diffey: ultraviolet radiation physics and the skin · pdf fileregion of the solar spectrum...

Phys . Med . Biol., 1980. V01 . 25 . No . 3. 405-426 . Printed in Great Britain

Review Article

Ultraviolet radiation physics and the skin

B L Diffey: Medical Physics Department. Kent and Canterbury Hospital. Canterbury CT1 3NG . England

This review was completed in June 1979

Contents

1 . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 . Physics and measurement of natural and artifical uv sources

2.1. The nature of ultraviolet radiation . . . . . . . . . . . . . . . . . . . . .

2.2. The production of ultraviolet radiation . . . . . . . . . . . . . 2.3. Artificial sources of ultraviolet radiation used in photobiology . . . 2.4. Solar radiation . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Measurement of ultraviolet radiation . . . . . . . . . . . . . . 2.6. Ultraviolet radiation dosimetry . . . . . . . . . . . . . . . . 2.7. Isolation of spectral regions . . . . . . . . . . . . . . . . . .

3 . Theskin . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. The structure of the skin . . . . . . . . . . . . . . . . . . . 3.2. Optics of the skin . . . . . . . . . . . . . . . . . . . . . .

4.1. Ultraviolet erythema . . . . . . . . . . . . . . . . . . . . . 4.1.1. The mechanism of u v ~ erythema . . . . . . . . . . . . 4.1.2. The action spectrum for erythema . . . . . . . . . . . .

4.2. Melanin pigmentation . . . . . . . . . . . . . . . . . . . . 4.3. Production of vitamin D . . . . . . . . . . . . . . . . . . . 4.4. Aging of the skin . . . . . . . . . . . . . . . . . . . . . . 4.5. The carcinogenic nature of ultraviolet radiation . . . . . . . . . 4.6. Models of skin cancer incidence . . . . . . . . . . . . . . . .

5 . The current involvement of medical physics with ultraviolet radiation . . 5.1. Protection against ultraviolet radiation . . . . . . . . . . . . . 5.2. Photochemotherapy . . . . . . . . . . . . . . . . . . . . .

4 . The biological effects of ultraviolet radiation in normal skin . . . . . .

1 . Introduction

405 406 406 407 408 409 411 412 412 414 414 415 416 416 417 418 418 419 419 419 420 42 1 421 422

The role of sunlight in the causation of biological effects in the human skin has been evident for many centuries . The importance of sunlight for the maintenance of health was realised by the ancient Assyrians. Babylonians. Egyptians. Greeks and Romans. and the worship of the sun as a health-bringing deity was practised by the ancient Germans and by the Incas of South America (Ellinger 1957) . However. it is only in the

f Present address: Regional Medical Physics Department. Durham Area Unit. Dryburn Hospital. Durham DH1 5TW. England .

0031-9155/80/030405 +22$01.50 @ 1980 The Institute of Physics 405

406 B L Diffey

last hundred years or so that the scientific investigation of the effects of light, both beneficial and harmful, has begun.

In 1666 Isaac Newton '. . . procured me a Triangular glass-Prisme, to try therewith the celebrated Phaenomena of Colours' and thereby explained the polychromatic nature of light. It was not until 1801 that Johann Ritter discovered the ultraviolet region of the solar spectrum (Meyer 1952) and showed that chemical action was caused by some form of energy in the dark portion beyond the violet. Modern ultraviolet photobiology started with the work of Niels Finsen, who by sound experimentation during the years 1893-96 proved that sunburn was caused by the ultraviolet radiation (UVR) of the sun's spectrum and not by the radiant heat, as the name implies. Although a clinician, whose name is especially associated with the first successful therapy for Lupus vulgaris, his approach was very much that of a scientist. His goal was the efficient exploitation of UVR therapy in medicine, but he also spent much time examining the basic physical characteristics of UVR (Magnus 1978).

Following the pioneering work of Finsen, the early part of the twentieth century saw the rapid expansion of heliotherapy and actinotherapy throughout Europe and the United States of America. In his book on sunlight and health published in 1926, Saleeby discusses with much enthusiasm the benefits of heliotherapy and in particular describes the work of Dr Rollier at Leysin in the Alpes Vaudoises, whose experience in twenty years of successful treatment of surgical tuberculosis by natural sunlight has included '. . . many extreme cases of spinal tuberculosis, with paralysed lower limbs, tuberculosis of every other part of the body, of course, including the lungs, rickets, many skin diseases, varicose ulcers of the longest standing, wounds of war, non-healing operative wounds, osteomyelitis, bed-sores and so on' (Saleeby 1926). In the preface to the third edition of his book Saleeby notes the formation of the Sunlight League in London in 1924 and states as one of its aims "the education of the public to the appreciation of sunlight as a means of health; teaching the nation that sunlight is Nature's universal disinfectant, as well as a stimulant and tonic.' It is interesting to contrast this statement with the view of many leading dermatologists at the VIIth International Congress of Photobiology held in Rome in 1976, where it was maintained that excessive or even moderate exposure to sunlight was harmful. It was suggested that when we get up in the morning, after we have cleaned our teeth, we apply our sun barrier lotion!

The interaction of ultraviolet radiation in the skin leading to a macroscopic biological effect encompasses several seemingly unconnected areas of knowledge; from meteorology through photophysics, photochemistry and cellular biology to clinical dermatology and oncology. It is not appropriate or possible in the present review to include the photochemistry and cellular biology associated with uv interactions in the skin but for a good, recent monograph see Jagger (1967). Instead the review will limit itself to the beginning and end of the chain of events: the specification of UVR sources and optics of the skin, and the observable biological effects.

2. Physics and measurement of natural and artificial uv sources

2.1. The nature of ultraviolet radiation

Ultraviolet radiation is part of the electromagnetic spectrum and lies between the visible spectrum and the x-ray region. The Commission Internationale de 1'Eclairage (1970), amongst others, has divided the wavelengths between 400 and 100 nm into


three regions:

407

UVA 400-315 nm

UVB 3 15-280 nm

uvc 280-100 nm.

These regions have widely differing physical properties and potential for causing biological damage. UVR in the region 200 to 100 nm is readily absorbed in air and so has little opportunity to produce biological effects, apart from the indirect effects resulting from the production of ozone and of oxides of nitrogen in the air. In the case of mercury vapour lamps with transparent vitreous silica envelopes, most the ozone production is probably due to the 185 nm line (Koller 1952).

2.2. The production of ultraviolet radiation

Ultraviolet radiation may be produced either by the heating of a body to an incandescent temperature or by the excitation of a gas discharge (Henderson and Marsden 1972).

A body heated to a high temperature radiates as a result of its constituent particles becoming excited by numerous interactions and collisions. For a perfect black body the power radiated at any wavelength from unit surface area of such a body is determined solely by its temperature in accordance with Planck's law, which is

A A '[exp(B/AT) - l] M e , =

where M,, is the spectral radiant exitance at wavelength A , T is the absolute temperature of the radiator, and A and B are constants.

As the temperature is raised, not only does the maximum power radiated increase rapidly, but the peak of the emission curve moves to a shorter wavelength. The sun, of course, is the most celebrated source of incandescent UVR. However artificial incandescent sources are not efficient emitters of UVR: the ultraviolet emission from a general purpose tungsten filament lamp is only 0.08% of the rated power for a 40 W lamp, rising to 0.1% for a 100 W lamp and 0.17% for a 1 kW lamp (Summer 1962).

In a discharge tube, across which there is an electric field, the electrons drift towards the anode and the positive ions towards the cathode. In a low pressure discharge (a few Torr) such as occurs in a fluorescent lamp for example, one of three events may take place when a free electron collides with a neutral gas atom. The electron may undergo an elastic collision, the atom may be excited or the atom may be ionised. In a fluorescent lamp containing a mixture of mercury and argon, the mercury is pre- ferentially ionised since its ionisation potential (10.4 eV) is lower than that of argon (15.7 eV). The inert gases present in most practical discharge lamps act to reduce ion losses to the wall by ambipolar diffusion, to control the mobility of the electrons, to provide easier breakdown at a lower striking voltage and to prolong the life of the electrodes by reducing sputtering and evaporation (Henderson and Marsden 1972).

For a discharge to operate in a steady condition the rate of ionisation must exactly balance the rate of loss of electrons and ions by ambipolar diffusion to the walls. The consequence of this is that there is no simple relationship between voltage and current. The electrical characteristic of the discharge is very complex and dependent on all the

408 B L Diffey

constituents of the discharge and conditions of operation. In general the discharge has a ‘falling characteristic’: the volt-ampere curve has a negative slope (Koller 1952).

Most of the radiation from the majority of discharge lamps is from the positive column, i.e. the large uniform region of the discharge between the electrodes. The energetic electrons which produce the ionisation also produce excitation of the gas atoms which subsequently radiate at their characteristic frequencies. Figure 1 shows a few of the excitation and radiating transitions in lamps containing mercury vapour.

Ionlsailon Excltoilon Radlatlon

7 73 - > a - 5 L 6 x L 89 P L 6 7 C W

0

Green Blue Vlolet 5L61 L35 8 LOL 7

A

t ‘ H I 1

I

Ultravloiet

radlatlon resonance

2 5 3 7nm

I nm Exclted states

Ground state

Figure 1. Simplified energy transition diagram for mercury (from Henderson and Marsden 1972).

As the pressure in a discharge tube is raised to a few atmospheres, two principal changes occur:

(1) the gas temperature increases due to the increasing number of collisions (mainly elastic collisions) with the energetic electrons; and

(2) the high temperature becomes localised at the centre of the discharge, there now being a temperature gradient towards the walls, which are much cooler.

The wall becomes much less important at high pressures, and not altogether essential: discharges can operate between two electrodes without any restraining wall, and are then referred to as arcs. At high pressures the characteristic lines present in the low pressure discharge spectrum broaden to give a virtually continuous spectrum.

For a much more detailed description of lamp performance and technology the reader is referred to the brief article by Beeson (1978) or to the textbooks by Henderson and Marsden (1972), Summer (1962) and Koller (1952).

2.3. Artificial sources of ultraviolet radiation used in photobiology

The most common artificial sources of UVR currently used in photobiology are the xenon arc, mercury vapour arcs operating at various pressures and mercury discharge lamps used with or without fluorescent coatings. For a historical review of the artificial production of UVR see Schafer (1969).

The spectral emission from a xenon arc lamp operating at several atmospheres is similar to that of the uv component of the solar spectrum. An apparatus consisting of a 150 W xenon lamp, collecting optics, spectral shaping components and a refocusing lens

Ultraviolet radiation physics and the skin 409

has been developed as a ‘solar simulator’ (Berger 1969) and has been used for phototesting human skin (Urbach 1969a). The output spectrum has 70% of its energy at wavelengths below 400 nm and resembles the spectral distribution of global UVR for a 70” sun elevation. Xenon lamps are also used in conjunction with monochromators for action spectrum studies (Cripps and Ramsay 1970, MacKenzie and Frain-Belll973, Magnus et a1 1959, Sayre et a1 1965, Satoh et al 1974).

Medium-pressure mercury vapour arc lamps are commonly available in hospital physiotherapy departments; the Hanovia Sunlamp and the Kromayer lamps are typical. The Kromayer lamp is water-cooled to allow contact with the skin and consequently short exposure times. Both of these lamps emit copious amounts of energy in the UVB and uvc. Recently two clinical irradiation units incorporating high-pressure mercury lamps have been described. Alsins et a1 (1975) describe a water-cooled 500 W lamp used in conjunction with cut-off and interference filters, whilst an apparatus consisting of a 200 W super-pressure mercury lamp and a flexible light guide with a liquid light-conducting core is presented by Plewig et a1 (1978).

Low-pressure mercury discharge lamps with no phosphor coating have little potential in dermatological photobiology (Magnus 1976). These lamps emit a line spectrum with more than 90% of the radiated energy at a wavelength of 253.7 nm. This has resulted in an almost monochromatic lamp and as such it has been used for much microbiological work. However since the 253.7 nm radiation is not contained in terrestrial sunlight, of which the lower wavelength limit is about 295 nm, it is usually considered irrelevant to diagnostic and clinical photobiology.

The fluorescent lamp is a low-pressure mercury discharge lamp but with a phosphor coating on the inside of the lamp envelope. The chemical nature of the phosphor, which is excited by the 253.7 nm line, determines the spectral emission from the lamp. The fluorescent sunlamp, which emits a continuous spectrum from about 270 to 400 nm with a fluorescent emission peak at 310 nm (Mullen et a1 1975), is used extensively in actinotherapy where the normal sunburn wavelengths are required. More recently a fluorescent lamp with a continuum from about 315 to 400 nm accompanied by several lines in the uv and visible (Diffey and Challoner 1978) has been employed in whole body irradiation units for the photochemotherapy of psoriasis (see, for example, Parrish et a1 1974).

The emission spectrum and clinical suitability of many of the light sources referred to in this section have been summarised in a report by the Task Force on Photobiology of the National Program for Dermatology (1974).

2.4. Solar radiation

The radiant energy received from the sun is responsible for the development and continued existence of life on earth. The spectral distribution as well as the total amount of energy reaching the earth’s surface are both important factors in our environment.

The intensity of the solar radiation on a surface normal to the sun’s direction, outside the earth’s atmosphere and at the earth’s mean distance from the sun is called the solar constant. Recent determinations of the solar constant have yielded a value of 1.351 k0.028 kWm-2 (Thekaekara 1970). About two-thirds of this energy actually reaches the surface of the earth, the remainder being reflected, scattered or absorbed in the atmosphere. Of this energy about 50% lies in the visible spectrum and less than 5 % is in the ultraviolet (Koller 1952).

410 B L Diffey

The solar radiation which reaches the surface of the earth consists of a direct component (sunlight) and a diffuse component (skylight). The total radiation, sunlight plus skylight, is designated global radiation.

The global ultraviolet radiation is attenuated in the atmosphere, principally by the following effects (Gates 1966):

(a) Absorption by atmospheric ozone which is concentrated in a layer between 10 and 50 km above sea level with a concentration maximum of about 10 parts per million at an altitude of about 25 km (Henderson 1970). The total amount of atmospheric ozone is variable but generally equivalent to a layer about 0.3 cm thick at standard temperature and pressure (STP). The absorption of UVR by ozone is important for wavelengths less than 330 nm where the values of the ozone absorption coefficient increase rapidly with decreasing wavelength so that there is practically no radiation with wavelengths less than 295 nm which reach the earth’s surface.

(b) Rayleigh scattering caused by oxygen, nitrogen and other molecular components of the atmosphere where the scattering particle is small compared with the wavelength of the radiation.

(c) Mie scattering caused by dust, aerosols, water droplets and other particles of diameter comparable to the wavelength of the radiation.

In addition to the above effects, the degree of cloud cover and ground reflection will also affect the diffuse component of the global radiation (Robertson 1972).

It is evident that the theoretical determination of the spectral distribution of global UVR is extremely complex if all the above factors are to be considered. In particular, the computation of multiple scattering of sunlight by a molecular atmosphere is mathematically very difficult (Chandrasekhar 1950, Deirmendjian and Sekera 1954). More recently, theoretical radiative transfer calculations which lead to a computer model of the uv environment have been described by Green et a1 (1974). McCullough and Porter (1971) have described a theoretical approach to the calculation of global radiation at 11 1 wavelengths over the entire solar spectrum from 290 to 4000 nm. A set of empirical equations based upon experimental measurements of global radiation (Bener 1972) have been developed by Diffey (1977) to enable approximate values of the spectral distribution of natural UVR to be calculated for any time of day, day of year and geographical latitude.

Table 1. Approximate values for the relative erythemal effectiveness of sunlight at different months of the year in the United Kingdom.

Month Relative erythemal effectiveness

January February March April May June July August September October November December

0.02 0.07 0.27 0.62 0.89 1.00 0.96 0.69 0.40 0.16 0.04 0.01


Reviews of the experimental methods of determining global radiation have been given by Henderson (1970) and Stair (1969), whilst Drummond and Wade (1969) have discussed the instrumentation needed for the measurement of solar UVR.

If the spectral distribution of global radiation is multiplied by the International Commission on Illumination (ICI) erythema action spectrum (Coblentz and Stair 1934, IC1 1935) then it is found that the maximum of the resultant curve lies very close to 307.5 nm for both summer and winter values (Schulze and Grafe 1969). These authors have tabulated monthly sums of global UVR at 307.5 nm which are based upon measurements by Bener (1972). Interpolation of their results for the United Kingdom (geographical latitude 50" to 59"N) yields the values given in table 1 for the relative erythemal effectiveness of sunlight at different months in the year. From these data it is apparent that sunlight is about 100 times more effective in producing sunburn in the summer than in the winter.

2.5. Measurement of ultraviolet radiation

Techniques for the measurement of UVR may be divided into three classes: physical, chemical and biological. In general, physical devices measure dose rate, although exceptions are thermoluminescent dosemeters (Nambi and Higashimura 1971, McCullough et a1 1972, Bassi et a1 1975) and polymer film dosemeters (Davis et a1 1976, Diffey et a1 1977, Diffey and Davis 1978). The use of chemical methods, which measure the chemical change produced by the radiation, is called actinometry. The standard method is that due to Hatchard and Parker (1956) and uses potassium ferrioxalate as the liquid actinometer. Biological techniques of measurement are sometimes useful, and are generally limited to the use of viruses (Jagger 1967).

The most fundamental physical instrument for measuring radiant power is the thermopile (Gillham 1961) which provides the basis for calibrating all other types of uv measurement systems. The principle of operation of the thermopile is based on the Seebeck or thermoelectric effect whereby an EMF is generated when heat is applied to the junction of two dissimilar metals. Construction of practical devices of this nature was first described by Schwarz (1952), and more recently a device for measuring the irradiance from extended, linear uv sources has been described (Diffey and Challoner 1978). Byrne and Farmer (1972) have developed a self-calibrating black body radiometer which operates on the principle of balancing a Wheatstone bridge circuit, of which the radiometer forms one arm.

Photoelectric detectors, unlike thermopiles, do not have a linear wavelength response, but they have a much faster response, usually a higher sensitivity for certain spectral regions and are less prone to mechanical shock. The simplest of detection or measurement devices is the photovoltaic cell. These detectors have formed the basis of portable photometers (Latarjet et a1 1953, Jagger 1961, Ruff 1970).

Some instruments which have been considered more reliable and sensitive for routine use have relied upon vacuum phototubes and photomultiplier tubes. Robert- son (1969, 1972) has developed an instrument incorporating a phototube with appropriate correction filters and phosphor to approximately match the erythema action spectrum for use as an 'erythema monitor' for sunlight. A vacuum photodiode has been incorporated into an instrument (the IL730A uv actinic radiometer, International Light Inc, Newburyport, USA) which, with suitable filters, matches the wavelength response of the detector to the 'hazard action spectrum' for occupational exposure to UVR (NIOSH 1972). This instrument has been used by Hughes (1978) to make an

412 B L Diffey

objective assessment of the potential hazard of several items of laboratory equipment which emit UVR.

Photoconductive cells have also been used as uv detectors. The cheapest is cadmium sulphide, which when suitably prepared, is an insulator in the dark but becomes a conductor when light falls upon it. The principal disadvantage of CdS for use in the uv is its high sensitivity in the visible and near infrared compared to the ultraviolet (Mullard 1969). Schottky barrier type silicon photodiodes have found more application as uv detectors in, for example, so-called PUVA meters (e.g. the IL442A phototherapy radiometer, International Light, Inc). More recently GaAsP photodiodes have been developed which, unlike CdS and Si photodiodes, have no infrared response. A simple combination of such a photodiode with a ‘black glass’ filter (e.g. Schott UG1 glass filter, Jenaer Glaswerk Schott and Gen, Mainz, W Germany) can result in a uv selective detector (e.g. the ‘Uvichek’, Rank Hilger, Margate, England).

2.6. Ultraviolet radiation dosimetry

In. uv photobiology it is more eustomary to use the terminology of radiometry rather than that of photometry, since, strictly speaking, a source that emits only UVR has a zero intensity in photometric terms (Jagger 1967).

Terms relating to a beam of radiation passing through space are the ‘radiant energy’, the ‘radiant flux’ and the ‘radiant flux density’. Terms relating to a source of radiation are the ‘radiant intensity’ and the ‘radiance’. The term ‘irradiance’, which is most commonly used in photobiology, relates to the object struck by the radiation. These terms together with appropriate formulae and units are listed in table 2.

Table 2. Radiometric terms and units

Pertains to Term Formula Unit

Beam Radiant energy E J Beam Radiant flux P = dE/dt W Beam Radiant flux density dP/(dA cos 0) W cm-’ Source Radiant intensity d P j d n W sr-l Source Radiance dP/(dR dA cos 8) W sr” cm-’ Object Irradiance dPJdA W cm-’

The time integral of the irradiance is termed the ‘dose’ and is normally expressed in units of J cm-’ or multiples thereof. The term ‘dose’ is photobiology is analogous to the term ‘exposure’ in radiobiology and not to ‘absorbed dose’. As yet the problems of estimating the energy absorbed by the critical target in the skin (whatever that might be) remain unsolved. For a summary of some common units and terms used in photobiology see Pathak (1974), and for discussions of terminology see Withrow and Withrow (1956) and Seliger and McElroy (1965).

2.7. Isolation of spectral regions

One of the most fundamental measurements in the investigation of the biological effects of UVR is the determination of the action spectrum, which indicates which wavelengths produce most change in a system. Since there exist no truly monochromatic sources in


the uv, some means must be used to eliminate or minimise the undesired radiation in the source spectrum. Although lasers have been used for biological studies in the uv (Fine and Klein 1969), their expense at present is restrictive (MacKenzie 1978).

The simplest way to modify the source spectrum is by the use of spectrally selective filters based on salt solutions, plastic sheet, or, more generally, optical glasses. The majority of useful filters are of the ‘cut-off’ type. Typically there is high transmission in the longer wavelength region of the spectrum. At the so-called cut-off wavelength, transmission falls rapidly so that in the short-wave region the transmission is negligible. The ideal transmission for this type of filter would be a step function but this is not possible to attain in practice. It should be noted that cut-off filters that transmit short and absorb long wavelengths in the ultraviolet do not exist (Magnus 1976).

A useful class of filters for use in the ultraviolet is the uv bandpass filter (often called ‘black glass’ or ‘Wood’s glass’) which has negligible transmission in the visible. However many of this type of filter may have significant transmission in the red and infrared regions of the spectrum (MacKenzie 1978). A useful compilation of the transmittance curves of about 800 coloured filter glasses made by 13 manufacturers has been produced by Dobrowlski et a1 (1977).

Advances in filter technology since 1950 have resulted in the development of interference filters (Harris 1970). An interference filter is designed to transmit a narrow band of wavelengths with extremely low transmission at the long and short wavelength sides of the transmission band. The simplest forms of these filters contain a layer of dielectric material between two semi-transparent metal deposits. The optical thickness of the dielectric layer is usually made equal to a multiple or submultiple of the wavelength at which maximum transmission is desired. More refined interference filters may have many layers of dielectric materials. The main disadvantages of interference filters is their relatively small size and the strong dependence of transmission upon the angle of incidence of the radiation.

In addition to coloured glass filters and interference filters there exist several other types of filter that are less commonly used including liquid filters, gelatine filters, Christiansen filters and polarisation filters (Magnus 1976). For a detailed account of the use of coloured glass filters and interference filters for the isolation of radiant energy see Harris (1970).

Monochromators for irradiation of the skin may be regarded as adjustable filters which are rather expensive and inefficient. Nevertheless, they are indispensable to much dermatological photobiology because of the simplicity of isolating narrow spectral bands (Magnus 1976). The detailed design and theory of monochromators is well covered elsewhere (James and Sternberg 1969, Thorne 1974), but for brief articles on the theory and design of high-intensity uv monochromators for photobiology see Johns and Rauth (1965a) and Berger et a f (1969).

Monochromators, incorporating both prisms and diffraction gratings as the dispersion device, have been used extensively in dermatological photobiology, principally for measuring the erythema action spectrum (Magnus et a1 1959, Magnus 1964, Sayre et a1 1965, Everett etal 1965, Freeman etal 1966, Berger etal 1968, Cripps and Ramsay 1970, MacKenzie and Frain-Bell 1973, Satoh et a1 1974). These irradiation systems almost invariably use a xenon arc lamp as the source of radiation. The energy output of a monochromator is dependent upon factors including the radiance of the source, the geometry of optical components (e.g. lenses) and the area of the entrance and exit slits. The spectral distribution and bandwidth of the radiation leaving the monochromator depends upon the reciprocal dispersion of the prism or grating and the width of the

414 B L Difey

entrance and exit slits (Johns and Rauth 1965a). MacKenzie and Frain-Bell (1973) have compared the performance of monochromators which have been used for dermatological work, and a comparison of the spectral purity and intensity of different uv monochromators has been made by Johns and Rauth (1965b). The realisation that 100 times the energy is required for narrow band irradiation at 315 nm to produce a minimal erythema response of the skin compared with that required at 300 nm, highlights the importance of spectral purity. The spectral purity of the exit beam is to a large extent dependent on the stray radiation scattered towards the exit slit from inside the monochromator, which in turn is influenced by the design and operation of the monochromator and upon the condition of the optical surfaces such as the grating and mirrors (MacKenzie and Frain-Bell 1973).

3. The skin

3.1. The structure of the skin

The skin consists of a superficial layer, the epidermis, and a deeper vascular connective tissue layer termed the dermis or corium (Green 1968). A schematic cross-section of the superficial layers of the human skin perpendicular to the skin surface is given in figure 2.

T 1 6 0 1

l - J C

Prickle cells

Melanocyte1

Capillary loop

Venule

Arteriole

Figure 2. Schematic representation of the structure of human skin.

The epidermis contains no blood or lymphatic vessels and is composed of stratified squamous epithelium which varies in thickness in different parts of the body. The superficial layer of the epidermis, termed the stratum corneum or horny layer, is a mechanically tough and chemically resistant layer. For the general body surface Kligman (1964) has found the thickness of the stratum corneum to be in the range 11 - 15 km. The epidermal cells are continually being manufactured by the viable


layer (basal and prickle cell layers) which lie on the dermis. Published values on the thickness of the viable layer have been variable (Van der Leun 1966). In a study on diffusion processes in human skin, Van der Leun (1966) used a value of 65 km for the thickness of the viable layer for the flexor side of the forearm and 36 km for the anterior side of the thigh.

Below the epidermis lies the dermis which has well-marked ridges and projections, called papillae, on its superficial surface. This prevents a separation of the two layers by shearing. The dermis consists of dense connective tissue with blood vessels and lymphatics, and merges into the less dense subcutaneous tissue. It is the elastic fibres present in the dermis that give the skin its characteristic elasticity.

3.2. Optics of the skin

Studies on the optics of the skin present severe experimental problems. The skin is not only reflecting and absorbing, it is inhomogeneous, and even its dimensions are difficult to measure (Daniels 1969). Ultraviolet radiation incident on the skin may be reflected, refracted, absorbed, scattered, transmitted or produce fluorescence.

Early work on the penetration of visible and uv radiation in human skin was carried out by Bachem and Reed (1930) and Bachem (1929). These workers placed frozen sections of plantar skin over the slit of a spectrograph and determined the transmittance of the mercury lines through the stratum corneum, viable layer of the epidermis and dermis. Their results are summarised in table 3, which gives the percentage of incident radiation transmitted through skin layers of defined thickness.

Table 3. Percentage transmission of visible and ultraviolet radiation at different depths in skin (Bachem and Reed (1930); adapted from Daniels (1969)).

Layer Thickness of Wavelength (nm) layer (mm) 200 2.50 280 300 400 5.50 7.50

Horny layer 0.03 0 19 1.5 34 80 87 7 8 +Viable epidermis 0.0.5 0 11 9 16 S7 77 6.5 +Dermis 2.0 0 0 0 0 1 S 2 1

A comprehensive experimental study on the transmission through skin of visible and uv radiation was reported by Hansen (1948). As a radiation source Hansen used a super high-pressure mercury arc lamp cooled by circulating distilled water. Spectral regions were separated with the help of optical fluid filters, solid filters or by a double monochromator and dosimetry was carried out photographically. The results of the investigation showed that transmission of UVR through skin, both (depilated) mouse skin and human skin, and through frozen sections of different layers of both types of skin, decreases uniformly with wavelengths from 500 to 300 nm. In the study Hansen looked at the influence of scattering of UVR in the skin, reflection of UVR from the skin surface, fluorescence of the skin and width of the spectral interval. He concluded that the introduction of the various corrections needed did not alter the shape of the transmission curve, but only the numerical value of the transmission in the spectral region 280-500 nm. Apart from a minimum at 415 nm, due to absorption by the haemoglobin, the various layers of the skin did not exhibit any characteristic feature as regards the absorption of radiation.

416 B L Diffey

Kirby-Smith et a1 (1942) found it required 1-10% of 300 nm UVR to penetrate the epidermis of untanned and variously tanned white skin. Tronnier and Merten (1956) have also measured transmittance at 300 nm. They took 25 pm sections parallel to the skin surface and found 32% transmittance of the outer section (considered to be stratum corneum), 35% through the next section (considered to be the viable layer of the epidermis) and 43% through the third 25 pm section (considered to be dermis).

Caucasian and negro skin specimens treated in various ways to separate the stratum corneum and entire epidermis have been studied by Everett et a1 (1966). Like previous authors, these investigators found transmission through the stratum corneum from 17 to 66% in the sunburn range of 290-320 nm.

To date there have been no theoretical studies on the transport of UVR in skin, probably because of its complex nature. However, if skin can be modelled as a turbid medium, then there exist many reports on the application of turbid medium theory, which is a particular case of the phenomenon of radiative transfer, to areas such as astrophysical studies of stellar and intergalactic atmospheres and visibility in fog, rain and snow (Atkins 1969). Turbid medium theory may be conveniently divided into two categories; those based upon continuum models and those based upon statistical models. In 1931 Kubelka and Munk published what has probably become the most widely used continuum model. The Kubelka-Munk treatment starts with the assump- tion of a plane parallel medium of infinite sideways extent but of finite thickness. The theory is based upon a model in which the radiation field is approximated by two fluxes, one travelling from the illuminated surface and the other moving in the opposite direction. As radiation travels from the surface, its intensity is decreased by scattering and absorption processes, both assumed to be proportional to the thickness of the medium traversed. The Kubelka-Munk theory has found widespread application in the colorant ,industry. An appraisal of this model together with other continuum and statistical models is given by Hecht (1976).

4. The biological effects of ultraviolet radiation in normal skin

The normal responses of the skin to UVR can be classed under two headings: acute effects and chronic effects. The acute reactions considered will be erythema (sunburn), delayed melanin pigmentation (suntan) and vitamin D production. Skin aging and skin cancer will be discussed as those reactions produced by prolonged or repeated UVR

exposure. For a review of these and other aspects of the biological response of human skin to UVR see Johnson et a1 (1968).

4.1. Ultraviolet erythema

Exposure to UVR, particularly from wavelengths less than 315 nm, can result in erythema or sunburn. The redness of the skin which is characteristic of erythema is attributable to an increased blood content by dilation of the superficial blood vessels, mainly the subpapillary venules (Rothman 1954). The papillary capillaries themselves add little to the redness seen (Magnus 1976). Natural sunburn has a latent period of a few hours and once present, may persist for several hours or even a few days. Erythema induced by artificial sources is strongly dependent on the wavelength of radiation. At 300 nm an average threshold dose or minimal erythema dose (MED) in white skin lies at about 10-20mJ cm-2 (see, for example, Freeman et a1 1966), whereas for UVA

radiation the MED is about a thousandfold higher (Parrish et a1 1974). Large doses of


UVB may result in oedema, pain and blistering (Magnus 1976) although blistering never occurs with uvc (Warin 1978).

4.1.1. The mechanism of UVR erythema. The vascular response due to UVR could be considered to arise from two different types of mechanism. It could be from a direct action on the vessel wall itself, or indirectly from a photochemical reaction via a diffusing chemical mediator arising in the epidermis (Magnus 1976). In a comment on ultraviolet erythemas in man, Warin (1978) has examined the evidence that prostaglandins act as mediators or modulators of inflammation in UVR erythema. He postulates that uvc erythema is closely associated with the formation of prostaglandins as a result of damage to the epidermis, which then diffuse into the dermis and cause vasodilation, and that UVA erythema is entirely due to a direct effect on dermal vasculature. UVB erythema is considered to be a mixture of both effects but with the prostaglandin formation taking the major,role in erythema production.

The evidence that UVR erythema entails diffusion of a chemical mediator from the epidermis to the dermis includes the following:

(a) the erythemally reactive UVR is mostly absorbed in the epidermis (Blum 1955). (b) there is a latent period between exposure and appearance of erythema (Kimmig

and Wiskemann 1959). (c) the phenomenon known as ‘diffusion flush’ (Lewis 1927) which is that an

erythema, elicited with a large dose of UVR, is after some time surrounded by a diffuse reddening, less intense than the central erythema and slowly spreading outward. This effect was also noted by Van der Leun (1966) who termed it a ‘lateral extension’, and by Diffey and Goldin (1979, unpublished data) after UVA irradiation of the forearm previously photosensitised with 8-methoxypsoralen.

The diffusion theory of UVR erythema has been examined mathematically by Van der Leun (1966). His first approximation was to consider the skin to be a semi-infinite, uniform and isotropic medium, bounded only by a plane horny layer. He further assumed that the mediating substance was formed at the junction of the horny layer and viable layer of the epidermis (x = 0), and instantaneously during radiation (t = 0). The diffusion process is then described by the differential equation given by Fick (1855)

dC(x, t)/dt = D[d2C(x, t)/dx2] (2) where C(x, t) is the concentration of the mediator substance at a depth x normal to the skin surface and at time t after irradiation, and D is the effective diffusion coefficient. The solution of equation (2) which describes the spreading of the substance through the skin is

C(x, t ) =A(rDt)-l” exp(-xZ/4Dt) (3)

where A is the total amount of mediator substance formed per unit area of skin and is assumed to be directly proportional to the uv dose. Van der Leun then applied this equation to predict the time of appearance and disappearance of erythema as a function of uv dose. He found substantial agreement between theory and observation with respect to 300 nm UVR erythema, though not with 250 nm UVR, for which he proposed a direct dermal action. However, the results of Ramsay and Challoner (1976) favour one mechanism for the erythema from both wavelengths. They explain the darker colour of 300 nm UVR erythema, compared with 250 nm, as being due to blood flow in more and deeper vessels and a greater flow. Recent work on skin temperature changes following uv irradiation by Challoner and Woodrough (1978) has supported this theory.

418 B L Diffey

4.1.2. The action spectrum for erythema. An action spectrum is a plot of the reciprocal of the dose required for a given effect versus wavelength (Loofbourow 1948), and strictly applies only if the dose-response curves are similar at all wavelengths, which implies that the mechanism of action is the same at all wavelengths (Jagger 1967).

The erythema action spectrum has long been the subject of controversy. The earlier form of the curve had a major peak of activity at 297 nm, a minimum at 280 nm, and a second but lesser peak at 250 nm (Hausser and Vahle 1922,1927, Luckiesh et a1 1930, Coblentz et a1 1931). These different curves showed close agreement from approximately 270 nm to 310 nm and so from these reports a 'standard erythemal curve' was proposed by Coblentz and Stair (1934) and adopted by the International Commission on Illumination (1935). This double peaked curve was accepted as standard for many years but more recent work suggests that the peak at 250 nm is in fact higher than that at 297 nm (Rottier 1954, Magnus 1964, Everett etal 1965, Freeman etal 1966, Berger et a1 1968, Cripps and Ramsay 1970, MacKenzie and Frain-Bell 1973). Figure 3 compares the IC1 action spectrum with more recent estimates.

Wavelength inm i

Figure 3. Comparison of the standard erythemal curve (IC1 1935) with the erythema action spectra determined by Everett et al(1965), Freeman etal (1966) and Cripps and Ramsay (1970). - Standard erythemal curve; - - - Everett et a l ; . ' ' ' Freeman e t a l ; - - . Cripps and Ramsay.

Methods to quantitate the erythemal response have involved use of both red- coloured papers (Hausser and Vahle 1927) and red photographic filters (Berger et a1 1968) to which the reaction could be compared and graded, and also reflectance spectrophotometry (Henschke and Schulze 1938, Langen 1938, Daniels and Imbrie 1958).

4.2. Melanin pigmentation

The familiar suntan that is seen about two days after exposure to unfiltered sunlight is due not only to the formation of new melanin but also depends upon the distribution of


the newly formed pigment throughout the cells of the epidermis (Johnson 1978). The threshold dose for melanogenesis and its range has not been well defined; obviously it varies considerably from one subject to another. In European stock perhaps the threshold dose, on visual assessment, lies somewhere around 100 mJ cm-’ for monochromatic 300 nm radiation (Magnus 1976). However longer wavelengths than those required for erythema can produce some suntanning, even those wavelengths extending well into the visible region (Pathak 1964, Quevedo et a1 1974), although wavelengths in the UVB region are still the most effective in initiating melanogenesis.

The manner in which melanin affords protection against sunburn is not entirely understood. Daniels (1964) reviewed the relation between pigmentation and human adaptation to environmental UVR and stated that it was unlikely that a darkly pig- mented skin was required solely as a shield against natural UVR. Recent studies have shown that the thickening of the epidermis that occurs after mild exposure to UVR can also afford protection against damage by UVR.

4.3. Production of vitamin D

The skin absorbs UVB radiation in sunlight to convert sterol precursors in the skin such as 7-dehydrocholesterol to vitamin D. The metabolism and biochemical production of vitamin D are outside the scope of this review, but for a brief survey of these aspects together with the importance of sunlight see Beadle (1978).

The vitamin D status in long-stay geriatric patients is known to be poor (Corless et al 1975) but a recent study in this group of patients has shown that irradiation with UVB lamps gave increases sufficient to bring the plasma 25-hydroxyvitamin D levels into the normal range with uv doses less than those required to produce erythema (Corless et a1 1978).

4.4. Aging of the skin

Chronic exposure to UVR can result in skin aging, particularly in those who do not have a well-melanised skin. The clinical changes associated with skin aging include increased dryness and cracking of the horny layer, increased pigmentation, decreased elasticity of the skin, localised hyperkeratoses, and localised vascular dilatations (Magnus 1976, Giese 1976).

4.5. The carcinogenic nature of ultraviolet radiation

The idea that chronic exposure to sunlight is a cause of skin cancer is not an old one. The earliest suggestions appear to have come from Unna (1894) who referred to a precancerous development which he coined as ‘Seaman’s skin’. Since then it has generally been accepted that sunlight, or more precisely those wavelengths less than about 320 nm, plays an aetiological role in the formation of skin cancer in man (Blum 1959, Freeman et a1 1970, Setlow 1974, Freeman 1975). There appears to be little question now that the majority of squamous cell carcinomas are caused by chronic exposure to sunlight (Urbach 1969b), although such a clear relationship between the incidence of basal cell carcinomas and sunlight exposure has not been demonstrated (Urbach et a1 1974, Brodkin et a1 1969, Diffey et a1 1977). In general the incidence of skin cancer is highest in fair-skinned people who live in tropical and subtropical areas (Belisario 1972).

420 B L Diffey

The evidence for the carcinogenic effects of UVR is of different kinds, some of it more convincing than others, but all converging.

(a) there is strong evidence in the anatomical distribution of skin cancers, which in light-complexioned peoples appear predominantly on areas not ordinarily protected by hair or clothing: chiefly on the face (Lacassagne 1933, Blum 1940, Brodkin eta1 1969).

(b) Negroes are notably free from skin cancer, this corresponds to their insensitivity to sunburn. Both factors may be accounted for, in part at least, by the greater opacity of the horny layer of the epidermis. This greater opacity may be due to greater thickness of the horny layer, as well as to a greater amount of melanin, although the exact role of this pigment is not clear, and correlation of sunburn or skin cancer with skin colour is not to be trusted (Daniels 1964).

(c) it is widely accepted that persons who are much exposed to the ultraviolet of sunlight, because of occupation or geographical location, are more likely to get skin cancer; but, although this generalisation seems sound enough, it must be admitted that statistics to establish it beyond question are lacking (Urbach 1969b, Urbach et a1 1974).

(d) perhaps the most convincing evidence comes from laboratory experiments: cancer can be induced in 100% of albino mice with repeated doses of UVR (Blum 1959).

4.6. Models o f skin cancer incidence

At present there are no firm data on the relationship between incidence of skin tumours and sunlight exposure. Neither the appropriate dose rate nor what the induction time might be is known in the case of human skin cancer apparently provoked by sunlight.

The model proposed by Blum (1969) based upon the results of tumour development time in albino mice led to the tentative suggestion that the incidence of cancers of the skin in man should be proportional to the square root of the average uv dose. On the other hand, correlation of observed ultraviolet exposure and skin cancer incidence in Australia (Robertson 1969) resulted in the finding that in going from Brisbane (latitude 27’s) to Cloncurry (21”s) an increase of roughly three times the skin tumour frequency corresponded to an exposure increase of wavelengths less than 320 nm of about 1.5 to 1.6. This suggests that the tumour incidence is proportional to the square of the environmental uv dose. Furthermore, from detailed epidemiological investigations, Urbach et a1 (1974) have estimated that the incidence of skin cancer approximately doubles for every 10 degrees decrease of latitude, provided that the population is of reasonably similar genetic stock. Schulze and Grafe (1969) have calculated that the erythemally effective and presumably carcinogenic global uv radiation increases with the square of decrease in latitude. This result, coupled with Urbach’s estimate, leads to a rather complex relationship between skin cancer incidence and sunlight exposure.

More recently the ultraviolet dose dependence of non-melanoma skin cancer incidence has been studied by Green er a1 (1976) and Fears et a1 (1977). These authors have correlated epidemiological skin cancer data with both measured and calculated estimates of the uv environment. In both instances the age-adjusted data has been modelled by a power law relationship of the form

skin cancer incidence a (annual uv dose)’. (4)

A power law representation of age-adjusted incidence data versus uv dose has the convenience that the power p serves as a constant ‘biological amplification factor’ in the


sense that dI/ I

p = d D I D

42 1

where I is skin cancer incidence and D is annual uv dose. Green et a1 (1976) have found that p varies from 0.0 f 2.0 to 3.8 f 2.0 depending upon geographical location, although a consolidated fit to the complete ensemble of data led to a value of p = 2.5 f 0 . 3 . Fears et a1 (1977) obtained values of p equal to 2.96h0.59 and 2.45 f 0.63 for non-melanoma skin cancer in males and females respectively. These authors correlated the incidence of skin cancer among white people in the United States using data contained in the Third National Cancer Survey (Cutler and Young 1975), with field measurements of biologically effective UVR at 10 locations in the USA (Scotto et a1 1975).

However Green (1978) has recently pointed out that the spectral response of the uv meter used in this survey (Robertson 1972) falls off more slowly at longer wavelengths than the deoxyribonucleic acid action spectrum (Green and Miller 1975) which is thought to be the relevant biophysical mechanism for skin cancer (Setlow 1974). By appropriate correction of the uv dose measurements of Fears et a1 (1977), Green has concluded that the risk of non-melanoma skin cancer depends upon a factor of uv dose raised to the power of 1.8.

5. The current involvement of medical physics with ultraviolet radiation

Probably the two main areas of growing involvement of hospital physicists with ultraviolet radiation are radiation protection and photochemotherapy.

5.1. Protection against ultraviolet radiation

Equipment which emits UVR is commonplace in many hospitals. Examples include (a) low-pressure mercury discharge lamps with no phosphor coating for sterilisation

(the so-called ‘bacteriacidal lamp’); (b) fluorescent lamps emitting either UVB, UVA or a combination of both, for

dermatological therapy; (c) medium and high pressure mercury arc lamps for irradiation of, for example,

superficial ulcers in physical medicine; and (d) xenon lamps used in conjunction with optical filters or a monochromator for

diagnostic photodermatology. It may also be noted that there is an increasing domestic use of sun-lamps, solaria and sunbeds (Which? 1979).

The biological effects of overexposure to UVR are dependent on the wavelength range of the radiation. As the penetration of the radiation is small, effects are mainly limited to the unprotected skin and the surface of the eyes. The effects on the skin of exposure to UVR have already been discussed. The principal effect of excessive uv exposure on the eyes is kerato-conjunctivitis. The main symptoms are pain and an aversion to bright light. The length of time between exposure and the occurrence of symptoms is similar to skin erythema; the symptoms tend to disappear after 36 h and permanent corneal damage is very rare. The most effective wavelength for producing the effect is at 270 nm (Pitts 1973). Radiation in the UVA range is absorbed less by the

422 B L Diffey

cornea, but its absorption in the aqueas humour and in the lens may produce a harmless, transient fluorescence effect during exposure. Absorption of UVA in the lens probably contributes to its progressive yellowing with increasing age and may be a factor in producing cataracts.

Threshold limit values for occupational exposure to uv radiation are published by the American Conference of Governmental Industrial Hygienists (1976). These values have been adopted by the UK Health and Safety Executive, the UK National Radiolo- gical Protection Board and the US Department of Health, Education and Welfare and are intended to protect the eyes and the skin from acute effects.

For radiation in the UVA region (315-400 nm), the threshold limit value is set at an irradiance qf 1 mW cm-2 for periods of exposure greater than 1000s. For actinic uv radiation (200-315 nm), the threshold limit value is given as a radiant exposure per eight-hour period of exposure; the actual value depends greatly on the wavelength as it reflects the large variation in the effectiveness of the radiation with wavelength. The lowest value is 3 mJ cm-2 for an eight-hour period and occurs at a wavelength of 270 nm. This corresponds to an average irradiance of mW cm-2 over the eight- hour period.

For broad-band radiation in the actinic region, a weighted summation is necessary to give the effective irradiance relative to a monochromatic source at 270 nm. Instru- mentation is available which makes this weighted summation (the IL730A uv actinic radiometer) and has been used by Hughes (1978) and McKinlay (1978) in the assessment of hazards from UVR sources. McKinlay (1978) has also measured absolute irradiance using a portable spectral scanning radiometer which incorporates integrating sphere input optics, a double grating monochromator and a photomultiplier detector.

Protection against unnecessary exposure to UVR may be achieved by a combination of administrative and engineering controls, and personal protection, details of which are contained in the NRPB booklet (1977) entitled Protection against Ultraviolet Radiation in the Workplace.

5.2. Photochemotherapy

In recent years a treatment for the skin disease, psoriasis, has been developed which requires the topical or oral administration of the photoactive drug 8-methoxypsoralen followed by exposure to long-wave ultraviolet radiation (Parrish et a1 1974). This treatment, known colloquially as photochemotherapy or PUVA, is proving as effective as more conventional forms of treatment (Rogers et a1 1979).

However, statements of uv dose or irradiance quoted by the majority of authors are subject to some degree of uncertainty in their interpretation. The uv sources used in PUVA are usually low pressure mercury discharge tubes with a fluorescent coating on the inside of the lamp envelope. A recent survey of several commercially available treatment units (Diffey et a1 1979) has shown the emission spectrum to be qualitatively the same for all units studied, that is, a continuous spectrum from about 315 to 400 nm accompanied by several mercury lines in the uv and visible. The uv detectors that are generally used for monitoring the output of these lamps consist of a wavelength- dependent photocell filtered by a uv bandpass filter (and possibly an infrared blocking filter). The spectral response of such a detector varies from one manufacturer to another. Also it is often not clear from the manufacturer’s documentation how the instrument is calibrated. Consequently any measured value of irradiance in physical units such as mW cmp2 will represent a convolution between the emission spectrum of


the source and the spectral response of the detector. This dependence of statements of irradiance upon both spectral emission and spectral response may partly account for the large differences in the total uv dose needed to clear psoriatic lesions which have been reported in the literature (Diffey 1978).

Rationalisation of measurement techniques in radiotherapy led to the development and universal acceptance of instruments such as the Farmer ion chamber and dose- meter. Perhaps collaboration between dermatologist, medical physicist and manufacturer could result in similar rationalisation of instrumentation in clinical ultraviolet photobiology.

References

ACGIH 1976 Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment: Ultraviolet Radiation (American Conference of Governmental Industrial Hygienists, Cincinnati, Ohio 45201)

Alsins J, Claesson S , Fischer T and Juhlin L 1975 Acta Dermatovenerol 55 261 Atkins J T 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F Urbach

Bachem A 1929 A m . J. Physiol. 91 58 Bachem A and Reed C I 1930 Arch. Phys. Ther. 11 49 Bassi P, Busuoli G and Rimondi 0 1975 Health Phys. 28 470 Beadle P C 1978 in Ultraviolet Radiation and its Medical Applications HPA Report CRS-28 (HPA. 47

Beeson E J G 1978 Ultraviolet Radiation and its Medical Applications HPA Report CRS-28 (HPA. 47

Belisario J C 1972 Int. J. Dermatol. 11 148 Bener P 1972 Final Technical Report DAJA37-68-C-1017 (European Research Office, United States Army,

Berger D S 1969 J. Invest. Dermatol. 53 192 Berger D, Magnus I A, Rottier P B, Sayre R M and Freeman R G 1969 in The Biologic Effects of Ultraviolet

Berger D, Urbach F and Davies R E 1968 in Proc. 13th Int. Congr. of Dermatology, Munich, 1967 (New York:

Blum H F 1940 J. Natl Cancer Inst. 1 397

- 1959 Carcinogenesis by Ultraviolet Light (New Jersey: Princeton University Press) - 1969 in The Biologic Effects of Ultraviolet Radiation With Emphasis on the Skin ed F Urbach (Oxford:

Brodkin R H, Kopf A W and Andrade R 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis

Byrne P 0 and Farmer F T 1972 J. Phys. E : Sci. Znstrum., 5 590 Challoner A V J and Woodrough R E 1978 Br. J. Dermatol. 98 537 Chandrasekar S 1950 Radiative Transfer (Oxford: Clarendon) Coblentz W W and Stair R 1934 J. Res. Narl Bur. Stand. 12 13 Coblentz W W, Stair R and Hogue J M 1931 Proc. Natl Acad. Sci USA 17 401 Commission Internationale de I’Eclairage 1970 International Lighting Vocabulary 3rd ed (Paris) Corless D Beer M, Boucher B J, Gupta S P and Cohen R D 1975 Lancet June 28 1404 Corless D, Gupta S P, Switala S , Barragry J M, Boucher B J, Cohen R D and Diffey B L 1978 Lancet 23

Cripps D J and Ramsay C A 1970 Br. J. Dermatol. 82 584 Cutler S J and Young J L 1975 Natl Cancer Inst. Monograph 41 DHEW Publ. No (NIH) 75-787 Daniels F 1964 in Handbook of Physiology-ZVAdaptation to the Environment ed D B Dill (Washington DC:

(Oxford: Pergamon) p 141

Belgrave Square, London, SWlX 8QX) p 34

Belgrave Square, London SWlX 8QX) p 17

London)

Radiation with Emphasis on the Skin ed F Urbach (Oxford: Pergamon) p 125

Springer-Verlag) p 11 12

- 1955 in Radiation Biology ed A Hollaender (New York: McGraw-Hill) 2 p 487

Pergamon) p 543

on the Skin ed F Urbach (Oxford: Pergamon) p 581

September 649

Am. Physiol. Soc.) p 969

Pergamon) p 151 - 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F Urbach (Oxford:

Daniels F and Imbrie J D 1958, J. Invest. Dermatol. 30 295

424 B L Diffey

Davis A, Deane G H W and Diffey B L 1976 Nature (London) 261 169 Deirmendjian D and Sekera Z 1954 Tellus 6 382 Diffey B L 1977 Phys. Med. Biol. 22 309 - 1978 Br. J. Dermarol. 98 703 Diffey B L and Challoner A V J 1978 Phys. Med. Biol. 23 1124 Diffey B L, Challoner A V J and Key P J 1979 Br. J. Dermatol. (in press) Diffey B L and Davis A 1978 Phys. Med. Biol. 23 318 Diffey B L, Davis A, Johnson M and Harrington T R 1977 Br. J. Dermatol. 97 127 Diffey B L, Kerwin M and Davis A 1977, Br. J. Dermatol. 97 407 Dobrowlski J A, Marsh G E, Charbonneau D G, Eng J and Josephy P D 1977 A p p l . Opt. 16 1491 Drummond A J and Wade H A 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin

Ellinger F 1957 Medical Radiation Biology (Springfield: Charles C. Thomas) p 614 Everett M A, Olson R L and Sayre R M 1965 Arch. Dermatol. 92 713 Everett M A, Yeargers E, Sayre R M and Olson R L 1966 Photochem. Phorobiol. 5 533 Fears T R, Scotto J and Schneiderman M A 1977 A m . J. Epidemiol. 105 420 Fick A 1855 A n n . Phys. Chem. 94 59 Fine S and Klein E 1969 in The Biologic Effects of Ulrravioler Radiation with Emphasis on the Skin ed F

Freeman R G 1975 J. Narl Cancer Inst. 55 11 19 Freeman R G, Hudson H T and Carnes R 1970 Int. J. Dermatol. 9 232 Freeman R G, Owens D W, Knox J M and Hudson H T 1966 J. Invest. Dermatol. 47 586 Gates D M 1966 Science ( N Y ) 151 523 Giese A C 1976 Living with our sun’s ultraciolet rays (New York: Plenum) Gillham E J 1961 Radiometric Standards and Measurements Notes on Applied Science No. 23 (London:

Green A E S 1978 A m . J. Epidemiol 107 277 Green A E S, Findley G B, Klenk K F, Wilson W M and MO T 1976 Photochem. Photobiol. 24 353 Green A E S and Miller J H 1975 CIAPMonograph 5 ed A Grobecker, (Washington DC: Department of

Green A E S, Sawada T and Shettle E P 1974 Photochem. Photobiol. 19 251 Green J H 1968 A n Introduction to Human Physiology (London: Oxford University Press) 2nd edn p 119 Hansen K G 1948 Acta Radio[. Suppl. 71 Harris L 1970 Eppley Foundation for Research, Monograph No. 2 (Newport RI: Eppley Foundation) Hatchard C G and Parker C A 1956 Proc. R . Soc. A 235 518 Hausser K W and Vahle W 1922 Strahlentherapie 13 41 - 1927 Wiss. Ver6ff. Siemens-Konzern 6 101 Hecht H G 1976 J. Res. Natl Bur. Stand. 80A 567 Henderson S T 1970 Daylight and its Spectrum (New York: American Elsevier) Henderson S T and Marsden A M 1972 Lamps and Lighting (London: Edward Arnold) 2nd edn Henschke U and Schulze R 1938 Srrahlentherapie 63 236 Hughes D 1978 Hazards of Occuparional Exposure to Ultraviolet Radiation, Occupational Hygiene Mono-

International Commission on Illumination, Berlin 1935 Compt. Rend. 9 596 Jagger J 1961 Radiat. Res. 14 394 - 1967 Introduction to Research in Ultraviolet Photobiology (New Jersey: Prentice-Hall) James J F and Sternberg R S 1969 The Design of Optical Spectrometers (London: Chapman and Hall) Johns H E and Rauth A M 1965a Photochem. Photobiol. 4 673 - 1965b Photochem. Photobiol. 4 693 Johnson B E 1978 in Ultraviolet Radiation and its Medical Applications HPA Report CRS-28 (HPA: 47

Johnson B E, Daniels F and Magnus I A 1968 in Current topics in Photophysiology IV ed A C Giese (New

Kimmig J and Wiskemann A 1959 in Handbuch der Haut- und Geschlechtskrank-heiten. Erganzungswerk

Kirby-Smith J S, Blum H F and Grady H G 1942 J. Narl Cancer Inst. 2 403 Kligman A M 1964 in The Epidermis eds W Montagna and W C Lobitz, Jr (New York: Academic) pp

ed F Urbach (Oxford: Pergamon) p 391

Urbach (Oxford: Pergamon) p 11 5

HMSO)

Transportation)

graph No. 1 (Uni:rersity of Leeds Industrial Services Ltd)


York: Academic)

(Berlin: Springer-Verlag) 5 part 2 1021

387-433


Koller L R 1952 Ultraviolet Radiation (New York: John Wiley and Sons) Kubelka P and Munk F 1931 Z. Tech. Phys. 12 593 Lacassagne A 1933 Ann. Dermatol. Syph. 4 497 Langen D 1938 Strahlentherapie 63 142 Latarjet R, Morenne P and Berger R 1953 Ann. Inst. Pasteur 85 174 Lewis T 1927 The Blood Vessels of the Human Skin and their Responses (London: Shaw) p 17 Loofbourow J R 1948 Growth Symposium 12 75 Luckiesh M, Holliday L L and Taylor A H 1930 J. Opt. Soc. A m . 20 423 McCullough E C, Fullerton G D and Cameron J R 1972 J. A p p l . Phys. 43 77 McCullough E C and Porter W P 1971 Ecology 52 1008 MacKenzie L A 1978 Ultraviolet Radiation and its Medical Applications HPA Report CRS-28 (HPA: 47

MacKenzie L A and Frain-Bell W 1973 Br. J. Dermatol. 89 251 McKinlay A F 1978 in Ultraviolet Radiation and its Medical Applications HPA Report CRS-28 (HPA: 47

Magnus I A 1964 Br. J. Dermatol. 76 245

- 1978 Ultraviolet Radiation and its Medical Applications HPA Report CRS-28 (HPA: 47 Belgrave Square, London SWlX 8QX) p 5

Magnus I A, Porter A D, McCree K J, Moreland J D and Wright W D 1959 Br. J. Dermatol. 71 261 Meyer A E H 1952 Strahlentherapie 88 481 Mullard 1969 General Explanatory Notes on Cadmium Sulphide Photoconductive Cells Mullen P A, Kinmonth R A and Searle N Z 1975 J. Test. Eval. 3 15 Nambi K S and Higashimura T 1971 3rd Int. Conf. on Luminescence Dosimetry, Riso (Danish AEC, Riso,

National Institute for Occupational Safety and Health 1972 Criteria for a recommended standard., .

National Radiological Protection Board 1977 Protection against Ultraviolet Radiation in the Workplace

Parrish J A, Fitzpatrick T B, Tanenbaum L and Pathak M A 1974 New Engl. J. Med. 291 1207 Parrish J A, Ying C Y, Pathak M A and Fitzpatrick T B 1974 in Sunlight and Man ed T B Fitzpatrick (Tokyo:

Pathak M A 1964 in Recent Progress in Photobiology (New York: Academic) p 381 - 1974 in Sunlight and Man ed T B Fitzpatrick (Tokyo: University of Tokyo Press) p 815 Pitts D G 1973 Health Phys. 25 559 Plewig G, Hofmann C, Braun-Falco 0, Nath G and Kreitmair A, 1978 Br. J. Dermatol. 98 15 Quevedo W C, Fitzpatrick T B, Pathak M A and Jimbow K 1974 in Sunlight and Man ed T B Fitzpatrick

Ramsay C A and Challoner A V J 1976 Br. J. Dermatol. 94 487 Robertson D F 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F Urbach



- 1976 Dermatological Photobiology (Oxford: Blackwell)

Denmark)

Occupational exposure to ultraviolet radiation (US DHEW, Washington DC)

(HMSO) ISBN 0 85951 063 8

University of Tokyo Press) p 131

(Tokyo: University of Tokyo Press) p 165

(Oxford: Pergamon) p 619 - 1972 Ph.D. Thesis University of Queensland, Brisbane, Australia Rogers S, Marks J, Schuster S, Vella Briffa D, Warin A and Greaves M 1979 Lancet March 3, 455 Rothman S 1954 Physiology and Biochemistry of the Skin (Chicago: University of Chicago Press) pp 69-70 Rottier P B 1954 in Proceedings of the First International Photobiology Congress (Wageningen: H Veenman

Ruff H R 1970 Light Res. Technol. 2 43 Saleeby C W 1926 Sunlight and Health (London: Nisbet) 3rd edn p 72 Satoh Y, IrimajiriT, Okawara S, Shimao K and Seiji M 1974 in SunlightandMan e d T B Fitzpatrick (Tokyo:

Sayre R M, Straka E R, Anglin J H and Everett M A 1965 J. Invest. Dermatol. 45 190 Schafer V 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F Urbach

Schulze R and Grafe K 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F

Schwarz E 1952 Research 5 407 Scotto J, Fears T R and Gori G B 1975 DHEW Pub1 No (NIH) 76-1029 Seliger H H and McElroy W D 1965 Light: physical and biological action (New York: Academic) pp 8-17

and Zonen) p 152

University of Tokyo Press) p 575


Urbach (Oxford: Pergamon) p 369

426 B L Diffey

Setlow R B 1974 Proc. Natl Acad. Sci. ( U S A ) 71 3363 Stair R 1969 in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F Urbach (Oxford:

Summer W 1962 Ultraviolet and Infrared Engineering (London: Pitman) Task Force on Photobiology of the National Program for Dermatology 1974 Arch. Derm. 109 833 Thekaekara M P 1970 N A S A Technical Report TR-R-351 (Washington DC) Thorne A P 1974 Spectrophysics (London: Chapman and Hall) Tronnier H and Merten W 1956 Zeit. Haut- U Geschlechtskr. 21 157 Unna P G 1894 Die Hisropathologie der Hautkrankheiten (Berlin: Hirschwald) p 725 Urbach F 1969a in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F Urbach

-- 1969b in The Biologic Effects of Ultraviolet Radiation with Emphasis on the Skin ed F Urbach (Oxford:

Urbach F, Epstein J H and Forbes P D 1974 in Sunlight and Man ed T B Fitzpatrick (Tokyo: University of

Van der Leun J C 1966 Ph.D. Thesis University of Utrecht, The Netherlands Warin A P 1978 Br. J. Dermatol. 98 473 Which? 1979 May p 268 Withrow R B and Withrow A P 1956 in Radiation Biology Vol ZZZ ed A Hollaender (New York:

McGraw-Hill) p 125

Pergamon) p 377


Pergamon) p 635

Tokyo Press) p 259

diffey: ultraviolet radiation physics and the skin · pdf fileregion of the solar spectrum...

Documents