Environ Sci Pollut Res DOI 10.1007/s11356-013-2465-4
RESEARCH ARTICLE
Manganese concentrations in maternal–infant blood and birth weight Limei Chen & Guodong Ding & Yu Gao & Pei Wang & Rong Shi & Hong Huang & Ying Tian
Received: 6 October 2013 / Accepted: 13 December 2013 # Springer-Verlag Berlin Heidelberg 2014
Abstract Manganese (Mn) is an essential mineral nutrient in mammals. The physiological role of Mn in animal models is well documented, but little is known about the adverse effects of Mn deficiency or overexposure in humans, including pregnancy outcomes such as birth weight. We examined the relationship of the maternal and cord blood Mn levels with birth weight in a cohort of 172 mother–infant pairs born in Shanghai, China. Non-linear spline and quadratic regression models were used to test the hypothesis of an inverted U-shaped association between the Mn levels and birth weight. The median (range) levels of Mn in the maternal and cord blood were 5.38 (2.34–30.37) μg/dL and 7.66 (2.57–34.23) μg/dL, respectively. An inverted U-shaped relationship was observed between maternal Mn and birth weight after adjusting for potential confounders. The birth weight increased with Mn levels up to 4.18 μg/dL, and a slight reduction in weight was observed at higher levels. The cord blood Mn levels were not found to be associated with birth weight. Both lower and higher Mn exposures are associated with lower birth weight, which may influence important developmental parameters;
Responsible editor: Philippe Garrigues Limei Chen and Guodong Ding contributed equally to this work. L. Chen : Y. Gao : R. Shi : Y. Tian (*) Department of Environmental Health, School of Public Health, Shanghai Jiao Tong University, Shanghai 200025, China e-mail: [email protected] L. Chen : G. Ding : H. Huang (*) : Y. Tian MOE and Shanghai Key Laboratory of Children’s Environmental Health, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200092, China e-mail: [email protected] P. Wang Shanghai Pudong Center for Disease Control and Prevention, Shanghai 200136, China
the association of higher Mn levels with lower weight was weak and imprecise. Keywords Trace element . Manganese . Maternal whole blood . Cord blood . Birth weight . China
Introduction Manganese (Mn) is a trace element that has been shown to be an essential nutrient in several animal species, particularly during gestation and early infancy (Tyrala 1992). Mn is required for normal amino acid, lipid, protein, and carbohydrate metabolism (Wood 2009). Because it is fairly widespread in the diet, Mn deficiency has not been recognized as a significant problem in human nutrition. The recommended adequate dietary Mn levels are approximately 2.2 mg/day, and the upper limits of Mn ingestion are 11 mg/day (Donohue et al. 2005). As with other essential minerals, there is a dichotomy to the response of the body to Mn. Adverse health effects can occur when the body stores of Mn are either too low (deficiency) or too high (toxicity) (Aschner et al. 2005). Mn deficiency has been recognized in experimental animals and is associated with impaired growth, skeletal defects, reduced reproductive function, abnormal glucose tolerance, and altered lipid and carbohydrate metabolism (FreelandGraves et al. 1994; Keen et al. 1999). The physiological role of Mn in the developing organism in animal models is well documented (Keen et al. 1999), but little is known about the adverse effects of Mn deficiency or overexposure in humans (Vigeh et al. 2008a, b). Under certain high-dose exposure conditions, Mn may accumulate within the human body, and it can induce adverse neurological, reproductive, and respiratory effects (Aschner et al. 2005). Some epidemiological data are available on the effects of in utero Mn exposure on pregnancy outcomes, including birth weight, which is a
Environ Sci Pollut Res
predictor of survival and developmental outcomes during childhood. The maternal blood Mn concentrations rise during pregnancy because of increased fetal demand and Mn accumulation through active transport across the placenta (Tholin et al. 1995). There have been some common characteristics in recent studies: lower maternal blood Mn levels were associated with poorer birth outcomes (e.g., increased risk of IUGR and lower birth weight) (Vigeh et al. 2008a, b; Zota et al. 2009). The association between higher concentrations of Mn and birth weight has not been found. The concern over developmental susceptibility to Mn toxicity has been framed largely by data suggesting a developmental period prior to weaning during which the newborn is unable to eliminate Mn (Kontur and Fechter 1985; Tran et al. 2002). Some additional light has been shed on the relationship between the maternal blood Mn concentrations and birth weight in a recent report. A study by Zota et al. (2009) suggested that a non-linear relationship between the maternal blood Mn and infant birth weight was found, and little or excessive Mn exposure can interfere with the birth weight (an inverted U-shaped relationship). In the present study, we aimed to further examine the specific association between the Mn concentration and birth weight. Using a cohort of 172 mother–infant pairs born in Shanghai, China, we tested the hypothesis that lower and higher Mn exposures would be associated with lower birth weights.
Methods Participants and recruitment Pregnant women were recruited from the Departments of Gynecology and Obstetrics of Ruijin Hospital and No. 7 People’s Hospital, two major obstetric hospitals in Shanghai, between September 2006 and April 2007. Eligible women with singleton pregnancy were ≥18 years, reported no preexisting or gestational diabetes, no chronic or pregnancyassociated hypertension, and no HIV infection or AIDS. A total of 199 women met the eligibility criteria. Of these, the following subjects were excluded: 26 women without maternal or cord blood samples and 1 woman with missing values for major confounders. A total of 172 women were enrolled in the analysis. Specially trained nurses administered a 20-min questionnaire to the mothers shortly after delivery. The questionnaire included the following: demographic and socioeconomic information (maternal age, height, prepregnancy weight, body mass index (BMI, the body weight divided by the height squared), education level, household income, and address) and maternal characteristics (cigarette smoking and alcohol use during pregnancy, employment). Other relevant information such as previous pregnancies, current pregnancy complications, prenatal care, and self-reported last menstrual
period (LMP) was obtained by interview and confirmed by medical records. Anthropometric measurements of the newborns were performed by the delivery room staff using standard anthropometric procedures. Written informed consent was obtained from each participating woman, and the study was approved by the Shanghai Jiao Tong University School of Medicine Institutional Review Board. Blood Mn measurements The maternal blood samples were obtained during the hospital admission for delivery (within 3 days before delivery); the cord blood samples were collected at the time of spontaneous delivery or cesarean section. The whole blood samples were aliquoted into 4 mL in low trace element EDTA blood collection tubes. The samples were immediately stored at −80 °C until analysis. The blood samples were weighed (1 g) and digested in 2 mL of concentrated HNO3 acid for 24 h and diluted to 10 mL with deionized water after the addition of 1 mL of 30 % H2O2. Mn was analyzed by an Agilent 7500CE inductively coupled plasma mass spectrometer (Agilent, Inc., USA). A 2-ng/mL solution of indium in 5 % HNO3 was used as the internal standard. We used CORTOX Lot Number HMB59311 (Kaulson Laboratories, Inc., USA) for the quality control. All the measurements were conducted in duplicate, and if the results were dubious, repeat measurements were made. The Mn blood sample concentrations in our study population were all above the limit of detection for this procedure (0.02 μg/dL). For instrument calibration throughout the measurements, at least 15 % of the analyses were controls and 10 % were blanks. Statistical analysis All the variables were examined by univariate and bivariate summary statistics and distributional plots. Linear and quadratic regression models, analysis of covariance, and Pearson correlations were performed using the SAS software v9.1 (SAS Institute, Cary, NC). Because of skewed distributions, the maternal and cord blood Mn were transformed to their decimal logarithms, and subsequent geometric means were calculated. We used a univariate analysis to identify all the potential confounders, including the mother (age, education, parity, household monthly income, passive smoking during pregnancy, prepregnant BMI, and hemoglobin concentration) and infant (gestational age, sex), to be considered for the relationship between the birth weights and the maternal and cord blood Mn concentrations. To isolate an optimal subset of these confounders, a forward stepwise procedure was used with a 40 % significance level of the F-statistics (Yazbeck et al. 2006). Multiple linear regression models were used to describe the relationships between the infant birth weights,
Environ Sci Pollut Res
manganese concentrations, and the main confounders. Nonlinear associations were initially examined with generalized additive models in R software(R Development Core Team 2004), using penalized spline terms for the manganese biomarkers, and their difference from a linear term was assessed with a likelihood ratio test. If this indicated a potentially nonlinear association between manganese exposure and birth weight, it was then modeled with a quadratic polynomial regression function to obtain coefficients for the effect estimates (Zota et al. 2009). The relationship between the maternal and cord blood Mn concentrations and the newborn birth weight was finally adjusted for maternal age, education level, pre-BMI, and hemoglobin. The research method for a peak level of Mn corresponding to a maximal activity of birth weight was similar to the method described by Zota et al. (2009). The relationship between the birth weight and the Mn concentration was analyzed by introducing the Mn and its square in the general model to estimate the peak level. The confidence interval of this peak level was calculated by the delta method (Armitage 1998).
Results The sociodemographic characteristics of the study sample are shown in Table 1. The data from 172 pregnant women were analyzed. The majority of the women (84.9 %) were primiparous, approximately four fifths (81.4 %) of the women were younger than 30 years old at the child’s birth, 77.9 % had a normal BMI (18.5–24 kg/m2) before pregnancy, 41.9 % had graduated from high school or college, and 71 % lived in households with a monthly income less than RMB (¥) 5,000 Yuan. The mean hemoglobin of the women was 109.2 g/dL (SD=19.6). Of the newborn infants, 51.2 % were male. The mean birth weight was 3,350.3 g (SD=488.9). Only 7 % of the newborns (n=12) weighed less than 2,500 g at birth, and 3.5 % (n=6) of the newborns were preterm. The mean concentrations of Mn in the maternal and cord blood were 6.6 μg/dL (SD=4.7) and 8.5 μg/dL (SD=4.6), respectively. The maternal and umbilical cord blood Mn levels were significantly correlated (Spearman correlation=0.175, P=0.022). The Mn concentrations observed in our study were higher than those in other published studies from North America and Europe (Table 2) (Takser et al. 2003, 2004; Zota et al. 2009). Figure 1 describes the relationship between the birth weight and the Mn concentrations in maternal and cord blood and a quadratic fit line. The relationship between the birth weight and the Mn exposure in maternal blood is shown in Fig. 1a. Unadjusted smoothed plots suggested a non-linear relationship between the maternal blood Mn and the birth weight (r2 = 0.025). The cord blood Mn was not associated with the birth weight in the quadratic analyses (Fig. 1b).
Table 1 Main characteristics of the study population (n=172) Variable Maternal characteristics Age (years)
RESEARCH ARTICLE
Manganese concentrations in maternal–infant blood and birth weight Limei Chen & Guodong Ding & Yu Gao & Pei Wang & Rong Shi & Hong Huang & Ying Tian
Received: 6 October 2013 / Accepted: 13 December 2013 # Springer-Verlag Berlin Heidelberg 2014
Abstract Manganese (Mn) is an essential mineral nutrient in mammals. The physiological role of Mn in animal models is well documented, but little is known about the adverse effects of Mn deficiency or overexposure in humans, including pregnancy outcomes such as birth weight. We examined the relationship of the maternal and cord blood Mn levels with birth weight in a cohort of 172 mother–infant pairs born in Shanghai, China. Non-linear spline and quadratic regression models were used to test the hypothesis of an inverted U-shaped association between the Mn levels and birth weight. The median (range) levels of Mn in the maternal and cord blood were 5.38 (2.34–30.37) μg/dL and 7.66 (2.57–34.23) μg/dL, respectively. An inverted U-shaped relationship was observed between maternal Mn and birth weight after adjusting for potential confounders. The birth weight increased with Mn levels up to 4.18 μg/dL, and a slight reduction in weight was observed at higher levels. The cord blood Mn levels were not found to be associated with birth weight. Both lower and higher Mn exposures are associated with lower birth weight, which may influence important developmental parameters;
Responsible editor: Philippe Garrigues Limei Chen and Guodong Ding contributed equally to this work. L. Chen : Y. Gao : R. Shi : Y. Tian (*) Department of Environmental Health, School of Public Health, Shanghai Jiao Tong University, Shanghai 200025, China e-mail: [email protected] L. Chen : G. Ding : H. Huang (*) : Y. Tian MOE and Shanghai Key Laboratory of Children’s Environmental Health, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200092, China e-mail: [email protected] P. Wang Shanghai Pudong Center for Disease Control and Prevention, Shanghai 200136, China
the association of higher Mn levels with lower weight was weak and imprecise. Keywords Trace element . Manganese . Maternal whole blood . Cord blood . Birth weight . China
Introduction Manganese (Mn) is a trace element that has been shown to be an essential nutrient in several animal species, particularly during gestation and early infancy (Tyrala 1992). Mn is required for normal amino acid, lipid, protein, and carbohydrate metabolism (Wood 2009). Because it is fairly widespread in the diet, Mn deficiency has not been recognized as a significant problem in human nutrition. The recommended adequate dietary Mn levels are approximately 2.2 mg/day, and the upper limits of Mn ingestion are 11 mg/day (Donohue et al. 2005). As with other essential minerals, there is a dichotomy to the response of the body to Mn. Adverse health effects can occur when the body stores of Mn are either too low (deficiency) or too high (toxicity) (Aschner et al. 2005). Mn deficiency has been recognized in experimental animals and is associated with impaired growth, skeletal defects, reduced reproductive function, abnormal glucose tolerance, and altered lipid and carbohydrate metabolism (FreelandGraves et al. 1994; Keen et al. 1999). The physiological role of Mn in the developing organism in animal models is well documented (Keen et al. 1999), but little is known about the adverse effects of Mn deficiency or overexposure in humans (Vigeh et al. 2008a, b). Under certain high-dose exposure conditions, Mn may accumulate within the human body, and it can induce adverse neurological, reproductive, and respiratory effects (Aschner et al. 2005). Some epidemiological data are available on the effects of in utero Mn exposure on pregnancy outcomes, including birth weight, which is a
Environ Sci Pollut Res
predictor of survival and developmental outcomes during childhood. The maternal blood Mn concentrations rise during pregnancy because of increased fetal demand and Mn accumulation through active transport across the placenta (Tholin et al. 1995). There have been some common characteristics in recent studies: lower maternal blood Mn levels were associated with poorer birth outcomes (e.g., increased risk of IUGR and lower birth weight) (Vigeh et al. 2008a, b; Zota et al. 2009). The association between higher concentrations of Mn and birth weight has not been found. The concern over developmental susceptibility to Mn toxicity has been framed largely by data suggesting a developmental period prior to weaning during which the newborn is unable to eliminate Mn (Kontur and Fechter 1985; Tran et al. 2002). Some additional light has been shed on the relationship between the maternal blood Mn concentrations and birth weight in a recent report. A study by Zota et al. (2009) suggested that a non-linear relationship between the maternal blood Mn and infant birth weight was found, and little or excessive Mn exposure can interfere with the birth weight (an inverted U-shaped relationship). In the present study, we aimed to further examine the specific association between the Mn concentration and birth weight. Using a cohort of 172 mother–infant pairs born in Shanghai, China, we tested the hypothesis that lower and higher Mn exposures would be associated with lower birth weights.
Methods Participants and recruitment Pregnant women were recruited from the Departments of Gynecology and Obstetrics of Ruijin Hospital and No. 7 People’s Hospital, two major obstetric hospitals in Shanghai, between September 2006 and April 2007. Eligible women with singleton pregnancy were ≥18 years, reported no preexisting or gestational diabetes, no chronic or pregnancyassociated hypertension, and no HIV infection or AIDS. A total of 199 women met the eligibility criteria. Of these, the following subjects were excluded: 26 women without maternal or cord blood samples and 1 woman with missing values for major confounders. A total of 172 women were enrolled in the analysis. Specially trained nurses administered a 20-min questionnaire to the mothers shortly after delivery. The questionnaire included the following: demographic and socioeconomic information (maternal age, height, prepregnancy weight, body mass index (BMI, the body weight divided by the height squared), education level, household income, and address) and maternal characteristics (cigarette smoking and alcohol use during pregnancy, employment). Other relevant information such as previous pregnancies, current pregnancy complications, prenatal care, and self-reported last menstrual
period (LMP) was obtained by interview and confirmed by medical records. Anthropometric measurements of the newborns were performed by the delivery room staff using standard anthropometric procedures. Written informed consent was obtained from each participating woman, and the study was approved by the Shanghai Jiao Tong University School of Medicine Institutional Review Board. Blood Mn measurements The maternal blood samples were obtained during the hospital admission for delivery (within 3 days before delivery); the cord blood samples were collected at the time of spontaneous delivery or cesarean section. The whole blood samples were aliquoted into 4 mL in low trace element EDTA blood collection tubes. The samples were immediately stored at −80 °C until analysis. The blood samples were weighed (1 g) and digested in 2 mL of concentrated HNO3 acid for 24 h and diluted to 10 mL with deionized water after the addition of 1 mL of 30 % H2O2. Mn was analyzed by an Agilent 7500CE inductively coupled plasma mass spectrometer (Agilent, Inc., USA). A 2-ng/mL solution of indium in 5 % HNO3 was used as the internal standard. We used CORTOX Lot Number HMB59311 (Kaulson Laboratories, Inc., USA) for the quality control. All the measurements were conducted in duplicate, and if the results were dubious, repeat measurements were made. The Mn blood sample concentrations in our study population were all above the limit of detection for this procedure (0.02 μg/dL). For instrument calibration throughout the measurements, at least 15 % of the analyses were controls and 10 % were blanks. Statistical analysis All the variables were examined by univariate and bivariate summary statistics and distributional plots. Linear and quadratic regression models, analysis of covariance, and Pearson correlations were performed using the SAS software v9.1 (SAS Institute, Cary, NC). Because of skewed distributions, the maternal and cord blood Mn were transformed to their decimal logarithms, and subsequent geometric means were calculated. We used a univariate analysis to identify all the potential confounders, including the mother (age, education, parity, household monthly income, passive smoking during pregnancy, prepregnant BMI, and hemoglobin concentration) and infant (gestational age, sex), to be considered for the relationship between the birth weights and the maternal and cord blood Mn concentrations. To isolate an optimal subset of these confounders, a forward stepwise procedure was used with a 40 % significance level of the F-statistics (Yazbeck et al. 2006). Multiple linear regression models were used to describe the relationships between the infant birth weights,
Environ Sci Pollut Res
manganese concentrations, and the main confounders. Nonlinear associations were initially examined with generalized additive models in R software(R Development Core Team 2004), using penalized spline terms for the manganese biomarkers, and their difference from a linear term was assessed with a likelihood ratio test. If this indicated a potentially nonlinear association between manganese exposure and birth weight, it was then modeled with a quadratic polynomial regression function to obtain coefficients for the effect estimates (Zota et al. 2009). The relationship between the maternal and cord blood Mn concentrations and the newborn birth weight was finally adjusted for maternal age, education level, pre-BMI, and hemoglobin. The research method for a peak level of Mn corresponding to a maximal activity of birth weight was similar to the method described by Zota et al. (2009). The relationship between the birth weight and the Mn concentration was analyzed by introducing the Mn and its square in the general model to estimate the peak level. The confidence interval of this peak level was calculated by the delta method (Armitage 1998).
Results The sociodemographic characteristics of the study sample are shown in Table 1. The data from 172 pregnant women were analyzed. The majority of the women (84.9 %) were primiparous, approximately four fifths (81.4 %) of the women were younger than 30 years old at the child’s birth, 77.9 % had a normal BMI (18.5–24 kg/m2) before pregnancy, 41.9 % had graduated from high school or college, and 71 % lived in households with a monthly income less than RMB (¥) 5,000 Yuan. The mean hemoglobin of the women was 109.2 g/dL (SD=19.6). Of the newborn infants, 51.2 % were male. The mean birth weight was 3,350.3 g (SD=488.9). Only 7 % of the newborns (n=12) weighed less than 2,500 g at birth, and 3.5 % (n=6) of the newborns were preterm. The mean concentrations of Mn in the maternal and cord blood were 6.6 μg/dL (SD=4.7) and 8.5 μg/dL (SD=4.6), respectively. The maternal and umbilical cord blood Mn levels were significantly correlated (Spearman correlation=0.175, P=0.022). The Mn concentrations observed in our study were higher than those in other published studies from North America and Europe (Table 2) (Takser et al. 2003, 2004; Zota et al. 2009). Figure 1 describes the relationship between the birth weight and the Mn concentrations in maternal and cord blood and a quadratic fit line. The relationship between the birth weight and the Mn exposure in maternal blood is shown in Fig. 1a. Unadjusted smoothed plots suggested a non-linear relationship between the maternal blood Mn and the birth weight (r2 = 0.025). The cord blood Mn was not associated with the birth weight in the quadratic analyses (Fig. 1b).
Table 1 Main characteristics of the study population (n=172) Variable Maternal characteristics Age (years)
