) to < 5 years; in maternal urine from 5 to 12 years; in child urine at time of recruitment 14–17 years	Mean (SD) maternal urine As from 1 year prior to birth (in utero) to < 5 years (μg/g creatinine): 264.5 (247.5) Mean (SD) maternal urine As from 5 to 12 years (μg/g creatinine): 235.6 (234.2) Mean (SD) child urine As at 14–17 years (μg/g creatinine): 158.4 (207.1)	Systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP) in adolescents 14–17 years of age	Sex, age at time of assessment, BMI	Every doubling of urinary As at time of assessment (current exposure) and maternal urinary As (in utero/early childhood) was positively associated with a difference of 0.7 mmHg (95% CI: 0.1, 1.3) and 0.7 mmHg (95% CI: 0.05, 1.4) in SBP, respectively. Associations were stronger in adolescents with a BMI above the median than those below the median. No associations were observed for DBP. BKMR identified current urinary As as a significant contributor to SBP and DBP independent of other concurrent blood levels of Cd, Pb, Mn, and Se
Stratakis et al., 2020; Association of Fish Consumption and Mercury Exposure During Pregnancy With Metabolic Health and Inflammatory Biomarkers in Children	5 European countries (France, Greece, Norway, Spain, and the UK)	805 mother–child pairs	Mean maternal age at study inclusion or delivery: 31.3 years; Mean child age 8.4 years; 91.2% of sample was of White race/ethnicity	HELIX Project (Consortium of 5 European cohorts: UK (BIB cohort), France (EDEN cohort), Spain (INMA cohort), Norway (MoBa cohort), Greece (Rhea cohort)); Enrollment between 2003 and 2009, studies conducted between 2003and 2016	Prospective cohort	Prenatal Hg levels measured in either maternal whole blood (mid-pregnancy for MoBa & RHEA cohorts; late pregnancy for BiB cohort or in infant cord blood samples (INMA cohort)). Estimated Hg exposure also was assessed by reported fish intake during pregnancy (FFQ categorized as low (< 1x/week), medium (1–3x/week), and high (> 3x/week))	Median (IQR) Maternal Hg or infant cord:2.5 μg/L (1.5–4.2)	Systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP) in children ages 6–12 years. Child levels of high-density lipoprotein (HDL) cholesterol and triglyceride levels were assessed in serum	Maternal age, pre-pregnancy BMI, parity, maternal/parental educational level, and child’s race/ethnicity. Effect modification by maternal educational level, gestational diabetes, and child sex	Maternal mercury concentrations and fish intake during pregnancy were positively correlated (Spearman r = 0.2, no p-value given). Children with medium and high maternal fish intake had significantly higher HDL cholesterol levels than children with low maternal fish intake. No significant association was found between medium or high maternal fish intake and child SBP, DBP, or triglyceride levels
Farzan et al., 2021; Prenatal and postnatal mercury exposure and blood pressure in childhood	New Hampshire, US	395 children; n = 301 maternal prenatal (GW 24—28) and/or n = 322 postpartum (6 weeks) toenail measurements	Mean maternal age at enrollment: 30.9 years;. Mean child age at time of assessment: 5.5 years	New Hampshire Birth Cohort Study; ongoing recruitment since January 2009	Prospective cohort	Mercury [Hg] levels measured in maternal toenail clippings collected at gestational week (GW) 24–28 and 6 weeks postpartum. Children’s Hg levels were measured in toenail clippings at 3 years of age and in urine samples at 5–6 years of age	Mean (SD) Child 3-year toenail Hg: 0.055 μg/g (0.087); Mean (SD) Child 5–6 year urine Hg: 0.071 μg/L (0.119); Mean (SD) Maternal GW 24–28 toenail Hg: 0.129 μg/g (0.139) Maternal 6 week postpartum toenail Hg: 0.128 μg/g (0.157)	Systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP) in children 5–6 years of age	Child age, sex, height, weight, birth weight, gestational age and maternal smoking during pregnancy and educational attainment. Additional adjustment for urine specific gravity was included in models with urine biomarkers only	Higher child mercury exposure was associated with higher DBP (toenail β: 0.53 mmHg; 95% CI: 0.02, 1.07, p = 0.06 urine β: 0.48 mmHg; 95% CI: 0.10, 0.86, p = 0.01) and higher MAP (toenail β: 0.67 mmHg; 95% CI: 0.002, 1.33, p = 0.049 urine β: 0.55 mmHg; 95%CI: 0.10, 1.01, p = 0.02). When stratified by sex, stronger associations were observed in males, but did not reach statistical significance
Chan et al., 2021; Prenatal methylmercury exposure is associated with decreased heart rate variability in children	Hong Kong, China	604 children, enrolled as part of prior birth cohort study	Mean maternal age at enrollment: Not reported; Mean (SD) child age at time of assessment: 8.1 (0.9) years	Hong Kong Birth Cohort, enrolled between July 2000–December 2001 at Prince of Wales Hospital	Prospective cohort	Cord blood and current whole blood total Hg concentration were used as biomarkers of prenatal and recent MeHg exposure. Recent child fish consumption was measured by FFQ (estimate kg/month)	Mean (SD) infant cord blood Hg (nmol/L): 50.12 (23.9); current mean (SD) whole blood Hg (nmol/L): 15.94 (9.94)	Heart rate variability (HRV), resting heart rate (RHR) and blood pressure in children 7–8 years of age	Child sex, age, body mass index, gestational age at birth, maternal smoking and alcohol consumption during pregnancy, and maternal hypertension during pregnancy. Additional adjustment for recent fish intake was included as a sensitivity analysis	Prenatal MeHg exposure was associated with multiple measures of decreased HRV. Recent blood Hg showed no association with any outcomes Adjustment of recent fish consumption further increased the significance and magnitude of the adverse associations of MeHg
Lead
Sanders et al., 2018; “Prenatal lead exposure modifies the effect of shorter gestation on increased blood pressure in children”	Mexico City, Mexico	565 mother–child pairs	Mean (SD) maternal age: 27.6(5.6) years; mean (SD) child age at time of assessment: 4.8(0.6) years	PROGRESS; enrolled between 2007 and 2011	Prospective cohort	Lead [Pb] levels measured in maternal blood at 2nd trimester visit	Mean (SD) Maternal Pb (βg/dL): 3.7 (2.7)	Systolic and diastolic blood pressure of child (avg. 2 measurements) between 4 and 6 years of age	Maternal SES, tobacco smoke at home and child age, height, sex	Increased blood Pb levels were associated with higher systolic BP measurements in children with shorter gestational ages
Liu et al. 2019; “Prenatal Lead Exposure, Type 2 Diabetes, and Cardiometabolic Risk Factors in Mexican Children at Age 10–18 Years”	Mexico City, Mexico	369 mother–child pairs	Mean (SD) maternal age: 26.7 (5.6) years; mean (SD) child ageat time of assessment: 13.7 (1.9)	ELEMENT; enrolled between 1997 and 2003	Prospective cohort	Lead [Pb] levels measured in maternal blood at 1st trimester visit	Mean (95% CI) Maternal Pb (βg/dL): 4.3 (4.0–4.6)	Fasting glucose serum sample, insulin serum sample, lipids serum sample (triglycerides, total cholesterol, HDL-C, LDL-C) between 10 and 18 years of age (peripubertal)	Maternal age, marital status, education, smoking history and child age, sex, BMI, number of siblings at birth	Increased Pb concentrations were associated with decreased total cholesterol scores (B = − 0.76, 95% CI: − 1.38, − 0.13) and decreased LDL scores (B = − 0.96, 95% CI: − 1.59, − 0.33). Association only remained in males after adjusting for child sex
Wang et al., 2021; “Associations between prenatal and postnatal lead exposure and preschool children humoral and cellular immune responses”	Wuhan, China	394 mother–child pairs	Mean (SD) maternal age: 28.98 (3.54) years; child age caetgories at time of assessment: < 36 months (10.91%), 36–48 months (82.23%), > 48 months (6.85%)	Wuhan Medical and Healthcare Center for Women and Children birth cohort; enrolled between 2013 and 2015	Prospective cohort	Lead [Pb] levels in maternal urine at third trimester; normalized to urine creatinine	Mean (SD) Maternal Pb (βg/g): 5.82 (6.47)	Biomarkers of immune cytokines and T lymphocytes via child blood serum at age 3 years	Maternal age, parity, passive smoking, education and child sex	Increased prenatal Pb exposure is associated with reductions in two anti-atherosclerotic cytokines IL-4 (B = − 5.62, 95%: − 10.44, − 0.80) and IL-10 (B = − 5.93, 95%: − 11.82, − 0.03)
Muciño-Sandoval et al., 2021; “Prenatal and Early Childhood Exposure to Lead and Repeated Measures of Metabolic Syndrome Risk Indicators From Childhood to Preadolescence”	Mexico City, Mexico	601 mother–child dyads	Mean (SD) maternal age: 27.1 (5.5) years; no average child age reported	PROGRESS birth cohort in Mexico enrolled between 2007 and 2011	Prospective cohort	Prenatal Lead [Pb] levels in maternal blood at 2nd and 3rd trimester visit & cord blood, and bone (patella and tibia) at 1 month postpartum. Postnatal Pb in child blood samples at ages 1/2/4 years	Median (IQR) 2nd trimester Pb (μg/dL): 2.9 (1.9–4.4). Median (IQR) 3rd trimester Pb (μg/dL): 3.1 (2.0–4.8). Median (IQR) Umbilical cord Pb (μg/dL):(1.4–3.7). Median (IQR) patella Pb (μg/dL): 3.4 (1.3–8.9). Median (IQR) tibia Pb (μg/dL): 3.0 (1.1–7.5)	Metabolic syndrome: biomarkers (fasting glucose, HbA1C) via blood, lipids (triglycerides, total cholesterol, HDL, LDL) via blood, anthropometric measures (BMI, waist circumference, body fat%), systolic and diastolic blood pressure between ages 6 to 12 years	Maternal: age, SES, parity. Child: sex, size for gestational age, age	Children with higher prenatal Pb exposure had lower odds of higher total cholesterol, compared to children with lower prenatal Pb exposure (OR = 0.53, 95% CI: 0.31–0.99) & (OR = 0.67, 95% CI: 0.43–1.02, results in table reported as “marginal significance”). Children with higher prenatal Pb exposure had lower odds of high triglycerides (OR = 0.65, 95% CI: 0.44–0.95), high diastolic BP (OR = 0.60, 95% CI: 0.37–0.98), compared to children with lower prenatal Pb exposure
Cadmium
Chatzi et al., 2018; “Associations of prenatal exposure to cadmium with child growth, obesity, and cardiometabolic traits”	Heraklion, Greece	515 mother–child pairs	Mean (SD) maternal age: 29.8 (5.0) years; mean (SD) child age at time of assessement: 4.2 (0.2) years)	Rhea Cohort; recruited between 2007 and 2012	Prospective cohort	Cadmium [Cd] levels in maternal urine during pregnancy; Median (IQR) 13 (12–15) weeks of gestation	Median (IQR) Maternal Cd (μg/L): 0.5 (0.3–0.7)	Weight, height, waist circumference, skinfold thickness, systolic and diastolic blood pressure of child (avg. 5 measurements), lipids (total cholesterol, HDL-C), CVD biomarkers (leptin, C-reactive protein) measured at age-4-year examination	Maternal age, education, national origin, parity, pre-pregnancy BMI, gestational diabetes, tobacco smoking, pollutant exposure and child sex, gestational age at birth, weight, breastfeeding duration	Cd concentrations and change in BP measures (systolic [B = 0.01, 95%: − 0.14, 0.16]; diastolic: [B = 0.04, 95%: − 0.06, 0.13]), lipids (total [B = − 3.07, 95%: −8.79, 2.64]; HDL: [B = − 0.07, 95%: − 2.43, 2.28]), and CVD biomarkers (CRP [B = − 11.1, 95%: − 33.0, 18.0]; Leptin: [B = 0.6, 95%: − 15.0, 10.0]), were not statistically significantly associated (null)
Co-exposure to two metals
Farzan et al., 2018; “Prenatal lead exposure and elevated blood pressure in children”	New Hampshire, USA	323 mother–child pairs	Mean maternal age: 31.3 years; Mean child age at time of assessment: 5.5 years	New Hampshire Birth Cohort Study; ongoing recruitment since January 2009	Prospective cohort	Arsenic [As] and lead [Pb] levels in maternal toenails at 24–28 weeks gestation and 6 weeks postpartum	Mean (SD) prenatal maternal toenail Pb (μg/g): 0.3 (0.6); Mean (SD) postpartum maternal toenail Pb (μg/g): 0.3 (1.4); Mean (SD) prenatal maternal toenail As: 0.1 μg/g (0.1); Mean (SD) postpartum maternal toenail As: 0.1 μg/g (0.1)	Systolic and diastolic blood pressure of child (avg. 5 measurements)	Maternal tobacco use during pregnancy and child sex, age, height, weight	Prenatal Pb concentrations were associated with increased systolic BP for children (B = 0.58, 95% CI = 0.05, 1.11). Postpartum Pb was not associated with child BP. There was no significant associations between prenatal or postpartum arsenic exposure and child’s systolic blood pressure or diastolic blood pressure
Akhtar et al. 2021; “A longitudinal study of rural Bangladeshi children with long-term arsenic and cadmium exposures and biomarkers of cardiometabolic diseases”	Matlab, Bangladesh	540 mother–child pairs	Maternal age not reported; Mean child age at time of assessment: 8.9 years	Multiple micronutrient supplementation (MINIMat) trial; enrolled 2001–2003	Prospective cohort	Arsenic [As] and cadmium [Cd] levels measured in maternal urine at 8 weeks gestation; normalized to urine creatinine	Median (IQR) Cd (μg/g): 0.58 (0.31, 0.96); As (μg/L): 76.07 (37.62, 218.64): maternal urine AS: Median, 76.07 μg/L	Systolic and diastolic blood pressure measured at two time points (4.5 years and 9 years of age). Lipids (total cholesterol, triglycerides, HDL-C, LDL-C), and metabolic biomarkers (glucose, a1C, adipokines, eGFR) measured in child blood serum at 9 years	Maternal educational level, food supplementation, micronutrient supplementation and child SES, sex, age, height-for-age z-score, BMI-for-age z-score	Increased Cd concentrations were associated with increased systolic BP (exact B not given). Increased Cd concentrations were associated with decreased HDL scores (B = − 1.22, 95% CI: − 2.02, − 0.42). Association remained when accounting for Cd-As interaction. CVD biomarkers (Leptin) not associated with Cd concentrations. Maternal As exposure was not associated with child total cholesterol at age 9 or with child blood pressure at either time point. A negative association was found between maternal arsenic exposure and child HDL levels at age 9 (β = − 1.02, 95% CI: 1.56, − 0.48, p < 0.001)