Registration Dossier

Administrative data

Description of key information

Lead has been documented in observational human studies to produce toxicity in multiple organ systems and body function includine the haemotopoetic sytem, kidney function, reproductive function and the central nervous system.

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion
Dose descriptor:
NOAEL
Study duration:
chronic

Mode of Action Analysis / Human Relevance Framework

Additional information

Documentation of repeated dose toxicity from lead exposure is extensive and includes detailed evaluations of effects in humans. Research in animals serves to support causality in epidemiology studies and/or the elucidation of mechanisms of toxicity. The animal studies described here are thus superceded by the human epidemiological data describing health effects. The dose response for health effects in human can usually be defined with precision and indexed to internal measures of systemic exposure such as the concentration of lead in blood. Effects assessment indexed to external dose is thus not necessary. Although repeated dose toxicity can be demonstrated in animals, adverve effects have been more precisely characterised in observational human studies and multiple aspects of repeated dose toxicity are decribed in section 7.10.2 Epidemiological data. These human studies define effects in terms of blood lead levels - an indicator of systermic or internal exposure that represents the integrated total of exposure that results from oral, dermal and inahalation exposure routes. Reliance upon blood lead as an exposure indicator eliminates the need for route specific metrics of toxicity that are by their very nature inherently more imprecise and variable.

In many instances the toxic manifestations of lead are mediated by interaction of the lead ion with cellular proteins (e.g. metalloenzymes) in place of essential trace minerals such as calcium, zinc or iron. As such, the initial effects of lead upon the function of specific metabolic, cellular, or organ system processes will be detectable as increases or decreases in the activity of specific metalloenzymes. The biological and clinical significance of such interactions requires careful assessment of the extent of enzyme activity alteration and whether or not the alteration alters metabolic pathways to an extent that is of health significance. For example, inhibition of an enzyme present in excess may be evident as an effect of lead, but may not be considered as adverse if the affected enzyme is not involved in rate limiting steps of metabolism and the overall function of the metabolic pathway is not adversely impacted.

As the intensity of exposure increases, the magnitude of specific lead induced effects may increase and/or yet additional steps in specific biosynthetic pathways may be altered. An assessment must thus be made when the severity of impacts results in affects that are considered as adverse. Put another way, evaluation of the toxicological data base must attempt to discriminate between “effect levels” and “adverse effect” levels. This can, in some instances result in the identification of No Observable Effect Levels (NOELs), Lowest Observable Effect Levels (LOELs, No Observable Adverse Effect Levels (NOAEL’s) and Lowest Observable Adverse Effect Levels (LOAELs). In some instances the demarcation points between these different effect levels can be defined with both precision and with a high degree of certainty that the mechanistic basis of toxicity is sufficiently well understood.

NOAEL identification can become complicated if the effects under study are sufficiently subtle as to elude detection at the level of the individual. Effects of this nature may have no material impact upon the individual, but may have significance if large numbers of individuals are potentially exposed. This situation is applicable to multiple health endpoints and significantly complicates discrimination between NOEL’s and NOAEL’s. The following effects are generally considered to be the most important health endpoints defined in observational human studies.

Haematological Effects:

The effects of lead upon the haeme biosynthetic pathway have been extensively studied and provides one of the best illustrations of a graduated continuum of effects that can be defined with both mechanistic certainty and with precision for dose-effect relationships. Effects of lead can be detected at low levels of exposure but are enzymatic or biochemical changes not deemed to be adverse. As exposure intensity increases, the constellation of effects observed becomes increasingly diverse until impacts upon haeme synthesis results in an inhibition of haemoglobin production. Decreased haemoglobin production can be observed at blood lead levels above 40 µg/dL in children and above 50 µg/dL in adults. Impacts upon haemoglobin production sufficient to cause anaemia are associated with blood lead levels of 70 µg/dL or higher.

Clinical anaemia is an unambiguous adverse health effect.  Impacts upon the haematopoietic system that are clinical precursors to anaemia (diminished haemoglobin production) can be considered as adverse effects. Diminished haemoglobin production thus serves as a conservative indicator of when the impacts of lead upon haeme biosynthesis make the transition from nonadverse biochemical effects to adverse effects. NOAEL’s for haematopoietic system function of 40 µg/dL for children and 50 µg/dL for adults are thus suggested by this analysis.

Renal Effects

Numerous studies have evaluated the effects of lead upon renal function under environmental and occupational exposure conditions. A continuum of responses has been observed with enzymatic changes being reported at lower levels of exposure and clinical lead nephropathy at high levels of exposure. Clinical lead nephropathy exhibits slow onset after prolonged high level exposure to lead and is characterised by a progressive decrease in glomerular filtration rate and a subsequent rise in serum creatinine. These functional changes are associated with apparently irreversible degenerative changes in kidney (interstitial fibrosis) affecting the glomerulus. Early mortality resulting from these changes has been documented by several cohort mortality epidemiology studies of occupationally exposed individuals heavily exposed to lead prior to the promulgation of modern occupational exposure standards.

A small number of recent general population studies have observed a correlation between low levels of lead in blood and biomarkers of kidney function such as creatinine clearance. The significance of such correlations is difficult to reconcile with studies of occupationally exposed individuals that have not observed such effects at far higher levels of exposure. Risk of impaired renal function from environmental levels of lead exposure is thus not expected for adults or children.

Occupational studies indicate that individuals with blood lead levels maintained below 60 µg/dL have renal function (e.g. glomerular filtration rates) equal or superior to individuals without occupational exposure. As initially observed by Buchet et al. (1980), and subsequently confirmed multiple studies, maintenance of blood lead levels at or below 60 µg/dL appears to guard against the onset of lead nephropathy. The collective studies indicate a threshold for significant renal effects that is in excess of 60 µg/dL lead in blood and with a requirement for prolonged (five years or more) lead exposure. A NOAEL of 60 µg/dL, combined with five years or more of lead exposure, is thus adopted for adverse effects upon renal function in adults.

Impaired Vitamin D metabolism that may be mediated at the level of the kidney is of potential concern and could influence the growth and development of children. Blood lead levels below 25 μg/dL appear to be without affect, at least in children with adequate nutrition (Koo et al. 1991). The safety of blood lead levels above 25 μg/dL is unclear and, on a precautionary basis, should not be exceeded. Although other health effects manifest at lower levels of exposure in children, 25 µg/dL can be considered as a NOAEL relevant to renal effects in children.

 

Blood Pressure and Cardiovascular Effects

Reviews and meta-analyses of the current literature on the blood lead/blood pressure relationship indicate that there is at best a weak positive association between blood lead and blood pressure in general population and occupational studies with average blood lead levels below 45 µg/dL. IPCS (1995), in reflecting upon the conclusions of most other meta-analyses) concluded that the magnitude of this association for a twofold increase in blood lead (i.e., from 0.8 to 1.6 µmol/l or 16-32 µg/dL) is a mean 1.0 mm Hg increase in systolic blood pressure and a 0.7 mm Hg increase in diastolic blood pressure. Slightly smaller estimates of the effect of lead upon blood pressure have been calculated in more recent meta-analyses. The increase in blood pressure statistically associated with an increase in blood lead is small and would not result in a meaningful increase in the risk of the individual to cardiovascular disease. However, it can be hypothesised that a modest increase in blood pressure would increase the overall incidence of cardiovascular disease in a large population of individuals. This consideration of “societal risk” as opposed to “individual risk” thus merits careful evaluation.

Evidence for causality in the relationship between blood lead and pressure is primarily derived from animal studies but the doses used in such studies are much higher than typical human exposures and are thus of uncertain relevance. Evidence of cardiovascular disease has not been found in most observational epidemiology studies, leading many to suggest that the association is not causal but rather the effect of residual confounding in statistical analyses of blood lead –blood pressure relationships. In the small number of studies reporting excess cardiovascular disease risk, relationships between blood lead and blood pressure were either absent or not reported. Although it is possible to make quantitative estimates of the impact of blood lead upon blood pressure and subsequent population–based increases in cardiovascular disease risk, in the absence of a consistent relationship between blood lead, blood pressure, and cardiovascular disease the validity of such calculations is not supported by the observational data.  

Indeed, the more recently conducted studies have used more sophisticated blood pressure measurement tools and attempted more precise correction for the multiple confounders known to impact upon blood pressure. These most recent studies have generally failed to observe a relationship between blood lead and blood pressure.  A meta-analysis conducted for the purpose of this Assessment, restricted to the highest quality recent studies, further observed that the effect of lead upon systolic blood pressure declines from 1.0 to 0.38 mm Hg for a doubling of lead in blood. No statistically significant associations with diastolic pressure was found. Such a significant shift in effect size as a function of study quality indicates that residual confounding, as opposed to causality, is likely responsible for earlier correlations observed in studies of humans. Due to lack of a quantifiable estimate of effect or disease risk, this endpoint not appropriate for derivation of a NOAEL.

Recent studies suggest that lead body burden may play a role in hypertensive nephropathy. However, the association remains hypothetical and needs to be substantiated with a larger number of cases. Cardiovascular effects secondary to renal damage have long been suggested as a consequence of high levels of occupational exposure, but cohort mortality studies do not provide independent evidence of risk for cardiovascular disease or derivation of a NOAEL via this alternate mechanistic pathway.

Neurological Effects in Adults

Prospective and cross-sectional studies, and associated meta-analyses of neurological function and neuropsychological performance suggest that effects of lead in the areas of sensory motor slowing, coupled with difficulties in remembering recently acquired information are the most sensitive indicators of impact upon the nervous system of adults. The severity of the effect observed appears to increase as a function of the intensity and duration of exposure.

Peripheral nervous system effects (as reviewed by Araki et al., 2000) are first detected as subtle decreases in nerve conduction velocity at blood lead levels as low as 30 µg/dL. Effects observed in the blood lead range of 30 – 40 µg/dL are subtle and well within clinical norms – no known decrements in performance (e.g. changes in reaction time, loss of motor coordination or strength) are associated with these changes induced by lead exposure. Effects further appear to be freely reversible upon the cessation of exposure. Effect reversibility, combined with the absence of functional changes and a lack of long-term clinical sequelae indicates that peripheral nervous system effects in this exposure range should not be considered as adverse effects.

As the duration and intensity of lead exposure increases, impacts upon nerve conduction velocity become more severe and begin to manifest as a peripheral neuropathy characterised by muscle weakness in the upper limbs. The precise intensity and duration of lead exposure required to induce lead neuropathy is somewhat unclear in that most cases were observed prior to the 1920’s (Ehle, 1986). This older literature also suggests that elevated lead exposure for a duration of at least one year is required for neuropathy. The exposure intensity required to produce effects was likely significantly in excess of 70 µg/dL. For the purposes of this assessment, peripheral nervous system effects are not considered further since the central nervous system appears to be more sensitive to the effects of lead and thus provides the basis of a NOAEL for neurological effects in adults.

The effects of lead upon the central nervous system have been evaluated in numerous studies, most of which have been comprehensively evaluated in recent meta-analyses by Meyer-Baron and Seeber (2000) and Goodman et al. (2001). These evaluations of neurobehavioural testing have reached disparate conclusions. Whereas Goodman et al (2001) conclude that there is little consistent evidence of lead effects at blood lead levels below 70 µg/dL, Meyer-Baron and Seeber (2000) maintain that effects can be observed in cohorts with average blood lead levels that approach 40 µ/dL. A comparison of the two different analyses permits the following conclusions to be drawn.

1.     Meyer-Baron and Seeber (2000) included fewer study results in their analysis (12 studies were included as opposed to 22 included by Goodman et al., 2001), in part reflective of more rigorous inclusion criteria for study quality. Restriction of the analysis to higher quality studies likely contributed to suggestions of an effect at lower blood lead levels.

2.     Both analyses were conducted based upon cohort performance as opposed to the performance of individuals. Within each cohort individuals have concurrent and past lead exposures above and below the cohort average at the time of psychometric testing. As a generalisation, effects seen in a given study will be greater in those individuals with blood lead levels above the cohort average.

3.     For cohorts with blood lead averages < 70 µg/dL, the 95th percentile confidence interval for “lead effect size” includes “zero” for performance on most neurobehavioral tests. For those tests with effect sizes significantly different from zero, effects generally attenuate as lead exposure decreases and, within individual studies, the performance of lead exposed cohorts with average blood lead levels of 40µg/dL is similar to that of controls.

4.     Subtle effects upon cohort performance become evident as average cohort blood lead levels approach 50 µg/dL. Effects observed at this average exposure level may be reversible, have little clinical significance for the individual and are likely largely reflective of effects in individuals with blood lead levels higher than the average (e.g. 60 µg/dL). These effects can be interpreted as the initial manifestations of lead neurotoxicity in the central nervous system and thus can be considered adverse effects.

5.     Given the impacts of confounding, small effects sizes and other technical limitations, precise definition of a LOAEL for lead effects upon the adult central nervous system is difficult. Nor is there agreement as to what level of impact can be considered as adverse. However, adverse effects upon the individual likely require exposures in excess of 50 µg/dL.

6.     Given the relative absence of effects in cohorts with average blood lead levels of 40 µg/dL or less, despite the presence of individuals with higher blood lead levels, a blood lead level of 40 µg/dL is identified as a NOAEL for nervous system impacts in individual adults. Establishment of 40 µg/dL as the upper limit for individuals further means that occupational cohorts will have average blood lead levels well below 40 µg/dL and thus is the range where no effects can be detected.

Neurological Effects in Children The impacts of post-natal lead exposure upon the development of children have been studied and multiple endpoints evaluated using psychometric testing tools. Confounding factors that influence complex behavioural endpoints are imperfectly understood, and quantitative assessment of impacts at low blood lead levels is highly imprecise. Given the inherent difficulty of assessing effects upon behavioural endpoints, intelligence (IQ) is the endpoint for which the most robust measurement tools are available and for which confounder correction can be most comprehensively implemented. The underlying mechanisms for lead-induced IQ decrements is not known, nor have epidemiology studies identified a specific “syndrome” of neuropsychological deficits that would permit more precise quantitation of low-level lead effects independent of residual confounding and other sources of uncertainty in effects assessment.

The effects of lead upon IQ in children are thus summarised as follows:

1.     Meta-analyses of human observational epidemiology data show a statistical association between post-natal blood lead and IQ that is small and most likely between a one to three IQ point deficit for a change in mean blood lead level from 10 µg/dL to 20 µg/dL. Meta-analyses performed by Pocock et al. (1994) and IPCS (1995) serve as the basis for this estimate of effect size.

2.     IQ decrements on the order of 1 – 3 points are smaller than the standard error of measurement of IQ tests. As such they cannot be detected at the level of the individual and have no known functional significance for the individual. Any effects that occur in the blood lead range of 10 – 20 µg/dL are not regarded as adverse for the individual since they cannot be detected or measured. A NOAEL for the effects of lead upon the IQ of the individual can thus be regarded as 10 µg/dL.

3.     Based upon an estimated effect size of 1 – 3 IQ points, and an IQ test Standard Error of Measurement of 5 IQ points, effects hypothetically discernable at the level of the individual may not occur until blood lead levels exceed 20 µg/dL. Even though variability in the multiple variables that contribute to IQ would likely preclude detection of an effect of lead upon IQ in the individual, effects that fall within the precision range of existing psychometric measurement tools can be considered as adverse for the individual. Given that effects could, at least in theory, be detected in the individual at a blood lead level of 20 µg/dL, lead effects for the individual can provisionally be regarded to have a LOAEL of 20 µg/dL.

4.     Evidence is suggestive of an effect of blood lead upon IQ at blood lead levels less than 10 µg/dL but is difficult to interpret because of limitations in analytical and psychometric measurement techniques. In the absence of adequate data defining the nature and extent of effects at blood lead levels lower than 10 µg/dL, such effects are difficult to apply to risk assessment in a quantitative fashion.

5.     Any IQ decrements that might occur at blood lead levels below 10 µg/dL would not be detectable in the individual. Although such effects may lack significance for the individual, an impact upon large numbers of children may have societal significance. 

6.     Available data also do not permit the identification of a threshold for lead’s effects upon children. Observational data suggests that population effects may occur at blood lead levels as low as 5 µg/dL. This level of exposure also represents a point where current science is not capable of resolving further effects and any effects that might occur are secondary in magnitude relative to other factors that influence child development. A blood lead level of 5 µg/dL can thus be considered as an epistemic threshold that both recognizes the high degree of scientific uncertainty regarding effects at blood lead levels less than 10 µg/dL and at the same time establishes an exposure benchmark goal that is protective of public health

7.     Designation of 5 µg/dL as an epistemic threshold and a “societal blood lead target” also dramatically reduce the probability that individual children might exceed a blood lead level of 10 µg/dL. Maintenance of blood lead levels for the majority of the population below 10 µg/dL would require average population blood lead levels less than 5 µg/dL. For purposes of Risk Characterisation, 10 µg/dL post-natal lead in blood can be considered as a NOAEL for the individual child but that a general population blood lead average of 5 µg/dL is required to minimize the probability that individual blood lead levels will exceed 10 µg/dL.

Justification for classification or non-classification

Current classifications for lead compounds not otherwise specified in Annex 1 for the DSD is:

R33 (danger of cumulative effects)

 

Current classification under CLP is STOT RE2

Available data indicate that, under DSD, a more appropriate classification may be:

Xn; R48/20/22 (danger of serious damage to health by prolonged exposure through inhalation and if swallowed

Demonstration of toxicty in humans provides the basis for consideration of classification under CLP of STOT RE1.