Opinion on the results of the Risk Assessment of: ACRYLAMIDE (Human Health and the Environment) - CAS No. 79-06-1 - EINECS No. 201-173-7. Report version : October 2000 carried out in the framework of Council Regulation (EEC) 793/93 on the evaluation and control of the risks of existing substances1. Opinion expressed at the 22nd CSTEE plenary meeting, Brussels, 6/7 March 2001.
Terms of reference
In the context of
Regulation 793/93 (Existing
Substances regulation), and on
the basis of the examination of
the Risk Assessment Report the
CSTEE is invited to examine the
following issues:
1. Does the CSTEE agree
with the conclusions of the
Risk Assessment Report?
2. If the CSTEE
disagrees with such
conclusions, the CSTEE is
invited to elaborate on the
reasons for this divergence of
opinion
Introduction
Acrylamide is produced
at 80,000 - 100,000 tonnes per
year in the EU. Approximately
99.9 % of Acrylamide is used in
the production of
polyacrylamides, most of which
is used in wastewater treatment
and paper and pulp processing.
Other uses of polyacrylamides
include cosmetic additives such
as body lotion and shampoo and
polyacrylamide gel
electrophoresis. Acrylamide can
also be used in the formulation
of grouting agents.
GENERAL COMMENTS
The environmental risk
assessment has been performed
on the basis of a sufficient
database, both on effects and
on exposure. For most use
patterns, there are no reasons
of concern for environmental
risk but hot spots are likely
to occur in the aquatic
environment in relation to some
particular uses
(Acrylamide-based grouts in
construction applications).
Therefore, the CSTEE agrees
with conclusions ii) and iii)
of the report.
Nevertheless there is
the possibility that use of
grouts in pipeline and sewer
repairs and manhole sealing
operations may produce, in some
cases, emissions comparable
with the use of grouts in
construction applications, even
if the theoretical PEC
calculation indicates low
figures. Thus there is a need
for more information on
potential exposure due to this
use.
The conclusion iii) for
the use of acrylamide in based
grouts should be extended to
the terrestrial compartment and
secondary poisoning.
The human health risk
assessment is based on a
sufficient toxicological
database. The main effects of
Acrylamide in animal
experiments are neurotoxicity,
genotoxicity to both somatic
and germ cells, carcinogenicity
and reproductive toxicity.
There are several case studies
reporting neurotoxic effects
also in humans. The CSTEE
agrees with the overall
conclusion iii) for workers,
consumers and indirect exposure
via the environment.
In light of the overall
conclusion iii) for all three
exposure categories, it is
difficult to understand the
conclusion ii) for "Human
health effects" on page 17 of
the RAR.
There is a general
problem in estimating human
exposure, as Acrylamide is
readily taken up by all routes.
Dermal exposure may be
important in the case of
contaminated drinking water if
the water is also used for
bathing etc. Measurement of
haemoglobin adducts can give an
integrated measure of the
exposure.
The report is of good
quality, but a general comment
is that references should be
checked. Some are missing in
the references list. The
literature does not seem to be
fully updated, either.
SPECIFIC COMMENTS
Environment
Exposure assessment
According to use
patterns and to
physical-chemical properties
and environmental behaviour,
the aquatic environment is the
most important compartment in
relation to environmental
exposure to Acrylamide. A
surprisingly high vapour
pressure (2500 Pa) is given for
aqueous solutions (50%) of
Acrylamide as compared to that
over the pure substance (0.9
Pa) (page 23). This will,
however, not influence the
final conclusions.
Calculation of PEC at
local, regional and continental
level have been correctly
performed for the aquatic
(water and sediment) and
terrestrial (soil and air)
environments. However, in the
estimations of PEClocalwater,
the highest value (<82.4
µg/L, page 59) is not
considered, as this release is
to an estuary. Even if a
dilution factor of 100 is used
this emission would, however,
be one of the highest. Another
site (H) showing a release
concentration of 41 µg/L is
disregarded in the assessment
and the reason for this is not
clear. All measured data in
emissions from production sites
are more than 20 years old, and
may not represent the present
situation.
Available experimental
data are in reasonable
agreement with calculated PECs,
indicating a low or negligible
background level, as well as
relatively low figures (usually
in the range of a few m g/L) in
most cases at local level.
Nevertheless, there is
evidence of some hot spots with
concentrations many orders of
magnitude higher (up to
hundreds of mg/L). All these
cases are related to the use of
Acrylamide-based grouts during
construction of pipelines or
tunnels.
Effects assessment
There is enough evidence
to demonstrate that toxicity of
Acrylamide on aquatic organisms
(fish, crustaceans and algae)
is relatively low.
Some reason for concern
may derive from the possible
effects on reproductive
activity. No reproduction tests
are available on freshwater
fish and crustaceans.
Nevertheless, a reproduction
test on marine crustaceans
indicate a NOEC > 4.4 mg/L.
Taking into account the
amount of toxicological
information available, the
procedure adopted to calculate
the PNEC for the aquatic
environment can be considered
conservative enough to cover
some uncertainty related to
reproduction effects.
The information on soil
dwelling organisms is limited
to effects on terrestrial
plants, the validable assays
were conducted using hydroponic
systems. Additional studies
indicate effects on plant
exposed to contaminated soils.
The report concludes that
slight effects on plant growth
can be observed at 10 mg/kg
soil. No information on soil
micro-organisms and
invertebrates is available.
Under these
circumstances, the
extrapolation of the PNEC from
the EC50 obtained from
hydroponic plant exposures is
not acceptable. The CSTEE
proposes the derivation of the
PNEC for terrestrial organisms
using a combined approach:
1. Applying a factor of
1000 to the reported value of
10 mg/kg soil (assuming an
EC50>10 mg/kg soil)
2. Using the equilibrium
partitioning method to cover
the other taxonomic groups
3. The lowest estimation
must be used as PNEC
Risk characterisation
Under normal operating
conditions, for most use and
emission patterns, PEC/PNEC
ratios at local level are below
1 indicating negligible risk
for the environment.
Although the PNEC
terrestrial organisms must be
recalculated considering the
low PECsoil, estimations of
PEC/PNEC ratios below 1 are
expected.
The only exceptions are
the already mentioned hot spots
related to the use of
Acrylamide-based grouts during
construction of pipelines or
tunnels. In these cases, a
threshold of concern may be
largely exceeded, even if
limited in space and time.
This conclusion is
mostly based on incidents
following grout applications in
Scandinavia, showing effects on
aquatic and terrestrial
organisms exposed to
contaminated water. The levels
in downstream water exceeded
the PNEC and were several
orders of magnitude higher than
those expected for other uses.
The environmental conditions
favoured the movement of
Acrylamide to aquatic systems.
The contamination of soils
around the area cannot be
excluded for this use,
particularly when run-off and
draining are limited due to
environmental conditions. In
addition, poisoning of
terrestrial mammals using the
contaminated water as drinking
water was observed. Therefore,
conclusion iii) should cover
not only the aquatic
compartment but also the
terrestrial compartment and
secondary poisoning.
Human health
Exposure assessment
One of the occupational
exposures considered is during
the large-scale use of
Acrylamide based grouting
agents. Here the construction
of the Hallandsås tunnel in
Sweden is described. It could
have been mentioned that the
dermal exposure was judged to
be more important than the
inhalation exposure.
Measurement of haemoglobin
adducts in workers showed a
good correlation to the
estimated exposure. In the
estimated high exposure group
the mean haemoglobin adduct
level was 0.44 nmoles/g globin
compared to 0.24 nmoles/g in
the whole group of exposed
workers and 0.04 nmoles/g in a
control group.
As polyacrylamide is
used in cosmetics (in
concentrations up to 2%), the
CSTEE supports the recent
recommendation from the
Scientific Committee on
Cosmetic products and Non-Food
Products intended for consumers
(SCCNFP) on a maximum
Acrylamide level in
polyacrylamide. Their
recommended tolerable level for
non-rinse products is <0.1
ppm and for rinse-off products
<0.5 ppm, which is
considerably lower than the
common level to-day, < 0.1 %
of residual monomer in the
polymer.
Effects assessment
Sensitisation
Animal data provide
clear evidence of skin
sensitisation following
exposure to Acrylamide. There
is also some evidence of skin
sensitisation in humans
following skin contact.
Currently, there are no data
available regarding respiratory
sensitisation. In view of these
findings and the fact that
inhalation is an important
route of exposure and
Acrylamide shows reactivity for
proteins, the possibility of
inhalatory sensitisation should
be considered.
Repeated-dose toxicity
Peripheral neuropathy
has been demonstrated in
several animal species, and
human case reports are also
described. Under "Workplace
surveys" the Swedish Hallandsås
accident is described on page
154-156. This description is
not in conformity with the most
recent report from that study
(Lars Hagmar et al, manuscript
submitted to Scandinavian J
Work Environ Health). Blood
samples for the analysis of
haemoglobin adducts were taken
from more than 200 workers (but
were not analysed at the same
time). Thus, there was good
individual exposure
information. Clear-cut
dose-response associations were
seen between Hb-adduct levels
and prevalence of symptoms from
the peripheral nerve system.
For 23 workers there were
strong evidence of impairment
of peripheral nerve system
functions due to occupational
exposure to Acrylamide. All but
two had recovered 18 months
after exposure cessation.
Mutagenicity
There is a large body of
evidence clearly demonstrating
that Acrylamide is genotoxic in
vivo to both somatic and germ
cells. In the case of germ
cells, Acrylamide has been
demonstrated to induce
heritable mutations.
Carcinogenicity
Acrylamide is
carcinogenic in animals,
producing increased incidences
in a number of benign and
malign tumours identified in a
variety of organs. Acrylamide
was upgraded by the IARC in
1994 from Group 2B to Group 2A
(probably carcinogenic to
humans) taking into
consideration its genotoxic
properties.
Norway has classified
Acrylamide as a high-potency
carcinogen (K1), any
preparation containing greater
than 0.01% W/W Acrylamide would
require classification and
labelling as a carcinogen.
Considering that
Acrylamide is clearly a
genotoxic carcinogen, available
quantitative risk assessments
should have been cited in the
RAR.
Toxicity to reproduction
Impaired fertility, that
may have been associated with
effects on sperm count and
sperm motility parameters, have
been demonstrated in male rats.
Risk characterisation
Repeated-dose
toxicity
The identification of
the NOAEL is based on a 2-year
rat carcinogenicity study in
which histological peripheral
nerve lesions were seen at 2
mg/kg/d but not at 0.5 mg/kg/d.
However, in the comprehensive
90-day study by Burek et al.,
1980, slight changes in nerve
tissue (visualised only by
electron microscopy) were seen
at 1 mg/kg/d but not at 0.2
mg/kg/d. As the Burek et al.
study was more extensive with
respect to neurotoxicity as a
critical endpoint, the CSTEE
concludes that the alternative
NOAEL 0.2 mg/kg/d is warranted.
The CSTEE agrees with
the conclusion iii) with regard
to the risk for neurotoxic
effects in workers from the use
of Acrylamide-based grouts in
construction applications. This
conclusion should also apply to
pipeline and sewer repairs and
manhole operations. The very
low MOS of around 1 for workers
indicates a clear cause for
concern. According to the
section on risk
characterisation in the report,
the exposure estimate for small
scale use of grouts is likely
to be an overestimate, but on
the other hand there are case
studies reporting effects such
as skin peeling in workers who
perform such operations.
Concerning indirect
exposure via the environment,
the CSTEE agrees with the
conclusion iii) with regard to
the risk for neurotoxic effects
in connection with large-scale
grouting. For small scale
grouting activities, the
reported MOS of 4500 is
seemingly reassuring. However,
the case report by Igisu et al.
(1975) on neuropathy in a
family caused by local grouting
activities that contaminated
the drinking water, calls for
attention to potential risks
for indirect exposure due to
small scale use of grouts. It
should also be noted that only
drinking of the contaminated
water was considered when
calculating the MOS . Skin
contact with the water
(washing-up, bathing etc) is
also an important exposure
pathway.
Carcinogenicity
The CSTEE agrees with
the overall result iii) of the
risk assessment for workers,
consumers and indirect exposure
via the environment with regard
to genotoxic/carcinogenic
effects. The risk
characterisation conclusions
iii)a and iii)b are used
without explanation or
definition. Conclusion iii)a
seems to be used when the
cancer risk is considered to be
very small and iii)b when the
risk is a cause for concern.
However, there are no
quantitative cancer risk
assessments accompanying the
different conclusions.
Two carcinogenicity studies in rats describe tumours in various organs. Together with the clearly shown mutagenicity in vivo it is reasonable to assume that acrylamide can act via a genotoxic mechanism. Unit risks were calculated by EPA to be 1.3 x 10-3 (inhalation) and 1.3 x 10-4 per µg/l (drinking water), the slope factor is 4.5 per mg/kg/day (EPA 1993). The study of Johanson et al. (1986) with female rats was used. However, these calculations included mammary tumours in female rats, which contributed at large to the excess tumour incidences. The relevance of these tumours for man is doubtful. Many of the tumours in the carcinogenicity study of Johanson et al. (1986) developed in organs that are hormonally controlled (thyroid, uterus, mammary). It is unclear whether acrylamide acts in these organs exclusively via a genotoxic mechanism. Therefore, these calculated values may be conservative. Nevertheless to obtain a crude estimate of the cancer risk, the slope factor together with estimated exposure data (mg/kg/day) could be used to calculate excess cancer risks for the various exposure scenarios described in the risk assessment report. It has to be pointed out that exposure estimates are also based on worst case scenarios and therefore the calculated cancer risks are additionally conservative.
Summary of risk
characterisation for
workers
The conclusion of the
risk assessment report that "it
is not possible to reliably
identify a threshold level of
exposure below which there is
no increased risk and the
magnitude of the risk of cancer
at occupationally relevant
exposure levels is not clear"
is supported. Exposure has to
be reduced as far as possible.
Summary of risk
characterisation for
consumers
For consumer exposure
the excess cancer risk may be
calculated to be 4.5 x 1000 x
10-6 = 4.5 x 10-3. Given the
uncertainties of the risk
estimate as outlined above, the
carcinogenic risk for consumers
may be probably lower.
The EU has no guidelines
for what should be regarded as
an acceptable/non acceptable
cancer risk. However, the
calculated risk for consumers
is higher than what is used
within the framework of the WHO
Drinking Water Guidelines.
There, guideline values are set
at a level theoretically
corresponding to an excess
lifetime cancer risk of 1 x
10-5. Calculated risks for
small and large scale use of
acrylamide grouts also exceed
that level.
It is surprising that in
the section on risk
characterisation for consumers,
the reports states that it is
plausible that a threshold for
carcinogenicity may exist. The
reasoning behind this statement
is not discussed further and it
is not brought forward in the
sections on risk
characterisation for workers or
from indirect exposure via the
environment. It is not
supported by the CSTEE.
With regard to the
inherent toxic properties of
Acrylamide (neurotoxicity,
genotoxicity to both somatic
and germ cells, carcinogenicity
and reproductive toxicity), the
CSTEE concludes that the
exposure to humans should be
kept as low as possible.
----------------------------------------
1 Regulation 793/93
provides a systematic framework
for the evaluation of the risks
to human health and the
environment of those substances
if they are produced or
imported into the Community in
volumes above 10 tonnes per
year. The methods for carrying
out an in-depth Risk Assessment
at Community level are laid
down in Commission Regulation
(EC) 1488/94, which is
supported by a technical
guidance document.