Cancer of the kidney and renal pelvis account for approximately 4% of all new cancer cases and 2% of cancer deaths among US men and women.1 Worldwide, the incidence of kidney cancer in men is twice as high as in women.2 Incidence and mortality rates for renal cell carcinoma have increased over the past 30 years.3,4 Cigarette smoking, obesity, and hypertension are the most commonly reported risk factors.2,3,5,6 Inconsistent associations have been reported for exposure to asbestos, polycyclic aromatic hydrocarbons, diesel exhaust, cadmium, and trichloroethylene.2–5,7
Trichloroethylene (also known as TCE) has been widely used as a solvent since the early 1900s.8 It has been used primarily in industrial degreasing involved with metal fabricating and manufacturing, especially in the aircraft/aerospace industry.8 Trichloroethylene is also used in dry cleaning, textiles, leather processing, electronics, health services, agriculture, and the food and chemical industry.8–10 However, workers engaged in metal degreasing appear to be the most heavily exposed.8,11
Trichloroethylene has been shown in animals to produce higher rates of cancer of the kidney, liver, lung, and hematopoietic tissue.11–17 However; these findings are not consistent, and some effects appear to be specific only to certain species or strains of animals.18,19 Reviews of the epidemiologic evidence regarding the carcinogenicity of trichloroethylene have not produced a consensus.20–23 In this meta-analysis, we included recently published cohort studies of Danish workers,24,25 aerospace workers,26,27 and electronics manufacturing workers,28 and 3 case-control studies7,29,30 not evaluated in previous reviews. Our objectives were as follows (1) to calculate summary relative risks and (2) to examine potential sources of heterogeneity across studies. We also conducted sensitivity and influence analyses, and assessed the potential role of confounding, information bias, selection bias, and publication bias.
We searched Medline and EMBASE, specifying trichloroethylene and human studies (including occupational studies); trichloroethylene or organic solvents with case-control or cohort studies; degreasers and cancer; hazardous waste sites and cancer outcomes; and trichloroethylene, organic solvents, or chlorinated solvents and kidney cancer or renal cell cancer. We also searched the bibliographies of recent review articles.20–23,31–34
Study Inclusion Criteria
We included epidemiologic studies that (1) used a cohort or case-control study design; (2) examined occupational trichloroethylene exposure and kidney cancer among workers; (3) specifically identified trichloroethylene exposure by reference to industrial hygiene records, individual biomarkers, job exposure matrices, or industrial processes that involved the use of trichloroethylene (cohort studies), or included questions regarding trichloroethylene exposure (case-control studies); (4) reported results specifically for kidney cancer or renal cell carcinoma; and (5) presented associations as relative risk estimates with corresponding measures of variability (eg, confidence intervals). Case-control studies included in this meta-analysis assessed trichloroethylene from self-reports of exposure or from job exposure matrices.7,29,30,35–38 Case-control studies that collected nonspecific or general exposure information such as “solvents” or “chemicals,” or that analyzed findings by a single job title only that is, without a job exposure matrix were not included. Studies of dry cleaners and laundry workers were excluded because of exposure classification and study design limitations. Prior to the 1960s, dry cleaning work involved some exposure to trichloroethylene in addition to carbon tetrachloride and Stoddard solvent; however, perchloroethylene has been the predominant agent used in the dry cleaning industry since the early 1960s.8,39,40 Dry cleaners did comprise a small proportion (<1%) of the study populations in 2 “multiple industry” studies included in this meta-analysis.24,25 We identified a total of 25 peer-reviewed epidemiologic studies that assessed the relation between trichloroethylene exposure and kidney cancer and met our inclusion criteria. Two of those studies reported on largely overlapping cohorts26,27 leaving 24 studies in our primary analyses.
Classification of Studies and Data Extraction
Cohort studies were categorized into 2 groups. Group I studies (n = 10) specifically identified trichloroethylene as a workplace exposure through biomonitoring, industrial hygiene data, identified work practices, or job titles that involved trichloroethylene. Group I studies also demonstrated sufficiently complete enumeration of the workforce.24–27,41–46 Three cohort studies24,45,46 obtained urinary trichloroacetate measurements from workers, whereas other studies25–27,41–44 assessed exposure to trichloroethylene through chemical inventory, industrial hygiene, or other records (eTable, http://links.lww.com/EDE/A354). Group II cohort studies (n = 7) were more limited in that they lacked specific, detailed trichloroethylene exposure information. For the group II studies,28,47–52 the reports mentioned trichloroethylene but provided little or no documentation of actual exposure28,47,50 or identified trichloroethylene but did not specify an exposed subcohort51,52; alternatively, the cohort consisted of aerospace or aircraft workers who were presumed to be exposed to trichloroethylene (eTable, http://links.lww.com/EDE/A354).48,49
Relative risk (RR) estimates and associated 95% confidence intervals (CIs) were abstracted from each publication for (1) analyses that reflected the entire cohort under study (“total cohort”), or the exposure category that included all trichloroethylene exposed workers; (2) the “Subcohort” of workers in the group I cohort studies that were identified specifically as more likely exposed to trichloroethylene; and (3) exposure-response results (cumulative exposure or duration of exposure/employment). For studies based on urinary trichloroacetate monitoring,24,45,46 the same data were used in both the total cohort and subcohort analyses. One group I cohort study did not specifically report data for an exposed subcohort and kidney cancer,44 and was included only in the total cohort analyses. Two studies26,27 analyzed largely overlapping populations from the same occupational cohort in Southern California (Rocketdyne): Zhao et al27 evaluated kidney cancer incidence whereas Boice et al26 evaluated kidney cancer mortality among the total cohort and a smaller group of workers identified as being more likely to have been exposed. The Boice et al study cohort was larger (n = 8372) and included more follow-up time (1948–1999) than the Zhao et al study (n = 5049 male workers follow-up from 1988–2000). Data from Boice et al were used in our primary meta-analysis, and we conducted sensitivity analyses using Zhao et al results were used in the separate analyses of kidney cancer incidence. For cohort studies that had been updated, we used the most recently published findings.43,45 Results for men and women were combined.
Most studies reported results for malignant neoplasms of the kidney and renal pelvis together (ICD versions 8 and 9 codes 189.0 and 189.1, respectively). In 2 group II cohort studies, cancer of the kidney was combined with cancer of other urinary organs.48,49 Excluding these studies from the meta-analyses, however, did not materially change summary estimates.
We conducted separate meta-analyses for group I cohort studies (total cohort and subcohort), group II cohort studies, group I and II cohort studies combined, case-control studies, and all types of studies combined. In addition, we conducted separate analyses of studies of (1) cancer incidence studies, (2) workers who were biomonitored for trichloroethylene exposure, (3) workers in the aircraft/aerospace industry, (4) workers from multiple industrial groups that used trichloroethylene, (5) duration of exposure/employment, and (6) cumulative exposure. Random effects meta-analysis methods were used to calculate summary estimates of relative risk and 95% CIs, with estimates of individual studies weighted by the inverse of their variance. We also conducted univariate meta-regression analyses by study classification and study design. The relative influence of each study on the summary RR was examined by calculating a summary RR that included all studies, followed by recalculation of the summary RR after removing each study one at a time. Publication bias was assessed by evaluation of funnel plots and use of the Begg and Matzumdar test and Egger regression method.53
All analyses were performed using Comprehensive Meta-Analysis version 2.2.04654 and verified by Episheet.55
A table in the electronic appendix summarizes characteristics of the studies, including the size of the study population, type of industrial/occupational group, follow-up time or enrollment periods, exposure assessment methods, and approximate trichloroethylene exposure levels (eTable, http://links.lww.com/EDE/A354). Study-specific kidney cancer results are also summarized in this table. The group I cohort studies consisted of 5 US studies of aerospace manufacturing or aircraft maintenance workers, 1 study of uranium processing workers, and 4 European studies of workers from various industries using trichloroethylene. Three of the European studies used urinary trichloroacetic acid measurements to identify trichloroethylene-exposed workers. The 7 group II cohort studies consisted of aircraft, electronics, Coast Guard, military, and cardboard manufacturing workers. Of the 7 case-control studies, one was nested within a cohort of transformer assembly workers37 and one analyzed French workers from a region where trichloroethylene was used in a screw cutting industry.30 Three case-control studies were hospital-based and one identified cases from a statewide cancer surveillance system.
All Studies Combined
The summary RR for all studies combined (group I [total cohort data], group II cohort, and case-control studies) was 1.30 (95% CI = 1.04–1.61). Strong heterogeneity was present (test for heterogeneity: P < 0.001). This heterogeneity remained when analysis was restricted to the trichloroethylene-exposed subcohorts from group I and all other studies (1.42 [1.13–1.77; test for heterogeneity: P = 0.001]) (Table, Fig. 1). Three studies were limited in study design, exposure assessment, or control selection procedures.36,47,52 After excluding these 3 studies, the summary RR was 1.24 (1.06–1.45 [test for heterogeneity: P = 0.62]).
The summary RR for the group I cohort studies was 1.34 (1.07–1.67; test for heterogeneity: P = 0.85) (Table, Fig. 2). The Raaschou-Nielsen et al study25 contributed the greatest relative weight (57%) and other studies each contributed a relative weight of 9% or less. Removal of this large study slightly reduced the estimate to 1.27 (0.90–1.78; test for heterogeneity: P = 0.79). The result for the 3 studies with biomonitoring was 1.02 (0.59–1.77; test for heterogeneity: P = 0.90) (Table). Among aerospace/aircraft worker cohorts, the estimate was 1.44 (0.94–2.21) for group I studies and 1.14 (0.84–1.57) for group I and II studies combined (Table). Among workers from various industries (all European studies), the summary RR was 1.31 (1.01–1.69, test for heterogeneity: P = 0.75) (Table).
Meta-analysis of the 7 group II cohort studies resulted in a summary RR of 1.58 (0.75–3.32); (test for heterogeneity: P < 0.001) (Table, Fig. 1). After excluding the 2 potential outlier studies, the estimate was 0.88 (0.58–1.33) with no heterogeneity (P = 0.99) (Table).
The summary RR for the 7 case-control studies was 1.57 (1.06–2.30), test for heterogeneity: P = 0.003 (Table, Fig. 1). Excluding one study with many limitations36 lowered the summary odds ratio and decreased heterogeneity (1.33 [1.02–1.73, test for heterogeneity: P = 0.14]). Six studies adjusted for smoking, 3 studies adjusted for BMI, and 2 studies adjusted for hypertension.
Analysis of Exposure Response and Publication Bias
Analysis of 10 studies with data on exposure categories showed no apparent exposure response patterns by duration or cumulative exposure (Table, Figs. 3, 4). Using funnel plots to assess potential publication bias, we observed slight asymmetry and identified 3 outlier studies36,47,52 (Fig. 5). We conducted meta-analyses including and excluding these studies. Overall, statistical evaluations did not confirm publication bias, although the power to demonstrate this bias is generally low.53
A recent committee review by the National Research Council recommended that meta-analysis of trichloroethylene and cancer should (1) identify and include all relevant studies and if studies are excluded, provide objective criteria for exclusion, (2) avoid use of subjective quality scoring, (3) assess heterogeneity, (4) analyze all studies combined (unless this introduces heterogeneity), and (5) conduct sensitivity analyses.56 All of these standard procedures for meta-analysis have been applied in this review.57 In addition, this meta-analysis included (1) recently published studies; (2) evaluated subgroups of studies defined by type of work, exposure assessment methods, and study design; (3) assessed exposure-response patterns; and (4) evaluated potential biases.
Group I Cohort Studies
The group I cohort studies used better exposure assessment methods compared with group II studies and had potentially more homogeneous trichloroethylene exposure compared with the case-control studies. However, the number of kidney cancer cases in most group I cohort studies was low. The 3 European cohort studies with biomonitoring had 4–7 cases each; the US aerospace/aircraft workers studies had 7–15 cases in the exposed subcohorts and 30–125 cases in the total cohorts. The Danish study25 had a total of 103 kidney cancer cases; the standardized incidence ratio was 1.2 (95% CI = 0.9–1.5) among all blue-collar workers (76 cases) and 1.4 (1.0–1.8) among the subcohort of blue-collar workers who were potentially more highly exposed (53 cases) (eTable, http://links.lww.com/EDE/A354). One of the aircraft-maintenance-worker studies43 appeared to have some of the highest trichloroethylene exposures and reported no overall excess risk or exposure-response pattern (eTable). This study was recently updated with follow-up through the year 2000, with similar results.58 The group I cohort studies had summary relative risk estimates ranging from 1.0 (biomonitoring) to 1.44 (aerospace workers), with an overall estimate of 1.34 (1.07–1.64).
Group II Cohort Studies
The group II cohort studies had less-specific trichloroethylene exposure information. Some of those studies had other design limitations including short latency,28 lack of individual exposure data, and limited exposure information (eg, material safety data sheets information only47). Henschler et al52 conducted a small cohort study of 169 workers exposed to trichloroethylene at a cardboard factory in Germany, where exposures are reported to be relatively high (eTable). Numerous limitations of this study have been discussed in previous reviews.20,21,23 After excluding 2 studies that introduced considerable heterogeneity and had serious design limitations, the group II studies do not indicate excess kidney cancer risk among workers potentially exposed to trichloroethylene (Table).
The findings of the case-control studies were mixed—some reported positive associations and others reported equivocal or no associations. Two studies in one region of Germany identified an increased risk of kidney cancer related to self-reported trichloroethylene exposure or work history.29,36 Vamvakas et al conducted a case-control study using cases defined as all renal cell cancer patients from the Urology Department of a specific hospital in Germany. However, shortcomings in case identification procedures, selection of controls, potential interview bias, and problems with matching may have led to biased results.20,21 The Bruning et al study29 improved upon several study design aspects of the Vamvakas et al study; however, hospital referral patterns in the region and the use of hospital-based controls may have introduced bias. All other studies collected data on work histories and job titles that were incorporated into a job-exposure matrix for trichloroethylene. Studies of self-reported trichloroethylene exposure reported higher relative risks than those based on a matrix, which could either indicate a cancer effect or reflect potential recall bias. A more recent investigation of occupational exposures in a wider region of Germany did not find substantial increases in renal cell cancer risk among men with “high exposure” to trichloroethylene.7 Among French workers there was an exposure-response trend, with an odds ratio of 2.2 in the high cumulative-exposure group.30 Greenland et al37 reported no increased risk of kidney cancer among workers from a US transformer assembly plant. Similarly, Siemiatycki35 evaluated self-reported trichloroethylene exposure as well as specific occupations and industries, and did not find elevated odds ratios for kidney cancer. In summary, the interpretation of the case-control studies depends largely on the weight given to, and the validity of, the 2 German studies. If these are included, the estimate is 1.59 (1.00–2.54; test for heterogeneity: P = 0.002); if excluded, the estimate is 1.14 (0.93–1.76; test for heterogeneity: P = 0.80).
A recent comprehensive review of trichloroethylene exposure in the United States reported several general trends relevant to this meta-analysis (1) exposures are higher in degreasing and vapor degreasing activities; (2) metal workers will likely have higher exposures; (3) trichloroethylene use was very limited in the dry-cleaning industry; (4) trichloroethylene use dropped dramatically in the 1980s; and (5) for many industries, trichloroethylene is reported as “having been used,” but there are only limited data available that quantify the potential exposures to workers. Many job groups that have often been listed as “trichloroethylene-exposed” (eg, electric components, rubber industry, paints, and lacquers) likely have had very limited trichloroethylene exposures, which suggests the possibility that case-control studies overestimate exposure.8
Exposure assessment has been a critical limitation among epidemiologic studies of trichloroethylene-exposed workers. Most studies have relied on some type of qualitative job exposure matrix, often with a dichotomous yes/no classification. Three European studies used biomonitoring data to establish their study cohorts, but the number of observed cancers was small, thus limiting the opportunity to use the more quantitative exposure information.24,45,46,59 In addition, the use of the urinary trichloroacetate biomarker may be limited in assessing long term/historical trichloroethylene exposure because the half-life of urinary trichloroacetate is relatively short (ie, days)46 and because, for nearly all participants, only a few biomarker measurements were available to represent an entire work history. The recent case-control study in France integrated more semi-quantitative exposure data, thus representing an improvement over most previous case-control studies.
Although based on limited data, there were no overall exposure-response patterns for duration of employment or cumulative exposure (Table, Figs. 3, 4). Similarly, study-specific exposure estimates did not show a consistent pattern of exposure response (eTable, http://links.lww.com/EDE/A354).
Several studies relied on job exposure or facility exposure matrices to assign individual worker exposure, which can be limited by unaccounted-for within-worker variability. For example the largest study25 classified all workers at trichloroethylene companies with 200 or fewer employees as “high trichloroethylene” assuming that a high proportion of workers at such facilities were exposed. In fact, the percentage of workers actually exposed at these smaller facilities varied from 1% to 100%.59,60
Thus it appears that exposure classification protocols in the Danish and perhaps US studies labeled more workers as exposed to trichloroethylene than were truly exposed. If this exposure misclassification was nondifferential it could lead to underestimating the summary relative risks; if differential, it could lead to either over- or underestimation of summary relative risks.
Confounding and Other Potential Biases
Smoking has been consistently identified as a risk factor for kidney cancer, with an average 2-fold relative risk and increasing risk with duration of smoking and total pack-years.6,61 Although data on smoking were not available in the trichloroethylene cohort studies, general surveys of smoking prevalence by industries labeled as “metal industries,” “machinery,” and occupational categories such as “fabrication,” “crafts workers,” “machine operators,” and “extraction-precision production” indicate that smoking prevalence is higher among these groups as compared with other occupational groups and the general population.62,63 The Danish investigators also noted that smoking was likely higher among blue collar workers compared with the general population.25 Thus, lack of control for smoking may have upwardly biased the relative risk estimates in the cohort studies that used external comparisons, but may not have affected case-control studies or cohort studies using internal comparison groups.
The impact of uncontrolled confounders (such as obesity) on meta-analysis results remains unknown. Diagnostic bias among employed populations, who may have better access to health care and thus may be more likely to have cancer diagnosed, may contribute to upwardly biased risk estimates, especially in areas such as the United States where most health insurance coverage depends on employment. Conversely, potential nondifferential disease misclassification when relying on mortality data could bias risk estimates towards the null value.
We found no consistent indication of publication bias based on various assessments, although removal of 3 studies (all reporting positive results) reduced heterogeneity considerably. All 3 studies were apparently initiated in response to reported disease clusters, and although researchers attempted to conduct a full epidemiologic investigation, the results will be predictably elevated given that the cohort was defined (or cases and controls were selected) after identification of the cluster.
While a simple summary meta-analysis of occupational exposure to trichloroethylene suggests an association with kidney cancer, more careful analyses of subgroups of studies indicate no association or, at best, moderately elevated associations. Exposure response analysis, although limited, did not show a pattern of higher risk with higher exposure. In addition, sources of systematic bias, such as confounding due to smoking and diagnostic bias, could plausibly explain the modest association between trichloroethylene and kidney cancer. Despite modest elevations in some summary relative risk estimates, insight into the role of trichloroethylene exposure and its association with kidney cancer is limited by the consideration of unmeasured confounding, the general lack of quantitative exposure assessment, and the absence of clear exposure-response trends.
We thank Betty Dowd for graphical assistance.
1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin
2. Janout V, Janoutova G. Epidemiology and risk factors of kidney cancer. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub
3. McLaughlin JK, Lipworth L, Tarone RE, Blot WJ. Renal cancer. In: Schottenfeld D, Fraumeni JF Jr, eds. Cancer Epidemiology and Prevention
. New York: Oxford University Press; 2006:1087–1100.
4. Moore LE, Wilson RT, Campleman SL. Lifestyle factors, exposures, genetic susceptibility, and renal cell cancer risk: a review. Cancer Invest
5. Lindblad P, Adami HO. Kidney cancer. In: Adami HO, Hunter D, Trichopoulos D, eds. Textbook of Cancer Epidemiology
. New York: Oxford University Press; 2002:467–485.
6. Kuper H, Boffetta P, Adami HO. Tobacco use and cancer causation: association by tumour type. J Intern Med
7. Pesch B, Haerting J, Ranft U, et al; MURC Study Group. Occupational risk factors for renal cell carcinoma: agent-specific results from a case-control study in Germany. Multicenter urothelial and renal cancer study. Int J Epidemiol
8. Bakke B, Stewart PA, Waters MA. Uses of and exposure to trichloroethylene in U.S. industry: a systematic literature review. J Occup Environ Hyg
9. Wong O. Carcinogenicity of trichloroethylene: an epidemiologic assessment. Clin Occup Environ Med
10. Blair A. Mortality among workers in the metal polishing and plating industry, 1951–1969. J Occup Med
11. International Agency for Research on Cancer. Trichloroethylene: IARC Monographs on the Evaluation of Carcinogenic Risks to Humans.
Vol 63. Lyon, France. World Health Organization; 1995.
12. US Department of Health and Human Services, Public Health Service, National Toxicology Program. Report on Carcinogens, Eleventh Edition.
Research Triangle Park, NC: National Institute of Environmental Health Sciences, 2005.
13. Maltoni C, Lefemine G, Cotti G, Perino G. Long-term carcinogenicity bioassays on trichloroethylene administered by inhalation to Sprague-Dawley rats and Swiss and B6C3F1 mice. Ann NY Acad Sci
14. Fukuda K, Takemoto K, Tsuruta H. Inhalation carcinogenicity of trichloroethylene in mice and rats. Ind Health
15. National Toxicology Program. NTP carcinogenesis studies of trichloroethylene (without epichlorohydrin) (CAS No. 79–01–6) in F344/N rats and B6C3F1 mice (Gavage studies). Natl Toxicol Program Tech Rep Ser
16. Rhomberg LR. Dose-response analyses of the carcinogenic effects of trichloroethylene in experimental animals. Environ Health Perspect
. 2000;108(suppl 2):343–358.
17. National Toxicology Program. NTP Toxicology and Carcinogenesis Studies of Trichloroethylene (CAS No. 79–01–6) in Four Strains of Rats (ACI, August, Marshall, Osborne-Mendel) (Gavage Studies). Natl Toxicol Program Tech Rep Ser
18. Henschler D, Elsasser H, Romen W, Eder E. Carcinogenicity study of trichloroethylene, with and without epoxide stabilizers, in mice. J Cancer Res Clin Oncol
19. Henschler D, Romen W, Elsasser HM, et al. Carcinogenicity study of trichloroethylene by longterm inhalation in three animal species. Arch Toxicol
20. Mandel JS, Kelsh MA. A review of the epidemiology of trichloroethylene and kidney cancer. Hum Ecol Risk Assessment
21. McLaughlin JK, Blot WJ. A critical review of epidemiology studies of trichloroethylene and perchloroethylene and risk of renal-cell cancer. Int Arch Occup Environ Health
22. Wartenberg D, Reyner D, Scott CS. Trichloroethylene and cancer: epidemiologic evidence. Environ Health Perspect
. 2000;108(suppl 2):161–176.
23. Weiss NS. Cancer in relation to occupational exposure to trichloroethylene. Occup Environ Med
24. Hansen J, Raaschou-Nielsen O, Christensen JM, et al. Cancer incidence among Danish workers exposed to trichloroethylene. J Occup Environ Med
25. Raaschou-Nielsen O, Hansen J, McLaughlin JK, et al. Cancer risk among workers at Danish companies using trichloroethylene: a cohort study. Am J Epidemiol
26. Boice JD Jr, Marano DE, Cohen SS, et al. Mortality among Rocketdyne workers who tested rocket engines, 1948–1999. J Occup Environ Med
27. Zhao Y, Krishnadasan A, Kennedy N, Morgenstern H, Ritz B. Estimated effects of solvents and mineral oils on cancer incidence and mortality in a cohort of aerospace workers. Am J Ind Med
28. Chang YM, Tai CF, Yang SC, et al. A cohort mortality study of workers exposed to chlorinated organic solvents in Taiwan. Ann Epidemiol
29. Bruning T, Pesch B, Wiesenhutter B, et al. Renal cell cancer risk and occupational exposure to trichloroethylene: results of a consecutive case-control study in Arnsberg, Germany. Am J Ind Med
30. Charbotel B, Fevotte J, Hours M, Martin JL, Bergeret A. Case-control study on renal cell cancer and occupational exposure to trichloroethylene. Part II: epidemiological aspects. Ann Occup Hyg
31. Alexander DD, Mink PJ, Mandel JH, Kelsh MA. A meta-analysis of occupational trichloroethylene exposure and multiple myeloma or leukaemia. Occup Med (Lond)
32. Mandel JH, Kelsh MA, Mink PJ, et al. Occupational trichloroethylene exposure and non-Hodgkin's lymphoma: a meta-analysis and review. Occup Environ Med
33. Lynge E, Anttila A, Hemminki K. Organic solvents and cancer. Cancer Causes Control
34. Alexander DD, Kelsh MA, Mink PJ, et al. A meta-analysis of occupational trichloroethylene exposure and liver cancer. Int Arch Occup Environ Health
35. Siemiatycki J. Risk Factors for Cancer in the Workplace
. Boca Raton, FL: CRC Press; 1991.
36. Vamvakas S, Bruning T, Thomasson B, et al. Renal cell cancer correlated with occupational exposure to trichloroethene. J Cancer Res Clin Oncol
37. Greenland S, Salvan A, Wegman DH, Hallock MF, Smith TJ. A case-control study of cancer mortality at a transformer-assembly facility. Int Arch Occup Environ Health
38. Dosemeci M, Cocco P, Chow WH. Gender differences in risk of renal cell carcinoma and occupational exposures to chlorinated aliphatic hydrocarbons. Am J Ind Med
39. Blair A, Petralia SA, Stewart PA. Extended mortality follow-up of a cohort of dry cleaners. Ann Epidemiol
40. Travier N, Gridley G, De Roos AJ, et al. Cancer incidence of dry cleaning, laundry and ironing workers in Sweden. Scand J Work Environ Health
41. Boice JD Jr, Marano DE, Fryzek JP, Sadler CJ, McLaughlin JK. Mortality among aircraft manufacturing workers. Occup Environ Med
42. Morgan RW, Kelsh MA, Zhao K, Heringer S. Mortality of aerospace workers exposed to trichloroethylene. Epidemiology
43. Blair A, Hartge P, Stewart PA, McAdams M, Lubin J. Mortality and cancer incidence of aircraft maintenance workers exposed to trichloroethylene and other organic solvents and chemicals: extended follow up. Occup Environ Med
44. Ritz B. Cancer mortality among workers exposed to chemicals during uranium processing. J Occup Environ Med
45. Axelson O, Selden A, Andersson K, Hogstedt C. Updated and expanded Swedish cohort study on trichloroethylene and cancer risk. J Occup Med
46. Anttila A, Pukkala E, Sallmen M, Hernberg S, Hemminki K. Cancer incidence among Finnish workers exposed to halogenated hydrocarbons. J Occup Environ Med
47. Sinks T, Lushniak B, Haussler BJ, et al. Renal cell cancer among paperboard printing workers. Epidemiology
48. Costa G, Merletti F, Segnan N. A mortality cohort study in a north Italian aircraft factory. Br J Ind Med
49. Selden A, Ahlborg G Jr Mortality and cancer morbidity after exposure to military aircraft fuel. Aviat Space Environ Med
50. Blair A, Haas T, Prosser R, et al. Mortality among United States Coast Guard marine inspectors. Arch Environ Health
51. Garabrant DH, Held J, Langholz B, Bernstein L. Mortality of aircraft manufacturing workers in southern California. Am J Ind Med
52. Henschler D, Vamvakas S, Lammert M, et al. Increased incidence of renal cell tumors in a cohort of cardboard workers exposed to trichloroethene. Arch Toxicol
53. Rothstein H, Sutton A, Borenstein M. Publication Bias in Meta-Analysis: Prevention, Assessment, and Adjustment
. Chichester, UK; John Wiley and Sons, Ltd; 2005.
54. Comprehensive Meta-Analysis version 2.2.046. Englewood, NJ: Biostat; 2007.
56. National Research Council of the National Academies. Assessing the Human Health Risks of Trichlroroethylene: Key Scientific Issues.
Washington, DC: National Academies Press; 2006.
57. Blair A, Burg J, Foran J, et al. Guidelines for application of meta-analysis in environmental epidemiology. ISLI Risk Science Institute. Regul Toxicol Pharmacol
58. Radican L, Blair A, Stewart P, Wartenberg D. Mortality of aircraft maintenance workers exposed to trichloroethylene and other hydrocarbons and chemicals: an extended follow-up. J Occup Environ Med
59. Hansen J. Cohort studies of cancer risk among Danish workers exposed to TCE. In: Symposium on new scientific research related to the health effects of trichloroethylene; 2004; Washington, DC.
60. Raaschou-Nielsen O, Hansen J, Thomsen BL, et al. Exposure of Danish workers to trichloroethylene, 1947–1989. Appl Occup Environ Hyg
61. Dhote R, Thiounn N, Debre B, Vidal-Trecan G. Risk factors for adult renal cell carcinoma. Urol Clin North Am
62. Leigh JP. Occupations, cigarette smoking, and lung cancer in the epidemiological follow-up to the NHANES I and the California Occupational Mortality Study. Bull NY Acad Med
63. Bang KM, Kim JH. Prevalence of cigarette smoking by occupation and industry in the United States. Am J Ind Med
64. Fevotte J, Charbotel B, Muller-Beaute P, Martin J, Hours M, Bergeret A. Case-control study of renal cell cancer and occupational exposure to trichloroethylene. Part 1: Exposure Assessment. Ann Occup Hyg