4
Health Effects of Radon Progeny on Non-Lung-Cancer Outcomes

This report focuses on the lung-cancer risk associated with radon progeny. However, other health effects of exposure to radon progeny and to uranium mining in general have also been of concern. The health effects mentioned in the literature are nonmalignant respiratory diseases in underground miners, cancers other than lung-cancer in miners and in the general population, and adverse reproductive outcomes of pregnancies in the wives of uranium miners and in communities adjacent to areas where uranium is mined and milled. The BEIR IV report addressed evidence available through 1987 on these potential health effects. Its appendix V, on nonmalignant respiratory and other diseases in miners, also covered effects on reproductive outcomes.

The evidence available to the BEIR IV committee on each of those issues was sparse. For nonmalignant respiratory diseases, animal studies provided some relevant findings, and a few epidemiologic studies had been reported. The committee noted that silicosis had been documented in silica-exposed populations of uranium miners; the evidence in other fibrotic lung disorders, particularly interstitial fibrosis, was limited. With regard to reproductive outcomes, the BEIR IV report found the data to be "sparse" and characterized the associations as "weak." The report provided little comment on cancers other than lung-cancer.

Since the publication of BEIR IV, additional data on risks of cancers other than lung-cancer have become available from an analysis of pooled data in underground miners (Darby and others 1995). Reports based on ecologic analyses have suggested that indoor radon is associated with diverse malignancies, including leukemias and some solid cancers (Henshaw and others 1990). Only limited information has been reported on nonmalignant respiratory disease and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI 4 Health Effects of Radon Progeny on Non-Lung-Cancer Outcomes This report focuses on the lung-cancer risk associated with radon progeny. However, other health effects of exposure to radon progeny and to uranium mining in general have also been of concern. The health effects mentioned in the literature are nonmalignant respiratory diseases in underground miners, cancers other than lung-cancer in miners and in the general population, and adverse reproductive outcomes of pregnancies in the wives of uranium miners and in communities adjacent to areas where uranium is mined and milled. The BEIR IV report addressed evidence available through 1987 on these potential health effects. Its appendix V, on nonmalignant respiratory and other diseases in miners, also covered effects on reproductive outcomes. The evidence available to the BEIR IV committee on each of those issues was sparse. For nonmalignant respiratory diseases, animal studies provided some relevant findings, and a few epidemiologic studies had been reported. The committee noted that silicosis had been documented in silica-exposed populations of uranium miners; the evidence in other fibrotic lung disorders, particularly interstitial fibrosis, was limited. With regard to reproductive outcomes, the BEIR IV report found the data to be "sparse" and characterized the associations as "weak." The report provided little comment on cancers other than lung-cancer. Since the publication of BEIR IV, additional data on risks of cancers other than lung-cancer have become available from an analysis of pooled data in underground miners (Darby and others 1995). Reports based on ecologic analyses have suggested that indoor radon is associated with diverse malignancies, including leukemias and some solid cancers (Henshaw and others 1990). Only limited information has been reported on nonmalignant respiratory disease and

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI reproductive outcomes. This chapter updates the BEIR IV report's coverage of outcomes other than lung-cancer. DOSES TO ORGANS OTHER THAN LUNG Interpretation of the epidemiologic findings on effects of radon other than in the lungs needs to be based on an understanding of the dosimetry of radon and its progeny. Dosimetric models used for this purpose have extended beyond deposition of progeny in the lung to distribution to and absorption in other organs. Although there has been little work on this aspect of radon dosimetry, the models suggest potentially important doses to skin and lymphocytes. Table 4-1 shows estimated annual absorbed doses to various adult tissues from Rn-222 and its short-lived daughters, for a domestic concentration of 20 Bqm-3 (0.54 pCiL-1). The estimated dose for the basal cells in exposed skin depends heavily on assumed deposition velocity. The upper estimates are comparable with lung doses and might be responsible for a substantial number of skin cancers, most of which would be nonfatal. With assumptions of 125 µGy y-1 to exposed skin of the face and neck, 7.5 µGy y-1 to skin of all other regions, an RBE of 5 for alpha- TABLE 4-1 Estimated annual absorbed doses to adult tissues from Rn-222 and its short-lived progeny for domestic radon concentration of 20 Bqm-3 (0.54 pCiL-1) Tissue Annual dose (µGy y-1) Reference Lung 500 ICRP65 1993; Simmonds and others 1995 Bronchial basal cell nuclei 740 James, personal communication Bronchial secretory cell nuclei 1,680 James, personal communication Skin (basal cells at 50 µm in exposed skin) 50–1,000 85–850 Harley and Robbins 1992 Eatough and Henshaw 1992 Red marrow 0.5–1.3 Harley and Robbins 1992   4–6 Richardson and others 1991   4.5 Simmonds and others 1995   2.2–2.6 James 1992 Bone surface 0.4–1 Harley and Robbins 1992   1.2 ICRP65 1993; Simmonds and others 1995   4.4 James 1992 Breast ~ 1.5 Harley and Robbins 1992   1.2 ICRP65 1993; Simmonds and others 1995 T-lymphocytes: circulating 1 Harley and Robbins 1992 in bronchial epithelium 1,000 Harley and Robbins 1992 Blood 1.1 Harley and Robbins 1992 Liver 2.5 James 1992 Kidney 14.4 James 1992

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI particles, and ICRP59 risk factors for low-LET radiation, Eatough and Henshaw (1995) have estimated that about 1–10% of nonmelanoma skin cancers in the UK can be theoretically associated with domestic radon, including possible interaction with UV radiation. Calculated doses from thoron, available only for lung and bone marrow, tend to be lower than those from radon (Table 4-2). Doses to red marrow from radon are substantially lower but nevertheless lead, by conventional risk estimation, to expectations of contributions to the natural incidence of leukemia. Richardson and others (1991) suggested that about 6–12% of myeloid leukemia in the U.K. could be from radon. Similarly, Simmonds and others (1995) have applied age-dependent radon doses and risks to a national study population of children in Seascale, U.K. From their evaluations, it can be deduced that some 34% of the expected natural childhood-leukemia incidence is attributable to natural radiation, made up of about 20% from natural low-LET radiation and about 14% from natural high-LET radiation (Simmonds and others 1995; COMARE 1996). About 3% of the 14% is due to radon and thoron. James (1992) has estimated from ICRP (1991) risk factors that 1.25% of the additional cancers caused by indoor radon and thoron are expected to be leukemia; the vast majority (98%) of the additional cancers are expected to be in the lung. However, no excess leukemia has been observed in the studies of the mining cohorts. NONMALIGNANT RESPIRATORY DISEASES In addition to radon, potential causes of nonmalignant respiratory diseases in the underground miners include silica (well documented as causing silicosis) γ, blasting fumes, and, in some mines, diesel exhaust. Persons with silicosis are at TABLE 4-2 Estimated doses to adult tissues from thoron and its short-lived decay products for typical domestic thoron concentration corresponding to 212Pb-progeny concentration of 0.3 Bqm-3 (0.008 pCiL-1) Tissue Annual dose (μGy y-1) Reference Lung 34 Simmonds and others 1995 Red marrow 1.5 Stather and others 1986   0.5 James 1992a Bone surface 22 Simmonds and others 1995   3.0 James 1992a Liver 1.7 James 1992a Kidney 10 James 1992a a The doses presented here are one third of those tabulated in James (1992), so as to correspond to a typical PAEC for indoor thoron progeny of 21 nJM-3 (1 mWL) (Nero and others 1990) and 212Pb concentration of 0.3 Bqm-3 (0.008 pCiL-1) (James, personal communication).

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI increased risk for tuberculosis and the development of silicotuberculosis, a fibrotic disorder. On reviewing the epidemiologic evidence on nonmalignant respiratory diseases, the BEIR IV committee (NRC 1988) concluded that the effects of the various inhaled agents in the mines could not be readily separated, and it did not attribute nonmalignant respiratory diseases in underground miners to radon specifically. The principal evidence considered came from surveys of Colorado Plateau and New Mexico uranium miners. A case series recently reported by Archer (1996) provides evidence that uranium miners may develop a fibrotic lung disease distinct from silicosis. About 400 uranium miners had been referred to Archer and colleagues for evaluation of lung disease and of exposure, either for special studies or for the purpose of seeking compensation, and 22 of these were selected as possible examples of radiation-related diffuse interstitial fibrosis. All had evidence of diffuse abnormalities in both lower lung fields. Lung-biopsy specimens from five of the 22 were examined. The findings were not those of silicosis. The 5 men had evidence of interstitial fibrosis and honeycombing, a pattern of destruction found in advanced lung disease. Birefringent crystals consistent with silica were present in trace quantities at most. Archer and others conclude that there is a radiation-caused diffuse interstitial fibrosis in uranium miners. The New Mexico Miners' Outreach Program provides free screening to active and retired miners for mining-related disease (Mapel and others 1996). Cross-sectional data from 1,359 former uranium miners were analyzed by Mapel and colleagues, who examined predictors of lung function and of radiographic abnormality with regression methods. Duration of underground mining was negatively associated with lung function, but the association was statistically significant only for the American Indian miners. Underground uranium-mining experience was a significant predictor of having pneumoconiosis for Hispanic and American Indian former uranium miners. Quantitative estimates of radon-progeny exposure were not available, so the investigators could not assess further the contribution of radon-progeny exposure to the reduction of lung function associated with duration of underground mining. MALIGNANCIES OTHER THAN LUNG-CANCER The dramatic excess of lung-cancer in underground miners exposed to radon progeny has focused emphasis in research, in risk assessment, and in risk management on this cancer. At the time of the BEIR IV report, there was limited information on cancers of sites other than the lung, and it came primarily from the cohort study of Colorado Plateau miners (Waxweiler and others 1976). Although statistical power related to this single cohort was limited, cancers were not in apparent excess for sites other than the lung (Waxweiler and others 1976). Nor had understanding of radon dosimetry led to concern about cancers at other sites.

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI Consequently, the findings of general-population-based ecologic studies indicating associations between population estimates of indoor radon exposures and rates for cancers other than lung-cancer were not expected, and they were not readily explicable on a biologic basis. A 1989 letter to the editor by Lucie (1989) described an association between county-level leukemia incidence and average radon concentration. A 1990 report by Henshaw and colleagues (1990), indicating that radon might be associated with myeloid leukemia, cancer of the kidney, melanoma, and some childhood cancers brought public and scientific concern to the issue. Henshaw and collaborators conducted an ecologic analysis of cancer rates and estimated background radon exposure; positive and statistically significant associations were found. However, after those publications, methodologic biases affecting interpretation of the results were noted, and correspondence called for caution in interpreting the results. Furthermore, the highest radon levels in that report were quite low (near average indoor concentrations) and the study is subject to the flaws frequently associated with ecologic studies as described in appendix G. Additional ecologic analyses were reported, and case-control studies are now in progress. These findings provide a rationale for further consideration of malignancies other than lung-cancer in the underground miner data, and Darby and others have evaluated these malignancies through combined analyses of data from 11 studies of underground miners. Studies of Underground Miners Since the BEIR IV report, initial analyses of data on a number of cohorts of underground miners have been reported, thereby increasing the extent of information available on malignancies other than lung-cancer. Reports of several studies have provided information on all malignancies (Morrison and others 1988; Darby and others 1995; Tomásek and others 1993; Tirmarche and others 1993). Tomásek and others (1993) reported on the Czech cohort of 4,320 miners. At an average of 25 years of follow-up, statistically significant excesses of deaths from cancers of the liver, gallbladder, and extrahepatic bile ducts were noted, but there were no excess deaths for other individual sites. There was not a significant excess of all cancers other than lung-cancer (observed: expected ratio, or O/E, = 1.11; 95% confidence interval, CI 0.98–1.24). In the Newfoundland fluorspar miners, excess occurrence of deaths from cancers of the buccal cavity and pharynx and of the salivary glands was noted, but small numbers of cases limited dose-response analysis and interpretation of the findings (Morrison and others 1988). Tirmarche and others (1993) found significant excess occurrence of cancer of the larynx (O/E = 2.35; 95% CI, 1.37–3.76). In the Cornish tin miners, there were nonsignificant excesses of deaths from cancer of the stomach (O/E, 1.41) and from leukemia (O/E, 1.73). These analyses of data from cohorts did not indicate any consistent patterns across the cohorts,

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI and the possibility of false-positive findings when several associations are investigated in several cohorts makes the results difficult to interpret. The most-informative analysis on cancers other than lung-cancer was based on pooling of data from 11 studies of underground miners (Darby and others 1995). Darby and colleagues assembled data from 11 cohorts—10 of those included in the lung-cancer pooled lung-cancer analysis and a replacement of the Radium Hill cohort with the Cornish tin miners because of the small number of deaths and incomplete follow-up in the Radium Hill group. For each cohort, the expected number of deaths was calculated on the basis of external comparison rates; the external comparisons excluded the China cohort because appropriate national or regional rates were not available. In addition, internal comparisons were made by evaluating the association for specific cancers with cumulative exposure to radon progeny. Both types of comparisons were carried out separately for time since first employment categories of less than 10 years and equal to or greater than 10 years. External comparisons were also made for the combined periods. Overall, there was no excess of deaths from cancers other than lung-cancer (O/E, 1.01; 95% CI, 0.95–1.07). Of the 28 individual cancer categories evaluated, there were significant excesses for stomach cancer (O/E, 1.33; 95% CI, 1.16–1.52) and for primary liver cancer (O/E, 1.73; 95% CI, 1.29–2.28) (Table 4-3). The ratios of observed to expected deaths were greater than unity for all leukemias combined and for each of the principal subtypes, although none of the excesses was statistically significant. Leukemia also showed a significant increase when analyses were restricted to less than 10 years since first employment. There were statistically significant decreases for some sites. In comparisons by extent of exposure, none of the cancers showing excesses (stomach, liver, and leukemia) was positively associated with exposure, so increased rates for these cancers were judged unlikely to be related to exposure to radon and radon progeny. For the leukemia elevation in the first 10 years since start of employment, the possibility of exposure to gamma radiation in the mines was considered. Both cancer of the pancreas and the category of all cancers other than lung showed statistically significant positive associations with cumulative exposure, although the latter association was found only for the first 10 years since start of employment. The finding for cancer of the pancreas was interpreted as likely to be spurious. The finding for all cancers other than lung was attributed to 2 deaths in workers with cumulative exposure exceeding 5.25 Jhm-3 (1,500 WLM); these deaths were in the "other and unspecified cancers" category, and review of available records resulted in the conclusion that the true underlying cause of death was very likely lung-cancer. With deaths in the "other and unspecified" category excluded, the category of all cancers excluding lung-cancer was not found to be significantly associated with cumulative Jhm-3 (WLM). Overall, the study of Darby and others (1995) found no substantial evidence

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI TABLE 4-3 Numbers of deaths observed (O), ratio of observed to expected deaths (O/E), and 95% confidence interval (Cl) for selected sites of cancer, analysis of pooled data on miners (all studies except China), by time since first employment Cancer site (ICD-9 code) O O/Ea 95% CI Tongue and mouth (141, 143–145) 11b 0.52 0.26–0.93 Salivary gland (142) 4 1.41 0.39–3.62 Pharynx (146–149) 9c 0.35 1.16–0.66 Esophagus (150) 45 1.05 0.77–1.41 Stomach (151) 217c 1.33 1.16–1.52 Colon (152–153) 95b 0.77 0.63–0.95 Rectum (154) 60 0.86 0.66–1.11 Liver, primary (155.5, 155.1) 50c 1.73 1.29–2.28 Liver, unspecified (155.2) 3 0.43 0.09–1.26 Gallbladder (156) 19 1.23 0.74–1.92 Pancreas (157) 91 1.05 0.85–1.29 Nose (160) 3 0.69 1.14–2.02 Larynx (160) 38 1.21 0.86–1.67 Bone (170) 10 1.04 0.50–1.91 Connective tissue (171) 5 0.82 0.27–1.91 Malignant melanoma (172) 18 0.92 0.54–1.45 Other skin (173) 9 1.60 0.73–3.03 Prostate (185) 83 0.88 0.70–1.09 Testis (186) 6 0.72 0.26–1.57 Bladder (188, 189.3–189.9) 39 0.85 0.61–1.16 Kidney (189.0–189.2) 44 0.91 0.66–1.22 Brain and central nervous system (191, 192) 52 0.95 0.71–1.25 Thyroid gland (193) 2 0.47 0.06–1.71 Non-Hodgkin's lymphoma (200, 202) 36 0.80 0.56–1.10 Hodgkin's disease (201) 17 0.93 0.54–1.48 Multiple myeloma (203) 26 1.30 0.85–1.90 Leukemia (204–208) 69 1.16 0.90–1.47 Leukemia excluding chronic lymphatic (204–208 except 204.1)d 36 1.11 0.78–1.54 Myeloid leukemia (205–206)d       Acute myeloid leukemia (205.0, 205.2, 206.0, 206.2)d 27 1.41 0.93–2.05   12 1.16 0.60–2.02 Other and unspecified 118 1.12 0.93–1.35 All cancers other than lung (140–161, 163–208) 1,179 1.01 0.95–1.07 a Expected deaths calculated from national or local mortality rates. b 0.05 = P > 0.01 c P œ 0.001 (2-sided tests). d For each study, only the period for which the 8th or 9th ICD revisions was in use nationally is included. Source: Darby and others (1995).

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI of increased risks of cancers other than lung-cancer in the 11 miner cohorts. The authors concluded that the study provided strong evidence that high levels of exposure to radon do not cause a ''material risk" of mortality from cancers other than lung-cancer and that protection standards for radon should continue to be based on consideration of lung-cancer risk alone. Studies of the General Population The hypothesis that radon might cause cancers other than lung-cancer in the general population originated in ecologic studies reported after the BEIR IV report. Those analyses were conducted with units of analysis ranging from counties to countries (Table 4-4). Lucie (1989) published one of the initial reports, showing a positive correlation between acute myelogenous leukemia incidence for counties in the U.K. and county average radon concentration. Henshaw and colleagues (1990) followed with their report which provided estimates of radiation dose to the red marrow that indicated a significant dose to the marrow at typical indoor radon concentrations. For example, at an exposure of about 185 Bqm-3 (5 pCiL-1), the estimated dose to marrow from radon was estimated to be similar to that from low-LET radiation. Henshaw and colleagues described ecologic correlations between estimated mean exposures of residents of 15 countries and incidence of leukemias, childhood cancers, and selected additional cancers in these countries. Further analyses were reported for provinces of Canada. Henshaw and others did not find correlations for lung-cancer or for skin cancer. Bridges and colleagues (1991) later reported a correlation between frequency of a mutation of the hypoxanthine guanine phosphoribosyl transferase gene (hprt) and indoor radon exposure, supporting the plausibility of the ecologic associations found by Lucie and Henshaw and colleagues. Bridges and colleagues selected 20 persons, mostly never-smokers, from homes that had been monitored for radon; they were selected to provide a range of exposure from below about 111 to 740 Bqm-3 (3 to 20 pCiL-1). The logarithm of the mutation frequency was significantly associated with radon concentration in the homes, which was measured twice—for 1 month and for 3 months. This result suggested a mutation frequency that was much greater than would be expected from estimated radon dose to blood and previous in vitro hprt mutation data in numerous cell types. However, in a larger follow-up study by the same group (Cole and others 1996), no significant association was found. The later study was of 65 persons from 41 houses in the same small rural town, with measured radon of 19–484 Bqm-3 (0.51–13.08 pCiL-1). BCL-2 t (14;18) translocations, chromosomal events associated with leukemia and lymphoma, also showed no association with radon exposure. Bauchinger and others (1994) performed conventional chromosomal analyses in blood lymphocytes of 25 persons living continuously in houses with indoor radon concentrations exceeding the average in German houses of 50 Bqm-3 (1.35

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI TABLE 4-4 Ecologic studies of cancer other than lung-cancer Reference Location Exposure measure Outcomes Findings Lucie 1989 United Kingdom County average radon concentration AMLa incidences for 23 countries Positive correlation Henshaw and others 1990 a,b 15 countries and regions in countries National survey data Cancer and leukemia incidence Positive correlation with myeloid leukemia, melanoma, kidney cancer, prostatic cancer, and some childhood cancers Alexander and others 1990 United Kingdom County average radon concentration Leukemia and lymphoma incidences Positive correlation with multiple outcome groups Butland and others 1990 Various countries County average radon concentration Incidence, of leukemia and selected cancers Positive, significant correlation for all childhood cancers Muirhead and others 1994 United Kingdom County and district average concentration from surveys Incidence, of leukemia and NHLb Variable pattern of correlation comparing surveys county and district-level analyses a Acute myelogenous leukemia. b Non-Hodgkin's lymphoma.

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI pCiL-1) by a factor of 5–60. The mean frequencies of dicentrics and rings per cell (1.5 ± 0.4 × 10-3) were significantly higher than control levels (0.54 ± 0.11 × 10-3). Grouping the persons by estimated cumulative exposure showed a tendency for λ an exposure-effect relationship. Estimated radon exposures ranged from 700 to 6300 Bqm-3 (18.92 to 170.3 pCiL-1). Subsequently, painting of 3 chromosomes by fluorescence in situ hybridization (FISH) techniques was carried out on the same lymphocyte samples (Bauchinger and others 1996). The mean frequency of symmetrical translocations was slightly (1–5 fold), but not significantly (p <0.1), raised in the radon group compared to the controls. Only for males separately did the raised level reach statistical significance. For dicentric chromosomes, scoring of FISH-painted chromosomes gave results that were consistent (3-fold increase) with those previously obtained by the conventional methods. The apparent lower sensitivity of the FISH-translocation measurements to discriminate radon exposure was ascribed to the much higher control frequencies for translocations compared to dicentrics and to the lower doses received by the hemopoietic compartments such as bone marrow that should contribute most to stable symmetrical aberrations, as compared to mature blood lymphocytes that are the direct target cells for observed dicentric aberrations. The report of Henshaw and colleagues (1990) attracted substantial attention and criticism, as shown in correspondence to the Lancet (Mole 1990; Bowie 1990; Prentice and Copplestone 1990; Baverstock 1990; Butland and others 1990) and other journals (Peto 1990; Wolff 1991). Mole (1990) raised questions concerning the dosimetric model, and Butland and others (1990) noted that even Henshaw's high estimate of dose to the red marrow from radon is only about 1% of that to the lung. Butland and colleagues repeated the analysis of Henshaw and colleagues but used the data from cancer registries that they regarded as most satisfactory. They found a significant positive correlation for all childhood cancers combined. Results for other sites and for the leukemias were positive but not statistically significant. However, the numbers of countries were small and power was limited. None of the analyses showed a significant association with lung-cancer. Other critics raised concerns about confounding and biologic plausibility. Wolff and Stern (1991) presented analyses that suggested that the observed correlations might have resulted from confounding by socioeconomic factors. Muirhead and others (1992) conducted an ecologic analysis of childhood leukemia and non-Hodgkin's lymphoma rates based on small areas (districts) of the U.K. and found no significant associations with radon exposure, even though analyses by aggregated areas (counties) showed a significant positive correlation with radon levels. The correlation between districts within counties was negative; that between counties was positive. The Henshaw and others (1990) analysis of U.K. data was similar to that based on the aggregated areas. For the United States, Cohen (1993) examined ecologic correlations between estimated average radon concentrations in 1,600 counties and cancer mortality, in unspecified years, in males and females. Positive and statistically significant

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI associations were found for a number of sites, including, in men, cancer of the lip, salivary gland, nasopharynx, nose, nasal cavity, middle ear, breast, eye, thyroid, thymus and endocrine glands, and multiple myeloma and, in women, cancer of the salivary gland, nasopharynx, large intestine, liver, gallbladder, bile ducts, larynx, bone, connective tissue, kidney and uterus, eye, thymus, and endocrine glands, and lymphosarcoma, reticular sarcoma, Hodgkin's disease, multiple myeloma, and leukemia. The consequences of adjusting for smoking at the state level were also examined. Mifune and colleagues (1992) described cancer mortality for 1952–1988 in inhabitants of the Misasa spa area in Japan, comparing standardized mortality ratios based on national data with those in a control area. In the spa area, there are 90 hot-spring sources, and average radon concentration was reported to be 26 mBqL-1 in outdoor air and 35 mBqL-1 in indoor air. Routes of exposure included use of the hot springs for bathing and the medical treatment of patients. Overall, there was no excess of all cancers, and the risk of lung-cancer death in the Misasa area was only 55% of that in the control area. REPRODUCTIVE OUTCOMES The committee identified only a single new investigation relevant to concern about reproductive outcomes. Shields and others (1992) conducted a case-control study of congenital abnormalities, stillbirths, developmental disorders, and deaths from causes other than injuries. The case series included 266 cases and an equal number of controls with a normal birth. Exposure variables included the occupations of the parents and grandparents; the nearness of the subject's residences to uranium mines, mine dumps, and mill tailings; and living in a home constructed with uranium-mine rock. There was no evident effect of the fathers' being employed in a uranium mine or mill. There was increased risk for adverse pregnancy outcome if the mother lived near tailings or mine dumps. Interpretation of the findings is limited by lack of statistical power, particularly within the categories of specific adverse outcomes. CONCLUSIONS Although it has been nearly 10 years since the BEIR IV committee reviewed the evidence on health effects of radon-progeny exposure other than lung-cancer, the database is still limited. Nevertheless, the committee found several conclusions to be warranted. In regard to cancers other than lung-cancer, the committee interpreted the pooled analysis reported by Darby and colleagues (1995) as not indicating excess risk for cancers other than cancer of the lung in radon-exposed miners. Although 95% confidence limits are wide for some sites, the data provide evidence that radon and its progeny are not a major cause of non-lung-cancers and leukemias in the general population, as suggested by some ecologic studies.

OCR for page 117
Health Effects of Exposure to Radon: BEIR VI The committee concluded that the findings in the miners could be reasonably extended to the general population; there is no basis for considering that effects would be observed in the range of typical exposures of the general population that would not be observed in the underground miners exposed at generally much higher levels. The studies in the general population are ecologic and subject to many potential biases. The committee agrees with the conclusion of Darby and others that there is no need to consider cancers other than lung-cancer in setting protection standards and guidelines for radon. The dose calculations suggest that radon and progeny could contribute to some proportion of skin-cancer cases. Only 2 new studies had been reported on nonmalignant respiratory diseases. The report from New Mexico again documented that uranium mining adversely affects lung function (Mapel and others 1996). Archer and colleagues (1996) described an intriguing series of cases that support the possibility that exposure to radon progeny cause fibrosis of the pulmonary interstitium, often referred to as pulmonary or interstitial fibrosis. However, this clinical case series is insufficient to establish the link to radon progeny specifically, and there is a need for more research on the persistent question of the existence of radon-related pulmonary fibrosis. The new case-control study of reproductive outcomes in Shiprock, New Mexico, was limited by sample size and the possibility of measurement error because of the reliance on self-reported exposure measures. The committee was unable to reach any conclusion with regard to adverse effects of radon exposure on reproductive outcome.