Abstract
Background:
Dry cleaning workers are commonly exposed to tetrachloroethylene, a suspected bladder carcinogen, and other organic solvents. The health risks associated with solvent exposures in this industry are unclear.
Methods:
We extended mortality follow-up of 5369 dry cleaning union members in St. Louis to further investigate solvent-related risks. We added 22 years of follow-up, through 2014, via linkage to the National Death Index. Using Cox proportional hazards modeling, we computed hazard ratios (HRs) and 95% confidence intervals (CIs) relating cause-specific mortality with levels of a solvent exposure index previously developed by an industrial hygienist based on workers’ job titles from union records. The models were fit adjusting for age, sex and decade of union enrollment, and assuming different exposure lags.
Results:
In internal analyses of estimated solvent exposure with a twenty-year lag, we observed exposure-response relationships for bladder cancer [HR 4.2, 95% CI 0.7 to 24.5 and 9.2, 1.1 to 76.7 for medium and high exposure respectively vs. no exposure; Ptrend=0.08] and kidney cancer (HR 4.1, 95% CI 0.7 to 22.5 and 24.4, 2.9 to 201.6; Ptrend=0.004). High exposure was also associated with heart disease (HR 1.6, 95% CI 1.1 to 2.2) and lymphatic/hematopoietic malignancies (HR 4.3, 95% CI 1.4 to 13.6).
Conclusions:
These findings are, to the best of our knowledge, the first cohort evidence relating solvent exposure levels among dry cleaners to elevated risks of selected cancers and heart disease. Additional studies employing solvent-specific exposure assessment are needed to clarify cancer risks associated with tetrachloroethylene.
Keywords: dry cleaning, tetrachloroethylene, mortality, bladder cancer, kidney cancer, cohort study
INTRODUCTION
The International Agency for Research on Cancer (IARC) has classified exposures related to employment in the dry cleaning industry as possibly carcinogenic to humans (Group 2B).1 The exposures of concern involve organic solvents used in the dry cleaning process. Over the years, the dry cleaning industry has used several different solvents.2, 3 Stoddard solvent dominated the industry until the 1930s, when chlorinated solvents such as carbon tetrachloride and trichloroethylene became popular. Both Stoddard solvent and carbon tetrachloride were largely replaced in the 1960s by tetrachloroethylene (also commonly referred to as perchloroethylene), which has become the predominant solvent used by retail dry cleaners in the United States. Tetrachloroethylene is a widespread environmental pollutant due to industrial releases,4 with particularly high concentrations in ambient indoor air of buildings containing dry cleaning facilities.5, 6
There is concern over the health effects of tetrachloroethylene, classified by IARC as a probable human carcinogen.7 Experimental studies have demonstrated excesses of liver tumors in mice, mononuclear cell leukemia in one strain of rats, and renal cell adenomas and adenocarcinomas in male rats.7 The epidemiologic evidence, however, is limited. Some epidemiologic studies among dry cleaners, but not others, have reported excesses for bladder cancer and, less frequently, cancers of the esophagus, NHL, cervix, and kidney,7 although most lacked assessments of workers’ solvent exposure levels.
To further investigate cancer associations with solvent exposure in the dry cleaning industry, we extended mortality follow-up within a cohort of 5369 dry cleaning union members. In particular, we utilized solvent exposure assessment metrics previously developed within this cohort to conduct internal comparisons of relative exposure level among workers.
METHODS
Details of the study have been described.8, 9 The cohort was assembled from historical dues records maintained by Local No 161 (St Louis) of the Laundry, Dry Cleaning, and Dye House Workers’ International Union. Members of Local 161 worked exclusively in dry cleaning establishments. Only union members admitted before 1978 were included in the cohort. Investigators abstracted the following information from dues records for union members admitted: name, Social Security Number, date of birth, year and age of entry into the union, number of dues-paying months by calendar year, race, sex, job titles (usually from around the time of union entry), and most recent firm where employed. When not available from dues records, race, sex, and date of birth were sought from other sources including driver’s license records, social security files, health care finance administration records, and credit bureaus.
In the initial follow-up of the cohort through 1978, vital status was determined using information from union records, the Social Security Administration, motor vehicle departments, credit bureaus, state bureaus of social services, and telephone and street directories.8 Death certificates were coded by an experienced nosologist. In a subsequent follow-up through 1992, vital status and cause of death were ascertained through linkage to the National Death Index (NDI).9 For this update, we extended follow-up until December 31, 2014 through linkage to the NDI and Social Security Administration Death Master File. Underlying and contributing causes of death, coded according to the International Classification of Diseases, Ninth Revision (ICD-9) system, were translated to the Eighth Revision (Adapted) Classification (ICDA-8) coding system to harmonize with previously collected mortality data from earlier cohort follow-up.
A solvent exposure score was developed by the study industrial hygienist (PAS) based on published monitoring studies of the dry cleaning industry10-14 and applied to job titles abstracted from union records. Mean tetrachloroethylene exposure levels reported in the published monitoring studies ranged from 25ppm to 280ppm. For context, the current Occupational Safety and Health Administration tetrachloroethylene personal exposure limit is an eight-hour time-weighted average of 100ppm.15 Monitoring data indicated that levels were highest for jobs performed at the washers and progressively decreased with distance from the washing machines. In our study, cleaners were assigned a score of 40 (high exposure) for an eight-hour time-weighted average, while persons working as pressers, sewers, or at the counter were given a score of seven (medium exposure). Cohort members employed at pick-up stations where no dry cleaning occurred had little or no exposure and were assigned a score of zero (although they would have had a higher exposure than the general population due to off-gassing from the cleaned garments). For most subjects, only one job title was recorded. Of the 978 subjects with a second job title, the exposure assessment category was identical for 956 of them. In addition to the workers’ assessed solvent exposure level, other analyzed exposure metrics included duration of union membership through 1978 (as a surrogate for exposure duration) and cumulative exposure (the product of exposure score and duration of union membership through 1978), categorized using tertiles.
Person-year accumulation began on date of entry into the union, or January 1, 1948, whichever came later, and ended at date of death, age 90, or end of follow-up, whichever came first. For external comparisons, we calculated standardized mortality ratios (SMRs) and 95% confidence intervals (CI), with US general population mortality rates by age, sex, race and calendar period as a reference, both overall and stratified by sex, race, and year of enrollment in the union (<1960, ≥1960). For internal comparisons, we computed hazard ratios (HRs) and 95% CIs relating different levels of the exposure metrics to mortality by fitting Cox proportional hazards models using attained age as the time scale, adjusting for sex and decade of union enrollment, and applying exposure lags of 0, 10 and 20 years. We evaluated the proportional hazards assumption through inspection of Kaplan Meier curves and testing the significance of interaction terms relating exposure and attained age. Tests of trend were computed by modeling the solvent exposure level scores (0, 7, 40), intra-category medians of union membership duration, or cumulative exposure as continuous variables. As a sensitivity analysis, we also performed trend tests using alternative integer values (1, 2, 3) for the exposure levels. Internal analyses were also restricted to subjects entering the cohort during or after 1960.
RESULTS
Selected characteristics of the study cohort are summarized in Table 1. Workers joined the union in the mid-1950s on average with a mean age at entry of 36.7 years, and remained as members through 1978 an average of 6.6 years. Three-quarters of workers were female and 55% were white. With this latest follow-up period, the total person-years of follow-up increased to 178,459 (no/low exposure: 84,428; medium exposure: 69,891; high exposure: 8,641; unknown exposure: 15,498), and the number of deaths increased from 2,278 to 3,543. A plot of the cumulative distribution of accumulated person-time across calendar years of follow-up is provided in Supplementary Figure 1.
Table 1.
Characteristics of dry cleaners in cohort (n= 5,369)
| Characteristic | Mean (SD) |
|---|---|
| Year of birth | 1918 (15.4) |
| Age at entry | 36.7 (12.1) |
| Year at entry | 1956 (8.5) |
| Years of union membership | 6.6 (6.0) |
| Sex | n (%) |
| Male | 1320 (25) |
| Female | 4049 (75) |
| Race | |
| White | 2973 (55) |
| Black/other | 2396 (45) |
| Estimated solvent exposure | |
| No/low | 2526 (47) |
| Medium | 2079 (39) |
| High | 278 (5) |
| Unknown | 486 (9) |
Abbreviations: SD, standard deviation.
SMR results for selected causes of death are presented in Table 2 (complete results in Supplemental Table 1). Mortality excesses among workers were observed for cancer overall (SMR 1.2, 95% CI 1.1 to 1.3); cancers of the esophagus (SMR 2.0, 95% CI 1.4 to 2.9), larynx (SMR 2.0, 95% CI 1.0 to 3.8), lung (SMR 1.5, 95% CI 1.3 to 1.7), and cervix (SMR 1.6, 95% CI 1.0 to 2.2); heart disease (SMR 1.1, 95% CI 1.0 to 1.1); and emphysema (SMR 1.8, 95% CI 1.2 to 2.5). The SMR results for bladder cancer and kidney cancer were null overall (SMR 1.1 for both), although suggestive excesses were observed among workers who joined the union in 1960 or later (SMR 1.5, 95% CI 0.4 to 3.7 and 1.8, 0.7 to 3.9, respectively). The SMR findings did not notably differ by sex or race (results not shown).
Table 2.
Mortality among dry cleaners for selected causes of death by compared to the US population
| Entire cohort | Year Joined the Union | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| <1960 | ≥1960 | |||||||||||
| Cause of death | O | E | SMR | (95%CI) | O | E | SMR | (95%CI) | O | E | SMR | (95%CI) |
| All causes | 3543 | 3403 | 1.0 | (1.0, 1.1) | 2674 | 2632 | 1.0 | (1.0, 1.1) | 869 | 770 | 1.1 | (1.1, 1.2) |
| All cancers | 879 | 741 | 1.2 | (1.1, 1.3) | 636 | 549 | 1.2 | (1.1, 1.3) | 243 | 192 | 1.3 | (1.1, 1.4) |
| Buccal cavity and pharynx | 13 | 12 | 1.1 | (0.6, 1.8) | 10 | 9 | 1.1 | (0.5, 2.0) | 3 | 3 | 1.1 | (0.2, 3.1) |
| Esophageal | 32 | 16 | 2.0 | (1.4, 2.9) | 26 | 12 | 2.1 | (1.4, 3.1) | 6 | 4 | 1.7 | (0.6, 3.7) |
| Stomach | 23 | 30 | 0.8 | (0.5, 1.2) | 17 | 24 | 0.7 | (0.4, 1.1) | 6 | 5 | 1.1 | (0.4, 2.4) |
| Colon | 88 | 76 | 1.2 | (0.9, 1.4) | 65 | 58 | 1.1 | (0.9, 1.4) | 23 | 18 | 1.3 | (0.8, 1.9) |
| Rectum | 20 | 15 | 1.3 | (0.8, 2.1) | 17 | 12 | 1.4 | (0.8, 2.3) | 3 | 3 | 1.0 | (0.2, 2.9) |
| Liver | 16 | 20 | 0.8 | (0.5, 1.3) | 13 | 15 | 0.9 | (0.5, 1.5) | 3 | 5 | 0.6 | (0.1, 1.8) |
| Pancreas | 49 | 42 | 1.2 | (0.9, 1.5) | 31 | 31 | 1.0 | (0.7, 1.4) | 18 | 11 | 1.6 | (1.0, 2.6) |
| Larynx | 10 | 5 | 2.0 | (1.0, 3.8) | 7 | 4 | 1.9 | (0.7, 3.8) | 3 | 1 | 2.7 | (0.6, 8.0) |
| Lung | 232 | 153 | 1.5 | (1.3, 1.7) | 155 | 109 | 1.4 | (1.2, 1.7) | 77 | 45 | 1.7 | (1.4, 2.1) |
| Skin | 5 | 7 | 0.7 | (0.2, 1.6) | 5 | 5 | 1.0 | (0.3, 2.2) | 0 | 2 | 0.0 | (0.0, 1.8) |
| Breast | 88 | 90 | 1.0 | (0.8, 1.2) | 60 | 62 | 1.0 | (0.7, 1.2) | 28 | 28 | 1.0 | (0.7, 1.5) |
| Cervix uteri | 30 | 19 | 1.6 | (1.0, 2.2) | 24 | 15 | 1.6 | (1.0, 2.4) | 6 | 4 | 1.4 | (0.5, 3.1) |
| Corpus uteri | 22 | 19 | 1.1 | (0.7, 1.7) | 17 | 14 | 1.2 | (0.7, 1.9) | 5 | 5 | 1.0 | (0.3, 2.4) |
| Prostate | 28 | 30 | 1.0 | (0.6, 1.4) | 23 | 26 | 0.9 | (0.6, 1.4) | 5 | 4 | 1.2 | (0.4, 2.9) |
| Bladder | 15 | 14 | 1.1 | (0.6, 1.8) | 11 | 11 | 1.0 | (0.5, 1.8) | 4 | 3 | 1.5 | (0.4, 3.7) |
| Kidney | 14 | 13 | 1.1 | (0.6, 1.9) | 8 | 9 | 0.9 | (0.4, 1.7) | 6 | 3 | 1.8 | (0.7, 3.9) |
| Brain | 7 | 12 | 0.6 | (0.2, 1.2) | 5 | 8 | 0.6 | (0.2, 1.4) | 2 | 3 | 0.6 | (0.1, 2.1) |
| Lymphatic and hematopoietic | 64 | 64 | 1.0 | (0.8, 1.3) | 55 | 47 | 1.2 | (0.9, 1.5) | 9 | 17 | 0.5 | (0.2, 1.0) |
| Non-Hodgkin's lymphoma | 21 | 21 | 1.0 | (0.6, 1.5) | 18 | 15 | 1.2 | (0.7, 1.9) | 3 | 6 | 0.5 | (0.1, 1.5) |
| Hodgkin's disease | 5 | 3 | 1.8 | (0.6, 4.1) | 4 | 2 | 1.8 | (0.5, 4.5) | 1 | 1 | 1.7 | (0.0, 9.6) |
| Multiple myeloma | 14 | 15 | 0.9 | (0.5, 1.5) | 13 | 11 | 1.2 | (0.6, 2.0) | 1 | 4 | 0.3 | (0.0, 1.4) |
| Leukemia | 20 | 23 | 0.9 | (0.5, 1.3) | 16 | 17 | 0.9 | (0.5, 1.5) | 4 | 6 | 0.7 | (0.2, 1.7) |
| Heart disease | 904 | 859 | 1.1 | (1.0, 1.1) | 694 | 694 | 1.0 | (0.9, 1.1) | 210 | 165 | 1.3 | (1.1, 1.5) |
| Emphysema | 33 | 18 | 1.8 | (1.2, 2.5) | 24 | 14 | 1.7 | (1.1, 2.5) | 9 | 4 | 2.1 | (0.9, 3.9) |
Abbreviations: O, observed number of deaths; E, expected numbers of deaths; SMR, standardized mortality ratio; CI, confidence interval.
The results for selected causes of death from internal comparisons of the solvent exposure score are shown in Table 3 (complete results in Supplemental Table 2). In 20-year lagged analyses, exposure-response relationships were observed for bladder cancer (medium: HR 4.2, 95% CI 0.7 to 24.5; high: HR 9.2, 95% CI 1.1 to 76.7; Ptrend = 0.08) and kidney cancer (4.1, 0.7 to 22.5; 24.4, 2.9 to 201.6; Ptrend = 0.004). High exposure was also associated with elevated mortality from heart disease (HR 1.6, 95% CI 1.1 to 2.2; Ptrend = 0.01) and lymphatic/hematopoietic malignancies (4.3, 1.4 to 13.6; Ptrend = 0.01). The alternative trend tests were statistically significant for cancers of the bladder and kidney (Ptrend = 0.04 and 0.006, respectively assuming a 20-year lag) but not for heart disease or the blood malignancies (0.10 and 0.18, respectively). Of the other endpoints elevated in SMR analyses, most were null in the internal comparisons, except a weak association for lung cancer (HR 1.4, 95% CI 0.8 to 2.6; Ptrend = 0.28) and an inverse association for esophageal cancer (HR 0.2, 95% CI 0.0 to 1.8; Ptrend = 0.15). When 20-year lagged analyses were restricted to subjects entering the cohort during or after 1960, the association with high exposure and heart disease persisted, but was imprecise (HR 1.4, 95%CI 0.7 to 2.8; Ptrend=0.30) and there were too few kidney or bladder cancer deaths to generate stable HR estimates (results not shown).
Table 3.
Risk of death from selected causes by estimated solvent exposure, Missouri Dry Cleaners cohort
| Cause of death |
Solvent
exposure |
No Lag | 10-year lag | 20-year lag | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| N events | HR a | (95%CI) | N events | HR a | (95%CI) | N events | HR a | (95%CI) | ||
| Anyb | No/low | 1525 | 1.0 | 1433 | 1.0 | 1200 | 1.0 | |||
| Medium | 1508 | 1.1 | (1.0, 1.2) | 1408 | 1.1 | (1.0, 1.2) | 1196 | 1.1 | (1.0, 1.2) | |
| High | 217 | 1.2 | (1.0, 1.4) | 194 | 1.1 | (1.0, 1.3) | 147 | 1.1 | (0.9, 1.3) | |
| Ptrend | 0.05 | 0.09 | 0.83 | |||||||
| All cancers | No/low | 351 | 1.0 | 321 | 1.0 | 264 | 1.0 | |||
| Medium | 384 | 1.2 | (1.0, 1.4) | 360 | 1.2 | (1.0, 1.4) | 307 | 1.1 | (1.0, 1.4) | |
| High | 54 | 1.1 | (0.8, 1.6) | 52 | 1.1 | (0.8, 1.6) | 47 | 1.2 | (0.9, 1.8) | |
| Ptrend | 0.43 | 0.38 | 0.23 | |||||||
| Esophageal cancer | No/low | 9 | 1.0 | 9 | 1.0 | 9 | 1.0 | |||
| Medium | 17 | 0.8 | (0.3, 2.0) | 16 | 0.7 | (0.3, 1.7) | 14 | 0.7 | (0.3, 1.7) | |
| High | 2 | 0.4 | (0.1, 2.0) | 2 | 0.4 | (0.1, 2.0) | 1 | 0.2 | (0.0, 1.8) | |
| Ptrend | 0.26 | 0.30 | 0.15 | |||||||
| Lung cancer | No/low | 82 | 1.0 | 81 | 1.0 | 72 | 1.0 | |||
| Medium | 110 | 1.4 | (1.0, 2.0) | 106 | 1.4 | (1.0, 1.9) | 95 | 1.4 | (1.0, 2.0) | |
| High | 15 | 1.1 | (0.6, 2.0) | 15 | 1.1 | (0.6, 2.1) | 15 | 1.4 | (0.8, 2.6) | |
| Ptrend | 0.14 | 0.64 | 0.28 | |||||||
| Bladder cancer | No/low | 6 | 1.0 | 5 | 1.0 | 2 | 1.0 | |||
| Medium | 6 | 1.7 | (0.5, 5.8) | 6 | 2.3 | (0.6, 8.3) | 5 | 4.2 | (0.7, 24.5) | |
| High | 3 | 3.2 | (0.6, 17.1) | 3 | 5.2 | (0.9, 30.4) | 3 | 9.2 | (1.1, 76.7) | |
| Ptrend | 0.19 | 0.09 | 0.08 | |||||||
| Kidney cancer | No/low | 3 | 1.0 | 2 | 1.0 | 2 | 1.0 | |||
| Medium | 6 | 2.4 | (0.5, 10.6) | 6 | 3.4 | (0.6, 18.6) | 6 | 4.1 | (0.7, 22.5) | |
| High | 3 | 13.2 | (1.9, 90.8) | 3 | 17.8 | (2.2, 143.7) | 3 | 24.4 | (2.9, 201.6) | |
| Ptrend | 0.004 | 0.004 | 0.004 | |||||||
| Lymphatic/hematopoietic malignancies | No/low | 30 | 1.0 | 27 | 1.0 | 21 | 1.0 | |||
| Medium | 21 | 0.8 | (0.4, 1.5) | 17 | 0.7 | (0.4, 1.4) | 17 | 0.9 | (0.5, 1.8) | |
| High | 7 | 2.8 | (1.0, 7.5) | 7 | 3.2 | (1.1, 9.1) | 6 | 4.3 | (1.4, 13.6) | |
| Ptrend | 0.02 | 0.02 | 0.01 | |||||||
| Non-Hodgkin lymphoma | No/low | 9 | 1.0 | 7 | 1.0 | 5 | 1.0 | |||
| Medium | 11 | 1.7 | (0.6, 4.4) | 8 | 1.5 | (0.5, 4.6) | 8 | 1.9 | (0.6, 6.6) | |
| High | 1 | 1.7 | (0.2, 16.6) | 1 | 1.6 | (0.2, 16.4) | 1 | 2.1 | (0.2, 23.6) | |
| Ptrend | 0.57 | 0.17 | 0.55 | |||||||
| Cervical cancer | No/low | 13 | 1.0 | 9 | 1.0 | 5 | 1.0 | |||
| Medium | 13 | 1.2 | (0.5, 2.7) | 8 | 1.0 | (0.4, 2.8) | 5 | 1.2 | (0.3, 4.5) | |
| High | 0 | 0 | 0 | |||||||
| Ptrend | 0.91 | 0.78 | 0.99 | |||||||
| Heart disease | No/low | 412 | 1.0 | 393 | 1.0 | 329 | 1.0 | |||
| Medium | 357 | 1.0 | (0.9, 1.2) | 340 | 1.0 | (0.9, 1.2) | 289 | 1.0 | (0.9, 1.2) | |
| High | 66 | 1.5 | (1.2, 2.1) | 61 | 1.6 | (1.2, 2.1) | 46 | 1.6 | (1.1, 2.2) | |
| Ptrend | 0.005 | 0.006 | 0.01 | |||||||
| Emphysema | No/low | 17 | 1.0 | 16 | 1.0 | 12 | 1.0 | |||
| Medium | 11 | 0.7 | (0.3, 1.6) | 11 | 0.7 | (0.3, 1.6) | 10 | 0.9 | (0.3, 2.3) | |
| High | 2 | 0.4 | (0.1, 2.0) | 2 | 0.4 | (0.1, 2.0) | 2 | 0.6 | (0.1, 3.2) | |
| Ptrend | 0.28 | 0.30 | 0.59 | |||||||
Abbreviations: HR, hazard ratio; CI, confidence interval. aHRs calculated from Cox proportional hazards models using age as the underlying time-scale adjusted for race, sex, and decade of enrollment.
Interaction terms with exposure and time tested the proportional hazards assumption. This assumption was violated only for the analysis of overall mortality (p=0.004); however, results adjusted for the interaction between exposure and time were similar to the results presented.
Analyses of cumulative solvent exposure found associations with cancers of the bladder and kidney similar to the associations by exposure level (Supplemental Table 3) but no associations with heart disease or lymphatic/hematopoietic malignancies. Analyses of union membership duration were null (results not shown).
DISCUSSION
A strength of this study that is unique among cohort studies of dry cleaners is the availability of specific job titles from union records, which enabled the development of semi-quantitative estimates of workers’ solvent exposure levels. In earlier analyses of this cohort there was an insufficient number of deaths to support formal comparisons of mortality rates across workers’ relative solvent exposure levels. With this 22-year update we accrued enough deaths to conduct such internal analyses. From these analyses, we observed that solvent exposure levels were associated with increases in mortality due to heart disease and cancers of the bladder, kidney and lymphatic/hematopoietic system. These findings are, to the best of our knowledge, the first cohort evidence directly relating solvent exposure levels among dry cleaners to excess mortality for these endpoints. As such, they offer new evidence supporting workplace solvent exposures as a possible explanation for reported excesses of these cancers among dry cleaners.7, 16 Dry cleaning employment was associated with elevated bladder cancer mortality in a meta-analysis of seven cohort studies,16 while findings for cancers of the kidney and lymphatic system have been inconsistent. The only other dry cleaning cohort to have investigated non-cancer mortality also observed an elevated SMR (1.2, 95% CI 1.0 to 1.5) for ischemic heart disease among workers exposed exclusively to tetrachloroethylene.17
The lack of information on subjects’ smoking habits is a limitation of our study. Given that data from other sources suggest that dry cleaners smoked more than workers in other occupations,18, 19 it is plausible that the elevated SMRs for emphysema and cancers of the esophagus, larynx, lung and cervix, all tobacco-related endpoints, may at least partly reflect different smoking habits within the cohort compared to the general population. We note, however, that these smoking-related endpoints were not associated with solvent exposures in our internal-comparison analyses, thus arguing against confounding from smoking as an explanation for the observed exposure-response patterns for cancers of the bladder, kidney and lymphatic and hematopoietic system. Additionally, suggestive associations with dry cleaning for kidney and bladder cancer have been reported in case-control studies that did adjust for smoking.16, 20
Other limitations in the information available from union records affected our exposure assessment. We did not have information on workers’ potential dry cleaning industry employment prior to their membership in the union. Further, as we only had union records available through 1978, subjects who worked in dry cleaning establishments beyond this year would have truncated estimates of exposure duration and cumulative exposure, which may have introduced bias into results for these metrics. The absence of direct exposure monitoring in subjects’ workplaces and the lack of a specific exposure scale for tetrachloroethylene, or other individual dry cleaning solvents are additional limitations of our study. We note that the SMRs for bladder cancer, kidney cancer and heart disease, but not lymphatic/hematopoietic malignancies, were elevated among workers joining the union during or after 1960, when tetrachloroethylene became the overwhelming solvent of choice in the industry,2, 3 and null among those joining prior to 1960. These results are compatible with a tetrachloroethylene-specific effect for these outcomes. However, we were unable to directly assess when members of this cohort were first exposed to tetrachloroethylene. Two case-control studies employing detailed exposure assessment methods to investigate occupational tetrachloroethylene exposure have also reported associations with high exposure for cancers of the bladder and kidney.21, 22
Our study was also limited by the relatively small sample size. The number of kidney and bladder cancer deaths was too small to assess solvent exposure in internal analyses restricted to subjects who entered the cohort after 1960. We also observed lower rates of several causes of death (e.g. esophageal cancer) among the high versus the low exposure group that were based on very few of deaths overall and particularly among highly exposed subjects (n<3). As these inverse associations have wide confidence limits, they may have arisen due to chance.
In conclusion, our findings from this updated analysis of a cohort of dry cleaning workers suggest that workplace exposure to tetrachloroethylene in the dry cleaning industry is associated with an increased risk of bladder cancer, kidney cancer, lymphatic/hematopoietic malignancies, and heart disease. Further research is needed to clarify the cancer and non-cancer risks associated with occupational exposure to tetrachloroethylene, as well as the possible health risks associated with residential proximity to dry cleaning facilities 5, 6 and the presence of dry cleaned clothing in the home.23, 24
Supplementary Material
Funding:
This research was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Division of Cancer Epidemiology, and Genetics.
Footnotes
P.S. is employed by Stewart Exposure Assessments, LLC (Arlington, VA, USA). The remaining authors declare they have no actual or potential competing financial interests.
Data are available upon request.
REFERENCES
- 1.IARC. Dry Cleaning, Some Chlorinated Solvents and Other Industrial Chemicals. IARC Monogr Eval Carcinog Risks Hum; 1995;63. [PMC free article] [PubMed] [Google Scholar]
- 2.Doherty RE. A History of the Production and Use of Carbon Tetrachloride, Tetrachloroethylene, Trichloroethylene and 1,1,1-Trichloroethane in the United States: Part 1--Historical Background; Carbon Tetrachloride and Tetrachloroethylene. Environmental Forensics 2000;1:69–81. [Google Scholar]
- 3.Gold LS, De Roos AJ, Waters M, Stewart P. Systematic literature review of uses and levels of occupational exposure to tetrachloroethylene. J Occup Environ Hyg 2008;5:807–39. [DOI] [PubMed] [Google Scholar]
- 4.NTP. National Toxicology Program: Report on Carcinogens, Fourteenth Edition Research Triangle Park: U.S. Department of Health and Human Services, 2016. [Google Scholar]
- 5.McDermott MJ, Mazor KA, Shost SJ, et al. Tetrachloroethylene (PCE, Perc) levels in residential dry cleaner buildings in diverse communities in New York City. Environ Health Perspect 2005;113:1336–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Roda C, Kousignian I, Ramond A, Momas I. Indoor tetrachloroethylene levels and determinants in Paris dwellings. Environ Res 2013;120:1–6. [DOI] [PubMed] [Google Scholar]
- 7.IARC. Trichloroethylene, Tetrachloroethylene, and Some Other Chlorinated Agents. IARC Monogr Eval Carcinog Risks Hum; 2014;106. [PMC free article] [PubMed] [Google Scholar]
- 8.Blair A, Stewart PA, Tolbert PE, et al. Cancer and other causes of death among a cohort of dry cleaners. Br J Ind Med 1990;47:162–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Blair A, Petralia SA, Stewart PA. Extended mortality follow-up of a cohort of dry cleaners. Ann Epidemiol 2003;13:50–6. [DOI] [PubMed] [Google Scholar]
- 10.Materna BL. Occupational exposure to perchloroethylene in the dry cleaning industry. Am Ind Hyg Assoc J 1985;46:268–73. [DOI] [PubMed] [Google Scholar]
- 11.NIOSH. Criteria for a Recommended Standard: Occupational Exposure to Tetrachloroethylene (Perchloroethylene) Cincinnati, OH: HEW publ No (NIOSH) 76-185., 1976. [Google Scholar]
- 12.NIOSH. Engineering control technology assessment of the dry cleaning industry, vol. DHHS (NIOSH) publ No 80-136; Cincinnati, OH, 1980. [Google Scholar]
- 13.Ludwig HR, Meister MV, Roberts DR, Cox C. Worker exposure to perchloroethylene in the commercial dry cleaning industry. Am Ind Hyg Assoc J 1983;44:600–5. [DOI] [PubMed] [Google Scholar]
- 14.Ludwig HR. Occupational exposure to perchloroethylene in the dry cleaning industry. In: NIOSH, ed. Internal Report, DHHS, 1981. [Google Scholar]
- 15.Administration OSaH. Reducing Worker Exposure to Perchloroethylene (PERC) in Dry Cleaning In: Labor USDo, ed., 2005. [Google Scholar]
- 16.Vlaanderen J, Straif K, Ruder A, et al. Tetrachloroethylene exposure and bladder cancer risk: a meta-analysis of dry-cleaning-worker studies. Environ Health Perspect 2014;122:661–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Calvert GM, Ruder AM, Petersen MR. Mortality and end-stage renal disease incidence among dry cleaning workers. Occup Environ Med 2011;68:709–16. [DOI] [PubMed] [Google Scholar]
- 18.Sterling TD, Weinkam JJ. Smoking characteristics by type of employment. J Occup Med 1976;18:743–54. [DOI] [PubMed] [Google Scholar]
- 19.Nelson DE, Emont SL, Brackbill RM, et al. Cigarette smoking prevalence by occupation in the United States. A comparison between 1978 to 1980 and 1987 to 1990. J Occup Med 1994;36:516–25. [PubMed] [Google Scholar]
- 20.Karami S, Colt JS, Schwartz K, et al. A case-control study of occupation/industry and renal cell carcinoma risk. BMC Cancer 2012;12:344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pesch B, Haerting J, Ranft U, et al. Occupational risk factors for renal cell carcinoma: agent-specific results from a case-control study in Germany. MURC Study Group. Multicenter urothelial and renal cancer study. Int J Epidemiol 2000;29:1014–24. [DOI] [PubMed] [Google Scholar]
- 22.Purdue MP, Stewart PA, Friesen MC, et al. Occupational exposure to chlorinated solvents and kidney cancer: a case-control study. Occup Environ Med 2017;74:268–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sherlach KS, Gorka AP, Dantzler A, Roepe PD. Quantification of perchloroethylene residues in dry-cleaned fabrics. Environmental Toxicology and Chemistry 2011;30:2481–87. [DOI] [PubMed] [Google Scholar]
- 24.Thomas KW, Pellizzari ED, Perritt RL, Nelson WC. Effect of dry-cleaned clothes on tetrachloroethylene levels in indoor air, personal air, and breath for residents of several New Jersey homes. J Expo Anal Environ Epidemiol 1991;1:475–90. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.