Ординатура / Офтальмология / Английские материалы / Progress in Lens and Cataract Research_Hockwin_2002

to the people who were recruited as the subjects of the ophthalmologic epidemiological study carried out in several climatically different areas in Japan, Singapore and Iceland. All of these field studies were conducted by the Kanazawa Medical University, Department of Ophthalmology (Prof. Kazuyuki Sasaki).

To discuss the relationship between UV exposure and health effects epidemiologically, it is essential to evaluate the lifetime UV exposure of the subjects. According to the data collected by the Japan Meteorological Agency (JMA) [1], the UV-B irradiance in Japan was characterized by (1) distinct gradation corresponding to latitude with about 2 times higher annual mean UV-B irradiance in Okinawa compared to that in Sapporo, (2) strong in summer and weak in winter; with 70–80% of yearly UV-B irradiance observed within the summer season from April to September, and (3) uni-modal intensity curve with a peak near noon, and 60% (summer) to 70–80% (winter) of daily UV-B irradiance are observed during midday, 2 h around the local noon [2].

UV exposure of people is influenced by many factors, and it is recognized that the UV irradiance at ground level is important, but not a sufficiently accurate index of UV exposure [3, 4]. In addition to the UV irradiance at ground level, the time spent outdoors by the subjects is an important factor for their UV exposure. Behavioral characteristics of subjects, such as clothing and whether or not a hat and glasses are worn, also play important roles. And the cumulative exposure to UV radiation of individuals depends on their background, such as the location of residence, type of job and lifestyle [3, 5].

Estimation of Daily UV Exposure

Daily UV exposure is estimated by formula 1 and cumulative UV exposure is calculated by summing it up in people’s lifetime according to the residential history. In formula 1, UV intensity at time t is obtained as observed data or estimated values for all places. Outdoor activity at each time of day is not easy to determine for the subjects throughout their whole lifetime. We propose the following modified and simplified estimation of their outdoor activity.

From the results of surveys of the outdoor activity patterns of schoolchildren [6], we discussed the comparability of a simplified method and an original method. We found that the estimates using the number of hours spent outdoors in a whole day gave a good correlation, but smaller estimates than those obtained using the original outdoor activity pattern, and the number of hours spent outdoors between 9:00 and 15:00 gave the best estimates and a good correlation with those obtained using the original outdoor activity pattern (fig. 1) [2]. So the estimates of daily UV exposure were obtained using formula 2.

Fig. 1. Comparison of the UV estimates between different methods. x-axis represents estimation using exact outdoor activity pattern (number of hours and time of day). y-axis represents estimation only using the number of hours spent outdoors. Results estimated using the following time periods are shown: whole day (a), 9:00–17:00 (b), 9:00–15:00 (c) and 10:00–14:00 (d ).

The simple question, ‘How long do you usually spend outdoors between 9:00 and 15:00?’, will provide estimates as good as asking the question, ‘How long and when are you usually outdoors?’.

t 0

where UVest estimated UV exposure, UVt UV intensity at time t and OUTt outdoor activity at time t (out 1, in 0).

where UVest* estimated UV exposure, Toutt number of hours spent outdoors between 9 and 15 h and UVave average UV intensity between 9 and 15 h.

Fig. 2. Setup of the mannequin experiment. These two mannequins with (left) and without (right) protective measures are controlled by computer and move synchronously following the scenario of actual movement of a human head which was input into the computer in advance. In the experiment, UV exposure levels were continuously measured using small UV sensors attached at different points on the mannequins. The UV transmittancy of the lens used in the experiment is less than 1%.

UV Exposure to the Eye and Protective Measures

Although there are many protective measures against UV exposure to the eye, such as wearing glasses or sunglasses and/or a hat [7, 8], there has not yet been sufficient evidence of their efficacy. With the experiments using a newly developed system (mannequin model, fig. 2) we reevaluated the effects of protective measures, i.e. wearing a hat and/or glasses, against UV exposure to the eye.

As shown in figure 3, compared to the ocular UV exposure level without glasses or a brimmed hat, the ocular UV exposure levels with glasses and with a hat were averages of 8.8 and 80.8% over 30 min, respectively. This means that the use of glasses and hats could reduce the UV exposure the eye by 90 and 20%, respectively. From these results the UV exposure to the eye can be estimated as in formula 3.

where UVest* UV estimates with protective measure, UVest UV estimates without protective measure, Gl use of glasses or not (always 1, sometimes 0.5, no 0) and Hat use of hat or not (always 1, sometimes 0.5, no 0).

Fig. 3. Time series of UV exposure of the eye (results of mannequin experiment). Each line shows the UV irradiance at the eye with and without protective measure. The dotted line shows the UV irradiance at ground level as a reference.

Estimation of Ocular UV Exposure for the Participants of Epidemiological Survey

For participants of our epidemiological survey in one area (Kikai Island, Kagoshima) [9], UV exposure was assessed using the above-described method. As shown in figure 4, the estimated cumulative UV exposure was widely distributed. Differences in UV exposure levels were observed for age (higher in older people) and jobs (higher in farmers, people in transportation and sales) but not for gender. There were large differences in estimated UV exposure levels between individuals on this small island, even though all of them had nearly the same residential history except a few who had spent a short time in their life off the island. This suggests that these differences between the subjects were mainly caused by the differences in their outdoor activity patterns.

Evaluation of the Estimates of UV Exposure [30]

To evaluate the validity of estimated levels of UV exposure, we compared two variables, the observed and estimated one based on the field experiment. UV exposure was measured using a simple watch-like device (Model SUB-T, Toray Techno Inst.), in four areas throughout the year, and 20 outdoor workers (caddies at golf links) in each area were selected as participants. Averaged level

%

50

Male

Fig. 4. Frequency distribution of the lifetime UV exposure. The lifetime UV exposure was estimated for the study subjects in Kikai Island using the method described in this article. For the subjects aged 60 years and over, lifetime UV exposure was estimated as the total amount of UV exposure until aged 80 years.

Fig. 5. Correlation between observed and estimated UV exposure – Tsukuba, 1994.09.

of personal UV exposure of the subjects in each area showed (1) seasonal fluctuations, (2) daily fluctuations depending mainly on meteorological conditions, and (3) a geographic gradient according to latitude, the same as those of JMA’s monitoring data. There are many factors, which affect personal UV exposure, but our results indicate that personal UV exposure on the whole corresponds well to UV irradiance at ground level.

Correlation coefficients between the two variables, observed and estimated values of each subjects in each month, ranged from 0.252 to 0.741 and they all

NIES site

Network site

Fig. 6. Location map of UV monitoring network.

were statistically significant. As shown in figure 5 observed and estimated personal UV exposure levels correlate well with each other. The regression coefficients differed little in the study areas and with seasons. There was on the whole a good correlation between observed and estimated UV exposures, and time spent outdoors is thought of as an important explanatory variable for personal UV exposure.

According to similar data for skiers, the observed UV exposure level was higher than the estimated level predicted from the results for caddies, and the correlation between observed and estimated values was smaller than that for caddies. These results might indicate the effect of ground reflection on UV exposure.

UV Monitoring Network in Japan

At the end of this article I will introduce an interesting program: UV Monitoring Network in Japan. This monitoring network started in January 2000

in response to a call by the National Institute for Environmental Studies (NIES, Chair: M. Ono). The members of the network were from 16 organizations, i.e. NIES (6 stations), 9 universities, 5 national and/or prefectural institutions and 1 NGO from all around Japan (fig. 6). The selection of network members was based upon geographic location with regard to UV-related epidemiological studies.

Broadband UV radiometers, type MS-210W (280–315 nm) and type MS210A (315–400 nm, EKO), were used for UV measurement and the MS-210W will be routinely calibrated every year. The amount of total solar radiation was measured using a pyranometer except in a few stations. The data collected in those monitoring stations will be sent to the NIES for data analysis.

Many efforts are being made for quality control of UV measurements and data analysis following the advice of a special committee on UV monitoring. The network consists of 21 stations from 16 organizations at present, but I believe that it will become more complete in the near future and the data and UV information will be given to researchers and/or the public.

References

1Japan Meteorological Agency: Annual Report on Monitoring the Ozone Layer. No 1–10 (1990–1999).

2Ono M: Assessment of exposure to ultraviolet radiation within a lifetime. Environ Sci, in press.

3WHO: The Effects of Solar Ultraviolet Radiation on the Eye. WHO/EHG/94. Geneva, WHO, 1994.

4WHO: Environmental Health Criteria 160 Ultraviolet Radiation, 45–55. Geneva, WHO, 1994.

5MaCarty CA, Lee SE, Livingston PM, Taylor HR: Assessment of lifetime ocular exposure to UV-B. Dev Ophthalmol 1997;27:9–13.

6Munakata N, Ono M, Watanabe S: Monitoring of solar-UV exposure among schoolchildren in five Japanese cities using spore dosimeter and UV-coloring labels. Jpn J Cancer Res 1998;89:235–245.

7WHO: Protection against Exposure to Ultraviolet Radiation. WHO/EHG/95.17. Geneva, WHO, 1995.

8WHO: Environmental Health Criteria 160 Ultraviolet Radiation, 235–252. Geneva, WHO, 1994.

9Sasaki H, Asano K, Kojima M, Sakamoto Y, Kasuga T, Nagata M, Takahashi N, Sasaki K, Ono M, Katoh N: Epidemiological survey of ocular diseases in K Island, Amami Islands. J Jpn Ophthalmol Soc 1999;103:556–563.

10Ono M: Preliminary study on exposure measurement of ultra-violet radiation. Dev Ophthalmol 1997;27:81–88.

Masaji Ono, Environmental Health Sciences Division, National Institute for Environmental Studies, Onogawa 16-2, Tsukuba 305-8506 (Japan)

Tel. 81 298 50 2421, Fax 81 298 50 2588, E-Mail ono@nies.go.jp

determination of UVR exposure of the eye. The degree of lid opening limits ocular exposure to rays entering at angles near the horizon. Clouds redistribute overhead UVR to the horizon sky. Mountains, trees and building shield the eye from direct sky exposure. Most ground surfaces reflect little UVR. The result is that highest UVR exposure occurs during light overcast where the horizon is visible and ground surface reflection is high. By contrast, exposure in a high mountain valley (lower ambient temperature) with green foliage results in a much lower ocular dose. Other findings of these studies show that retinal exposure to light and UVR in daylight occurs largely in the superior retina.

Copyright © 2002 S. Karger AG, Basel

Introduction

The great latitudinal variation of ‘the world’s most blinding disease’, cataract [1–16], if examined along with a wide variety of laboratory studies, would suggest that environmental factors should be a major determinant of the time of onset. However, most epidemiological evidence points to ultraviolet radiation (UVR) in sunlight as a significant risk factor only in cortical cataract [2, 5, 17]. The evidence for UVR as an etiological factor in droplet keratopathies and pterygium is stronger [1, 2, 5, 6, 10, 18, 19]. Prof. K. Sasaki has provided us with a very important insight into the geographical variation in the incidence of cataract by examining the incidence of different types of cataract with latitude [3]. Nuclear cataract is more common in the tropics; cortical cataracts are more common in midlatitudes, and posterior subcapsular cataracts were not so clearly related to latitude. Despite these latitudinal variations in type and overall incidence, many epidemiological studies do not appear to show a relation between UVR and cataract. On the other hand, a wide variety of scientific evidence, from laboratory studies of the UV photochemistry of lens proteins to a number of different animal exposure studies, all provide support for the hypothesis that UVR should play a far greater role in cataractogenesis [1, 20–31]. Most age-related changes in the skin (from accelerated aging to skin cancer) have been conclusively shown to result from excessive exposure to solar UVR (or ‘sunlight exposure’) [1]. Although no one questions that UVR exposure produces the acute effects of ‘sunburn’ (erythema) and snowblindness (photokeratitis), some have questions whether cataracts, pterygium and droplet keratopathies are clearly related to UVR exposure [2, 5]. Even more under debate are theories that suggest that UVR and light may affect retinal diseases such as age-related macular degeneration [32–35]. A better resolution of these questions requires far better ocular dosimetry for both heat and UVR. Epidemiological studies can arrive at erroneous conclusions if assignments of exposure are seriously in error, and assumptions regarding relative exposures have been argued to be incorrect [4, 9].

A major objective of my laboratory at the US Army Center for Health Promotion and Preventive Medicine (USACHPPM) has been to determine the most effective protective measures for sunlight exposure as well as to recommend human exposure limits to UVR. To that end, it has been necessary to characterize the primary sources of UVR exposure to the eye, emphasizing the importance of geometrical exposure factors by measuring UVR distributions in skylight under different environmental conditions. We have also studied the impact of lid opening upon optical radiation exposure dose to the cornea, lens and retina, and how scene luminance (brightness) affects the degree of lid opening and ocular exposure in outdoor subjects. These studies have shown the importance of good ocular dosimetry in any epidemiological study of the relationship between sunlight exposure and ocular disease.

With the geographical variations in the incidence of nuclear cataract, it is surprising that most epidemiological studies show only weak or no apparent relation between UV or sunlight exposure and the incidence of nuclear cataract. Perhaps it is necessary to consider the other important environmental parameter – environmental temperature. Both sunlight and ambient temperature have been cited as potential etiological factors in several age-related ocular diseases [4]. However, the combined roles of these physical factors and their possible synergisms have generally not been carefully examined. It is therefore worthwhile to examine these environmental cofactors together along with geometrical factors to provide some suggestions toward the etiology of these ocular diseases. It is worthwhile also to compare the knowledge of UVR photodamage of the skin and subsequent, localized accelerated ageing changes in skin tissues.

Heat and UVR both damage proteins and biological tissue; however, thermal effects are generally less localized. Because of heat flow and the physiological mechanisms designed to achieve constant thermal equilibrium of adjacent tissues, the changes in local temperature would not normally produce highly localized areas of increased temperature, with one exception: the strong absorption of light and near-infrared radiant energy by the pigmented iris and the direct infrared absorption by the lens. The latter factors play a role in industrial heat cataract. By contrast, the penetration and absorption of UVR lead to a highly localized variation in dose. Thus, by examining highly localized changes in ocular tissues, it may be possible to elucidate the relative roles of UV and IR in certain ocular changes.

Both physical factors (UVR and temperature) could be argued to play causative roles in cataractogenesis on purely theoretical grounds. The purpose of this paper is to better quantify the ambient UVR exposure and thermal environment of the lens to permit more accurate epidemiological studies of these two potential etiological factors in the development of age-related cataract and other changes in the anterior segment.