XA9745549 IAEA_CN-67/10

LEUKEMIA IN THE PROXIMITY OF A GERMAN BOILING WATER NUCLEAR REACTOR: EVIDENCE OF POPULATION EXPOSURE BY CHROMOSOME STUDIES AND ENVIRONMENTAL RADIOACTIVITY

Bettina Dannheim, Anna Heimers, Boris Oberheitmann, Inge Schmitz-Feuerhake, Heike Schroder, Heiko Ziggel

University of Bremen, Dept. of Physics, P.O. Box 330440, D-28334 Bremen,

Abstract The detection of an exceptional elevation of in children appearing 5 years after the start-up of the nuclear power plant KnJmmel in 1983, accompagnied by a significant increase of leukemia cases in adults gave rise for investigations of radiation exposures of the population living near to the plant. The rate of dicentric chromosomes in peripheral blood lymphocytes of 7 parents of leukemia children and 14 other inhabitants in the proximity of the plant was significantly elevated and showed ongoing exposures over the years of operation. This finding gives rise to the hypothesis that chronic leakages by the reactor had occured. This assumption is supported by the identification of artificial radioactivity in air, rain water, soil, and vegetation registered by the regular environmental monitoring programme of the nuclear power plant. Calculations of the corresponding source terms show that the originating emissions must have been well above authorized annual limits. The bone marrow dose is supposed to be originated mainly by incorporation of bone-seeking 0- and a-emitters.

Introduction A cluster of childhood leukemia was observed in the community of Elbmarsch in Northern Germany in 1989-91 (1, 2, 3). All five cases had been living within a distance of 5 kilometers from the nuclear power plant Kriimmel (KKK), located at the river about 35 km south-east of the city of . Since 1994, four additional cases in the 5 km surrounding of KKK (Fig. 1) appeared. All these cases occured at ages £ 10 years, with five of them at very early ages (1.8-4 years), see Table 1. Furthermore, a leukemia in a young adult and an aplastic anaemia in a child had been diagnosed there. The mean number of children (up to the age of 14) living in the 5 km surrounding of KKK between 1986-95 was about 5400 (4). Corresponding to the West German Registry of Childhood Cancer (5), the mean incidence of acute leukemia between 1980 and 1990 was 4.3-10-5 a1 in individuals aged < 15 years. The expected number of leukemia cases in children in the 5 km region around KKK for the period 1990-96 would be 1.6, therefore the elevation of childhood leukemia in that region corresponds to a factor of 5.6. The nuclear power plant is a 1300 MW« boiling water reactor - the largest one in the world - and started operation in 1983. In the vicinity of this plant the nuclear research institute GKSS (Gesellschaft fur Kemenergieverwertung in Schiffbau und Schiffahrt), which was established in 1958, operated two research reactors of 5 and 15 MW (Fig. 1). The governments of the Federal States of and Schleswig-Holstein had set up a board of experts to identify possible causes for the observed leukemia cluster. Potential risk factors such as X-rays, chemicals and previous diseases of the affected families were investigated and could be excluded. None of the parents had been exposed to occupational or unusual medical irradiation. One family, however, was found to be living in a house with a mean radon concentration of 450 Bq/m3 in their living rooms. However, the only common factor remaining was proximity to the two nuclear establishments. With respect to KKK not only the spatial relationship of the observed leukemia cluster to the nuclear power plant is remarkable but also the temporal correspondence. The first case of malignant blood diseases was diagnosed 5 years after KKK had been commissioned. In 1994, a retrospective epidemiological study by Hoffmann and Greiser (6) showed an elevation of leukemia also in the whole population in the vicinity of KKK. For the period 1984-93 they found a statistically significant elevation for males (+56%) within a radius of 5 km around the nuclear power plant. The increase was restricted to the period 1989-93, i.e. beginning 5 years after the start-up of KKK as well. Due to these facts, a careful investigation of the possible relationship between the observed cluster and the local nuclear facilities was imperative, although whole body counting of inhabitants performed in 1991 gave negative results and analyses concerning the emission of long-lived fission products failed to demonstrate a consistent pattern of increased radionuclide concentrations in soil and vegetation (7). However, there were several isolated findings of elevated activities in different environmental media at different times (8). To examine whether the population living in the vicinity of KKK was subjected to elevated exposures of ionising radiation, an analysis of dicentric chrotjiosome aberrations in lymphocytes of the peripheral blood was carried out in a group of inhabitants of the Elbmar&h community. In parallel, analyses of data on environmental radioactivity measurements were conducted.

37 Figure 1: Map of the surrounding of the nuclear power plant Krummel. Black Points indicate homes of leukemia cases and aplastic anaemia with year of diagnosis. Circles show the distance to KKK. (One more case of 1994 is not registered because the location is unknown.)

Assessment of radiation exposurej)y investigation of dicentric chromosome aberrations The retrospective assessment of radiation exposure by chromosome aberration analysis comprised 21 individuals (19 females and 2 males), inhabitants or former inhabitants of the Elbmarsch community. All of them were residing within a distance of 5 km in the southern direction from KKK. At the time the blood samples were taken (in 1992,1993 and 1995) these volunteers were aged 28-45 years. Four women were selected because of the fact that they had settled in the village after 1986 (Table 2A, nos. 8-11), as had one of the leukemia families (Table 2A, no. 4). Seven subjects were parents of the children having diagnosed leukemia in the community of Elbmarsch. The control group consisted of 25 healthy adult persons (9 females and 16 males) living in the city of Bremen, 100 km south-west of Hamburg. At the time of blood sampling between 1988 and 1994 they were aged 17-57 years. A detailed questionnaire was completed by each of the subjects. Exclusion criteria were previous occupational exposures, greater than average diagnostic medical irradiation, or exposure to chemical mutagens. Smokers (more than 10 cigarettes per day) were also excluded. To evaluate the rate of dicentric chromosomes and centric ring chromosomes lymphocyte cultures and slide preparations were made according to standard cell cycle controlling methods (9). The collection of metaphases was facilitated by a semi-automatic computerized system which includes data management (METASYSTEMS, Sandhausen, Germany). Only first division metaphases with 46 centromeres were analysed and all structural aberrations were registered. The results are given in Table 2.

38 Compared to the laboratory control (0.46x1 Or3) the investigated population of Elbmarsch showed a highly significant elevation of dicentric chromosomes with a rate of 1.77X10*3 dicentric chromosomes/metaphase (p<0.01 (10)). The highest individual number of dicentrics (excluding case no. 3 of Table 2A) was found to be 5 dicentric chromosomes in 1,000 metaphases. Particularly noteworthy was the existence of cells with two dicentrics in the Elbmarsch group (Table 2A, subjects 5, 8, 10, and 12), in contrast to the control group (no cells with more than one dicentric). Additionally, there was one metaphase with 6 dicentrics (Table 2A, subject 3). The aberrations did not follow Poisson distribution but showed a significant overdispersion (11). This was also valid for the entire Elbmarsch population which was investigated, even if the multiaberrant cell is excluded (Table 3). Although the results of the chromosome aberration analysis give evidence that the population of the Elbmarsch community has been severely exposed to ionising radiation in the past (12) an extrapolation of the whole accumulated dose is not adequate since the exposure conditions are unknown.

Fission and activation products in environmental samples Several media in the environment of KKK are routinely controlled for long-lived gamma emitters (13, 14). The concentration in rain water is measured periodically at three locations near KKK. One of these sampling stations which is situated in a distance of 2.2 km in the main wind direction of KKK shows a decreasing but permanent elevation of the Cs-isotopes 134 and 137 since 1986. These isotopes were elevated compared to the other locations of the KKK monitoring programme and the Cs 137-concentration was at least lOtimes higher than that at other measuring points in Northern Germany from the 3rd quarter of 1986 until 1993 (15). Thus, an exclusive causation by Chernobyl is unlikely. In addition, the existence of the isotope Cs 134 in these samples shows that the contaminations can not go back to the fallout from nuclear weapons testing because of the short half-life of Cs134(2.1y).

Calculations and measurements demonstrate that the permitted releases of radioactivity by German nuclear facilities will not lead to detectable contamination in the surrounding (16). Therefore, the Cesium findings in rain water are an indicator for repeated releases of nuclides above permitted limits. Elevated Cs-emissions are confirmed by the results of measurements of dry fallout at different locations near KKK (13). Compared to other locations in Germany the Cs 137 aerosol concentration in air near the ground in the vicinity of KKK was significantly elevated in different quarters in the period 1985-95. It should be noted that these measurements were carried out by three different laboratories; thus, systematic measurement errors can be excluded. Additionally, remarkable concentrations of Sr 90 were found in 1984 and 1988 which are also not explainable by other sources than by emissions of KKK.

Table 1: Observed cases of leukemia and aplastic anaemia in the 5 km surroundings of the nuclear power plant

Date of Age at No. Born Sex Disease Diagnosis Diagnosis

1 9/82 f AA 12/89 7 2 8/86 f ALL 2/90 3.5 3 2/81 m ALL 3/90 9 4 3/81 m ALL 4/90 9 5 3/89 f ALL 1/91 1.8 6 1970 m AML 4/91 21 7 9/88 m ALL 5/91 2.7 8 m ALL 1994 9 ca. 1991 m ALL 1995 4 10 1985 m ALL 6/95 10 11 1993 m ALL 6/96 3

AA: aplastic anaemia ALL: acute lymphatic leukemia AML: acute myeloid leukemia m: male f: female

39 Table 2: Results of chromosome aberration analysis in adult residents of Elbmarsch (A) and in controls (B)

A) Elbmarsch group

No. parent of resident in date of blood No. of No. of rate of leuk. Elbmarsch sampling analysed dicentrics dicentrics x child metaphases 10-3

1 yes since <1984 Jan. 1992 1005 2 2.0 2 yes since <1984 Jan. 1992 390 2 5.1 3 yes since <1984 Jan. 1992 1000 3* 3.0 4 yes <1984-1991 Jan. 1992 1001 2 2.0 5 yes since < 1984 Jan. 1992 664 2« 3.0 6 yes since <1884 Oct. 1995 1010 2 2.0 7 yes since <1984 Oct. 1995 1010 0 0,0 8 no since 1988 Apr. 1993 1002 5*** 5.0 9 no since 1987 Apr. 1993 1014 0 0.0 10 no since 1987 Apr. 1993 1097 3.6 11 no since 1987 Apr. 1993 1034 1 1.0 12 no since <1984 Jun. 1993 1005 3*** 3.0 13 no since <1984 Jun. 1993 1110 0 0.0 14 no since <1984 Jun. 1993 1003 1 1.0 15 no since <1984 Jun. 1993 1005 1 1.0 16 no since < 1984 Jun. 1993 1002 2 2.0 17 no since <1984 Jun. 1993 1011 2 2.0 18 no since <1984 Jun. 1993 1008 0 0.0 19 no since <1984 Jun. 1993 1007 2 2.0 20 no since <1984 Jun. 1993 1004 2 2.0 21 no since <1984 Jun. 1993 1009 0 0.0

Total 20391 36 1.77 ± 0.33s

*) Excluding one multiaberrant cell with 6 dicentrics. **) One cell contained 1 tricentric which was counted as 2 dicentrics ***) Including one cell with 2 dicentrics. §) Standard error of the mean

B) Control group

No. date of blood No. of analysed No. of dicentrics rate of dicentrics x sample metaphases 10-3

1 -25 1988-1995 19775 0.46 ± 0.15s

§) Standard error of the mean

40 Table 3: Intercellular distribution of dicentric chromosomes in Elbmarsch inhabitants

No.

0 1 2 3 4 5 6 S2/Y*

incl. multiaberrant cell observed 20358 28 4 0 0 0 1 1.90 92.26 expected 20349 41.91 0.04 0 0 0 0

excl. multiaberrant cell observed 20358 28 4 0 0 0 0 1.22 22.58 expected 20354 35.94 0.03 0 0 0 0

*) relative variance (Y is the mean value) **) u > 1.96: overdispersion is significant (p < 0.05)

Since the start-up of KKK measurements of environmental samples show elevated activities of the nuclides Cs 137 and Sr 90 in soil and plants (13, 14). The Chernobyl accident had only lead to marginal increases of Sr 90 in the German environment (16). Nevertheless, the concentration of this isotope rose repeatly by several Bq/kg (dry mass) against the former fallout background in the grass around KKK. The highest Sr 90 contamination was registered in August 1987 at a location in 10 km distance to KKK, containing 30 Bq/kg (dry mass) which is more than 10fold of the normal concentration in Germany at that time (17). The highest concentration of Cs 137 in grass (103 Bq/kg) was measured in 1988 in a distance of 2 km from KKK (13). This corresponds to an about 100fold increase against the normal vallje in the vegetation (17). The environmental monitoring programme for KKK does not demand nuclide specific and continuous measurements of pure p- and a-emitters in the controlled media, except Sr 90 in the air, nor any measurements of short-lived gamma and p-emitters. Nevertheless, there have been occasional findings of short-lived fission and activation products as well as radioactive corrosion products which are also not explainable by the Chernobyl accident. Only after damage of the fuel rods these nuclides would be detectable outside the plant. In these cases leakages of the primary cooling water will be accompanied by releases of Pu and other transuranium isotopes (18).

Discussion Leukemia elevations have been found repeatedly in the population in the vicinity of nuclear establishments (19, 20, 21). In the case of the Pilgrim nuclear reactor in Massachusetts, a correlation between the leukemia rates in the vicinity of the plant and official information on releases of radioactivity could be established and a cause- effect relationship was derived (22). In Western Germany, an incidence study of childhood leukemia near nuclear establishments in the former countries of the federal republic lead to negative results (24). However, with respect to the innermost of the different circular regions considered (< 5km) and the group of children < 5 years at diagnosis there was a 3fold significant elevation compared to the control regions (23). The study of Keller and coworkers (23) was restricted to the period 1980-1990, consequently most of the cases of childhood leukemia near KKK were not included. These children became diseased predominantly in very young ages (Table 1). The hypothesis of radiation induction is thus supported by the well-known fact of a higher sensivity of individuals in the prenatal stage and in very early ages/Furthermore, the predominance of diseases in boys (Table 1) corresponds to the findings in the A-bomb survivors showing a male to female ratio of 2:1 in the radiation-induced excess cases (24), while the normal ratio is 1.3:1 (5). Further evidence of the hypothesis of radiation-induced leukemia in the Elbmarsch region is added by investigations of the latency periods for leukemia found in exposed cohorts. Children at the age of 0-15 years receiving X-ray therapy for ringworm of the scalp showed a maximum rate of leukemia about 4 years after irradiation and 50 % of all cases occured up to this point (28, 29). This corresponds to the distribution seen in the Oxford Survey of Childhood Cancers for leukemia after prenatal exposure to diagnostic X-rays (30), and to considerations about the age-dependency of latencies in the Japanese A-bomb survivors (31). Assuming 4 years of latency and referring to the first appearance of childhood leukemia in the proximity of KKK in 1990 (Table 1), a single exposure event in about 1986 could be supposed. However, the ongoing occurence of malignancies in Elbmarsch points to continuous contamination and/or the incorporation of radioactivity with a long biological half-life.

41 By the medical and biological findings discussed in this paper, there is evidence that the population was affected by repeated exposures, because: (i) at least 4 of the children with leukemia were bom later than 1987 (Table 1) (ii) of the well-known decline of the dicentric chromosomes after irradiation (of about 40 % per year) (12) the rates measured in persons of the Elbmarsch do not represent an accumulated exposure. Although - as discussed below - it is not adequate to derive a whole body dose from the results given in Table 2A, a dating back of the measured rates of dicentric chromosomes to a single event in 1986 or earlier would yield unrealistic high original rates (iii) the elevated rate of dicentric chromosomes is rather similar in persons who settled at Elbmarsch before and after 1986 (before 1986: cases nos. 1-3 and 5 with 2.9 103 dicentrics/metaphase; after 1986: cases nos. 4 and 6-9 with 2.3 10"3 dicentrics/metaphase; see Table 2A). Dose estimations based on the observed rate of dicentric chromosomes in the inhabitants of Elbmarsch are not sensible, because neither the time nor the condition of exposures are known. The findings about overdispersion in the distribution of aberrations per cell give, however, additional information about the kind of exposure. Overdispersion of dicentric chromosomes can be caused either by non-uniform or by high LET irradiation. Such distributions have been found in blood samples of persons occupationally exposed to e.g. Plutonium (25), Tritium (26) or Uranium (27). In case of the Elbmarsch population this can not be proved by available data or measurements. The observed overdispersion in the present study is also not to be explained by external gamma- irradiation or other kinds of low dose low LET exposure. The supervising ministry in Schleswig-Holstein has denied a relevant contribution of KKK to the exposure of the population by KKK. They refer to the results of the extensive gamma monitoring in the surrounding. Although there is, indeed, no indication for a severe overexposure by external gamma-irradiation in that region the supposed bone marrow dose may have been caused by incorporation of a- and p-aerosols while the gamma- submersion was negligible. Scenarios which lead to such a composition of organ doses are described for accidents caused by cooling water effluent (18). In addition, fuel rod failures are a well-known problem in nuclear reactors. Being the greatest boiling water reactor of the world, the Kriimmel nuclear power plant is a unique installation. Last year, it was reported in a TV special that there had been severe problems during construction of the pressure vessel related to welding of single elements. The board of experts named for the leukemia study found out information on significant leakages of cooling media from the pressurised system nearly all over the years of operation. The quantities rose to 300 I per hour in the year 1986 but increased also to more than 200 I per hour in 1993 and 1996. Further investigations will consider the relevance of these leakages by analysing the pattern of observed nuclides in the surrounding.

References

1. Dieckmann H. HSufung von Leukamieerkrankungen in der Elbmarsch. Gesundh-Wes 54: 592-596 (1992) 2. Schmrtz-Feuerhake I, Schroder H, Dannheim B, GreU-Buchtmann 1, Heimers A, Hoffmann W, Nahrmann A, Tomalik P. Leukaemia near water nuctear reactor. Lancet 342:1484 (1993) 3. Hoffmann W, Schmitz-Feuerhake I, Dieckmann Hayo, Dieckmann Helga. A cluster of childhood leukemia near a nuclear reactor in Northern Germany. Arch Environ Health, in press 4. Kaatsch P. Pers. communication 1996, Inst. for Medical Statistics and Documentation, Johannes Gutenberg-University Mainz 5. Haaf HG, Kaatsch P, Keller B, Michaelis J. Jahresbericht 1990 des Kinderkrebsregisters Mainz. Institute for Medical Statistics and Documentation, Johannes Gutenberg-University Mainz 1993 6. Hoffmann W, Grosser E. Retrospektive Inzidenzstudie Elbmarsch. Bremen: Bremer Instrtut for Praventjonsforschung und Sozialmedizin (BIPS), 1994 7. Fachbeamtenkommission Niedersachsen/Schleswig-Holstein. Untersuchungen zur Frage der Ursache-Wirkungs-Beziehung zwischen dem Betrieb der kemtechnischen Anlagen KKK und GKSS und dem Auftreten von Kinderteukamien in der Elbmarsch. Report to the Ministry of Social Affairs of the Federal State of Lower Saxony, 1992 8. KKK. Statusbericht zur Umweitradioaktivita't der kraftwerksnahen Umgebung fur den Zeitraum 1981 bis 1990, TeHe Nil. KrOmmel: Kemkraftwerk Krummel GmbH, 1992 9. Heimers A, Schroder H, Lengfekter E, Schmitz-Feuerhake I. Chromosome aberration analysis in aircrew members. Radiat Protection Dos 60:171-175(1995) 10. Kastenbaum MA, Bowman KO. Tables for determining the statistical significance of mutation frequencies. MutRes9: 527-49 (1970) 11. Edwards AA, Lloyd DC, Purrot J. Radiation induced chromosome aberrations and the Poisson distribution. Radiat Environm Biophys 16: 89- 100(1979) 12. Hoffmann W., Schmitz-Feuerhake I. Zur Strahlenspezifitat der angewandten Bkjlogischen Dosimetrie. Berichte des Otto Hug Strahteninstituts, No. 7, Munchen: Munchenar Medizm Verlag, 1993 13. Kemkraftwerk Krummel. Jahresberichte zur Umgebungsuberwachung 1981 -1995 14. Energiesysteme Nord GmbH. RadioaMivttatsuberwachung in der Umgebung von Kemkraftwerken. BehordenmeBprogramm 1983-1993 15. Schmitz-Feuerhake I. Edelgasszenario - Bewertung kurzlebiger Spaltproduktemissionen des KKK im Rahmen der Edelgasthese zur AufklSrung des LeukaTnievorkommerts in der Elbmarsch. Report to the Ministry of the Environment of the German Federal State of Lower Saxony in Hannover. Sept. 6,1996 16. Bundesminister fur Umwett, Naturechutz und Reaictorsicherheit. Allgemeine Verwaltungsvorechriften zu § 45 Strahtenschutzverordnung. Ermittlung der Strahlenexposition durch die AWeitung radioaktiver Stoffe aus kemtechnischen Anlagen Oder Einrichtungen vom 21.2.1990. Bundesanzeigei Nr. 64a vom 31. Mdrz 1990 17. Bundesminister fur Umwelt, Naturschutz und Reaktorsicherheit. UmweltradioaktivitSt und Strahlenbelastung. Jahresberichte Bonn 1987 18. Technischer Oberwachungsverein Norddeutschland. Storfalle mit Aktwitatsfreisetzungen. Gutachten fiber die Sicherheit des Kemkraftwerks KrOmmel. Hamburg, Dezember 1982 19. Roman E, Beral V, Carpenter L, Watson A, Barton C, Ryder H, Aston D. Childhood leukaemia in the West Berkshire and Basinstoke and North Hampshire Health Authorities in relationt o nuclear establishments in the vicinity. Br Med J 294: 597-602 (1987) 20. Gardner MJ, Snee MJ, Hall AJ, Powell CA, Dowries S, Ten-ell JD. Results of a case-control study of leukemia and lymphoma among young people near SeKafield nuclear processing plant in West-Cumbria. Br Med J 300: 423-429 (1990) 21. Goldsmith JR. Nuclear installations and childhood cancer in the UK: mortality and incidence for 0-9-year-old children. 1971-1980. Sci Total Environ 127:13-35 (1992) 22. Morris MS, Knorr RS. Adult leukemia and proximity-based surrogates for exposures to Pilgrim plant's nuclear emissions. Arch Environ Health 51:266-274 (1996)

42 23. Keller B, Haaf G, Kaatsch P. Michaelis J. Untersuchung der Haufigkeit von Krebserkrankungen im Kindesarter in der Umgebung westdeutscher kemtechnischer Anlagen. Schriftenreihe Reaktorsicherheit und Strahlenschutz. BMU-1992-326, ISSN 0724-3316,1992 24. Finch SC, Finch CA. Summary of the studies at ABCC-RERF concerning the late hematologic effects of atomic bomb exposure in Hiroshima and Nagasaki. Radiation Effects Research Foundation. Technical Report TR 23-68. Hiroshima: 1990 25. Tawn EJ, Hall JW. SchofieW 6B. Chromosome studies in plutonium workers. Int J Radiat Biol 47: 599-610 (1985) 26. RommH. Stephan G. Chromosomenanalyse nach beruflicher Tritiumexposition. Institut fur Strahlenhygiene Neuherberg. ISH Heft 126: 104- 105(1988) 27. Scheid W. Traut H: Biologische Dosimetrie und Strahlenqualitat. Strahlenschutz in Forschung und Praxis 29: 149-155 (1987) 28. Albert RE, Omran AR. Follow-up study of patients treated by X-ray epilation for tinea capitis. Arch Environ Health 17: 899-918 (1968) 29. Ron E, Modan B, Boice JD. Mortality after radiotherapy for ringworm of the scalp. Am J Epidemiol 127: 713-725 (1988) 30. Stewart AM, Kneale GW. Age-distribution of cancers caused by obstetric X-rays and their relevance to cancer latent periods. Lancet i: 4-8 (1970) 31. Ichtmaru M, Ishimaru T, Beteky JL, Tomiyasu T, Sadamori N, Hoshino T, Tomonaga M, Shimizu N, Okada H. Incidence of leukemia in atomic bomb survivors, Hiroshima and Nagasaki 1950-71, by radiation dose, years after exposure, age, and type of leukemia. RERF Technical Report 10-76, Hiroshima: Radiation Reserach Effect Foundation, 1977

43