Article Text
Abstract
Background Ambient particulate air pollution has been linked to cardiovascular disease. Occupational particle exposure levels may be several times higher than ambient levels but has been less studied.
Objectives The authors investigated the association between occupational exposure to particles and the incidence of ischaemic heart disease (IHD).
Methods The cohort included all manual workers in the Swedish national census of 1980 with information on demographic data and occupation. Information on hospital admissions for acute myocardial infarction or other IHDs and cause of death were obtained from nation-wide registers. A job-exposure matrix for exposure to small (<1 μm) and large (>1 μm) particles was developed. HRs were calculated with Cox regression with adjustment for sex, age, socioeconomic group and urban/rural residential area.
Results Exposure to small particles was associated with an increased HR for acute myocardial infarction of 1.12 (95% CI 1.09 to 1.15), and HR for exposure to large particles was 1.14 (95% CI 1.10 to 1.18). The association was somewhat stronger for workers exposed to small particles for more than 5 years, 1.21 (95% CI 1.11 to 1.31), but no trend with exposure intensity was found. The risk associated with exposure to small particles was higher among women than among men, 1.30 (95% CI 1.12 to 1.51) and 1.10 (95% CI 1.07 to 1.14), respectively. Findings were essentially similar for other IHDs.
Conclusions This explorative study gives some support to the hypothesis that occupational exposure to particles increases the risk of acute myocardial infarction and other IHD. The findings must be interpreted cautiously due to lack of smoking data.
- Epidemiology
- cardiovascular disease
- retrospective exposure assessment
- particulates
- public health
- hygiene/occupational hygiene
- cardiovascular
- allergy
- biological monitoring
- training and education
- toxicokinetics
- exposure monitoring
- exposure assessment
- solvents
- pesticides
- painters
- noise
- developing countries
- respiratory
- renal
- occupational asthma
- cancer
- wood dust
- asbestos
- occupational health practice
- risk assessment
- preventive medicine
- man-made mineral fibres
- driving
- diesel fumes
Statistics from Altmetric.com
- Epidemiology
- cardiovascular disease
- retrospective exposure assessment
- particulates
- public health
- hygiene/occupational hygiene
- cardiovascular
- allergy
- biological monitoring
- training and education
- toxicokinetics
- exposure monitoring
- exposure assessment
- solvents
- pesticides
- painters
- noise
- developing countries
- respiratory
- renal
- occupational asthma
- cancer
- wood dust
- asbestos
- occupational health practice
- risk assessment
- preventive medicine
- man-made mineral fibres
- driving
- diesel fumes
What this paper adds
Ambient particulate air pollution has been linked to cardiovascular disease, but results from previous studies on occupational particle exposure are inconclusive.
This study shows an increased risk of IHD among workers occupationally exposed to particles. The risk was somewhat higher in women than in men and for exposure to particles less than 1 μm in size than to particles larger than 1 μm for workers exposed in two consecutive censuses, that is, more than 5 years. The results indicate that women might be more susceptible.
The results underline the need to explore exposure to different types of particles and the occurrence of IHD, as well as differences in sensitivity among men and women.
Introduction
In 2008, 21 500 cases of first-time acute myocardial infarction were diagnosed among men and 15 700 cases were diagnosed among women in Sweden. There were 25 000 deaths from acute myocardial infarction and other ischaemic heart diseases (IHDs) among men and 19 500 among women.1 A growing number of epidemiological studies indicate that exposure to particulate matter in ambient air is a risk factor for IHD.2–4 In the mid-1990s, inflammation and altered coagulation was hypothesized as a link between particulate air pollutants and IHD.5 ,6 Today, the epidemiological evidence has grown stronger and there is strong mechanistic evidence from animal and human studies supporting a systemic inflammatory response as the pathway.7 Certain occupational groups have a high exposure to particulate air pollution, for example, silica dust, welding fumes and combustion particles. Elevated risks of acute myocardial infarction have been reported in some particle-exposed occupational groups, for example, chimney sweeps,8 construction workers,9 welders,10 ,11 miners,12 livestock workers13 and cooks.14 A population-based case–referent study, adjusted for smoking, showed an increased risk for IHD after occupational exposure to combustion particles.15
Although the available evidence suggests negative health effect on the cardiovascular system from occupational exposure to particles, little is known on the effect of particles of various sizes and from various sources. Small particles, with a particle size below 0.1 μm, due to their larger surface area possess a greater potential for interaction with biological targets and consequently may result in more significant impact on health.16
This study, the PARrticles and Cardio- and Cerebrovascular diseases (PARCC) study, was designed as a population-based cohort study applying a job-exposure matrix (JEM) to link occupational particle exposure to job titles. No smoking data were available, but by restriction to manual workers, confounding due to socioeconomic differences was reduced. In this explorative study, we have chosen to compare highly exposed to unexposed workers. This is possible due to the large study population resulting in a high statistical power. The overall aim was to investigate the association between occupational exposure to particles from various sources and of various sizes with the incidence of cardiovascular diseases. This presentation focuses on the influence of particle size on the risk of acute myocardial infarction and other IHD.
Methods
Study base
The cohort comprises all manual workers in the Swedish National Census in 1980, who were alive on 1 January 1987, in total 984 040 men and 741 631 women. No smoking data were available, and in order to limit potential confounding from smoking and other lifestyle factors as well as problems due to misclassification of exposure, the cohort was restricted to manual workers. This group comprises four subgroups: skilled and unskilled manual workers in the production or service sector, respectively. White-collar workers, professionals, self-employed or farmers were not included in the cohort.
First-time events of acute myocardial infarction (International Classification of Diseases, ICD9 code 410 and ICD10 code I21) or other IHDs (ICD9 codes 411-414 and ICD10 codes I20, I22-I25) during 1987–2005 were identified through linkage to the Hospital Discharge Register and the National Cause of Death Register.
The study was approved by the regional ethics committee in Stockholm, Sweden.
Exposure assessment
Information on occupation and socioeconomic status was obtained from the population censuses in 1980, 1985 and 1990. The censuses provide occupational codes corresponding to the Nordic version of the International Standard Classifications of Occupations.17 A job-exposure matrix (PARCC-JEM) was developed by experienced occupational hygienists (PW, NP and GN) by combining relevant occupational particle exposure information from a Swedish JEM developed for the Nordic Occupational Cancer Study18 and an Airway Irritant-JEM.19 The estimated levels in the PARCC-JEM were based on exposure measurements from Sweden when available and otherwise from other Nordic countries. Exposure to cooking fumes was added to the PARCC-JEM since it was not included in any of the two other JEMs. The PARCC-JEM includes 20 groups of particle exposure. It is generic and time specific, including two time periods: 1975–1984 and 1985–1994.
Each cell of the JEM shows the proportion of exposed workers and the exposure level (air concentration) for each particle group. Exposure levels were assessed on a semi-quantitative scale with four categories, expressed as proportions of the Swedish occupational exposure limits (OELs) 200520: level 0<1/30 (no exposure), level 1=1/30–1/10 (low exposure), level 2=1/10–1/3 (medium exposure) and level 3>1/3 (high exposure). To achieve as great contrast as possible between exposed and unexposed workers, the main analyses were based on occupations with an exposure prevalence of at least 80% and medium or high exposure intensity. Additional analyses were performed to investigate the risk in occupations with a low, medium or high exposure intensity and prevalence ≥80%. Occupations with a prevalence >0 but <80% were not included in the analyses. To assess the effect of exposure duration, we investigated subjects with exposure to particles in at least two consecutive censuses, that is, subjects who had the same occupational code, exposure status and socioeconomic group in at least two consecutive censuses in 1980, 1985 and 1990. The corresponding restriction was applied to the unexposed subjects.
To investigate the influence of particle size, the particle exposures in the JEM were classified as small particles, <1 μm, or large particle size, >1 μm (table 1). The cut-off at 1 μm was selected in order to differentiate between particles formed during combustion and particles generated by mechanical processes.
The PARCC-JEM included 104 occupations with particle exposure, but only 31 occupations fulfilled our criteria for exposed occupations. In 50 occupations, no exposure assessment was feasible, mostly due to undefined working tasks or an inhomogeneous group of tasks with different exposures. Unexposed occupations were defined as occupations without exposure to any of the particles in the JEM, in total 128 occupations. In census 1980, workers in four occupations were exposed to small particles and 29 to large particles. Thus, 379 351 men and 355 651 women were included in the analyses for acute myocardial infarction and 379 354 men and 355 641 women in analyses for other IHDs.
Statistical analysis
We calculated person-years from 1 January 1987 until first episode of acute myocardial infarction, death, emigration, or 31 December 2005, whichever occurred first. The same procedure was performed in analyses of other IHD. In case of immigration of a formerly emigrated person in the study base, the person contributed with person-years from the immigration date until any of the events mentioned above occurred. Exposure status was determined from the 1980 census and was updated according to the 1985 and 1990 censuses if exposure status had changed. The association between particle exposure and the outcome was estimated through Cox proportional hazards modelling. HRs are presented with 95% CIs. All risk estimates were adjusted for age (continuous variable), sex, socioeconomic group (using four indicator variables: skilled/unskilled manual workers in production or service sector, respectively) and residential area (categorized into five classes based on the number of citizens).
Results
During the follow-up period 1987–2005, 54 336 cases of acute myocardial infarction and 91 553 cases of other IHDs were registered (table 2). The main analyses included 5024 cases of acute myocardial infarction exposed to small particles, 38 866 cases exposed to large particles and 14 567 unexposed cases. The corresponding overall analysis of IHDs included 8241 cases exposed to small particles, 62 696 exposed to large particles and 27 381 unexposed (table 3).
HRs were adjusted for age, sex and particle size. Further adjustment for socioeconomic group and residential area had generally a minor effect on the point estimates, except among women with low numbers of exposed cases. Workers ever exposed to small or large particles showed a slightly increased risk for acute myocardial infarction. In workers ever exposed to small particles, fully adjusted HR for acute myocardial infarction was 1.12 (95% CI 1.09 to 1.15) and for large particles the HR was 1.14 (95% CI 1.10 to 1.18) (table 3). The association between small particles and acute myocardial infarction was somewhat stronger for subjects exposed at two consecutive censuses, 5 years apart, HR 1.21 (95% CI 1.11 to 1.31). Women had a higher RR of acute myocardial infarction than men; HR for women ever exposed to small particles was 1.30 (95% CI 1.12 to 1.51) and for men 1.10 (95% CI 1.07 to 1.14). Generally, point estimates associated with exposure to small particles were somewhat higher than point estimates for exposure to large particles in subjects exposed at two consecutive censuses, although this difference was not statistically significant.
There were no indications of a positive dose–response relationship between exposure intensity and risk of acute myocardial infarction, neither for small or large particles. The HRs for exposure to small particles in low, medium and high exposed were 1.26 (95% CI 1.22 to 1.30), 1.08 (95% CI 0.99 to 1.18) and 1.13 (95% CI 1.09 to 1.17). The HRs for ever exposure to large particles were 1.19 (95% CI 1.16 to 1.21), 1.14 (95% CI 1.10 to 1.17) and 1.15 (95% CI 1.11 to 1.19).
Medium or high exposure to small particles with a prevalence of ≥80% occurred in firefighters; welders and furnacemen, metal annealers, temperers and case-hardeners. We investigated to which extent each of these occupations contributed to the overall risk. The HR for acute myocardial infarction was significantly increased in welders; furnacemen and metal annealers, temperers and case-hardeners but not in firefighters (table 4).
Workers in occupations assessed as exposed to particles also had a significantly increased risk for other IHDs. The results were essentially similar to that of acute myocardial infarction, with principally somewhat higher point estimates for exposure to small particles and for a longer period of time and higher risk estimates in women than in men (tables 3 and 4).
Discussion
Our main finding was an increased risk of IHD among workers occupationally exposed to particles. The risk was somewhat higher for exposure to particles less than 1 μm in size than to particles larger than 1 μm for workers exposed in two consecutive censuses, that is, more than 5 years, but there were no indications of an association between exposure intensity and risk. We found somewhat higher risk estimates among women than among men.
The long tradition of population registration in Sweden, with unique personal identity numbers, makes it possible to perform high-quality epidemiological studies. In addition, the present study is large, including the entire occupationally active Swedish population, and use prospectively collected exposure information and health outcome data obtained from population registers, thus preventing bias due to influence of the outcome on the classification of exposure status or from selection mechanisms. The large data set reduces random errors in the findings.
The main limitation of the study is the lack of information on tobacco smoking habits. Exposures are assessed based on a JEM applied to occupational titles from censuses—meaning that there is a risk for misclassification in the exposure assessment and problems to rule out confounding. Against this background, we decided to make the restriction to manual workers since there is a large variation in IHD between socioeconomic groups in Sweden—mainly because of variations in lifestyle.21 Furthermore, the prevalence of exposed subjects in other socioeconomic groups than manual workers is very low, which will increase the risk for misclassification of the exposure in these groups. The restriction to manual workers resulted in a more homogeneous study base regarding socioeconomic factors including tobacco smoking. In addition, all analyses were adjusted for the four socioeconomic subgroups among manual workers.
The second ‘restriction’ (defining exposed as occupations with >80% prevalence) was a way to maximise the contrast in exposure and also avoiding to make ungrounded assumptions on how to combine prevalence and exposure intensity in a combined exposure metric. In this explorative study, we chose to compare highly exposed with unexposed workers, which was possible due to the large study population, implying a high statistical power.
Welders; furnacemen and metal annealers, temperers and case-hardeners were the three job-titles contributing to the high risk of IHD among those exposed to small particles. These groups are all exposed to combustion fumes rich in particles and to heat. The chronic effect of work in a hot environment is debated and there are few epidemiological studies on this topic. No excess of cardiovascular disease was found among heat-exposed stainless-steel producing workers.22 Wild et al23 found an increased mortality from cardiovascular diseases among potash miners exposed to heat. Particle exposure was, however, not controlled for in the analyses, and hence, it is possible that the effect may be due to particle exposure. A hypothesis states that exposure to noise can contribute to increased blood pressure levels.24 The association between occupational noise exposure and IHD is less studied. Among lumber mill workers in British Columbia, an exposure–response trend was found between occupational noise and mortality due to acute myocardial infarction,25 but workers in two English nuclear power plants with long-term exposure to noise showed no increased risk for IHD mortality.26 Considering earlier findings from epidemiological air pollution studies as well as mechanistic data, it seems more likely that the present findings are caused by occupational exposure to particles than by exposure to heat or noise.
A JEM is a powerful tool being a relatively cost- and time-efficient method compared with other exposure assessment methods. However, JEMs, as well as other methods for exposure assessment, will give some misclassification of exposure. Furthermore, exposure assessment in this study was based on occupations registered in 5-year intervals and hence is not representing the entire work history for most subjects. However, misclassification of exposure in this study is non-differential and will bias the HRs towards the null. JEMs are optimal when specificity is favoured over sensitivity. In the present study, the specificity was high since a job was classified as exposed only if the probability was above 80% and the level of exposure was above 1/10 of the OEL.
We defined level of exposure in relation to Swedish OELs, in order to take the toxic effect of each particle type into account. This approach would not be appropriate if the biological effect is related solely to an unspecific particle effect. However, the use of absolute weight (milligrams per cubic metre) to classify intensity would underestimate the exposure to particles of small size and was not considered as an appropriate metric when comparing the effects of small versus large particles.
Our results are in accordance with other studies on occupational exposure to combustion particles. A Swedish study of occupational exposure to particulate air pollution among male construction workers9 showed that exposure to small particles (diesel exhaust) was associated with an increased risk for IHD (RR 1.18, 95% CI 1.13 to 1.24) than large particles of inorganic dust (RR 1.07, 95% CI 1.03 to 1.12), after controlling for tobacco smoking. In a study of occupational exposure to combustion products,15 the RR of myocardial infarction was 2.11 (95% CI 1.23 to 3.60) in exposed subjects after controlling for tobacco smoking. A recently published study of Danish welders found an increased standardised incidence ratio for acute myocardial infarction of 1.12 (95% CI 1.01 to 1.24). The risk estimate was adjusted for smoking, alcohol and medicine use.10 A systematic review of 37 articles27 came to the conclusion that occupational exposure to particulate matter (mainly from traffic and welding) may be associated with IHD mortality and myocardial infarction.
In the present study, women had a higher RR for IHD than men. There is some evidence in the literature that the association between exposure to airborne particulate matter and the risk of fatal coronary heart disease is stronger in women.28 ,29 Studies including only women have also demonstrated more pronounced effects of particle matter exposure than studies including both sexes.3 Several hypotheses have been suggested to explain these sex-specific differences. There are more non-smokers among women and effects of air pollution may be stronger in non-smokers than in smokers.30 Oxidative and inflammatory effects of smoking may dominate to such an extent that the additional exposure to air pollutants may not further enhance effects along the same pathways in smokers. In addition, women have greater airway reactivity than men, as well as smaller airways,31 the latter resulting in differing particulate deposition patterns in women and men. This may partly explain the differences between the sexes and are probably important explanatory factors in a 22-year follow-up study of cigarette smoking and risk for myocardial infarction.32 Smoking women had significantly increased risks at a consumption of 3–5 g of tobacco per day with RR of 2.14 (95% CI 1.11 to 4.13). Corresponding RR for smoking men were 1.03 (95% CI 0.53 to 2.01). Four occupations were assessed as exposed to small particles according to the JEM: welders; metal annealers, temperers and case-hardeners; furnacemen and firefighters. Workers in these occupations also had a high risk for IHD, except for firefighters. Employment as a firefighter in Sweden requires a mandatory and demanding exercise test, repeated annually for persons older than 50 years, and thus, there is a strong selection of healthy individuals in this occupation that may mask a harmful effect of the particle exposure.
The three occupations that were associated with an increased risk for IHD involve a mixture of exposures, but according to our definition of exposed occupations, welders are exposed to welding fume and iron dust, metal annealers, temperers and case-hardeners are exposed to benzo(a)pyrene, and furnacemen to benzo(a)pyrene and quartz dust. These exposures are formed in combustion processes from heating of metals. Primary combustion particles are <0.01 μm in size but coalesce within seconds to form larger aggregates with a size of 0.01–1 μm. Small particles with a large surface area per mass have been found to induce a more pronounced inflammatory response than large particles.16 Ultrafine particles (<100 nm or 0.1 μm) are to a greater extent deposited in the alveolar region of the lung and are not as readily phagocytised by alveolar macrophages as are larger particles and may enter interstitial sites or even the blood circulation. Furthermore, these particles have a relatively large surface area per given mass, resulting in a high biological reactivity possibly associated with a toxic effect. The increased surface area may be a carrier for co-pollutants as metals and organic compounds such as polycyclic aromatic hydrocarbons (PAHs). Some species of PAHs have a high potential to induce inflammation, and exposure to PAH have been associated with elevated risks of mortality from IHD in occupations such as aluminium smelting,33 chimney sweeping8 and waste incineration.34 However, it is not fully understood how much of the effect that is attributable to the particles or to particle bound compounds.
Previous studies of ambient air pollution have shown an increased risk of IHD. Experimental evidence supports the hypothesis that particulate matter cause alterations in blood rheology that favour thrombosis, cardiac arrhythmias, acute vascular dysfunction, plaque instability, and the long-term development of atherosclerosis. Additional pathways such as changes in autonomic balance via lung neural reflex arcs triggered by inhaled particulate matter may also be involved.35 There is mechanistic evidence that particulate air pollution cause a low-grade pulmonary inflammation associated with release of proinflammatory cytokines.7 This may result in increased coagulability of the blood, triggering cardiovascular events in susceptible subjects.36 Similar mechanisms have been suggested for occupationally exposed workers.37 Exposed tunnel workers had increased blood concentrations of interleukin 6 and fibrinogen.38 Interleukin 6 is released from the bronchial mucosa and stimulates the production of fibrinogen. There is also an association between respiratory symptoms and IHD.39
In conclusion, our results give some support to the hypothesis that occupational exposure to particles increases the risk of acute myocardial infarction and other IHD. The results indicate that women may be more susceptible. The findings from this explorative study must be interpreted cautiously since there were no data on tobacco smoking, although potential confounding should be reduced by the limitation of the cohort to manual workers, adjustment for socioeconomic subgroups and for residential area. Further studies are needed to investigate the effects caused by specific particles and exposure levels in the work environment.
References
Footnotes
Funding Financial support from the Swedish Council for Working Life and Social Research (grant number 2006-0849).
Competing interests None.
Ethics approval Ethics approval was provided by the regional ethics committee in Stockholm, Sweden.
Provenance and peer review Not commissioned; externally peer reviewed.