Grantee Research Project Results
Final Report: Meconium Analysis - A Promising Tool to Detect Fetal Exposure to Environmental Toxins
EPA Grant Number: R829395Title: Meconium Analysis - A Promising Tool to Detect Fetal Exposure to Environmental Toxins
Investigators: Ostrea, Enrique M. , Bielawski, Dawn , Villanueva-Uy, Esterlita , Ager, Joel
Institution: Wayne State University
EPA Project Officer: Aja, Hayley
Project Period: April 1, 2002 through April 1, 2006
Project Amount: $726,411
RFA: Children's Vulnerability to Toxic Substances in the Environment (2001) RFA Text | Recipients Lists
Research Category: Children's Health , Human Health
Objective:
The overall goal of this research project was to develop the meconium test as a diagnostic tool to detect fetal exposure to heavy metals and common pesticides (heretofore, referred to as environmental toxins) during pregnancy. The specific objectives of this research project were to: (1) compare the prevalence and amount (concentration) of fetal exposure to environmental toxins, as detected by the analysis of meconium, cord (umbilical) blood, and neonatal hair and to determine the degree of agreement between these three methods; and (2) determine the relationship between the prevalence and amount (concentration) of maternal exposure to environmental toxins during pregnancy, as determined by maternal hair and blood analyses and the prevalence and amount of fetal exposure to environmental toxins, as determined by meconium, cord blood, and neonatal hair analyses.
Objective #1
Hypothesis 1a: Meconium analysis will detect a higher prevalence and amount of fetal exposure to environmental toxins as compared to cord blood or neonatal hair analyses
Hypothesis 1b: There will be a positive relationship between meconium analysis and cord blood or neonatal hair analyses on the prevalence and amount of fetal exposure to environmental toxins.
Objective #2
Hypothesis 2: The prevalence and amount of maternal exposure to environmental toxins, as determined by maternal hair and blood analyses, will correlate more to the prevalence and amount of fetal exposure to environmental toxins, as determined by meconium analysis, compared to cord blood and neonatal hair analyses.
Summary/Accomplishments (Outputs/Outcomes):
Subject Enrollment
Pregnant women were recruited prospectively at midgestation from an Outpatient Clinic in the Bulacan Provincial Hospital (BPH) in Malolos, an agricultural town in the province of Bulacan, Philippines. Informed consent was obtained from the subjects and maternal blood and hair (sample A) were obtained upon recruitment. Women were followed up until delivery. At birth, their infants were included in the study and maternal blood and hair samples (sample B) as well as infant hair, umbilical cord blood, and meconium were obtained. A total of 936 women were recruited initially but 25 women voluntarily dropped out before delivery. The remaining 911 women subsequently delivered 875 live births: 595 live births were delivered at the study site hospital (BPH) and all except for 2 infants were included in the study. A total of 280 infants were born outside of BPH and 200 (71.4%) were included successfully in the study. The 80 infants were excluded as we were not informed of their delivery. A total of 793 mother/infant dyads were eligible at birth; only 781, however, were enrolled in the study because sufficient meconium was obtained from these infants for analysis. The remaining 12 were excluded for insufficient meconium samples, one of our criteria for exclusion. For this report, we only used singleton births (N = 772) for the comparison and statistical analysis.
Maternal/Infant Characteristics
The antenatal and birth characteristics of the mothers in the study were as follows: mean maternal age was 25.8 years; mean gravidity and parity were 2.4 and 1.2, respectively. The common problems noted during pregnancy were a high rate of meconium stained fluid (12.2%) and antibiotic use (19.4%) caused by a high rate of urinary tract infections (25.4%). According to maternal report, direct exposure to alcohol and tobacco was not common. By meconium testing, there was minimal fetal exposure to opiates, cocaine, and marijuana. Methamphetamine was found in 9 percent of the infants. The majority of the infants were born at or near term. Mean gestational age was 38.5 weeks (54.4% male). Mean Apgar scores at both 1 and 5 minutes were above seven. Physical examination was normal in 97.2 percent of the infants. The most common postnatal problems included suspect sepsis, the need for oxygen at birth, jaundice, hyaline membrane disease, and transient tachypnea of the newborn. Neonatal mortality rate was 0.9 percent of the live births, mostly caused by prematurity.
Home and Environment
A survey of the family, home, and the child’s environment was conducted for each subject. Results indicated that 68.9 percent of the mothers and 70.4 percent of the fathers obtained at least a high school degree. The father’s mean age was 28.5 years and occupation was predominantly nonskilled labor (72.9%). The mean age of the mother was 25.8 years and 77 percent were homemakers. Average monthly household income was 5,599 pesos (≈ US$108). The average number of people and families per household was 5.3 and 1.6, respectively. The mean number of children under the family’s support was 1.1 (range of 0-11), and the average child age was 2.9 years. There were other dependents (children of relatives) receiving family support, with a mean number of 0.35 (range of 0-11) and mean age of 4.8 years. The predominant religion was Catholic (89.2%). About 57.9 percent of the mothers lived in their own homes, although 6.9 percent of these 57.9 percent were squatters living in makeshift homes.
Socioeconomic status (SES) was assessed using the Roberto Scale, which is a widely used socioeconomic scale in the Philippines. SES scores are based on home structure and appearance (Roberto, 2002). The standard test of SES commonly used in the United States (i.e., the Hollingshead Four Factor Index of Social Status) was not applicable to the Philippine population because of cultural differences. The Roberto Scale ranges from A (highest) to E (lowest). As very few households were classified as A or B, these two classes were combined (high). Class C was intermediate and classes D and E belonged to low SES. About 61 percent of the homes were in the class D and E category. A lead recycling plant was located near 7 percent of the homes. The cleanliness of the home and surroundings was rated mostly as fair (70.4%). The toilet was predominantly water seal (81.2%); water source was either piped in (50.7%) or from a well (42%); waste disposal was via sewer (25%) or canal (63.1%); and 58.7 percent had organized garbage collection. Telephone communication was predominantly by cellular phone (44.7%). Most homes had problems with flies (91.7%), roaches (90.4%), and mosquitoes (97.4%). Pesticide spray was used in 37.1 percent of the homes; the principal pesticide used was Baygon™ (contains propoxur and cyfluthrin). Spray of home pesticides was done by 37 percent of mothers. Reentry time after spraying of a room was less than or equal to 60 minutes in 76.6 percent of the households. The mosquito coil Katol™ (containing bioallethrin) was used in 52.5 percent of the households. Farm pesticides were used by 15.3 percent and only 4.3 percent used gloves to handle pesticides.
Pesticide Measurement
The methods used to analyze various matrices (maternal hair, maternal blood, cord blood, infant hair, and meconium) for pesticides and their metabolites using gas chromatography/mass spectrometry (GC/MS) have been published (Bielawski, et al., 2005; Corrion, et al., 2005; Ostrea, et al., 2006). The following pesticides were analyzed based on a preliminary survey of the study site for the common pesticides that were used in the homes and farms: cyfluthrin, propoxur, chlorpyrifos, cypermethrin, pretilachlor, bioallethrin, malathion, diazinon, and transfluthrin (Bielawski, et al., 2005). Lindane and DDT also were analyzed as our previous study in Manila, Philippines, showed significant exposure to these pesticides (Ostrea, et al., 2002). Samples also were analyzed for several pesticide metabolites, but very few samples were positive for them (Bielawski, et al., 2005; Corrion, et al., 2005; Ostrea, et al., 2006)
Hair. In hair, matrix-spiked calibration curves were linear for all parent pesticides and pesticide metabolites, with coefficients of linearity greater than 0.998. Optimum recovery rates using our 6-hour hexane extraction method ranged from 87-112 percent at a spiked concentration of 31.25 µg/g. The interassay and intra-assay coefficients of variability for the analysis of parent pesticides were below 11 percent. Limits of detection (LOD) by empirical approach ranged from 30.50 to 488.00 ng/g hair. Recovery rate of the metabolites by liquid-liquid extraction of the acid digest ranged from 87 percent-103 percent using a spiked concentration of 46.86 µg/g. Interassay and intra-assay coefficients of variability for the analysis of metabolites were less than 11 percent, and the LOD ranged from 0.18-5.88 µg/g (Ostrea, et al., 2006).
Blood. Similar effective separation of parent compounds and metabolites was shown for blood (Corrion, et al., 2005). The matrix-spiked calibration curves were linear for all parent pesticides and pesticide metabolites with coefficients of linearity greater than 0.994. Optimum recovery rates ranged from 84-142 percent for parent compounds and 54-122 percent for the metabolites. The interassay and intra-assay coefficients of variability for the analysis of parent pesticides and metabolites were below 10 percent. LOD by the empirical approach ranged from 0.03-0.49 µg/mL.
Meconium. Effective extraction and GC/MS analyses of parent pesticides and metabolites also were developed for meconium (Bielawski, et al., 2005). The matrix-spiked calibration curves were linear for all parent compounds and metabolites, with coefficients of linearity greater than 0.991. The LOD in matrix-spiked standards for the parent compounds ranged from 0.098 µg/g for propoxur to 1.56 µg/g for lindane. Interassay variability was less than 11.4 percent for all compounds. LODs for pesticide metabolites ranged from 0.312 µg/g for most compounds, to 4.15 µg/g for 3,5,6 trichloro-2-pyridinol. Parent pesticides recovery ranged from 82.4-109.3 percent; metabolites from 72.3-108.0 percent.
Prevalence and Concentration of Fetal Exposure to Environmental Pesticides as Determined by Analysis of Meconium, Infant Hair, and Cord Blood
Prevalence. Meconium showed the highest number and prevalence (percent positivity) of pesticides, compared to cord blood and infant hair: propoxur (21.2%) and the pyrethroids (2.7%) consisting of bioallethrin (0.3%), cyfluthrin (0.8%), and cypermethrin (1.6%). Other pesticides detected in meconium were malathion (0.3%), pretilachlor (1.8%), and DDT (0.5%). Cord blood was only positive for propoxur (1.6%), whereas infant hair was positive for propoxur (0.1%) and chlorpyrifos (0.1%). Diazinon, lindane, and transfluthrin were not found in any of the matrices. By the Cochran Q test, there was a significant difference in the prevalence of the following pesticides in meconium compared to cord blood or infant infant: propoxur (p < 0.001), pretilachlor (p < 0.001), DDT (p = 0.037), cyfluthrin (p = 0.012), and cypermethrin (p < 0.001).
Concentration. The concentrations (µg/mL) of the various pesticides in the positive samples are shown as median and interquartile range (Table 1) as a result of a skewed distribution. Meconium and hair concentrations are shown as µg/mL to allow uniformity of units for comparison among the different matrices. The concentrations of pesticides were high in meconium compared to the other matrices.
Substrate (µg/mL) |
|||
Toxin |
Meconium |
Cord Blood |
Infant |
Propoxur |
0.33 (0.24 – 1.51) |
0.77 (0.77 – 0.77) |
0 |
Malathion |
4.15 (2.92 – 5.38) |
0 |
0 |
Chlorpyrifos |
0 |
0 |
2.16 |
Bioallethrin |
1.20 (0.61 – 1.79) |
0 |
0 |
Pretilachlor |
0.52 (0.37 – 1.38) |
0 |
0 |
DDT |
1.75 (1.08 – 3.13) |
0 |
0 |
Cyfluthrin |
2.22 (1.21 – 5.14) |
0 |
0 |
Cypermethrin |
2.33 (1.91 – 2.54) |
0 |
0 |
Table 1. Concentration of Pesticides in Meconium, Cord Blood, and Infant Hair
Relationship. The relationship between meconium, cord blood, or infant hair analyses on the prevalence and amount of fetal exposure to environmental toxins could be tested only for propoxur because the other pesticides have been detected only in at most one substrate; thus, no statistical test is possible. For propoxur prevalence, infant hair is not included because there was no detection for that substrate when analysis for all three substrates (meconium, cord blood, and infant hair) was made. (Note: the only infant hair sample that was positive for propoxur had no corresponding cord blood sample for analysis.) The agreement between meconium and cord blood prevalence for propoxur was significant (Cohen’s kappa = 0.065, p < 0.011, N = 638). Similarly, there was significant agreement between meconium and cord blood concentrations for propoxur (Kendall’s coefficient of concordance W = 0.057, p < 0.001, N = 638).
Conclusion. As a biomarker of fetal exposure to propoxur, meconium is a better matrix to analyze compared to cord blood. Meconium analysis yielded a higher prevalence and amount of fetal exposure as compared to cord blood or neonatal hair, supporting Hypothesis 1a. There was a positive relationship between the prevalence and amount of pesticides detected in meconium and that of cord blood (Hypothesis 1b). Unfortunately, because of the extremely low detection in infant hair, we were not able to compare it with the other two fetal substrates. Overall, these data provide support of our hypotheses and we successfully accomplished Objective 1.
Prevalence and Concentration of Maternal Exposure to Environmental Pesticides by Analysis of Maternal Hair and Maternal Blood
Prevalence. The prevalence of pesticides in maternal hair and maternal blood taken at midgestation (sample A) and at birth (sample B) is shown in Table 2. The highest number and prevalence of pesticides were seen in maternal hair compared to maternal blood: propoxur (10.1%) in maternal hair A and 13.2 percent in maternal hair B compared to 0.4 percent in maternal blood A and 2.7 percent in maternal blood B, respectively. The prevalence of bioallethrin also was high in maternal hair A (9.2% vs. 0%) and B (8.2% vs. 0%). The other pesticides were of low prevalence in maternal hair or blood. Diazinon, lindane, transfluthrin, cyfluthrin, and cypermethrin were not detected and are not shown in the table. Based on the type of matrix (maternal hair vs. maternal blood), propoxur was found at a significantly higher frequency in maternal hair compared to blood, both at midgestation (10.1% vs. 0.4%, p < 0.001) and at delivery (13.2% vs. 2.7%, p < 0.001) by the McNemar test for dichotomous outcome and two related samples or the exact test. Bioallethrin also was detected at significantly higher frequency in maternal hair compared to blood, both at midgestation (9.2% vs. 0%, p < 0.001) and at birth (8.2% vs. 0%, p < 0.001).
Concentration. The concentration (median and interquartile ranges for positive cases only) of pesticides in maternal hair and blood are shown in Table 2. The relationship between the matrices for pesticide concentration was tested only for propoxur because the other pesticides have been detected only in at most one substrate. There was no significant difference in the concentration of propoxur in maternal hair at midgestation (A) as compared to delivery (0.31 vs. 0.24, µg/mL, p = 0.484, Wilcoxon sign test); there was a significant difference, however, between the concentration of propoxur in maternal hair versus maternal blood at midgestation (0.31 versus 0.67, p < 0.001, Wilcoxon sign test) and at birth (0.24 versus 0.77, p < 0.001, Wilcoxon sign test).
Concentration (µg/mL) |
||||
Toxicants |
Hair A |
Hair B |
Blood A |
Blood B |
Propoxur |
0.31 |
0.24 |
0.67 |
0.77 |
Malathion |
1.72 (1.61-2.17) |
0 |
0 |
0 |
Chlorpyrifos |
1.83 |
1.80 (1.77-1.82) |
0 |
0 |
Bioallethrin |
2.08 (0.84-2.68) |
0.87(0.41-1.81) |
0 |
0 |
Pretilachlor |
1.07 |
0 |
0 |
0 |
DDT |
0.41 (0.17-0.65) |
0.81 (0.48-2-21) |
0.56 |
0.53 |
Table 2. Concentration of Pesticides in Maternal Hair and Blood at Midgestation and Birth
These findings indicate that the prevalence and amount of maternal exposure to propoxur and bioallethrin during pregnancy is detected more sensitively by the analysis of maternal hair compared to maternal blood.
Relationship Between Maternal and Fetal Matrices
A comparison of the prevalence or concentration of pesticides in maternal hair, maternal blood, meconium, and cord blood could be done only for propoxur because it was the only pesticide found in all four matrices.
Prevalence. There was no relationship between propoxur in maternal hair A and propoxur in meconium or cord blood. There was a significant relationship, however, between propoxur in maternal hair B, with propoxur in meconium (odd’s ratio [OR] = 1.90, p = 0.006) but not for cord blood.
There was no relationship between propoxur in maternal blood A and propoxur in meconium or cord blood. There was a significant relationship, however, between propoxur in maternal blood B and propoxur in cord blood (OR = 19.34, p < 0.001).
For the other toxins, where tests are possible (for maternal hair A malathion, bioallethrin, pretilachlor, and DDT, and for hair B bioallethrin and DDT), there was no relationship between meconium and maternal hair. No tests were done for other toxins for maternal blood because there was no prevalence or the prevalence is too small.
Concentration. Because of the skewed distribution of the variables, the Spearman’s rho was used to assess the relationship.
There was a significant correlation between concentrations of propoxur in maternal hair B and meconium (rho = 0.08, p = 0.034, N = 764). There also was significant correlation between propoxur in cord blood and maternal blood B (rho = 0.24, p < 0.001, N = 647).
There was significant correlation between concentrations of propoxur in cord blood and meconium (rho = 0.17, p < 0.001, N = 648), between maternal hair A and maternal hair B (rho = 0.09, p < 0.009, N = 765), and between maternal blood A and maternal blood B (rho = 0.121, p < 0.001, N = 647).
Conclusions:
The prevalence of maternal exposure to propoxur and bioallethrin was detected more sensitively through the analysis of maternal hair compared to maternal blood. Similarly, the prevalence and amount of propoxur in maternal hair and blood (B) correlated more to the prevalence and amount of propoxur in meconium compared to cord blood. These findings support Hypothesis 2 (for Objective 2).
Comparison of Pesticide Prevalence in Meconium Versus Other Matrices
Comparison was made only for propoxur and bioallethrin where there was adequate prevalence. The overall test used was the Cochran Q and the post hoc was the McNemar test for a dichotomous outcome and two related samples. Meconium had a significantly higher prevalence (p < 0.001) for propoxur compared to the other matrices. For bioallethrin, maternal hair had a higher prevalence than meconium.
Comparison Among Matrices for Propoxur and Bioallethrin
For the detection of propoxur, the best matrix to analyze was meconium (p < 0.001), followed by maternal hair and maternal blood. Infant hair was the least sensitive matrix to analyze for propoxur. Maternal hair was the best matrix to analyze for exposure to bioallethrin compared to meconium (p < 0.001).
Comparison Among Matrices for Propoxur and Bioallethrin Concentrations
The overall test used was the Friedman test and the post hoc test was the Wilcoxon sign test. As in the prevalence results, meconium significantly (p < 0.001) showed the highest concentration of propoxur compared to cord blood, infant hair, maternal hair, and maternal blood. For bioallethrin, maternal hair significantly (p < 0.001) showed the highest concentration compared to meconium.
Pesticide Metabolites
Pesticide metabolites were detected rarely in any of the matrices examined. Of 489 meconium samples analyzed, 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE) was seen in only one sample and in cord blood, no metabolite was found. Because of a limited sample available, infant hair was tested only for the parent pesticides. In 666 maternal hair A samples, no metabolite was seen and in 499 maternal hair B samples, DDE was detected only in 1 sample. Similarly, in 861 maternal blood A samples, 3-phenoxybenzoic acid (3-PBA) was seen only in 1 sample and in 521 maternal blood B samples, 3-PBA was found in 2 samples and DDE in 1 sample. Thus, pesticide metabolites were detected rarely in the different matrices at the limits of sensitivity of the assay and, with permission from the project officer of the study, further analysis of metabolites in the remaining samples was discontinued.
Heavy Metals
Heavy metals (lead, cadmium, mercury, and arsenic) were analyzed in maternal hair and blood collected at midgestation and at delivery and in infant hair, cord blood, and meconium. The heavy metals were found mostly in maternal hair (25.4% for lead, 0.3% for cadmium, 24% for mercury, and 7.3% for arsenic). Maternal blood showed only mercury (1.0-1.4 %). No heavy metals were found in infant hair, cord blood, or meconium. In an area where a lead battery recycling plant was located (Marilao, Bulacan), 48 percent of the women were positive for lead compared to a mean of 27.7 percent in other towns.
Technical Effectiveness and Economic Feasibility of the Methods and Explanation of How the Research is of Benefit to the Environment and Human Health
The methods developed to extract and analyze quantitatively meconium, blood, and hair for pesticides by GC/MS were effective in that they showed high recovery, low limits of detection, and low variability in spiked samples. These methods have been published in peer-reviewed journals. The cost of the analysis is approximately $35 per sample for the 11 pesticides.
This research has allowed the comparison of the pesticide presence and concentrations across several matrices in a large, prospective cohort known to have a high rate of exposure. Meconium showed the highest rate of positivity out of the infant samples, and of the maternal matrices, hair was most often positive. This research is of benefit to human health as the development of multipesticide methods modeling actual exposure patterns is important. Furthermore, analysis of metabolites is of limited use, as they are transient in most tissues and direct exposure cannot be determined. Looking at the larger picture, the exposure data from this study now are being used to determine the effects of the pesticide exposure on physical and neurodevelopmental outcomes of the children.
There are several broader impacts of the research:
- The principal investigator (PI) was invited to participate in the International Workshop on Advances on the Use of Biomarkers in Children, sponsored by the World Health Organization in Buenos Aires, Argentina, on November 17, 2005, to present our research data on biomarkers of fetal exposure to environmental toxicants.
- The PI was invited to speak in a-1 hour radio program on April 12, 2006, at the National Broadcasting Station in the Philippines to present information arising from the research on the prevalence of exposure of the pregnant women and their infants to environmental pesticides in the Philippines, their potential harm, and preventive measures to minimize further exposure. The radio broadcast is heard throughout the Philippines.
- We distributed flyers to women in our study cohort on the safe use of pesticides, especially at home. We have no way of determining the impact of this educational program except to say that our subjects became more aware of the danger of exposure to pesticides and how to avoid unnecessary exposure to pesticides.
References:
Roberto N. The marketer’s guide to socioeconomic classification of consumers: Insights and challenges of target marketing in the Philippines using SEC indicators. Thousand Oaks, CA: Sage Publications, 2002.
Ostrea Jr. EM, Morales V, Ngoumgna E, Prescilla R, Tan E, Hernandez E, Ramirez GB, Cifra HL, Manlapaz, ML. Prevalence of fetal exposure to environmental toxins as determined by meconium analysis. Neurotoxicology 2002;23(3):329-339.
Journal Articles on this Report : 4 Displayed | Download in RIS Format
Other project views: | All 20 publications | 4 publications in selected types | All 4 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Bielawski D, Ostrea Jr. E, Posecion Jr. N, Corrion M, Seagraves J. Detection of several classes of pesticides and metabolites in meconium by gas chromatography/mass spectrometry. Chromatographia 2005;62(11-12):623-629. |
R829395 (2004) R829395 (Final) R826408 (Final) |
Exit |
|
Corrion ML, Ostrea Jr. EM, Bielawski DM, Posecion Jr. NS, Seagraves JJ. Detection of prenatal exposure to several classes of environmental toxicants and their metabolites by gas chromatography/mass spectrometry in maternal and umbilical cord blood. Journal of Chromatography B 2005;822(1-2):211-229. |
R829395 (2004) R829395 (Final) |
Exit Exit |
|
Ostrea Jr. EM, Bielawski DM, Posecion Jr. NC. Meconium analysis to detect fetal exposure to neurotoxicants. Archives of Disease in Childhood 2006;91(8):628-629. |
R829395 (Final) |
Exit |
|
Ostrea Jr. EM, Villanueva-Uy E, Bielawski DM, Posecion Jr. NC, Jin Y, Janisse JJ, Ager JW. Maternal hair—an appropriate matrix for detecting maternal exposure to pesticides during pregnancy. Environmental Research 2006;101(3):312-322. |
R829395 (Final) |
Exit Exit |
Supplemental Keywords:
children’s health, disease and cumulative effects, health risk assessment, physical processes, risk assessments, susceptibility, sensitive population, toxicology, genetic susceptibility, children, cord blood, detecting fecal matter, detection, diagnostic tool, environmental toxicant, exposure, exposure assessment, fetal exposure, fetus, heavy metals, human exposure, human health risk, infants, maternal exposure, meconium, meconium analysis, neonatal hair, pesticide exposure, pesticides, pregnancy, pregnant women,, RFA, Scientific Discipline, Health, PHYSICAL ASPECTS, Health Risk Assessment, Risk Assessments, Susceptibility/Sensitive Population/Genetic Susceptibility, Disease & Cumulative Effects, Physical Processes, Children's Health, genetic susceptability, Biology, pesticide exposure, diagnostic tool, fetal exposure, infants, detection, cord blood, pesticides, exposure, children, fetus, Human Health Risk Assessment, meconcium, human exposure, meconium analysis, environmental toxicant, detecting fecal matter, pregnant women, pregnancy, toxicants, exposure assessment, maternal exposure, heavy metals, neonatal hairProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.