Formaldehyde Science Conference, Madrid 19-20 Apr 2012
Occupational Exposure Limits
Hermann M. Bolt
Leibniz-Institut fr Arbeitsforschung an der TU Dortmund Leibniz Research Centre for Working Environment and Human Factors
WHO Collaborating Centre for Occupational Health
The Positioning of Advisory Bodies
The general discourse and interplays leading to OEL/BLV regulations are similar at national and EU levels :
Political authorities
Legislation on occupational standards
Scientific experts
Social partners: employers/industry, trade unions
Chemical desasters with long-term influence on policy and legislation on chemicals in the EU
1974 Flixborough/UK 1976 Seveso/I 1984 Bhopal/IND 1986 Schweizerhalle/CH 1993 Frankfurt-Hchst/D 2000 Baia Mare/RO 2000 Enschede/NL 2001 Toulouse/F
Cyclohexane explosion: 28 deaths,89 wounded TCDD desaster Methylisocyanate desaster Environmental desaster; Rhine pollution Accidents: communication desaster Cyanide (50-100 t) in Tisa and Danube Explosion: 23 deaths, 947 wounded Ammonium nitrate explosion: 31 deaths
http://www.sust-chem.ethz.ch/teaching/courses/MaterialCuG/Unfaelle.pdf (modified)
OELs and BLVs: Development of the SCOEL Mandate
2011: New mandate
1980 1988 1989 1990 1995 1997 2004
2007
SCOEL strategy for carcinogens
Revision/extension of Biological Monitoring EUR 19253Methodology of derivation of OELs Commission Decision 95/320/EC setting up SCOEL Council Directive 90/394/EEC protection of workers to carcinogens Council Directive 89/391/EEC introduction of measures/safety and health of workers Council Directive 88/642/EEC amending the Directive of 1980 Council Directive 80/1107/EEC protection of workers from the risk related to exposure to chemical, physical and biological agents at work
Commission Decision (95/320/EC) of 12 July 1995, setting up a Scientific Committee for Occupational Exposure Limits to Chemical Agents (SCOEL)
Article 2 (1) ... The Committee shall in particular give advice on the setting of Occupational Exposure Limits (OELs) based on scientific data and, where appropriate, shall propose values which may include: the eight-hour time-weighted average (TWA), short-term limits/excursion limits (STEL), biological limit values. The OELs may be supplemented, as appropriate, by further notations. The Committee shall advise on any absorption of the substance in question via other routes (such as skin and/or mucous membranes) which is likely to occur. [Nota bene: no explicit reference to carcinogens!]
Review: European aspects of standard setting in occupational hygiene and medicine. Rev.Environ.Health 16:81-86
Methodology
Evaluations on a case by case basis Recommendations with clear justifications Critical effects and mechanisms of action to be described as detailed as possible NOAEL and/or LOAEL, extrapolation model used and quantitative considerations Systematic update of key scientific criteria (e.g. genotoxicity)
6
Official performance data in 2008
156 Recommendations in total 18 Carcinogens 96 IOELVs (+10) D2000/39/CE & D2006/15/CE (91/322/CE)
Binding Values:
Benzene VCM Wood dust Lead Asbestos
D 2004/37/CE D 98/24/CE D2003/18/CE
Triggering Discussions for new Criteria (2000-2007) (Threshold Effects for Carcinogens ?)
Induction of aneuploidy Topoisomerase II poisons Oxidative stress Inhibition of DNA synthesis Steep dose-effect curve, cytotoxicity involved Endogenous carcinogens, within limits of homeostasis Clastogens (being discussed)
Kirsch-Volders et. al: Mutation Res. 464:3-11, 2000 Madle et. al: Mutation Res. 464:117-121, 2000 Pratt & Baron: Toxicol. Lett. 140/141: 53-62, 2003
The dose-response relationship for a number of such agents is generally accepted to show a threshold, however, the degree of acceptance of the threshold effect differs in different EU regulatory systems.
Dose-Effect Relations in the Low Dose Range and Risk Evaluation
(Concept adopted by SCOEL - see Archives of Toxicology 82: 61-64, 2008)
Chemical carcinogen, causing tumours in humans and/or experimental animals Genotoxic
DNA reactive, causing mutations
Non-genotoxic
Genotoxicity only on chromosome level (e.g. spindle, topoisomerase)
Clearly DNA-reactive & initiating A: No threshold,
Borderline cases
Weak genotoxin, secondary mechanisms important
B: Situation not clear C: Practical/apparent D: Perfect/statistical
threshold likely threshold likely
LNT model to apply LNT as default
Numerical risk assessment,
risk management procedures
NOAEL
health-based exposure limits
Summary of the SCOEL Strategy for Carcinogens
The scientific development allows to identify carcinogens with a threshold-type mode of action. For these compounds health based OELs (and BLVs, where appropriate) can be derived. Such a mechanism-based assignment is independent of the formal classification of carcinogens (i.e., former EU categories 1, 2 or 3, equivalent to GHS categories 1A, 1B, 2)! When derivation of a health-based OEL/BLV is not possible, SCOEL assesses the quantitative cancer risk, whenever data are sufficient. When data are not sufficient for a risk assessment, SCOEL gives recommendations on possible strategies for risk minimisation, if possible.
Results of SCOEL Discussions (Examples)
A
No threshold, LNT (Linear Non-Threshold) model to apply: vinyl chloride / vinyl bromide (risk assessment) MDA dimethyl / diethyl sulfate 1,3-butadiene (risk assessment) LNT as default assumption: acrylonitrile benzene (provisional assignment) arsenic naphthalene hexavalent chromium o-anisidine 2,6-dimethylaniline (insuff. Data) Practical/apparent threshold: vinyl acetate nitrobenzene pyridine lead (provisional OEL); lead chromate TRI DCM glyceryl trinitrate Perfect/statistical threshold: carbon tetrachloride chloroform
Distinction between B and C is most important !
SCOEL: Formaldehyde - B oder C ?
Major points of general discussion
Classical case since the 1980s of nasal tumours in rats Sublinear dose-response curve (accepted since the 1980s) Cytotoxicity as relevant/necessary influencing factor IARC (2005): Sufficient evidence of human nasopharyngeal carcinomas (local effect in humans)
Discussions by SCOEL (2005-2007):
A: Cell profiferation/irritation necessary for tumour formation B: No straightforward evidence for systemic effects
Group C: Carcinogen with practical threshold
OEL derivation by SCOEL (I)
Local carcinogenicity requires cell proliferation. Irritancy on the upper respiratory tract is therefore a precondition that must be avoided. For this effect, the database is insufficient to establish an OEL. The database is much better for eye irritation. Avoidance of eye irritation also avoids irritancy on the upper respiratory tract and provides an additional safety margin.
OEL derivation by SCOEL (II)
Evaluation by Paustenbach et al. (1997): at 0.3 ppm practically all workers protected against eye irritation DECOS/Nordic (2003): 0.24 ppm as LOAEL
(difference to Paustenbach: interpretation of 2 studies from Scandinavia)
Lang et al. (2008): NOAEL for eye irritation at 0.5 ppm, or 0.3 ppm with peaks of 0.6 ppm
OEL derivation by SCOEL (III)
Proposal of an OEL of 0.2 ppm (TWA) and 0.4 ppm (STEL), because:
Consideration of particularly sensitive persons Safety margin to the onset of irritation-induced cytotoxicity / cell proliferation necessary
Arguments by DFG/MAK (2006)
Proposal of an OEL of 0.3 ppm (TWA) because: DPX and cell proliferation in nasal epithelia experimentally increased only at > 2 ppm Consideration of CIIT (1999) risk assessment: at 0.3 ppm -> risk for non-smokers 1.3 x 10-8, for smokers 3.8 x 10-7 (as secondary argument) Evaluation of Paustenbach et al (1997): practical NOAEL for irritation (workers) of 0.3 ppm
OEL Recommendations for Formaldehyde TWA
(ppm) ACGIH (USA, 2006) DECOS (NL)+Nordic (2003) 0.12 DFG/MAK (D, 2006) 0.3 SCOEL (EU, 2006) 0.2
(Health-based OELs recommended)
STEL
0.3 0.42 0.6 0.4
It may be concluded:
There has been significant progress in research on modes of carcinogenic action. The recognition of genotoxic and carcinogenic thresholds will allow the assignment of health-based limit values for an increasing number of relevant carcinogens.
Further examples: Recent SCOEL recommendations on inorganic compounds
A (no threshold): Cr(VI) - numerical risk assessment B (situation not clear): Be no OEL recommended C (practical threshold): Crystalline silica, resp. dust: OEL = 50 g/m3 Cd: OEL = 4 g/m3; BLV = 2 g/g creatinine Ni: OEL = 10 g/m3 inh., 5 g/m3 resp. dust; BLV = 3 g/L D (perfect threshold): RCF; OEL = 0.3 fibres/mL
Example: Silica, crystalline, respirable dust
Minimizing silicosis will minimize cancer risk
Sigmoidal doseresponse for silicosis
0.05 mg/m3 will reduce ILO 1/1 to less than 5%
OEL should be below 0.05 mg/m3 respirable dust
[Carcinogen with threshold, category C]
Example: Threshold Arguments for Cd
Experimentally: tumours at 12.5 g/m3 (inhal.) Epidemiological risk assessment difficult (coexposures!) Genotoxixity threshold in workers: 25 g/m3 for 40 years exp.
Indirect mechanisms of genotoxicity likely
No cancer excess in workers at exposures without lung/renal toxicity
Recommendation: Carcinogenicity group C (practical threshold) OEL: 4 g Cd/m3 BLV: 2 g Cd/g creatinine
Most recent example: Nickel and inorganic Ni compounds (I) SCOEL/SUM 85 [2011]
Basic features of Ni
Cancer risk in lung and nasal cavity; inflammatory responses/fibrosis in the lung Increased chromosomal aberrations in humans at exposure levels >0.5 mg/m3
At low concentrations Ni-ions do not directly interact with DNA Indirect genotoxic effects by interference with DNA repair and DNA methylation, leading to clastogenicity and genomic instability Indirect genotoxicity of Ni2+
Example: Ni and Ni compounds (II)
Metallic Ni: no positive evidence of carcinogenicity, but clear inflammation response in rats at 0.1 mg/m3 NOAEL for Ni-sulfate; local inflammation: 0.03 mg/m3 Carcinogenicity group C (practical threshold) Proposed OELs: 0.01 mg Ni/m3 (inhalable fraction) 0.005 mg/m3 (respirable fraction) BLV: 3 g Ni/l urine
General conclusions:
There has been consistent progress in research on modes of carcinogenic action. Secondary genotoxicity is receiving more attention! The recognition of genotoxic and carcinogenic thresholds will allow the assignment of health-based limit values for an increasing number of relevant carcinogens.
Discussions on formaldehyde have been a key for the treatment of other chemicals!
