Criteria for skin notation in an international perspective
Jesper B Nielsen PhD1, Philippe Grandjean MD, DMSc1,2
1
Institute of Public Health,
University of Southern Denmark,
DK-5000 Odense C, Denmark
2
Department of Environmental Health,
Harvard School of Public Health,
665 Huntington Avenue Boston, MA 02115, USA
Running head: Criteria for skin notation
Corresponding author:
Jesper B Nielsen
Institute of Public Health
University of Southern Denmark,
DK-5000 Odense C, Denmark
Fax: +45 65 91 14 58
e-mail: jbnielsen@health.sdu.dk
1
Abstract
Background: The skin notation was originally introduced as a qualitative indicator of a
hazard related to dermal absorption at work. However, risk is not a qualitative term, but
needs a quantitative measure to be useful for prevention. For this reason, some countries
have developed alternative criteria for assigning skin notations to industrial chemicals.
Methods: The present study compares the use of skin notations on lists of exposure limits
for industrial chemicals in six countries.
Results: Up to one-third of industrial chemicals listed now have a skin notation. The
criteria for assigning skin notation differ but cannot explain the substantial discrepancies
between countries that otherwise have very comparable occupational exposure limits.
Conclusions: The increasing number of chemicals with a skin notation requires a new
approach to differentiate between degrees of risk. The first step will be to address the
problems related to criteria for assigning skin notations, and the effect of penetration
enhancers and mixtures.
Keywords: skin notation, dermal penetration, occupational exposure limits, penetration
enhancers
2
Introduction
As pulmonary exposures to workplace chemicals have decreased, the relative importance
of percutaneous uptake has increased. Because lowered limits for airborne exposures may
not necessarily decrease the potential for dermal absorption, a ‘skin’ notation is of
increasing relevance for compounds that may cause a systemic toxicity hazard due to
uptake via this route. Skin notations were originally introduced as qualitative indicators
of hazards related to dermal absorption. Thus, skin notations were to be used as warning
signs only, indicating that a certain chemical might penetrate the human skin with the
potential for systemic toxicity.
One of the main obstacles in this regard is that the skin notation is often used as
an instrument for risk management. However, work-place exposures need to be assessed
in quantitative terms, and a qualitative hazard indicator is neither intended nor very useful
for risk assessment or risk management. Thus, if a chemical is listed with a skin notation,
must all cutaneous contact necessarily be fully avoided? And, vice versa, is skin contact
not a problem if a chemical is not listed? These questions suggest that skin notation needs
to be supplemented by some kind of quantitative measure to be useful in most
occupational health settings. Thus, the skin notation does not allow differentiation
between chemicals depending on their relative toxicity or percutaneous penetration rates.
Recent emphasis on occupational hazards has lead to lowered limits for respiratory
exposures. Increased emphasis on cutaneous exposures would therefore seem warranted.
As the understanding of percutaneous absorption processes has improved, some
countries have developed alternative and more sophisticated guidelines for assigning skin
notations to chemicals (Fiserova-Bergerova et al. 1990, Bos et al. 1998). These guidelines
3
also aim at taking into account the relative importance of the dermal absorption compared
to inhalation exposure at the exposure limit. Thus, a skin notation is assigned if dermal
absorption is expected to cause a significantly increased total systemic exposure when
pulmonary exposure equals the current exposure level. Although this approach is highly
attractive, several issues must be considered. First, how much is a ‘significantly’
increased absorption? Second, should this assessment take into regard that percutaneous
absorption may be considerably increased in certain dermatological diseases and after
defattening of the skin? Third, many cutaneous exposures involve chemical mixtures
where one agent may enhance or attenuate the penetration of another. Should skin
notation include substances that enhance the penetration of other, more toxic, chemicals?
Such issues may be difficult to address with as crude an administrative instrument as the
notation in official lists of exposure limits (Sartorelli 2000). An interesting notion is that
one of the documents most often quoted and cited in discussions on methods for
assigning skin notations, the ECETOC document 31 entitled “Strategy for assigning a
´skin notation´”, has never been published in final form by the European chemical
industry association and still exists only in draft versions (ECETOC, pers. comm.).
The guidelines used for assigning skin notations have developed over time and
differences between countries are known to occur (Grandjean 1990). Today, some
countries also include skin irritation and dermatoses as a reason for assigning skin
notations on chemicals (Czerczak and Kupczewska 2002). Still, these countries otherwise
have comparable airborne exposure limits. A possible reason for uncertainty and
controversy is that proper data on dermal penetration of chemicals are very often missing,
of questionable quality, or they exist only for the pure chemical and not for the mixture
4
present in commercial products. This lack of relevant information on penetration
characteristics for chemical products will need to be addressed by authorities when
approving new products for marketing.
The latest thorough comparison on the use of skin notations on occupational
relevant chemicals in different countries dates back some ten years (Grandjean, 1990).
We now provide a partial update of this information with a discussion of current issues
related to assigning skin notations to industrial chemicals and products.
5
Material and methods
The most recent lists of occupational exposure limits (OELs) with skin notation from
Denmark (Arbejdstilsynet 2000), Germany (DFG 2000), Netherlands (MSZW 2001),
Poland (Czerczak and Kupczewska 2002), Slovakia (Directive 45/2002), and the USA
(ACGIH 2001) were used for comparisons. The countries were selected to represent a
country outside Europe with a long tradition for independent evaluation of occupational
exposure limits (USA) and three European Union (EU) countries (Denmark, Germany,
Netherlands) with independent assessment of exposure limits. Further, two countries
from Eastern Europe, now scheduled to become EU-members (Poland and Slovakia),
were included in the comparison.
Based on names of chemicals on the lists of OELs, a comparison was made
between the use of skin notation in these six countries. Some substances were listed
under different chemical names, and CAS-numbers were then used for identification.
Some countries grouped isomeric compounds, whereas other countries used individual
OEL values for each isomer. We have adjusted our data for such differences by grouping
isomeric chemicals; if a single isomer had a skin notation, the combined group of isomers
was rated likewise. This adjustment means that the numbers stated for each country may
deviate slightly from those on the national lists of OELs. However, these differences are
small and of minor importance for a discussion on the considerable differences between
countries. Specific chemicals were selected from the national lists to illustrate some of
the problems related to the use of skin notation. These examples are provided in lieu of a
more comprehensive comparison, but are thought to be representative of major
discrepancies.
6
Results
Chemicals with skin notation
Germany, USA, Denmark, and the Netherlands include approximately the same number
of chemicals on their OEL lists, whereas Poland and Slovakia have fewer entries (Table
I). At least in Slovakia, the lower number of chemicals reflects the ongoing process of
establishing new independent national OELs. The largest number of chemicals with a
skin notation occurs in Denmark where over 200 chemicals have a skin notation. In
general, however, some 30% of all chemicals on the OEL lists in these countries have a
skin notation (Table I).
Agreement between skin notations in six countries
Similar proportions of chemicals with skin notation on official OEL lists in the six
countries (Table I) would suggest a reasonably good agreement on the use of skin
notation. However, substantial differences exist (Figure 1). For example, Denmark and
USA have comparable numbers of chemicals with a skin notation; 204 and 192,
respectively. However, only in 151 cases do the two lists agree, leaving 41 chemicals on
the US list though not on the Danish list, and 53 on the Danish list that are not on the US
list (Figure 1). For the remaining countries included in this comparison, disagreement
also seems to prevail. As the documentation on which the skin notation is based is
probably identical, the reason for the lack of agreement between countries would likely
relate to the criteria used for assigning a skin notation.
Criteria for skin notation
7
The criteria used for assigning skin notation differ between countries, not only in
interpretation and intention, but also in regard to its consequence.
In Poland, the symbol Sk attached to the name of a chemical on the OEL list
indicates that to assess the exposure quantitatively, air sampling only is not sufficient and
must be accompanied by determination of the potential for dermal absorption (Czerczak
and Kupczeweska 2002). Thus, the Polish criterion is strictly related to exposure
assessment and does not include any requirement for specific prevention of health effects.
Thus, the Polish interpretation is in line with the original intentions of the skin notation.
The German criterion has recently been discussed by Drexler (1998): ‘When the
maximal allowable concentration (MAK) value for the substance is not sufficient to
protect dermally exposed persons from adverse effects on their health’. The German
criterion has removed its linkage to hazard, exposure, and potential for absorption, and is
instead linked directly to protection against adverse health effects. Thus, in contrast to
Poland, a chemical that does penetrate the skin but does not demonstrate significant
health effects will not be given a skin notation in Germany. Moreover, the German
provision includes a clear consequence of assigning a skin notation to a chemical. Thus,
when an employee has dermal contact with a substance that has been assigned a skin
notation, the employer has to arrange for biological monitoring (TRGS 150). The
advantage is that the total personal exposure must be assessed in quantitative terms,
thereby improving the basis for a risk assessment. The disadvantage is the lack of
validated methods for biological monitoring for many chemicals, perhaps excluding some
chemicals from a skin notation. However, while the German notion has introduced
adverse health effects as part of the criterion, it does not include quantitative terms.
8
The Dutch criterion focuses on the quantitative assessment of the relative
importance of the dermal exposure. The Dutch criterion is defined as follows: ‘When the
amount absorbed by both arms and forearms in 1 hour corresponds to more than 10% of
the amount absorbed via the lungs upon exposure to the OEL for 8 hours’ (de Cock
1996). Thus, the Dutch criterion is defined in terms of exposure area and exposure time
and is related to the OEL.
Criteria using quantitative measures for assigning skin notations
The toxic properties of chemicals given a skin notation differ, both regarding
toxicokinetic and toxicodynamic features including degrees of reversibility. Thus, to
assess risk and potential differences in risk, a joint measure or indicator ought to reflect
these differences in toxic properties.
One suggestion to include a quantitative term is to consider the dermal LD50 in the
criteria for skin notation (Kennedy et al. 1993, Czerczak and Kupczewska 2002).
Chemicals with dermal LD50 values below 2000 mg/kg are normally described as
constituting a health risk to humans, and an LD50 value below 2000 mg/kg has previously
been described as the ‘cut-off’ value for skin notation attribution (Scansetti et al. 1988).
However, the present use of skin notation does not reflect this situation, as only few
countries have included these chemicals on their lists of chemicals with a skin notation
(Table II). Further, an inherent problem with LD50 values is that they refer only to acute
toxicity with death as the single registered outcome. Thus, an LD50 value will not
accommodate non-lethal toxicity, accumulation, or effects of repeated dosing. For
example, ammonium perfluoro octanoate with an LD50 value of 4,300 mg/kg has a skin
9
notation in only two of the countries included in this comparison (DK and NL) despite
the demonstration of liver damage after 10 daily dermal dosages of only 20 mg/kg
(Kennedy 1985).
Another term that might be more suitable to include in the criteria for skin
notation is the OEL for respiratory exposures. However, recognizing that skin notations
depend on OELs also imply that differences between countries with respect to OELs will
consequently cause differences in regard to which chemicals that will receive a skin
notation (Table III). Even within the EU, where efforts have been directed toward more
uniform regulations, differences between OELs remain (DK, D, and NL in Table III).
Allergens and penetration enhancers
Consideration of only systemic effects excludes not only specific target organs but also
adverse effects such as allergy. This limitation might be worth considering from a
mechanistic as well as a toxicity viewpoint. Thus, an allergic response requires the
immune system to be initially sensitized followed by an exposure that will promote the
immune response observed as allergy. All processes occur within the body and could well
be considered systemic, because the allergen has to pass the dermal membrane in order to
reach the immune system targets.
Penetration enhancers represent a special problem. Typical examples of such chemicals
would be detergents, vehicles or solubilizers. These chemicals often do not cause
systemic toxicity themselves at the exposures occurring during normal use. Therefore, the
criteria for skin notation used in most countries will exclude these chemicals from a skin
10
notation. However, these penetration enhancers may promote penetration of other
chemicals through damage of the dermal barrier or by otherwise facilitating dermal
penetration of other chemicals (Nielsen 2000, Nielsen et al. 2000). In the strict definition
of skin notation, this specific group of chemicals would not receive a skin notation
because they are not causing systemic toxicity by themselves. However, as exemplified
with dimethyl sulfoxide (DMSO), several countries do not follow the strict criteria for
assigning skin notations to chemicals causing systemic toxicity only. Thus, with the
exception of USA and Poland, all countries included in this comparison have assigned
DMSO with a skin notation.
11
Discussion
The original intention with the skin notation was to provide a warning that the chemical
in question may cause systemic toxicity following dermal exposure. Such a hazard
indicator allows attention to be focused when the number of chemicals given this
classification is limited. However, with close to one third of all chemicals given a skin
notation, this qualitative approach is loosing its value as a warning signal. Under these
circumstances, risk management based on this hazard indicator gets blurred. A
quantitative term would be required to allow a differentiation between different degrees
of risk associated with cutaneous exposure to chemicals with a potential for dermal
absorption and systemic toxicity.
Acknowledging that skin notations would no longer be a qualitative indicator but
related to the OEL would require that skin notations be reconsidered whenever OELs are
changed or when inhalation exposures for other reasons are reduced, so that the relative
importance of the dermal exposure route increases. However, for many solvents, the OEL
has been reduced drastically during the last decade or so, whereas the number of solvents
with a skin notation has not increased to the extent expected.
Using OEL as part of the criteria for skin notation will also have to consider those
situations where the OEL aims at protecting against irritation of mucous membranes in
the eyes or the respiratory tract, i.e., target organs that are specifically excluded from the
present criteria for skin notation. A pragmatic solution would be to calculate an additional
OEL for systemic toxicity to be used for assigning skin notations for those chemicals for
which the critical effect (the first significant health effect occurring at increasing
exposure/dose) is not systemic.
12
For all of these purposes,, skin notation requires solid data for evaluating the
potential for systemic toxicity following dermal penetration. Diverse though scattered
data may originate from human experience, experimental in vivo data from humans or
laboratory animals, in vitro data based on trans-dermal penetration flux, or fluxes
estimated from structure-activity relationships and solubility characteristics (Bos et al.
1998, Fiserova-Bergerova et al. 1990). Despite continued efforts through recent decades
to generate more valid data on dermal penetration, the problem remains that the decision
whether or not to assign a skin notation often is based on only one or two studies of
reasonably high quality and relevance. Thus, there is an urgent need for more valid data
on dermal penetration of chemicals.
The present criteria focus on single chemicals, not mixtures. Chemicals do,
however, seldom occur as neat chemicals, but as part of industrial products and processes
with active ingredients as well as detergents, solubilizers etc. Thus, interaction between
chemicals may involve toxic mechanisms, but may also include potentiation of
transdermal transport. Compounds that act as carriers or penetration enhancers will
seldom receive a skin notation using the present criteria, despite significantly affecting
penetration and toxic risk of other chemicals. Therefore, interaction ought to be addressed
in the criteria for skin notation. However, data on penetration characteristics of mixtures
are very limited, though increasingly requested by regulatory agencies for products as
well as individual ingredients. It will probably be at least as difficult to develop tools to
handle mixtures in relation skin notation as it has been in relation to general OELs. A
first step could be to incorporate a notation for penetration enhancers.
13
The differences observed between countries with respect to skin notations
document that, despite equal access to the literature, conclusions differ between national
experts. Differences in the written criteria for skin notation obviously play a role. Even
though uniform guidelines for assigning skin notations may be agreed upon, the final
decision includes subjective elements, when experts decide which studies are sufficiently
good to consider and how to judge the findings reported. Decisions on assignment of skin
notations should therefore be accompagnied by specific reference to the documentation
as well as the arguments for and against assigning a skin notation. Such documentation
will provide more transparency in the decision process and may allow transfer of
assessments on skin notation from one country to another.
The inconsistencies indicate an urgent need for a more efficient evaluation process
regarding risk of toxicity following dermal exposure. One first step forward would be to
agree on common criteria for assigning skin notations and thorough guidelines on using
skin notation as a hazard indicator and as a potential tool for risk assessment and risk
management. This process is under way within EU, but deserves international
collaboration.
14
References
American Conference of Governmental Industrial Hygienists (ACGIH). 2001. Threshold
limit values for chemical substances and physical agents. ACGIH, Cincinnati, US.
Arbejdstilsynet. 2000. At-vejledning Nr. C.O.1 af oktober 2000. Grænseværdier for
stoffer og materialer. Arbejdstilsynet, Copenhagen, Denmark.
Bos PMJ., Brouwer DH, Stevenson H, Boogaard PJ, de Kort WLAM, van Hemmen JJ.
1998. Proposal for the assessment of quantitative dermal exposure limits in occupational
environments: part 1. Development of a concept to derive a quantitative dermal
occupational exposure limit. Occup Environ Med 55: 795-804.
Czerczak S, Kupczewska M. 2002. Assignment of skin notation for maximum allowable
concentration (MAC) list in Poland. Appl Occup Environ Hyg 17: 187-199.
De Cock J, Heederik D, Kromhout H, Boleij JSM. 1996. Strategy for assigning a ´skin
notation´: a comment. Ann Occup Hyg 40: 611-614.
Deutsche Forschungsgemeinschaft (DFG). 2000. List of MAK and BAT values, report
No 36. Wiley-VCH Verlag, Weinheim, Germany.
Directive of the Government of the Slovak Republic, No 45/2002 Z.z.
15
Drexler H. 1998. Assignment of skin notation for MAK values and its legal consequences
in Germany. Int Arch Occup Environ Health 71: 503-505.
Fiserova-Bergerova V, Pierce JT, Droz PO. 1990. Dermal absorption potential of
industrial chemicals: criteria for skin notation. Am J Ind Med 17: 617-635.
Grandjean P. 1990. Skin denotation. In: Grandjean P, editor. Skin penetration: Hazardous
chemicals at work. London UK: Taylor & Francis. p 20-26.
Kennedy, Jr., GL. 1985. Dermal toxicity of ammonium perfluorooctanoate. Tox Appl
Pharmacol 81: 348-355.
Kennedy, Jr., GL, Brock WJ, Banerjee AK. 1993. Assignment of skin notation for
threshold limit values chemicals based on acute dermal toxicity. Appl Occup Environ
Hyg 8: 26-30.
MSZW (Ministerie van Sociale Zaken en Werkgelegenheid). 2001. Nationale MAC lijst
2001. Sdu Uitgevers, Den Haag, The Netherlands.
Nielsen JB. 2000. Influence of four detergents on in vitro barrier function of human skin.
Int J Occup Environ Health 6: 143-147.
16
Nielsen, GD, Nielsen JB, Andersen KE, Grandjean P. 2000. Effect of industrial
detergents on the barrier function of human skin. Int J Occup Environ Health 6: 138-142.
Sartorelli P. 2000. La stima del rischio cutaneo in medicina del lavoro. Med Lav 91: 183191.
Scansetti G, Piolatto G, Rubino GF. 1988. Skin notation in the context of workplace
exposure standards. Am J Ind Med 14: 725-732.
TRGS 150. 1989;1996. Bundesarbeitsblatt 10/1989, pp 47-48, and Bundesarbeitsblatt
6/1996, pp 31-33.
17
Table I. Number of chemicals on national OEL lists and chemicals with skin notation.
Germany
Chemicals
in OEL lists
N
684
Chemicals with skin
notation
N
Per cent
176
26
USA
650
192
30
Denmark
634
204
32
The Netherlands
676
157
23
Poland
414
144
35
Slovakia
269
85
32
Table II. Presence or absence of skin notation for five chemicals with a dermal LD50
below 2000 mg/kg.
DK
USA
NL
D
P
Slo
Ethylamine
+
+
-
-
+
-
Cyanamide
-
-
-
-
+
-
Methacrylic
acid
Sodium
azide
Acroleine
-
+
-
-
-
-
-
-
-
-
-
+
-
+
-
-
-
-
Table III. Occupational exposure limits for four solvents (ppm, 8 hour TWA´s).
Chemical
DK
USA
NL
D
P
Slo
Xylene
25
100
50
100
25
50
n-hexane
25
50
25
50
30
50
Toluene
25
50
40
50
25
50
Perchlorethylene
10
25
35
n.a.*
10
50
* Perchlorethylene is rated as carcinogenic in Germany and not given an exposure limit.
18
US
US
US
US
US
41
81
91
112
137
151
111
101
80
55
DK
30
46
53
75
64
NL
D
SLO
P
Figure 1. The number of chemicals with a skin notation in the USA as compared with
five other countries. Hatched overlap indicates the number of chemicals given a skin
notation in both countries.
19