Chapter 34: Occupational Toxicology

The work environment has played a significant role in the occurrence of adverse human health effects due to chemical and biological hazards for centuries. Early writings by Ulrich Ellenbog (1435–1499), Agricola (1494–1555), and Paracelsus (1492–1541) revealed the toxic nature of exposures in mining, smelting, and metallurgy. A systematic treatise by Ramazzini (1633–1714) described the hazards as they applied to miners, chemists, metal workers, tanners, pharmacists, grain sifters, stonecutters, sewage workers, and even corpse bearers. Legislation to protect worker health began in England with the Factory Act of 1883, which established a factory inspectorate and limited child laborers nine to 13 years old to a 48-hour workweek, and 14 to 18 years olds to a 68-hour workweek. Today we continue to be concerned with occupational health and safety in a wide variety of work environments. Although occupational settings in developed countries are safer now than in the past, the levels of risk deemed acceptable have decreased while the recognition of the causal link of exposures to chronic diseases or diseases with long latencies has increased. As new hazards arise with the emergence of new technologies, we must be prepared to assess the risks and protect the health of workers. With increased globalization, there exists a responsibility to extend health protection to workers in developing nations who too often bear the burden of high exposures to occupational toxicants.

Occupational toxicology is the application of the principles and methodology of toxicology to understanding and managing chemical and biological hazards encountered at work. The objective of the occupational toxicologist is to prevent adverse health effects in workers that arise from exposures in their work environment. Because nonoccupational exposures can act as confounders or can increase the susceptibility of individual workers, occupational toxicologists must evaluate the entire spectrum of exposures experienced by the work force under consideration. Occupational toxicology is a discipline that draws on industrial hygiene, epidemiology, occupational medicine, and regulatory toxicology. The occupational toxicologist must have an intimate knowledge of the work environment and be able to recognize and prioritize exposure hazards. Because the work environment can present exposures to complex mixtures, the occupational toxicologist must also recognize those that are particularly hazardous when occurring in combination.

It is often difficult to establish a causal link between a worker's illness and job. First, the clinical expressions of occupationally induced diseases are often indistinguishable from those arising from nonoccupational causes. Second, there may be a protracted but biologically predictable latent interval between exposure and the expression of disease. Third, diseases of occupational origin may be multifactorial with other environmental factors or personal (eg, genetic or epigenetic) risk factors, contributing to the disease process. Nevertheless, studies have repeatedly shown that the dose of toxicant is a strong predictor of the likelihood, severity, and type of health effect.

The Nature of the Work Force

Approximately 40% of the global work force, some one billion people, work in agricultural production. Many of these workers are engaged in subsistence agriculture (International Labour Organization, 2011). In contrast, the demographics and distribution of the work force in industrialized nations have undergone a progressive shift over the past three decades, moving away from jobs in heavy industry and agriculture toward jobs in the service sector and high-technology industries. In some instances, these manufacturing jobs have moved to less developed countries with less stringent worker health protection. There are currently 134 million people in the United States in the civilian, paid work force (seasonally adjusted data). The civilian labor force participation rate is 64.1% (United States Department of Labor [USDL], 2011b), a figure that has grown markedly over the past 50 years owing to the increased entry of women into the labor force. The 27 countries of the EU have an estimated 222.3 million workers (Employment in Europe, 2010), and Japan has about 62.6 million (Japan Monthly Statistics, 2011). Fig. 34-1 shows the breakdown of employment in the United States. This illustrates that there are about 20.1 million workers (15.0%) engaged in manufacturing, construction, mining and logging, and agriculture—occupations that have the potential for significant exposure to chemical and biological agents. Furthermore, there are occupations in the service sector, such as automobile repair; other repair and maintenance; work in gasoline stations, pipeline transportation, truck and rail transportation; waste management and remediation services; and employment in botanical gardens that can also include exposures to hazardous chemicals. Service-producing occupations (including government jobs) account for the majority (85%) of US jobs. On average, employees in the service sector work 6.7 fewer hours per week than in the goods-producing sector. Farm work employs an estimated 2.11 million workers, 16 years of age and older, and the US Bureau of Labor Statistics has documented an average of 41.8 hours worked per week. Work in agriculture is markedly different from most other occupations in four fundamental ways. First, 44% of those employed in agriculture are self-employed or unpaid family members. Second, the overwhelming majority of farm establishments (90%) have fewer than 20 employees, and these work sites represent 35% of the agricultural work force (USDL, 2011a). Third, although the annual national average hours worked per week is 41.8, many farmers, ranchers, and farm workers have periods (eg, harvest time) when they work as many as 20 hours per day, seven days per week. Fourth, according to the US Bureau of the Census, there are in excess of 290,000 children who identify agricultural work as their major employment (GAO/HEHS, 1998). Department of Labor data indicate that, on average, 128,500 hired farm workers between the ages of 14 and 17 were working annually in crop production (GAO/HEHS, 1998). The presence of children in the work force has important ramifications for body burdens, disease latency, toxicokinetics, and biotransformation of toxicants.

Determinants of Dose

Dose is defined as the amount of toxicant that reaches the target tissue over a defined time span. In occupational environments, exposure is often used as a surrogate for dose. The response to a chemical is dependent on both host factors and dose. Fig. 34-2 illustrates the pathway from exposure to subclinical disease or to adverse health effect, and suggests that there are important modifying factors: contemporaneous exposures, genetic and epigenetic susceptibility, age, gender, nutritional status, and behavioral factors. These modifying factors can influence whether a worker remains healthy, develops subclinical disease that is repaired, or progresses to frank illness. Workplace health protection and surveillance programs (shown in blue) can reduce exposures, disrupt the exposure-dose pathway, or identify internalized dose and early effects before irreparable disease develops. These programs help to ensure a safe workplace and a healthier work force.

As illustrated in Fig. 34-2, dose is a function of exposure concentration, exposure duration, and exposure frequency. Individual and environmental characteristics can also affect dose. Exposure assessment is the process of quantifying the intensity, frequency, and duration of exposures and their determinants in order to better estimate dose. Table 34-1 indicates determinants of dose for exposure via the inhalation and dermal routes. For determining inhalation exposures, environmental conditions such as concentration, particle size distribution, and properties of the chemical are important. However, respiratory rate and breathing volume as well as other host factors contribute. Protection afforded by personal protective equipment (especially respirators) will reduce but not eliminate exposure. The degree of exposure reduction for a particular respirator (the workplace protection factor) varies with respirator design, fit, maintenance, manner of use, and environmental conditions from less than five to 10,000. A respirator with a workplace protection factor of 10, when used properly, will reduce particulate matter concentrations inside the mask to one-tenth that outside the mask. A commonly used two-strap fiber respirator for lowering particulate matter exposure, the N95, should reduce exposures by 95% but may fail to meet this criterion for submicron aerosols (Balazy et al., 2006).

Dermal exposures depend on toxicant concentration; work conditions, including the degree and duration of wetness; and the ambient conditions at the work site (Table 34-1). Some determinants of dermal dosing relate to the physicochemical properties of the chemical as they affect the percutaneous absorption rate. These include solubility, temperature, pH, molecular size, and chemical characteristics of the vehicle. Host factors also influence dermal absorption and distribution. Important factors include the surface area of the skin that is exposed, the integrity of the skin, blood flow, and biotransformation. Because the stratum corneum, the outer layer of the epidermis, is the principal barrier to dermal uptake, the thickness of this layer in the exposed area has great significance. As an example, the absorption of hydrocortisone through the plantar foot arch is 1/25 less than through the back and 1/300 less than through scrotal skin (Bason et al., 1991). A classic human in vivo study by Maibach et al. (1971) using the pesticide parathion demonstrated a 12-fold range in percutaneous absorption between the anatomical site with the highest (scrotum) to that with the lowest (forearm) percent of dose absorbed. For malathion there was a 12-fold difference in absorbed dose between the jaw angle and the palm. The use of protective gloves and clothing or aprons and the application of barrier creams can greatly reduce exposure. For maximal protection, it is important that the glove be constructed of material tailored to the toxicant(s) of concern.

Occupational Exposure Limits

One of the roles of the occupational toxicologist is to contribute data to the process of establishing standards, or determining the appropriateness of those standards. Workplace exposure limits exist for chemical, biological, and physical agents, and are recommended as guidelines or promulgated as standards in order to promote worker health and safety. For chemical and biological agents, exposure limits are expressed as acceptable ambient concentration levels (occupational exposure limits [OELs]), or as concentrations of a toxicant, its metabolites, or a specific marker of its effects in biological systems (biological exposure indices).

OELs are established as standards by regulatory agencies or as guidelines by research groups or professional organizations. In the United States, the Occupational Safety and Health Administration (OSHA) under the Department of Labor promulgates legally enforceable standards known as permissible exposure limits (PELs). These standards are determined and supported by the best scientific evidence available and assure “to the extent feasible… that no employee will suffer material impairment of health or functional capacity” with regular exposure “for the period of his working life.” OSHA has defined an eight-hour time-weighted average (TWA) PEL (US Federal Register, 1992) as the “employee's average airborne exposure in any eight-hour work shift of a 40-hour workweek which shall not be exceeded.” The TWA PEL is established as the highest level of exposure to which an employee may be exposed without incurring the risk of adverse health effects (OSHA, 1995). Although approximately 500 PELs have been promulgated, there are not enough to regulate exposures to the vast number of compounds to which workers are exposed, and some of the existing PELs do not reflect current knowledge.

The National Institute for Occupational Safety and Health (NIOSH), under the Centers for Disease Control and Prevention, publishes recommended exposure limits (RELs) that are more frequently updated and are generally more stringent than PELs. NIOSH also performs research and disseminates information on workplace hazards and their prevention. Most developed countries have governmental inspectorate agencies, analogous to OSHA, that are responsible for establishing and enforcing OELs. In some countries, the insurance system also plays a significant role.

The European Commission has established legally enforceable binding occupational exposure limit values (BOELVs) and biological limit values (BLVs) for the protection of health and safety in the workplace. Socioeconomic and technical feasibility factors are also considered in setting these values. The OELs are established based on recommendations of the Scientific Committee on Occupational Exposure Limits (SCOEL), which evaluates published scientific data on hazardous compounds for regulation and provides assessment of exposure limits that “it believes will protect workers from chemical risks” (European Commission, 1995, 1998). Indicative Occupational Exposure Limit Values (IOELVs) are nonbinding, health-based OELs that do not consider technical or economic feasibility. A first list of IOELVs was established in 2000 followed by a second in 2006. EU member countries may not promulgate OELs less stringent than these values.

The American Conference of Governmental Industrial Hygienists (ACGIH) is a nonprofit scientific association that publishes OELs for chemicals and for physical agents. These take the form of threshold limit values (TLVs) and biological exposure indices (BEIs). They are frequently revisited and generally reflect current knowledge in occupational toxicology and industrial hygiene. They are developed as guidelines and are not enforceable standards; however, many industries adopt TLVs and BEIs as internal OELs. As stated by the ACGIH, “The TLVs and BEIs represent conditions under which ACGIH believes that nearly all workers may be repeatedly exposed without adverse health effects” (ACGIH, 2010). These guidelines are health based and do not include consideration of technical feasibility or economic effects of implementation.

Three types of TLVs are suggested, depending on the time scale of adverse effects inducible by the toxicants. The time-weighted average TLV (TLV-TWA) is an OEL for exposures, averaged over an eight-hour day, five-day workweek regimen. These are generally applied to toxicants that exert their effects over long periods. The short-term exposure limit (TLV-STEL) is an OEL for a 15-minute measurement period. The TLV-STEL should not be exceeded in any 15-minute sampling window, and there should be 60 minutes or more between exposures in this range. The ceiling limit (TLV-C) represents a concentration that should never be exceeded. These are usually applied to toxicants that cause acute effects (such as asphyxia or potent sensory irritation) and for which real-time monitoring devices are available. BEIs are guidelines for biological monitoring and represent levels “most likely to be observed in specimens collected from healthy workers who have been exposed to chemicals to the same extent, as workers with inhalation exposure at the Threshold Limit Value” (ACGIH, 2010). BEIs are recommended for analysis of urine, blood, and exhaled air. While hair, fingernails, adipose tissue biopsies, and other specimens are used in research and forensic toxicology, there are no BEIs for these.

It is important to recognize that OELs do not correspond to exposure conditions devoid of health risk. The concept of acceptable exposure level must be understood as the level of exposure, below which, the probability of impairing the health of the exposed workers is acceptable. The process of deciding what is an acceptable risk to occupational or environmental hazards blends the scientific disciplines of exposure assessment and toxicology with often vexing policy issues. Historically, acceptable risk in a society is related to the general health of the population and to a host of factors that influence how risks are perceived. To determine that the risks from an occupational hazard are acceptable, it is necessary to characterize the hazard, identify the potential diseases or adverse outcomes, and establish the relationship between exposure intensity or dose and the adverse health effects. If biological markers of exposure or early reversible effects are identified, this can aid in the risk-assessment process.

Routes of Exposure

Diseases arising in occupational environments involve exposure, primarily through inhalation, ingestion, or dermal absorption. In the vast majority of work environments, inhalation of toxicants is a primary concern. Inhalation exposures can occur with gases, vapors, liquid aerosols, particulate aerosols, fumes, and mixtures of these. Dermal exposures are also important and can arise from airborne materials as well as liquids splashed onto the skin, immersion exposures, or from material handling.

Additional exposure hazards exist for infectious agents. Exposures leading to occupational infections may arise through inhalation or ingestion of microorganisms, but can also arise from needlesticks in health care workers or through insect bites among farmers, natural resource workers, and others employed out of doors. Additionally, poisonings from toxic plants or venomous animals can occur through skin inoculation (eg, zookeepers, horticulturists, or commercial divers).

Agents Associated With Diseases

There are a myriad of agents responsible for occupational diseases. While some act on a particular organ such as the liver or kidney, others can affect multiple organ systems. Table 34-2 presents a list of the major occupational diseases and examples of agents that cause them. This is not intended to be all-inclusive. Rather it is meant to highlight what are historically the most prevalent and widely recognized occupational diseases, plus those that continue to be prevalent in the workplace. The toxicants listed are those for which there is a strong association with the disease or the most conclusive data to support causality. Examples are shown for cancer and for diseases of the lung and airways, heart, liver, kidney, skin, nervous system, immune system, and reproductive system. Several examples of occupational infectious diseases are also listed to highlight the fact that, in many work settings, infectious agents may constitute the major hazard and may coexist with chemical hazards. Most of the occupational diseases listed in Table 34-2 are associated with industrial chemicals. These are discussed in other chapters throughout this book.

Table 34-3 lists forty agents that are known by the International Agency for Research on Cancer (IARC) to be carcinogens in humans (Group 1) and for which there has been or is currently extensive occupational exposure. This list includes agents such as asbestos, arsenic, benzene, vinyl chloride, and coal tars. IARC Group 1 refers to agents or mixtures that are known to be carcinogenic to humans based on sufficient epidemiological evidence usually accompanied by limited or sufficient animal evidence.

Occupational Respiratory Diseases

Because inhalation generally represents the most significant route of exposure, many of the major occupational diseases affect the lung and airways. These diseases have been studied extensively and are largely responsible for the creation of the occupational regulatory framework. Deaths due to occupational lung diseases such as asbestosis, coal workers' pneumoconiosis, silicosis, byssinosis, and occupational asthma have been recognized for centuries and led to important legislation such as the 1970 US Occupational Safety and Health Act. These occupational lung diseases continue to have significant associations with morbidity. Table 34-4 lists the crude US death rate and annual deaths for 2005 (the most recent data available) and illustrates that while the death rates are fairly low, there are still about 2460 deaths per year attributable to asbestos, silica, coal dust, and other pneumoconiotic dusts; 2704 malignant mesothelioma deaths; and 67 deaths from hypersensitivity pneumonitis. However, fatalities are just the tip of the iceberg as some occupational lung diseases are rarely fatal, yet may still be debilitating. Every year in the United States, there are 20,000 hospital discharges related to cases of asbestosis, and 188,000 coal workers receive federal Black Lung Benefits (NIOSH, 2006). The US Mine Safety and Health Administration has compiled data on inspector- and mine operator-collected samples in coal mining operations and found that 29.3% of these exceeded the PEL for respirable quartz and 7.5% were over the PEL for respirable coal dust.

Many of the diseases listed in Table 34-2 are known by other names that refer to a particular occupation or agent. One example is hypersensitivity pneumonitis, an allergic lung disease marked by interstitial lymphocytic pneumonitis and granulomatous lesions. Hypersensitivity pneumonitis is also known as extrinsic allergic alveolitis, farmer's lung disease, bagassosis (sugar cane), humidifier fever, Japanese summer house fever, pigeon breeder's lung, and maple bark stripper's lung, depending on the occupational setting in which it arises. Although we often think of these as the same disease, it is important to recognize that the exposures and physiological responses they induce are complex and may differ in the manifestation of the disease.

The US Bureau of Labor Statistics tracks data for nonfatal occupational injuries and illnesses, of which, there were 3.3 million in 2009. The vast majority of these were injuries, but there were still over 220,000 occupational illnesses reported in 2009 (USDL 2011c). There were 21,500 reported respiratory conditions arising from exposure to toxic agents and dust diseases of the lung. The former category includes pneumonitis, pharyngitis, rhinitis, and acute lung congestion. Table 34-5 shows data for rates and cases in 2009 by industry division. Poisonings were relatively rare, whereas skin diseases occurred at rates as high as 5.2 per 110,000 workers per year. The manufacturing sector accounted for approximately 29% of all occupational illnesses. Seven industry codes that had respiratory condition incidence rates that exceeded seven per 10,000 full-time workers were ambulance services, police and fire protection, urban transit systems, support activities for water transportation, adhesive manufacturing, grain and oil seed milling, and animal slaughtering. The relevant exposures for the first four of these are combustion products from fires and diesel engines. The rate for chemical manufacturing was much lower, 1.8 cases per 10,000 workers, down by 50% since a decade ago.

Toxic gas injuries are often characterized by leakage of both fluid and osmotically active proteins from the vascular tissue into the interstitium and airways. Important determinants of the severity and location of injury are the concentration and water solubility of the toxic gas or vapor. Anhydrous ammonia, with its extremely high solubility, primarily damages the eyes, sinuses, and upper airways. The vapors combine with water in the tissue and form ammonium hydroxide, quickly producing liquefaction necrosis. Chemicals with lower solubility, such as nitrogen dioxide, act more on the distal airways and alveoli and take longer to induce tissue damage.

Occupational asthma may be defined as a “disease characterized by variable airflow limitation, and/or airway hyperresponsiveness, and/or inflammation due to causes and conditions attributable to a particular occupational environment and not to stimuli encountered outside the workplace” (Bernstein et al., 2006). Data from the NIOSH Work-Related Occupational Lung Disease Surveillance System has shown that 20% of occupational asthma cases represented aggravation of preexisting asthma by workplace exposures and 68% of cases represented new-onset asthma caused by occupational exposures. The remaining 12% could not be classified (NIOSH, 2011). The US National Health Interview Survey has provided data on the prevalence of asthma based upon usual industry and smoking status (Table 34-6). These data indicate that asthma prevalence is highest in health services; general merchandise stores; food, bakery, and dairy stores; and eating and drinking places. Also high are hospitals; transportation equipment; printing, publishing, and allied industries; utilities and sanitary; furniture, lumber, and wood products industries; and chemicals and allied products. Interestingly, in some industries, current smokers demonstrate the higher prevalence, while in other industries, the higher prevalence is seen among nonsmokers. This may reflect that many people diagnosed with these conditions quit smoking.

There are a variety of industries in which there is increased risk of developing work-related asthma. In chemical-based industries, plastic and rubber polymer precursors, diisocyanates, reactive dyes, and acid anhydrides are recognized low-molecular-weight sensitizing compounds. Biocides and fungicides used in metal fabrication and machining, custodial services, lawn and turf growing, and agriculture are also chemicals associated with occupational asthma. A number of metals can induce sensitization and asthma, including chromium, cobalt, nickel, platinum, and zinc. Enzymes pose significant risks for occupational asthma (Heederik et al., 2002). Examples include α-amylase among bakery workers and subtilisin, a protease used in laundry detergents. The enzyme production industry has had to adopt strict environmental and process controls to reduce the incidence of occupational asthma in their production facilities.

Those working with animals or animal products are at increased risk of developing allergy (Elliott et al., 2005) and occupational asthma (Pacheco et al., 2003). Animal handlers, processors, and laboratory technicians who work with animals can become immunologically sensitized to urine or salivary proteins in many vertebrates; proteins in bat guano and bird droppings; animal dander; serum proteins in blood products; dust from horns, antlers, and tusks; or the shells of crustaceans. Very high rates of sensitization can occur in shellfish processors (Glass et al., 1998). Arthropods such as insect larvae, cockroaches, mites, or weevils are recognized inducers of work-related asthma. Plants and plant products (eg, soy flour, spices, and coffee beans) can also cause asthma among workers. In a variety of occupations, exposure to fungi, especially of the genera Aspergillus, Penicillium, Rhizopus, Mucor, and Paecilomyces, are associated with allergic rhinitis and asthma. These are especially present in sawmills, woodchip handling, and composting facilities (Duchaine et al., 2000; Eduard et al., 1992). Apart from the contaminating microorganisms, certain woods themselves produce chemical sensitizing agents. Examples include western red cedar, redwood, and some tropical hardwoods.

Asthma emerged in the 1980s as a major occupational health concern among health care workers. In order to reduce the risk of hepatitis B and other infectious diseases, health care workers adopted the use of natural rubber latex gloves for barrier protection. Proteins from the latex of the rubber tree, Hevea brasiliensis, led to immunological sensitization. Thirteen of these high-molecular-weight proteins have been characterized as allergens (Bernstein et al., 2003). Many other plants of less commercial value produce a similar milky fluid when cut, and have similar sensitizing properties. A shift to powder-free and nonlatex examination gloves has reduced the incidence of latex allergy among health care workers and patients.

Studies of asthma prevalence in occupational settings with exposure to low-molecular-weight chemicals have suggested prevalence rates of 5% to 10% for toluene diisocyanate (Baur, 1996; Becklake et al., 2006), 3.2% to 18% for anhydrides (Venables et al., 1985; Wernfors et al., 1986), 4% for plicatic acid from western red cedar (Chan-Yeung et al., 1984), and 54% in a platinum refinery (Venables et al., 1989). For high-molecular-weight allergens, prevalence among shellfish processors was estimated at 21% to 26% (Desjardins et al., 1995; Glass et al., 1998), 11% to 44% among lab animal workers (Cullinan et al., 1994; Fuortes et al., 1997; Hollander et al., 1997), and 5% to 7% among bakers exposed to wheat and α-amylase (Heederik et al., 2002).

Agricultural workers exposed to grain dust, cotton dust, or atmospheres in swine or poultry confinement barns are at risk for the development of an asthma-like syndrome. This syndrome is an acute nonallergic airway response characterized by self-limited inflammation with neutrophilic infiltrates and increased proinflammatory cytokines and chemokines (eg, TNFα, IL-6, and IL-8), but it does not include persistent airway hyperreactivity as in occupational asthma (Schenker et al., 1998). Asthma-like syndrome includes cough, mild dyspnea, fever, malaise, and cross-shift declines in lung function. Endotoxin in combination with other inflammatory bioaerosols is the likely etiological agent (Douwes et al., 2003; Schwartz et al., 1994, 1995).

An emerging area of concern is adverse effects of respiratory exposures to manufactured nanomaterials. There are currently over one thousand consumer products that contain nanotechnology-based materials (Nanotechnology Consumer Products Inventory, 2011). Nanomaterials are engineered structures that have a primary size less than 100 nm in at least one dimension. Airborne nanoparticles generally form agglomerates in which groups of particles are held together by relatively weak forces (eg, van der Waals). They may also aggregate forming clusters that are fused, sintered, or chemically bonded to one another (Pettibone et al., 2008). Manufactured nanomaterials find applications in electronics, construction materials including wall and floor coverings, cosmetics, drug delivery, medical imaging, food products, packaging materials, and textiles, to name a few. Occupational exposures occur in the manufacture of the nanomaterials and in their use in fabricating materials and consumer products. Exposures can also occur when nanomaterials are cut or shaped and when product waste is discarded. Engineered nanomaterials may be carbon-based, metal-based, or biological in nature (Adamcakova-Dodd et al., 2010). Biological actions of nanoscaled materials may differ from the comparable bulk materials or micron-sized particles (Nel et al., 2006; Grassian et al., 2007). Inhaled nanomaterials may induce pulmonary toxicity or they can cause adverse effects in other tissues through adsorption and transport, generation of toxic substances by their dissolution or degradation, or by crossing key physiological barriers, or cell and nuclear membranes (Kim et al., 2011). In early 2011, NIOSH issued a new guideline on TiO2 marking the first instance of two guidelines for the same compound based on particle size. In mid 2011, NIOSH has issued for peer review a draft Current Intelligence Bulletin on carbon nanotubes and nanofibers. Other research and regulatory bodies worldwide are studying nanomaterials and conducting human health risk assessments.

Other Occupational Diseases

Occupational diseases of the skin are common but less often fatal that those of the respiratory system. Irritant dermatitis and allergic contact dermatitis have the highest incidence of reported skin conditions. Table 34-5 shows data for rates of skin diseases or disorders in 2009 by industry division. These data indicate that agriculture, fishing, and forestry services; and manufacturing carry the highest risks. In the service sector, the education and health services division has the highest rate of skin diseases. This is due to the high rate of skin disorders in hospitals (7.2 per 10,000) and in nursing and residential care facilities (10.6 per 10,000). In 2009, 14% of reported occupational skin disorders were in manufacturing while 21% were in education and health services.

Occupational toxicants may induce diseases in a variety of body sites distant from the lung or skin. These include tumors arising in the liver, bladder, gastrointestinal tract, or hematopoietic system and are attributable to a variety of chemical classes. Further discussion of other occupational diseases and the toxicants listed in Table 34-2 can be found in the relevant chapters in Units IV and V of this text.

Nervous system damage can be central, peripheral, or both. It may be acute, as with some organophosphate exposures, or chronic, as with organomercury poisoning or acrylamide-induced neuropathy. Injury affecting the immune system may arise from the immunosuppressive effects of chemicals such as dioxins or toxic metals. Many occupational diseases of the immune system occur due to hypersensitivity leading to respiratory or dermal allergy or systemic hypersensitivity reactions. Autoimmune syndromes have been associated with occupational exposures to crystalline silica and vinyl chloride.

Occupational diseases of the cardiovascular system include atherosclerosis, a variety of arrhythmias, problems with coronary blood supply, systemic hypotension, and cor pulmonale (right ventricular hypertrophy usually due to pulmonary hypertension as with chronic obstructive pulmonary disease). Liver diseases such as carbon tetrachloride-induced fatty liver and hepatocellular death due to toxic concentrations of acetaminophen have classically been used to illustrate chemical mechanisms of cellular injury leading to organ failure. These are thoroughly discussed in Chap. 13. Occupational diseases of the reproductive system can be gender- and organ-specific; but several toxicants—including carbon disulfide, lead, and vinyl chloride—may affect both sexes.

Exposures to infectious agents are a part of a variety of occupations (Thorne and Duchaine, 2007). Veterinarians, health care workers, and biomedical researchers studying infectious agents have exposures that are largely known and infection control strategies can limit their risks. For others, such as farmers and foresters, specific risks may be less obvious. Zoonotic diseases such as Q-fever, rabies, leptospirosis, and brucellosis may affect abattoir workers, zookeepers, animal handlers, and veterinarians. Foresters, field biologists, and natural resource workers who spend time in wooded areas experience tick- and mosquito-borne illnesses at a higher frequency than that of the general population. These illnesses include the arboviral encephalitides, Rocky Mountain spotted fever, Lyme disease, and ehrlichiosis. Occupational infections may arise as a result of work settings, bringing people into close proximity with other people or animals, thus facilitating the transmission of microorganisms. Occupational infectious diseases attributable to the clustering of people affect workers in such facilities as day care centers, schools, health care settings, correctional facilities, dormitories, military barracks, or shelters for the homeless, among others. Industrial settings can place large numbers of workers in a shared space, leading to increased transmission of diseases. This is especially true for diseases with annual outbreaks, such as influenza and Norwalk-like viruses. Exposures to chemicals may increase the susceptibility of workers to infection through irritation of mucosa or the pulmonary epithelium or through immunosuppression leading to impaired host defense.

Both industrial and nonindustrial occupational environments may pose occupational hazards due to the presence of chemical or biological agents. Reports of work environments with ineffective ventilation or decreased ventilation rates and increased utilization of synthetic building materials have demonstrated a rise in complaints associated with occupancy in buildings. In some cases, service sector workers in a problem building develop specific clinical conditions with recognized etiology. This is defined as building-related illness. In other cases, symptoms are nonspecific and disappear when the worker leaves the problem building. When this occurs with sufficient prevalence, it is termed sick building syndrome. This can arise from volatile and semivolatile chemicals released from office materials, building materials, floor coverings, furniture, cleaning products, and microorganisms. Office buildings and residential settings comprise a complex ecology consisting of people, molds, mites, volatile organic compounds of microbial and nonmicrobial origin, and sometimes plants, pets, cockroaches, and other vermin (Thorne and Heederik, 1999b). Molds, house dust mites, and animal proteins are potent human allergens that can lead to allergy and asthma. Exposure to chemicals and biomolecules such as endotoxin may enhance this process (Douwes et al., 2003). Workers in laboratory animal facilities exhibit a high prevalence of allergy to rodent urinary proteins. In some cases, the occupied space of a building may be clean and dry, but local amplification sites for molds may develop. These may arise in ventilation systems, utility closets, subfloors or basements that serve as return air plenums, or in local sites of water damage. Such sites can become sources of microorganisms and aeroallergens of sufficient volume to generate significant bioaerosol exposures throughout the environment. Airborne viruses, bacteria, and fungi are responsible for a variety of building-related illnesses arising from organisms that are pathogenic to humans. Nonpathogenic microorganisms may induce symptoms or diseases through inflammatory processes, by stimulating innate or adaptive immune response, or by releasing noxious odors, allergenic compounds, or bioactive macromolecules. These may combine with industrial chemicals released into the air to create complex exposure environments.

Evaluation of Occupational Risks

In most instances of prolonged exposure to low levels of chemicals, there is a continuum between being healthy and being ill. The exposures may impart biochemical or functional changes that are without signs or symptoms, subclinical changes, or in cases of more significant toxicity, manifest as clinical disease. The health significance of the identified changes resulting from exposure to a particular agent must be assessed in order to determine which effects are adverse. Following this assessment, one must consider the interindividual variability or susceptibility factors that influence the risks. There is no single dose–effect relationship but a distribution of responses. Therefore, in order to recommend an acceptable exposure level to an industrial chemical, one must attempt to define the risk associated with adverse effects in the most sensitive populations exposed. It then remains to be determined what proportion of exposed subjects may still develop an adverse effect at the proposed acceptable exposure level. This acceptable risk level will vary according to a value judgment of the severity, permanence, and equality of the potential adverse effects, and the characteristics of the most susceptible population. Clearly, inhibition of an enzyme without functional consequences will be viewed as more acceptable than a more serious toxic effect, such as teratogenicity leading to a congenital malformation in the offspring of the exposed individual.

Establishing Causality

In complex occupational environments, it may be difficult to establish a causal relationship between a toxic substance and a disease. For this reason, a number of systematic approaches have been devised to help define causation. In 1890, Robert Koch proposed postulates for “proving” that a specific organism caused a specific disease. TM Rivers extended this approach to viruses in 1937. Sir Austin Bradford Hill suggested epidemiological criteria for assessing causality in 1965 considering strength, specificity, consistency, temporality, exposure period, exposure gradients, and biological plausibility of the associations. These schema and modern weight-of-evidence determination criteria were later combined to suggest a set of postulates for the evaluation of evidence for disease agents in organic dust (Donham and Thorne, 1994). A matrix was developed and is extended here to evaluate the weight of evidence for a causal association between a toxicant and an occupational disease (Fig. 34-3). Evidence from well-conducted in vitro studies, animal studies, human challenge studies, case reports, and epidemiological investigations are evaluated with regard to data quality and clarity of evidence in support of the establishment of causality. This evaluation is guided by seven criteria (shown in blue). If a chemical were thoroughly studied in animals, humans, and in vitro studies, and produced clear and convincing evidence of an exposure–response relationship in controlled studies that used appropriate models and relevant endpoints, that would constitute compelling evidence of a causal relationship between that chemical and that disease. Fig. 34-3 reminds us that a consortium of study types contributes data used for the evaluation of occupational hazards. These are discussed below.

To evaluate with some degree of confidence, the level of exposure at which the risk of health impairment is acceptable, a body of toxicological information is required. Five sources of data may be available to inform the occupational risk-assessment process.

• In vitro assays

• Animal toxicology studies

• Human challenge studies

• Case reports

• Epidemiology studies

In Vitro Assays

A number of useful in vitro assays have been developed over the past several decades in order to provide screening data and, in some cases, mechanistic insight without the need or expense of exposing animal or human subjects. While at this time there are few validated methods to determine complex toxicological responses such as immune hypersensitivity or peripheral neuropathy, there are validated and very useful screening assays. Notable examples are the Salmonella typhimurium reverse mutation assay, or Ames test, the Corrositex assay for dermal corrosivity potential of chemicals (Interagency Coordinating Committee on the Validation of Alternative Methods [ICCVAM], 2003; NIH-99-4495, 1999), the in vitro ocular toxicity test for identifying severe irritants and corrosives (ICCVAM, 2006), and use of three in vitro test methods for assessment of eye irritation of antimicrobial cleaning products (ICCVAM, 2010). In addition, quantitative structure–activity relationships can help suggest potential toxicological effects for an unstudied compound if structurally similar compounds have been evaluated.

Animal Toxicology Studies

Animal toxicology studies serve an important function in terms of identifying adverse effects, providing mechanistic data, establishing dose–response relationships, and aiding the process of establishing standards. Because animal studies can be conducted before there is any human exposure, these studies play an important role in hazard identification and prevention of human disease. There are numerous animal models for occupational injury and illness, which are described throughout this textbook in the context of the affected organ system and the classes of toxicants. Generation of animal toxicology data to predict health effects in workers is a central function of experimental toxicologists.

Toxicological investigations using animals often serve to establish a tentative acceptable exposure level. Other important information that may also be derived from these investigations concerns the relationships between the metabolic handling of the chemical and its interactions with target molecules (mechanism of action), identification of methods for biological monitoring of exposure and early health effects, and identification of preexisting pathological states that may increase susceptibility to the chemical. However, animal testing can provide only an estimate of the toxicity of a chemical for humans. Animals do not always respond to a chemical exposure in the same way as humans. For instance, there are very significant species and strain differences in responsiveness to aryl hydrocarbon receptor agonists, such as polychlorinated dibenzo-p-dioxins (Abnet et al., 1999; Boverhof et al., 2006). In some instances, interspecies differences in metabolism or mechanism of action cause certain chemicals to induce cancer in rodents, but not in humans. One such example is kidney cancer, attributable to the accumulation of a rat-specific protein (α_2u-globulin) in proximal tubular cells, and produced in male rats chronically exposed to unleaded gasoline (Hard et al., 1993). There are a few compounds for which predictive animal models have not been found. As discussed further, skin and internal cancers caused in humans by excessive oral exposure to inorganic arsenic were not reproduced during classic carcinogenicity studies in animals (Agency for Toxic Substances and Disease Registry [ATSDR], 2007).

Human Challenge Studies

Human challenge studies, or clinical exposure studies, are a useful approach for verifying findings from animal toxicology studies in humans and for establishing whether biotransformation pathways in the animal models represent those in exposed humans. Human challenge studies with occupational toxicants are usually designed to answer very specific questions regarding rates of uptake, biotransformation pathways, the time course of metabolite excretion during and after exposure, evaluation of the threshold concentration for sensory responses (odor, irritation of the nasal mucosa, etc), and acute effects of toxicant exposure on perception, vigilance, and function. Human challenge studies help to establish biomarkers of exposure. For reversible conditions, they can be useful for testing therapeutic options. They may also be useful for investigating bronchial hyperresponsiveness to inhaled agents. Extreme caution must be exercised to ensure the safety of research subjects. Idiosyncratic responses can cause a subject to be exceptionally sensitive. For inhalation studies, equipment malfunction can result in overexposure, so real-time exposure monitoring is a necessity. In the past 25 years, there have been several serious injuries and fatalities associated with human challenge studies. These have been attributed to hypersensitivity reactions and generally used mock workplace simulations without rigorous control and monitoring of exposures, as is the current standard of practice. Thus, such studies should be undertaken only when the required data cannot be obtained through other means and under circumstances in which the risk for volunteers can reasonably be estimated as negligible.

Case Reports

When new toxicants, new combinations of toxicants, or changes in process conditions occur in the workplace, a case or outbreak of cases can occur. These may be identified through workplace surveillance systems or through workers associating their disease with workplace exposures. In some cases, the problem is identified quickly and resolved, while others take years to resolve. These are often published as case reports and may give rise to animal or epidemiological studies. Hypersensitivity pneumonitis among machinists exposed to metalworking fluids contaminated with mycobacteria is one example (Thorne et al., 2006; Weiss et al., 2002). Another example is diacetyl-induced bronchiolitis obliterans that was first recognized in workers manufacturing artificially flavored microwave popcorn (Kreiss et al., 2002). NIOSH has a Health Hazard Evaluation (HHE) program in place in which employees or their authorized representatives or employers at a job site can request an investigation to evaluate a potentially hazardous situation. This program issues HHE Reports to disseminate information regarding the hazard. NIOSH also publishes NIOSH Alerts, Criteria Documents, Special Hazard Reviews, Occupational Hazard Assessments, and Current Intelligence Bulletins. In addition, NIOSH publishes Joint Occupational Health Documents with similar entities in foreign governments. While useful for hazard identification and mitigation, case reports and HHE Reports generally do not establish incidence or prevalence of diseases associated with an occupational hazard.

Epidemiology Studies

Epidemiology studies help to unravel the associations between occupational diseases, exposures, and personal risk factors. Exposure may be characterized using a surrogate measure such as job classification, or via questionnaire, or more directly through exposure monitoring or biomonitoring. Adverse effects may be expressed in terms of mortality, incidence or prevalence of clinical disease, irreversible or reversible functional changes, or critical biological changes.

Several types of epidemiological studies are used to gather data on the association of workplace exposures with human disease. Cross-sectional studies compare disease prevalence or health status between groups of workers classified according to job title, work site, or exposure status. Cohort studies compare exposed workers versus unexposed or less exposed workers either prospectively or retrospectively in order to associate the occurrence of disease with exposure. Because many occupational diseases have a long induction period or occur only rarely, prospective cohort studies may require a long time and need a large number of subjects to establish significant findings. Retrospective cohort studies can resolve the latency problem but require that relevant exposure data have been collected over time. In the absence of measured exposures, job titles, tasks routinely performed, and years of employment may allow exposure categorization (eg, low, medium, high). Exposure misclassification is frequently a problem in retrospective studies that reduces the likelihood of associating a disease with a particular exposure.

Case–control studies are useful for investigating rare diseases or diseases with long induction periods. As the name suggests, case–control studies compare workers with disease to workers without disease with regard to their past exposure intensity, frequency, and duration, plus other postulated risk factors. In some instances, where the exposure–disease relationship is not understood, it may be difficult to identify an appropriate control group. Case–control studies are strongest when accompanied by a rich data set of measured exposures to the candidate causative agents spanning the relevant exposure period. If exposure history is assessed by questionnaire after workers have developed adverse health effects, there arises the potential for recall bias in which those with disease may recall past exposures differently than those who are free of disease. Error can also arise due to selection bias if those who agree to participate in the study are not representative of the population of interest.

Characteristics of observational epidemiology studies are listed in Table 34-7. Occupational epidemiology studies assess relationships between exposures and human health outcomes and, therefore, are particularly useful for risk assessment. Confounding may arise due to exposures of risk factors not associated with the work environment.

Because measures of effect may be subtle and may overlay a background level of incidence, results generally require sophisticated statistical comparisons between a group of exposed workers and a similar group of workers without the exposure of interest. Ideally, the group of unexposed workers should be matched on variables such as age, race, gender, socioeconomic status, and smoking habits. They should also undergo the same standardized clinical, biological, or physiological evaluation at the same time as the exposed group. Comparison with the general population is ill-advised because an employed population is a highly selected group and may have a higher degree of physical fitness. Because occupational epidemiological studies often last for several years, all methods of investigation—such as questionnaires, measurement instruments, and analytic techniques—must be validated and standardized before the start of the study. If results will be compared to prior or contemporaneous studies, the same questionnaires should be used. The number of subjects under study should be chosen based on a sample size calculation to be able to detect a difference between exposed and unexposed subjects (should there be a difference) and should take into account labor turnover and those declining participation in any aspects of the study. If exposures are high enough to induce an adverse effect, it is expected that these studies may permit establishment of the relationship between integrated exposure (intensity × time) and frequency of abnormal results and, consequently, a redefinition of the OEL.

In most cases, occupational epidemiological studies encompass the collection of samples from the subjects or data obtained through interaction with the subjects. Because this often includes identifiable private information, the confidentiality of the data must be protected. United States, European Union, and international laws require that subjects must always have the right to refuse participation and investigators must have written consent from a duly informed volunteer obtained without coercion.

It is evident that certainty as to the complete safety of a chemical can never be obtained, regardless of the extent of toxicological investigations performed on animals. Nevertheless, animal studies provide valuable data from which to estimate the level of exposure at which the risk of health impairment is acceptable. Table 34-7 compares the information gained from animal studies to epidemiology studies. To the extent possible, animal studies should employ species for which the metabolic pathways and disease processes reflect those of humans. Guidelines and protocols for assessing experimentally the toxicological hazards of chemicals have been formulated by various national and international agencies. These tests include local and systemic acute toxicity tests, tests of toxicity following repeated exposure, investigations of metabolism and mechanism of action, short-term tests for detecting potential mutagens and carcinogens, studies of effect on reproduction and of teratogenic activity, chronic studies to detect carcinogenesis and other long-term effects, interaction studies, tests for immunosuppression, and dermal and pulmonary hypersensitivity tests. The need for performing these testing protocols should be carefully evaluated for the inclusion of any occupational toxicant to which workers will be exposed. In selecting the studies most appropriate for safety evaluation, the toxicologist should be guided by an understanding of the following:

• physicochemical properties of the chemical

• potential for the generation of toxic derivatives when the chemical is submitted to heat, pH changes, and UV light

• conditions of use and route of exposure

• type of exposure (continuous, intermittent, or incidental)

• degree of exposure

Toxicological information already available on other chemicals with similar chemical structure and reactive chemical groups can suggest potential hazards and reactivity.

Conclusions drawn from any toxicological investigation are useful only if the composition and physical state of the tested preparation is known. This would include the nature and concentration of impurities or degradation products, speciation of inorganic compounds, characterization of physicochemical properties for inhaled materials, and characterization of the vehicle (the carrier, diluent or excipient of the toxicant). Sensitive and specific methods of analysis of the chemical in solution, air, and biological material should also be available. The assessment of the toxicity of malathion illustrates this point. Malathion is an organophosphate insecticide that normally has relatively low human toxicity. This pesticide was responsible for an episode of mass poisoning among malaria workers in Pakistan because the specific product contained impurities (mainly isomalathion) capable of inhibiting tissue and plasma carboxyesterases (Aldridge et al., 1979) and inactivating acetylcholinesterase via formation of adducts (Doorn et al., 2003). The toxicity evaluation for malathion had not anticipated isomalathion coexposure. A recent study evaluated the pulmonary toxicity of inhaled aluminum nanowhiskers (Adamcakova-Dodd et al., 2012). The manufacturer of this material stated they were Al₂O₃ whiskers. Careful physicochemical analysis prior to toxicity evaluation using x-ray photoelectron spectroscopy and transmission electron microscopy revealed they were actually mixed phase aluminum hydroxy structures containing Al(OH)₃ and λ-AlOOH. Further, the whiskers were 100 to 400 nm in length, considerably shorter than the 2800 nm length claimed by the manufacturer.

The duration of tests necessary to establish an acceptable level for occupational exposure is primarily a function of the type of toxic action suspected. It is generally recognized that for systemically acting chemicals, subacute and short-term toxicity studies are usually insufficient for proposing OELs. Subacute and short-term toxicity tests are usually performed to find out whether the compound exhibits immunotoxic properties and cumulative characteristics. They also aid in selection of the doses for long-term-exposure studies and the kind of tests that may be most informative when applied during long-term exposures. A number of studies have drawn attention to the fact that the reproductive system may also be the target organ of industrial chemicals (eg, glycol ethers, styrene, lead, dibromochloropropane). Thus, studies designed to evaluate reproductive effects and teratogenicity should also be considered during routine toxicological testing of occupational toxicants.

Information derived from exposure routes similar to those experienced by workers is clearly the most relevant. For airborne pollutants, inhalation exposure studies provide the basic data on which provisional OELs are based. Experimental methodology is much more complicated for inhalation studies than for oral administration experiments and requires more specialized equipment and expertise (Thorne, 2000). For example, in the case of exposure to an aerosol, particle size distribution must be evaluated, and the degree of retention in the respiratory tract of the animal species under study should be established. Ideally, particle size should be selected according to the deposition pattern of dry or liquid aerosols in the particular animal species used in order to represent human lung deposition with occupational exposures. Particle deposition and retention curves have been published for human, monkey, dog, guinea pig, rat, and mouse (Asgharian et al., 2003; Hsieh et al., 1999; Schlesinger, 1985). Recent research coupling asymmetric multiple-path models of the bronchial tree and ventilation parameters can provide more accurate prediction of site-specific particle deposition (Asgharian and Price, 2006). It should also be kept in mind that the concentration of the material in the air and the duration of exposure do not give a direct estimate of the dose, because retained dose is also dependent on the minute volume and the proportion of inhaled particles retained. Measurement of pulmonary dust retention following exposure to a radiolabeled or fluorescently tagged test aerosol should be performed prior to conducting acute, subchronic, or chronic studies. This allows one to assess deposition and determine whether the selected levels of exposure may overwhelm pulmonary clearance mechanisms (ILSI, 2000; Oberdörster, 2002).

The choice of studies to perform and their routes of administration must be evaluated scientifically for each toxicant. Important considerations include its target sites and mechanism of action, metabolism, the nature of its adverse effects, and how workers are exposed to the toxicant. The morphological, physiological, and biological parameters that are usually evaluated, either at regular intervals in the course of the exposure period or at its termination, are described in Units IV and V of this text. Investigations that can make use of specific physiological or biochemical tests, based on knowledge of the principal target organ or function, produce highly valuable information and increase confidence in the OEL derived from them.

The primary objective of worker health surveillance programs is to provide both periodic screening of general health and wellness plus health and exposure monitoring tailored to recognized hazards of the workplace. The monitoring of exposures to toxicants in the workplace may play an important role in detecting excessive exposures before the occurrence of significant biological disturbances and health impairment. A scheme for biological monitoring of exposure and of early biological effects is possible only when sufficient toxicological information has been gathered from in vitro, animal, or human studies on the mechanism of action and the metabolism of xenobiotics to which workers are exposed. When a new chemical is being used on a large scale, the careful clinical surveillance of workers and monitoring of workplaces should be instituted in order to address three aims: (1) to identify overexposure or adverse effects on the health of the workers and quickly intervene, (2) to evaluate the validity of an existing or proposed OEL, and (3) to test the validity of a proposed method for biological monitoring.

Evaluation of the validity of the proposed OEL derived from animal experiments through workplace surveillance is the major aim because studies and observations on humans are the final basis for deciding whether an OEL set originally on the basis of animal toxicity testing is truly acceptable as one that will not produce excess health risks. This means that sensitive clinical, biochemical, physiological, or behavioral tests for detecting an adverse effect of a toxicant should ideally be performed on the workers concurrent with exposure assessment. It is helpful if health surveillance programs can include the same biomarkers as used in prior animal or human exposure studies. Occupational toxicologists and occupational physicians cannot rely solely on the standard diagnostic tools used in clinical medicine, as they were established primarily to reveal advanced pathological states and not to detect early adverse effects at a stage when they are still reversible. For example, the measurement of serum creatinine is still a widely used clinical test for assessing renal integrity, yet it is known that the glomerular filtration rate of the kidney must be reduced by more than 50% before serum creatinine rises significantly.

The main limitation of current OELs or BEIs is that some are based on limited experimental data or clinical studies in which only late effects have been investigated and correlated with past exposure. Furthermore, several BEIs are derived from the study of external–internal exposure relationships and not from relationships between internal dose and early adverse effects. The validity of an OEL is much stronger if it is based on the study of dose–response relationships in which the dose is expressed in terms of the cumulative target dose and the monitored effect reflects a critical biological event. However, for some chemicals and some adverse effects (eg, induction of hypersensitivity and possibly genotoxic effects), the frequency of peak exposure may be more important for health risk assessment than the integrated dose. For example, long-term low-level exposures to commercial enzymes rarely induce sensitization. However, a single exposure to a high concentration can produce hypersensitivity and occupational asthma.

In cases where a surveillance program was not instituted before the introduction of a new chemical, it is more difficult to establish the efficacy of the OEL. In this situation, evaluation depends on retrospective cohort studies or case–control studies on workers who have already sustained exposure. Evaluation of a “no observed adverse effect level” (NOAEL) is difficult because information on past exposures is often incomplete, and frank effects are generally the focus of retrospective or case–control studies. Provided that a satisfactory assessment of past exposure is possible, cross-sectional studies that rely on preclinical signs of toxicity may, to a certain extent, overcome these difficulties. Whether or not clinical investigations are planned from the introduction of a new chemical or process, it is essential to keep standardized records of occupational histories and exposure. The need may arise for mortality or case history studies in order to answer an urgent question on a suspected risk.

Careful investigation of overexposures resulting from specific incidents such as containment breaches, chemical spills, or vessel or pipe ruptures can provide useful information. Although, such observations are usually not helpful for determining the NOAEL in humans, they may indicate whether human symptomatology is consistent with medical signs found in animals and may suggest functional or biological tests that might prove useful for routine monitoring of exposed workers.

Linkage of Animal Studies and Epidemiological Studies

In the field of occupational toxicology, perhaps more than in other areas of toxicology, cooperation between those conducting animal studies and studies of workers is essential for examining risks associated with overexposure to chemicals and other toxicants. A few examples will serve to illustrate the complementarity of these disciplines.

Several occupational carcinogens have been identified clearly through combined epidemiological and experimental approaches. For example, the carcinogenicity of vinyl chloride was first demonstrated in rats (Viola et al., 1971), and a few years later, epidemiological studies confirmed the same carcinogenic risk for humans (Creech and Johnson, 1974; Monson et al., 1974). This observation stimulated several investigations on the metabolism of vinyl chloride in animals and on its mutagenic activity in in vitro systems. Identification of vinyl chloride metabolites led to the conclusion that there is microsomal oxidation leading to the formation of an epoxide derivative, which acts as a proximate carcinogen (ATSDR, 2006). This finding triggered further studies on the biotransformation of structurally related halogenated ethylenes, such as vinyl bromide, vinylidene chloride, 1,2-dichloroethene, trichloroethylene, and perchloroethylene. Comparison of their oncogenic activity in relation to their metabolism suggested that an interplay between the stability and reactivity in reaching the DNA target and reacting with it after being formed would determine their genotoxic risk. It is now recognized that vinyl chloride is metabolized to 2-chloroethylene oxide, which rearranges to 2-chloroacetaldehyde and produces promutagenic etheno-DNA adducts including etheno-guanine, etheno-cytosine, and lesser amounts of etheno-adenine (ATSDR 2006; Bolt, 2005; National Toxicology Program [NTP], 2011).

1,3-Butadiene is a known human carcinogen with more than 75% of its use in the manufacture of synthetic rubber products. Experimental studies in rats and mice demonstrated carcinogenicity, with mice being particularly sensitive. Subsequent to these findings, 1,3-butadiene was shown to follow the same metabolic pathway in humans as in rats and mice, forming mutagenic and carcinogenic epoxides. That led to cohort and case–control studies establishing 1,3-butadiene as a human carcinogen (NTP, 2011). Recent studies have considered the relative importance of the metabolic pathway leading to formation of the reactive metabolites, 3,4-epoxy-1-butene, 1,2:3,4-diepoxybutane, and 3,4-epoxy-1,2-diol, which react with DNA and with nuclear proteins. Under the assumption of a genotoxic mechanism and cross-species comparisons of DNA binding, these data facilitate a more informed cancer risk assessment. In vitro studies showed that human lymphocytes were less sensitive to 1,2:3,4-diepoxybutane-induced chromosomal abberations than mouse or rat cells (ATSDR, 2009). In vivo studies revealed a higher rate of formation of epoxides in mice than rats. Inhaled 1,3-butadiene produced N7-guanine DNA adducts in lung tissue due to 3,4-epoxy-1,2-diol (Koivisto and Peltonen, 2001) and 1,2:3,4-diepoxybutane (Goggin et al., 2009). These cross-links were more prevalent in mice than rats, and females than males, which agreed with the enhanced susceptibility to cancers of female mice. On this basis, rats were judged to be a more predictive species than mice for human risk assessment. This work illustrates that the measurement of DNA adducts is a sensitive method for monitoring 1,3-butadiene metabolism via the epoxide-forming pathway in workers.

In 1973, an outbreak of peripheral neuropathy occurred in workers exposed to the solvent methyl butyl ketone (2-hexanone, MBK) (McDonough, 1974; Allen et al., 1975). The same lesion was reproduced in animals (Mendell et al., 1974; Spencer et al., 1975). Bio-transformation studies were then undertaken in rats and guinea pigs, and some MBK metabolites (2,5-hexanedione, 5-hydroxy-2-hexanone) were also found to possess neurotoxic activity (Spencer and Schaumburg, 1975; DiVincenzo et al., 1976, 1977). Similar oxidation products are formed from n-hexane, the neurotoxicity of which is probably due to the same active metabolite as that produced from MBK. Because methyl ethyl ketone (MEK) cannot give rise to 2,5-hexanedione, it was suggested as a replacement solvent. Recent work has demonstrated the potentiation of neurotoxic effects of n-hexane upon coexposure with methyl ethyl ketone (Yu et al., 2002). Nevertheless, MEK remains a much less toxic solvent than MBK.

These examples demonstrate that studies of the metabolic handling of occupational toxicants in animals are instrumental in the characterization of reactive intermediates, and may suggest unsuspected risks or indicate new methods of biological monitoring. Conversely, clinical observations of workers may stimulate studies of the metabolism or the mechanism of toxicity of a toxicant in animals, thereby revealing the health significance of a biological disturbance.

Arsenic is one of the very few compounds for which there are limited data of predictive value from animal studies to human health effects. Arsenic has been used as a medicine since the time of Hippocrates. Initially used to treat ulcers, arsenicals achieved notoriety as medicinals for a wide variety of ailments, and then, in the first half of the 20th century, for the treatment of syphilis and parasites. Many foods and beverages contaminated with arsenic have been associated with accidental and intentional poisonings. Inorganic pentavalent arsenic (arsenate) is readily absorbed across tissues and converted to the trivalent form (arsenite). This is then methylated to form monomethyl arsenic acid and dimethyl arsenic acid (ATSDR, 2007). These are primarily transported in the blood bound to sulfhydryl groups in proteins. The half-life in humans for arsenic compounds is two to four days and the major excretion is via urine (Nriagu, 1994).

Inorganic arsenic was first noted as a human carcinogen by Hutchinson in 1887 (Hutchinson, 1887). Epidemiological studies led to classification of arsenic by the IARC as a skin and lung carcinogen in 1980 (IARC, 1980). Since then studies among occupationally exposed populations and populations with high arsenic in their drinking water have shown conclusively, that arsenic causes human cancers of the skin, lung, bladder, kidney, liver, nasal tissue, and prostate. There is also evidence for arsenic-associated cutaneous effects, cardiovascular and cerebrovascular disease, diabetes mellitus, and adverse reproductive outcomes (EPA, 2000; WHO, 2001).

A large number of carefully executed cancer bioassays in mice, rats, beagles, and monkeys have been performed using sodium arsenate, sodium arsenite, lead arsenite, arsenic trioxide, and dimethylarsinic acid. These studies have been uniformly negative for cancer. A number of subsequent studies that tested for tumor-promotion activity following dosing with recognized tumor initiators also yielded negative results. The ATSDR Toxicological Profile for arsenic states, “First and most importantly, no animal model exists for the health effect of greatest concern for human exposure: carcinogenicity in skin and other organs after oral exposure” (ATSDR, 2007). Recent studies using transgenic mice, high-dose in utero exposures, or administration of a cocarcinogen have yielded tumors in mice (Hughes, 2006). However, negative results in standard animal carcinogenicity screens have been problematic in the face of unquestionable oncogenic activity in humans.

The examples above demonstrate that the occupational toxicologist cannot rely solely on animal or epidemiological studies. A combined approach is necessary in order to identify, elucidate, and prioritize risks, and to develop interventions and techniques for worker health surveillance.

Two important applications of occupational toxicological investigations are compared below: environmental monitoring and biological monitoring. As described above under “Occupational Health Standards,” both are important in worker health surveillance, and are essential elements of toxicology studies with dosing via the inhalation or dermal routes.

Environmental Monitoring for Exposure Assessment

An important objective of experimental and clinical investigations in occupational toxicology is the proposal of safe levels of exposure. OELs must be reevaluated at regular intervals as new information on the toxicity of industrial chemicals develops. Adherence to OELs may not protect everyone and, therefore, cannot supplant close medical surveillance of workers. Various private and official institutions regularly review the toxicological information on chemicals in order to propose or update permissible levels of exposure. These include governmental organizations worldwide and professional organizations such as the ACGIH. A critical element of establishing OELs is the accurate and uniform assessment of exposure. Methodology for exposure assessment must be specifically tailored to the agent under study and the environment in which it appears. To assess airborne exposures for compliance purposes, personal samples taken in the breathing zone are generally used. In a few specific environments, area samples form the basis of an exposure standard (eg, the OSHA standard for exposure to raw cotton dust specifies use of the vertical elutriator or an equivalent method). Occupational environmental surveys may employ area sampling to determine areas with higher or lower toxicant concentrations. However, concentrations determined from personal samples typically exceed area concentrations depending on the work practices and environmental controls. For example, geometric mean concentrations of inhalable dust assessed from 159 personal samples in dairy barns were 1.78 mg/m³, compared with 0.74 mg/m³ for 252 area samples collected simultaneously in two locations in the same barns (Kullman et al., 1998). Thus, in this environment, area sampling alone would underestimate personal exposures by a factor of 2.4.

Repeated random sampling is theoretically the best approach to developing unbiased measures of exposure. However, this is rarely the approach that is taken. Variability in exposure, especially variability over time, is often large; therefore, a considerable number of repeated measurements are needed to obtain an accurate proxy of the true exposure. When the number of repeats is insufficient, the slope of the exposure–response relationship will be biased, usually leading to considerable underestimation of the relationship (Heederik and Attfield, 2000). Recent studies have demonstrated that approaches assessing exposures to groups rather than to individuals are more efficient in terms of measurement effort for obtaining a desired level of accuracy (Vermeulen and Kromhout, 2005). In a group-based approach, workers are grouped by job title, task performed, or through exposure modeling studies to elucidate determinants of exposure, and the group mean is used as the average exposure for each worker (Kromhout et al., 1996). Further statistical modeling of the exposure data can reduce problems of bias and large temporal and spatial variability (Preller et al., 1995; Tielemans et al., 1998). Whereas this approach is gaining acceptance among occupational epidemiologists for evaluating exposure–response data and assessing risks, it is not accepted for compliance monitoring.

Although, one cannot assess dose directly through exposure monitoring, it does have several distinct advantages over biomonitoring. Exposure monitoring allows one to quantify workplace exposure by route through selective air monitoring in the breathing zone of the worker and dermal dosimetry using absorptive material affixed to the workers' skin or clothing. Biomonitoring generally does not provide route-specific exposure data. Environmental monitoring techniques are generally less expensive and less invasive than techniques involving the collection and analysis of biological samples such as blood or urine. Thus, a larger population of workers can be studied for the same amount of money. New personal sampling devices incorporate global positioning system, accelerometers, and cellphone technology to provide enhanced data on location and activity. Workers are accustomed to wearing personal samplers for exposure assessment and are generally quite willing to do so. However, they are often unwilling to give a blood or urine sample, fearing that the sample will be surreptitiously used for drug testing, DNA testing, or experimentation. Another benefit of air sampling in the workplace is that spatial, temporal, and work practice associations can be established, and can suggest better interventions and engineering controls to reduce exposures. Finally, analytic interferences and variabilities are generally lower with environmental samples than with biological samples.

A fully validated sampling and analysis method requires specification of the sampling methods; sample duration, handling, and storage procedures; the analytic method and measurement technique; the range, precision, accuracy, bias, and limits of detection; quality assurance issues; and known interferences. It is also important to document intralaboratory and interlaboratory variability. Once a standard method is established, it must be closely followed in every detail in order to assure consistency of results.

The development of accurate and precise analytic methods for environmental assessment is an ongoing effort. NIOSH publishes the extensive NIOSH Manual of Analytical Methods (NIOSH, NMAM, 2011) and these are widely used. ASTM International has also developed a rigorous system for the establishment of methods. It generally requires five or more years to establish a new ASTM method for exposure assessment. The International Organization for Standardization (ISO) is a global federation of national standards bodies with 162 member countries. The subcommittee on workplace atmospheres is administered by the American National Standards Institute (ANSI). ISO has completed harmonization on a number of air sampling methods—for example, the determination of the number concentration of airborne inorganic fibers by phase contrast optical microscopy. The American Industrial Hygiene Association (AIHA) and the ACGIH publish compilations with descriptions of analytic devices and methodology (Anna, 2010; Cohen and McCammon, 2001). Methods have also been developed for bioaerosol exposure assessment and these have been reviewed (Heederik et al., 2003; Thorne and Heederik, 1999a).

Biological Monitoring for Exposure Assessment

Biomonitoring consists of the measurement of toxicants, their metabolites or molecular signatures of effect in specimens from humans or animals, including urine, blood, feces, exhaled breath, hair, finger or toenails, bronchial lavage, breast milk, and adipose tissue. These may serve as biomarkers of exposure, biological effect, or susceptibility. New technologies are emerging that will allow measurement and monitoring of chemicals in the body and transmission of the data from indwelling biosensors. Biomonitoring data provide a measurement of exposure based upon internalized dose and, thus, account for all exposures by all routes for the assessed analyte.

Depending on the chemical and the analyzed biological material, the term internalized dose may have different meanings. The measured biomarker may reflect the amount of chemical absorbed shortly before sample collection, as with the concentration of a solvent in exhaled air or in a blood sample obtained during the work shift. It may reflect exposure during the preceding day, as with the measurement of a metabolite in blood or urine collected after the end of exposure. For toxicants with a long biological half-life, the measured parameter may reflect exposure accumulated over a period of weeks or months, as with arsenic in toenails. Internal dose may refer to the amount of chemical stored in one or in several body compartments or in the whole body (the body burden).

When biological measurements are available to assess the internal dose, they offer important advantages over monitoring the air of the workplace or measuring deposition of chemicals onto dermal patches. The greatest advantage is that the biological measure of exposure is more directly related to the adverse health effects than environmental measurements, because it reflects the amount of toxicant absorbed. Therefore, it may offer a better estimate of the risk than can be determined from ambient monitoring. Biological monitoring accounts for uptake by all exposure routes. Many industrial chemicals can enter the organism by absorption through the skin or the gastrointestinal tract, as well as the lung. For example, some solvents (eg, dimethylformamide) and many pesticide formulations exhibit substantial exposure via the dermal route. In these situations, exposures determined through monitoring airborne concentrations alone underestimate true exposure. Monitoring of an early biomarker of effect is useful when the subjects are exposed to a complex mixture and the agents responsible for the effect are unknown or several agents are working in synergy.

Several factors can influence uptake. Personal hygiene habits vary from one person to another, and there is some degree of individual variation in the absorption rate of a chemical through the lungs, skin, or gastrointestinal tract. Use of size-selective air sampling to determine the inhalable or respirable fraction can strengthen the exposure estimate. However, biological factors such as ventilatory parameters can affect the strength of such a correlation because increased workload can markedly increase the respiratory uptake of an airborne toxicant. Because of its ability to encompass and evaluate the overall exposure (whatever the route of entry), biological monitoring can also be used to test the overall efficacy of personal protective equipment such as respirators, gloves, barrier creams, or aprons. Another consideration with biological monitoring is the fact that nonoccupational exposures (through hobbies, residential exposures, dietary habits, smoking, second jobs) may also be expressed in the biological sample. The organism integrates the total external (environmental and occupational) exposure into one internal load. Whereas this is beneficial for worker health and safety, it may be confounding in epidemiological studies or compliance monitoring. Thus, while biomonitoring is an important exposure measurement tool for health risk assessment, it generally does not allow one to relate sources and levels of exposure to adverse health effects (Albertini et al., 2006).

The value of biological monitoring is heightened when the relationships between external exposure, internal dose, and adverse effects are established. Normally, biological monitoring of exposure cannot be used for assessing exposure to substances that exhibit their toxic effects at the sites of first contact and are poorly absorbed. Examples include dermally corrosive compounds and primary lung irritants. In this situation, the only useful relationship is that between external exposure and the intensity of the local effects.

Relationships between air monitoring and biological monitoring may be modified by genetic or external factors that influence the fate of an occupational toxicant in vivo. Metabolic interactions can occur when workers are exposed simultaneously to chemicals that are biotransformed through identical pathways. Exposure to chemicals that modify the activity of the biotransformation enzymes (eg, microsomal enzyme inducers or inhibitors) may also influence the fate of another compound. Furthermore, metabolic interferences may occur between occupational toxicants and alcohol, tobacco, food additives, prescription drugs, natural product remedies, or recreational drugs. Changes in any of several biological variables (weight, body mass, pregnancy, diseases, immune status, etc) may modify the metabolism of an occupational chemical. These factors have to be taken into consideration when the results of biomonitoring are interpreted. Whatever the parameter measured, whether it is the substance itself, its metabolite, or an early biomarker of effect, the test must be sufficiently sensitive and specific to provide meaningful data in the range of workplace exposures.

Some chemicals have a long biological half-life in various body compartments (eg, hydrophobic persistent organic pollutants), and the time of sampling may not be critical. For other chemicals, the time of sampling is critical because, following exposure, the compounds or their metabolites may be rapidly eliminated from the body. In these cases, the biological sample is usually collected during exposure, at the end of the exposure period, or sometimes just before the next work shift. When biological monitoring consists of sampling and analysis of urine, it is usually performed on “spot” urine specimens or on the first morning void. It is a standard practice for most organic compounds to adjust the results for urine output by expressing the results per gram of creatinine in the urine. Analyses performed on very dilute urine samples are not reliable. The WHO has specified acceptable limits for urine specimens of between 0.3 and 3.0 g/L creatinine or 1.010 and 1.030 specific gravity (ACGIH, 2010). When samples exhibit large interindividual variability or high “background” levels, the interpretation of a single measurement may be difficult. In such cases, it may be useful to analyze biological material collected before and after the exposure period and gauge exposure based upon the cross-shift change.

The majority of BEIs listed by the ACGIH refer to analysis of the parent compound or its phase I metabolite. Analytical advances over the past decade have yielded methods for detecting reactive intermediates of metabolism and macromolecular adducts that may induce mutations or cell cycle disruption. These methods are important for comparing human and animal toxicity data for risk assessment and may help to explain differential responses in animal models and susceptibility in human populations. These methods are also useful for occupational health surveillance programs.

Environmental monitoring plays an important role in the evaluation and prevention of excessive exposure to toxicants in the workplace. However, the prevention of acute toxic effects on the respiratory tract, skin, or eye mucosa can only be achieved by keeping the concentration of the irritant substance below a certain level or by eliminating the exposure. Local acute effects of chemicals do not lend themselves to a biological surveillance program. Likewise, biological monitoring is usually not indicated for detecting peak exposure to dangerous chemicals such as arsine (AsH₃), carbon monoxide, or prussic acid (HCN). Furthermore, identification of emission sources and the evaluation of the efficiency of engineering control measures are usually best performed by ambient air analysis. Table 34-8 lists the approaches most useful for controlling inhalation exposures in the workplace. These include process changes, engineering controls, and use of personal protective equipment. It should be emphasized that process changes and application of engineering controls are preferable to reliance on personal protective equipment.

In summary, environmental and biological monitoring should not be regarded as opposites but as complementary elements of an occupational health and safety program. They should be integrated as much as possible to ensure low levels of contaminants and optimal health for workers.

The working environment will always have the potential to overexpose workers to various toxicants. Recognition of these risks should not wait until epidemiological studies have uncovered hazardous levels. A combined experimental, clinical, and epidemiological approach is most effective for evaluating and managing the potential risks. One can then promulgate scientifically based occupational health standards, apply effective workplace controls to ensure adherence to those standards, and institute worker health surveillance programs to identify unexpected effects in susceptible individuals.