lead exposure in children living in a smelter ...

lournal of Tox~cologyand Envrronmental Health. Part A, 6'2.417-429, 2001 Copyr~ghrG'2001 Taylor & Franc15 1528-7394101 $1100 + 00

LEAD EXPOSURE IN CHILDREN LIVING IN A SMELTER COMMUNITY IN REGION LAGUNERA, MEXICO C . G. Garcia Vargas Facultad de Medicina, Universidad Juirez del Estado de Durango, G6mez Palacio, Durango, Mexico

M. Rubio ~ n d r a d e Facultad de Medicina, Universidad Juirez del Estado de Durango, and lnstituto Mexicano del Seguro Social, UMF-10, G6mez Palacio, Durango, Mexico

1. M. Del Razo Secci6n de Toxicologia Ambiental, Centro de lnvestigacion y de Estudios Avanzados del IPN, Mexico D.F., Mexico vr&

V. Borja Aburto Secci6n de Toxicologia Ambiental, Centro de Investigaci6n y de Estudios Avanzados del IPN, Mexico D.F., and Centro Nacional de Sal ud Am biental, Secretaria de Sal ud, Metepec, Estado de Mexico, Mexico

E. Vera Aguilar, M. E. Cebrian Secci6n de Toxicologia Ambiental, Centro de Investigaci6n y de Estudios Avanzados del IPN, Mexico D.F., Mexico /ndustrialgrowth has created the potential for environmental problems in Mexico, sinc.r attention to environmental controls and urban planning has lagged behind the p x e of industrialization. The aim of this cross-sectional study was to assess lead exposurr i l l children aged 6-9 yr attending 3 primary schools and living in the vicinity of the l a r p , smelter complex in ~Mexico.One of the schools i.s located 650 177 distant from a snreltcv. complex that includes a lead smelter (close school); the second is located 1750 m , . l ~ n y from the complex and at the side of a heavy traffic road (intermediate school) in Torretin, Coahuila. The third school is located in Gdmez Palacio, Durango, 8 100 171 away from the smelter complex and distant from heavy vehicular traffic or i n d i ~ s t r i ~ ~ l areas (remote school). Lead was measured in air, soil, dust, and well water. Lead ~ I J blood (F'bB) was determined in 394 children attending the above mentioned schoo/s.

Received 6 J u n e 2000; sent for revision 1 1 J u l y 2000; revision received and accepted 1 l September 2000. This work was supported b y the Sistema de lnvestigacion Regional "Francisco Villa" (SIVILLA; from ('ONACYT, Mexico, grant 90502 156. The assistance o f Aristides GGmez with statistical analvs~s and the secretarial assistance o f Kosalinda Flores and Guadalupe Favela are also acknowledged. Address correspondence t o G. G. Garcia Vargas, Facultad de Medicina, Universidad Juarez d d Estado d e Durango, LaSalle 1 y Sixto Ugalde S/N, Grimez Palacio, Durango 35050, Mkxico. k-rnLlil: [email protected]

Determinations were performed by atomic absorptiot~spectrometry. Diet, socioeconon~ic: status, hygienic habits, and other variables were assessed by questionnaire. iMedi,ll~ (range) PbE values were 7.8 pg/dl i3.54-29.67) in the remote school, 2 7.8 t ~ g / r l l (8.37-52.08) in the intermediate school and 27.6 pg/dl (7.37-.58..53) in chilrlr-rvi attending t l ~ eclose school. The percentage of children with PbE >1.5 p d d l was h.r'il:l,, 84.9%, and 92.7 % respectively. In this order, the geometric means (range) a / Ph ( . ( I / ] centrations in air were 2.5 pg/m3 (7.7-7..5), 5.8 pg/m' (4.3-8.5), and 6 . 7 pg/m' ( 7 . 0 74.9). The Ph concentrations in dust from playgrounds areas in the intermedintc ;i11r1 close school settings ranged from 74.57 to 4 7 h2..5 m d k g . Pb concentrations in drinking water were less than 5 p d L . Soil and dust ingestion and inl~alationappear to 11r (hrl main routes of exposure. Our resi~ltsindicate that environmental contamination 11a.s I Y sulted in an increased body h u r d e ~of~ Pb, suggesting that children living in the vic-inity of the smelter complex are at high risk for adverse effects of lead.

Lead i s one of the most ubiquitous toxic heavy metals and has heen detected in virtually all phases of the environment (air, soil, sediment, surface and ground water) and biological systems (ATSDR, 1998). Lead in the. environment occurs both naturally and from human activities. Industrial emissions, mainly smelting factories, serve as important sources of c o n t i mination, and lead exposure can be through direct inhalation of fumcs containing lead, or indirectly through soil and dust contaminated with lead (Roels et al., 1980). Exposure through foods and drinks prepared it1 lead-glazed ceramics i s particularly important in southern Mexico (Albcrt & Badillo, 1991; Hernsndez-Avila et al., 1991 ). Lead poisoning in childhood i s an important public-health problem (CDC, 1991). In 1991 the Centers for Disease Control (CDC) in Atlant,~, GA, defined a level of 10 pg/dl of Pb in blood (PbB) as an action or intervention level in children (CDC, 1991). Children are more likely than adulls to be exposed to contaminants present in soils and dust, and they are m o r ~ susceptible to neurological damage (National Research Council, 19931. Perhaps even more importantly, it is now suggested that there may bc no level of PbB that does not produce a toxic effect, particularly in the t l ~ veloping central nervous system; this concern applies to the fetus in utero and to women of childbearing age (Coyer, 1993). A more complete rc:view of lead toxicity can be found in the Lead Toxicological Profile of the ATSDK (1998). Industrial growth has created the potential for environmental pro11lems in Mexico, since attention to environmental controls and urban planning has lagged behind the pace of industrialization. Lack of municipal planning has meant that industrial sites such as ore smelters are at presenl located in residential areas (Diaz-Barriga et al., 1993). As a consequence, there are increasing risks of exposure to heavy metals through inhcllcltion and ingestion of contaminated soil and dust in Mexico. The cities of Torre6n and C6mez Palacio are located at the center ot northern Mexico, represent the main urban area of Region Lagunera with more than 800,000 inhabitants (Figure 1). This area has several conditions leading to lead contamination, such as heavy vehicular traffic and 111e

LEAD EXPOSURE IN CHILDREN FROM MEXICO

41 9

FIGURE 1. M a p of Region Lagunera. The region i s located at center of northern Mexico at t l i r boundary of the states of Coahuila and Durango. The state line between Durango (left) and Coahu~ln (right) is represented by Nazas River (shadowed area). Smelter complex is showed as X. The locat~ons of selected schools representing the levels of exposure groups are C (close group), I (intermed~atcl group), and R (remote group). The arrow shows the predominant wind direction. The numbers show the atmospheric monitor locations. See text for complete details.

largest nonferrous smelter complex in Latin America, the fourth largest in the world. The complex is located in Torrebn, State of Coahuila (Coah), ,ind in 1996 produced 166,000 metric tons of lead and 123,000 metric tons of zinc, and includes a lead smelter, a lead-silver refinery, and a zinc electrolytic refinery plant, surrounded by residential neighborhoods (Industrias Penoles, 1996). Several studies have shown evidence of lead contamination in Torreon. Albert and Badillo (1991) describe a former study (Albert & Garcia, 1977) that examined lead concentrations in the hair of children from five different regions of Mexico. This study included: (a) the north of Mexico City; (b) Puebla, State of Puebla (Pue); (c) a small town (Matamoros, Coah) that served as a control; (d) Torrebn, Coah, mentioned already; and (e) its twin city G6mez Palacio, State of Durango (Dgo). Lead concentrations in sdniples from the control zone were the lowest in the study (Matamoros, Coah, 4.2 pg/g), whereas the hair lead concentrations in hair found in samples from urban areas of Mexico City, Puebla, Pue, and G6mez Palacio, Dgo, were 12.1, 17.7, and 12.8 pglg, respectively. In contrast, in those

from Torrebn, Coah, the mean value was o f 55.1 l g / g w i t h a m a x i m u m of 2 2 0 pg/g. Hernandez-Serrano (1982) studied lead concentrations i n bloocl in medical undergraduate students residing in Torrebn, Coah, and reporte(l that 10% (9/90) o f students presented PbB values higher than 25 pg/dl. 111 1984, Calder6n-Salinas et al. (1 996) carried o u t a study on 9 8 children (aged 7-12 yr) living w i t h i n 1-km o f a smelter complex. The PbB averagcl concentration found in these participants was 17.3 pgldl. Although thcrcl are several studies showing evidence o f human lead exposure in urban areas o f the Region Lagunera, there i s a lack o f an integrated survey lo assess lead exposure i n children, including environmental and b o d y b u r den measures. The aim o f this w o r k was to assess the impact o f environmental contamination o n the lead body burden o f children living i n tho vicinity o f the largest smelter complex in Mexico.

METHODS General Design For area characterization, the relevant information about site history, potential lead sources, geographical features, and population distribution was considered, particularly background air lead concentrations, thc Icad refinery plant production, heavy traffic streets, and seasonal w i n d direction. Three elementary schools were selected for this study. The first was locate(l in Torreon, Coah, 650 m from the smelter complex, and students were considered representative o f children living within 1 k m o f the smelter ( t l o w school). The second school was also located in Torrebn, 1 7 5 0 m aw,iy from the smelter complex at the side o f a road with heavy traffic, and w a s considered representative o f children living in d o w n t o w n Torre6n (i1ltc.rmediate school). The third school was located i n the neighboring city of G 6 m e z Palacio, Dgo, 8 1 0 0 m away from t h e smelter and distant from roads w i t h heavy traffic or industrial areas (remote school) (Figure 1). Thcl study group consisted o f 3 9 4 elementary school students, aged 6-9 yr, attending f r o m first t o t h i r d grades. Informed written consent was o b tained f r o m children's parents before collecting b l o o d and urine san1plc5 and administering the exposure assessment questionnaire. The qucstionnaire information obtained from each child's parent included age, oc,c LIpational history o f their parents, use of lead-glazed pottery, and d i e t c t ~ r habits. For investigation of source exposure, environmental samples wercl collected i n adequate plastic containers previously washed w i t h nitric acid (1 0% w/v) and rinsed w i t h deionized water.

Environmental Measurements Historical data o n atmospheric lead o f total suspended particles (TSP) were obtained from the smelter company that collected the samples from three h i g h - v o l u m e air monitors located t o the northwest ( m o n i t o r 1 ) , north (monitor 2), and east side (monitor 3) o f the smelter and 2 Icm or

L E A D E X P O S U R E IN C H I L D R E N F R O M M E X I C O

42 1

less from the refinery (see Figure 1). The pertinent data during the period January 1992 to May 1996 are shown in Figure 2. Atmospheric Lead Air samples from each school's playgrounds wprc3 collected monthly during the period of this study (April-September 1907) in fiberglass filters (2.2 pm) exposed during 24 h, using high-volume air monitors (Gras Eby, GMW, model RAS 464, Village of Leaves, O H ) thcit represent a high-volume monitoring system for TSP and suspended parlicles of size < I 0 p m (PM-10). All filters remained dry a t room temperaturc until analysis. Fiberglass filters for measuring lead in air were prepared according to procedures NOM-033-STPS (1 993) and NOM-084-STPS (1 994) (Mexican regulations mainly based o n NlOSH methods); filters were wet digested with nitric acid. Lead content was determined by illductively coupled plasma emission spectrometry. These measuremenls were carried out in the metal laboratory of Met-Mex Penoles. Lead Determination in Dust and Soil Soil was obtained in 3 strata, 1 0 c m deep. Dust was obtained by brushing 50 g of material in school playground areas. Samples were stored at room temperature until analysis. Duplicate samples were prepared by drying, homogenizing, and sieving

0MONITOR 1

MONITOR 3

Q

MONITOR 2

Average per month, 1992-1996 F I G U R E 2. Historical lead concentrations i n air. Values of lead i n air are plotted and related lo monthly averages from 1992 to 1996. Air values come from the northwest (monitor I), north irnc~nilor 2), and east side (monitor 3 ) of the smelter. The line represents 1.5 pg/m', which is the maximum permitted level by Mexican laws (Diario Oficial de la Federacion, 1994). (Data were provided by i h t r Environmental Division of Industries Penoles, S.A. de C.V.)

422

G. G. C A R C ~ AVARGAS ET AL.

(~600 pm). Samples (1 g) were pretreated with 20 m l of 32.5% w/v nitric acid (Merck, Mexico), digested via microwave oven (CEM Corporation, 1990), and analyzed for lead content by flame atomic absorption spectroscopy (AAS, Perkin-Elmer, model 3100). Water Lead Measurements Water samples were collected from school drinking fountains and stored at 4°C until analysis. Samples were acidified (0.5%) with nitric acid (Merck, Mexico), and duplicate lead analysis were analyzed concurrently with the blood samples by furnace AAS (Apt IA, 1989). Quality control for environmental lead analysis included the analysis of Pb in the standard reference materials S R M 2704C, Buffalo River Sediment (NIST, Gaithersburg, M D ) containing 161 +- 17 pg Pb/g and S R M 1643c, trace elements in water (NIST, Gaithersburg, MD), containing 35.3 + 0.9 pg Pb/L, concurrently with environmental samples. The accuracy obtained ranged from 93 to 11 7%, and the coefficient of variation for 8 samples was between 3.7 and 9.5%.

Biological Measurements Blood Sampling The venipuncture area was washed with soap dntl water and cleaned with 70°/0 ethanol-water. Ten milliliters of venous blood were taken from each child using two 5-ml sealed ethylenediamine tetraacetic acid (EDTA) Vacutainer lead-free tubes (Becton Dickinson, IntJianapolis, IN). One sample was immediately used for hematological studies and second stored at 4°C until blood lead analyses. Urine Sampling First void urine samples were collected. Urinalysis was carried out immediately, and samples were stored at -20°C until further analysis. Lead in Blood Duplicate prepared blood samples were analyzed according Miller et al. (1 987), using a Zeeman graphite furnace AAS (PerkinElmer, model 5000). Quality control for lead in blood (PbB) included thc. analysis of lead in blood standard reference material ( S R M 955a) with three different concentrations (5.01, 13.53, and 30.63 mg/dl) concurrently with blood samples from participants. Recoveries ranged from 104 to 112% and the coefficient of variation from 3 to 10%. In addition, for external quality control, the ClNVESTAV (Centro de Investigation y de Estudios Avanzados) laboratory participates in two external quality assessment schemes for lead in blood: for trace elements and for occupational exposure. These programs were coordinated by the Robens Institute, U n versity of Surrey, Guildford, and the National Institute of Security and Labor Hygiene, Zaragoza, Spain, respectively. The global accuracy obtained in 1997 was 102.5 -t_ 4.4% (n = 36). The correlation coefficient between obtained values and real values was .998. Statistical Analysis Descriptive statistics analyses was done for all variables. Regression analyses were performed to assess the association between Pb exposure and the independent variables. All statistical an'ily-


423

ses were performed using procedures available in STATA version 6.0 (Stata Corporation, College Station, TX).

RESULTS Figure 2 shows historical data (1 992-1 996) on the lead content of dir total suspended particles; the lead concentration in monitor 1 (Torrccin city, downtown) was most of the time higher than 1.5 p g ~ b / r n 3 (U.S. National Ambient Air Quality Standard). This value i s also the maximum concentration allowed in Mexico (Diario Oficial de la Federacibn, 1994). During the period of this study (April-September 19977, the Pb content ill TSP and PM-10 showed a trend of progressive increase from the remot(. school to the nearest school with respect to the smelter complex (Table 1 ) . Lead concentrations in dust and soil samples obtained from plc-lygrounds of all schools sites were high in surface layers and decreased according with the depth of the strata sampled (Table 2), suggesting airborne deposition of lead on dust and soil. Thus, lead concentrations ill soil and dust from close school surpassed U.S. Environmental Protection Agency (EPA) cleanup goals (200-500 pglg soil) for Superfund sites under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) (U.S. EPA, 1997). Lead concentrations i n all water samples were lower than 5 pg/L, suggesting that its contribution to total Iclr~tl intake i s unlikely to be significant. In order to assess the impact of the environmental contamination, PbK was determined in 394 children attending the studied schools. Median PbK values were 7.8, 21.8, and 27.6 pgldl in children attending the remote school, the intermediate school, and the nearest school to the lead refillery, respectively (Figure 3). A more detailed data analyses i s shown in Table 3, where children attending each school were classified in the exposure and recomniended action categories proposed by the CDC (1 991). Most children attending the remote distance school (75.1 5%) had PbB values lower than 1 0 pgldl and were classified in Class I (not considered indicative of lead poisoning), TABLE 1. Lead Concentrations i n Air Total Suspended Particles (TSP) and Suspended Particles o f SIN < I 0 vn (I'M-1 0) Remote school (pg/m3)

Intermediate school (pplm3)

Close school (pg/m3)

Parameter

n

Geometric mean

Range

Geometric mean

Range

Geometric mean

Range

TSP I'M-I 0

5 5

2.5 0.8

1 .I-7.5 0.3-5.2

5.8 2.6

4.3-8.5 2.1-3.9

6.1 3.8

1.6- 1 4.0 2.3-5 .I\

Note. Monitors were placed in representative playgrounds of the studied schools (see Method\)

424

G. G. G A R C ~ AVARGAS ET AL.

TABLE 2. Lead Concentrations (pg Pb/g) i n Soil (cm Sampling Depth) and Surface Dust i n 131;lygrounds o f Schools According to Distance From the Smelter Complex in Region Lagunera, Mexico -

Kemote

-

-

Intermediate

Close

Sample

n

t ~ gPb/g

n

pg Pb/g

n

C]g PbIg

Soi 1 0-1 0 c m Soil 10-20 ern Soil 20-30 c m Dust

1 1 1 1

124.8 50.0 48.4 208.1

3 3 3 2

639.3 (209-1 101) 175.3 ( 5 8 4 2 9 ) 229.5 (216-348) 21 56.0 (1457-285.5)

3 3 2 2

2772.8 2374.7 1 196.9 3471.3

(648.3-3221 .O) (375.5-2772.li1 (207.5-2186.3 1 ( 2 7 8 0 4 1 62.5)

Note. Values are given at median and (range).

whereas the remaining had values above 10 pgldl and were classifiecl i t 1 Classes I1 and Ill (where an environmental intervention i s indicated a n d children should be rescreened). Most children attending the intermediate distance school had PbB 1 1 5 pg/dl and were classified in Class IIB (26.9%) and 111 (56.3%). Children in Class Ill should receive environmental evaluation and remediation, a medical evaluation, and may need a pharmacological treatment of lead poisoning. Most children attending the school nearest to the lead smelter were classified in Class 111 (67.5%), whereas 9.65% of children from this school were classified in Class IV. Children i l l

REMOTE

INTERMEDIATE

CLOSE

FIGURE 3. Lead i n b l o o d b y level of exposure. Box-plot graphs for lead i n b l o o d concentrations according to level of exposure as defined i n Methods. The outer bounds o f the boxes represent thv interquartile range; the median is represented by the midline. The whiskers represent the adjacol-11 values, w h i c h are not more than 1.5 times the interquartile range beyond the 75th percentile. l3al.i points outside the high adjacent values are plotted.


425

TABLE 3. Lead Blood Levels According to CDC Exposure Categories i n School Children From Kc.,"'ion Lagunera, Mexico Remote school CDC Class.'

PbB (pg Pb/dl)

Total

Intermediate school

Close school

Total

n

YO

n

(YO

n

YU

17

'%I

161

100.0

119

100.0

114

100.0

394

100.0

Class IV will need both medical and environmental interventions, inclucling chelation therapy. N o children with levels 270 pg Pb/dl (Class V, considered as a medical emergency) were found in any of the groups studied in this work. The analyses of PbB levels in children and distance from home to thc smelter complex are shown in Figure 4, which shows that PbB concenir,ltions decreased exponentially as function of distance (i= .668, p < .0001). The distance was plotted in logarithm form to obtain a better representation of the dot plot. In this figure are clearly two populations: those living within a 2500-m radius of the smelter (close and intermediate schools), and those living outside this distance (remote school). There was a child living about 10,000 m with a PbB value of 43.62 pg/dl who was studying in the close school. There were no statistical associations between PbB i n children and the use of glazed ceramics, which was mainly because the LISP of glazed ceramics is uncommon in the urban area of the Region Laguner,~ (only 10.7°/0 of questionnaires reported the use of glazed pottery to prepare and store food and drinks).

DISCUSSION The main findings of this study were that heavy metal contamination in soil and dust around the smelter site by far exceeded background levels and that a high proportion of children living in the neighborhood had PbK levels above 10 pgPb/dl. Lead concentrations in soil and dust were inversely related to the distance from the lead refinery and the depth of the strata sampled. These data support the fact that the smelter complex i s and has been the main source of lead emission in the city, since the smelter h,~s been working there for almost 100 yr. Ingestion appears to be the maill route of lead entry to the organism, since soil and dust were the most important contaminated environmental media having concentrations abovc.

G. G. G A R C ~ AVARGAS ET AL.

PbB=109.9-35.9* In distance +2.7 * (In distance)' & = 0.668 p