1. Introduction
The assessment of internal exposure of emergency responders at nuclear accidents is important for evaluating the occupational doses of them as well as the effectiveness of countermeasures, such as pharmacologic thyroid blocking by oral potassium iodide. In addition, the results of the assessment with environment monitoring, at the same time and in the same place, are invaluable for evaluating radiation doses in the general population after nuclear accidents, and occasionally provides important and fundamental information about the intake of biokinetic radionuclides. Indeed, we previously reported that the thyroid iodine-131 (
131I) activities derived from thyroid counter measurements were in good agreement with environment monitoring data, the results also implied a faster kinetics of iodine metabolism in younger adults as previously suggested by Hänscheid et al. [
1,
2].
Recently, however, a few groups have reported a difference between the intake amounts derived from direct measurements and the estimated intake from the air concentrations [
3,
4,
5]. In their reports, estimated intake ratios of
131I to radiocesium (cesium-137 (
137Cs) or cesium-134 (
134Cs)) from in-vivo thyroid measurements were significantly smaller than the ratios derived from environmental measurements. They concluded that the estimated intake amounts of
131I from environmental monitoring data were overestimated because the transfer coefficient, 0.3, from blood to thyroid useby the International Commission on Radiation Protection (ICRP) in the biokinetic model was too high for the relatively low thyroid iodine uptake in the Japanese population. Meanwhile, Morita et al. reported that when the intake scenarios were taken into account, their internal radioactivity assessed by a whole body counter examination comparatively agreed with the predicted airborne radioactivity [
6]. As regards to the intake scenario, Kurihara et al. and Kim et al. assumed an acute intake of 60% elemental iodine vapor and 40% particulate aerosols via inhalation on 15 March 2011 [
3,
4]; the assessments made by Hosoda et al., assumed a similar acute intake scenario using only inhalation of elemental iodine vapor [
5]. Morita et al. also assumed acute inhalation, however, the date and duration of the particulate aerosols inhaled during the stay in Fukushima were specified by using each subject’s behavioral record [
6]. All the groups used the same ICRP biokinetic model, hence the relationship between in-vivo measurements and environment monitoring data remained unclear at the Fukushima Daiichi Nuclear Power Plant accident.
To clarify these inconsistencies, we investigated the
131I and tellurium-132 (
132Te) body burdens of the same subjects used in our previous report, whose intake scenario was clearly available [
2]. Technically, the gamma-rays from
132I were strongly affected to determine
134Cs and
137Cs body burdens. Although the interference declines with time, it is an issue for early-stage measurements, these were shown by the measurements from the whole body counter that was equipped with NaI(Tl) detectors, immediately after the Fukushima accident [
7]. Additionally, the gamma ray 637 keV from
131I might have an effect that may determine the
134Cs and
137Cs activities [
8,
9]. On the other hand,
132Te body burdens could be more clearly determined. The half-lives of
131I and
132Te are similar when compared to
134Cs or
137Cs, so that the estimated intakes from environment monitoring data could be less susceptible to intake scenarios if the ratios of
131I to
132Te were used. Therefore, the ratio of
131I to
132Te was better suited for this investigation than the ones of
131I to
134Cs or
137Cs. In consideration of the consistency of the thyroid
131I activities, previously reported by us, biokinetic calibration which will be described later in detail, was applied to the measurement data of the whole body counter to determine the
131I and
132Te body burdens. The consistency between the whole body counter measurements and several environment monitoring data are discussed using the ratios of
131I to
132Te and the 24-h thyroid uptakes.
4. Discussion
In the present study, we have determined the 131I and 132Te body burdens of five men of a six-member team that comprised the second DMAT of Fukui Prefectural Hospital. The biokinetic calibration was introduced to consider the thyroid autoregulation that plays an important role on the subjects who are on a rich iodine diet. Using the biokinetic calibration results, the arithmetic and weighted averages of the ratio of 131I to 132Te derived from the subjects were 1.63 ± 0.55 and 1.02 ± 0.05 from the biokinetic calibration 131I and 132Te, and 2.02 ± 1.32 and 1.00 ± 0.05 from the biokinetic (131I) and homogenous (132Te) calibrations.
In the previous study, the soil measurement was performed at Tokiwa Junior High School, which is approximately 4 km to the east of the Tamura City Sports Park, the decay corrected ratio of
131I to
132Te was 1.92 ± 0.03 [
21]. On the other hand, the air dose rate measurement that was conducted at 15:51 on 15 March 2011 in the Abukumakogen Service Area (SA), which is approximately 6 km to the south of the Tamura City Sports Park, the decay corrected ratio of
131I to
132Te was 0.78 [
22]. In the present study, the arithmetic and weighted averages of the ratio of
131I to
132Te derived from the subjects were 1.63 ± 0.55 and 1.02 ± 0.05 from the biokinetic calibration
131I and
132Te, and 2.02 ± 1.32 and 1.00 ± 0.05 from the biokinetic (
131I) and homogenous (
132Te) calibrations. Here, we assume that the physicochemical form of
131I was 60% elemental iodine vapor and 40% particulate aerosols (absorption type F) [
4] and the physicochemical form of
132Te was 100% particulate aerosols (absorption type F). Information regarding the rate of absorption in the lungs for
132Te was not available, so that the deposited
132Te in the lungs could have remained there [
23]. Therefore, in this case, we should adopt the ratios from the biokinetically calibrated
131I to the homogenous calibrated
132Te, because the homogenous calibration results included the total residual activities in the lungs independent of the physicochemical forms. We never considered the respiratory tract, the factors 0.8 (=0.4 from aerosol mixing rate times 0.5 from exhaled correction plus 0.6 from vapor) for
131I and 0.5 for
132Te were smaller than the ratio [
4], this can be compared to the air concentration data from the environmental monitoring. As the result, arithmetic and weighted averages of the ratio of
131I to
132Te were 1.26 ± 0.83 and 0.63 ± 0.03, where 1.26 = 2.02/(0.8/0.5) and 0.63 = 1.00/(0.8/0.5), respectively. Therefore, the result was in good agreement with the result of the air dose rate measurement at Abukumakogen, SA. Although, our result was slightly lower, it was not inconsistent with the soil measurement collected at Tokiwa Junior High School. Accordingly, this result supports that environmental measurements could be applied for estimating the intake amounts of the human subjects as in the past, even if the diet of the subjects customarily includes moderate to large quantities of foods rich in iodine.
The estimated total uptake of
131I was not significantly dependent on the models, within a range of 1000 to 7000 Bq; however, each value fluctuated within 0.4- to 2.1-fold range. Morita et al. reported their whole body counter was calibrated by using the BOMAB phantoms, so that their results seemed to be homogenously calibrated [
6,
7]. They mentioned the results of the difference of the
131I/
137Cs ratios between their human measurements and the Worldwide Version of System for Prediction of Emergency Dose Information (WSPEEDI-II) simulations, which predicted the spatiotemporal distribution of activity concentrations in ground-level air from estimated release rates of radionuclides from the reactors. Whereas the accrual internal radioactivity assessed by the whole body counter examination comparatively agreed with the predicted airborne radioactivity by the WSPEEDI-II simulation. Our results suggest that the difference of the
131I/
137Cs ratios may be caused by the calibration model. We should also mention that, as described in the introduction section, earlier measurements by Morita et al. might be influenced by the overlap of gamma ray peaks from
132I and
134Cs, and
132I and
137Cs. This may have contributed to the discrepancy.
The inconsistency in the direct comparison between the in-vivo measurements and the soil sampling data was pointed out by Hosoda et al. [
5]. In the present study, the deposition density 5.33 × 10
2 kBq/m
2 derived from the soil measurements at Tokiwa Junior High School by Endo et al. [
21] could be a comparable result of soil measurements. Under the assumption that the deposition was dry, i.e., the deposition rate was 10
−3–10
−2 m/s, the estimated time-integrated concentration in air could be evaluated to be in the range of 5.33 × 10
4 to 5.33 × 10
5 kBq s/m
3. The other comparable data are from the United States National Oceanic and Atmospheric Administration (NOAA) simulation results published in the UNSCEAR 2013 report, which estimated dispersion and deposition from reverse modelling of environmental measurements of concentrations in the air, deposition densities, dust samples, and air dose rates [
24]. The NOAA estimated values of the deposition density and the time-integrated concentration in the air at Denso Higashinihon, which is approximately 10 km west of the Tokiwa Junior High School, were 3.6 × 10
2 kBq/m
2 and 9.4 × 10
3 kBq s/m
3, respectively. The results of the deposition density corroborate in both cases; however, the estimated time-integrated concentrations in the air were significantly different. When compared to the present results, the NOAA simulation results are likely to be a better estimation because the inhalation intake was calculated 3.9 kBq under the assumption that the duration of the deposition was the same as the time of exposure to the radioactive plume (2 h) and the ventilation rate was 1.5 m
3/h [
2,
25]. This assumption is consistent with Morita et al. that the accurate estimation of internal doses in the first week after the radiological accident was critical [
6]. The soil was sampled not less than 12 h after the subjects in this study withdrew from the area. The weather data of Japan Meteorological Agency showed that it rained at midnight [
26,
27] after they left. Consequently, the additional wet deposition should be considered when using the soil measurement data. Meanwhile, the air concentration data estimated by an atmospheric dispersion model, such as the one used by NOAA, are comparatively robust to the intake scenario of the human subjects, so that the adjusted integration period, to match the intake scenario of the subjects, may provide a better estimation value, at the same time, the deposition density were in good agreement with both results.
Results of in-vivo measurements
131I, and environment monitoring, coinciding in time and place, were quite limited. To our knowledge, excluding our previous study, only one report from Kurihara et al. is available [
3]. They estimated that the intakes of three workers were 3990–6190 Bq which were derived from the thyroid
131I residual activities, 183–285 Bq, under the assumption of acute inhalation on the morning of 15 March. The air dose rate at the time of exposure to the radioactive plume at their location was increasing from 0.25 to 4.8 μGy/h during 3 h and the atmospheric
131I concentration from the air sampling results at that time was 1.6 kBq/m
3 [
3,
28]. The atmospheric concentration
CA can be calculated by using the following formula [
29]:
where
I0 represents the intake,
B is the individual’s breathing rate, and
is the exposure duration. As the result, we should note that the estimated atmospheric
131I concentrations of 1.4–2.2 kBq/m
3 which were obtained from the thyroid measurements, with the daily-averaged ventilation volume for adult males 0.925 m
3/h [
3], agreed very well with the air sampling measurement. These results imply that there is little doubt about the consistency between the in-vivo measurements and the air sampling results. On the other hand, in the present study, the estimated total uptake of
131I is within the range of 2000–7000 Bq and the air dose rate at the exposed to the radioactive plume was around 2.2 μGy/h [
2]. The relationship between
131I intake and air dose rate was in good agreement with both the Kurihara et al. results and the present study. Consequently, these results may indicate that the atmospheric
131I concentration from either the results of actual measurements, such as air sampling or of the simulation based on the environment monitoring data, could provide optimal estimations of the inhaled
131I body burdens. Additionally, the air dose rate is likely to be a more useful indicator of the
131I inhaled amounts at the Fukushima accident.
The conventional estimate of the protective effect of pharmacologic thyroid blockage by using stable iodine pills is 70% [
2], however this could be estimated to be about 10% if the reduction of the coefficient
λ1 occurs by the downregulation of NIS-mediated iodine uptake. This relatively low protective effect might be the result of the faster kinetics caused by the thyroid autoregulation [
18,
19]. Further investigation in the internal radiation dosimetry may be warranted to assess how many doses would be influenced by this phenomenon, a premature ingestion of the stable iodine may cause a reduction of its protective effect on the persons receiving regular iodine sufficient diets. This result suggests that the timely decision based on the change of the air dose rate may be essential and be more effective in the prophylaxis of the stable iodine administration, as conducted in Miharu town, which is a neighboring town of Tamura City [
30].
We acknowledge that there are several limitations in this study. First, the biokinetic calibration was performed phenomenologically. Indeed, several possible parameter sets exist when only self-consistency on each subject is required. Because there was no information about the relationships between age and NIS-mediated iodine uptake (
λ1) or iodide leak (
λ5), this could suggest the possibility that several separate parameter sets could apply to each subject. However, the parameter set we adopted was fully considered as systematic consistency, such as age ordering of
λ1, within the normal Japanese range of 24-h thyroid uptake, and agreement with each other in estimated absorption. In addition, minimal parameter changes from the default values were conducted. The potential importance of the model is to show that accounting of the iodide leak is the key in obtaining a rational explanation for every relationship between the in-vivo thyroid measurements and the environment monitoring results even if the thyroid uptake in Japanese populations is relatively low, as discussed above. Second, the influence of the stable iodine prophylaxis and nutritional iodine intake does not reflect the biokinetic calibration. However, generalized parameter value of
λ5 is yet to be known. Unusual modifications were needed for the transfer coefficient from the urinary bladder contents to urine, causing large differences in the total body burdens in
Table 2 and
Table 3. Meanwhile, it is well known that urinary iodine excretion exhibits large variations [
31], and recently an intravesical urine vanishment phenomenon has been reported [
32]. The estimated urinary excreted
131I was about 1000–5000 Bq, which was in the range of 0.2–1.1 pg in 21.5 h. This was negligibly small compared to the amount of the stable iodine excretion at the same time, which was in the range of 10 μg to 10 mg per 24 h in general [
31]. These arguments may not be sufficient, and further developments, such as the extension of the range of the Leggett model application to children, were recently reported [
33]. Therefore, further research in this area, especially to determine
λ5 value for the subjects who are regularly receiving iodine sufficient diets, is warranted. Third, the effect of shelter-in-place had never been considered. As described previously, the team advised the residents to take refuge, as soon as possible, immediately after they were aware of the sudden increase of the air dose rate [
2]. It is also possible that the amount of exposure to the subjects depended on where the team members were: in the gymnasium hall or outside [
2]. This might be caused in part to the dispersion of the total body burdens. Finally, the respiratory tract was never considered, so that the estimated intakes reported are realistic absorbed amounts. However, the physicochemical form of the inhaled iodine is not clear, and the estimated intakes were in good agreement with the previous study as shown above. Further investigations are needed.