Environmental Research 133 (2014) 232–238
Contents lists available at ScienceDirect
Environmental Research journal homepage: www.elsevier.com/locate/envres
Prenatal exposure to manganese at environment relevant level and neonatal neurobehavioral development Xiao-Dan Yu, Jun Zhang, Chong-Huai Yan n, Xiao-Ming Shen MOE-Shanghai Key Laboratory of Children's Environmental Health, XinHua Hospital affiliated to Shanghai Jiao Tong University School of Medicine, 1665 Kong Jiang Road, Shanghai 200092, China
art ic l e i nf o
a b s t r a c t
Article history: Received 16 September 2013 Received in revised form 11 April 2014 Accepted 16 April 2014
Background: Effects of prenatal Manganese (Mn) exposure at an environmental relevant level on neonatal neurodevelopment remains unclear. Objectives: In the multi-center study, we assessed the impact of low level prenatal Mn exposure on neonatal behavioral neurological assessments (NBNA), and explore a threshold umbilical cord blood Mn on neonatal neurological development. Methods: We investigated 933 mother-newborn pairs in Shanghai, China, from 2008 through 2009. Umbilical cord serum concentrations of Mn were measured and NBNA tests were conducted. The NBNA contains five clusters: behavior, active tone, passive tone, primary reflexes and general assessment with a maximal total score of 40. The scoreo37 is defined as low. Results: The median serum Mn concentration was 4.0 μg/L. Of the 933 infants, 44 (4.7%) had low NBNA. After adjusting for potential confounders, a high level of Mn ( Z 75th percentile ) was associated with a lower NBNA score (adjusted ß ¼ 1.1, 95% CI: 1.4–0.7, p o0.01) and a higher risk of low NBNA (adjusted OR ¼9.4, 95% CI: 3.4–25.7, po 0.01). A nonlinear relationship was observed between cord serum Mn and NBNA after adjusting for potential confounders. NBNA score decreased with increasing Mn levels after 5.0 μg/L(LgMn Z0.7). The cord serum MnZ5.0 μg/L had adverse effects on behavior, active tone and general reactions of clusters (p o0.001). Conclusions: High prenatal Mn exposure even at an environmental relevant level, is associated with poor fetal neurobehavioral development in a nonlinear pattern. A threshold cord serum Mn of 5.0 μg/L existed for lower neonatal behavioral neurological assessments. & 2014 Published by Elsevier Inc.
Keywords: Prenatal exposure Mn Neonatal neurobehavioral development Epidemiology
1. Introduction Manganese (Mn) is an essential nutrient to human, but it also has the potential to produce neurotoxic effects when accumulating in an organ, especially the brain (World Health Organization, 1981). Studies of Mn exposure in workplace have demonstrated that exposure to high doses of Mn is associated with irreversible neurodegenerative disorders resembling idiopathic Parkinson disease (Furbee, 2011). More recently, child environmental exposure to Mn has gained interest. Postnatal Mn exposure was found to interfere with development of brain functions (Zoni et al. 2007; Bouchard et al., 2007). However, few epidemiologic data are available on the effects of in utero Mn exposure on child neuropsychological development. In a birth cohort study in France, cord
Abbreviations: Mn, Manganese; NBNA, neonatal behavioral neurological assessments; LOD, limit of detection; ICP-MS, Inductively Coupled Plasma Mass Spectrometry n Corresponding author. Fax: þ86 21 25078875. E-mail address: [email protected] (C.-H. Yan). http://dx.doi.org/10.1016/j.envres.2014.04.012 0013-9351/& 2014 Published by Elsevier Inc.
blood Mn was negatively associated with attention and non-verbal memory and boys' manual ability at 3 years, after adjusting for mother's educational level (Takser et al., 2003). In a U.S. study, Mn was analyzed in the enamel of deciduous teeth. Its concentration in tissue formed during the intra-uterine phase was significantly associated with disinhibitory behavior evaluated at 36 and 54 months of age (Erikson et al., 2007). The history of neonatal neurobehavioral assessment began in early last century. Sarnat (1984); Amiel-Tison (2002) and Dubowitz et al. (2005) gradually developed neurological assessment in the newborn via examining tone and reflexes. After realizing that a newborn can regulate his behavior, Brazelton developed the Neonatal Behavioral Assessment Scale (NBAS), the first truly standardized, comprehensive assessment of newborn neurobehavior (Brazelton, 1973). Canals et al. (2011) confirmed that neonatal self-regulation behaviors were the best predictors of infant development and intelligence and that NBAS could be a useful tool to observe behaviors related to later development in healthy infants. Based on the method of Brazelton and Amiel-Tison for behavioral neurological measurement in newborns as well as
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
their own experience, Bao et al. (1991) formulated Neonatal Behavioral Neurological Assessments (NBNA), and subsequently repeated measurements in children showed that the total scores increased with child age and the NBNA had distinct stability and reliability, and was not influenced by geographic location. Therefore, it is convenient for a large survey (Bao et al. 1993).We used NBNA to measure neurobehavioral development of neonates in the study. To our best knowledge, the potential association between prenatal Mn exposure on neurobehavior of neonates have not been studied, though it is known that neonatal irritability and selfregulation are related to cognitive development and intelligence both short term and long term (Canals et al., 2011). Furthermore, there are no data available suggesting “safe” level of cord serum Mn. We conducted a study of prenatal Mn exposures and neonatal development in a mother-infant cohort in Shanghai, China. The objectives of the present study are (1) to determine the cord blood level of Mn and the corresponding effects on neurobehavioral development in neonates; and (2) to explore possible nontoxic level of Mn for neonatal neurodevelopment.
233
the Medical Ethics Committee of Shanghai XinHua Hospital affiliated to Shanghai Jiao Tong University School of Medicine (No.2008.7). After the women had signed the consent form, an in-person interview was conducted to collect information on social and demographic characteristics and potential sources of metal exposure. Anthropometric measurements of the newborns were made by delivery room staff according to a standard anthropometric protocol. Information on gestational age along with characteristics of the birth and newborn were extracted from the medical records. NBNA was administered when the infants were 3 days old as previously described (Gao et al., 2007). NBNA assesses functional abilities, most reflexes and responses, and stability of behavioral status during the examination. It contains five clusters: behavior (six items), passive tone (four items), active tone (four items), primary reflexes (three items), and general assessment (three items). Each item has three levels (0, 1 and 2). Twenty items have a maximal total score of 40. Neonates with a total score of equal to or more than 37 are considered well developed while below 37 are considered low NBNA (Bao et al., 1991). NBNA assessments were conducted by ten examiners who were rigorously trained and certified by the creator of Chinese NBNA, Professor Bao (Bao et al., 1991). Umbilical cord blood was collected and serum was separated. All the samples were immediately frozen at 40 1C and shipped in batches to the central laboratory. Serum Mn concentration was measured by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) (7500 CE, Agilent) as described previously (Yu et al., 2013). The limit of detection (LOD) for Mn was 0.13 μg/L.
2.2. Statistical analysis
2. Methods 2.1. Study subjects This multicenter study recruited 1377 healthy pregnant women who came to the hospital to deliver a term (37–42 weeks of gestation) singleton infant in 10 maternity hospitals in Shanghai, China, from 2008 to 2009, as we previously described (Yu et al., 2013). A total of 48 infants with disorders closely associated with adverse severe neurodevelopment such as traumatic brain injury, meningitis, epilepsy, severe neonatal jaundice were excluded. Of the 1329 infants, 933 parents signed the consent form to do the NBNA test. The study protocol was approved by
Cord serum concentrations of Mn were expressed as μg/L. Values below LOD were imputed using the default value of ½ LOD (6 of 933 subjects were below LOD). We first examined the distribution of NBNA score and Mn level. The latter was severely skewed towards the left. Thus, we performed log transformation before analysis. We divided study population into four groups based on quartiles of LgMn (Table 1) and performed univariate analysis and multiple regression examining group difference of both mean NBNA score and percent of low NBNA (Tables 2 and 3). We then applied a generalized additive model to estimate the independent relationship between LgMn and NBNA scores, with adjustment for potential confounders (Fig. 1).
Table 1 Characteristics of mothers-infants and the associations with cord serum manganese. P value
Lg Mn quartiles
N NBNA Score Behavior Active tone Passive tone Primary reflexes General reactions Birth weight (g) Gestational age (d) Maternal age (y) Gender(%) male female Household incomes (Yuan/m/Person)(%) o 2000 2000–5000 45000 Maternal education(%) Middle school or lower High school Bachelor degree Higher than bachelor degree Paternal education (%) Middle school or lower High school Bachelor degree Higher than bachelor degree Maternal occupation (%) White collar Technician Blue collar Housewife
o P25 (Mn o2.7 μg/L)
P25–50 (Mn:2.7–4.0 μg/L)
P50–75 (Mn:4.1–9.0 μg/L)
Z P75 (Mn Z9.1 μg/L)
231 39.4 7 0.9 11.8 7 0.5 7.7 70.5 7.9 70.3 6.0 7 0.2 6.0 7 0.2 3454.8 7 410.7 275.9 7 8.6 27.1 73.9
228 39.5 7 0.9 11.8 7 0.5 7.8 7 0.4 7.9 7 0.3 6.0 70.2 6.0 70.2 3396.4 7429.1 276.7 7 7.8 27.3 74.2
239 39.4 7 1.0 11.8 7 0.6 7.7 7 0.6 7.9 7 0.3 6.0 7 0.1 6.0 7 0.2 3347.2 7 393.3 275.8 7 7.6 27.0 74.7
235 38.2 7 1.6 11.2 7 1.1 7.2 70.9 7.9 70.4 6.0 7 0.2 6.0 7 0.0 3415.6 7 418.2 275.3 7 7.1 28.3 7 4.0
53.7 46.3
49.8 50.2
57.1 42.9
51.1 48.9
52.1 34.3 13.6
51.0 35.3 13.7
48.0 34.4 17.6
28.5 35.7 35.7
39.3 22.8 33.9 4.0
41.3 418.8 37.2 2.7
40.4 18.3 38.7 2.6
21.5 14.9 54.8 8.8
36.9 21.2 37.4 4.5
34.7 21.2 38.7 5.4
35.2 18.9 42.5 3.4
15.2 16.6 58.3 9.9
22.1 27.1 49.7 1.1
22.2 29.2 47.6 1.1
24.4 27.4 46.7 1.5
13.3 15.0 69.9 1.7
o 0.001 o 0.001 o 0.001 0.235 0.849 0.050 0.047 0.341 0.010 0.396
o 0.001
o 0.001
o 0.001
0.001
234
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
Table 2 The unadjusted association between factors and neonatal behavioral neurological assessments score. Low NBNAn
NBNA score ß (95%CI) Lg Mn quartiles (μg/L) oP25(Mn o 2.7) P25–50(Mn:2.7–4.0) P50–75(Mn:4.1–9.0) ZP75(Mn Z 9.1) Maternal age Gestational age Birth weight LgPb (μg/L) LgHg (μg/L) Gender Male Female Household incomes o2000 Yuan/m/per person 2000–5000 Yuan/m/per person 45000Yuan/m/per person Maternal education Middle school or lower High scool Bachelor degree Higher than Bachelor degree Paternal education Middle school or lower High school Bachelor degree Higher than Bachelor degree Maternal occupation White collar Technician Blue collar Housewife Paternal occupation White collar Technician Blue collar Unemployed Maternal smoking No Yes Maternal passive smoking Never Rarely Often Always n
p-value
OR (95%CI)
p-value
0.482 0.961 o 0.001 0.002 0.265 0.192 0.802 0.226
1.0 1.1 1.9 13.0 1.0 1.0 1.0 1.2 2.5
0.792 0.133 o 0.001 0.457 0.852 0.970 0.762 0.044
0 0.1 ( 0.3, 0.1)
0.295
1.0 1.1 (0.6, 2.1)
0.687
0 0.1 ( 0.3, 0.1) 0.5 ( 0.7, 0.2)
0.407 o 0.001
1.0 1.0 (0.5, 2.1) 1.9 (0.9, 4.0)
0.927 0.100
0 0.1 ( 0.4, 0.1) 0.2 ( 0.3, 0.0) 0.7 ( 1.1, 0.3)
0.295 0.103 0.001
1.0 1.6 (0.7, 3.9) 1.4 (0.6, 2.9) 4.9 (1.7, 14.0)
0.313 0.446 0.003
0.579 0.006 0.004
1.0 0.9 (0.3, 2.6) 1.7 (0.8, 3.8) 4.6 (1.6, 13.0)
0.805 0.175 0.004
0.912 0.133 0.175
1.0 1.2 (0.4, 3.8) 1.7 (0.6, 4.5) 3.3 (0.3, 31.0)
0.793 0.313 0.302
0.032 0.467 0.265
1.0 0.7 (0.2, 2.3) 1.0 (0.5, 2.2) 1.3 (0.4, 4.6)
0.522 0.999 0.700
0.0 ( 0.5, 0.6)
0.928
1.0 1.0 (0.3, 3.5)
0.958
0.0 ( 0.2, 0.2) 0.0 ( 0.3, 0.3) 0.4 ( 0.5, 1.4)
0.737 0.874 0.368
1.0 0.7 (0.8, 1.1) 0.9 (0.49, 2.0) 0.9 (0.50, 2.1)
0.125 0.965 0.980
0 0.1 0.0 1.2 0.0 0.0 0.0 0.1 0.2
( 0.1, 0.3) ( 0.2, 0.2) ( 1.4, 1.0) ( 0.1, 0.0) ( 0.0, 0.0) (0.0, 0.0) ( 0.4, 0.5) ( 0.6, 0.2)
0 0.1 ( 0.2, 0.3) 0.3 ( 0.5, 0.1) 0.5 ( 0.9, 0.2) 0 0.0 ( 0.3, 0.3) 0.2 ( 0.4, 0.1) 0.6 ( 1.3, 0.2) 0 0.3 (0.0, 0.6) 0.1 ( 0.1, 0.3) 0.2 ( 0.6, 0.2) 0
0
(0.5, 2.8) (0.8, 4.3) (6.3, 26.6) (1.0, 1.1) (1.0, 1.0) (1.0, 1.0) (0.4, 3.5) (1.0, 6.0)
low NBNA: NBNA Scoreo37.
Table 3 The unadjusted and adjusted association between maternal education and cord serum manganese. Lg Mn(ug/L) Crude
Maternal education(%) Middle school or lower High school Bachelor degree Higher than bachelor degree
Lg Mn(ug/L) Adjusted
ß (95%CI)
p value
ß (95%CI)
p value
0 0.0 ( 0.2, 0.2) 0.4 (0.2, 0.5) 0.6 (0.2, 0.9)
0.782 o 0.001 0.002
0 0.2 ( 0.5, 0.1) 0.1 ( 0.2, 0.5) 0.0 ( 0.5, 0.6)
0.257 0.425 0.929
*Adjusted: adjust for maternal age, family incomes, maternal occupation, paternal occupation, dietary intakes.
We further applied two-piece-wise linear regression model to examine the threshold effect of LgMn on NBNA scores according to the smoothing plot (Table 5). The turning point of LgMn where the relationship between NBNA scores and LgMn started to change and became eminent was determined using trial method, which was to move the trial turning point along the pre-defined interval and picked up the one which gave maximum model likelihood. We also conducted log likelihood ratio test comparing one-line linear regression model with two-piece-wise linear model. Student t-test was used for the effects of Mn on NBNA five clusters (Table 6).
All analysis was done using Empower(R) (www.empowerstats.com, X&Y solutions, inc. Boston MA) and R (http://www.R-project.org).
3. Results In total, 933 term newborns (494 males and 439 females) and their mothers were recruited. The median serum concentration of
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
Mn was 4.0 μg/L. Cord serum Mn was associated with NBNA score, maternal age, household incomes, maternal education, paternal education maternal occupation and birth weight (Table 1).
Fig. 1. Cord Serum Magenese Level and NBNA Score*. The nonlinear relationship between LgMn and NBNA scores was observed, and a threshold cord serum Mn of 5.0 μg/L existed for lower neonatal behavioral neurological assessments. *Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg exposure.
235
Table 2 shows the unadjusted associations between cord serum Mn and NBNA score, risk of low NBNA. Of the 933 infants, 44 (4.7%) had low NBNA. High level of Mn ( Z75th percentile) was associated with lower NBNA scores (ß¼ 1.2, 95% confidence interval [CI]¼ 1.4–1.0, p o0.001) and a higher risk of low NBNA (OR¼ 13.0, 95%CI ¼ 6.3–26.6, p o0.01). NBNA scores were also associated with maternal age, household incomes, maternal education and paternal education. We discovered a strange phenomenon which the babies with higher socioeconomic levels had higher cord Mn concentrations than those with lower levels. We took a multiple regression analysis to explore the association between maternal education and cord Mn concentrations after adjusting maternal age, family incomes, maternal occupation, paternal occupation, dietary intakes. Table 3 showed that the association between maternal education and cord Mn was not significant after adjusting the confounders. We found that there was interaction between materntal education and family income (p o0.05). Thus, we adjusted maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure as the confounders. Table 4 shows the adjusted associations between cord serum Mn and NBNA score, risk of low NBNA.Compared to low Mn level, high level (Z75th percentile) was negatively correlated with NBNA score (β ¼ 1.1, 95% CI: 1.4, 0.7), and high level (Z 75th percentile) was associated with an increased risk of low NBNA score (relative risk ¼9.4, 95% CI: 3.4, 25.7). Fig. 1 shows the nonlinear relationship between LgMn and NBNA scores adjusting for maternal age, maternal education family incomes, paternal education, maternal occupation, paternal
Table 4 The adjusted association between cord serum manganese and neonatal behavioral neurological assessments score. NBNA Scoren
Low NBNAn
ß (95%CI) Lg Mn quartiles (μg/L) oP25(Mn o 2.7) P25–50(Mn :2.7–4.0) P50–75(Mn :4.1–9.0) ZP75(Mn Z9.1)
p value
OR (95%CI)
p value
0.311 0.743 o 0.001
1.0 0.6 (0.2, 2.2) 1.3 (0.4, 3.9) 9.4 (3.4, 25.7)
0.457 0.676 o 0.001
0 0.2 ( 0.2, 0.5) 0.1 ( 0.3, 0.4) 1.1 ( 1. 4, 0.7)
n Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.
Table 5 The threshold effect of fetal manganese level on neonatal behavioral neurological assessments. Adjustedn
Crude
NBNA score (μg/L) Lg Mn o 0.7 (Mn o5.0) Lg MN Z 0.7 (Mn Z 5.0)
ß (95%CI)
p-value
ß (95%CI)
p-value
0.1 ( 0.3, 0.5) 1.6 ( 1.8, 1.3)
0.554 o 0.001
0.0 ( 0.7, 0.7) 1.4 ( 1.8, 1.0)
0.945 o 0.001
n Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.
Table 6 The adjusted effect of high level manganese on five neonatal behavioral neurological assessments clustersn.
Lg Mn o 0.7 (Mn o5.0 μg/L) Lg Mn Z 0.7 (Mn Z5.0 μg/L) P value
Total score
Behavior
Active tone
Passive tone
primary reflexes
General reactions
39.4 70.8 38.3 71.7 o 0.001
11.7 7 0.5 11.2 7 1.0 o 0.001
7.9 70.5 7.4 70.9 o 0.001
7.9 7 0.3 7.9 7 0.3 40.05
6.0 7 0.2 6.0 7 0.2 4 0.05
6.0 7 0.1 5.9 7 0.2 o 0.001
n Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.
236
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.. A nonlinear relationship was observed between cord serum Mn and NBNA score. The latter decreased with increasing Mn levels after the turning point (LgMn ¼0.7, Mn ¼ 5.0 μg/L). The threshold effect of LgMn on NBNA scores was significant after adjusting for potential confounders. The regression coefficient was 0.0 (95% CI: 0.7, 0.7) for Lg MNo 0.7 (Mn o5.0 μg/L) while 1.4 (95% CI: 1.8, 1.0) for Lg MNZ 0.7 (MnZ 5.0 μg/L) (Table 5). Table 6 shows that a high level of Mn (MnZ5.0 μg/L) adversely affects the total NBNA score as well as three NBNA clusters, including behavior, active tone and general reactions (Po0.001).
4. Discussion Our multicenter study indicates that higher in utero Mn exposure was significantly associated with lower NBNA score and a higher risk of low NBNA. We further revealed a threshold effect where a “safe level” of cord serum Mn may be considered based on the neonatal neurodevelopment assessment. The median level of Mn (4.0 μg/L) in our study was higher than the level reported in Japan (males: 1.6370.61 μg/L, females: 1.6871.04 μg/L) (Urushidate et al., 2010) and similar to the level reported in Brazil (3.572.2 μg/L) (Farias et al., 2010). In the present study, we discovered a strange phenomenon which the babies with higher socioeconomic levels had higher cord Mn concentrations than those with lower levels. We speculate that the phenomenon maybe due to the older age in the educated parents and dietary factor. We took a multiple regression to analyze the association between maternal education and cord Mn concentrations, and found that the association between maternal education and cord Mn was not significant after adjusting the confounders. Bao et al. (1991) formulated NBNA, and subsequently validated in 714 normal newborns in 12 provinces of China. Among the 2142 times examination, 90% of the 714 newborns had a score of 39% or 40% and 97% were equal to or greater than 37%, and none below 35%. The mean NBNA score and the prevalence of low NBNA in this study were similar to those obtained in other cities in China (Bao et al., 1991, 1993). The brain undergoes two periods of rapid maturation: first in utero and then during the first few months of life. The developing nervous system is a prime target for the disrupting effects of Mn. Mn-related embryotoxic and fetotoxic effects have been observed. In animal fetus, cerebral Mn accumulation is dose dependent (Lai et al., 1999). In humans, Mn easily crosses the placenta via active transport mechanisms (Krachler et al., 1999). Thus, fetal life can be regarded as a period of great vulnerability to Mn toxicity at low environmental levels (Takser et al., 2003). Several mechanisms of Mn neurotoxicity may be involved, including the disruption of mitochondrial metabolism (Zhang et al., 2008), oxidative stress (Milatovic et al., 2009), alteration of iron homeostasis (Zheng et al., 2000), inflammation (Milatovic et al., 2009), and altered glutamate and dopamine metabolism (Erikson et al., 2008). The globus pallidus and substantia nigra pars reticulate are most affected by Mn while injury to the caudate nucleus and the putamen are severe (Olanow et al., 1996; Stanwood et al., 2009). Oxidative stress in the striatum is associated with impairment of motor activity (de Oliveira et al., 2007). It has also been demonstrated that Mn-dependent increased oxidative stress formation can interfere with the removal of glutamate from the synaptic cleft (Erikson et al., 2002), resulting in excitotoxicity (Xu et al., 2010). However, epidemiologic studies on the effects of prenatal Mn exposure on child neuropsychological development are limited. Most epidemiologic research focused on the postnatal Mn exposure. Bouchard et al. (2007) reported a significant association
between hair Mn levels and hyperactive and oppositional behavior in children exposed to Mn. Decrements in IQ scores in Korean children were associated with elevated blood Mn levels in a population-based study (Kim et al., 2009). Zoni et al. (2007) in a review of recent studies on neurobehavioral performance and Mn exposure across the lifespan, suggested that children's cognitive functions might be particularly vulnerable to Mn. Previous studies have indicated that Mn was more strongly associated with Performance IQ than with Verbal IQ in children (Takser et al., 2003; Erikson et al., 2007). For instance, postnatal exposure to environmental level of Mn has been associated with a slower rate of development, diminished intellectual function and poorer learning and recall (Bouchard et al., 2007; Wasserman et al., 2006). Wasserman et al. (2006) and Bouchard et al. (2011) observed a stronger association of water Mn level with Performance IQ than with Verbal IQ in school-age children. Meanwhile, some epidemiologic data on the effects of prenatal Mn exposure on old children's IQ also indicated that the exposure to high in- utero Mn levels can affect children's psychomotor development, including non-verbal memory and boy's manual ability and disinhibitory behavior (Takser et al., 2003; Erikson et al., 2007). Our study found that behavior, active tone and general reactions but not the passive tone and primary reflexes, were inversely associated with Mn level. More studies are needed to explore the potential relationship between behavior, active tone, general reactions in newborns and non-verbal memory, manual ability, and disinhibitory behaviors in older children in older children. Many studies have examined the effects of high levels of acute or chronic occupational exposure to Mn (10–150 μg/L). However, little is known about exposure at subtoxic levels (o10 μg/L) (Aschner et al., 2007). To date, the reference intervals of what is considered a ‘normal’ Mn level vary considerably. Takagi et al. (2002) measured whole blood and plasma Mn levels in healthy volunteers and consider a range of 1.9–5.8 μg/L in plasma as normal. In contrast, a UK study quotes a reference range of 7–27 nmol/L (0.38–1.4 μg/L) as the normal plasma Mn level (Reynolds et al., 1998). An Australian reference laboratory uses an average of intervals published by Australasian hospitals and the Tietz's Textbook of Clinical Chemistry (Tietz, 1998) and suggested that the normal values range between 5–33 nmol/L (0.3–1.8 μg/L) in serum. However, Mn levels and requirements are known to vary by lifestage, especially for the fetal period. For fetal development, the possible “safe level” of Mn remains unknown. In our study, none of the women were exposed to Mn at work, and their exposure likely resulted from the general environment. We found that environmental level of Mn exposure in utero may still affect neurobehavioral development in children and a threshold cord serum Mn higher than 5.0 μg/L existed for lower NBNA score. Previously, we also found that cord serumMn levels less than 5.0 μg/L may be considered safe with respect to neonatal ponderal index assessment (Yu et al., 2013). This two results concerning Mn and neonatal's growth and development, furtherly indicated that a threshold cord serum of 5.0 μg/L existed. Ours is a relative large, cohort study of prenatal Mn exposure on neonatal neurodevelopment. However, the limitations are worth noting. First, dietary Mn intake and environmental risk factors were not investigated. Second, the study lacked the longterm follow-up investigation into the impact of prenatal Mn on childhood development. However, now we are following up these children's neurodevelopment. Third, the maternal serum Mn during pregnancy was not analyzed to further explore the relationship between maternal serum Mn and neonatal development.
5. Conclusions In conclusion, our study showed that prenatal exposure to Mn at environment relevant level was significantly and negatively
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
associated with fetal neurodevelopment. The “safety” threshold appears at or below 5.0 μg/L in umbilical cord serum Mn based on the neonatal neurodevelopment.
Information on the approval by the medical ethics committee The study protocol was approved by the Medical Ethics Committee of Shanghai XinHua Hospital, affiliated with the Shanghai Jiao Tong University School of Medicine (No. 200807).
The study protocol was approved by the Medical Ethics Committee(No. 200807).
Acknowledgments The authors would like to thank Dr. Chengzhong Chen (DanaFarber Cancer Institute, Harvard Medical School, Boston, MA) for providing statistical guidance on Empower(R). This study was funded by the National Natural Science Foundation of China (No. 81373004), the National Basic Research Program of China (“973” Program, No. 2012CB525001), and the Shanghai Committee of Science and Technology, China (No.114119A1600 and No. 11ZR1429800). References Amiel-Tison, C., 2002. Update of the Amiel-Tison neurologic assessment for the term neonate or at 40 weeks corrected age. Pediatr. Neurol. 27, 196–212. Aschner, M., Guilarte, T.R., Schneider, J.S., Zheng, W., 2007. Manganese: recent advances in understanding its transport and neurotoxicity. Toxicol. Appl. Pharmacol. 221, 131–147. Bao, X.L., Yu, R.J., Li, Z.S., Zhang, B.L., 1991. Twenty-item behavioral neurological assessment for normal newborns in 12 cities of China. Chin. Med. J. (Engl.) 104, 742–746.
237
Bao, X.L., Yu, R.J., Li, Z.S., 1993. 20-item neonatal behavioral neurological assessment used in predicting prognosis of asphyxiated newborn. Chin. Med. J. (Engl.) 106, 211–215. Bouchard, M., Laforest, F., Vandelac, L., Bellinger, D., Donna, Mergler, 2007. Hair manganese and hyperactive behaviors: pilotstudy of school-age children exposed through tap water. Environ. Health Perspect. 15, 122–127. Bouchard, M.F., Sauvé, S., Barbeau, B., Legrand, M., Brodeur, M.È., Bouffard, T., Limoges, E., et al., 2011. Intellectual impairment in school-age children exposed to manganese from drinking water. Environ. Health Perspect. 119, 138–143. Brazelton, T.B., 1973. Neonatal behavioral assessment scale. Clin. Dev. Med. 50, 1–66. Canals, J., Hernández-Martínez, C., Esparó, G., Fernández-Ballart, J., 2011. Neonatal behavioral assessment scale as predictor of cognitive development and IQ in full-terminfants: a 6-year longitudinal study. Acta Paediatr. 100, 1331–1337. de Oliveira, M.R., de Bittencourt Pasquali, M.A., Silvestrin, R.B., EST, Mello, Moreira, J.C., 2007. Vitamin A supplementation induces a prooxidative state in the striatum and impairs locomotory and exploratory activity of adult rats. Brain Res. 1169, 112–119. Dubowitz, L., Ricciw, D., Mercuri, E., 2005. The Dubowitz neurological examination of the full-term newborn. Ment. Retard. Dev. Disabil. Res. Rev. 11, 52–60. Erikson, K.M., Dorman, D.C., Lash, L.H., Aschner, M., 2008. Duration of airbornemanganese exposure in rhesus monkeys is associated with brain regional changes in biomarkers of neurotoxicity. Neurotoxicology 29, 377–385. Erikson, K.M., Suber, R.L., Aschner, M., 2002. Glutamate/aspartate transporter (GLAST), taurine transporter and metallothionein mRNA levels are differentially altered in astrocytes exposed to manganese chloride, manganese phosphate or manganese sulfate. Neurotoxicology 23, 281–288. Erikson, K.M., Thompson, K., Aschner, J., Aschner, M., 2007. Manganese neurotoxicity: a focus on the neonate. Pharmacol. Ther. 113, 369–377. Farias, A.C., Cunha, A., Benko, C.R., McCracken, J.T., Costa, M.T., Farias, L.G., Cordeiro, M.L., 2010. Manganese in children with attention-deficit/hyperactivity disorder: relationship with methylphenidate exposure. J. Child Adolesc. Psychopharmacol. 20, 113–118. Furbee, B., 2011. Welding and parkinsonism. Neurol. Clin. 29, 623–640. Gao, Y., Yan, C.H., Tian, Y., Wang, Y., Xie, H.F., Zhou, X., et al., 2007. Prenatal exposure to mercury and neurobehavioral development of neonates in Zhoushan City, China. Environ. Res. 105, 390–399. Kim, Y., Kim, B.N., Hong, Y.C., Shin, M.S., Yoo, H.J., Kim, J.W., et al., 2009. Co-exposure to environmental lead and manganese affects the intelligence of school-aged children. NeuroTox 30, 564–571. Krachler, M., Rossipal, E., Micetic-Tuck, D., 1999. Trace element transfer from the mother to the newborn—investigations on triplets of colostrum, maternal and umbilical sera. Eur. J. Clin. Nutr. 53, 486–494. Lai, J.C.K., Minski, M.J., Chan, A.W.K., Leung, T.K.C., Lim, L., 1999. Manganese mineral interactions in brain. NeuroToxicology 20, 433–444. Milatovic, D., Zaja-Milatovic, S., Gupta, R.C., Yu, Y., Aschner, M., 2009. Oxidative damage and neurodegeneration in manganese-induced neurotoxicity. Toxicol. Appl. Pharmacol. 240, 219–225. Olanow, C.W., Good, P.F., Shinotoh, H., Hewitt, K.A., Vingerhoets, F., Snow, B.J., et al., 1996. Manganese intoxication in the rhesus monkey: a clinical, imaging, pathologic, and biochemical study. Neurology 46, 492–498. Reynolds, N., Blumsohn, A., Baxter, J.P., Houston, G., Pennington, C.R., 1998. Manganese requirement and toxicity in patients on home parenteral nutrition. Clin. Nutr. 17, 227–230. Sarnat, H.B., 1984. Anatomic and physiologic correlates of neurologic development in prematurity. Grune & Stratton, Orlando, pp. 1–25 (Topics in neonatal neurology). Stanwood, G.D., Leitch, D.B., Savchenko, V., Wu, J., Fitsanakis, V.A., Anderson, D.J., et al., 2009. Manganese exposure is cytotoxic and alters dopaminergic and GABAergic neurons within the basal ganglia. J. Neurochem. 110, 378–389. Takagi, Y., Okada, A., Sando, K., Wasa, M., Yoshida, H., Hirabuki, N., 2002. Evaluation of indexes of in vivo manganese status and the optimal intravenous dose for adult patients undergoing home parenteral nutrition. Am. J. Clin. Nutr. 75, 112–118. Takser, L., Mergler, D., Hellier, G., Sahuquillo, J., Huel, G., 2003. Manganese, monoamine metabolite levels at birth, and child psychomotor development. Neurotoxicology 24, 667–674. Tietz, N.W. (Ed.), 1998. Textbook of Clinical Chemistry, 3rd ed. WB Saunders, Philadelphia. Urushidate, S., Matsuzaka, M., Okubo, N., Iwasaki, H., Hasebe, T., Tsuya, R., et al., 2010. Association between concentration of trace elements in serum and bronchial asthmaamong Japanese general population. J. Trace Elem. Med. Biol. 24, 236–242. Wasserman, G.A., Liu, X., Parvez, F., Ahsan, H., Levy, D., Factor-Litvak, P., et al., 2006. Water manganese exposure and children's intellectual function in Araihazar, Bangladesh. Environ. Health Perspect. 114, 124–129. World Health Organization, 1981. Manganese. Environmental Health Criteria, 17. WHO, Geneva. Xu, B., Xu, Z.F., Deng, Y., 2010. Manganese exposure alters the expression of Nmethyl-D-aspartate receptor subunit mRNAs and proteins in rat striatum. J. Biochem. Mol. Toxicol. 24, 1–9. Yu, X., Cao, L., Yu, X., 2013. Elevated cord serum manganese level is associated with neonatal high ponderal index. Environ. Res. 121, 79–83.
238
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
Zhang, F., Xu, Z., Gao, J., Xu, B., Deng, Y., 2008. in vitro effect of manganese chloride exposure on energy metabolism and oxidative damage of mitochondria isolated from rat brain. Environ. Toxicol. Pharmacol. 26, 232–236. Zheng, W., Kim, H., Zhao, Q., 2000. Comparative toxicokinetics of manganese chloride and methylcyclopentadienyl manganese tricarbonyl (MMT) in Sprague-Dawley rats. Toxicol. Sci. 54, 295–301.
Zoni, S., Albini, E., Lucchini, R., 2007. Neuropsychological testing for the assessment of manganese neurotoxicity: a review and a proposal. Am. J. Ind. Med. 50, 812–830.
Contents lists available at ScienceDirect
Environmental Research journal homepage: www.elsevier.com/locate/envres
Prenatal exposure to manganese at environment relevant level and neonatal neurobehavioral development Xiao-Dan Yu, Jun Zhang, Chong-Huai Yan n, Xiao-Ming Shen MOE-Shanghai Key Laboratory of Children's Environmental Health, XinHua Hospital affiliated to Shanghai Jiao Tong University School of Medicine, 1665 Kong Jiang Road, Shanghai 200092, China
art ic l e i nf o
a b s t r a c t
Article history: Received 16 September 2013 Received in revised form 11 April 2014 Accepted 16 April 2014
Background: Effects of prenatal Manganese (Mn) exposure at an environmental relevant level on neonatal neurodevelopment remains unclear. Objectives: In the multi-center study, we assessed the impact of low level prenatal Mn exposure on neonatal behavioral neurological assessments (NBNA), and explore a threshold umbilical cord blood Mn on neonatal neurological development. Methods: We investigated 933 mother-newborn pairs in Shanghai, China, from 2008 through 2009. Umbilical cord serum concentrations of Mn were measured and NBNA tests were conducted. The NBNA contains five clusters: behavior, active tone, passive tone, primary reflexes and general assessment with a maximal total score of 40. The scoreo37 is defined as low. Results: The median serum Mn concentration was 4.0 μg/L. Of the 933 infants, 44 (4.7%) had low NBNA. After adjusting for potential confounders, a high level of Mn ( Z 75th percentile ) was associated with a lower NBNA score (adjusted ß ¼ 1.1, 95% CI: 1.4–0.7, p o0.01) and a higher risk of low NBNA (adjusted OR ¼9.4, 95% CI: 3.4–25.7, po 0.01). A nonlinear relationship was observed between cord serum Mn and NBNA after adjusting for potential confounders. NBNA score decreased with increasing Mn levels after 5.0 μg/L(LgMn Z0.7). The cord serum MnZ5.0 μg/L had adverse effects on behavior, active tone and general reactions of clusters (p o0.001). Conclusions: High prenatal Mn exposure even at an environmental relevant level, is associated with poor fetal neurobehavioral development in a nonlinear pattern. A threshold cord serum Mn of 5.0 μg/L existed for lower neonatal behavioral neurological assessments. & 2014 Published by Elsevier Inc.
Keywords: Prenatal exposure Mn Neonatal neurobehavioral development Epidemiology
1. Introduction Manganese (Mn) is an essential nutrient to human, but it also has the potential to produce neurotoxic effects when accumulating in an organ, especially the brain (World Health Organization, 1981). Studies of Mn exposure in workplace have demonstrated that exposure to high doses of Mn is associated with irreversible neurodegenerative disorders resembling idiopathic Parkinson disease (Furbee, 2011). More recently, child environmental exposure to Mn has gained interest. Postnatal Mn exposure was found to interfere with development of brain functions (Zoni et al. 2007; Bouchard et al., 2007). However, few epidemiologic data are available on the effects of in utero Mn exposure on child neuropsychological development. In a birth cohort study in France, cord
Abbreviations: Mn, Manganese; NBNA, neonatal behavioral neurological assessments; LOD, limit of detection; ICP-MS, Inductively Coupled Plasma Mass Spectrometry n Corresponding author. Fax: þ86 21 25078875. E-mail address: [email protected] (C.-H. Yan). http://dx.doi.org/10.1016/j.envres.2014.04.012 0013-9351/& 2014 Published by Elsevier Inc.
blood Mn was negatively associated with attention and non-verbal memory and boys' manual ability at 3 years, after adjusting for mother's educational level (Takser et al., 2003). In a U.S. study, Mn was analyzed in the enamel of deciduous teeth. Its concentration in tissue formed during the intra-uterine phase was significantly associated with disinhibitory behavior evaluated at 36 and 54 months of age (Erikson et al., 2007). The history of neonatal neurobehavioral assessment began in early last century. Sarnat (1984); Amiel-Tison (2002) and Dubowitz et al. (2005) gradually developed neurological assessment in the newborn via examining tone and reflexes. After realizing that a newborn can regulate his behavior, Brazelton developed the Neonatal Behavioral Assessment Scale (NBAS), the first truly standardized, comprehensive assessment of newborn neurobehavior (Brazelton, 1973). Canals et al. (2011) confirmed that neonatal self-regulation behaviors were the best predictors of infant development and intelligence and that NBAS could be a useful tool to observe behaviors related to later development in healthy infants. Based on the method of Brazelton and Amiel-Tison for behavioral neurological measurement in newborns as well as
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
their own experience, Bao et al. (1991) formulated Neonatal Behavioral Neurological Assessments (NBNA), and subsequently repeated measurements in children showed that the total scores increased with child age and the NBNA had distinct stability and reliability, and was not influenced by geographic location. Therefore, it is convenient for a large survey (Bao et al. 1993).We used NBNA to measure neurobehavioral development of neonates in the study. To our best knowledge, the potential association between prenatal Mn exposure on neurobehavior of neonates have not been studied, though it is known that neonatal irritability and selfregulation are related to cognitive development and intelligence both short term and long term (Canals et al., 2011). Furthermore, there are no data available suggesting “safe” level of cord serum Mn. We conducted a study of prenatal Mn exposures and neonatal development in a mother-infant cohort in Shanghai, China. The objectives of the present study are (1) to determine the cord blood level of Mn and the corresponding effects on neurobehavioral development in neonates; and (2) to explore possible nontoxic level of Mn for neonatal neurodevelopment.
233
the Medical Ethics Committee of Shanghai XinHua Hospital affiliated to Shanghai Jiao Tong University School of Medicine (No.2008.7). After the women had signed the consent form, an in-person interview was conducted to collect information on social and demographic characteristics and potential sources of metal exposure. Anthropometric measurements of the newborns were made by delivery room staff according to a standard anthropometric protocol. Information on gestational age along with characteristics of the birth and newborn were extracted from the medical records. NBNA was administered when the infants were 3 days old as previously described (Gao et al., 2007). NBNA assesses functional abilities, most reflexes and responses, and stability of behavioral status during the examination. It contains five clusters: behavior (six items), passive tone (four items), active tone (four items), primary reflexes (three items), and general assessment (three items). Each item has three levels (0, 1 and 2). Twenty items have a maximal total score of 40. Neonates with a total score of equal to or more than 37 are considered well developed while below 37 are considered low NBNA (Bao et al., 1991). NBNA assessments were conducted by ten examiners who were rigorously trained and certified by the creator of Chinese NBNA, Professor Bao (Bao et al., 1991). Umbilical cord blood was collected and serum was separated. All the samples were immediately frozen at 40 1C and shipped in batches to the central laboratory. Serum Mn concentration was measured by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) (7500 CE, Agilent) as described previously (Yu et al., 2013). The limit of detection (LOD) for Mn was 0.13 μg/L.
2.2. Statistical analysis
2. Methods 2.1. Study subjects This multicenter study recruited 1377 healthy pregnant women who came to the hospital to deliver a term (37–42 weeks of gestation) singleton infant in 10 maternity hospitals in Shanghai, China, from 2008 to 2009, as we previously described (Yu et al., 2013). A total of 48 infants with disorders closely associated with adverse severe neurodevelopment such as traumatic brain injury, meningitis, epilepsy, severe neonatal jaundice were excluded. Of the 1329 infants, 933 parents signed the consent form to do the NBNA test. The study protocol was approved by
Cord serum concentrations of Mn were expressed as μg/L. Values below LOD were imputed using the default value of ½ LOD (6 of 933 subjects were below LOD). We first examined the distribution of NBNA score and Mn level. The latter was severely skewed towards the left. Thus, we performed log transformation before analysis. We divided study population into four groups based on quartiles of LgMn (Table 1) and performed univariate analysis and multiple regression examining group difference of both mean NBNA score and percent of low NBNA (Tables 2 and 3). We then applied a generalized additive model to estimate the independent relationship between LgMn and NBNA scores, with adjustment for potential confounders (Fig. 1).
Table 1 Characteristics of mothers-infants and the associations with cord serum manganese. P value
Lg Mn quartiles
N NBNA Score Behavior Active tone Passive tone Primary reflexes General reactions Birth weight (g) Gestational age (d) Maternal age (y) Gender(%) male female Household incomes (Yuan/m/Person)(%) o 2000 2000–5000 45000 Maternal education(%) Middle school or lower High school Bachelor degree Higher than bachelor degree Paternal education (%) Middle school or lower High school Bachelor degree Higher than bachelor degree Maternal occupation (%) White collar Technician Blue collar Housewife
o P25 (Mn o2.7 μg/L)
P25–50 (Mn:2.7–4.0 μg/L)
P50–75 (Mn:4.1–9.0 μg/L)
Z P75 (Mn Z9.1 μg/L)
231 39.4 7 0.9 11.8 7 0.5 7.7 70.5 7.9 70.3 6.0 7 0.2 6.0 7 0.2 3454.8 7 410.7 275.9 7 8.6 27.1 73.9
228 39.5 7 0.9 11.8 7 0.5 7.8 7 0.4 7.9 7 0.3 6.0 70.2 6.0 70.2 3396.4 7429.1 276.7 7 7.8 27.3 74.2
239 39.4 7 1.0 11.8 7 0.6 7.7 7 0.6 7.9 7 0.3 6.0 7 0.1 6.0 7 0.2 3347.2 7 393.3 275.8 7 7.6 27.0 74.7
235 38.2 7 1.6 11.2 7 1.1 7.2 70.9 7.9 70.4 6.0 7 0.2 6.0 7 0.0 3415.6 7 418.2 275.3 7 7.1 28.3 7 4.0
53.7 46.3
49.8 50.2
57.1 42.9
51.1 48.9
52.1 34.3 13.6
51.0 35.3 13.7
48.0 34.4 17.6
28.5 35.7 35.7
39.3 22.8 33.9 4.0
41.3 418.8 37.2 2.7
40.4 18.3 38.7 2.6
21.5 14.9 54.8 8.8
36.9 21.2 37.4 4.5
34.7 21.2 38.7 5.4
35.2 18.9 42.5 3.4
15.2 16.6 58.3 9.9
22.1 27.1 49.7 1.1
22.2 29.2 47.6 1.1
24.4 27.4 46.7 1.5
13.3 15.0 69.9 1.7
o 0.001 o 0.001 o 0.001 0.235 0.849 0.050 0.047 0.341 0.010 0.396
o 0.001
o 0.001
o 0.001
0.001
234
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
Table 2 The unadjusted association between factors and neonatal behavioral neurological assessments score. Low NBNAn
NBNA score ß (95%CI) Lg Mn quartiles (μg/L) oP25(Mn o 2.7) P25–50(Mn:2.7–4.0) P50–75(Mn:4.1–9.0) ZP75(Mn Z 9.1) Maternal age Gestational age Birth weight LgPb (μg/L) LgHg (μg/L) Gender Male Female Household incomes o2000 Yuan/m/per person 2000–5000 Yuan/m/per person 45000Yuan/m/per person Maternal education Middle school or lower High scool Bachelor degree Higher than Bachelor degree Paternal education Middle school or lower High school Bachelor degree Higher than Bachelor degree Maternal occupation White collar Technician Blue collar Housewife Paternal occupation White collar Technician Blue collar Unemployed Maternal smoking No Yes Maternal passive smoking Never Rarely Often Always n
p-value
OR (95%CI)
p-value
0.482 0.961 o 0.001 0.002 0.265 0.192 0.802 0.226
1.0 1.1 1.9 13.0 1.0 1.0 1.0 1.2 2.5
0.792 0.133 o 0.001 0.457 0.852 0.970 0.762 0.044
0 0.1 ( 0.3, 0.1)
0.295
1.0 1.1 (0.6, 2.1)
0.687
0 0.1 ( 0.3, 0.1) 0.5 ( 0.7, 0.2)
0.407 o 0.001
1.0 1.0 (0.5, 2.1) 1.9 (0.9, 4.0)
0.927 0.100
0 0.1 ( 0.4, 0.1) 0.2 ( 0.3, 0.0) 0.7 ( 1.1, 0.3)
0.295 0.103 0.001
1.0 1.6 (0.7, 3.9) 1.4 (0.6, 2.9) 4.9 (1.7, 14.0)
0.313 0.446 0.003
0.579 0.006 0.004
1.0 0.9 (0.3, 2.6) 1.7 (0.8, 3.8) 4.6 (1.6, 13.0)
0.805 0.175 0.004
0.912 0.133 0.175
1.0 1.2 (0.4, 3.8) 1.7 (0.6, 4.5) 3.3 (0.3, 31.0)
0.793 0.313 0.302
0.032 0.467 0.265
1.0 0.7 (0.2, 2.3) 1.0 (0.5, 2.2) 1.3 (0.4, 4.6)
0.522 0.999 0.700
0.0 ( 0.5, 0.6)
0.928
1.0 1.0 (0.3, 3.5)
0.958
0.0 ( 0.2, 0.2) 0.0 ( 0.3, 0.3) 0.4 ( 0.5, 1.4)
0.737 0.874 0.368
1.0 0.7 (0.8, 1.1) 0.9 (0.49, 2.0) 0.9 (0.50, 2.1)
0.125 0.965 0.980
0 0.1 0.0 1.2 0.0 0.0 0.0 0.1 0.2
( 0.1, 0.3) ( 0.2, 0.2) ( 1.4, 1.0) ( 0.1, 0.0) ( 0.0, 0.0) (0.0, 0.0) ( 0.4, 0.5) ( 0.6, 0.2)
0 0.1 ( 0.2, 0.3) 0.3 ( 0.5, 0.1) 0.5 ( 0.9, 0.2) 0 0.0 ( 0.3, 0.3) 0.2 ( 0.4, 0.1) 0.6 ( 1.3, 0.2) 0 0.3 (0.0, 0.6) 0.1 ( 0.1, 0.3) 0.2 ( 0.6, 0.2) 0
0
(0.5, 2.8) (0.8, 4.3) (6.3, 26.6) (1.0, 1.1) (1.0, 1.0) (1.0, 1.0) (0.4, 3.5) (1.0, 6.0)
low NBNA: NBNA Scoreo37.
Table 3 The unadjusted and adjusted association between maternal education and cord serum manganese. Lg Mn(ug/L) Crude
Maternal education(%) Middle school or lower High school Bachelor degree Higher than bachelor degree
Lg Mn(ug/L) Adjusted
ß (95%CI)
p value
ß (95%CI)
p value
0 0.0 ( 0.2, 0.2) 0.4 (0.2, 0.5) 0.6 (0.2, 0.9)
0.782 o 0.001 0.002
0 0.2 ( 0.5, 0.1) 0.1 ( 0.2, 0.5) 0.0 ( 0.5, 0.6)
0.257 0.425 0.929
*Adjusted: adjust for maternal age, family incomes, maternal occupation, paternal occupation, dietary intakes.
We further applied two-piece-wise linear regression model to examine the threshold effect of LgMn on NBNA scores according to the smoothing plot (Table 5). The turning point of LgMn where the relationship between NBNA scores and LgMn started to change and became eminent was determined using trial method, which was to move the trial turning point along the pre-defined interval and picked up the one which gave maximum model likelihood. We also conducted log likelihood ratio test comparing one-line linear regression model with two-piece-wise linear model. Student t-test was used for the effects of Mn on NBNA five clusters (Table 6).
All analysis was done using Empower(R) (www.empowerstats.com, X&Y solutions, inc. Boston MA) and R (http://www.R-project.org).
3. Results In total, 933 term newborns (494 males and 439 females) and their mothers were recruited. The median serum concentration of
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
Mn was 4.0 μg/L. Cord serum Mn was associated with NBNA score, maternal age, household incomes, maternal education, paternal education maternal occupation and birth weight (Table 1).
Fig. 1. Cord Serum Magenese Level and NBNA Score*. The nonlinear relationship between LgMn and NBNA scores was observed, and a threshold cord serum Mn of 5.0 μg/L existed for lower neonatal behavioral neurological assessments. *Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg exposure.
235
Table 2 shows the unadjusted associations between cord serum Mn and NBNA score, risk of low NBNA. Of the 933 infants, 44 (4.7%) had low NBNA. High level of Mn ( Z75th percentile) was associated with lower NBNA scores (ß¼ 1.2, 95% confidence interval [CI]¼ 1.4–1.0, p o0.001) and a higher risk of low NBNA (OR¼ 13.0, 95%CI ¼ 6.3–26.6, p o0.01). NBNA scores were also associated with maternal age, household incomes, maternal education and paternal education. We discovered a strange phenomenon which the babies with higher socioeconomic levels had higher cord Mn concentrations than those with lower levels. We took a multiple regression analysis to explore the association between maternal education and cord Mn concentrations after adjusting maternal age, family incomes, maternal occupation, paternal occupation, dietary intakes. Table 3 showed that the association between maternal education and cord Mn was not significant after adjusting the confounders. We found that there was interaction between materntal education and family income (p o0.05). Thus, we adjusted maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure as the confounders. Table 4 shows the adjusted associations between cord serum Mn and NBNA score, risk of low NBNA.Compared to low Mn level, high level (Z75th percentile) was negatively correlated with NBNA score (β ¼ 1.1, 95% CI: 1.4, 0.7), and high level (Z 75th percentile) was associated with an increased risk of low NBNA score (relative risk ¼9.4, 95% CI: 3.4, 25.7). Fig. 1 shows the nonlinear relationship between LgMn and NBNA scores adjusting for maternal age, maternal education family incomes, paternal education, maternal occupation, paternal
Table 4 The adjusted association between cord serum manganese and neonatal behavioral neurological assessments score. NBNA Scoren
Low NBNAn
ß (95%CI) Lg Mn quartiles (μg/L) oP25(Mn o 2.7) P25–50(Mn :2.7–4.0) P50–75(Mn :4.1–9.0) ZP75(Mn Z9.1)
p value
OR (95%CI)
p value
0.311 0.743 o 0.001
1.0 0.6 (0.2, 2.2) 1.3 (0.4, 3.9) 9.4 (3.4, 25.7)
0.457 0.676 o 0.001
0 0.2 ( 0.2, 0.5) 0.1 ( 0.3, 0.4) 1.1 ( 1. 4, 0.7)
n Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.
Table 5 The threshold effect of fetal manganese level on neonatal behavioral neurological assessments. Adjustedn
Crude
NBNA score (μg/L) Lg Mn o 0.7 (Mn o5.0) Lg MN Z 0.7 (Mn Z 5.0)
ß (95%CI)
p-value
ß (95%CI)
p-value
0.1 ( 0.3, 0.5) 1.6 ( 1.8, 1.3)
0.554 o 0.001
0.0 ( 0.7, 0.7) 1.4 ( 1.8, 1.0)
0.945 o 0.001
n Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.
Table 6 The adjusted effect of high level manganese on five neonatal behavioral neurological assessments clustersn.
Lg Mn o 0.7 (Mn o5.0 μg/L) Lg Mn Z 0.7 (Mn Z5.0 μg/L) P value
Total score
Behavior
Active tone
Passive tone
primary reflexes
General reactions
39.4 70.8 38.3 71.7 o 0.001
11.7 7 0.5 11.2 7 1.0 o 0.001
7.9 70.5 7.4 70.9 o 0.001
7.9 7 0.3 7.9 7 0.3 40.05
6.0 7 0.2 6.0 7 0.2 4 0.05
6.0 7 0.1 5.9 7 0.2 o 0.001
n Adjusted: maternal age, maternal education family incomes, paternal education, maternal occupation, paternal occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.
236
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
occupation, second-hand smoking, gestational age, gender, birth weight, Pb and Hg expoure.. A nonlinear relationship was observed between cord serum Mn and NBNA score. The latter decreased with increasing Mn levels after the turning point (LgMn ¼0.7, Mn ¼ 5.0 μg/L). The threshold effect of LgMn on NBNA scores was significant after adjusting for potential confounders. The regression coefficient was 0.0 (95% CI: 0.7, 0.7) for Lg MNo 0.7 (Mn o5.0 μg/L) while 1.4 (95% CI: 1.8, 1.0) for Lg MNZ 0.7 (MnZ 5.0 μg/L) (Table 5). Table 6 shows that a high level of Mn (MnZ5.0 μg/L) adversely affects the total NBNA score as well as three NBNA clusters, including behavior, active tone and general reactions (Po0.001).
4. Discussion Our multicenter study indicates that higher in utero Mn exposure was significantly associated with lower NBNA score and a higher risk of low NBNA. We further revealed a threshold effect where a “safe level” of cord serum Mn may be considered based on the neonatal neurodevelopment assessment. The median level of Mn (4.0 μg/L) in our study was higher than the level reported in Japan (males: 1.6370.61 μg/L, females: 1.6871.04 μg/L) (Urushidate et al., 2010) and similar to the level reported in Brazil (3.572.2 μg/L) (Farias et al., 2010). In the present study, we discovered a strange phenomenon which the babies with higher socioeconomic levels had higher cord Mn concentrations than those with lower levels. We speculate that the phenomenon maybe due to the older age in the educated parents and dietary factor. We took a multiple regression to analyze the association between maternal education and cord Mn concentrations, and found that the association between maternal education and cord Mn was not significant after adjusting the confounders. Bao et al. (1991) formulated NBNA, and subsequently validated in 714 normal newborns in 12 provinces of China. Among the 2142 times examination, 90% of the 714 newborns had a score of 39% or 40% and 97% were equal to or greater than 37%, and none below 35%. The mean NBNA score and the prevalence of low NBNA in this study were similar to those obtained in other cities in China (Bao et al., 1991, 1993). The brain undergoes two periods of rapid maturation: first in utero and then during the first few months of life. The developing nervous system is a prime target for the disrupting effects of Mn. Mn-related embryotoxic and fetotoxic effects have been observed. In animal fetus, cerebral Mn accumulation is dose dependent (Lai et al., 1999). In humans, Mn easily crosses the placenta via active transport mechanisms (Krachler et al., 1999). Thus, fetal life can be regarded as a period of great vulnerability to Mn toxicity at low environmental levels (Takser et al., 2003). Several mechanisms of Mn neurotoxicity may be involved, including the disruption of mitochondrial metabolism (Zhang et al., 2008), oxidative stress (Milatovic et al., 2009), alteration of iron homeostasis (Zheng et al., 2000), inflammation (Milatovic et al., 2009), and altered glutamate and dopamine metabolism (Erikson et al., 2008). The globus pallidus and substantia nigra pars reticulate are most affected by Mn while injury to the caudate nucleus and the putamen are severe (Olanow et al., 1996; Stanwood et al., 2009). Oxidative stress in the striatum is associated with impairment of motor activity (de Oliveira et al., 2007). It has also been demonstrated that Mn-dependent increased oxidative stress formation can interfere with the removal of glutamate from the synaptic cleft (Erikson et al., 2002), resulting in excitotoxicity (Xu et al., 2010). However, epidemiologic studies on the effects of prenatal Mn exposure on child neuropsychological development are limited. Most epidemiologic research focused on the postnatal Mn exposure. Bouchard et al. (2007) reported a significant association
between hair Mn levels and hyperactive and oppositional behavior in children exposed to Mn. Decrements in IQ scores in Korean children were associated with elevated blood Mn levels in a population-based study (Kim et al., 2009). Zoni et al. (2007) in a review of recent studies on neurobehavioral performance and Mn exposure across the lifespan, suggested that children's cognitive functions might be particularly vulnerable to Mn. Previous studies have indicated that Mn was more strongly associated with Performance IQ than with Verbal IQ in children (Takser et al., 2003; Erikson et al., 2007). For instance, postnatal exposure to environmental level of Mn has been associated with a slower rate of development, diminished intellectual function and poorer learning and recall (Bouchard et al., 2007; Wasserman et al., 2006). Wasserman et al. (2006) and Bouchard et al. (2011) observed a stronger association of water Mn level with Performance IQ than with Verbal IQ in school-age children. Meanwhile, some epidemiologic data on the effects of prenatal Mn exposure on old children's IQ also indicated that the exposure to high in- utero Mn levels can affect children's psychomotor development, including non-verbal memory and boy's manual ability and disinhibitory behavior (Takser et al., 2003; Erikson et al., 2007). Our study found that behavior, active tone and general reactions but not the passive tone and primary reflexes, were inversely associated with Mn level. More studies are needed to explore the potential relationship between behavior, active tone, general reactions in newborns and non-verbal memory, manual ability, and disinhibitory behaviors in older children in older children. Many studies have examined the effects of high levels of acute or chronic occupational exposure to Mn (10–150 μg/L). However, little is known about exposure at subtoxic levels (o10 μg/L) (Aschner et al., 2007). To date, the reference intervals of what is considered a ‘normal’ Mn level vary considerably. Takagi et al. (2002) measured whole blood and plasma Mn levels in healthy volunteers and consider a range of 1.9–5.8 μg/L in plasma as normal. In contrast, a UK study quotes a reference range of 7–27 nmol/L (0.38–1.4 μg/L) as the normal plasma Mn level (Reynolds et al., 1998). An Australian reference laboratory uses an average of intervals published by Australasian hospitals and the Tietz's Textbook of Clinical Chemistry (Tietz, 1998) and suggested that the normal values range between 5–33 nmol/L (0.3–1.8 μg/L) in serum. However, Mn levels and requirements are known to vary by lifestage, especially for the fetal period. For fetal development, the possible “safe level” of Mn remains unknown. In our study, none of the women were exposed to Mn at work, and their exposure likely resulted from the general environment. We found that environmental level of Mn exposure in utero may still affect neurobehavioral development in children and a threshold cord serum Mn higher than 5.0 μg/L existed for lower NBNA score. Previously, we also found that cord serumMn levels less than 5.0 μg/L may be considered safe with respect to neonatal ponderal index assessment (Yu et al., 2013). This two results concerning Mn and neonatal's growth and development, furtherly indicated that a threshold cord serum of 5.0 μg/L existed. Ours is a relative large, cohort study of prenatal Mn exposure on neonatal neurodevelopment. However, the limitations are worth noting. First, dietary Mn intake and environmental risk factors were not investigated. Second, the study lacked the longterm follow-up investigation into the impact of prenatal Mn on childhood development. However, now we are following up these children's neurodevelopment. Third, the maternal serum Mn during pregnancy was not analyzed to further explore the relationship between maternal serum Mn and neonatal development.
5. Conclusions In conclusion, our study showed that prenatal exposure to Mn at environment relevant level was significantly and negatively
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
associated with fetal neurodevelopment. The “safety” threshold appears at or below 5.0 μg/L in umbilical cord serum Mn based on the neonatal neurodevelopment.
Information on the approval by the medical ethics committee The study protocol was approved by the Medical Ethics Committee of Shanghai XinHua Hospital, affiliated with the Shanghai Jiao Tong University School of Medicine (No. 200807).
The study protocol was approved by the Medical Ethics Committee(No. 200807).
Acknowledgments The authors would like to thank Dr. Chengzhong Chen (DanaFarber Cancer Institute, Harvard Medical School, Boston, MA) for providing statistical guidance on Empower(R). This study was funded by the National Natural Science Foundation of China (No. 81373004), the National Basic Research Program of China (“973” Program, No. 2012CB525001), and the Shanghai Committee of Science and Technology, China (No.114119A1600 and No. 11ZR1429800). References Amiel-Tison, C., 2002. Update of the Amiel-Tison neurologic assessment for the term neonate or at 40 weeks corrected age. Pediatr. Neurol. 27, 196–212. Aschner, M., Guilarte, T.R., Schneider, J.S., Zheng, W., 2007. Manganese: recent advances in understanding its transport and neurotoxicity. Toxicol. Appl. Pharmacol. 221, 131–147. Bao, X.L., Yu, R.J., Li, Z.S., Zhang, B.L., 1991. Twenty-item behavioral neurological assessment for normal newborns in 12 cities of China. Chin. Med. J. (Engl.) 104, 742–746.
237
Bao, X.L., Yu, R.J., Li, Z.S., 1993. 20-item neonatal behavioral neurological assessment used in predicting prognosis of asphyxiated newborn. Chin. Med. J. (Engl.) 106, 211–215. Bouchard, M., Laforest, F., Vandelac, L., Bellinger, D., Donna, Mergler, 2007. Hair manganese and hyperactive behaviors: pilotstudy of school-age children exposed through tap water. Environ. Health Perspect. 15, 122–127. Bouchard, M.F., Sauvé, S., Barbeau, B., Legrand, M., Brodeur, M.È., Bouffard, T., Limoges, E., et al., 2011. Intellectual impairment in school-age children exposed to manganese from drinking water. Environ. Health Perspect. 119, 138–143. Brazelton, T.B., 1973. Neonatal behavioral assessment scale. Clin. Dev. Med. 50, 1–66. Canals, J., Hernández-Martínez, C., Esparó, G., Fernández-Ballart, J., 2011. Neonatal behavioral assessment scale as predictor of cognitive development and IQ in full-terminfants: a 6-year longitudinal study. Acta Paediatr. 100, 1331–1337. de Oliveira, M.R., de Bittencourt Pasquali, M.A., Silvestrin, R.B., EST, Mello, Moreira, J.C., 2007. Vitamin A supplementation induces a prooxidative state in the striatum and impairs locomotory and exploratory activity of adult rats. Brain Res. 1169, 112–119. Dubowitz, L., Ricciw, D., Mercuri, E., 2005. The Dubowitz neurological examination of the full-term newborn. Ment. Retard. Dev. Disabil. Res. Rev. 11, 52–60. Erikson, K.M., Dorman, D.C., Lash, L.H., Aschner, M., 2008. Duration of airbornemanganese exposure in rhesus monkeys is associated with brain regional changes in biomarkers of neurotoxicity. Neurotoxicology 29, 377–385. Erikson, K.M., Suber, R.L., Aschner, M., 2002. Glutamate/aspartate transporter (GLAST), taurine transporter and metallothionein mRNA levels are differentially altered in astrocytes exposed to manganese chloride, manganese phosphate or manganese sulfate. Neurotoxicology 23, 281–288. Erikson, K.M., Thompson, K., Aschner, J., Aschner, M., 2007. Manganese neurotoxicity: a focus on the neonate. Pharmacol. Ther. 113, 369–377. Farias, A.C., Cunha, A., Benko, C.R., McCracken, J.T., Costa, M.T., Farias, L.G., Cordeiro, M.L., 2010. Manganese in children with attention-deficit/hyperactivity disorder: relationship with methylphenidate exposure. J. Child Adolesc. Psychopharmacol. 20, 113–118. Furbee, B., 2011. Welding and parkinsonism. Neurol. Clin. 29, 623–640. Gao, Y., Yan, C.H., Tian, Y., Wang, Y., Xie, H.F., Zhou, X., et al., 2007. Prenatal exposure to mercury and neurobehavioral development of neonates in Zhoushan City, China. Environ. Res. 105, 390–399. Kim, Y., Kim, B.N., Hong, Y.C., Shin, M.S., Yoo, H.J., Kim, J.W., et al., 2009. Co-exposure to environmental lead and manganese affects the intelligence of school-aged children. NeuroTox 30, 564–571. Krachler, M., Rossipal, E., Micetic-Tuck, D., 1999. Trace element transfer from the mother to the newborn—investigations on triplets of colostrum, maternal and umbilical sera. Eur. J. Clin. Nutr. 53, 486–494. Lai, J.C.K., Minski, M.J., Chan, A.W.K., Leung, T.K.C., Lim, L., 1999. Manganese mineral interactions in brain. NeuroToxicology 20, 433–444. Milatovic, D., Zaja-Milatovic, S., Gupta, R.C., Yu, Y., Aschner, M., 2009. Oxidative damage and neurodegeneration in manganese-induced neurotoxicity. Toxicol. Appl. Pharmacol. 240, 219–225. Olanow, C.W., Good, P.F., Shinotoh, H., Hewitt, K.A., Vingerhoets, F., Snow, B.J., et al., 1996. Manganese intoxication in the rhesus monkey: a clinical, imaging, pathologic, and biochemical study. Neurology 46, 492–498. Reynolds, N., Blumsohn, A., Baxter, J.P., Houston, G., Pennington, C.R., 1998. Manganese requirement and toxicity in patients on home parenteral nutrition. Clin. Nutr. 17, 227–230. Sarnat, H.B., 1984. Anatomic and physiologic correlates of neurologic development in prematurity. Grune & Stratton, Orlando, pp. 1–25 (Topics in neonatal neurology). Stanwood, G.D., Leitch, D.B., Savchenko, V., Wu, J., Fitsanakis, V.A., Anderson, D.J., et al., 2009. Manganese exposure is cytotoxic and alters dopaminergic and GABAergic neurons within the basal ganglia. J. Neurochem. 110, 378–389. Takagi, Y., Okada, A., Sando, K., Wasa, M., Yoshida, H., Hirabuki, N., 2002. Evaluation of indexes of in vivo manganese status and the optimal intravenous dose for adult patients undergoing home parenteral nutrition. Am. J. Clin. Nutr. 75, 112–118. Takser, L., Mergler, D., Hellier, G., Sahuquillo, J., Huel, G., 2003. Manganese, monoamine metabolite levels at birth, and child psychomotor development. Neurotoxicology 24, 667–674. Tietz, N.W. (Ed.), 1998. Textbook of Clinical Chemistry, 3rd ed. WB Saunders, Philadelphia. Urushidate, S., Matsuzaka, M., Okubo, N., Iwasaki, H., Hasebe, T., Tsuya, R., et al., 2010. Association between concentration of trace elements in serum and bronchial asthmaamong Japanese general population. J. Trace Elem. Med. Biol. 24, 236–242. Wasserman, G.A., Liu, X., Parvez, F., Ahsan, H., Levy, D., Factor-Litvak, P., et al., 2006. Water manganese exposure and children's intellectual function in Araihazar, Bangladesh. Environ. Health Perspect. 114, 124–129. World Health Organization, 1981. Manganese. Environmental Health Criteria, 17. WHO, Geneva. Xu, B., Xu, Z.F., Deng, Y., 2010. Manganese exposure alters the expression of Nmethyl-D-aspartate receptor subunit mRNAs and proteins in rat striatum. J. Biochem. Mol. Toxicol. 24, 1–9. Yu, X., Cao, L., Yu, X., 2013. Elevated cord serum manganese level is associated with neonatal high ponderal index. Environ. Res. 121, 79–83.
238
X.-D. Yu et al. / Environmental Research 133 (2014) 232–238
Zhang, F., Xu, Z., Gao, J., Xu, B., Deng, Y., 2008. in vitro effect of manganese chloride exposure on energy metabolism and oxidative damage of mitochondria isolated from rat brain. Environ. Toxicol. Pharmacol. 26, 232–236. Zheng, W., Kim, H., Zhao, Q., 2000. Comparative toxicokinetics of manganese chloride and methylcyclopentadienyl manganese tricarbonyl (MMT) in Sprague-Dawley rats. Toxicol. Sci. 54, 295–301.
Zoni, S., Albini, E., Lucchini, R., 2007. Neuropsychological testing for the assessment of manganese neurotoxicity: a review and a proposal. Am. J. Ind. Med. 50, 812–830.
