Table 2 Summary of results from studies examining manganese concentrations in hair (Hair-Mn or H-Mn)

Barlow et al. (1983) [62]

<16 years

Hyperactive: 0.84 (0.64)

Control: 0.68 (0.45)

None measured

Hyperactivity was more prevalent in hyperactive children (mean age: 7.6 years) but at lower levels of statistical significance (90 % confidence) using bivariate analyses.

Lower levels of zinc (Zn) were associated with hyperactive children: 83.4 (32.3) compared to controls: 99.1 (54.3), 95 % confidence using bivariate analysis. Other metals were nonsignificant in association with the outcome including: (Cadmium (Cd), Copper (Cu), Iron (Fe), Lead (Pb), and Magnesium (Mg).

Collipp et al. (1983) [34]

Ages 7–10 years Learning- disabled:0.43 Control: 0.27

None measured

Significantly higher hair-Mn levels from learning disabled children, 7–10 years old, compared to children without the condition.

Not applicable

Age 4-months

Breastfed: 0.33 Formula fed: 0.685

Significantly greater hair-Mn in formula-fed infants.

None applicable

Not applicable

Takser et al. (2003) [43]

Newborns to 6 years 0.75¹

None measured

No association was found between hair-Mn post-childbirth and general psychomotor developmental indices at 9 months and a general cognitive index at 3 and 6 years in models adjusted for maternal age and education, smoking, labor duration, children’s sex and cord blood lead levels and other confounders.

Not applicable

Wright et al. (2006) [16]

11–13 years

0.47

None measured

Lower full-scale IQ, verbal learning and memory scores were associated with higher concentrations of hair-Mn from children, on average 12.6 years old, in analyses adjusted for maternal education, child sex and concentrations of lead PbH.

Higher arsenic (As) levels, particularly in combination with higher Mn levels, associated with lower IQ, verbal learning, and memory scores. No associations found with Cd levels.

Bouchard et al. (2007) [17]

6–15 years

5.1 (4.3)

Greater MnH concentrations from children who drank well water with higher Mn-water.

Greater hyperactive and oppositional classroom behavior was associated with higher hair-Mn from children, on average, 11 years old, in analyses adjusted for age, sex and income. No interaction between

hair-Mn and child sex.

Not applicable

Hernández-Bonilla et al. (2011) [48]

7–11 years

12 (exposed)

0.57 (nonexposed)

Respiratory Mn exposures were associated with residential proximity to Mn mines, but specific measures were not reported.

Hair-Mn was not associated with neuromotor outcomes (grooved pegboard, finger tapping repetition and Santa Anna test in children, on average, 9 years old in analyses adjusted for Pb in blood, hemoglobin, sex, age and maternal education.

Not applicable

Menezes-

Filho et al. (2011) [35]

6–12 years

5.83¹ (11.5)

Hair-Mn levels were

6 times higher than those in the general Brazilian population (mean 0.47 μg/g, range 0.89–2.15 μg/g)

None measured, but Mn exposures were from residential proximity to Mn alloy production plant.

Lower full scale and verbal IQ scores in children, on average, 8.8 years old, in analyses adjusted for maternal education and nutritional status. A ten-fold increase of hair-Mn was associated with a 6.7 - point loss in Verbal IQ score.

Children with iron deficiency had higher hair-Mn (15.94 ± 19.68 μg/g; p = 0.06) compared to those with FeS in normal range (8.69 ± 8.23 Mn/g).

Riojas-

Rodriguez et al. (2010) [31]

7–11 years

Exposed: 12.13

Unexposed: 0.57

Median airborne concentration of Mn in PM10 of exposed (0.13 μg/m³) versus unexposed (0.02 μg/m³) communities, but personal exposures were not reported.

Lower full scale, verbal and performance IQ scores in children, on average, 9 years old, in analyses adjusted for blood-Pb, hemoglobin, age, sex and nutritional status. Sex significantly modified the association with the strongest inverse association in young girls. There was little evidence of an association in boys.

Not applicable

Bouchard et al. (2011) [32]

6–13 years

Median: 0.7

Hair-Mn levels were associated with higher Mn in water (Water-Mn) (mean: 98 μg/L, GM:¹20 μg/L), but not in diet.

Lower full-scale IQ scores were associated with increased hair-Mn concentrations in children, on average, 9 years old, in analyses adjusted for maternal intelligence and education, income, sex and age of children, Fe concentrations in water and other confounders. A 10-fold increase in

water-Mn was associated with a decrease of 2.4 IQ points (95 % CI:-3.9 to -0.9, p < 0.01) adjusting for maternal intelligence and other confounders. Sex stratification showed a slightly higher impact of

hair-Mn for girls’ full-scale IQ, but the interaction term was nonsignificant. Water-Mn was more strongly associated with performance than verbal IQ.

Not applicable

Lucchini, et al. (2012) [33]

11–14 years

0.16 Median

Significant differences for Mn concentrations in soil (soil-Mn) and air (air-Mn) by proximity to industrial sites with historical Mn emissions, but not for tap water, diet or hair.

Impairment of motor coordination, hand dexterity and odor identification was associated with median concentrations of soil-Mn in exposed (897 ppm) versus reference (409 ppm) communities.

Tremor intensity in dominant hand was positively associated with hair-Mn in children, on average, 12.9 years old, in analyses adjusted for age, gender, SES, family size, parity order, parents’ education, smoking habits and soil concentrations of Pb and other metals. Boys had increased tremor intensity relative to girls.

Not applicable

Lucchini, et al. (2012) [41]

11–14 years

0.17

No association between concentrations of hair-Mn with soil- or air-Mn

No association between hair-Mn concentrations with full-scale, verbal or performance IQ, or behavioral and attention deficit hyperactivity scores for children, on average, 12.9 years old, with Mn exposure modeled as a main effect or an interactive term with blood-Pb in analyses adjusted for age, gender, family size, SES, area of residence, hemoglobin, ferritin and confounders.

Torres-

Agustín et al. (2013) [54]

7–11 years

Exposed: 14.2

Unexposed: 0.73

Greater hair-Mn concentrations in children in exposed group. Mn concentrations in outdoor air from Mn mining and ranged from the median: 0.08, μg/m³ in the exposed location compared to the median: 0.02, μg/m³ in the control location.

Lower long-term memory and learning scores were associated with increased hair-Mn in children, on average, 9 years old, in analyses adjusted for children’s sex, blood-Pb, age, hemoglobin and maternal education. The negative association was stronger for girls.

Not applicable

Rink et al. (2014) [46]

14–45 months

0.98 (0.74)

None measured

Lower scores in cognitive and expressive language tests in children, on average, 28.8 months old, but only in unadjusted models. Boys had a significantly positive association between hair-Mn concentrations and receptive languages scores in analyses adjusted for hair-Pb concentrations, child hemoglobin and age, paternal education, maternal IQ, SES, and other confounders.

Not applicable

Carvalho et al. (2014) [36]

7–12 years

14.6 (11.8)

None were measured, but Mn exposure was related to residential location and air emissions from an iron-Mn alloy plant.

Lower full-scale IQ, and lower scores on Vocabulary, Block Design, and Digit Span tests were associated with increased hair-Mn for children, on average, 9.4 years old, in analyses adjusted for maternal education and children’s age. Each 1 μg/g increase in hair-Mn was associated with a decrease of approximately 1 full-scale IQ point and lower test scores for executive function, strategic visual formation and verbal working memory. No significant sex differences for hair-Mn concentrations.

Menezes-

Filho et al. (2014) [53]

7–12 years

Boys: 15.3 (9.9) Girls: 13.9 (13.4)

None were measured, but Mn exposure was related to residential location and air emissions from an iron-Mn alloy plant.

Externalizing behaviors and attention problems on the Child Behavior Checklist (CBC) for girls was significantly associated with higher hair-Mn. No significant association was found between CBC scores for boys in sex-stratified models adjusted for age (with boys) or maternal IQ (with girls).

Oulhote et

al. (2014) [47]

6–13 years

Boys: 0.75

Girls: 0.80

Greater water-Mn (mean: 99 μg/L; GM: 20 μg/L) & hair.

Mn exposure was associated with significant decrements in memory (hair and water) and attention (hair), and motor function (water) adjusted for maternal education and nonverbal intelligence, tobacco consumption, child sex, age and other confounders. Estimates of associations by sex were similar.

do Nascimento et al. (2015) [37]

6–12 years

Rural: 2.07 (2.6)

Urban: 0.45 (0.2)

Greater Mn in drinking water (mean: 20 μg/L) rural sites and (mean: 1.0 μg/L) urban sites) associated with greater Mn levels in hair.

Lower (nonverbal) IQ scores were

associated with hair-Mn and water-Mn concentrations for children, on average, 8.5 years old using models adjusted for age, gender and parental education.

Additional, similarly specified models were tested for the association of Pb, Cr, As, Hg, and Fe in hair on cognitive outcomes. Only hair-Fe showed a significant and inverse association with outcomes.

Haynes et al. (2015) [15]

7–9 years

0.42¹ (0.002)

Air-Mn associated with home proximity to ferromanganese refinery.

Lower full-scale IQ and perceptual

reasoning scores were associated with hair-Mn for children, 7–9 years old in analyses adjusting for blood-Pb, blood-Mn, serum creatinine, community of residence, child sex, parents’ IQ, education and parenting confidence. A U-shaped association was observed as children with hair-Mn concentrations > 747 μg/g had significantly lower IQ than children with hair-Mn concentrations between 207.2–747 μg/g (ß -3.66, 95 % CI: -6.9, -0.43). Children with hair-Mn levels < 207 had lower, but nonsignificant associations with full scale IQ than those with concentrations between 207.2–747 μg/g.

Not applicable

Shin et al. (2015) [51]

6–16 years

Case: 0.31 (0.46)

Control: 0.22 (0.10)

None measured

No association between hair-Mn and ADHD was found in children on, average, 9.7 years old when analysis was adjusted for confounders of age, sex and full-scale IQ.

Not applicable