INTRODUCTION

The most susceptible individuals to the development of both melanoma and nonmelanoma skin cancers are those with very pale skin (1). They are typically subjects with red hair and freckles, who burn easily but tan poorly, and are classified as having Fitzpatrick skin type I or II (2). Previous studies have investigated the minimal erythema doses (MED) of populations with higher skin types (3,4), but little information, however, exists on the amount of sunlight required to cause sunburn or tanning in this particular group of individuals. It is important to know the amount of natural sunlight that causes damage to these susceptible skin types so that sun protection strategies can be implemented and optimized.

Various solar simulators have been used to approximate the spectrum of natural sunlight and have been shown to be useful substitutes for natural sunlight (5). They have been used in the determination of sunscreen sun protection (SPF) factor measurements and for investigating the photobiological effects of UV radiation. It is important, however, to ensure that similar physical doses deliver similar erythemogenic doses.

The determination of the MED causing sunburn and minimal melanogenic dose (MMD) causing tanning is currently determined visually. Previous studies have shown the ability of instruments such as the chromameter and the erythema meter to quantitate more sensitive and objective measures of skin changes (6,7). We have previously shown that the use of a reflectance spectrometer fitted with a fiber-optic probe is able to quantitatively monitor such biological changes (8).

In this study we have shown that artificial UV sources underestimate but provide a good correlation for the MED of fair-skinned individuals to natural sunlight and that the use of mathematical modeling and instruments such as the erythema meter and the chromameter can provide a more sensitive and objective method for measuring erythemal responses.

MATERIALS AND METHODS

Volunteer data. The trial was conducted over a 2 week period in early autumn, between 12 and 26 March 2001 at the Department of Dermatology, Royal Prince Alfred Hospital, Sydney, Australia. Informed consent was obtained from all volunteers, and ethical approval was granted by the Central Sydney Area Health Service Human Ethics Committee.

Thirty-one healthy pale-skinned volunteers were recruited by advertisement from the University, Hospital and local communities. There were 26 females and 5 males (average age, 27.8 years; range, 18-56). They were eligible for the study if they had very pale skin and on history were unable to tan despite exposure to sunlight. They had no history of photosensitivity and were not on any photosensitizing or anti-inflammatory medications. Their mid-lower backs were not directly exposed to any sunlight for the preceding 4 weeks and for the duration of the trial.

Quantitation of erythema and skin pigmentation. Each volunteer initially had baseline readings taken at the test sites of their constitutive skin pigmentation and erythema using both a reflectance spectrometer or an erythema meter (Diastron, Hampshire, UK) and a chromameter (Minolta CR-300, Osaka, Japan). The erythema meter calculates an erythema index (EI), which is mainly dependent on the blood content of the superficial dermis and hence measures skin erythema by comparing the amount of reflected light at 546 nm with a reference signal at 632 nm. There is a linear relationship between skin erythema and El measured by the erythema meter (9). A melanin index (MI) is calculated by comparing the reflectance at 632 nm with a reference signal at 905 nm. The chromameter measures erythema and skin color based on the Commission Internationale de L'eclairage L*a*b* color space. The a* value is well recognized to linearly correlate with skin erythema. Each volunteer was also skin typed according to the Fitzpatrick classification based on the volunteer's previous experience of sunburn and tanning