Tobacco Documents Online

LEAD AND HUMAN HEALTH 1I following appropriate adjustments, with deficits in reaction time under varying intervals, which is an index of attentiveness, and with reaction behavior. The 1986 EPA criteria document concluded that these findings argue for probable effects of lead on attention and vigilance functions at PbB levels extending below 30 µg/dl, and possibly, down to as low as 15-20 p.g/dl. There is also evidence that low levels of lead may be associated with effects on some complex cognitive functions including learning, visual-perception skills, and IQ scores. The studies on children have attracted controversy because of d'afficulties associated with attributing subtle deficits in child development to lead exposure rather than to effects due to genetics, nutrition, medical history, access to education, and parental and social_influences, all of which interact in potentially complex ways to mold an individual. On the basis of five methodological criteria (adequate markers of lead exposure, sensitive measures of neurobehavioral function, appropriate subject selection, control of confounding covariates, and appropriate statistical analysis), the 1986 EPA criteria document identified a group of neurobehavioral studies that were conducted rigorously enough to warrant at least some consideration. The general indication from the better investigations is that PbB levels persistently elevated in the range of 50-70 µg/dI tend to be associated with about a 5-point reduction in IQ, even among asymptomatic children and after controlling for potentially confound- ing variables. However, considerable uncertainty has existed regarding lead's impact on IQ scores of children with PbB levels below 40 µg/dl. This uncertainty stems largely from the complex interaction between lead exposure over time, social factors, and intelligence scores, from the statistical and methodological limitations of cross- sectional studies to untangle these variables, and the range of interpretations that result from these studies. 1°he 1986 EPA criteria document concluded from the Needleman et ol. (1979) stpdy and subsequent reanalyses (Needleman, 1984) that, after controlling for coitfounding variables including pica, average IQ decrements of about 4 points and other neurobehavioral deficits appear to be associated with lead exposures of U.S. children resulting in dentine lead values that exceed 20-30 ppm and likely average PbB levels in the 30-SO µg/dl range. Needleman et al. (1982) calculated that a 4-point decrement in the mean IQ of a normal population distribution would be'associated with a threefold increase in the number of children with severe defcits (IQ < 80) along with a 5% reduction in the number of children who attain superior function (IQ > 125) (see Fig. 6). In order to avoid the normal array of confounding factors, Bellinger rt al. (1987) performed a longitudinal analysis of prenatal and postnatal lead exposure and early cognitive development in 249 children. In general, the infants were healthy products of unremarkable pregnancies, with few of the characteristics of infants at increased risk of developmental handicap. Eighty-seven percent of the families were white, and 92% were intact. The differences among the families with infants in the three cord-blood lead groups were slight and generally not in the direction expected on the basis of studies of the social correlates of childhood lead expo- sure. On the basis of lead levels in umbilical-cord blood, children were assigned

12 100 0 MORTON LIPPMANN 0 1E] 60 80 100 VERBAL 1.A. 0 120 140 FtG. 6. Cumulative frequency distribution of verbal IQ scores in subjects with low or hi$h levels of lead (Pdeedleman et aL, 1982). to one of three prenatal-exposure groups: low (<3 µg/dl), medium (6 to 7 µg/dl), or high (;~--10 µg/dl). Development was assessed semiannually, beginning at the age of 6 months, with use of the Mental Development Index of the Bayley Scales of Infant Development. Regression methods for longitudinal data were used to evaluate the association between infants' lead levels and their development scores after adjustment for potential confounders. At all ages, infants in the high-prenatal-exposure group scored lower than infants in the other two groups. The results are summarized in Fig. 7. Scores were not related to infants' postnatal blood lead levels. McMichael rt al. (1988) studied the effect of environmental exposure to lead on children's abilities at the age of 4 years in a cohort of 537 children born during 1979 to 1932 to women living in a community situated near a lead smelter at Port Pirie in Australia. Samples for measuring blood lead levels were obtained from the mothers antenatally, at delivery from the mothers and umbilical cords, and at the ages of 6, 15, and 24 months and then annually from the children. Concurrently, the mothers were interviewed about personal, family, medical, and environmental factors. Maternal intelligence, the home environment, and the children's mental development (as evaluated with use of the McCarthy Scales of Children's Abili- ties) were formally assessed. The mean blood lead concentration varied from 9.1 µg/dt in midpregnancy to a peak of 21.2 µg/dl at the age of 2 years. The blood lead concentration at each age, particularly at 2 and 3 years, and the integrated postnatal average concentration were inversely related to development at the age of 4. Results of multivariate

LEAD AND HUMAN HEALTH 120 104 13 6 12 18 24 AGE AT TESTING (monttts) Fae. 7. Prenatal exposure to lead, as measured by umbilical cord blood lead levels vs early mental development index. Low is c3 µg/dl, medium is 6.7 µp'dt, and high is a 10 ;Wdl (Bellinger et al.. 1987). analysis are illustrated in Fig. 8. Within the range of exposure studied, no thresh- old:dose for an effect of lead was evident. This cohort study indicates that a raised blood lead concentration in early childhood has an independent deleterious effect on mental development as eval- uated at the age of 4 years. This effect was not accounted for by the known and measurable influences of obstetrical, parental, family, and social environmental factors on mental development. The results of this analysis and those of an earlier analysis of the children at the age of 2 years suggest that increased exposure to lead results in a developmental deficit, not just developmental delay. Bhattacharya et al. (1988) found that abnormalities in children's abilities to maintain physical balance were significantly associated with PbB. Their postural sway on a balance increased by -2.8 cm2lµg/dl. These data suggest that low levels of PbB affect the peripheral nervous system as well as the central nervous system. A sample of their results are illustrated in Fig. 9. Schwartz and Otto (1987) used the large database available from the Second National Health and Nutrition Examination Survey (NHANES II), conducted between E976 and 1980 on population samples selected as being representative of the civilian, noninstitutionalized U.S. population. For a subsample of 4519 youths ages 4-19 years, there were data available on blood Pb, audiometry, and various indicators of neurological development, such as age at which a child first sat up, walked, and spoke. The presence of speech difficulties and hyperactivity was also

14 MORTON LIPPMANN 120 110 I , % Mccarthy scal.: m ! V 1 .. . _ ~ ...._~_. ~ 1M .a QUal1tI~wi W tatIVe n D.....~ e~~ ~ (GCi) 90 Mmory Mctor 50 40 0.5 10.3 1.0 1.5 2.0 µmot/1 20.6 30.9 41.2 µg / d Blood Lead Concentration „,,,,,,,,,,,,k,,ch,,,,,, n,U) Fte. 8. McCarthy scales of children's abilities (MSCA) scores at the age of 4 vs blood lead con- centration at 3 years of a=e (McMichael et ot.. 1988). examined to determine if they were significantly related to lead exposure. The probability of elevated hearing thresholds a4 500, 1000, 2000, and 4000 Hz in- creased significantly (P < 0.0001) with increasing PbB (Fig. 10). PbB levels were also significantly related to delays in the age at which children first sat up (Fig. 11), walked, and spoke and to the probability that a child was hyperactive. Lead was not related to the probability of a child having a previously diagnosed speech impairment. Table 3 shows the variables considered in the stepwise multiple regressions, while Table 4 shows the levels of significance of the associations between blood lead and the developmental variables. The results of this large population study are clearly consistent with, and strongly supportive of, the validity of the asso- ciations between blood lead and neurobehavioral effects in the smatler popula- tions reviewed earlier. In another examination of NHANES 11 data, Schwartz tt ol. (1986) incorpo- rated medical history, physical examination, anthropometric measurements, di- etary information (24-hr recall and food frequency), labbratory tests, and radio- graphs in linear regressions of adjusted data from 2695 children ages 7 years and younger. They reported that 91% of the variance in height, 72% of the variance in weight, and 58% of the variance in chest circumference (Fig. 12) were explained by six variables: age, race, sex, blood lead level, total calories or protein, and hcmatocrit or transferrin saturation level. In summary, there are a number of well-designed studies which indicate that very low levels of exposure to lead affect neurobehavioral function and develop- ment in young children. nese various effects appear to be consistent with the

LEAD AND HUMAN HEALTH ~ TEST CONDITION: EYES CLOSED, NO FOAM (p . 0.0025, r Z. 0.41) 0 • • •• 1.6 0 • M • ® ® (uea of swayl-1.49 - o.00002 fntirt w presw.t ~-01 1 (lop maz PDe i+ i a yr.l + o.e9lbp m= Pb8 n 2td yr.1 I 1 I I I I I I I I r I 1.4 2.0 2.4 2.8 3.2 3.6 Log Max. PbB Durinp S®cond Year of Life (µp / d) 4.0 15 Fto. 9. National logarithm of postural sway of children at 6 years of age vs natural log blood lead concentration during second year of tife after controlling for blood lead during first year of life (Bhattacharya et al., 1988). effects of lead on heme biosynthesis which have been postulated to lead to eryth- ropoaetic, neural, renal endocrine, and hepatic effects in the body, as summarized in Fig. 13 from the 1986 EPA criteria document. Effects ojl.ead in the Blood on Blood Pressure Tlae 1986 EPA criteria document on lead also provided a critical review of ; studies showing associations between blood lead concentrations less than 40 µg/dl and blood pressure. It reviewed the influence of a number of environmental and nutritional factors affecting blood pressure in experimental and epidemiological studies. Among environmental factors that have been associated with blood pres- sures are lead (Pb) and noise. Among dietary factors associated with blood pres- sure are calcium (Ca), zinc (Zn), phosphorus (P), alcohol consumption, and vita- mins A and C. The role of Pb as a pollutant stressor for elevated blood pressure could well be confounded by the well-established role of Ca as a suppressor of blood pressure. It is possible that persons with high Ca consumption have both decreased blood pressure and reduced blood Pb due to the competition of both Pb and Ca for the same binding sites. The influence of the other cofactors known to affect blood pressure further complicates the task of establishing the extent to which Pb con- stitutes a significant risk factor for elevated blood pressure. A consistent pattern of results emerges from recent investigations of the rela- tions between lower-level lead exposures and increases in blood pressure or hy- pertension. pertension. Khera et a!. (1980) reported higher blood lead levels in hypertensive

16 MORTON LIPPMANN ~+ 11 m v • a 4 * • • • Z 31 • t 1 t 1 1 0 10 20 30 40 50 BLOOD LEAD LEVEL. µQ / d Fto. 10. Relationship of 2 kHz pure tone hearing threshold (right eu) and blood levels in 4519 NHANES II subjects ages 4-19 years. Each point represents the mean hearing threshold of all persons in a 3 µ" blood lead range, except for the last point, which represents the mean hearing threshold and ysuan blood lead for all children with blood lead levels over 35 Wdl (Schwartz and Otto, 1987). patients and those with other cardiovascular diseases than for hospital control subjects. Kromhout and Couland (1984) and Kromhout et al. (1985) reported associations between hypertension and blood lead among elderly men in the Neth- erlands. Batuman et al. (1983) reported an association between hypertension and chelatable lead burdens in veterans. Moreau et al. (1982) reported significant associations (P < 0.001) between blood lead levels and a continuous measure of blood pressure among French policemen after controlling for important potential confounding variables such as age, body mass index, smoking, and drinking. Weiss et al. (1986) reported that after correction for previous systolic blood pres- sure, body mass index, age, and smoking, a high level of blood lead was a sig- nificant predictor of subsequent elevation of systolic pressure in policemen in Boston. Sharp et a!. (1988) examined relationships between blood lead concen- tration and blood pressure in San Francisco bus drivers. The analysis was limited to subjects not on treatment for hypertension (n = 288). The blood lead concen- tration varied from 2 to 15 µg/dl. While the findings were not statistically signif- icant, they did suggest effects of lead exposure at lower blood lead concentrations than those previously linked with increases in blood pressure. In a follow-up study, Sharp et al. (1989) examined the relationship between blood pressure and blood lead concentration in 51 bus drivers who were treated for hypertension. •'These drivers were a subset of a representative sample (n = 342) of the driver population (n r--2000) and were not selected for hypertension or lead exposure. Blood lead concentrations ranged from 2 to 24 µg/dl. There were 33 subjects

LEAD AND HUMAN HEALTH ~ < W < 5.5 .3 • ® • • • • • • • • 17 • ® • • • 1 1 1 t t t t 1 1 20 40 60 80 100 LEAD PERCENTILE RANK FtG. I 1. Relationship of developmental mileston: attainment and blood lead levels in NHANES 11 subjects. Plots of age at which a child first sat up (in months) vs PbB, after adjusting for other significant covariates. Each point represents the mean adjusted developmental index for 99 consecu- tive observations ordered by mean blood lead level. Regression lines were derived from individual data (Schwarsz and Otto, 1987). treated primarily with diuretics, and 18 subjects were treated with beta blockers. There was a significant mean difference of 12 mm Hg in diastolic BP over the range of observed Pb in blood (2.0 to 11.4 µg/dl) in subjects treated with beta blockers (see Fig. 14). Thus, beta blocker therapy may be less effective in reduc- TABLE 3 VARIAa1.ES C/SED /Ft STEPWISE REGRESSIOIiS Rac.e t.eM Ear discharge Cold in last 2 weeks Other ear condition Chronic ear discharge Income Dietary calcium Race Lead Size Dietary protein . A. Audiometric analyses Sex Current cold Ringing in ear{s) Earache Previous running ear Diagnosed hearing impairment Degree of urbanization Head of household education level Developmental milestone analyses Sex Income Head of household education level Total iron binding capacity Transferrin saturation Serum iron Hemoglobin Dietary calories Weight

18 MORTON LIPPMANN TABLE4 RESULTS OF DEVELOPMENTAL MILESTONE ANALYSES Effect Coefficient P value A. A41e (in years) at first word Intercept 1.25 Sex -0.027 0.0277 Lead rank 0.0024 0.0094 B. Age (in months) when first walked Intercept 10.E8 Rare -0.655 0.0006 Lead rank 0.0070 0.0020 C. Age (in months) when first tat up Intercept 5.68 Protein intake -0.0039 0.0361 Lead rank 0.0061 0.0239 D. Probability of being hyperactive (log;stic regression) Intercept -4.503 - Lead rank _ 0.0116 0.0130 ing diastolic pressure in individuals with elevated PbB, even at PbB levels asso- ciated with exposures below the current ambient standard and far below the current occupational standard. In a large population study, Pocock et al. (1984) evaluated the relationships between blood lead concentrations, hypertension, and renal function indicators in a clinical survey of 7735 middle-aged men from 24 British towns. The association ~ W 55'50 U Z W Q ~ 55.00 ~ ~ U Q U 5450 H ~ Wy ® 54.00 y 50 ~ 53 S 10 15 20 ' ADJUSTED t3LOOD LEAD (Juq / d) 25 FIG. 12. Adjusted chest eircumference and adjusted blood lead levels for children ages 7 years and younger in NHANES 11. Both chest circumference and blood lead level have been adjusted by regression for effects of age, sex, and all other variables significant at 0.05 level. Each point is mean chest circumference and mean blood lead level for approximately 95 consecutive observations, or- dered by blood lead levels. Regression line reflects slope of coefficient obtained from multiple regres- sion analyses of all 2671 points with no missinp data (Schwartz et al., 1986).

LEAD AND HUMAN HEALTH EXACERBATION OF HYPOXfC EFFECTS OF OTHER STRESS AGENTS EFFECTS ON MEUHOMS,AXOP6 AND SCHwANN CELLS hicur~l Effscts REDUCED HEMOPROTEINS (e.g. CYTCCHROSIES) F-++ I~PAJRED IMPAIRED CELLULAR lfYEUFlATqN AND ENERGET~CS kERVE CONDUCTION t4lPA1RED DEVELOPMENT OF tdERVWS SYSTEM R.w Erdoaiiw Eff~cta REDUCED 125 {OH): - ViTAMSN D DISTURBED ttdMUHO f~IPAIRED REGULATORY ROLE MINERAL T1SSW OF CALCIUM FObtEOSTASfS ~---~+ DISTUABED RCIE W TUMORIGENESIS CONTROL Hcpatk Etf~ REDUCED HEME FOR HEME REGULATED TRANSFORIdATiONS •-a IMPAIRED G1LCtUY ROIE AS SECOt~D MESSENGER CARDIOVASCULAR DYSFUNCTION AND OTHER HYPOXIC EFFECTS utPAJRED BONE AND TOOTH DEVELOPIIENT IMPAlRED CALCriJY RO(E Dl CT'CLIC MUCLEOTIDE METABOLISM IMPAIRED DETOxnCAT10!! OF XENO8JOTICS IMPAIRED 61ETA80USIi CF EHDOGENOtlS AGOAnSTS IMPAIRED DETOXIFICATIOt OF EHVIRONAtENTA1 TDX7kS IMPAIRED DETOXtFICAT10/1 OF DRUGS r'1 ALTERED METABOLISM OF TRYPTOPMJW ucPAfRED 1 HYDROXYLATBON CF CORTISOL ELEVATED BRAW LEVELS OF TRYPTOPF{AN, SEROTONiN AAD MAA I DISTURBED INDOLEAlAkE NEUROTRANSAUTTER FUNCTION 19 Fto. 13. Multiorgan impact of reductions of heme body pool by lead. Impairment of heme synthesis by lead results in disruption of a wide variety of important physiotogical processes in many organs and tissues. Particularly weTl documented are erythropoietic, neural, renal-endocrine, and hepatic effects ii,dicated above by solid arrows. Plausible further consequences of heme synthesis interference by k:ad are indicated by dashed arrows (EPA, 1956). DISTURBED CALCIUM METABOLISM

MORTON LIPPMANN . 6 s 0 0=14.3 tS.69ft" 10 - ADJUSTED Pa@.sg/dL (log scaii) rto.14. Plot of adjusted diastolic blood pressure vs tdjusted natural lo& of blood lead concentration in male bus drivets treated for hypertension with beta blockers. Adjusted for age, aYe=, race, body mass index, and frequencies of catfeine, akohol, and tobacco use. Diastolic blood pressure is the average of three measured diastolic blood pressures on each subject (Sharp et at., 1989). between systolic blood pressure and blood lead levels, though small in magnitude, was statistically significant (P < 0.01). Analyses of data for men categorized according to blood level concentrations indicated increases in blood pressure only at lower blood lead levels; no further significant increments in blood pressure were observed at higher blood lead levels. • • ~t t t I I I I I I t 5 15 25 35 45 BLOOD LEAD LEVEL (Np / d) 6'ia. 15. Adjusted diastolic blood pressure and adjusted blood lead levels for males ages 20 to 74 from NHANES Il. Both blood pressure and blood lead were adjusted by regression for the effects o[ age, age', body mass, and other significant variables. Each point represents the mean blood pressure and mean blood lead for 50 consecutive observations, sorted in increasing order of blood lead (Schwartz, t488).