In this issue (pages 21-29), Hunt et al report studies in which volunteers consumed lobster from the coastal area of west Cumbria and measurements were made of the absorption of technetium-99. Previously, Hunt and colleagues (1986, 1990, 1993, 1998) have published measurements of the absorption of plutonium-239/240 and americium-241 from Cumbrian winkles, mussels and cockles, and the naturally occurring alpha emitter, polonium-210, from crabmeat. Their principal motivation has been to improve estimates of radiation doses to specific critical groups of seafood consumers in the coastal communities; in so doing they have provided valuable information for more general assessments of dose and risk.

Marine discharges of the alpha-emitting actinide nuclides from Sellafield have been dramatically reduced since the 1970s and 1980s and remain low following the introduction of the Enhanced Actinide Removal Plant in 1994 (BNFL 2000). However, the use of this plant to treat stored liquid wastes has led to an increase in discharges of ⁹⁹Tc (half-life 2.1 × 10⁵ y). While the actinides and ²¹⁰Po are concentrated particularly by molluscs, including winkles and mussels, ⁹⁹Tc is seen in greatest concentrations in lobsters (BNFL 2000, FSA/SEPA 2000). Studies suggest that Tc in lobster flesh occurs predominantly as a complex with the metal-binding protein, metallothionein (Knowles et al 1998, Goudard et al 1998).

Hunt et al measured ⁹⁹Tc in urine and faeces from eight volunteers ingesting lobster flesh. Total faecal elimination of ⁹⁹Tc was shown to be similar to intake while urinary excretion was low. Attempts to estimate absorption to blood on the basis of differences between intake and faecal elimination were largely unsatisfactory; clearly, this method works best when absorption is high and subsequent excretion occurs predominantly in urine. To estimate absorption on the basis of urinary excretion measurements, it was necessary to interpret results using data reported by Beasley et al (1966) for the urinary excretion of ^{95 m}Tc by volunteers after administration as pertechnetate. These data showed that urinary excretion over 7 days after intravenous injection accounted for 34% of the administered activity. Thus, an estimate of absorption of ⁹⁹Tc from lobster flesh could be made on the assumption that the total reaching the blood was three times the levels measured in bulked 7 day urine samples. However, as discussed by Hunt et al, the time-course of excretion reported by Beasley et al (1966) differed from that observed for the two volunteers ingesting lobster for whom 24 h urine samples had been analysed separately. Urinary excretion during the first day after administration as the pertechnetate was a substantially greater proportion of total urinary excretion over 7 days after administration of ^{95 m}Tc as pertechnetate than after consumption of ⁹⁹Tc in lobster flesh. This difference is largely attributable to rapid excretion of Tc entering the circulation as pertechnetate before binding to proteins and other ligands in blood and other tissues can occur. In addition, a slow rate of absorption to blood from lobster could have contributed; urinary excretion by one of the volunteers given lobster was lower on the day of the meal than on the following day. Hunt et al present two estimates of absorption of ⁹⁹Tc from lobster: first, applying the pertechnetate data directly (i.e. assuming 34% excretion over 7 days) and second, assuming that the rapid component of excretion from the pertechnetate study does not apply (resulting in an assumed 6.7% excretion over 7 days). The fractional absorption (f₁) values obtained were 0.2 ± 0.1 and 0.05 ± 0.02 (rounded to one significant figure).

Following their normal practice, Hunt et al give full details of experimental results to facilitate data analysis and interpretation. An alternative approach to the use of the data of Beasley et al (1966) might be to compare ratios of urinary excretion during the first 2 days and the following 5 days for the two studies. These ratios were 3-5 for the lobster study and 8 for the pertechnetate study. On this basis, perhaps half of the early rapid excretion in the pertechnetate study can be regarded as peculiar to pertechnetate, and excluding this would suggest a value of 19% total excretion over 7 days. The f₁ value for ⁹⁹Tc from lobster would then be 0.1 ± 0.03, which lies between the two values given by Hunt et al.

The International Commission on Radiological Protection (ICRP 1993) currently uses an f₁ value of 0.5 to calculate dose coefficients for the ingestion of radioisotopes of Tc by adult members of the public. This estimate of absorption was based on data from animal studies showing that while Tc administered as pertechnetate was well absorbed, incorporation into plant or animal tissues reduced absorption (Sullivan et al 1977, Gerber et al 1989).

A change in f₁ from 0.5 to 0.1 would result in only a 30% reduction in committed effective dose per unit intake (CED) from ingestion of ⁹⁹Tc from 6.4 × 10^-10 Sv Bq^-1 to 4.5 × 10^-10 Sv Bq^-1. This is because the effective dose is dominated in each case by equivalent doses to regions of the alimentary tract: the stomach and colon contribute 87% of the CED with an f₁ of 0.5 and 96% with an f₁ of 0.1. Uncertainties in the calculation of doses to the stomach and colon include assumptions made regarding transit times through these regions and the depth of sensitive cells in the epithelial lining. These issues are being addressed in a revision of the ICRP model of the alimentary tract (Metivier 1998).

Although CED values for ⁹⁹Tc are insensitive to changes in f₁ value, doses to tissues from absorbed ⁹⁹Tc (pure ß emitter) are directly proportional to the fraction reaching blood. Clearly, interpretation of bioassay data is also dependent on knowledge of absorption to blood.

In the calculation of doses to Sellafield critical groups, account can now be taken of specific f₁ data obtained by Hunt et al for ⁹⁹Tc, isotopes of Pu, ²⁴¹Am and ²¹⁰Po. It is noteworthy that in the most recently reported dose estimates for the most exposed Sellafield group of seafood consumers (FSA/SEPA 2000), the contributions from ⁹⁹Tc and actinide nuclides from Sellafield (<100 µSv) was less than that from ²¹⁰Po attributable to discharges from the Whitehaven phosphate processing plant and probably less than the dose from naturally occurring background levels of ²¹⁰Po.