Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Nov 1.
Published in final edited form as: Arthritis Rheumatol. 2014 Nov;66(11):3105–3112. doi: 10.1002/art.38786

Systemic Lupus Erythematosus is Associated with Uranium Exposure in a Community Living Near a Uranium Processing Plant: A Nested Case-Control Study

Pai-Yue Lu-Fritts 1, Leah C Kottyan 2, Judith A James 3, Changchung Xie 4, Jeanette M Buckholz 4, Susan M Pinney 4,*, John B Harley 2,*
PMCID: PMC4211941  NIHMSID: NIHMS630194  PMID: 25103365

Abstract

Objective

Explore the hypothesis that cases of SLE will be found more frequently in community members with high prior uranium exposure in the Fernald Community Cohort (FCC).

Methods

A nested case control study was performed. The FCC is a volunteer population that lived near a uranium ore processing plant in Fernald, Ohio, USA during plant operation and members were monitored for 18 years. Uranium plant workers were excluded. SLE cases were identified using American College of Rheumatology classification criteria, laboratory testing, and medical record review. Each case was matched to four age-, race-, and sex-matched controls. Sera from potential cases and controls were screened for autoantibodies. Cumulative uranium particulate exposure was calculated using a dosimetry model. Logistic regression with covariates was used to calculate odds ratios (OR) with 95% confidence intervals (CI).

Results

The FCC includes 4,187 individuals with background uranium exposure, 1,273 with moderate exposure, and 2,756 with higher exposure. SLE was confirmed in 23 of 31 individuals with a lupus ICD9 code, and in 2 of 43 other individuals prescribed hydroxychloroquine. The female:male ratio was 5.25:1. Of the 25 SLE cases, 12 were in the higher exposure group. SLE was associated with higher uranium exposure (OR 3.92, 95% CI 1.131-13.588, p = 0.031).

Conclusion

High uranium exposure is associated with SLE relative to matched controls in this sample of uranium exposed individuals. Potential explanations for this relationship include possible autoimmune or estrogen effects of uranium, somatic mutation, epigenetic effects, or effects of some other unidentified accompanying exposure.

Introduction

Systemic lupus erythematosus (SLE) is a chronic, heterogeneous autoimmune disease that affects between 20 and 70 individuals per 100,000 of the general population [1]. Women of child-bearing age are affected nine times more often than men. The development of SLE involves both genetic and environmental factors. This genetic contribution is illustrated by an SLE heritability of 66% and a sibling risk ratio of 8-29 [2, 3]. Many environmental factors have been investigated, and, for example, compelling evidence has associated silica quartz dust and Epstein Barr virus exposure with increased SLE risk [4-7]. More recently, meta-analyses have shown an association with smoking [8, 9]. Other studies have assessed the role of solvents, insecticides, and lipstick, though their outcomes have not yielded reliable associations [10-13]. Identification of previously unassociated exposures often requires cohort or community studies [14-16].

Fernald, OH was the location of a United States Department of Energy (DOE) Feed Materials Production Center (FMPC) until 1988. The primary function of the FMPC was to convert uranium ore concentrates and recycled uranium materials to uranium metal. The FMPC processed approximately 3.62 × 108 kg of uranium, along with smaller amounts of thorium [17]. During plant operation between 1951 and 1988, approximately 310,000 kg of uranium were released into the air, and 99,000 kg were released into the water [17]. Other products released during uranium processing included naturally occurring radionuclides in uranium ores (protactinium-234m, uranium isotopes, radium isotopes, and thorium isotopes); [17]. Radium-226-rich K-65 waste material was also stored in large quantities onsite, and it was a major source of radon-222 [17].

The community living around the plant filed a class action lawsuit when they learned of the contamination, since they feared cancer resulting from the radiation exposure [18]. The settlement resulted in one of the largest medical monitoring programs of its kind in the country: the Fernald Medical Monitoring Program (now known as the Fernald Community Cohort (FCC)), as has been previously described [17, 18]. Community residents living near the plant were recruited through local media to voluntarily enroll in the study. Ultimately, 9,782 individuals were enrolled, which represents approximately a quarter of the eligible population with high, moderate, and minimal exposure to uranium. Medical monitoring of individuals included eight cycles of physical exams, questionnaires, and lab tests. ICD-9 codes were assigned to medical records by a certified coder. Prior investigation of this cohort has yielded statistically significant elevations in bladder disease and chronic kidney disease [17].

Upon an initial International Classifications of Diseases, 9th revision (ICD-9) code search of the FCC for systemic lupus erythematosus (ICD-9 code 710.0), 26 individuals were identified. This more than 3-fold excess of reported SLE cases in a carefully monitored community provided an opportunity to evaluate the relationship between environmental uranium exposure and SLE.

Our objective was to explore the hypothesis that SLE patients will be found more frequently in community members with high prior uranium exposure in the Fernald Community Cohort (FCC).

Patients and Methods

A nested case control study was performed with data from the FCC.

Population

The FCC is comprised of voluntarily enrolled individuals who lived within 5 miles of an active uranium ore processing facility in Fernald, Ohio for at least two consecutive years between January 1, 1952 and December 18, 1984 [18]. These individuals were recruited through television, radio, and newspaper announcements between 1990 and 1991, and they were followed until 2008. No uranium plant workers are included in this cohort.

Case Identification

Potential FCC SLE cases were identified with searches of the FCC electronic database for ICD-9 codes associated with lupus and a medication code search for hydroxychloroquine. Lupus ICD-9 codes investigated included 710.0 (systemic lupus erythematosus) and 695.4 (lupus erythematosus). Individuals identified through the medication code search excluded those who also fulfilled lupus ICD-9 code search criteria. FCC records and medical charts were retrieved on potential cases and reviewed by a trained physician, where items pertaining to lupus were graded as absent, patient-reported, or physician-reported.

Cases were confirmed using an operational definition that included American College of Rheumatology (ACR) classification criteria, presence of autoantibodies, and medical record documentation. Cases were considered either “definite”, “possible”, or not SLE. Major criteria of the operational definition included physician or death certificate documentation. Minor criteria included documentation of at least 2 ACR criteria, serologic confirmation of autoantibodies, and documentation of the individual taking hydroxychloroquine or chloroquine. A definite SLE case was defined as an individual with at least one major and one minor criteria, self-report with two minor criteria, or three minor criteria. A possible SLE case was defined as an individual with a history of self-reported SLE and one minor criterion. Individuals who failed to meet these operational definitions were not considered to have SLE. For definite and possible SLE cases, the date of diagnosis was extracted or estimated from Fernald records and outside medical records.

Control Selection

A pool of potential controls was isolated from adults in the cohort. This pool excluded individuals who had been considered as potential cases. Additional exclusion criteria included non-white race (all cases were white), and certain abnormal laboratory findings (total white count less than 4 × 103/mm3, absolute lymphocyte count less than 1.5 × 103/mm3, or platelet count less than 100 × 103/mm3). These laboratory parameters were used as exclusion criteria to avoid including controls that could have possibly had undiagnosed lupus or criteria that would have satisfied a diagnosis of lupus. From this pool, four age-, race-, and sex-matched controls were selected for every case, except for 1 case, where only three matched controls were identified. Age-matching was based on age at diagnosis (Age +/- 2 years) and year of birth (+/- 2 years).

For comparison, analysis with rheumatoid arthritis (RA), another connective tissues disease with female predominance, was also performed. These cases were identified in the FCC through a search for the ICD-9 code 714.0 for RA.

Serum Analysis

Sera of potential cases and matched controls were screened for the presence of auto-antibodies, using serum from the FCC biobank. Attempts were made to select serum obtained closest to the diagnosis date. Over the past five years, the Bioplex™ 2200 multiplex immunoassay has become a new method for autoantibody screening. Using Luminex xMap® technology, 13 antibodies are detected by binding of antigen-coated fluorescent microbeads through precipitation [19]. Additional serum analysis for autoantibody testing was performed by the College of American Pathologists-certified Oklahoma Medical Research Foundation Clinical Immunology Laboratory. Anti-nuclear antibodies were detected with indirect immunofluorescence of HEp-2 cells and anti-double-stranded DNA antibodies were detected with indirect immunofluorescence of Crithidia cells. Enzyme-linked immunosorbent assays were used to detect anti-cardiolipin antibodies. Methods for these tests have been described previously [20-22].

Uranium Exposure Calculation

A cumulative uranium exposure estimate was used as the primary exposure measure. This estimate was calculated developed by FCC investigators using an exposure algorithm developed by the Centers for Disease Control and Prevention (CDC) as part of the Fernald Dose-Reconstruction Project as previously described [23]. The CDC researched records of plant emissions, meteorological data, modeling of the dispersion and deposition of uranium-containing particulate matter, simulation studies, and comparison with results of exposure assessments performed at currently operating radiation-producing facilities. Numerous validation studies were performed as part of this extensive dosimetry reconstruction project [23]. The FCC exposure domain was the 5-mile radius eligibility area for the study population. This domain was divided into 100 segments, each measuring approximately one square mile. By applying the exposure estimation methods developed by CDC to the FCC cohort, the average concentration of airborne uranium for each segment in the FCC exposure domain was calculated for each year of plant operation. FMPC emissions during the period of plant operations were the source of airborne uranium. The algorithm did not incorporate exposure through uranium-contaminated drinking water, another significant source of exposure. However, in the Fernald exposure domain, exposure to contaminated water, such as through private drinking water wells or cisterns, is highly correlated with airborne exposure. Inhalation exposure of uranium particulates has been estimated to contribute almost entirely to the body burden of uranium for those who never drank contaminated water, and up to 50% of the body burden for those who continuously drank contaminated water [23].

Elements included in the calculation were the source term, particle size, dispersion and deposition, and distance and direction from the plant. Annual estimated airborne exposure was calculated using the participant's location, calendar year, and duration of each place of residence. Exposure was assumed to be limited to the time that the plant was in operation, from 1951 to 1989, and the cumulative exposure estimates include only exposure during that time period. Cumulative airborne uranium exposure in μg/m3 was calculated for each individual by summing yearly/partial year estimates of exposure. Uranium exposure was then translated into approximate equivalent cumulative exposure to ionizing radiation for categorizing exposure into three groups; minimal (participants with minimal exposure, i.e., an estimated lifetime cumulative uranium exposure with an equivalent of <0.25 Sievert [Sv]); moderate (0.25 to 0.50 Sv); or high (>0.50 Sv). Radiation dose estimates from nine CDC exposure scenarios were used to establish the cut points of uranium exposure for the three groups. The annual average dose of background radiation for the United States population is approximately 0.003 Sv. Both the discrete exposure estimate values and the categorical exposure assignments were used in the statistical analysis.

Statistical Analysis

Median uranium exposure indices were calculated for each exposure group among SLE cases, controls, and RA cases. An association between covariates and uranium exposure category (minimal, moderate, high) was evaluated using the chi-square test, both overall and between specific exposure groups. The association between uranium exposure (minimal, moderate, high) and the incidence of lupus was evaluated using conditional logistic regression. Potential confounding variables (family history of lupus, alcohol use, history of smoking, age at menopause) were obtained via FCC questionnaires or medical exams. Alcohol use was also categorized into three groups: no alcoholic beverages/week, 1-2 alcoholic beverages/week, and ≥ 3 alcoholic beverages/week. Smoking history was quantified as cumulative pack-years. Age at menopause was categorized into “< 50 years old”, “≥ 50 years old”, and “not yet”. Measures of association for conditional logistic regression analyses were calculated as odds ratios (OR) with 95% percent confidence intervals (CI). All analyses were repeated in gender-specific strata to evaluate potential gender differences in the relationship between uranium exposure and the study outcomes. Logistic regression analysis using the log transformed continuous exposure calculation was also performed. The association between uranium exposure and rheumatoid arthritis was evaluated with a Fisher's exact test. Analyses were performed using SAS version 9.2 (SAS Institute: Cary, NC).

Results

SLE was confirmed in 21 of 26 cases with an ICD-9 code of 710.0, in 2 of 5 cases with an ICD9 code of 695.4, and in 2 of 43 other cases prescribed hydroxychloroquine (Table 1). Of the 74 individuals who were considered potential cases, 22 are “definite” cases and 3 are “possible” cases. Twelve of the “definite cases” fulfilled at least 4 ACR criteria. The remaining 10 definite cases fulfilled the operational definition criteria for a “definite case” and were agreed upon by a group of rheumatologists. The female to male ratio among the 25 cases is 5.25 to 1. The earliest date of diagnosis of any of the cases was 1974 and five others also had dates of diagnosis prior to the end of follow-up in 1988. However, only two had any exposure after diagnosis, as others had moved out of the exposure domain. Both of those persons were in the lowest exposure group, and their exposures were extremely low, with cumulative exposures of 0.009 μg/m3-years and 0.010 μg/m3-years after diagnosis. These amounts represent only 2% of the median value for cases. Of the SLE cases, 5 are in the low exposure group, 8 in the moderate exposure group, and 12 in the high exposure group. Among the controls, 41 are in the low exposure group, 27 in the moderate exposure group, and 31 in the high exposure group (Table 2). Cases and matched controls are not different by cumulative pack years of smoking or by alcohol intake (Table 2). None of the cases were related to each other. We did not have information on history of lupus in second or third degree relatives. Family history of first degree relatives was obtained and there was one report of family history of SLE in each group (Table 2). This was factored into the analysis, but it was not retained in the final model. Figure 1 illustrates median exposure index scores in the SLE cases, controls, and RA cases. Among the 25 cases and 99 controls included in the analysis, the median exposure index scores are 0.47 (mean 1.18, range 0.01-4.31) and 0.29 (mean 0.75, range 0-6.99), respectively. Among the 150 RA cases, the median exposure index score is 0.28 (mean 0.61, range 0-4.72). The distribution of exposure index scores in the RA cases more closely approximates that of the controls than the SLE cases.

Table 1. SLE case determination outcomes.

SLE Case Determination Total considered “Definite” “Possible” “No”*
ICD9 code 710.0 26 18 3 5
ICD9 code 695.4 5 2 0 3
HCQ medication code 43 2 0 41
Total 74 22 3 49
Open in a new tab
*

“No”: Exclusion based on failure to meet operational definition criteria. These persons were ineligible to be controls.

Table 2. Characteristics of cases and controls.

Characteristic Cases Controls
n = 25 n = 99
Age, years, median (range) 50 (29-78) 50 (29-79)
Females, number (%) 21 (84) 83 (83.8)
Smoking, cumulative pack-years, median (range) 0.25 (0-56.44) 10 (0-138)
Alcohol intake
 none, n (%) 15 (60) 70 (70.7)
 1-2 drinks/week, n (%) 5 (20) 14 (14.1)
 ≥ 3 drinks/week, n (%) 5 (20) 15 (15.2)
Family history of systemic lupus erythematosus, n (%) 1 (4) 1 (1)
Exposure group
 Minimal, n (%) 5 (20) 41 (41.4)
 Moderate, n (%) 8 (32) 27 (27.3)
 High, n (%) 12 (48) 31 (31.3)
Open in a new tab

Figure 1.

Figure 1

Open in a new tab

Uranium exposure index scores have a higher mean (◆) and median (horizontal line within the boxes) in SLE cases than in controls and RA cases. The interquartile range for uranium exposure index scores is presented with the upper and lower limits of the boxes. The box whiskers indicate the absolute range of scores.

The distribution of exposure groups among controls, RA cases, and SLE cases is illustrated graphically in Figure 2. Controls and RA cases are similarly distributed, whereas the SLE cases have a heavier representation in the high exposure group. In the FCC overall, RA occurs at the predicted United States (U.S.) prevalence (1.53%), while SLE is increased by 3.7-fold over the expected U.S. prevalence (0.256%) [1, 24].

Figure 2.

Figure 2

Open in a new tab

Comparison of exposure groups among controls, RA cases, and SLE cases in the FCC. The distribution of exposure groups is distorted in SLE relative to controls and RA cases.

Frequency of positive ANAs is increased in the cases compared to the controls (Table 3). Autoantibodies measured with the Bioplex™ 2200 multiplex immunoassay followed a similar distribution and trend (results not shown).

Table 3. Presence of anti-nuclear antibodies (ANA)* in cases and controls.

Cases Controls
ANA tested, n (%) 25 (100) 85 (85.9)
Negative, n (%) 3 (12) 59 (69.4)
≥ 1:40, n (%) 22 (88) 26 (30.6)
≥ 1:120, n (%) 17 (68) 17 (20)
≥ 1:360, n (%) 11 (44) 5 (5.9)
≥ 1:1080, n (%) 6 (24) 4 (4.7)
Open in a new tab
*

ANA results based on Hep-2 ANA indirect immunofluorescence

Following conditional logistic regression modeling, SLE was found to be associated with high exposure (OR 3.92, 95% CI 1.13-13.59, p = 0.031). When the three “possible” cases are excluded from the analysis, an association with high exposure persists (OR 3.62, 95% CI 1.05-12.56, p = 0.042). The OR for moderate uranium exposure and SLE was elevated, but not statistically significant. Smoking, alcohol intake, and family history of SLE are not significantly associated with risk of SLE in this small study (Table 4). The analysis of female cases and controls also shows an increased OR for SLE in women with high exposure compared to those with minimal exposure (OR 7.15, 95% CI 1.52-33.73, p = 0.01). Logistic regression using the log-transformed continuous exposure variable also yielded the statistically significant result of increased risk of SLE with increased exposure (OR 1.38, 95% CI 1.03-1.86, p = 0.03).

Table 4. Outcomes of logistic regression analysis.

A. Exposure Group Comparison with All Cases and Controls (n = 124)
Variables Odds Ratio 95% CI P-Values
Moderate vs Minimal exposure 2.65 (0.71, 9.84) 0.15
High vs Minimal exposure 3.92 (1.13, 13.59) 0.03
Alcohol 1-2 drinks/wkvs none 1.74 (0.47, 6.38) 0.41
Alcohol ≥ 3 drinks/wkvs none 2.11 (0.56, 8.00) 0.27
Smoking (cumulative pack-yrs) 0.98 (0.96, 1.01) 0.19
B. Exposure Group Comparison with Definite Cases and Controls (n = 109)
Variables Odds Ratio 95% CI P-Values
Moderate vs Minimal exposure 2.15 (0.54, 8.50) 0.27
High vs Minimal exposure 3.62 (1.05, 12.56) 0.04
Alcohol 1-2 drinks/wkvs none 2.15 (0.54, 8.62) 0.25
Alcohol ≥ 3 drinks/wkvs none 1.53 (0.40, 6.48) 0.50
Smoking (cumulative pack-yrs) 0.98 (0.96, 1.01) 0.14
Open in a new tab

Variables highlighted in bold indicate statistically significant results with a p-value < 0.05

Discussion

The Fernald Community Cohort provides an optimal setting to evaluate the effects of environmental exposures on disease. Our investigation of the relationship between environmental uranium exposure and lupus revealed a nearly four-fold increase in odds of lupus in individuals with high prior uranium exposure compared to subjects with minimal exposure. Other exposures, such as smoking and alcohol, had no significant effect. We were able to use banked serum to support the diagnoses in cases and the absence of disease in the controls.

The results of previous studies support an increased incidence of autoimmune disease and autoantibodies in uranium miners exposed to silica dust [4, 7, 25]. While specific effects of uranium on the development of autoimmunity have not been well studied, there has been some evidence regarding cadmium, a closely related heavy metal. In vivo, cadmium exposure has been related to immune-mediated glomerulonephritis and development of anti-nuclear antibodies in mice [26-28]. Furthermore, there could be several potential explanations for the relationship between uranium exposure and SLE based upon the molecular effects of uranium.

All uranium isotopes are radioactive and emit alpha particles, which are capable of inducing DNA damage [29-32]. Early investigations of somatic cell gene mutation in the FCC did not reveal a relationship between mutations and proximity to the FMPC [33]. However, quantitative estimates of exposure were not used in the analysis, and only older members of the cohort were selected, increasing background variability. Previous literature has shown that uranium can not only induce DNA damage through ionizing radiation, but it can also act directly through inherent properties [34, 35]. In fact, uranium itself and ionizing radiation may independently lead to epigenetic changes [36, 37]. Furthermore, radon, a radioactive decay product of uranium, has also been associated with abnormal DNA methylation [36, 38]. Interestingly, cadmium has also been shown to exert effects on the epigenome [39].

Another interesting finding in our study was a lower than expected female to male ratio of 5.25:1. The adult female to male ratio in SLE cases of European ancestry is generally around 9:1. This dilution of female predominance raises some questions, though its significance is uncertain given the small number of cases [40]. Uranium has been shown to affect estrogen receptor activity, similar to other heavy metals, such as cadmium [41, 42]. Furthermore, ionizing radiation itself can affect estrogen receptor activity [43]. Uranium miners have also been shown to have reduced levels of circulating testosterone [32]. X-chromosome dose effects have been identified by investigators as well [44-46]. Several immunomodulatory genes are present on the X chromosome [47]. It could be possible that uranium exposure led to increased expression of these genes, either through somatic mutation or demethylation. Whether or not this altered female to male ratio in our study might be a reflection of the altered sex-related risk from uranium and maybe other environmental exposures will require data from additional examples. We did not explore gene-environment interactions despite available whole blood and DNA from the cohort, as it was beyond the scope of our study aim.

The Fernald Dosimetry Reconstruction Project was focused primarily on exposure to uranium and its associated decay products (thorium, radium, and radon). Solvents, however, were also suspected to have been released into the ground water, though quantification did not reach a level of significance that justified further investigation [48]. The association of solvents with SLE development has been debated, but enough questions have been raised that further studies are warranted [5, 49, 50]. Importantly, silica is present in uranium ore and could have been released into the air during processing. Unfortunately, silica exposure related to the plant has not been quantified, and thus we are unable to analyze its potential effects on SLE in this cohort. Another potential explanation for our findings is that there may be other unidentified and unexplored factors related to living near industrial processing. No comparable industrial processing analysis has been done for other metal purifications that would permit an interpretation that would isolate the increased risk of SLE seen here with radiation, uranium itself, another component of uranium ore, or a chemical or pollutant generated by the processing.

This study has highlighted a relationship between environmental uranium exposure and the frequency of lupus in this cohort, but some limitations require consideration. This cohort was built to allow for the future study of different conditions and parameters. Special emphasis was placed on the confirmation of reported malignancies, given known effects of uranium and ionizing radiation exposure. It is important to note that questionnaires and exams were not built to assess formal criteria for SLE or other systemic autoimmune disease. Case identification and the data used for individual exposure estimation were all retrospectively gathered. However, longitudinal medical records for cases were obtained from the participant's personal physician's records. Stringent criteria were also used, which allowed us to eliminate 49 potential SLE cases.

Though factors could potentially lead us to overestimate our results, others could have led us to underestimate our results. Potential cases could have been missed with our screening methods. Also, we were unable to acquire medical records from several potential cases, which could have impaired our ability to confirm cases. With respect to the cohort itself, issues with reporting bias and selection bias are possible. FCC cohort members may have been more likely to report symptoms and diagnoses, although the community focus was on cancer. Alternatively, given the morbidity related to SLE and other chronic conditions, individuals with disease may also have been too ill to participate. This would alternatively foster a healthy volunteer effect [18].

High levels of environmental uranium processing exposure are associated with a nearly four-fold increase in lupus in the Fernald Community Cohort. Many explanations are possible. This association highlights the need to better understand 1) the impact of environmental uranium processing exposure on other autoimmune diseases, and 2) the interaction between environmental exposures and SLE.

Acknowledgments

We would like to thank the Fernald Community Cohort for their participation and contributions.

Financial Support: Grants: Dr. Lu and Dr. Kottyan received support through University of Cincinnati Center for Environmental Genetics New Investigator Scholar awards from the National Institute of Environmental Health Sciences (P30-ES006096). Dr. James has received support from Oklahoma Rheumatic Disease Research Cores Center (AR053483), Oklahoma Autoimmunity Center of Excellence (AI082714), Autoimmunity Prevention Grant (Understanding early events in lupus autoimmunity to aid prevention, AI101934), Science in a Culture of Mentoring (Centers of Biomedical Research Excellence, GM103510), and Oklahoma High-Troughput Serum Analyte System (S10, RR026735). Dr. Pinney, Dr. Xie, and Ms. Buckholz received support from the University of Cincinnati Center for Environmental Genetics (P30-ES006096). Dr. Harley has received support from the US Department of Veterans Affairs, the US Department of Defense (PR094002), and the National Institutes of Health (R37 AI024717, P01 AI083194, P01 AR049084, and U01 HG006828).

References

  • 1.Pons-Estel GJ, et al. Understanding the epidemiology and progression of systemic lupus erythematosus. Semin Arthritis Rheum. 2010;39(4):257–68. doi: 10.1016/j.semarthrit.2008.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Deapen D, et al. A revised estimate of twin concordance in systemic lupus erythematosus. Arthritis Rheum. 1992;35(3):311–8. doi: 10.1002/art.1780350310. [DOI] [PubMed] [Google Scholar]
  • 3.Alarcon-Segovia D, et al. Familial aggregation of systemic lupus erythematosus, rheumatoid arthritis, and other autoimmune diseases in 1,177 lupus patients from the GLADEL cohort. Arthritis Rheum. 2005;52(4):1138–47. doi: 10.1002/art.20999. [DOI] [PubMed] [Google Scholar]
  • 4.Conrad K, et al. Systemic lupus erythematosus after heavy exposure to quartz dust in uranium mines: clinical and serological characteristics. Lupus. 1996;5(1):62–9. doi: 10.1177/096120339600500112. [DOI] [PubMed] [Google Scholar]
  • 5.Finckh A, et al. Occupational silica and solvent exposures and risk of systemic lupus erythematosus in urban women. Arthritis Rheum. 2006;54(11):3648–54. doi: 10.1002/art.22210. [DOI] [PubMed] [Google Scholar]
  • 6.James JA, et al. Systemic lupus erythematosus in adults is associated with previous Epstein-Barr virus exposure. Arthritis Rheum. 2001;44(5):1122–6. doi: 10.1002/1529-0131(200105)44:5<1122::AID-ANR193>3.0.CO;2-D. [DOI] [PubMed] [Google Scholar]
  • 7.Rosenman KD, Moore-Fuller M, Reilly MJ. Connective tissue disease and silicosis. Am J Ind Med. 1999;35(4):375–81. doi: 10.1002/(sici)1097-0274(199904)35:4<375::aid-ajim8>3.0.co;2-i. [DOI] [PubMed] [Google Scholar]
  • 8.Kiyohara C, et al. Cigarette smoking, alcohol consumption, and risk of systemic lupus erythematosus: a case-control study in a Japanese population. J Rheumatol. 2012;39(7):1363–70. doi: 10.3899/jrheum.111609. [DOI] [PubMed] [Google Scholar]
  • 9.Costenbader KH, et al. Cigarette smoking and the risk of systemic lupus erythematosus: a meta-analysis. Arthritis Rheum. 2004;50(3):849–57. doi: 10.1002/art.20049. [DOI] [PubMed] [Google Scholar]
  • 10.Bengtsson AA, et al. Risk factors for developing systemic lupus erythematosus: a case-control study in southern Sweden. Rheumatology (Oxford) 2002;41(5):563–71. doi: 10.1093/rheumatology/41.5.563. [DOI] [PubMed] [Google Scholar]
  • 11.Cooper GS, et al. Occupational and environmental exposures and risk of systemic lupus erythematosus: silica, sunlight, solvents. Rheumatology (Oxford) 2010;49(11):2172–80. doi: 10.1093/rheumatology/keq214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Parks CG, et al. Insecticide use and risk of rheumatoid arthritis and systemic lupus erythematosus in the Women's Health Initiative Observational Study. Arthritis Care Res (Hoboken) 2011;63(2):184–94. doi: 10.1002/acr.20335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Wang J, et al. Is lipstick associated with the development of systemic lupus erythematosus (SLE)? Clin Rheumatol. 2008;27(9):1183–7. doi: 10.1007/s10067-008-0937-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Dahlgren J, et al. Cluster of systemic lupus erythematosus (SLE) associated with an oil field waste site: a cross sectional study. Environ Health. 2007;6:8. doi: 10.1186/1476-069X-6-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Balluz L, et al. Investigation of systemic lupus erythematosus in Nogales, Arizona. Am J Epidemiol. 2001;154(11):1029–36. doi: 10.1093/aje/154.11.1029. [DOI] [PubMed] [Google Scholar]
  • 16.Kardestuncer T, Frumkin H. Systemic lupus erythematosus in relation to environmental pollution: an investigation in an African-American community in North Georgia. Arch Environ Health. 1997;52(2):85–90. doi: 10.1080/00039899709602869. [DOI] [PubMed] [Google Scholar]
  • 17.Pinney SM, et al. Health effects in community residents near a uranium plant at Fernald, Ohio, USA. Int J Occup Med Environ Health. 2003;16(2):139–53. [PubMed] [Google Scholar]
  • 18.Wones R, et al. Medical monitoring: a beneficial remedy for residents living near an environmental hazard site. J Occup Environ Med. 2009;51(12):1374–83. doi: 10.1097/JOM.0b013e3181c558f1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Binder SR. Autoantibody detection using multiplex technologies. Lupus. 2006;15(7):412–21. doi: 10.1191/0961203306lu2326oa. [DOI] [PubMed] [Google Scholar]
  • 20.Arbuckle MR, et al. Development of anti-dsDNA autoantibodies prior to clinical diagnosis of systemic lupus erythematosus. Scand J Immunol. 2001;54(1-2):211–9. doi: 10.1046/j.1365-3083.2001.00959.x. [DOI] [PubMed] [Google Scholar]
  • 21.Arbuckle MR, et al. Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N Engl J Med. 2003;349(16):1526–33. doi: 10.1056/NEJMoa021933. [DOI] [PubMed] [Google Scholar]
  • 22.Heinlen LD, et al. Clinical criteria for systemic lupus erythematosus precede diagnosis, and associated autoantibodies are present before clinical symptoms. Arthritis Rheum. 2007;56(7):2344–51. doi: 10.1002/art.22665. [DOI] [PubMed] [Google Scholar]
  • 23.Killough GG. The Fernald Dosimetry Reconstruction Project: Task 6: Radiation doses and risks to residents from FMPC operations from 1951-1988. Radiological Assessments Corporation; Neeses, South Carolina: 1998. [Google Scholar]
  • 24.Alamanos Y, Voulgari PV, Drosos AA. Incidence and prevalence of rheumatoid arthritis, based on the 1987 American College of Rheumatology criteria: a systematic review. Semin Arthritis Rheum. 2006;36(3):182–8. doi: 10.1016/j.semarthrit.2006.08.006. [DOI] [PubMed] [Google Scholar]
  • 25.Conrad K, Mehlhorn J. Diagnostic and prognostic relevance of autoantibodies in uranium miners. Int Arch Allergy Immunol. 2000;123(1):77–91. doi: 10.1159/000024426. [DOI] [PubMed] [Google Scholar]
  • 26.Borgman RF, Au B, Chandra RK. Immunopathology of chronic cadmium administration in mice. Int J Immunopharmacol. 1986;8(7):813–7. doi: 10.1016/0192-0561(86)90019-6. [DOI] [PubMed] [Google Scholar]
  • 27.Griem P, Gleichmann E. Metal ion induced autoimmunity. Curr Opin Immunol. 1995;7(6):831–8. doi: 10.1016/0952-7915(95)80056-5. [DOI] [PubMed] [Google Scholar]
  • 28.Leffel EK, et al. Drinking water exposure to cadmium, an environmental contaminant, results in the exacerbation of autoimmune disease in the murine model. Toxicology. 2003;188(2-3):233–50. doi: 10.1016/s0300-483x(03)00092-1. [DOI] [PubMed] [Google Scholar]
  • 29.Bleise A, Danesi PR, Burkart W. Properties, use and health effects of depleted uranium (DU): a general overview. J Environ Radioact. 2003;64(2-3):93–112. doi: 10.1016/s0265-931x(02)00041-3. [DOI] [PubMed] [Google Scholar]
  • 30.Arruda-Neto JD, et al. Fragmentation of extracellular DNA by long-term exposure to radiation from uranium in aquatic environments. J Environ Monit. 2012;14(8):2108–13. doi: 10.1039/c2em30196b. [DOI] [PubMed] [Google Scholar]
  • 31.Zaire R, et al. Analysis of lymphocytes from uranium mineworkers in Namibia for chromosomal damage using fluorescence in situ hybridization (FISH) Mutat Res. 1996;371(1-2):109–13. doi: 10.1016/s0165-1218(96)90100-7. [DOI] [PubMed] [Google Scholar]
  • 32.Zaire R, et al. Unexpected rates of chromosomal instabilities and alterations of hormone levels in Namibian uranium miners. Radiat Res. 1997;147(5):579–84. [PubMed] [Google Scholar]
  • 33.Wones R, et al. Do persons living near a uranium processing site have evidence of increased somatic cell gene mutations? A first study. Mutat Res. 1995;335(2):171–84. doi: 10.1016/0165-1161(95)90053-5. [DOI] [PubMed] [Google Scholar]
  • 34.Barillet S, et al. Bioaccumulation, oxidative stress, and neurotoxicity in Danio rerio exposed to different isotopic compositions of uranium. Environ Toxicol Chem. 2007;26(3):497–505. doi: 10.1897/06-243r.1. [DOI] [PubMed] [Google Scholar]
  • 35.Stearns DM, et al. Uranyl acetate induces hprt mutations and uranium-DNA adducts in Chinese hamster ovary EM9 cells. Mutagenesis. 2005;20(6):417–23. doi: 10.1093/mutage/gei056. [DOI] [PubMed] [Google Scholar]
  • 36.Su S, et al. Aberrant promoter methylation of p16(INK4a) and O(6)-methylguanine-DNA methyltransferase genes in workers at a Chinese uranium mine. J Occup Health. 2006;48(4):261–6. doi: 10.1539/joh.48.261. [DOI] [PubMed] [Google Scholar]
  • 37.Miller AC, McClain D. A review of depleted uranium biological effects: in vitro and in vivo studies. Rev Environ Health. 2007;22(1):75–89. doi: 10.1515/reveh.2007.22.1.75. [DOI] [PubMed] [Google Scholar]
  • 38.Su SB, et al. p16 and MGMT gene methylation in sputum cells of uranium workers. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi. 2006;24(2):92–5. [PubMed] [Google Scholar]
  • 39.Wang B, et al. Cadmium and its epigenetic effects. Curr Med Chem. 2012;19(16):2611–20. doi: 10.2174/092986712800492913. [DOI] [PubMed] [Google Scholar]
  • 40.Weckerle CE, Niewold TB. The unexplained female predominance of systemic lupus erythematosus: clues from genetic and cytokine studies. Clin Rev Allergy Immunol. 2011;40(1):42–9. doi: 10.1007/s12016-009-8192-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Raymond-Whish S, et al. Drinking water with uranium below the U.S. EPA water standard causes estrogen receptor-dependent responses in female mice. Environ Health Perspect. 2007;115(12):1711–6. doi: 10.1289/ehp.9910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Hao Y, et al. The reproductive effects in rats after chronic oral exposure to low-dose depleted uranium. J Radiat Res. 2012;53(3):377–84. doi: 10.1269/jrr.11192. [DOI] [PubMed] [Google Scholar]
  • 43.Fucic A, Gamulin M. Interaction between ionizing radiation and estrogen: what we are missing? Med Hypotheses. 2011;77(6):966–9. doi: 10.1016/j.mehy.2011.08.021. [DOI] [PubMed] [Google Scholar]
  • 44.Dillon SP, et al. Sex chromosome aneuploidies among men with systemic lupus erythematosus. J Autoimmun. 2012;38(2-3):J129–34. doi: 10.1016/j.jaut.2011.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Sawalha AH, Harley JB, Scofield RH. Autoimmunity and Klinefelter's syndrome: when men have two X chromosomes. J Autoimmun. 2009;33(1):31–4. doi: 10.1016/j.jaut.2009.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Scofield RH, et al. Klinefelter's syndrome (47,XXY) in male systemic lupus erythematosus patients: support for the notion of a gene-dose effect from the X chromosome. Arthritis Rheum. 2008;58(8):2511–7. doi: 10.1002/art.23701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Bianchi I, et al. The X chromosome and immune associated genes. J Autoimmun. 2012;38(2-3):J187–92. doi: 10.1016/j.jaut.2011.11.012. [DOI] [PubMed] [Google Scholar]
  • 48.Voilleque PG. The Fernald Dosimetry Reconstruction Project: Tasks 2 and 3: Radionuclide source terms and uncertainties. Radiological Assessments Corporation; Neeses, South Carolina: 1995. [Google Scholar]
  • 49.Cooper GS, Parks CG. Occupational and environmental exposures as risk factors for systemic lupus erythematosus. Curr Rheumatol Rep. 2004;6(5):367–74. doi: 10.1007/s11926-004-0011-6. [DOI] [PubMed] [Google Scholar]
  • 50.Griffin JM, et al. Trichloroethylene accelerates an autoimmune response by Th1 T cell activation in MRL +/+ mice. Immunopharmacology. 2000;46(2):123–37. doi: 10.1016/s0162-3109(99)00164-2. [DOI] [PubMed] [Google Scholar]

RESOURCES