ICNIRPUV2004

INTERNATIONAL COMMISSION ON NON-IONIZING RADIATION PROTECTION
ICNIRP GUIDELINES
ON LIMITS OF EXPOSURE TO ULTRAVIOLET
RADIATION OF WAVELENGTHS BETWEEN
180 nm AND 400 nm (INCOHERENT OPTICAL
RADIATION)
PUBLISHED IN: HEALTH PHYSICS 87(2):171-186; 2004
ICNIRP PUBLICATION – 2004

ICNIRP Guidelines
GUIDELINES ON LIMITS OF EXPOSURE TO ULTRAVIOLET

RADIATION OF WAVELENGTHS BETWEEN 180 NM AND 400
NM (INCOHERENT OPTICAL RADIATION)
The International Commission on Non-Ionizing Radiation Protection*
INTRODUCTION radiation, so that they may serve as guidance to the

various international and national bodies or individual
SINCE THE publication of the ICNIRP Guidelines on UV experts who are responsible for the development of
Radiation Limits (ICNIRP 1996),† recent research has regulations, recommendations, or codes of practice to
made it appropriate to update the guidelines for protec- protect workers and the general public from the poten-
tion. While no significant changes are made in the values, tially adverse effects of UVR.
the biological basis can be strengthened, and the limita- The Committee recognized that when standards or
tions on use can be clarified. exposure limits (ELs) are established, various value
A document titled Environmental Health Criteria judgments are made. The validity of scientific reports has
160, Ultraviolet Radiation (UNEP 1994), was published to be considered, and extrapolations from animal exper-
in 1994 under the joint sponsorship of the United Nations
iments to effects on humans have to be made. Costs vs.
Environment Programme (UNEP), ICNIRP, and the
benefit analyses are necessary, including economic im-
World Health Organization (WHO). The document con-
pact of controls. The limits in these guidelines were
tains a review of the biological effects reported from
based on the scientific data, and no consideration was
exposure to ultraviolet radiation (UVR) and serves as the
given to economic impact or other non-scientific priori-
scientific rationale for the development of these guide-
ties. However, the limits represent conditions under
lines. In addition, the International Agency for Cancer
which it is expected that nearly all individuals may be
Research (IARC) published a monograph on UVR in
repeatedly exposed without acute adverse effects and,
1992 (IARC 1992) and published a monograph on
based upon best available evidence, without noticeable
sunscreens more recently (IARC/WHO 2001). Further-
risk of delayed effects (see paragraph on Special Con-
more, the National Radiological Protection Board
siderations). Although a single set of limits can apply for
(NRPB) has recently published a scientific review of the
exposure of the eye, it is not possible to provide a single
health effects of UVR (NRPB 2002). Reviews of relevant
exposure limit that applies to all skin phototypes. Addi-
UVR biological action spectra were published in a
tional guidance is required for applying guidelines for
monograph on the measurement of optical radiation
hazards (ICNIRP/CIE 1998). The important publications skin protection.
that relate most directly to the guidelines [some of which ICNIRP Subcommittee IV (Optical Radiation) pre-
have appeared since the Environmental Health Criteria pared the initial update of these guidelines after an
(EHC) document was drafted] are referenced in the extensive review of the current scientific evidence. The
rationale (Appendix). IRPA Associate Societies as well as a number of com-
The purpose of these guidelines is to provide basic petent institutions and individual experts were consulted
principles of protection against non-coherent ultraviolet in the preparation of the guidelines and their cooperation
is gratefully acknowledged.
In its review of the whole database, ICNIRP noted that
* ICNIRP, c/o BfS—R. Matthes, Ingolstaedter Landstr. 1, 85764
Oberschleissheim, Germany.
a substantial number of studies have been published since
†
The initial guidelines were published in Health Phys 49:331– 1989, when the last detailed rationale for the guidelines was
340; 1985, amended in Health Phys 56:971–972; 1989, and recon- published, and since the UNEP/ICNIRP/WHO EHC was
firmed by ICNIRP in Health Phys 71:978; 1996.
For correspondence or reprints contact: R. Matthes at the above published in 1994. Many of the biological effects, where
address or email at info@icnirp.org. only tentative data were available in 1994, have now been
(Manuscript received 5 February 2004; accepted 30 April 2004)
0017-9078/04/0 clarified. In particular, the understanding of UVA-induced
Copyright © 2004 Health Physics Society damage to DNA by indirect mechanisms, the involvement
171
172 Health Physics August 2004, Volume 87, Number 2
of new mechanisms for cell protection against the harmful and the role of both UVA and UVB in the development
effects of photosensitized reactions, and the participation of of different types of cataract (UNEP 1994). The Interna-
UVA in the chain of events believed to play a role in tional Agency for Research on Cancer (IARC) of the
melanocytic and non-melanocytic skin cancer provide a WHO recently reviewed the impact of sunscreens
better understanding of the risk of human exposure to UVR. (IARC/WHO 2001).
There is further evidence for the importance of early life ICNIRP concludes that, while significant clarification
(childhood and adolescence) irradiation for melanocytic has occurred with respect to health risk assessment from
skin cancer (IARC/WHO 2001) and probably for basal cell exposure to UVR, recent data do not provide any results
carcinoma (Kricker et al. 1995; Gallagher et al. 1995a, b). suggesting that the exposure limit values contained in Table
There has been significant improvement in the understand- 1 of the 1989 guidelines need to be amended. This conclu-
ing of the complex chain of events involved in photocarci- sion is supported by a review conducted by the National
nogenesis, e.g., the discovery of a UVR signature at the Radiological Protection Board (NRPB 2002). Thus, IC-
molecular level (i.e., the p53 gene mutation) (Mukhtar and NIRP reaffirms the 1989 guidelines on exposure limits to
Elmets 1996; IARC 1992). Progress has also been made in UVR as valid for current use. ICNIRP will continue to
standardizing several action spectra including those for monitor the scientific literature and amend the guidelines on
photocarcinogenesis and erythema by the International exposure limits as necessary.
Commission on Illumination (CIE 1999, 2000, 2002).
It was noted, however, that a number of issues still BACKGROUND
need further research before a more complete health risk
assessment can be made. These include the modulation Ultraviolet radiation (UVR) occupies that portion of
of the immune system by both UVA and UVB and their the electromagnetic spectrum from at least 100 to 400
interaction with several chromophores; the apparent role nanometers (nm). In discussing UVR biological effects,
of UVA in the development of melanocytic skin cancer; the International Commission on Illumination (CIE) has
Table 1. UV exposure limits and spectral weighting function.

ELd ELd ␭a ELd ELd
␭a (nm) (J m⫺2) (mJ cm⫺2) S(␭)b (nm) (J m⫺2) (mJ cm⫺2) S(␭)b
180 2,500 250 0.012 310 2,000 200 0.015
190 1,600 160 0.019 313c 5,000 500 0.006
200 1,000 100 0.030 315 1.0 ⫻ 104 1.0 ⫻ 103 0.003
205 590 59 0.051 316 1.3 ⫻ 104 1.3 ⫻ 103 0.0024
210 400 40 0.075 317 1.5 ⫻ 104 1.5 ⫻ 103 0.0020
215 320 32 0.095 318 1.9 ⫻ 104 1.9 ⫻ 103 0.0016
220 250 25 0.120 319 2.5 ⫻ 104 2.5 ⫻ 103 0.0012
225 200 20 0.150 320 2.9 ⫻ 104 2.9 ⫻ 103 0.0010
230 160 16 0.190 322 4.5 ⫻ 104 4.5 ⫻ 103 0.00067
235 130 13 0.240 323 5.6 ⫻ 104 5.6 ⫻ 103 0.00054
240 100 10 0.300 325 6.0 ⫻ 104 6.0 ⫻ 103 0.00050
245 83 8.3 0.360 328 6.8 ⫻ 104 6.8 ⫻ 103 0.00044
250 70 7 0.430 330 7.3 ⫻ 104 7.3 ⫻ 103 0.00041
254c 60 6 0.500 333 8.1 ⫻ 104 8.1 ⫻ 103 0.00037
255 58 5.8 0.520 335 8.8 ⫻ 104 8.8 ⫻ 103 0.00034
260 46 4.6 0.650 340 1.1 ⫻ 105 1.1 ⫻ 104 0.00028
265 37 3.7 0.810 345 1.3 ⫻ 105 1.3 ⫻ 104 0.00024
270 30 3.0 1.000 350 1.5 ⫻ 105 1.5 ⫻ 104 0.00020
275 31 3.1 0.960 355 1.9 ⫻ 105 1.9 ⫻ 104 0.00016
280c 34 3.4 0.880 360 2.3 ⫻ 105 2.3 ⫻ 104 0.00013
285 39 3.9 0.770 365c 2.7 ⫻ 105 2.7 ⫻ 104 0.00011
290 47 4.7 0.640 370 3.2 ⫻ 105 3.2 ⫻ 104 0.000093
295 56 5.6 0.540 375 3.9 ⫻ 105 3.9 ⫻ 104 0.000077
297c 65 6.5 0.460 380 4.7 ⫻ 105 4.7 ⫻ 104 0.000064
300 100 10 0.300 385 5.7 ⫻ 105 5.7 ⫻ 104 0.000053
303c 250 25 0.120 390 6.8 ⫻ 105 6.8 ⫻ 104 0.000044
305 500 50 0.060 395 8.3 ⫻ 105 8.3 ⫻ 104 0.000036
308 1,200 120 0.026 400 1.0 ⫻ 106 1.0 ⫻ 105 0.000030
a
Wavelengths chosen are representative; other values should be interpolated (see Eqns. 2a– c).
b
Relative spectral effectiveness.
c
Emission lines of a mercury discharge spectrum.
d
EL for a monochromatic source, but also limited by a dose-rate of 10 kW m⫺2 (1 W cm⫺2) for durations greater than 1 s as well in
order to preclude thermal effects.
Guidelines on limits of exposure to UV radiation ● ICNIRP 173
divided the UV spectrum into three bands. The band 315 arcs, gas and vapor discharges, fluorescent lamps, incan-
to 380 – 400 nm is designated as UVA, 280 to 315 nm as descent sources, and solar radiation. The limits do not
UVB, and 100 to 280 nm as UVC (CIE 1987, 1999). apply to lasers that emit UVR. Most incoherent UVR
Wavelengths below 180 nm (vacuum UV) are of little sources are broadband, although single emission lines
practical biologic significance since they are readily can be produced from low-pressure gas discharges.
absorbed in air. Ultraviolet radiation is used in a wide These values should be used as guides in the control of
variety of medical and industrial processes and for exposure to both pulsed and continuous sources where
cosmetic purposes. These include photocuring of inks the exposure duration is not less than 1 ␮s. These ELs are
and plastics (UVA and UVB), photoresist processes (all below levels that would be used for UV exposures of
UV), solar simulation (all UV), cosmetic tanning (UVA patients required as a part of medical treatment or for
and UVB), fade testing (UVA and UVB), dermatology elective cosmetic purposes. These ELs are exceeded for
(all UV), and dentistry (UVA). Even though the principal exposed skin by noonday summer sunlight overhead at
operating wavelengths for most of these processes are in 0 – 40° latitude within 5–10 min. The ELs should be
the UVA, almost always some shorter wavelength (UVB considered absolute limits for direct exposure of the eye
and UVC) radiation and violet light are emitted as well. and “advisory” for skin exposure because of the wide
Many industrial applications employ arc sources for heat range of susceptibility to skin injury depending on skin
or light (e.g., welding), which also produce UVR as an type. The ELs should be adequate to protect lightly
unwanted admixture for which control measures may be pigmented individuals.
necessary. While it is generally agreed that some low-
level exposure to UVR benefits health (UNEP 1994; BASIC CONCEPTS
Preece et al. 1975; Clemens et al. 1982; Holick 2000;
Webb et al. 1988, 1989; MacLaughlin and Holick 1985), This document makes use of the spectral band
there are adverse effects (de Gruijl 1997; UNEP 1994; designations of the CIE. Unless otherwise stated, UVA is
ICNIRP/CIE 1998) that necessitate the development and from 315 to 400 nm, UVB is from 280 to 315 nm, and
use of ELs for UVR. However, the development of UVR UVC is from 100 to 280 nm (CIE 1984, 1987). It should
EL poses a real challenge to achieve a realistic balance be noted that some specialists follow this general scheme
between beneficial and adverse health effects. but take the dividing line between UVA and UVB at 320
Until 1980, it was generally thought that the most nm. The UVR exposure should be quantified in terms of
significant adverse UVR health effects resulted from an irradiance E (W m⫺2 or W cm⫺2) for continuous
exposures at wavelengths below 315 nm; but today these exposure or in terms of a radiant exposure H (J m⫺2 or J
effects are recognized to be produced at longer wave- cm⫺2) for time-limited (or pulsed) exposures of the eye
lengths (UVA) at substantially higher doses. At one time, and skin. The geometry of exposure to UVR is very
wavelengths below 315 nm were collectively known as important. For example, the eyes (and to a lesser extent
“actinic radiation,” when it was thought that these effects the skin) are anatomically protected against UVR expo-
occurred only in the UVB and UVC. This guideline has sure from overhead sources such as the sun overhead
been limited to wavelengths greater than 180 nm where (Sliney 1995; UNEP 1994). The limits should be applied
UVR is transmitted through air. The most restrictive to exposure directed perpendicular to those surfaces of
limits are for exposure to radiation having those wave- the body facing the radiation source, measured with an
lengths less than 315 nm. instrument having cosine angular response (UNEP
1994). For highly non-uniform irradiation the irradiance
and radiant exposure need not be averaged over the area
PURPOSE AND SCOPE of a circular measurement aperture smaller than 1 mm in
diameter for pulsed exposures and 3.5 mm for lengthy
The purpose of this document is to provide guidance exposures.
on maximal limits of exposure to UVR in the spectral These ELs should be used as guides in the control of
region between 180 nm and 400 nm. The limits represent exposure to UV sources and as such are intended as
conditions under which it is expected that nearly all limits for non-therapeutic and non-elective exposure. The
individuals may be repeatedly exposed without acute ELs should be considered as absolute limits for ocular
adverse effects and, based upon best available evidence, exposure. The ELs were developed by considering
without noticeable risk of delayed effects (see paragraph lightly pigmented populations (i.e., white Caucasian)
on Special Considerations). These EL values for expo- with greatest sensitivity and genetic predisposition for
sure of the eye or the skin may be used to evaluate skin cancer. Exposure during sun bathing and tanning
potentially hazardous exposure from UVR; e.g., from under artificial sources may well exceed these limits but
exposed individuals should be advised that some health Permissible exposure time in seconds for exposure to
risk is incurred from such activity. Eye protection is UVR incident upon the unprotected skin or eye may be
always required during therapeutic exposures. Neverthe- computed by dividing 30 J m⫺2 by the value of Eeff in W
less, occasional exposures to conditioned skin may not m⫺2. The maximal exposure duration may also be deter-
result in adverse effects. The rationale for the UVR mined using Table 2, which provides representative
exposure limits is provided in the Appendix. exposure durations corresponding to effective irradiances
in W m⫺2 or ␮W cm⫺2.
Values of S(␭) for wavelengths that are not listed in
EXPOSURE LIMITS
Table 1 may be interpolated through the application of
For the EL for both general and occupational expo- the following three formulas (Wester 2000). The three
sure to UVR incident upon the skin or eye within an 8-h simple mathematical expressions apply in the range only
period, the following applies. from 210 – 400 nm:
Exposure of the eyes For the region

Ultraviolet radiant exposure in the spectral region
180 to 400 nm incident upon the unprotected eye(s) 210 ⱕ ␭ ⱕ 270 nm S共␭兲 ⫽ 0.959共 270⫺ ␭ 兲 (2a)
should not exceed 30 J m⫺2 effective spectrally weighted For the region
using the spectral weighting factors contained in Table 1,
and the total (unweighted) ultraviolet radiant exposure in
the spectral region 315 to 400 nm should not exceed 104 270 ⬍ ␭ ⱕ 300 nm S共␭兲 ⫽ 1 ⫺ 0.36x 冉␭ ⫺ 270
20 冊 1.64
J m⫺2.
(2b)
Exposure of the skin For the region

For the most sensitive, non-pathologic, skin photo-
types (known as “melano-compromised”), ultraviolet 300 ⬍ ␭ ⱕ 400 nm S共␭兲 ⫽ 0.3 ⫻ 0.736共 ␭ ⫺300 兲
radiant exposure in the spectral region 180 to 400 nm ⫹ 10共 2⫺0.0163 ␭ 兲 . (2c)
upon the unprotected skin should not exceed 30 J m⫺2
effective spectrally weighted using the spectral weight- The formulae interpolate between and substitute with
ing factors contained in Table 1. This limit should be reasonable accuracy the points along the action spectrum.
considered a desirable goal for skin exposure to mini-
mize the long-term risk, but it must be recognized that SPECIAL CONSIDERATIONS
this limit is difficult to achieve in sunlight and judgment
These EL values are intended to apply to UVR
must be used in its practical application. It has a very
exposure of the working population, but with some
substantial safety factor for dark skin phototypes (known
precaution also apply to the general population. How-
as “melano-competent”) and more generally for individ-
ever, it should be recognized that some rare, highly
uals who have been conditioned by previous, repeated
exposures (known as “melano-adapted,” i.e., tanned).
To determine the effective irradiance of a broadband Table 2. Limiting UV exposure durations based on exposure
source weighted against the peak of the spectral effec- limits.
tiveness curve (270 nm), the following weighting for- Effective irradiance
mula should be used: Duration of exposure
per day Eeff (W m⫺2) Eeff (␮W cm⫺2)
E eff ⫽ ⌺ E ␭ 䡠 S 共 ␭ 兲䡠 ⌬ ␭ , (1) 8h 0.001 0.1
4h 0.002 0.2
where: 2h 0.004 0.4
1h 0.008 0.8
Eeff ⫽ effective irradiance in ␮W cm⫺2 (␮J s⫺1 30 min 0.017 1.7
cm⫺2) or W m⫺2 (J s⫺1 m⫺2) normalized to a 15 min 0.033 3.3
10 min 0.05 5
monochromatic source at 270 nm; 5 min 0.1 10
E␭ ⫽ spectral irradiance from measurements in ␮W 1 min 0.5 50
cm⫺2 nm⫺1 or W m⫺2 nm⫺1; 30 s
10 s
1.0
3.0
100
300
S(␭) ⫽ relative spectral effectiveness (unitless); and 1s 30 3,000
⌬␭ ⫽ bandwidth in nanometers of the calculation or 0.5 s 60 6,000
0.1 s 300 30,000
measurement intervals.
photosensitive individuals exist who may react adversely hazardous UVR produced in many industrial applications
to exposure at these levels. These individuals are nor- such as the fade testing of materials, solar simulation,
mally aware of their heightened sensitivity. Likewise, if photoresist applications, and photocuring. For arc weld-
individuals are concomitantly exposed to photosensitizing, cabinets are not practical. Shields, curtains, barriers,
ing agents (Fitzpatrick et al. 1974; Johnson 1992), a and a suitable separation distance are used to protect
photosensitizing reaction can take place. It should be individuals against the UVR emitted by open-arc pro-
emphasized that many individuals who are exposed to cesses such as arc welding, arc-cutting, and plasma
photosensitizing agents (ingested or externally applied spraying. Dynamic-filter welding helmets and see-
chemicals, e.g., in cosmetics, foods, drugs, industrial through curtains have improved the safety of welding
chemicals, etc.) probably will not be aware of their operations in recent decades. There is a need for opera-
heightened sensitivity. Phototoxic reactions apply to all tional rules to protect potentially exposed individuals.
individuals and depend upon the quantity of photosensi- Operators should be trained to follow these general rules
tizing chemicals and the UVR exposure, whereas pho- properly. Ventilation may be required to exhaust ozone
toallergic reactions will be observed for much lower and other airborne contaminants produced by UVC
quantities of the substance in sensitized individuals. radiation.
Lightly pigmented individuals conditioned by previous
UVR exposure (leading to tanning and hyperplasia) and
heavily pigmented individuals can tolerate skin exposure MEASUREMENT
in excess of the EL without erythemal effects. However,
repeated tanning may increase the risk for those persons UV measurements for health risk evaluation are
later developing signs of accelerated skin aging and even sometimes of value for indoor exposure assessment.
skin cancer. Such risks should be understood prior to the However, they are generally not routinely performed for
use of UVR for medical phototherapy or cosmetic outdoor exposure conditions, except with regard to the
exposures. use of the Global UV Index (ICNIRP/WHO/WMO/
UNEP 2002; Gies et al. 1995).
PROTECTIVE MEASURES Although direct-reading UVR radiometers exist,
attempts to produce relatively inexpensive field safety
Protective measures will differ depending upon survey meters that respond directly to UVB and UVC
whether the UVR exposure results from sunlight or radiation [following the S(␭) function] have not been
from artificial sources. The use of hats, eye protectors, fully successful. However, relatively expensive instru-
clothing, and sun-shading structures are practical pro- ments exist which respond to UVB and UVC radiation
tective measures to reduce sunlight exposure. When according to the relative spectral effectiveness, S(␭).
these measures are inadequate, topical sunscreens Spectroradiometric measurements of the source which
should be applied to the skin. However, the value of can then be used with the S(␭) weighting function to
sunscreens has been questioned, and an IARC Work- calculate Eeff are often necessary for measurements more
ing Group on the Evaluation of Cancer-Preventive accurate than those with simple, direct-reading safety
Agents concluded that there was inadequate epidemi- meters. Whichever measurement technique is applied,
ological evidence in humans for a cancer-preventive the geometry of measurement is important. All the
effect of topical use of sunscreen formulations against preceding ELs for UVR apply to exposures that are
cutaneous malignant melanoma, or basal-cell carci- measured with an instrument having a cosine-response
noma, despite the experimental evidence in animal detector oriented perpendicular to the most directly
studies (IARC/WHO 2001). exposed surfaces of the body when assessing skin expo-
When exposure is to artificial sources, as in some sure. The detector is oriented along (or parallel to) the
industrial hazard situations, engineering control mea- line(s) of sight of each exposed individual when assess-
sures are preferable to protective clothing, goggles, and ing ocular exposure. The use of UV film badges makes it
procedural safety measures. Glass envelopes for arc possible to integrate UV exposure on specific body sites
lamps will filter out most UVB and UVC. Where lengthy which move with respect to the UVR source (Diffey et al.
exposure to high power glass-envelope lamps and quartz 1977; Saunders and Diffey 1995); however, the spectral
halogen lamps will occur at close proximity, additional response of such film badges still does not accurately
glass filtration may be necessary (McKinlay et al. 1989). follow S(␭).
Light-tight cabinets and enclosures and UVR absorbing For outdoor exposure, environmental UVR mea-
glass and plastic shielding are the key engineering surements may be of limited use for individual dose
control measures used to prevent human exposure to assessment because of geometrically changing exposure
conditions and human behavioral considerations. Per- D.H. Sliney (USA), Chairman
J-P. Césarini (France)
sonal dosimeters must properly take into consideration F. R. de Gruijl (The Netherlands)
the exposed sites of the individual, time of exposure, sun B. Diffey (U.K.)
angle, etc. The Global UV Index can be a useful tool in M. Hietanen (Finland)
M.A. Mainster (USA)
educating persons who are outdoors as to the changing T. Okuno (Japan)
level of overhead UVR. It is, however, not very predic- P. Söderberg (Sweden)
B.E. Stuck (USA)
tive of ocular exposure since it is a measure of the
overhead UVR incident on a horizontal surface. Ocular
exposure is highly dependent upon ground reflectance REFERENCES
factors and the upper lid and brow-ridge block most Anders A, Petry H, Fleming C, Petry K, Brix P, Luke W,
overhead UVR (Sliney 1995). Groger H, Schneider E, Kiefer J, Anders F. Increasing
melanoma incidence: Putatively explainable by retrotrans-
posons—Experimental contribution of the Xiphophorine
CONCLUDING REMARKS Gordon-Kosswig Melanoma System. Pigment Cell Res
Greater attention should be paid to the potential 7:433– 450; 1994.
Anders A, Altheide H, Knalmann M, Tronnier H. Action
hazards of UVR exposure. The increasing socially driven spectrum for erythema in humans investigated with dye
solar exposure as well as the increasing use of artificial lasers. Photochem Photobiol 61:200 –205; 1995.
UVR sources is a cause for concern. In many popula- Andreassi K, Simoni S, Fiorini P, Fiamiani M. Phenotypic
tions, skin cancer incidence continues to rise, due in large characters related to skin type and minimal erythema dose.
Photodermatol 4:43– 46; 1987.
part to a poor appreciation of the risk among the general Armstrong BK, Kricker, A. How much melanoma is caused by
population. Reduction of risk by avoidance of needless sun exposure? Melanoma Res 3:395– 401; 1993.
sunlight exposure and by physical means of protection Armstrong BK, Kricker A. Cutaneous melanoma. Cancer
should be an important public health goal. Improved Surveys 19:219 –240; 1994.
Azizi E, Lusky A, Kushelevsky AP, Schewach-Millet M. Skin
educational programs are needed for school children, for type, hair colour, and freckles are predictors of decreased
outdoor workers and the general public. The present minimal erythema ultraviolet radiation dose. J Am Acad
understanding of injury mechanisms and long-term ef- Dermatol 19:32–38; 1988.
fects of exposure to UVR is incomplete, and awaits Bastiaens MT, ter Huuren JA, Kielech C, Gruis NA, Westen-
dorp RG, Vermeer BJ, Bavinck JN. Melanocortin-1 recep-
further research. The above guidelines will be subject to tor gene variants determine the risk of non-melanoma skin
periodic review and amendment as appropriate. cancer independently of fair skin and red hair. Am J Hum
Genet 68:884 – 894; 2001.
Berger D, Urbach F, Davies RE. The action spectrum of
Acknowledgments—The support received by ICNIRP from the Interna-
erythema induced by ultraviolet radiation (Preliminary Re-
tional Radiation Protection Association, the World Health Organization, port XIII). In: Jadassohn W, Schirren CG, eds. Proceedings
the International Labor Office, the European Commission, and the German of the Congressus Internationalis Dermatologiae-Munchen
Government is gratefully acknowledged. 1967. New York: Springer-Verlag; 1968: 1112–1117.
Brash DE, Rudolph JA, Simon JA, Lin A, McKenna GJ, Baden
HP, Halperin AJ, Ponten J. A role for sunlight in skin
cancer: UV-induced p53 mutations in squamous cell carci-
During the preparation of these guidelines, the composition of the noma. Proc Natl Acad Sci USA 88:10124 –10128; 1991.
International Commission on Non-Ionizing Radiation Protection was as
follows: Cesarini JP. Ultraviolet radiation: Biological effects and health
A.F. McKinlay, Chairman (UK) consequences. In: Matthes R, Bernhardt JH, Taki M, eds.
J.H. Bernhardt, Vice-chairman (Germany) Non-ionizing radiation, Proceedings of the 3rd International
A. Ahlbom (Sweden) Non-Ionizing Radiation Workshop, Baden (Austria), April
J-P. Césarini (France) 22–26, 1996. Munich: ICNIRP; 1996: 55–76.
F. R. de Gruijl (The Netherlands) CIE. Comptes Rendues de la Commission Internationale de
M. Hietanen (Finland) l’éclairage. Berlin: CIE; 9:596 – 625; 1935.
R. Owen (USA) CIE. The spectroradiometric measurement of light sources.
D.H. Sliney (USA)
P. Söderberg (Sweden)
Vienna: Commission Internationale de l’Eclairage; Pub. No
A.J. Swerdlow (United Kingdom) 63; 1984.
M. Taki (Japan) CIE. International lighting vocabulary. Vienna: Commission
T.S. Tenforde (USA) Internationale de l’Eclairage (International Commission on
P. Vecchia (Italy) Illumination); Publication CIE No 17 (E-l.l); 1987.
B. Veyret (France) CIE. Erythema reference action spectrum and standard ery-
R. Matthes, Scientific Secretary (Germany) thema dose. Vienna: CIE; 1998.
M.H. Repacholi, chairman emeritus (Switzerland) CIE. Erythemal reference action spectrum and standard erythe-
mal dose. Vienna: CIE; CIE Standard S007–1998; also
available as ISO 17166; 1999a.
During the preparation of this document, the composition of the ICNIRP CIE. Standardization of the terms UV-A1, UV-A2 and UV-B.
Standing Committee IV and task group was: Vienna: CIE; Report CIE-134/1; 1999b.
CIE. Action spectrum for photocarcinogenesis (non-melanoma Gallagher RP, Hill GB, Bajdik CD, Fincham S, Coldman AJ,
skin cancers). Vienna: CIE; CIE 138/2; 2000. McLean DI, Threlfall WJ. Sunlight exposure, pigmentation
CIE. Action spectroscopy of skin with tunable lasers. Publica- factors, and risk of nonmelanocytic skin cancer: II Squa-
tion Vienna: CIE; CIE 148; 2002. mous cell carcinoma. Arch Dermatol 131:164 –169; 1995b.
Clemens TL, Adams JS, Henderson SL, Holick MF. Increased Gange RW, Park YK, Auletta M, Kagetsu N, Blackett AD,
skin pigment reduces the capacity of skin to synthesize Parrish JA. Action spectra for cutaneous responses to
vitamin D3. Lancet i:74 –76; 1982. ultraviolet radiation. In: Urbach F, Gange FW, eds. The
Coblentz WW, Stair R, Hogue JM. The spectral erythemic biological effects of UV-A radiation. New York: Praeger;
reaction of the human skin to ultraviolet radiation. Proc US 1986: 57– 65.
Nat Acad Sci 17:401– 403; 1931. Gezondheidsraad (Health Council of the Netherlands). Recom-
Cooper KD, Oberhelman L, Hamilton TA, Baardsgaard O, mendations concerning acceptable levels of electromagnetic
Terhune M, LeVee G, Anderson T, Koren H. UV exposure radiation in the wavelength range from 100 nm to 1 mm
reduces immunization rates and promotes tolerance to (micrometre radiation). The Netherlands: Ministry of
epicutaneous antigens in humans: relationship to dose, Health and Environmental Protection; Report 65E; 1978.
CD1a- DR⫹ epidermal macrophage induction and Langer- Gies HP, Roy CR, Toomey S, MacLennan R, Watson M. Solar
hans cell deletion. Proc Natl Acad Sci USA 89:8497– 8501; UVR exposures of three groups of outdoor workers on the
1992. Sunshine Coast, Queensland. Photochem Photobiol
Cox NH, Farr PM, Diffey BL. The relationship between 62:1015–1021; 1995.
chronological age and erythemal response to UVB radia- Green A, Williams G, Neale R, Hart V, Leslie D, Parsons P,
tion. Br J Dermatol 122:272–273; 1990. Marks GG, Gaffney P, Battistutta D, Frost C, Lang C,
Cullen AP, Perera SC. Sunlight and human conjunctival action Russell A. Daily sunscreen application and beta-carotene
spectrum. Proc SPIE, 2134B:24 –30; 1994. supplementation in prevention of basal-cell and squamous-
Dahaw-Barker P. Ocular photosensitization. Photochem Pho- cell carcinomas of the skin: A randomised controlled trail.
tobiol 46:1051–1055; 1987. The Lancet 354:723–729; 1999.
De Gruijl FR. Health effects from solar UV radiation. Radiat Ham WT Jr., Mueller HA, Ruffolo JJ Jr., Guerry D III, Guerry
Protect Dosim 72:177–196; 1997. RK. Action spectrum for retinal injury from near-ultraviolet
De Gruijl FR, van der Leun JC. Estimate of the wavelength radiation in the aphakic monkey. Am J Ophthalmol 93:299 –
dependency of ultraviolet carcinogenesis in humans and its 306; 1982.
relevance to the risk assessment of a stratospheric ozone Hamerski W. Studies on the histochemical changes in experi-
depletion. Health Phys 67:319 –325; 1994. mental corneal lesions induced with ultraviolet radiation
Despres S. Effets biologiques des infrarouges et des ultravio- and on prevention of photophthalmia. Klin Oczna 39:537–
lets. Radioprotection 13:11–21; 1978 (in French). 542; 1969 (in Russian); English translation in Pol Med J
Diffey BL. Observed and predicted minimal erythema doses: a 8:1469 –1476; 1969.
comparative study. Photochem Photobiol 60:380 –382; Hausser KW. Influence of wavelength in radiation biology.
1994. Strahlentherapie 28:25– 44; 1928 (in German).
Diffey BL. Human exposure to ultraviolet radiation. In: Hawk Hausser KW, Vahle W. Sonnenbrand und Sonnenbraunung.
JLM, ed. Photodermatology. London: Chapman and Hall; Wissenschaftliche Veroffentlichungen des Siemens
1998. Konzern 6:101–120; 1927.
Diffey BL, Kerwin M, Davis A. The anatomical distribution of
Hawk JLM, Parrish JA. Responses of normal skin to ultraviolet
sunlight. Br J Dermatol 7:407– 409; 1977.
radiation. In: Regan JD, Parrish JA, eds. The science of
Diffey BL, Farr PM, Oakley AM. Quantitative studies on UVA
photomedicine. New York: Plenum; 1982: 219 –260.
induced erythema in human skin. Br J Dermatol 117:57– 66;
1987. Health Council of the Netherlands. UV radiation: Human
Elwood JM, Jopson J. Melanoma and sun exposure: an exposure to ultraviolet radiation. The Hague: Health Coun-
overview of published studies. Int J Cancer 73:196 –203; cil of the Netherlands; Report 1986/93; 1986.
1997. Hiller R, Giacometti L, Yuen K. Sunlight and cataract: An
English DR, Armstrong BK, Kricker A, Winter MG, Heenan epidemiological investigation. Am J Epidemiol 105:450 –
PJ, Randell PL. Case-control study of sun exposure and 459; 1977.
squamous cell carcinoma of the skin. Int J Cancer 77:347– Holick MF. Sunlight and vitamin D: The bone and cancer
353; 1996. connections. Radiat Protect Dosim 91:65–71; 2000.
Everett MA, Olsen RL, Sayre RM. Ultraviolet erythema. Arch IARC. Solar and ultraviolet radiation. Monographs on the
Dermatol 92:713–729; 1965. evaluation of carcinogenic risk to humans. Vol. 55, Solar
Ferguson J. Drug and chemical photosensitivity. In: Hawk and UV Radiation. Lyon: International Agency for Re-
JLN, ed. Photodermatology. London: Chapman and Hall; search on Cancer; 1992.
1998: 155–169. IARC/WHO. Sunscreens, IARC handbooks of cancer preven-
Fitzpatrick TB, Pathak MA, Harber LC, Seiji M, Kutika A. tion. Volume 5. Lyon: IARC; 2001.
Sunlight and man. Tokyo: University of Tokyo Press; 1974. ICNIRP. Guidelines on UV radiation exposure limits. Health
Fitzpatrick TB. Soleil et peau. J Med Esthet 2:33–34; 1975. Phys 71:978; 1996.
Freeman RG, Owens DW, Knox JM, Hudson HT. Relative ICNIRP. Guidelines on limits of exposure to optical radiation
energy requirements for an erythemal response of skin to from 0.38 to 3.9 ␮m. Health Phys 73:539 –554; 1997.
monochromatic wavelengths of ultraviolet present in the ICNIRP. Health issues of ultraviolet tanning appliances used
solar spectrum. J Invest Dermatol 47:586 –592; 1966. for cosmetic purposes. Health Phys 84:119 –127; 2003.
Gallagher RP, Hill GB, Bajdik CD, Fincham S, Coldman AJ, ICNIRP/CIE. Measurements of optical radiation hazards. A
McLean DI, Threlfall WJ. Sunlight exposure, pigmentary reference book based on presentations by health and safety
factors, and risk of nonmelanocytic skin cancer: I Basal cell experts on optical radiation hazards. Matthes R, Sliney D,
carcinoma. Arch Dermatol 131:157–163; 1995a. eds. Gaithersburg MD: ICNIRP/CIE; 1998.
ICNIRP/WHO/WMO/UNEP. Global solar UV index. A prac- Mukhtar H, Elmets CA. Photocarcinogenesis: Mechanisms,
tical guide. Geneva, Switzerland: WHO; 2002. models and human health implications. Photochem Photo-
Jeevan A, Brown E, Kripke ML. UV and infectious diseases. biol 63:355– 447; 1996.
In: Photoimmunology. Krutmann J, Elmets CA, eds. Ox- Nakasawa H, English D, Randall PL, Nakasawa K, Martell N,
ford: Blackwell Science; 1995: 153–163. Armstrong BK, Yamasaki H. UV and skin cancer: specific
Johnson BE. Drug and chemical photosensitization. In: The p53 gene mutation in normal skin as a biologically relevant
environmental threat to the skin. Marks R, Plewig G, eds. exposure measurement. Proc Natl Acad Sci 91:360 –364;
London: Martin Dunitz; 1992: 57– 65. 1994.
Jose JG, Pitts DG. Wavelength dependency of cataracts in Noonan FP, Recio JA, Takayama H, Duray P, Anver MR, Rush
albino mice following chronic exposure. Exp Eye Res WL, de Fabo EC, Merlino G. Neonatal sunburn and
41:545–563; 1985. melanoma in mice. Nature 413:271–272; 2001.
Kelly DA, Walker SL, McGregor JM, Potten CS, Young AR. NRPB. Health Effects from Ultraviolet Radiation. Report of an
SSR-induced immunosupression in humans has a lower Advisory Group on Non-ionizing Radiation. Documents of
dose-threshold than erythema. Photochem Photobiol the NRPB. National Radiological Protection Board Vol 13
67:46S; 1998. (1). Chilton, Didcot, Oxon: NRPB; 2002.
Olson RL, Sayre RM, Everett MA. Effect of anatomic location
Kelly DA, Young AR, McGregor JM, Seed PT, Potten CS, and time on ultraviolet erythema. Arch Dermatol 93:211–
Walker SL. Sensitivity to sunburn is associated with sus- 215; 1966.
ceptibility to ultraviolet radiation-induced suppression of Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J.
cutaneous cell-mediated immunity. J Exp Med 191:561– Cancer incidence in five continents. Lyon: IARC; Vol VII,
566; 2000. IARC Sci Publ 143; 1997.
Kraemer KH. Commentary—Sunlight and skin cancer: An- Parrish J, Anderson R, Urbach F, Pitts D. UV-A biological
other link revealed. Proc Natl Acad Sci USA 94:11–14; effects of ultraviolet radiation with emphasis on human
1997. responses to longwave ultraviolet. New York: Plenum
Kricker A, Armstrong BK, English DR. Sun exposure and Press; 1978.
non-melanocytic skin cancer. Cancer Causes and Control Parrish JA, Jaenicke KF, Anderson RR. Erythema and mela-
5:367–392; 1994. nogenesis action spectra of normal human skin. Photochem
Kricker A, Armstrong BK, English DR, Heenan PJ. A dose- Photobiol 36:187–191; 1982.
response curve for sun exposure and basal cell carcinoma. Paul B, Parrish J. The interaction of UV-A and UV-B in the
Int J Cancer 60:482– 488; 1995. production of threshold erythema. J Invest Derm 78:371–
Kurtin WE, Zuclich J. Action spectrum for oxygen-dependent 374; 1982.
near-ultraviolet induced corneal damage. Photochem Pho- Perdiz D, Grof P, Mezzina M, Nikaido O, Moustacchi E, Sage
tobiol 27:329 –333; 1978. E. Distribution and repair of bipyrimidine photoproducts in
Luckiesh ML, Holladay L, Tavlor AH. Reaction of untanned solar UV-irradiated mammalian cells. Possible role of
human skin to ultraviolet radiation. J Opt Soc Am 20:423– Dewar photoproducts in solar mutagensis. J Biological
432; 1930. Chemistry 275:26732–26742; 2000.
Mackenzie LA. The analysis of the ultraviolet radiation doses Pirie A. Formation of N-formylkynurenine in proteins from
required to produce erythemal responses in normal skin. lens and other sources by exposure to sunlight. Biochem J
Br J Dermatol 108:1–9; 1983. 125:203–208; 1971.
MacLaughlin JA, Holick MF. Aging decreases the capacity of Pitts D. The ocular ultraviolet action spectrum and protection
human skin to produce vitamin D3. Clin Invest 76:1536 – criteria. Health Phys 25:559 –566; 1973.
1538; 1985. Pitts DG. Ocular effects of radiant energy. In: Pitts DG,
Mainster MA. The spectra, classification, and rationale of Kleinstein RN, eds. Environmental vision. Stoneham, MA:
ultraviolet-protective intraocular lenses. Am J Ophthalmol Butterworth-Heinemann; 1993: 151–220.
102:727–732; 1986. Pitts DG, Tredici TJ. The effects of ultraviolet on the eye. Am
Ind Hyg Assoc J 32:235–246; 1971.
Marrot L, Belaidi JP, Chaubo C, Meunier JR, Perez P, Pitts DG, Cullen AP, Hacker PD. Ocular effects of ultraviolet
Agapakis-Causse C. An in vivo strategy to evaluate the radiation from 295 to 365 nm. Invest Ophthalmol Vis Sci
phototoxicity of solar UV at the molecular and cellular 16:932–939; 1977.
level: application to photoprotection assessment. Eur J Ponten F, Berne B, Ren Z-P, Nister M, Ponten J. Ultraviolet
Dermatol 8:403– 412; 1998. light induces expression of p53 and p21 in human skin:
McKinlay AF, Diffey BL. A reference action spectrum for Effect of sunscreen and constitutive p21 expression in skin
ultraviolet induced erythema in human skin. CIE J 66:17– appendages. J Invest Dermatol 105:402– 406; 1995.
22; 1987. Preece MA, Tomlinson S, Ribot CA, Pietrek J, Korn HT,
McKinlay AF, Whillock MJ, Meulemans CCE. Ultraviolet Davies DM, Ford JA, Dunnigan MG, O’Riordan JLH.
radiation and blue-light emissions from spotlights incorpo- Studies of vitamin D deficiency in man. Quart J Med
rating tungsten halogen lamps. Didcot, UK: National Ra- 44:575–589; 1975.
diological Protection Board; Report NRPB-R228; 1989. Rees JL. The melanocortin-1 receptor (MC1R): more than just
Merriam GC, Lofgren S, Michael R, Soderberg PG, Dillon J, red hair. Pigment Cell Res 13:135–140; 2000.
Zheng L, Ayala M. An action spectrum for UVB radiation Ringvold A. In vitro evidence for UV-protection of the eye by
and the rat lens. Invest Ophthalmol Vis Sci 41:2642–2647; the corneal epithelium mediated by the cytoplasmic protein,
2000. RNA, and ascorbate. Acta Ophthalmol Scand 75:496 – 498;
Michael R, Soderberg PG, Chen E. Dose-response function for 1997.
lens forward light scattering after in vivo exposure to Ringvold A, Davanger M, Olsen EG, Changes of the cornea
ultraviolet radiation. Graefe’s Arch Clin Exp Ophthalmol endothelium after ultraviolet radiation. Acta Ophthalmo-
236:625– 629; 1998. logica 60:41–53; 1982.
Roberts JE. Ocular phototoxicity. J Photochem Photobiol Webb AR, Kline LW, Holick MF. Influence of season and
64:136 –143; 2001. latitude on the cutaneous synthesis of vitamin D3: exposure
Robinson ES, Hill RH, Kripke ML, Setlow RB. The monodel- to winter sunlight in Boston and Edmonton will not promote
phis melanoma model: initial report on large ultraviolet A vitamin D3 synthesis in human skin. J Clin Endocrinol
exposures of suckling young. Photochem Photobiol 71:743– Metab 67:337–338; 1988.
746; 2000. Webb AR, DeCosta BR, Holick MF. Sunlight regulates the
Sasaki H, Jonasson F, Shui YB, Kojima M, Ono M, Katoh N, cutaneous production of vitamin D3 by causing its photo-
Cheng HM, Takahashi N, Sasaki K. High prevalence of degradation. J Clin Endocrinol Metab 68:822– 827; 1989.
nuclear cataract in the population of tropical subtropical West SK, Duncan DD, Muoz B, Rubin GS, Fried LP, Bandeen-
area. Dev Ophthalmol 35:60 – 69; 2002. Roche K, Schein OD. Sunlight exposure and risk of lens
Saunders PJ, Diffey BL. Ambulatory monitoring of ultraviolet opacities in a population-based study: The Salisbury eye
erythema in photosensitive subjects. Photodermatol Photo- evaluation project. JAMA 280:714 –718; 1998.
immunol Photomed 11:22–24; 1995.
Schmidt K. On the skin erythema effect of UV flashes. Wester U. Analytic expressions to represent the hazard ultra-
Strahlentherapie 124:127–136; 1964. violet action spectrum of ICNIRP and ACGIH. Radiat
Setlow RB, Grist E, Thompson K, Woodhead AD. Wave- Protect Dosim 91:231–232; 2000.
lengths effective in the induction of malignant melanoma. Willis I, Kligman A, Epstein J. Effects of long ultraviolet rays
Proc Natl Acad Sci 90:6666 – 6671; 1993. on human skin: photoprotective or photoaugmentative. J In-
Sherashov SG. Spectral sensitivity of the cornea to ultraviolet vest Dermatol 59:416 – 420; 1972.
radiation. Biofizika 15:543–544; 1977 (in Russian). Young AR, Walker SL. Protection given by sunscreens. Radiat
Sliney DH. The merits of an envelope action spectrum for Protect Dosim 91:265–269; 2000.
ultraviolet exposure criteria. Am Ind Hyg Assoc J 33:644 – Young RW. The family of sunlight-related eye diseases.
653; 1972. Optometry Visual Sci 71:125–144; 1994.
Sliney DH. UV radiation ocular exposure dosimetry. J Photo-
Ziegler A, Jonason AS, Leffell DJ, Simon JA, Sharma HW,
chem Photobiol B 31:69 –771; 1995.
Kimmelman J, Remington L, Jacks T, Brash DE. Sunburn
Sliney DH. Geometrical gradients in the distribution of tem- and p53 in the onset of skin cancer. Nature 372:773–776;
perature and absorbed ultraviolet radiation in ocular tissues. 1994.
Dev Ophthalmol 35:40 –59; 2002.
Sliney DH, Wolbarsht ML. Safety with lasers and other optical Zigman S. Ocular light damage. Photochem and Photobiol
sources: A comprehensive handbook. New York: Plenum 57:1060 –1068; 1993.
Press; 1980. Zuclich JA. Cumulative effects of near-UV induced corneal
Sliney DH, Krueger RR, Trokel SL, Rappaport KD. Photok- damage. Health Phys 38:833– 838; 1980.
eratitis from 193 nm argon-fluoride laser radiation. Photo- Zuclich JA. Ultraviolet-induced photochemical damage in
chem Photobiol 53:739 –744; 1991. ocular tissues. Health Phys 56:671– 682; 1989.
Soderberg PG. Experimental cataract induced by ultraviolet Zuclich JA, Kurtin WE. Oxygen dependence of near UV-
radiation. Acta Ophthalmol Suppl 196:1–75; 1990. induced corneal damage. Photochem Photobiol 25:133–
Tapaszto I, Vass Z. Alterations in mucopolysaccharide com- 135; 1977.
pounds of tear and that of corneal epithelium, caused by
ultraviolet radiation. Ophthalmologica (Additamentum) Zuclich JA, Taboada J. Ocular hazard from UV laser exhibiting
158:343–347; 1969. self-mode locking. Appl Opt 17:1482; 1978.
Taylor HR, West SK, Rosenthal FS, Munoz B, Newland HS,
Abbey H, Emmett EA. Effect of ultraviolet radiation on
cataract formation. New Engl J Med 319:1429 –1433; 1988. APPENDIX: RATIONALE FOR THE LIMITS OF
UNEP. Ultraviolet radiation. Environmental Health Criteria 14, EXPOSURE TO UVR
United Nations Environment Programme, World Health
Organization, International Commission on Non-Ionizing Background
Radiation Protection. Geneva: WHO; 1979. Comprehensive reviews of UVR effects have been
UNEP. Ultraviolet radiation. Environmental Health Criteria
160. United Nations Environment Programme, World
published in conjunction with the United Nations Envi-
Health Organization, International Commission on Non- ronment Program and the World Health Organization
Ionizing Radiation Protection. Geneva: WHO; 1994. (UNEP 1979, 1994), and the interested reader is referred
Urbach F. The ultraviolet action spectrum for erythema— to those documents in particular. The CIE and ICNIRP
history. In: Matthes R, Sliney D, eds. Measurements of
optical radiation hazards. Munich: International Commis-
also reviewed UVR effects and action spectra in a
sion on Non-Ionizing Radiation Protection; 1998: 51– 62. monograph on optical radiation measurements (ICNIRP/
Urbach F, Epstein JH, Forbes PD. UV carcinogenesis. In: CIE 1998). In addition, the International Agency for
Fitzpatrick TB, Pathak MA, Harber LC, Seiji M, Kutika A, Cancer Research (IARC) published a monograph on
eds. Sunlight and man. Tokyo: University of Tokyo Press;
1974: 259 –283.
UVR in 1992 (IARC 1992) and published a monograph
Valverde P, Healy E, Sikkink S, Haldane F, Thody AJ, on sunscreens more recently (IARC/WHO 2001). Fur-
Carrothers A, Jackson IJ, Rees JL. The Asp84Glu variant of thermore, the National Radiological Protection Board
the melanocortin-1 receptor (MC1R) is associated with (NRPB) has recently published a scientific review of the
melanoma. Hum Mol Genet 5:1663–1666; 1996.
Van der Leun JC, Stoop T. In: Urbach F, ed. The biological
health effects of UVR (NRPB 2002). The following
effects of UV radiation. Oxford: Pergamon Press; 1969: discussion is a brief review of those physical and
251–254. biological factors used to derive the UVR guidelines.
General approach DNA lesions and errors in repair (mutations). UVR

Life has evolved under the daily exposure to solar exposure produces specific types of DNA damage that
radiation. Although UVR is only about 5% of the solar result in “UV signature” mutations, which can be de-
spectrum that reaches the earth’s surface, it plays a tected in skin carcinomas (Brash et al. 1991). Such
significant biological role since individual photon ener- mutations in the p53 gene can serve as a biomarker for
gies are the greatest within the optical spectrum. These past exposure and future carcinoma risk (Nakasawa et al.
shorter-wavelength, higher energy photons have suffi- 1994; Ziegler et al. 1994). The commonly experienced
cient energy to produce photochemical alterations that “sunburn” (erythema) may be used as a marker for the
may initiate biological effects that are potentially injuri- presence of substantial UVR-induced DNA damage
ous (sometimes referred to as “actinic effects”). Both (Young and Walker 2000). Through DNA damage,
beneficial and unwanted photobiological effects result erythema is related to skin cancer; and a knowledge of
from UVR exposure. The critical organs for UVR expo- erythema-dose response and action spectra are of value
sure are the eye and the skin since they may be readily for developing guidance that might reduce risk from skin
exposed. cancer.
The approach taken in the development of these
guidelines was to limit exposure to preclude any signif- Acute responses of the skin—Erythema
icant acute photobiologic effects and reduce the risks for Erythema (i.e., the reddening of the skin as in
delayed effects from chronic exposure as much as pos- sunburn) is the most commonly observed direct effect
sible, based upon the best available evidence. Thresholds observed in the skin following exposure to UVR. UVR-
for observed bioeffects vary strongly with wavelength. induced erythema results from photochemical damage—
Consequently, various spectral dose-response relation- principally to DNA—that leads to a cascade of molecular
ships and time-dependent dose-response relations have events resulting in redness due to an increased blood
been developed. In photobiology, the term “action spec- content of the skin by dilatation of the superficial blood
trum” refers to the relative spectral effectiveness of vessels. Unlike the erythema from a thermal insult, the
different wavelengths in eliciting a biological effect. erythema from UVR is delayed by some (1– 6) hours
Available data on the action spectra and dose response after the exposure. The duration of the delay is reduced
curves for each delayed effect were reviewed with the as the exposure dose increases.
goal of estimating risk at exposure levels below those Erythema action spectra have been the subject of
producing acute effects. Furthermore, the dose and action experimental and theoretical interest for over 70 y
spectra for beneficial effects, such as vitamin-D synthe- (Urbach 1998); and upon first review, there appear to be
sis, were examined to assure that the exposure guidelines many different, and even contradictory, action spectra.
did not lead to an inadequate level of exposure. The apparent differences relate to different methods of
Both the acute skin response (erythema) and long- assessment, types of monochromatic sources, clinical
term risk of skin cancer appear to be related to DNA endpoint, time of assessment, etc. The International
damage (de Gruijl and van der Leun 1994; Cesarini Commission on Illumination (CIE) reference action
1996; Kelly et al. 2000). In theory, only a single UV spectrum E(␭) as shown in Fig. A1, is routinely used
photon is required to alter directly a single DNA mole- today to convert absolute UV exposure levels into
cule (or indirectly through free-radical production). erythemally effective irradiance (CIE 1998; McKinlay
However, DNA repair mitigates most single-molecule and Diffey 1987), but this is based upon specific refer-
events. An enormous number of DNA lesions are pro- ence conditions. An historical perspective is useful to
duced in one single cell nucleus at the basal epidermal understand the differences in published action spectra
layer in skin irradiated at one minimal erythemal dose. and the significance of these differences.
The three major categories of photo-lesions induced in Hausser and Vahle (Hausser 1928; Hausser and
single cell in culture (hamster ovary mesenchyme) have Vahle 1927) were the first to document quantitatively
been identified and calculated after exposure to a solar erythema as a wavelength-dependent effect in the late
simulated source. The delivered dose corresponded to the 1920’s. By 1935, the CIE had recommended an early
dose received on the skin basal layer from 2 h exposure “standardized” erythemal action spectrum based upon
to a UV-index 6 (Paris mid-summer). The total number several studies using a limited number discrete mono-
of lesions was around 100,000 photolesions in a single chromatic emission lines of the mercury lamp (e.g., 254
cell for one standard erythema dose (SED) (Perdiz et al. nm, 280 nm, 297 nm, 303 nm, 313 nm, etc.), and this
2000). In practice, laboratory techniques exist to detect limited the full spectral detail (Coblentz et al. 1931;
UVR-induced changes down to the cellular and molec- Hausser 1928; Luckiesh et al. 1930; Urbach 1998). With
ular level (e.g., Marrot et al. 1998). These can detect the development of xenon-arc lamps and their use with
and assessment (Willis et al. 1972; van der Leun and

Stoop 1969; Paul and Parrish 1982; Parrish et al. 1978;
McKinlay and Diffey 1987; CIE 1999). In addition, the
variation in published threshold values results from
differences in the clinical definition and estimate of
“minimal erythema” and radiometric measurement tech-
niques. Despite the steeper slope between 300 nm and
315 nm found in laser studies (Anders et al. 1995), Diffey
concluded from a mathematical analysis of the action
spectra obtained with monochromators and lasers that the
CIE reference action spectrum was a valid predictor of
the erythemal effectiveness of different wavelengths of
ultraviolet radiation (Diffey 1998).
Fig. A1. Erythemal action spectra. The CIE (1998) reference Exposure to UVA alone can produce erythema, but
action spectrum for erythema in human skin (solid line), an only at very high radiant exposures (i.e., at ⬎10 J cm⫺2,
erythema action spectrum (Anders et al. 1995) determined using i.e., 100 kJ m⫺2) as shown by more recent studies (Diffey
dye lasers (triangles), and the CIE (1935) action spectrum
(squares) are shown with the action spectrum of human skin
et al. 1987; Parrish et al. 1982; Anders et al. 1995).
adapted from Parrish et al. (1982) for 8 h after irradiation Multiple-wavelength exposures (as within broad-
(diamonds). If measured at 24 h, the MED differs below 300 nm. spectrum exposures such as from sunlight) are additive in
producing erythema. Deviations from additivity have
been reported, such as photoaugmentation (Willis et al.
monochromators in the 1960’s, several groups conducted 1972), or the opposite effect, photoprotection (Van der
studies to fill in the missing spectral information (Everett Leun and Stoop 1969; Paul and Parrish 1982), but their
et al. 1965; Freeman et al. 1966; Berger et al. 1968), but importance is considered marginal.
these appeared to differ somewhat from the “classical” Individuals can be grouped into one of six sun-
studies of the 1930’s—particularly at wavelengths less reactive skin types as shown in Table A1 (Fitzpatrick
than 300 nm. The short-wavelength variations were the 1975). These skin types fall into three more significant
result of different end-points (Urbach 1998; Hausser groups: melano-compromised, melano-competent, and
1928). The use of high-pressure xenon-arc lamps and melano-protected (ICNIRP 2003). Individual susceptibil-
xenon-mercury arc lamps with monochromators filled in ity to both acute and delayed effects varies markedly with
missing spectral regions, but also introduced a new skin phototype and exposure history. Skin pigmentation
problem, since the 5–10 nm spectral bandwidths used in (“tanning”) and “conditioning” (thickening of the stra-
these studies lowered the spectral resolution. Later, tum corneum and tanning) may result in an increase of an
highly monochromatic studies with lasers were able to individual’s minimal erythemal doses (MED) by a factor
refine the spectral detail in the 300 –320 nm spectral of at least four for a UVB source (Gange et al. 1986). For
region where the action spectrum was rapidly changing individuals with melanocompromised skin, sunburn and
(Anders et al. 1995). These and other quantitative studies tanning are obtained for a single exposure dose, and there
have confirmed that the erythemal threshold varies with is no tanning without burning. However, for individuals
anatomical site, wavelength, and time between exposure with melanocompetent skin, a significant tan may be
Table A1. Skin phototypes.

Phototype Skin response to sunlight Typical appearance
I Burns easily and severely (painful People with very fair skin, blue eyes,
burn); tans little or none and peels freckles; unexposed skin is white
II Usually burns easily and severely People most often with fair skin, red or
(painful burn); tans minimally or blond hair, blue, hazel or even brown
lightly, also peels eyes; unexposed skin is white
III Burns moderately and tans People with white skin when unexposed;
generally dark hair
IV Burns minimally, tans easily People with white or light brown
unexposed skin, dark hair, dark eyes
V Rarely burns, tans easily and People with brown skin
substantially
VI Never burns and tans profusely People with black skin
obtained without burning—most notably for UVA wave- approximately one-third MED sunburn cells and immune
lengths. Skin color and other phenotype characteristics suppressive effects can be detected (Jeevan et al. 1995;
(hair color, eye color, and freckles) are associated with Kelly et al. 1998). Table A3 summarizes the cellular
the susceptibility to sunburn (Andreassi et al. 1987; Azizi responses to increasing dose levels at different MED
et al. 1988). Because the MED varies with each individ- values.
ual, the CIE standard erythema dose (SED) unit was
introduced for objective UVR dosimetry of skin effects Long-term effects on the skin
(CIE 1998). Erythemal thresholds as reported in studies Chronic exposure to the UVR in sunlight accelerates
for untanned, lightly pigmented skin, range from about the skin aging process and increases the risk of develop-
1.5 to 3 SED, i.e., 15 to 30 mJ cm⫺2 as weighted by the ing skin cancer (NRPB 2002). The solar spectrum is
CIE standard action spectrum for erythema (CIE 1998; greatly attenuated by the earth’s ozone layer, limiting
Everett et al. 1965; Freeman et al. 1966; Parrish et al. terrestrial UV to wavelengths greater than approximately
1982; Cox et al. 1990; Diffey 1994). The ICNIRP 290 nm. The UVB irradiance at ground level is a strong
guideline values are therefore approximately 2 to 4 times function of the sun’s elevation angle in the sky. This
less than these MED values. Fig. A1 also illustrates the results from the change of UV attenuation with atmo-
results of one study of the variation of erythema action spheric path length (time of day and season). Several
spectrum (Parrish et al. 1982). The six sun-reactive skin ecological epidemiologic studies showed that the inci-
types shown in Table A1 (Fitzpatrick 1975; Andreassi et dence of skin cancer is strongly correlated with latitude,
al. 1987) were based on a personal history of response to altitude, and cloud cover (UNEP 1979). Exact quantita-
45– 60 min of exposure to midday summer sun in early tive dose-response relationships have not yet been estab-
summer. lished although fair-skinned melanocompromised indi-
There are anatomical differences in erythemal sen- viduals, especially of Celtic origin, are much more prone
sitivity. The face, neck, and trunk are two to four times to develop skin cancer. Since the discovery by Valverde
more sensitive than the limbs (Olson et al. 1966). There and associates of polymorphism in the alpha-melanocytic
is no difference in sunburn susceptibility between sexes. stimulating hormone (␣-MSH) receptor associated with
Although there have been suggestions that erythemal red-haired phenotypes and extreme photosensitivity, it
sensitivity may change with age, and that young children has also been shown that polymorphisms in this receptor
and elderly people are said to be more sensitive (Hawk are an important risk determinant for all types of skin
and Parrish 1982), quantitative studies of erythemal cancers (Bastiaens et al. 2001; Rees 2000; Valverde et al.
sensitivity in subjects of these age groups have not 1996).
confirmed this (Cox et al. 1990). Prior to 1970, skin cancer was typically a disease of
The MEDs in a given spectral waveband and for a outdoor workers such as farmers and seamen routinely
normal population have a positively skewed distribution exposed to sunlight (Urbach et al. 1974), however,
(Mackenzie 1983). Values for the MED should therefore because of the change in social activity, it has become a
be expressed as the median, or geometric mean, rather disease of the general public whose exposure is largely
than the arithmetic mean. Examples of MEDs determined intermittent from recreational exposure (Cesarini 1996).
in a population of 252 subjects (skin types I, II and III) This change is important in interpreting epidemiological
are given in Table A2 (Diffey 1994). studies of skin cancer because of the different exposure
Cellular damage can be detected at levels below the patterns. Only a few quantitative studies have examined
MED. At approximately 0.1 MED, it is possible to detect
activation of p53 protein and p21 gene expression, which
Table A3. Dose response values.
indicate a cellular response (Ponten et al. 1995). At
Exposure level Effect Reference
0.1 MED p53 and p21 activated Ponten 1995
0.3 MED Sunburn cells just detectable Cesarini 1996
Table A2. Examples of minimal erythemal doses. 0.3 MED Immunosuppressive effect in Kelly 1998
Central wavelength Bandwidth Median MED 95% range melanocompromised
nm (FWHMa) nm J cm⫺2 J cm⫺2 individuals
0.5 MED Modification and depletion of Cooper 1992
300 5 0.027 0.015−0.051 Langerhans cells
320 10 1.9 1.0−3.4 1.0 MED 20 sunburn cells/cm2 Cesarini 1996
330 15 5.6 3.1−10 1.0 MED Immunosuppressive effect in Kelly 1998
350 30 19 11−35 melanocompetent individuals
370 30 27 16−47 2 MED 150 sunburn cells/cm2 Cesarini 1996
400 30 62 38−102 3 MED 400−500 sunburn cells/cm2 Cesarini 1996
a 6−10 MED Blistering Everett 1965
Full-width at half-maximum.
indoor working populations chronically exposed to arti- exposure to sunlight. Whilst SCC is strongly related to
ficial sources of UVB to determine whether there is an cumulative lifetime exposure to sunlight, this relation-
increased skin cancer risk in this occupational environ- ship is not so convincing for BCC (Gallagher et al.
ment. Squamous cell carcinoma is the most common type 1995b), and it may be that intermittent sun exposure and
in the outdoor working population. This is localized at perhaps exposure in childhood and adolescence may be
exposed sites (e.g., hands and back of the neck) and this critical for establishing adult risk for BCC (Kricker et al.
is suggestive of the importance of total cumulative 1995; Gallagher et al. 1995b).
exposure. Studies of the incidence of melanoma in
outdoor workers show a lower incidence than for indoor Action spectrum for non-melanoma skin cancer
workers (Armstrong and Kricker 1993; IARC 1992; At present, an action spectrum for skin cancer can
UNEP 1994). only be obtained from animal experiments. The most
extensive investigations to date are those from groups at
Types of skin cancer Utrecht and Philadelphia. These workers exposed a total
The three common forms of skin cancer, listed in of about 1,100 hairless albino mice to 14 different
ascending order of severity, are basal cell carcinoma broad-band ultraviolet sources and by a mathematical
(BCC), squamous cell carcinoma (SCC), and malignant optimization process derived an action spectrum referred
melanoma (MM). SCC is also known as spindle-cell to as the Skin Cancer Utrecht-Philadelphia (SCUP)
carcinoma. Around 90% of skin cancer cases are of the action spectrum (de Gruijl and van der Leun 1994). The
non-melanoma variety (BCC and SCC) with BCCs being SCUP action spectrum is that for skin tumor induction in
approximately four times as common as SCCs. The hairless mice, a species with a thinner epidermis than
overall lifetime risk of any type of skin cancer varies with humans. By taking into account differences in the optics
ethnic status and geography, but, as an example, the of human epidermis and hairless albino mouse epidermis,
cumulative lifetime risk of developing MM is 1:90 for a the experimentally determined action spectrum for tumor
white American. This risk increases to 1:7 for SCC and induction in mouse skin can be modified to arrive at a
BCC in the same population (Parkin et al. 1997). postulated action spectrum for human skin cancer (de
Exposure to UVR is considered to be a major Gruijl and van der Leun 1994). The resulting action
etiological factor for all three forms of skin cancer (IARC spectrum resembles the action spectrum for erythema
1992). For basal cell carcinoma and malignant mela- (Fig. 1). The CIE has recently published a “standardized”
noma, neither the wavelengths involved nor the exposure action spectrum based upon this work (CIE 1999).
pattern that results in risk have been established with
certainty; whereas for squamous cell carcinoma, UVB Malignant melanoma
and probably UVA are implicated and the major risk During the past 40 years or so there has been an
factors seem to be cumulative lifetime exposure to UV increase of the order of a doubling in each decade in the
radiation and a poor tanning response. incidence of cutaneous malignant melanoma in white
populations in several countries. There exists an inverse
Squamous cell cancer relationship between latitude and melanoma incidence;
The evidence that exposure to solar radiation is the and this, plus many other factors, has been taken as
predominant cause of squamous cell cancer in man is evidence for a possible role of sunlight as a cause of
very convincing. These cancers occur almost exclusively malignant melanoma. However, this pattern is not always
on sun-exposed skin such as the face, neck and arms, and consistent. In Europe, for example, the incidence and the
the incidence is clearly correlated with geographical mortality rates in Scandinavia are considerably higher
latitude, being higher in whites in the more equatorial than those in Mediterranean countries. This inconsis-
areas of the world (Kricker et al. 1994). Recent epide- tency may reflect ethnic differences in constitutional
miological studies and a randomized trial suggest that factors and customs. Also, the unexpectedly low inci-
sun exposure in the 10 years prior to diagnosis may be dence in outdoor workers, the sex and age distribution,
important in accounting for individual risk of SCC and the anatomical distribution have pointed to a more
(Gallagher et al. 1995a; English et al. 1996; Green et al. complex association (Armstrong and Kricker 1994).
1999). There is now growing evidence that intermittent sun
exposure, mainly from recreational activities, is associ-
Basal cell cancer ated with increased risk of developing malignant mela-
The relationship between basal cell carcinoma and noma. Several studies have found that a history of
sunlight is less compelling, but the evidence is suffi- sunburn is associated with risk for melanoma develop-
ciently strong to consider it also to be a consequence of ment, although in these studies a potential for recall bias
exists (Elwood and Jopson 1997) and may be con- spectrum) between 220 and 310 nm does not vary as
founded by skin type. Studies of migrants have led to the greatly as in the case of erythema with the thresholds
suggestion that sun exposure in childhood and adoles- varying from 4 –14 mJ cm⫺2. Sliney and colleagues used
cence is a particularly critical period in terms of mela- an excimer laser to determine the photokeratitis threshold
noma risk. at 193 nm (Sliney et al. 1991). Corneal injury from UVA
wavelengths requires levels exceeding 10 J cm⫺2 (Ham-
Action spectrum for melanoma erski 1969; Pitts 1993; Sherashov 1977; Tapaszto and
At one time, the only data that existed for an action Vass 1969; Zuclich and Kurtin 1977; Zuclich 1980;
spectrum for melanoma induction were those obtained Cullen and Perera 1994).
from irradiating hybrids of a small tropical fish with
different wavelengths of UVR (Setlow et al. 1993). This Cataract
fish action spectrum suggested that all wavelengths of Wavelengths above 295 nm can be transmitted
UV radiation could be important in melanoma, unlike through the cornea and are absorbed by the lens. Pitts et
non-melanoma skin cancer; however, at least one attempt al. (1977) have shown that both transient and permanent
to replicate this action spectrum was unsuccessful opacities of the lens (cataracts) can be produced in
(Anders et al. 1994). More recent studies in transgenic rabbits and primates by exposure to UVR having wave-
mice (Noonan et al. 2001) and in monodelphis domestica lengths in the 295–320 nm band. Similar findings were
(Robinson et al. 2000) indicate that neonatal UV expo- reported for the rat (Soderberg 1990). Thresholds for
sure is most significant. In contrast to small UVB doses, transient opacities ranged dramatically with wavelength,
Robinson and colleagues also found that large doses of from 0.15 to 12.6 J cm⫺2. Thresholds for permanent
UVA to neonates could not produce tumors (Robinson et opacities were typically twice those for transient opaci-
al. 2000). Melanoma incidence is also extremely high in ties (Pitts 1993). Experimental methods cannot readily
xeroderma pigmentosa (X-P) patients, who lack the show a threshold, since a measure of increased scatter is
capacity to repair UVB induced damage (Kraemer 1997). difficult when there is a background level of scattering
The weight of current evidence now suggests that UVB (Michael et al. 1998). The action spectrum for UVR
is the primary risk factor for MM. induced cataract was recently confirmed in the rat by use
of a quantitative criterion for light scattering (Merriam et
Ocular effects—Photokeratoconjunctivitis al. 2000). Opacities from chronic exposure at lower
Short-wavelength UVR (␭ ⬍ 300 nm) is strongly levels has been very difficult to show experimentally
absorbed by the cornea and conjunctiva. Excessive ex- (Jose and Pitts 1985; Zigman 1993). However, several
posure of these tissues causes photokeratoconjunctivitis, epidemiological studies show an association between the
commonly referred to as “welder’s flash,” “arc-eye,” etc. incidence of cortical opacities with ambient UVB expo-
Several research groups have characterized the course of sure (Hiller et al. 1977; Taylor et al. 1988; West et al.
ordinary clinical photokeratitis (Pitts 1993) and the 1998). Sasaki has shown a clear correlation of different
cellular changes in ocular tissues (Ringvold et al. 1982). forms of cataract with latitude, but does not explicitly
The latent period varies inversely with the severity of link this with the change of UVR exposure with latitude
exposure ranging from 1⁄2 to 24 h but usually occurs (Sasaki et al. 2002), although it was speculated that both
within 6 –12 h. Conjunctivitis tends to develop more temperature and UVR could be etiologic factors (Sliney
slowly and may be accompanied by erythema of the 2002). A number of biochemical studies of the effects of
facial skin surrounding the eyelids. The individual has UV irradiation of lens proteins have led to the theory that
the sensation of a foreign body or sand in the eyes and UVA radiation is a causal factor in cataract (Pirie 1971;
may experience photophobia, lacrimation, and blepharo- Roberts 2001; Young 1994). However, it has been
spasm to varying degrees. The acute symptoms last from difficult to link UVA radiation with cataract either
6 to 24 hours and discomfort usually disappears within epidemiologically or experimentally.
48 h. Although exposure rarely results in permanent
ocular injury, the individual is visually incapacitated Retinal effects
during this 48-h period. Pitts and Tredici (1971) reported The cornea and crystalline lens normally sufficiently
threshold data for photokeratitis in humans for 10 nm shield the retina from acute effects from UVR exposure.
wavebands from 220 to 310 nm (Pitts 1993). The Normally, less than 1% of UVA reaches the retina,
guideline ELs between 200 nm and 305 nm are about 1.3 shorter UVB wavelengths being totally attenuated except
to 4.6 times less than the threshold for minimal change. in neonates (UNEP 1994). Upon removal of the crystal-
The maximum sensitivity of the human eye was found to line lens, Ham and colleagues (Ham et al. 1982) demon-
occur at 270 nm. The wavelength response (action strated acute retinal injury (photoretinitis) at levels of the
⫺2
order of 5 J cm at the retina. Photoretinitis at these persons (i.e., melanocompromised skin phototypes I and II)
wavelengths is covered by the ICNIRP guidelines for it varies from about 3 to 20 depending on the spectral
exposure to incoherent optical radiation (ICNIRP 1997). composition of the radiation. Since there may be more than
one target molecule (chromophore) involved in erythema
Envelope action spectrum (and therefore more than one erythemal action spectrum),
Clearly, the development of UVR exposure limits the effect of radiations of two widely differing wavelengths
for workers and the general population must consider in the 180 nm to 315 nm range may not be simply additive.
two risks. These are the risks of acute and chronic injury The EL should be used with caution in evaluating sources
to both the eye and skin. The literature indicates that such as the sun and fluorescent lamps, having a rapidly
thresholds for injury vary significantly with wavelength increasing spectral irradiance in the 300–315 nm range.
for each effect. In the UVB and UVC regions, an action Large errors can arise because of the difficulty in making
spectrum curve can be drawn which envelops the thresh- accurate spectral measurements of such sources in this
old data for exposure doses (radiant exposures) in the region (Sliney and Wolbarsht 1980; ICNIRP/CIE 1998).
range of reciprocity (Schmidt 1964; Zuclich 1980) for The EL may not provide adequate protection for
acute effects obtained from recent studies of minimal photosensitive individuals or for normal individuals ex-
erythema and keratoconjunctivitis. Reciprocity means posed concomitantly to chemical, pharmaceutical, or phyto-
that irradiance E and exposure duration t have a recipro- photosensitizers, and special precautions must be taken for
cal relation, and a constant product of E and t (i.e., such cases (Dahaw-Barker 1987; Ferguson 1998).
exposure) results in a given effect. This EL curve does Based upon current knowledge, the EL should prevent
not differ significantly from the collective threshold data significant acute effects and reduce the magnitude of
considering measurement errors and variations in indi- chronic skin effects by limiting life-long UV exposure. The
vidual response (Sliney 1972; Sliney and Wolbarsht action spectrum for each type of skin cancer is still debated,
1980). Although the safety factor is minimal for just- although most research suggested that at least squamous-
detectable increases in corneal scatter, it is believed to cell carcinoma is probably related (both directly and indi-
range from 1.5 to 2.0 for acute keratitis. The curve is also rectly) to UV-induced molecular damage to DNA, and the
well below the acute UVB cataractogenic thresholds action spectrum is similar to that of the erythemal action
(Merriam et al. 2000; Pitts 1993). Repeated exposure of spectrum. Indeed, a Technical Committee of the CIE has
the eye to potentially hazardous levels of UV is not proposed a tentative action spectrum for photocarcinogen-
believed to increase significantly the protective capabil- esis (CIE 1999). The Dutch Health Council (Health Council
ity of the cornea as does skin tanning and thickening of of the Netherlands 1986) was the first to propose envelope
the stratum corneum [although some recent studies show limits similar to the guidelines developed for acute daily
a detectable change in threshold (Ringvold 1997)]. Thus, exposure (up to 10 y duration) and reduced levels for longer
this EL is more readily applicable to the eye and must be periods to protect against chronic effects. These should have
considered a limiting value for that organ (Sliney 1972). the same action spectrum in the UVB and UVC. In many
Any accumulation of UVB and UVC exposures causing cases, occupational exposure to UVB adds to an individu-
photokeratitis is limited to about 48 h since the outer al’s non-occupational exposure to solar UVB.
corneal epithelial layers are replaced in about 48 h by the It is worthy of note that in addition to the direct hazard
normal repair process of this tissue. Some slight additiv- of UV exposure, very intense UVC sources (particularly of
ity of UVA exposures exists beyond 48 hours because of wavelengths less than 230 nm) may also produce hazardous
the deeper penetration of UVA rays (Zuclich 1980). The concentrations of ozone and nitrogen oxides from the air
additivity factors were considered in deriving the mag- and of phosgene gas in the presence of degreasers; thus,
nitude of the safety factor built into the guidelines. On many UV germicidal lamps now have quartz-glass enve-
the basis of acute effects, the safety factor for UVA lopes that block wavelengths below ⬃230 nm.
guidelines is large, varying from about 7 at 320 nm to
more than 100 at 390 nm. UVA radiation effects
Because of the wide variations in threshold values and Studies of skin and ocular injury action spectra (Fig. A2)
exposure history (conditioning) among individuals, these in the UVA spectral region (315–400 nm) show very similar
guidelines should only be used as a starting point for thresholds for acute injury (Anders et al. 1995; McKinlay and
evaluating skin hazards (Despres 1978; Gezondheidsraad Diffey 1987; Parrish et al. 1982; Pitts et al. 1977; Zuclich
1978; NRPB 2002; Sliney and Wolbarsht 1980; UNEP 1989). These data are sufficient to define the relative spectral
1994). The envelope guideline has some margin of safety to effectiveness, S(␭), for exposure guidelines up to 400 nm.
protect all but the most sensitive individuals. An exact value However, if radiant energy were to be delivered to the skin or
for this margin cannot be given, but for lightly pigmented ocular tissues sufficiently fast for a substantial temperature
Fig. A2. Ocular action spectra. The ICNIRP UV guideline for exposure is depicted by the shaded, solid line. The data
for primate cornea of Pitts and Tredici (1971) are symbolized by a line with a closed circle, ●, of Kurtin and Zuclich
(1978) by a line containing an open circle, 䡬, and of Zuclich and Taboada (1978) by a line containing a closed square,
f. The data for rabbit cornea of Pitts and Tredici (1971) are represented by a line containing a closed diagonal square,
䉬, and of Pitts et al. (1977) by a line containing an open triangle, ‚. The human cornea data of Pitts (1973) are shown
by a line with an open square, e, and human conjunctiva data (Cullen and Perera 1994) by a line with a closed triangle,
Œ, but the outlier data point at 320 nm apparently resulted from thermal effects or experimental problems, as it is totally
inconsistent with environmental experience. Each data point was plotted after adjustment for spectral bandwidth used
for each exposure (Sliney and Wolbarsht 1980). A single 193-nm laser threshold point 1 J cm⫺2 is not shown.
increase, a thermal effect could result (Sliney and Wolbarsht experimental data, the Commission recommends a more
1980) at radiant exposures less than those required for photo- cautious approach for chronic ocular exposure.
chemically induced injury. However, few industrial sources In recent years there has been a rapidly growing
emit sufficient intensity in the UVA spectral region to cause population of individuals who have had one or both
adverse biologic effects, and only lasers may place tissues at crystalline lenses surgically removed as part of cataract
thermal risk. Nevertheless, a limit of 1 J cm⫺1 will protect surgery. Most of these patients have received artificial
against such effects. intraocular lenses of plastic. (Such individuals are fre-
There is a lack of evidence that the low levels of quently referred to as “pseudophakics”). Aside from a
UVA (of the order of 1–3 mW cm⫺2 or less) experienced few with implants that were not designed to absorb UVA
in sunlight or found in most indoor work environments to simulate the crystalline lens, or persons with no
present a hazard to either skin or eye. However, the implant (“aphakics”), all of these patients will be ade-
hypothesis originating from in vitro studies that UVA quately protected against retinal injury from UVA expo-
may be one causative agent for cataract (Roberts 2001), sure at the EL (Mainster 1986). Those without UV-
suggests the need for caution with regard to chronic absorbing IOLs should be fitted with UVA protective
low-level UVA ocular exposure. The EL for UVA should eyewear if working with sources of UVA radiation.
protect against potential photochemical injury; however,
experimental threshold data are lacking. In the absence of f f

ICNIRPUV2004

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

ICNIRPUV2004

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ICNIRPUV2004

Uploaded by

Copyright:

Available Formats

INTERNATIONAL COMMISSION ON NON-IONIZING RADIATION PROTECTION

PUBLISHED IN: HEALTH PHYSICS 87(2):171-186; 2004

ICNIRP PUBLICATION – 2004

GUIDELINES ON LIMITS OF EXPOSURE TO ULTRAVIOLET

The International Commission on Non-Ionizing Radiation Protection*

INTRODUCTION radiation, so that they may serve as guidance to the

Table 1. UV exposure limits and spectral weighting function.

Exposure of the eyes For the region

Exposure of the skin For the region

General approach DNA lesions and errors in repair (mutations). UVR

and assessment (Willis et al. 1972; van der Leun and

Table A1. Skin phototypes.

You might also like

ICNIRPUV2004

Uploaded by

Document Informationclick to expand document informationBueno

Document Informationclick to expand document information

Copyright:

Available Formats

ICNIRPUV2004

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ICNIRPUV2004

Uploaded by

Copyright:

Available Formats

INTERNATIONAL COMMISSION ON NON-IONIZING RADIATION PROTECTION

PUBLISHED IN: HEALTH PHYSICS 87(2):171-186; 2004

ICNIRP PUBLICATION – 2004

GUIDELINES ON LIMITS OF EXPOSURE TO ULTRAVIOLET

The International Commission on Non-Ionizing Radiation Protection*

INTRODUCTION radiation, so that they may serve as guidance to the

Table 1. UV exposure limits and spectral weighting function.

Exposure of the eyes For the region

Exposure of the skin For the region

General approach DNA lesions and errors in repair (mutations). UVR

and assessment (Willis et al. 1972; van der Leun and

Table A1. Skin phototypes.

You might also like