Contamination from Possible Solar Light Exposures in ESR Dosimetry Using Human Tooth Enamel

Transcription

1 Contamination from Possible Solar Light Exposures in ESR Dosimetry Using Human Tooth Enamel NORI NAKAMURA' *, JANINE F. KATANIC't and CHYUZO MIYAZAWA2 'Department of Genetics, Radiation Effects Research Foundation, 5-2, Hijiyama Park, Minami-ku, Hiroshima , Japan 2Department of Chemistry, Ohu University School of Dentistry, 31-1 Aza-Misumido, Tomita-machi, Koriyama , Japan (Received, April 27, 1998) (Revision received, July 6, 1998) (Accepted, July 13, 1998) ESR/ Tooth enamel/solar light exposure/a-bomb survivors/biodosimetry Electron spin resonance (ESR) measurement data of 98 teeth from atomic-bomb survivors who received various radiation doses were analyzed in terms of possible solar light exposure by tooth position. Each tooth was divided into buccal and lingual parts for independent ESR measurement. We found that average buccal doses were larger than their corresponding lingual doses by 0.48 ± 0.30 Gy (first incisors), 0.33 ± 0.38 Gy (second incisors), 0.20 ± 0.23 Gy (canines), 0.24 ± 0.26 Gy (first bicuspids), 0.17 ± 0.51 Gy (second bicuspids) and 0.04 ± 0.18 Gy (large molars and wisdom teeth). Such a clear declining trend follow ing tooth position in the mouth is readily appreciated as due to solar light exposures. Comparisons of lingual doses of multiple teeth from the same five donors suggest that lingual doses of first incisors were also overestimated by 0.34 ± 0.18 Gy. From the results presented, we deduce that the buccal doses of first incisors are, on the average, overestimated by nearly 0.8 Gy. Therefore, use of front teeth for biodosimetry requires special caution. How to estimate the contribution from exposure to solar light in ESR biodosimetry is discussed. INTRODUCTION Using the electron spin resonance (ESR) technique for tooth enamel, we have recently ex amined 100 teeth donated from Hiroshima atomic-bomb survivors to estimate individual doses'). Each tooth was divided into buccal and lingual parts for subsequent independent enamel isola tion and ESR measurement to evaluate possible contamination from dental X-ray exposures, which affect mostly buccal surfaces. Nearly 20 teeth showed considerably larger buccal doses than their corresponding lingual doses, at first appearing as if contribution from dental X-ray *Corresponding author: Tel , Fax , nnakamur@rerf.or.jp 'Present address: School of H ealth Sciences, Purdue University, 1338 Civil Engineering Building, West Lafayette, Indiana , USA

2 exposure was considerably large. However, it was found that most of these discrepant cases were incisors and canines. It is difficult to imagine that dental X-ray exposures affected mainly front teeth only. Instead, the results are most likely attributable to solar light exposures in that ultravio let (UV) light exposure, which was reported to induce an ESR signal2'3), is included. In the present report, we describe tooth-position-dependent differences in ESR-estimated dose between buccal and lingual parts of each tooth, differences most likely due to solar light exposures. MATERIALS AND METHODS Briefly, 100 teeth donated from 69 atomic-bomb survivors in Hiroshima who were at least ten years of age at the time of the bombing were chosen from over 300 teeth based on the avail ability of good enamel, DS86 estimated dose 4) of the donors, and multiple teeth from the same survivors to assess possible variation among teeth. Using a disk-shaped diamond cutter cooled with running water, each tooth was first cut into buccal and lingual parts. Subsequently, enamel was isolated and ESR measurement was performed with a Radical Biosensor FR-80 (JEOL, Tokyo). A selective saturation methods) with microwave power of 16 mw and 0.4 mw was used to isolate radiation-related signals from broad background signals ''6). To calculate 60Co gamma-ray equivalent dose from ESR signal intensity, an external calibration curve method, consisting of pooled enamel samples exposed to 0.1 to 4.0 Gy in air (corresponding to to 3.92 Gy in enamel), was utilized. Repeated measurements of 32 samples showed that the mean for the difference between the two measurements is 8.7%'). Additional details have been previously described'. Each ESR measurement of tooth samples from the survivors consisted of 6 consecutive scans. All the samples were measured once to produce the results shown in Figure 1 and Tables 1 and 4. As for multiple samples from the same donors, two additional measurements were performed almost 3 years later for better estimation of contributing dose by UV light. The results were in close agreement and means ± SDs were shown for these samples (Tables 2 and 3). RESULTS Among 100 teeth subjected to ESR measurement, two failed to produce ESR patterns suit able to extract the radiation-related signal. Consequently, results for 98 teeth were analyzed in this report. Although the teeth used in this study were not selected to match DS86 doses by tooth positions, the average lingual doses did not differ significantly among teeth at different positions (i.e., first incisors = 1.51 ± 0.48 Gy, second incisors = 1.17 ± 0.93 Gy, canines = 1.03 ± 1.05 Gy, first bicuspids = 1.48 ± 0.83 Gy, second bicuspids = 1.29 ± 1.17 Gy, large molars and wisdom teeth = 1.30 ± 0.95 Gy, and a grand average of 1.30 ± 0.93 Gy). Therefore, in the following analyses, differences between buccal and lingual doses were compared according to the tooth position but not to the dose that each tooth received.

3 Excess doses in buccal samples compared with corresponding lingual doses of the same teeth are shown in Fig. 1 according to the tooth position. As for front teeth (incisors and canines, Figs. IA, 1B, and 1C), distribution is highly skewed, namely, a modal peak lies at 0.1 to 0.2 Gy, but the majority of the cases (32 teeth out of 35 in total) showed positive values of excess buccal doses (Fig. 1D). In contrast, large molars and wisdom teeth showed a roughly symmetrical distri bution with a modal peak at 0 to 0.2 Gy and only slightly more than one-half (26 teeth out 41) showed positive values of excess buccal doses (Fig. 1G). Results for first bicuspids (Fig. 1E) were similar to those of front teeth, namely, 12 out of 13 teeth showed positive values of excess buccal doses, whereas results for second bicuspids (Fig. 1F) looked rather closer to those of large molars and wisdom teeth than those of front teeth except for one sample. The mean excess dose in buccal parts was largest in first incisors and decreased gradually from 0.48 ± 0.30 Gy (first incisors) to 0.04 ± 0.18 Gy (large molars and wisdom teeth) following tooth position in the mouth as summarized in Table 1. Such a declining trend is readily Figure 1. Histograms of excessive doses in buccal samples compared with lingual doses for each tooth estimated by ESR. A; first incisors (n = 15), B; second incisors (n = 10), C; canines (n = 10), D; all front teeth (incisors and canines) (n = 35), E; first bicuspids (n = 13), F; second bicuspids (n = 9), G; large molars and wisdom teeth (n = 41).

4 appreciated by exposure to UV light contained in solar light as recently reported for teeth from people without known history of exposure to radiation 3). Buccal doses were significantly larger than corresponding lingual doses for first incisors, second incisors, canines, and first bicuspids but not for second bicuspids, large molars and wisdom teeth. The grouped average of the buccal doses for front teeth (i.e., incisors and canines) was significantly larger than the corresponding lingual doses, but not for bicuspids or back teeth (i.e., bicuspids, large molars, and wisdom teeth) (Table 1). It is now clear that buccal parts of incisors, canines, and first bicuspids are affected by solar light exposures. Then, how do we know whether the lingual parts of these teeth are also affected by solar light exposures? Nine of the 69 donors provided two or more teeth, including at least one first incisor or bicuspid and one large molar or wisdom tooth. Lingual doses of the first incisors were compared with the same donors' lingual doses of large molars or wisdom teeth, which are least likely affected by factors other than atomic-bomb gamma rays and, thus, can be regarded as Table 1. Summary of excess buccal doses according to tooth positions Table 2. Summary of excess lingual doses of first incisors compared with lingual doses of large molars or wisdom teeth from the same five donors.

5 representative of true radiation dose. The mean difference was 0.34 ± 0.18 Gy, which is statisti cally significant by paired t-test (p < 0.01) (Table 2). Thus, apparently lingual parts of first inci sors also seem to be affected by solar light exposures. From the results, we deduce that dose overestimation for buccal parts of first incisors would be nearly 0.8 Gy (i.e., 0.48 Gy plus 0.34 Gy). Furthermore, lingual doses of first bicuspids had doses larger than those of large molars or wisdom teeth from the same donors by 0.07 ± 0.02 Gy (Table 3). Although the number of cases is only three, the difference was also statistically significant by paired t-test (p < 0.01). Conse quently, buccal parts of first bicuspids would have overestimated dose by nearly 0.3 Gy (i.e., 0.24 Gy plus 0.07 Gy). One might speculate that front teeth in upper jaw are exposed to solar light more often than those in lower jaw, giving rise to excess buccal doses larger in upper teeth, on the average, than in lower teeth. Present results did not support the notion, however (Table 4). In any of the front teeth, either separately or as a group, the excess buccal doses were not significantly different between upper and lower jaw teeth although the means tended to be slightly larger in upper teeth. Table 3. Summary of excess lingual doses of first bicuspids compared with lingual doses of large molars or wisdom teeth from the same three donors. Table 4. Comparison of excess buccal doses of front teeth between upper and lower jaws

6 DISCUSSION Present results clearly demonstrate that it is important to take into account tooth position when tooth enamel ESR is used for retrospective dosimetry. For incisors, canines, and first bicus pids at least, it is essential to consider that not only buccal doses but also lingual doses are contaminated by the effects of UV light exposures. As the dose detection limit by tooth enamel ESR is reported to be as small as 0.1 Gy'), it is crucial to exclude the effects by UV exposures for small-dose estimation. The first obvious choice is to avoid front teeth for ESR measurement. If this is not possible, however, we should use only lingual parts of such teeth as suggested by Ivannikov et a13) while keeping in mind that overestimation by as much as 0.5 Gy may occur in these samples (Table 2). Elimination of UV-affected parts of enamel may be another possibility to be pursued. As penetra tion of UV light is very limited compared with that of high-energy gamma rays, repeated gentle etchings of enamel surface followed by ESR measurements might result in preferential elimina tion of UV effects while leaving gamma-ray effects unchanged. As for 254-nm UV, the enamel thickness required to reduce the UV dose to 37% is reportedly 105 ± 10 µm3). Thus, dissolving 200 µm of enamel surface is expected to eliminate nearly 90% of UV-contaminated parts while preserving a substantial amount of enamel for ESR measurement of gamma-ray dose. Very unfortunately, however, UV light from the sun which reaches the earth's surface consists of wavelengths > 300 nm and is likely to penetrate more deeply, a corresponding enamel thickness reportedly 280 ± 70 µm3). Therefore, dissolving twice this thickness from the surface (i.e., 500 µm) may not be practical when we consider the thin structure of front teeth. An alternative approach is to measure buccal and lingual doses of front teeth from people without known history of radiation exposure, and use the excess buccal dose as an indicator of exposures to UV light contained in sunlight. Subsequently, when the excess buccal doses in front teeth from radiation-exposed individuals are negligible, we may disregard contamination by exposure to the UV light, whereas when the excess buccal doses are considerably large, contami nation by exposure to the UV light is obvious, and the excess buccal doses can be used to esti mate the contributing dose by the UV light. Actually, the first case reported in Table 2 showed a small but negative value of excess buccal dose in the first incisor (i.e., Gy) (this was a lower incisor), and lingual dose of the same tooth was very similar to that of his large molar (incisor dose was only 0.09 Gy larger than molar dose, Table 2). In contrast, the other four cases presented in Table 2 (lines 2 to 5) had, in their first incisors, excess buccal doses of 0.12 Gy, 0.64 Gy, 1.03 Gy, and 0.77 Gy, and their buccal doses of first incisors were larger than those of large molars by 0.50 Gy, 0.98 Gy, 1.29 Gy, and 1.31 Gy, respectively. Although the two parameters, excess buccal dose in first incisors and difference in buccal doses between first incisors and large molars, may not appear to be linearly related, further studies are required to test the possibility. Last, to eliminate UV-affected parts of the enamel, it should be mentioned that additive dose method (or internal calibration method), commonly used for dose estimation from ESR signal intensity, is destructive and cannot be applied. The present external calibration curve method, consisting of pooled enamel samples, is superior because the original ESR signals of the samples

7 can be maintained. However, the external calibration curve method requires minimizing sample to-sample variation of dentin contamination. The recent development of the alkaline method to denature dentin for subsequent isolation of enamel8) seems to be an easier way to reduce such variation. ACKNOWLEDGEMENTS This publication is based on research performed at the Radiation Effects Research Founda tion (RERF), Hiroshima and Nagasaki, Japan. RERF is a private foundation funded equally by the Japanese Ministry of Health and Welfare and the US Department of Energy (DOE) through the US National Academy of Sciences. JFK is a visiting scientist at RERF and is supported by grant DE-ACO5-76OR0003 from the US DOE Health Physics Faculty Research Award Program administered by Oak Ridge Associated Universities. REFERENCES 1. Nakamura, N., Miyazawa, C., Sawada, S., Akiyama, M. and Awa, A. A. (1998) A close correlation between elec tron spin resonance (ESR) dosimetry from tooth enamel and cytogenetic dosimetry from lymphocytes of Hiroshima atomic-bomb survivors. Int. J. Radiat. Biol. 73: Romanyukha, A. A., Wieser, A. and Regulla, D. (1996) EPR dosimetry with different biological and synthetic carbonated materials. Radiat. Prot. Dosim. 65: Ivannikov, A. I., Skvortzov, V. G., Stepanenko, V. F., Tikunov, D. D., Fedosov, I. M., Romanyukha, A. A. and Wieser, A. (1997) Wide-scale EPR retrospective dosimetry: Results and problems. Radiat. Prot. Dosim. 71: US-Japan Joint Reassessment of Atomic Bomb Radiation Dosimetry in Hiroshima and Nagasaki (1987) Final Report, Vol.1, Ed. W. C. Roesch, Radiation Effects Research Foundation, Hiroshima. 5. Ignatiev, E. A., Romanyukha, A. A., Koshta, A. A. and Wieser, A. (1996) Selective saturation method for EPR dosimetry with tooth enamel. Appl. Radiat. Isot. 47: Nakamura, N., Miyazawa, C., Akiyama, M., Sawada, S. and Awa, A. A. (1996) Biodosimetry: chromosome aber ration in lymphocytes and electron paramagnetic resonance in tooth enamel from atomic bomb survivors. World Health Statist. Quart. 49: Edgersdorfer, S., Wieser, A. and Muller, A. (1996) Tooth enamel as a detector material for retrospective method for EPR dosimetry. Appl. Radiat. Isot. 47: Nakamura, N. and Miyazawa, C. (1997) Alkaline denaturation of dentin (A simple way to isolate human tooth enamel for electron spin resonance dosimetry. J. Radiat. Res. 38: