- Split View
-
Views
-
Cite
Cite
Dominik D. Alexander, Pamela J. Mink, Jeffrey H. Mandel, Michael A. Kelsh, A meta-analysis of occupational trichloroethylene exposure and multiple myeloma or leukaemia, Occupational Medicine, Volume 56, Issue 7, October 2006, Pages 485–493, https://doi.org/10.1093/occmed/kql083
- Share Icon Share
Abstract
Background Trichloroethylene (TCE) has been widely used as an industrial solvent and degreasing agent.
Aims We conducted a meta-analysis of epidemiologic studies of occupational TCE exposure and multiple myeloma (MM) or leukaemia.
Methods We identified a total of eight cohort or case–control studies that enumerated a TCE-exposed study population and presented relative risk (RR) estimates for MM (n = 7) and/or leukaemia (n = 7). The individual studies included aerospace or aircraft workers (n = 3 studies), workers from a transformer manufacturing plant (n = 1 study) and workers from numerous occupations who, based on biomonitoring or extensive industrial hygiene exposure measurements, were likely exposed to TCE (n = 4). We used random effects models to calculate summary relative risk estimates (SRRE). In addition, we examined heterogeneity across studies and the relative influence of each individual study on the overall meta-analysis.
Results No association was observed for MM (SRRE = 1.05, 95% CI: 0.80–1.38; P value for heterogeneity = 0.94) or leukaemia (SRRE = 1.11, 95% CI: 0.93–1.32; P value for heterogeneity = 0.50), based on TCE-exposed subgroup meta-analyses. Study-specific RR estimates for MM ranged between 0.57 and 1.62. RRs for leukaemia ranged between 1.05 and 1.15 in five studies, while one study reported a 2-fold increased RR and another study reported an inverse association of 0.60. All confidence intervals (CIs) for study-specific estimates included 1.0.
Conclusions The results of this meta-analysis do not support an etiologic association between occupational TCE exposure and risk of MM or leukaemia.
Introduction
Trichloroethylene (TCE) is a non-flammable, colourless liquid that has been used extensively worldwide for >70 years primarily as a chemical intermediate and industrial solvent to remove grease from metal parts [1,2]. Occupational groups commonly exposed to TCE include aircraft manufacturers, printers, painters, solvent workers, electronic equipment cleaners, mechanics and dry cleaners [3–5]. Workers involved in metal degreasing, who are exposed to TCE via inhalation, are presumably the most heavily exposed occupational group [6].
TCE is rapidly absorbed from the stomach, intestines and lung and is distributed throughout the body, concentrating in fatty tissues, such as the liver, brain and body fat [7]. The carcinogenicity of TCE based on animal models and epidemiologic studies has been reviewed [1,3,5,8]. Increased risks of liver, lung and kidney tumours have been reported in laboratory mice and rats [7,9–12]; however, species disparities in TCE metabolism preclude extrapolation to human carcinogenesis. Furthermore, some studies reported no indication of carcinogenic potential in animals [13,14]. Findings from epidemiologic studies have been inconsistent, and TCE has most commonly been associated with increased risks of liver and renal malignancies, as well as with non-Hodgkin's lymphoma [1,3,5,6,8,15,16]. Interpretation is limited, however, by possible occupational or lifestyle confounding factors [6]. In occupational settings, TCE is often one of many possible workplace exposures, further complicating interpretive issues. Moreover, few studies have analytically isolated occupational TCE exposure, often relying on job title or occupational group as a surrogate for exposure, which also hinders interpretation [8].
Several studies have evaluated the association between TCE in community drinking-water and risk of leukaemia, with inconsistent results [17–21]. A study of drinking-water contamination in 75 municipalities in New Jersey reported relative risks (RRs) of 1.43 for total leukaemias (95% CI: 1.07–1.90) and 2.35 for acute lymphocytic leukaemia (95% CI: 1.03–5.45) among women in the >5.0 ppb TCE-exposure category [18]. No significant associations were reported for men. A California community cancer assessment found no association between long-term exposure to TCE and perchlorate in drinking-water and total leukaemia [standardized incidence ratio (SIR) = 1.02, 99% CI: 0.74–1.35] [17].
Multiple myeloma (MM) and leukaemia are rare outcomes; thus, combining study findings will allow for quantifiable RR estimates with greater precision than results from single studies. We, therefore, conducted a comprehensive meta-analysis of studies that evaluated occupational TCE exposure and risk of these two haematological malignancies [22–29]. In addition, we evaluated other epidemiological studies that evaluated TCE exposure and MM or leukaemia, but were not included in our meta-analysis.
Methods
We conducted a comprehensive literature search using PubMed and identified 284 studies using the keywords [trichloroethylene] or [TCE] and [cancer] or [multiple myeloma] or [leukaemia]. We included a total of eight epidemiological cohort or case–control studies that presented RR estimates for International Classification of Diseases (ICD) codes that corresponded to MM or leukaemia or with data that allowed for the calculation of RRs for a TCE-exposed worker population (e.g. TCE used as a solvent among aircraft maintenance workers). Of these studies, six presented results for both MM and leukaemia, while one cohort study presented findings for MM only, and one case–control study presented findings for leukaemia only.
We excluded ecological studies of TCE in community drinking-water because study design issues limit their use for making causal inferences. We excluded studies of dry cleaners and laundry workers due to exposure assessment and study design limitations. For example, TCE exposure was not identified specifically in these cohorts, there was little or no TCE exposure for a significant portion of these workers, there was a lack of a distinction between dry cleaners and laundry workers who generally have not been exposed to TCE, and workers were predominantly exposed to chemicals other than TCE, such as perchloroethylene or carbon tetrachloride [4,8,30]. Furthermore, proportionate mortality analyses were conducted in several of these studies, thus, limiting potential causal interpretation. In two of the multiple industry studies that were included, dry cleaners represented a relatively small proportion of the study population, however [28,29]. A total of eight studies met our inclusion/exclusion criteria and, thus, were considered the most informative for making aetiological assessments pertaining to occupational TCE exposure and risk of MM or leukaemia.
Random effects models were used to calculate summary relative risk estimates (SRRE), 95% confidence intervals (CIs), and corresponding P values for heterogeneity. This type of model assumes that the study-specific effect sizes come from a random distribution of effect sizes according to a specific mean and variance. In calculating SRREs, the estimates of the individual studies were weighted by the inverse of the variance, which takes into account the sizes of the study populations. The relative influence of each study on the overall SRRE was examined by generating an SRRE with all studies included in the model, followed by the removal of one study at a time to compare the overall SRRE with SRREs from models that each had one individual study removed. This allowed us to determine the robustness of each analytical model. All analyses were performed using ‘Episheet’, a spreadsheet-based analytical package designed for the analysis of epidemiologic data [31].
We generated separate meta-analysis models for MM and leukaemia using data from studies that presented findings for a TCE-exposed subgroup or a population with known occupational TCE exposure. If an individual study presented overall cohort results as well as findings for a TCE-exposed subgroup, we extracted the data pertaining to the specific TCE-exposed population. This was done to provide the most accurate estimate of TCE RR that was available in each study. The lack of data for other exposure metrics (e.g. intensity of exposure) across multiple studies precluded a meta-analysis evaluation of exposure–response. Two studies presented results based on a lowest and highest TCE exposure category [24,26], while one study presented findings for MM based on duration of TCE exposure [23], and another study presented cumulative TCE exposure findings for MM and leukaemia [22].
The study of Morgan et al. [24] did not provide results specifically for MM, however; MM data for the study period 1960–93 was provided by one of the study authors (M.K.). The standardized mortality ratio (SMR) was subsequently calculated for the TCE-exposed subgroup. In addition, we calculated SMRs for low and high TCE-exposed workers.
Results
The studies included in this meta-analysis are presented in Table 1. Three US studies included aerospace or aircraft worker cohorts [22–24], while four European studies included cohorts of workers from various occupational groups who were exposed to TCE [26–29]. In a case–control study of US transformer manufacturing plant workers, Greenland et al. [25] presented results for leukaemia, based on TCE job exposure matrices generated by industrial hygienists. In the aerospace and aircraft worker studies, TCE exposure was assessed based on job descriptions, work history information and available industrial hygiene data [22–24]. Biomonitoring surveillance for the presence of urinary trichloroacetic acid concentrations (U-TCA), a metabolite of TCE in urine, was performed in three [26–28] European worker cohort studies. In the other European worker cohort [29], historical measurement files were used to select TCE-using industries.
Authors and year . | Study location . | Study design . | Study population . | Disease end-points . | Exposure assessment method . | Exposure comments . |
---|---|---|---|---|---|---|
Blair et al., 1998 [22] | USA | Cohort | 14 457 aircraft maintenance workers | MM, leukaemia | TCE and other chemicals (industrial hygienists and management reviewed historical records, monitoring data, chemical inventories, organization charts, technical orders, job descriptions) | TCE was the primary solvent, and was used for degreasing and cleaning. TCE was used in degreasers until 1979 and in bench top work until 1968 |
Boice et al., 1999 [23] | USA | Cohort | 77 965 aircraft manufacturing workers | MM, leukaemia | TCE and other chemicals (walk-through surveys, interviews, review of available industrial hygiene data, job titles, work history information) | Workers classified as having routine TCE exposure jobs were potentially exposed as a normal part of their daily activities. TCE was the primary solvent used in vapour degreasers until 1966, when it was replaced by perchloroethylene |
Morgan et al., 1998 [24] | USA | Cohort | 20 508 aerospace workers | MM, leukaemia | TCE (job exposure matrices compiled by industrial hygienists) | The highest TCE-exposed jobs entailed work on degreaser machines |
Greenland et al., 1994 [25] | USA | Case control | 1821 eligible cancer cases from a transformer-manufacturing plant (the number of leukaemia-specific cases were not provided) | Leukaemia | TCE and other chemicals (industrial hygienist generated exposure ratings based on job exposure matrices) | TCE was used as a degreaser from about 1930 to 1977 |
Anttila et al., 1995 [26] | Finland | Cohort | 3974 workers in various occupations monitored for exposure to TCE, PCE or 1,1,1-trichloroethane | MM, leukaemia | TCE, PCE or 1,1,1-trichloroethane (U-TCA concentrations were recorded during 1965–82) | TCE was mainly used in degreasing or cleaning of metal surfaces. It was also used in rubber work, gluing, dry cleaning and in cleaning fluids |
Axelson et al., 1994 [27] | Sweden | Cohort | 1670 workers drawn from 115 companies who were offered free biomonitoring surveillance for U-TCA | MM | TCE (U-TCA concentrations) | Workers were included if their companies used the surveillance service at least once between 1955 and 1975 |
Raaschou-Nielsen et al., 2003 [29] | Denmark | Cohort | 40 049 workers from various companies with confirmed use of TCE | MM, leukaemia | TCE (historical measurement files, records of the national Supplementary Pension Fund, type of industry, historical records, job title) | Historical measurement files were used to select TCE-using industries. The primary industries representing the cohort of workers were iron and metal, followed by electronics and painting |
Hansen et al., 2001 [28] | Denmark | Cohort | 803 workers from various companies who were exposed to TCE | MM, leukaemia | TCE (work histories were reconstructed from the files of the National Pension Fund, and for workers prior to 1964, measurement dates were used) | TCE exposure was based on the presence of U-TCA |
Authors and year . | Study location . | Study design . | Study population . | Disease end-points . | Exposure assessment method . | Exposure comments . |
---|---|---|---|---|---|---|
Blair et al., 1998 [22] | USA | Cohort | 14 457 aircraft maintenance workers | MM, leukaemia | TCE and other chemicals (industrial hygienists and management reviewed historical records, monitoring data, chemical inventories, organization charts, technical orders, job descriptions) | TCE was the primary solvent, and was used for degreasing and cleaning. TCE was used in degreasers until 1979 and in bench top work until 1968 |
Boice et al., 1999 [23] | USA | Cohort | 77 965 aircraft manufacturing workers | MM, leukaemia | TCE and other chemicals (walk-through surveys, interviews, review of available industrial hygiene data, job titles, work history information) | Workers classified as having routine TCE exposure jobs were potentially exposed as a normal part of their daily activities. TCE was the primary solvent used in vapour degreasers until 1966, when it was replaced by perchloroethylene |
Morgan et al., 1998 [24] | USA | Cohort | 20 508 aerospace workers | MM, leukaemia | TCE (job exposure matrices compiled by industrial hygienists) | The highest TCE-exposed jobs entailed work on degreaser machines |
Greenland et al., 1994 [25] | USA | Case control | 1821 eligible cancer cases from a transformer-manufacturing plant (the number of leukaemia-specific cases were not provided) | Leukaemia | TCE and other chemicals (industrial hygienist generated exposure ratings based on job exposure matrices) | TCE was used as a degreaser from about 1930 to 1977 |
Anttila et al., 1995 [26] | Finland | Cohort | 3974 workers in various occupations monitored for exposure to TCE, PCE or 1,1,1-trichloroethane | MM, leukaemia | TCE, PCE or 1,1,1-trichloroethane (U-TCA concentrations were recorded during 1965–82) | TCE was mainly used in degreasing or cleaning of metal surfaces. It was also used in rubber work, gluing, dry cleaning and in cleaning fluids |
Axelson et al., 1994 [27] | Sweden | Cohort | 1670 workers drawn from 115 companies who were offered free biomonitoring surveillance for U-TCA | MM | TCE (U-TCA concentrations) | Workers were included if their companies used the surveillance service at least once between 1955 and 1975 |
Raaschou-Nielsen et al., 2003 [29] | Denmark | Cohort | 40 049 workers from various companies with confirmed use of TCE | MM, leukaemia | TCE (historical measurement files, records of the national Supplementary Pension Fund, type of industry, historical records, job title) | Historical measurement files were used to select TCE-using industries. The primary industries representing the cohort of workers were iron and metal, followed by electronics and painting |
Hansen et al., 2001 [28] | Denmark | Cohort | 803 workers from various companies who were exposed to TCE | MM, leukaemia | TCE (work histories were reconstructed from the files of the National Pension Fund, and for workers prior to 1964, measurement dates were used) | TCE exposure was based on the presence of U-TCA |
Authors and year . | Study location . | Study design . | Study population . | Disease end-points . | Exposure assessment method . | Exposure comments . |
---|---|---|---|---|---|---|
Blair et al., 1998 [22] | USA | Cohort | 14 457 aircraft maintenance workers | MM, leukaemia | TCE and other chemicals (industrial hygienists and management reviewed historical records, monitoring data, chemical inventories, organization charts, technical orders, job descriptions) | TCE was the primary solvent, and was used for degreasing and cleaning. TCE was used in degreasers until 1979 and in bench top work until 1968 |
Boice et al., 1999 [23] | USA | Cohort | 77 965 aircraft manufacturing workers | MM, leukaemia | TCE and other chemicals (walk-through surveys, interviews, review of available industrial hygiene data, job titles, work history information) | Workers classified as having routine TCE exposure jobs were potentially exposed as a normal part of their daily activities. TCE was the primary solvent used in vapour degreasers until 1966, when it was replaced by perchloroethylene |
Morgan et al., 1998 [24] | USA | Cohort | 20 508 aerospace workers | MM, leukaemia | TCE (job exposure matrices compiled by industrial hygienists) | The highest TCE-exposed jobs entailed work on degreaser machines |
Greenland et al., 1994 [25] | USA | Case control | 1821 eligible cancer cases from a transformer-manufacturing plant (the number of leukaemia-specific cases were not provided) | Leukaemia | TCE and other chemicals (industrial hygienist generated exposure ratings based on job exposure matrices) | TCE was used as a degreaser from about 1930 to 1977 |
Anttila et al., 1995 [26] | Finland | Cohort | 3974 workers in various occupations monitored for exposure to TCE, PCE or 1,1,1-trichloroethane | MM, leukaemia | TCE, PCE or 1,1,1-trichloroethane (U-TCA concentrations were recorded during 1965–82) | TCE was mainly used in degreasing or cleaning of metal surfaces. It was also used in rubber work, gluing, dry cleaning and in cleaning fluids |
Axelson et al., 1994 [27] | Sweden | Cohort | 1670 workers drawn from 115 companies who were offered free biomonitoring surveillance for U-TCA | MM | TCE (U-TCA concentrations) | Workers were included if their companies used the surveillance service at least once between 1955 and 1975 |
Raaschou-Nielsen et al., 2003 [29] | Denmark | Cohort | 40 049 workers from various companies with confirmed use of TCE | MM, leukaemia | TCE (historical measurement files, records of the national Supplementary Pension Fund, type of industry, historical records, job title) | Historical measurement files were used to select TCE-using industries. The primary industries representing the cohort of workers were iron and metal, followed by electronics and painting |
Hansen et al., 2001 [28] | Denmark | Cohort | 803 workers from various companies who were exposed to TCE | MM, leukaemia | TCE (work histories were reconstructed from the files of the National Pension Fund, and for workers prior to 1964, measurement dates were used) | TCE exposure was based on the presence of U-TCA |
Authors and year . | Study location . | Study design . | Study population . | Disease end-points . | Exposure assessment method . | Exposure comments . |
---|---|---|---|---|---|---|
Blair et al., 1998 [22] | USA | Cohort | 14 457 aircraft maintenance workers | MM, leukaemia | TCE and other chemicals (industrial hygienists and management reviewed historical records, monitoring data, chemical inventories, organization charts, technical orders, job descriptions) | TCE was the primary solvent, and was used for degreasing and cleaning. TCE was used in degreasers until 1979 and in bench top work until 1968 |
Boice et al., 1999 [23] | USA | Cohort | 77 965 aircraft manufacturing workers | MM, leukaemia | TCE and other chemicals (walk-through surveys, interviews, review of available industrial hygiene data, job titles, work history information) | Workers classified as having routine TCE exposure jobs were potentially exposed as a normal part of their daily activities. TCE was the primary solvent used in vapour degreasers until 1966, when it was replaced by perchloroethylene |
Morgan et al., 1998 [24] | USA | Cohort | 20 508 aerospace workers | MM, leukaemia | TCE (job exposure matrices compiled by industrial hygienists) | The highest TCE-exposed jobs entailed work on degreaser machines |
Greenland et al., 1994 [25] | USA | Case control | 1821 eligible cancer cases from a transformer-manufacturing plant (the number of leukaemia-specific cases were not provided) | Leukaemia | TCE and other chemicals (industrial hygienist generated exposure ratings based on job exposure matrices) | TCE was used as a degreaser from about 1930 to 1977 |
Anttila et al., 1995 [26] | Finland | Cohort | 3974 workers in various occupations monitored for exposure to TCE, PCE or 1,1,1-trichloroethane | MM, leukaemia | TCE, PCE or 1,1,1-trichloroethane (U-TCA concentrations were recorded during 1965–82) | TCE was mainly used in degreasing or cleaning of metal surfaces. It was also used in rubber work, gluing, dry cleaning and in cleaning fluids |
Axelson et al., 1994 [27] | Sweden | Cohort | 1670 workers drawn from 115 companies who were offered free biomonitoring surveillance for U-TCA | MM | TCE (U-TCA concentrations) | Workers were included if their companies used the surveillance service at least once between 1955 and 1975 |
Raaschou-Nielsen et al., 2003 [29] | Denmark | Cohort | 40 049 workers from various companies with confirmed use of TCE | MM, leukaemia | TCE (historical measurement files, records of the national Supplementary Pension Fund, type of industry, historical records, job title) | Historical measurement files were used to select TCE-using industries. The primary industries representing the cohort of workers were iron and metal, followed by electronics and painting |
Hansen et al., 2001 [28] | Denmark | Cohort | 803 workers from various companies who were exposed to TCE | MM, leukaemia | TCE (work histories were reconstructed from the files of the National Pension Fund, and for workers prior to 1964, measurement dates were used) | TCE exposure was based on the presence of U-TCA |
Multiple myeloma
There was no association between occupational TCE exposure and risk of MM (SRRE = 1.05, 95% CI: 0.80–1.38) (Table 2). The P value for heterogeneity was 0.94, indicating homogeneous findings across studies. All CIs for the individual studies included 1.0, and RR estimates ranged between 0.57 and 1.62. Of the seven studies included in this meta-analysis model, five reported six or fewer exposed MM cases or deaths. Raaschou-Nielsen et al. [29] included 31 MM cases and, based on our influence analysis, accounted for 57% of the relative weight of the SRRE. The exclusion of this study in our sensitivity analysis, however, did not appreciably change the result (SRRE = 1.08, 95% CI: 0.72–1.64; P value for heterogeneity = 0.88).
MM . | . | . | . | . | Leukaemia . | . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | ||||||||
Blair et al., 1998 [22] | RR | 14 | 1.30 | 0.50–3.40 | [22] | RR | 16 | 0.60 | 0.30–1.20 | ||||||||
Boice et al., 1999 [23] | SMR | 6 | 0.91 | 0.34–1.99 | [23] | SMR | 12 | 1.05 | 0.54–1.84 | ||||||||
Raaschou-Nielsen et al., 2003 [29] | SIR | 31 | 1.03a | 0.70–1.47 | [29] | SIR | 82 | 1.15a | 0.91–1.42 | ||||||||
Axelson et al., 1994 [27] | SIR | 1 | 0.57 | 0.01–3.17 | [25] | OR | NA | 1.10 | 0.46–2.66 | ||||||||
Anttila et al., 1995 [26] | SIR | 4 | 1.62 | 0.44–4.16 | [26] | SIR | 5 | 1.08 | 0.35–2.53 | ||||||||
Hansen et al., 2001 [28] | SIR | 1 | 0.71a | 0.02–3.98 | [28] | SIR | 6 | 2.00a | 0.73–4.35 | ||||||||
Morgan et al., 1998 [24] | SMR | 5 | 1.08b | 0.35–2.53 | [24] | SMR | 10 | 1.05 | 0.50–1.93 | ||||||||
Summary RR estimate | 1.05 | 0.80–1.38 | Summary RR estimate | 1.11 | 0.93–1.32 | ||||||||||||
P for heterogeneity | 0.94 | P for heterogeneity | 0.50 |
MM . | . | . | . | . | Leukaemia . | . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | ||||||||
Blair et al., 1998 [22] | RR | 14 | 1.30 | 0.50–3.40 | [22] | RR | 16 | 0.60 | 0.30–1.20 | ||||||||
Boice et al., 1999 [23] | SMR | 6 | 0.91 | 0.34–1.99 | [23] | SMR | 12 | 1.05 | 0.54–1.84 | ||||||||
Raaschou-Nielsen et al., 2003 [29] | SIR | 31 | 1.03a | 0.70–1.47 | [29] | SIR | 82 | 1.15a | 0.91–1.42 | ||||||||
Axelson et al., 1994 [27] | SIR | 1 | 0.57 | 0.01–3.17 | [25] | OR | NA | 1.10 | 0.46–2.66 | ||||||||
Anttila et al., 1995 [26] | SIR | 4 | 1.62 | 0.44–4.16 | [26] | SIR | 5 | 1.08 | 0.35–2.53 | ||||||||
Hansen et al., 2001 [28] | SIR | 1 | 0.71a | 0.02–3.98 | [28] | SIR | 6 | 2.00a | 0.73–4.35 | ||||||||
Morgan et al., 1998 [24] | SMR | 5 | 1.08b | 0.35–2.53 | [24] | SMR | 10 | 1.05 | 0.50–1.93 | ||||||||
Summary RR estimate | 1.05 | 0.80–1.38 | Summary RR estimate | 1.11 | 0.93–1.32 | ||||||||||||
P for heterogeneity | 0.94 | P for heterogeneity | 0.50 |
Combined data for men and women and recalculated using Fishers' exact method.
Recalculated using MM-specific data.
MM . | . | . | . | . | Leukaemia . | . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | ||||||||
Blair et al., 1998 [22] | RR | 14 | 1.30 | 0.50–3.40 | [22] | RR | 16 | 0.60 | 0.30–1.20 | ||||||||
Boice et al., 1999 [23] | SMR | 6 | 0.91 | 0.34–1.99 | [23] | SMR | 12 | 1.05 | 0.54–1.84 | ||||||||
Raaschou-Nielsen et al., 2003 [29] | SIR | 31 | 1.03a | 0.70–1.47 | [29] | SIR | 82 | 1.15a | 0.91–1.42 | ||||||||
Axelson et al., 1994 [27] | SIR | 1 | 0.57 | 0.01–3.17 | [25] | OR | NA | 1.10 | 0.46–2.66 | ||||||||
Anttila et al., 1995 [26] | SIR | 4 | 1.62 | 0.44–4.16 | [26] | SIR | 5 | 1.08 | 0.35–2.53 | ||||||||
Hansen et al., 2001 [28] | SIR | 1 | 0.71a | 0.02–3.98 | [28] | SIR | 6 | 2.00a | 0.73–4.35 | ||||||||
Morgan et al., 1998 [24] | SMR | 5 | 1.08b | 0.35–2.53 | [24] | SMR | 10 | 1.05 | 0.50–1.93 | ||||||||
Summary RR estimate | 1.05 | 0.80–1.38 | Summary RR estimate | 1.11 | 0.93–1.32 | ||||||||||||
P for heterogeneity | 0.94 | P for heterogeneity | 0.50 |
MM . | . | . | . | . | Leukaemia . | . | . | . | . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | Authors and year . | Type of risk estimate . | No. of exposed cases or deaths . | Risk estimate . | 95% CI . | ||||||||
Blair et al., 1998 [22] | RR | 14 | 1.30 | 0.50–3.40 | [22] | RR | 16 | 0.60 | 0.30–1.20 | ||||||||
Boice et al., 1999 [23] | SMR | 6 | 0.91 | 0.34–1.99 | [23] | SMR | 12 | 1.05 | 0.54–1.84 | ||||||||
Raaschou-Nielsen et al., 2003 [29] | SIR | 31 | 1.03a | 0.70–1.47 | [29] | SIR | 82 | 1.15a | 0.91–1.42 | ||||||||
Axelson et al., 1994 [27] | SIR | 1 | 0.57 | 0.01–3.17 | [25] | OR | NA | 1.10 | 0.46–2.66 | ||||||||
Anttila et al., 1995 [26] | SIR | 4 | 1.62 | 0.44–4.16 | [26] | SIR | 5 | 1.08 | 0.35–2.53 | ||||||||
Hansen et al., 2001 [28] | SIR | 1 | 0.71a | 0.02–3.98 | [28] | SIR | 6 | 2.00a | 0.73–4.35 | ||||||||
Morgan et al., 1998 [24] | SMR | 5 | 1.08b | 0.35–2.53 | [24] | SMR | 10 | 1.05 | 0.50–1.93 | ||||||||
Summary RR estimate | 1.05 | 0.80–1.38 | Summary RR estimate | 1.11 | 0.93–1.32 | ||||||||||||
P for heterogeneity | 0.94 | P for heterogeneity | 0.50 |
Combined data for men and women and recalculated using Fishers' exact method.
Recalculated using MM-specific data.
Categories of U-TCA concentrations or levels of TCE-exposure intensity were ascertained in some studies [27–29], but we were unable to conduct biological gradient meta-analyses across these studies due to small numbers of exposed cases or deaths. Two studies [24,26] reported elevated RR estimates for ‘high’ levels of TCE exposure, although findings were not significant and were based on four or fewer exposed cases (Table 3). Boice et al. [23] reported no increasing trend in MM risk based on time since first hire, duration of employment or increasing years of potential routine or intermittent TCE exposure (Table 3). Blair et al. [22] reported a 5-fold non-significant increased risk of incident MM among men in the highest category of cumulative TCE exposure (>25 unit/year: RR = 5.1, 95% CI: 0.60–43.7), based on five exposed cases, but did not find similar results based on mortality data (Table 3). The authors also reported an elevated RR for male workers exposed to chemicals other than TCE (RR = 3.7, 95% CI: 0.4–31.7) (Table 3).
Author and year . | Type of exposure analysis . | Exposure categories . | No. of exposed cases or deaths . | RR estimate (95% CI) . | Study notes . |
---|---|---|---|---|---|
MM | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 2 | 1.48 (0.18–5.35) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 2 | 2.41 (0.29–8.71) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 1 | 0.47 (0.01–2.61) | Findings were based on industrial hygiene job estimates and were recalculated using MM-specific data |
High | Approximately ≥50 ppm | 4 | 1.60 (0.44–4.10) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 6 | 1.08 (0.40–2.34) | There was no exposure–response relationship based on duration of potential routine or intermittent TCE exposure (P value for trend ≥0.20) |
10–19 years | 14 | 1.08 (0.59–1.82) | |||
≥20 years | 70 | 1.20 (0.94–1.52) | |||
Duration of employmenta | <10 years | 20 | 1.06 (0.65–1.63) | ||
10–19 years | 23 | 1.31 (0.83–1.96) | |||
20–29 years | 22 | 0.99 (0.62–1.50) | |||
≥30 years | 25 | 1.40 (0.90–2.07) | |||
Duration of potential TCE exposure (routine or intermittent) | 0 years | 24 | 1.00 | ||
<1 year | 3 | 0.45 (0.13–1.54) | |||
1–4 years | 8 | 1.48 (0.64–3.41) | |||
≥5 years | 3 | 0.51 (0.15–1.76) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 10 | 1.7 (0.5–5.5) | Other chemicals, such as methylene chloride, methyl ethyl ketone and carbon tetrachloride were positively associated with MM among workers in this cohort, thus the effects of confounding cannot be ruled out. The authors note that, ‘inconsistencies by sex and exposure level do not make a strong case for an association between these two cancers [multiple myeloma and non-Hodgkin's lymphoma] and exposure to trichloroethylene’ |
<5 unit/year | 4 | 1.0 (0.2–4.2) | |||
5–25 unit/year | 2 | 0.8 (0.1–4.4) | |||
>25 unit/year | 4 | 1.2 (0.3–4.7) | |||
Women: no exposure | 11 | 1.0 (0.1–9.9) | |||
<5 unit/year | 2 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 1 | 4.3 (0.4–43.4) | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 5 | 3.7 (0.4–31.7) | ||
<5 unit/year | 1 | 0.8 (01–12.7) | |||
5–25 unit/year | 3 | 3.8 (0.4–37.4) | |||
>25 unit/year | 5 | 5.1 (0.6–43.7) | |||
Women: no exposure | 1 | 1 observed/0 expected | |||
<5 unit/year | 2 | 2 observed/0 expected | |||
5–25 unit/year | 1 | 1 observed/0 expected | |||
>25 unit/year | 1 | 1 observed/0 expected | |||
Low level exposure | Men: intermittent | 4 | 0.5 (0.1–1.9) | ||
Continuous | 4 | 0.8 (0.2–3.2) | |||
Women: intermittent | 4 | 4.4 (1.0–20.4) | |||
Continuous | 1 | 2.4 (0.2–24.3) | |||
Leukaemia | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 1 | 0.39 (0.01–2.19) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 4 | 2.65 (0.72–6.78) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 3 | 0.85 (0.17–2.47) | Findings were based on industrial hygiene job estimates, and are not indicative of an exposure–response relationship |
High | Approximately ≥50 ppm | 7 | 1.17 (0.47–2.41) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 9 | 0.81 (0.37–1.53) | These findings represent the total cohort. TCE-specific analyses for duration or intensity of exposure were not performed |
10–19 years | 18 | 0.78 (0.46–1.23) | |||
≥20 years | 119 | 1.14 (0.94–1.36) | |||
Duration of employmenta | <10 years | 47 | 1.21 (0.89–1.61) | ||
10–19 years | 23 | 0.76 (0.48–1.14) | |||
20–29 years | 38 | 1.02 (0.72–1.40) | |||
≥30 years | 38 | 1.15 (0.82–1.59) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 9 | 1.0 (0.4–2.9) | The lack of an exposure–response relationship and gender differences in associations by exposure level do not support an etiologic association between TCE and leukaemia in this cohort of aircraft maintenance workers |
<5 unit/year | 7 | 1.0 (0.3–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 7 | 1.2 (0.4–3.6) | |||
Women: no exposure | 0 | 0 deaths | |||
<5 unit/year | 1 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 6 | 1.0 (0.3–3.7) | ||
<5 unit/year | 2 | 0.4 (0.1–2.0) | |||
5–25 unit/year | 0 | 0 cases | |||
>25 unit/year | 4 | 0.9 (0.2–3.7) | |||
Women: no exposure | 0 | 0 cases | |||
<5 unit/year | 0 | 0 cases | |||
5–25 unit/year | 1 | 2.4 (0.3–21.8) | |||
>25 unit/year | 0 | 0 cases |
Author and year . | Type of exposure analysis . | Exposure categories . | No. of exposed cases or deaths . | RR estimate (95% CI) . | Study notes . |
---|---|---|---|---|---|
MM | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 2 | 1.48 (0.18–5.35) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 2 | 2.41 (0.29–8.71) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 1 | 0.47 (0.01–2.61) | Findings were based on industrial hygiene job estimates and were recalculated using MM-specific data |
High | Approximately ≥50 ppm | 4 | 1.60 (0.44–4.10) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 6 | 1.08 (0.40–2.34) | There was no exposure–response relationship based on duration of potential routine or intermittent TCE exposure (P value for trend ≥0.20) |
10–19 years | 14 | 1.08 (0.59–1.82) | |||
≥20 years | 70 | 1.20 (0.94–1.52) | |||
Duration of employmenta | <10 years | 20 | 1.06 (0.65–1.63) | ||
10–19 years | 23 | 1.31 (0.83–1.96) | |||
20–29 years | 22 | 0.99 (0.62–1.50) | |||
≥30 years | 25 | 1.40 (0.90–2.07) | |||
Duration of potential TCE exposure (routine or intermittent) | 0 years | 24 | 1.00 | ||
<1 year | 3 | 0.45 (0.13–1.54) | |||
1–4 years | 8 | 1.48 (0.64–3.41) | |||
≥5 years | 3 | 0.51 (0.15–1.76) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 10 | 1.7 (0.5–5.5) | Other chemicals, such as methylene chloride, methyl ethyl ketone and carbon tetrachloride were positively associated with MM among workers in this cohort, thus the effects of confounding cannot be ruled out. The authors note that, ‘inconsistencies by sex and exposure level do not make a strong case for an association between these two cancers [multiple myeloma and non-Hodgkin's lymphoma] and exposure to trichloroethylene’ |
<5 unit/year | 4 | 1.0 (0.2–4.2) | |||
5–25 unit/year | 2 | 0.8 (0.1–4.4) | |||
>25 unit/year | 4 | 1.2 (0.3–4.7) | |||
Women: no exposure | 11 | 1.0 (0.1–9.9) | |||
<5 unit/year | 2 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 1 | 4.3 (0.4–43.4) | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 5 | 3.7 (0.4–31.7) | ||
<5 unit/year | 1 | 0.8 (01–12.7) | |||
5–25 unit/year | 3 | 3.8 (0.4–37.4) | |||
>25 unit/year | 5 | 5.1 (0.6–43.7) | |||
Women: no exposure | 1 | 1 observed/0 expected | |||
<5 unit/year | 2 | 2 observed/0 expected | |||
5–25 unit/year | 1 | 1 observed/0 expected | |||
>25 unit/year | 1 | 1 observed/0 expected | |||
Low level exposure | Men: intermittent | 4 | 0.5 (0.1–1.9) | ||
Continuous | 4 | 0.8 (0.2–3.2) | |||
Women: intermittent | 4 | 4.4 (1.0–20.4) | |||
Continuous | 1 | 2.4 (0.2–24.3) | |||
Leukaemia | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 1 | 0.39 (0.01–2.19) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 4 | 2.65 (0.72–6.78) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 3 | 0.85 (0.17–2.47) | Findings were based on industrial hygiene job estimates, and are not indicative of an exposure–response relationship |
High | Approximately ≥50 ppm | 7 | 1.17 (0.47–2.41) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 9 | 0.81 (0.37–1.53) | These findings represent the total cohort. TCE-specific analyses for duration or intensity of exposure were not performed |
10–19 years | 18 | 0.78 (0.46–1.23) | |||
≥20 years | 119 | 1.14 (0.94–1.36) | |||
Duration of employmenta | <10 years | 47 | 1.21 (0.89–1.61) | ||
10–19 years | 23 | 0.76 (0.48–1.14) | |||
20–29 years | 38 | 1.02 (0.72–1.40) | |||
≥30 years | 38 | 1.15 (0.82–1.59) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 9 | 1.0 (0.4–2.9) | The lack of an exposure–response relationship and gender differences in associations by exposure level do not support an etiologic association between TCE and leukaemia in this cohort of aircraft maintenance workers |
<5 unit/year | 7 | 1.0 (0.3–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 7 | 1.2 (0.4–3.6) | |||
Women: no exposure | 0 | 0 deaths | |||
<5 unit/year | 1 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 6 | 1.0 (0.3–3.7) | ||
<5 unit/year | 2 | 0.4 (0.1–2.0) | |||
5–25 unit/year | 0 | 0 cases | |||
>25 unit/year | 4 | 0.9 (0.2–3.7) | |||
Women: no exposure | 0 | 0 cases | |||
<5 unit/year | 0 | 0 cases | |||
5–25 unit/year | 1 | 2.4 (0.3–21.8) | |||
>25 unit/year | 0 | 0 cases |
Findings for total cohort.
Author and year . | Type of exposure analysis . | Exposure categories . | No. of exposed cases or deaths . | RR estimate (95% CI) . | Study notes . |
---|---|---|---|---|---|
MM | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 2 | 1.48 (0.18–5.35) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 2 | 2.41 (0.29–8.71) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 1 | 0.47 (0.01–2.61) | Findings were based on industrial hygiene job estimates and were recalculated using MM-specific data |
High | Approximately ≥50 ppm | 4 | 1.60 (0.44–4.10) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 6 | 1.08 (0.40–2.34) | There was no exposure–response relationship based on duration of potential routine or intermittent TCE exposure (P value for trend ≥0.20) |
10–19 years | 14 | 1.08 (0.59–1.82) | |||
≥20 years | 70 | 1.20 (0.94–1.52) | |||
Duration of employmenta | <10 years | 20 | 1.06 (0.65–1.63) | ||
10–19 years | 23 | 1.31 (0.83–1.96) | |||
20–29 years | 22 | 0.99 (0.62–1.50) | |||
≥30 years | 25 | 1.40 (0.90–2.07) | |||
Duration of potential TCE exposure (routine or intermittent) | 0 years | 24 | 1.00 | ||
<1 year | 3 | 0.45 (0.13–1.54) | |||
1–4 years | 8 | 1.48 (0.64–3.41) | |||
≥5 years | 3 | 0.51 (0.15–1.76) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 10 | 1.7 (0.5–5.5) | Other chemicals, such as methylene chloride, methyl ethyl ketone and carbon tetrachloride were positively associated with MM among workers in this cohort, thus the effects of confounding cannot be ruled out. The authors note that, ‘inconsistencies by sex and exposure level do not make a strong case for an association between these two cancers [multiple myeloma and non-Hodgkin's lymphoma] and exposure to trichloroethylene’ |
<5 unit/year | 4 | 1.0 (0.2–4.2) | |||
5–25 unit/year | 2 | 0.8 (0.1–4.4) | |||
>25 unit/year | 4 | 1.2 (0.3–4.7) | |||
Women: no exposure | 11 | 1.0 (0.1–9.9) | |||
<5 unit/year | 2 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 1 | 4.3 (0.4–43.4) | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 5 | 3.7 (0.4–31.7) | ||
<5 unit/year | 1 | 0.8 (01–12.7) | |||
5–25 unit/year | 3 | 3.8 (0.4–37.4) | |||
>25 unit/year | 5 | 5.1 (0.6–43.7) | |||
Women: no exposure | 1 | 1 observed/0 expected | |||
<5 unit/year | 2 | 2 observed/0 expected | |||
5–25 unit/year | 1 | 1 observed/0 expected | |||
>25 unit/year | 1 | 1 observed/0 expected | |||
Low level exposure | Men: intermittent | 4 | 0.5 (0.1–1.9) | ||
Continuous | 4 | 0.8 (0.2–3.2) | |||
Women: intermittent | 4 | 4.4 (1.0–20.4) | |||
Continuous | 1 | 2.4 (0.2–24.3) | |||
Leukaemia | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 1 | 0.39 (0.01–2.19) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 4 | 2.65 (0.72–6.78) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 3 | 0.85 (0.17–2.47) | Findings were based on industrial hygiene job estimates, and are not indicative of an exposure–response relationship |
High | Approximately ≥50 ppm | 7 | 1.17 (0.47–2.41) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 9 | 0.81 (0.37–1.53) | These findings represent the total cohort. TCE-specific analyses for duration or intensity of exposure were not performed |
10–19 years | 18 | 0.78 (0.46–1.23) | |||
≥20 years | 119 | 1.14 (0.94–1.36) | |||
Duration of employmenta | <10 years | 47 | 1.21 (0.89–1.61) | ||
10–19 years | 23 | 0.76 (0.48–1.14) | |||
20–29 years | 38 | 1.02 (0.72–1.40) | |||
≥30 years | 38 | 1.15 (0.82–1.59) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 9 | 1.0 (0.4–2.9) | The lack of an exposure–response relationship and gender differences in associations by exposure level do not support an etiologic association between TCE and leukaemia in this cohort of aircraft maintenance workers |
<5 unit/year | 7 | 1.0 (0.3–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 7 | 1.2 (0.4–3.6) | |||
Women: no exposure | 0 | 0 deaths | |||
<5 unit/year | 1 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 6 | 1.0 (0.3–3.7) | ||
<5 unit/year | 2 | 0.4 (0.1–2.0) | |||
5–25 unit/year | 0 | 0 cases | |||
>25 unit/year | 4 | 0.9 (0.2–3.7) | |||
Women: no exposure | 0 | 0 cases | |||
<5 unit/year | 0 | 0 cases | |||
5–25 unit/year | 1 | 2.4 (0.3–21.8) | |||
>25 unit/year | 0 | 0 cases |
Author and year . | Type of exposure analysis . | Exposure categories . | No. of exposed cases or deaths . | RR estimate (95% CI) . | Study notes . |
---|---|---|---|---|---|
MM | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 2 | 1.48 (0.18–5.35) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 2 | 2.41 (0.29–8.71) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 1 | 0.47 (0.01–2.61) | Findings were based on industrial hygiene job estimates and were recalculated using MM-specific data |
High | Approximately ≥50 ppm | 4 | 1.60 (0.44–4.10) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 6 | 1.08 (0.40–2.34) | There was no exposure–response relationship based on duration of potential routine or intermittent TCE exposure (P value for trend ≥0.20) |
10–19 years | 14 | 1.08 (0.59–1.82) | |||
≥20 years | 70 | 1.20 (0.94–1.52) | |||
Duration of employmenta | <10 years | 20 | 1.06 (0.65–1.63) | ||
10–19 years | 23 | 1.31 (0.83–1.96) | |||
20–29 years | 22 | 0.99 (0.62–1.50) | |||
≥30 years | 25 | 1.40 (0.90–2.07) | |||
Duration of potential TCE exposure (routine or intermittent) | 0 years | 24 | 1.00 | ||
<1 year | 3 | 0.45 (0.13–1.54) | |||
1–4 years | 8 | 1.48 (0.64–3.41) | |||
≥5 years | 3 | 0.51 (0.15–1.76) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 10 | 1.7 (0.5–5.5) | Other chemicals, such as methylene chloride, methyl ethyl ketone and carbon tetrachloride were positively associated with MM among workers in this cohort, thus the effects of confounding cannot be ruled out. The authors note that, ‘inconsistencies by sex and exposure level do not make a strong case for an association between these two cancers [multiple myeloma and non-Hodgkin's lymphoma] and exposure to trichloroethylene’ |
<5 unit/year | 4 | 1.0 (0.2–4.2) | |||
5–25 unit/year | 2 | 0.8 (0.1–4.4) | |||
>25 unit/year | 4 | 1.2 (0.3–4.7) | |||
Women: no exposure | 11 | 1.0 (0.1–9.9) | |||
<5 unit/year | 2 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 1 | 4.3 (0.4–43.4) | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 5 | 3.7 (0.4–31.7) | ||
<5 unit/year | 1 | 0.8 (01–12.7) | |||
5–25 unit/year | 3 | 3.8 (0.4–37.4) | |||
>25 unit/year | 5 | 5.1 (0.6–43.7) | |||
Women: no exposure | 1 | 1 observed/0 expected | |||
<5 unit/year | 2 | 2 observed/0 expected | |||
5–25 unit/year | 1 | 1 observed/0 expected | |||
>25 unit/year | 1 | 1 observed/0 expected | |||
Low level exposure | Men: intermittent | 4 | 0.5 (0.1–1.9) | ||
Continuous | 4 | 0.8 (0.2–3.2) | |||
Women: intermittent | 4 | 4.4 (1.0–20.4) | |||
Continuous | 1 | 2.4 (0.2–24.3) | |||
Leukaemia | |||||
Anttila et al., 1995 [26] | Low | <100 μmol/l | 1 | 0.39 (0.01–2.19) | Exposure classification was based on mean personal U-TCA concentrations |
High | 100+ μmol/l | 4 | 2.65 (0.72–6.78) | ||
Morgan et al., 1998 [24] | Low | Approximately <50 ppm | 3 | 0.85 (0.17–2.47) | Findings were based on industrial hygiene job estimates, and are not indicative of an exposure–response relationship |
High | Approximately ≥50 ppm | 7 | 1.17 (0.47–2.41) | ||
Boice et al., 1999 [23] | Time since first hirea | <10 years | 9 | 0.81 (0.37–1.53) | These findings represent the total cohort. TCE-specific analyses for duration or intensity of exposure were not performed |
10–19 years | 18 | 0.78 (0.46–1.23) | |||
≥20 years | 119 | 1.14 (0.94–1.36) | |||
Duration of employmenta | <10 years | 47 | 1.21 (0.89–1.61) | ||
10–19 years | 23 | 0.76 (0.48–1.14) | |||
20–29 years | 38 | 1.02 (0.72–1.40) | |||
≥30 years | 38 | 1.15 (0.82–1.59) | |||
Blair et al., 1998 [22] | Cumulative exposure (mortality) | Men: no exposure | 9 | 1.0 (0.4–2.9) | The lack of an exposure–response relationship and gender differences in associations by exposure level do not support an etiologic association between TCE and leukaemia in this cohort of aircraft maintenance workers |
<5 unit/year | 7 | 1.0 (0.3–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 7 | 1.2 (0.4–3.6) | |||
Women: no exposure | 0 | 0 deaths | |||
<5 unit/year | 1 | 0.4 (0.1–3.2) | |||
5–25 unit/year | 0 | 0 deaths | |||
>25 unit/year | 1 | 0.3 (0.1–2.4) | |||
Cumulative exposure (incidence) | Men: no exposure | 6 | 1.0 (0.3–3.7) | ||
<5 unit/year | 2 | 0.4 (0.1–2.0) | |||
5–25 unit/year | 0 | 0 cases | |||
>25 unit/year | 4 | 0.9 (0.2–3.7) | |||
Women: no exposure | 0 | 0 cases | |||
<5 unit/year | 0 | 0 cases | |||
5–25 unit/year | 1 | 2.4 (0.3–21.8) | |||
>25 unit/year | 0 | 0 cases |
Findings for total cohort.
Leukaemia
Similar to the MM findings, there was no association between occupational TCE exposure and risk of leukaemia (SRRE = 1.11, 95% CI: 0.93–1.32; P value for heterogeneity = 0.50) (Table 2). Furthermore, five of the individual studies reported associations ranging between 1.05 and 1.15 [23–26,29], while one study reported an inverse association (RR = 0.60) [22] and another reported a 2-fold positive association [28]. All CIs for these studies included 1.0. The study of Raaschou-Nielsen et al. [29] contributed 82 leukaemia cases, while the other studies each contributed 16 or fewer leukaemia cases or deaths. The study of Raaschou-Nielson et al. [29] accounted for 64% of the relative weight of the leukaemia meta-analysis, but its removal from the analysis did not markedly alter the overall result (SRRE = 1.05, 95% CI: 0.77–1.41; P value for heterogeneity = 0.40), nor did the removal of Hansen et al. [28], the study with the greatest magnitude of increased risk (SRRE = 1.08, 95% CI: 0.90–1.28). There was a lack of heterogeneity of findings across the remaining studies with the removal of Hansen et al. [28] (P value for heterogeneity = 0.690). Greenland et al. [25], the only case–control study included in our meta-analysis (leukaemia only), reported an odds ratio (OR) for TCE exposure among transformer manufacturing workers (OR = 1.10, 95% CI: 0.46–2.66). Removal of this study did not change the meta-analysis results (SRRE = 1.10, 95% CI: 0.90–1.35). Furthermore, the removal of any one of the individual studies in the MM or leukaemia meta-analyses did not alter the SRRE by >7%.
Anttila et al. [26] reported a non-significant positive association for the high category of U-TCA exposure (100+ μmol/l: SIR = 2.65, 95% CI: 0.72–6.78), based on four cases (Table 3). Among aerospace or aircraft workers, Morgan et al. [24] reported weak, non-significant associations for the high and low categories of TCE exposure, while Blair et al. [22] found no increasing risk of leukaemia among men or women, based on increasing categories of cumulative exposure to TCE (Table 3).
Discussion
The relative homogeneity of findings across studies of TCE-exposed workers for MM and leukaemia allowed for the application of meta-analysis modelling to more precisely estimate the association between occupational TCE exposures and risk of these rare malignancies. Other than the study of Raaschou-Nielsen et al. [29], the individual studies analysed were relatively small, with 16 or fewer TCE-exposed MM or leukaemia cases; thus, the aetiological interpretation of each individual study is limited by low statistical power to detect an association if one truly exists. By combining results from studies in a meta-analysis, we were able to produce relatively precise SRRE. Based on our meta-analyses, there is no epidemiological evidence to support a causal relationship between occupational TCE exposure and risk of MM or leukaemia (Table 2). The overall findings were robust, as they did not vary markedly with the addition and/or deletion of the study of Raaschou-Nielsen et al. [29], which contributed >49% of exposed cases in both meta-analyses. Findings from individual studies that analysed various categories of TCE exposure, such as cumulative duration of exposure or mean U-TCA concentration classifications, are not supportive of an aetiological association between TCE exposure and MM or leukaemia. These individual analyses, however, were limited by few or no exposed cases/deaths in the exposure subcategories (Table 3).
In a 2000 review of >80 published papers and letters, Wartenberg et al. [8] examined the cancer epidemiology of persons exposed to TCE. The authors combined results from individual cohort studies and reported no overall significant associations for MM [incidence: SIR = 1.50, 95% CI: 0.70–3.30; mortality: SMR = 1.00, 95% CI: 0.60–1.90 (note: after erratum corrected [32])] or leukaemia (incidence: SIR = 1.00, 95% CI: 0.50–2.10; mortality: SMR = 1.00, 95% CI: 0.70–1.30), based on the analysis of Tier I studies, which were considered the most informative for TCE characterization [8]. In our meta-analysis, we included data from two recent cohorts [28,29] and new data for MM from the study of Morgan et al. [24] that were not incorporated into the analysis of Wartenberg et al. [8]. In addition, Wartenberg et al. [8] found no statistically significant associations for MM or leukaemia, based on combined analysis of studies of dry cleaner and laundry workers, in which persons were exposed to various solvents including TCE. Recent cohort studies of dry cleaning and laundry workers have been inconsistent, with associations above and below 1.0 [30,33,34].
Other studies that evaluated worker populations that were potentially exposed to TCE, reported findings that are consistent with our results of no association between occupational TCE exposure and MM or leukaemia. These studies neither identified a TCE-exposed subpopulation nor provided results specific to TCE, therefore, were not included in our meta-analysis. In a cohort of >14 000 aircraft manufacturing company workers, Garabrant et al. [35] reported non-significant inverse associations for leukaemia and aleukaemia (SMR = 0.82, 95% CI: 0.47–1.34), and for neoplasms of ‘other’ lymphatic and haematopoietic tissue (SMR = 0.65, 95% CI: 0.21–1.52). There were no individual exposure measurements, but the authors reported that 37% of jobs entailed exposure to TCE. Ritz [36] reported no associations for leukaemia and aleukaemia, or for cancer of lymphatic tissue, a category that included MM, in a cohort of uranium-processing workers exposed to TCE, cutting fluids and kerosene. Chang et al. [37] reported non-significant inverse associations for cancer of the lymphatic and haematopoietic tissue among male (SIR = 0.73, 95% CI: 0.27–1.60) and female (SIR = 0.65, 95% CI: 0.37–1.05) workers in an electronics factory who were potentially exposed to TCE prior to 1968. Zhao et al. [38] reported no significant associations for combined categories of non-Hodgkin's lymphoma and leukaemia mortality (high exposure: RR = 1.30, 95% CI: 0.52–3.23) or incidence (high exposure: RR = 0.20, 95% CI: 0.03–1.46) among California aerospace workers exposed to TCE. Leukaemia was not analysed as a separate outcome; therefore, we did not include this study in our meta-analysis.
Interpretation of findings should be done with consideration of potential limitations in the studies included in our meta-analyses. The workforce cohorts from each study were presumably exposed to numerous and heterogeneous occupational exposures, which may have confounded individual study findings. Blair et al. [22], for example, reported elevated risks, albeit non-significant, for several chemicals including methylene chloride, freon, acetone and jet fuel. The study cohort in Raaschou-Nielsen et al. [29] was composed of a wide variety of TCE-using industries, including iron and metal, electronics, painting and printing. Despite these possible limitations, individual findings across studies were generally consistent, and none of the study-specific associations was statistically significant.
The evaluation of findings from individual epidemiological studies using meta-analysis techniques provided the opportunity to examine the relationship between occupational TCE exposure and the risk of developing two relatively rare haematological malignancies, MM and leukaemia. We were able to calculate a more precise summary RR estimate given the relative homogeneity of findings across individual studies. The results from our meta-analyses are not supportive of a causal role of occupational TCE exposure and risk of developing MM or leukaemia.
Conflicts of interest
The authors have consulted for a number of private and governmental clients on health issues related to occupational and environmental TCE exposure. This research was partially supported by the TCE Issues Group, a group of companies involved in TCE remediation.
References
Weiss NS. Cancer in relation to occupational exposure to trichloroethylene.
Feldman RG. Trichloroethylene. In: Occupational and Environmental Neurotoxicology. Philadelphia, PA: Lippincott-Raven,
Wong O. Carcinogenicity of trichloroethylene: an epidemiologic assessment.
Raaschou-Nielsen O, Hansen J, Thomsen BL et al. Exposure of Danish workers to trichloroethylene, 1947–1989.
Lynge E, Anttila A, Hemminki K. Organic solvents and cancer.
International Agency for Research on Cancer. Trichloroethylene. In: IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. World Health Organization,
US Department of Health and Human Services, Public Health Service. National Toxicology Program. Report on Carcinogens, Eleventh Edition,
Wartenberg D, Reyner D, Scott CS. Trichloroethylene and cancer: epidemiologic evidence.
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.
NTP. Carcinogenesis studies of trichloroethylene (without epichlorohydrin) (CAS No. 79-01-6) in F344/N rats and B6C3F1 mice (gavage studies).
NTP. Toxicology and carcinogenesis studies of trichloroethylene (CAS No. 79-01-6) in four strains of rats (ACI, August, Marshall, Osborne-Mendel) (gavage studies).
Rhomberg LR. Dose-response analyses of the carcinogenic effects of trichloroethylene in experimental animals.
Henschler D, Romen W, Elsasser HM, Reichert D, Eder E, Radwan Z. Carcinogenicity study of trichloroethylene by longterm inhalation in three animal species.
Henschler D, Elsasser H, Romen W, Eder E. Carcinogenicity study of trichloroethylene, with and without epoxide stabilizers, in mice.
McLaughlin JK, Blot WJ. A critical review of epidemiology studies of trichloroethylene and perchloroethylene and risk of renal-cell cancer.
Harth V, Bruning T, Bolt HM. Renal carcinogenicity of trichloroethylene: update, mode of action, and fundamentals for occupational standard setting.
Morgan JW, Cassady RE. Community cancer assessment in response to long-time exposure to perchlorate and trichloroethylene in drinking water.
Cohn P, Klotz J, Bove F, Berkowitz M, Fagliano J. Drinking water contamination and the incidence of Leukemia and non-Hodgkin's lymphoma.
Costas K, Knorr RS, Condon SK. A case-control study of childhood leukemia in Woburn, Massachusetts: the relationship between leukemia incidence and exposure to public drinking water.
Fagliano J, Berry M, Bove F, Burke T. Drinking water contamination and the incidence of leukemia: an ecologic study.
Vartiainen T, Pukkala E, Rienoja T, Strandman T, Kaksonen K. Population exposure to tri- and tetrachloroethylene and cancer risk: two cases of drinking water pollution.
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.
Boice JD Jr, Marano DE, Fryzek JP, Sadler CJ, McLaughlin JK. Mortality among aircraft manufacturing workers.
Morgan RW, Kelsh MA, Zhao K, Heringer S. Mortality of aerospace workers exposed to trichloroethylene.
Greenland S, Salvan A, Wegman DH, Hallock MF, Smith TJ. A case-control study of cancer mortality at a transformer-assembly facility.
Anttila A, Pukkala E, Sallmen M, Hernberg S, Hemminki K. Cancer incidence among Finnish workers exposed to halogenated hydrocarbons.
Axelson O, Selden A, Andersson K, Hogstedt C. Updated and expanded Swedish cohort study on trichloroethylene and cancer risk.
Hansen J, Raaschou-Nielsen O, Christensen JM et al. Cancer incidence among Danish workers exposed to trichloroethylene.
Raaschou-Nielsen O, Hansen J, McLaughlin JK et al. Cancer risk among workers at Danish companies using trichloroethylene: a cohort study.
Travier N, Gridley G, De Roos AJ, Plato N, Moradi T, Boffetta P. Cancer incidence of dry cleaning, laundry and ironing workers in Sweden.
Andersson T, Ahlbom A. Episheet Software: Spreadsheets for the Analysis of Epidemiologic Data, K. Rothman,
Blair A, Petralia SA, Stewart PA. Extended mortality follow-up of a cohort of dry cleaners.
Ji J, Hemminki K. Occurrences of leukemia subtypes by socioeconomic and occupational groups in Sweden.
Garabrant DH, Held J, Langholz B, Bernstein L. Mortality of aircraft manufacturing workers in southern California.
Ritz B. Cancer mortality among workers exposed to chemicals during uranium processing.
Chang YM, Tai CF, Yang SC et al. Cancer incidence among workers potentially exposed to chlorinated solvents in an electronics factory.