Biodosimetry after accidental radiation exposure by conventional chromosome analysis and FISH.

A 137Cs source was stolen from a radioactive waste depository in Estonia on 21 October 1994 and kept in a private house for 4 weeks. This resulted in the death of one person, acute radiation injuries to four people and exposure of several other people to lower doses of radiation. Analysis of chromosomal aberrations in peripheral blood lymphocytes was used in the assessment of radiation exposure of 18 people involved in the accident. Dose estimation assessment based on the frequencies of dicentric chromosomes was performed both by the standard method and by considering possible dose protraction and non-uniform exposure. Considerable differences in dose estimates were obtained depending on the approach used, ranging from about 1 Gy to almost 3 Gy in the patients most heavily exposed. In view of the deterministic health effects observed in some of the subjects, it was concluded that the dose estimates involving information on dose protraction were more realistic than those obtained by comparison with the standard high dose-rate calibration curve. Chromosome painting analyses using fluorescence in situ hybridization, with a probe cocktail for chromosomes 1, 2 and 4 and centromere detection, were performed in parallel. Good agreement on dicentric chromosome frequencies was observed between the conventional and painting analyses. The frequencies of complete translocations were comparable with the frequencies of dicentric chromosomes. In addition to the complete translocations, a pronounced increase in the frequency of incomplete translocations was observed. Dose estimates performed on the basis of FISH translocation frequencies were consistent with the dicentric analysis.