Occupational Asthma - technical

Article about occupational asthma

Respirable agents in the workplace are responsible for about 10% of asthma arising in adult life, and for an annual incidence of occupational asthma (OA) of 25–100 per million employed in industrially developed countries. For 5–10% of cases, toxic mechanisms are responsible, and asthma is the result of accidents that release agents such as chlorine, sulfur dioxide, and acetic acid into occupational environments. For the remaining 90–95%, asthma appears to arise through hypersensitivity mechanisms. Many of the several hundred causal ‘asthmagens’ are reactive chemicals of low molecular weight, though some are naturally occurring allergens of high molecular weight. Some agents (e.g., di-isocyanates, epoxy resins, flour) have such sensitizing potency that at current exposure levels in some occupations (e.g., spray painting and baking) the risk of OA exceeds that of asthma arising spontaneously. OA is thus important in an epidemiologic sense, and it serves as a useful model of asthma in general.

Asthma of occupational origin is like asthma resulting from any other cause. However, when due to hypersensitivity mechanisms, it has one unusual characteristic. If the diagnosis is recognized within 6–24 months and exposure ceases, there is a meaningful possibility that active disease will resolve. Therefore, the onus is on physicians to consider the possibility of an occupational cause in every adult presenting with asthma.


It was not until the twentieth century that investigatory techniques acquired the sophistication to distinguish airway disorders from those of the lung parenchyma/interstitium. The definition of asthma, the elucidation of its clinical characteristics and pathogenic mechanisms, and the evolution of strategies for its recognition, management, and prevention have consequently been comparatively recent events. Nevertheless, it is likely that asthma was the explanation for many of the respiratory disorders of antiquity and the middle ages that were recognized to afflict (and sometimes devastate) workers in certain trades and industries. The realization that asthma may arise as a direct consequence of inhaled occupational agents has been the focus of particular attention over the last three to four decades, and our understanding of occupational asthma (OA) has largely arisen during this period.

Occupational asthma has no recognized differences from asthma in general with respect to its pathology and genetic etiology, and so this article will focus on its features of special interest.

Definitions and Classification

Asthma is a disease of the intrathoracic airways that is characterized, and often defined, by its means of clinical expression (diffuse airway obstruction that varies in degree over time) and its underlying pathogenic basis (a state of enhanced airway responsiveness).

Occupational asthma is caused by exposure to agents, almost exclusively airborne, that are encountered primarily in the workplace. Such exposure in susceptible individuals causes the level of airway responsiveness to increase into the asthmatic range. Two distinct mechanisms are recognized. First, hypersensitivity mechanisms appear responsible since there is a latency period of presumed sensitization between exposure onset and symptom onset; this accounts for 90–95% of cases. Second, acute toxicity is the initiating event; this is generally a consequence of inhalation of toxic agents during an industrial accident. Asthma following the latter is sometimes identified as the reactive airways dysfunction syndrome (RADS) or, more simply, irritant asthma. Neither term is fully satisfactory. The first may be interpreted, incorrectly, to indicate a disorder other than asthma, while the second may be mistaken for pre-existing asthma that is exacerbated non-specifically at work (e.g., because of exertion in cold air); this is ‘work-aggravated’ not ‘occupational’ asthma, and is not associated with an increase in airway responsiveness.

It is currently unclear whether repeated exposures to toxic agents at dose levels insufficient to provoke clinical reactions might nevertheless cause minor increases in airway responsiveness. If so, the cumulative effect might elevate airway responsiveness to a level at which asthma eventually becomes inevitable (‘low-level RADS’). This would simulate the latency period association with presumed hypersensitivity mechanisms, and ongoing exposure might then provoke acute symptoms in a non-specific fashion. This would appear to simulate hypersensitivity mechanisms also, but such exposure should provoke reactions in any asthmatic individual with a sufficient level of airway responsiveness.

Key Clinical Features

When hypersensitivity mechanisms are responsible, further exposures to the particular inducing asthmagen may provoke specific asthmatic reactions. Symptoms are dependent on the degree of hypersensitivity, the magnitude of the provoking stimulus, the level of airway responsiveness, and the ease with which the affected individual perceives changing respiratory sensations.

Airway responsiveness varies in level from subject to subject over a considerable range, and can be quantified throughout the population at large. Its distribution (like that of most biologic parameters) is unimodal, and it follows the common biologic pattern of a ‘bell-shaped’ curve. The tail in which responsiveness is most marked gives rise to asthma, but the adjacent segment implies vulnerability, and the opposite tail implies comparative impunity. It is misleading to think of asthmatic and nonasthmatic subjects being fundamentally different because of the presence or absence of airway hyperresponsiveness; the issue, rather, is whether an individual’s level of airway responsiveness is sufficiently high to make meaningful bronchoconstriction likely when there are appropriate provoking stimuli.

Whether the resulting degree of bronchoconstriction is perceived to be distressing (or is perceived at all), is very dependent on psychological factors. This adds an important further level of complexity and variability, since at all degrees of asthmatic severity as defined physiologically, there will be a wide spectrum of perceived disability. This is particularly so in OA because of resentment, even anger, over the possible liability of a third party (the employer), and the possibility of compensation.

Longstanding occupational asthma, like asthma generally, is commonly associated with airway obstruction that has a reduced capability for reversal. It may come to simulate chronic obstructive pulmonary disease (COPD).


Over recent decades, occupational asthma has proved consistently to be the commonest type of newly diagnosed occupational lung disease in industrially developed countries, though the various disorders attributable to asbestos have an equal cumulative incidence. Between them, asthma and asbestos account for 65–70% of all incident respiratory disorders of occupational origin. In most outbreaks of occupational asthma no more than a few per cent of exposed workers become affected, but there are examples with both pathogenic pathways of prevalences approaching 50%.

The reported incidence of occupational asthma varies widely at a global level, depending on local employment patterns and diagnostic criteria. Estimates from statutory notification schemes, compensation registers, voluntary surveillance schemes, and general population surveys, suggest that 5–15% (median 9–10%) of asthma beginning in adult life is occupational. When asthma arises for the first time in a working adult, the background odds favoring an occupational cause over a nonoccupational cause are consequently of the order 1 in 10 only. This is consistent with the estimate from SWORD (Surveillance of Work-Related and Occupational Respiratory Disease) data that about 40 cases of occupational asthma per million employed workers arise in the UK each year.

SWORD has usefully considered the incidence of new cases within specific working groups for which there are particular occupational exposures. For example, among spray painters, who may use asthmagenic di-isocyanate, epoxy resin, and acrylic paints, the average annual incidence of occupational asthma from 1992 to 1997 was 1464 per million – more than threefold the UK national average for all cases of incident asthma.

Thus, in the case of spray painters developing asthma in adult life, the background odds favor an occupational cause over a coincidental cause. Other settings associated similarly with greater than even odds include baking, metal treatment, chemical processing, and plastics manufacture. Most national estimates range from 25 to 100 million per year.


Many occupational agents (some 400) have been reported to be definite or probable sensitizers capable of inducing asthma. Some of the most prominent are classified in Table 1, and the most common reports to SWORD over a 9-year period are listed in Table 2. Notable agents reported to cause RADS over recent years have been acetic acid, chlorine/chlorine dioxide/ hydrochloric acid, di-isocyanates, dinitrogen tetroxide, endotoxin, fire smoke, freons, hydrobromic acid, Iraq/Iran war gases, pentamidine, phosphoric acid, silo and swine confinement gases, sulfur dioxide/ sulfuric acid, tear gas, and welding fume.

Clinical Features

Once a diagnosis of asthma is suspected from the clinical history or physical examination and confirmed objectively (by the demonstration of reversible airway obstruction or the measurement of airway responsiveness), the diagnostic issue turns to whether it has arisen occupationally.

If the toxicity pathway has provided the mechanism, the diagnosis is clear and needs no further investigation. Asthma becomes evident during the recovery phase from what is usually a combination of conjunctivitis, rhinitis, pharyngolaryngitis, tracheobronchitis, and (possibly) pneumonitis induced by a major exposure to a toxic chemical or organic dust. In survivors, full recovery is the rule. If asthma arises, it is generally the only persisting respiratory problem, though occasionally bronchiectasis or an obliterative bronchiolitis is also detectable.

If hypersensitivity has provided the mechanism, the diagnosis is more challenging and may be extremely difficult. This is partly because the latency period during which sensitization occurs (usually 3–24 months) may be very short (days or weeks) or very prolonged (several years), and partly because in most working environments associated with OA, most cases of incident asthma are coincidental and quite unrelated to occupation. What follows addresses this diagnostically more challenging variety of occupational asthma.

Clinical History

The history provides an obvious starting point, but may be importantly distorted. Affected workers anxious to remain employed may deny or minimize relevant symptoms; they may also exaggerate or falsify critical aspects. An overwhelming belief that occupational exposure (and/or employer negligence) is responsible may, curiously, make a diagnosis of OA the ‘independent variable’ on which the symptoms depend: ‘‘if improvement during holidays/vacations is a cardinal diagnostic feature of occupational asthma then, yes, this must be so in my case since I know I have occupational asthma.’’

For a classical case, there is exposure to a known asthmagen and other exposed workers are affected. For the affected individual, there is recognition that symptoms arise or worsen after a minimum of 1–2 hours and a maximum of 8 h from the onset of each sufficiently strong exposure, and persist for hours or days. Such ‘late’ asthmatic reactions are characteristically associated with increases in airway responsiveness and so are much more definitive of OA than ‘immediate’ reactions, which may simply result non-specifically through ‘irritant’ mechanisms.

Mild late reactions may resolve within hours so that there is full recovery by the following day, but more commonly the rise in airway responsiveness is sufficient to worsen asthmatic severity for several days. A weekend away from work may therefore be insufficient to allow full recovery, and the association between occupational exposure and symptoms may not be recognized until there is a 2-week period of vacation (or sick leave). Even then gradual recovery, particularly cessation of disturbed sleep, may not become obvious until the second week, only to be reversed within a matter of days of returning to work.

Serologic Investigations

Laboratory investigation for diagnostic IgE antibodies to relevant asthmagenic agents has proved disappointing for two reasons. First, many occupational asthmagens are reactive chemicals of low molecular weight. They are not thought to act as sensitizers until coupled with appropriate haptogenic body proteins, and it has proved difficult to produce suitable complexes for antibody detection. Second, exposed individuals may develop IgE responses without any apparent ill effect. Antibodies appear to correlate more closely with exposure than disease. Nevertheless, when radioallergosorbant allergen testing to relevant asthmagens is available, positive tests increase the probability of OA, even if sensitivity and specificity are limited.

Peak Expiratory Flow

More popular and more readily available is peak expiratory flow (PEF) monitoring. Test subjects take measurements on several occasions each day for periods of several weeks, so that any differences in pattern between work days and rest days can be detected. Statistical software can aid interpretation. The unsupervised recordings may lack reliability and precision, and work-related changes may reflect nonspecific ‘irritant’ reactions rather than specific hypersensitivity responses. There is consequently some difference of opinion over the value of PEF monitoring. In practice, however, it provides the most widely used diagnostic tool for occupational asthma. 

Inhalation Challenge Tests

The greatest diagnostic confidence comes from laboratory inhalation challenge tests that are monitored by both supervised serial measurements of spirometry and repeated measurements of airway responsiveness. When there is deteriorating ventilatory function and increasing airway responsiveness, and the changes can be evaluated statistically, a diagnosis of OA can be considered ‘confirmed’ with considerable confidence, especially if the outcome is shown to be repeatable and the tests are carried out in a double- blind fashion. Thus, neither test subject nor the immediately supervising physician knows whether the challenge exposure involved the suspected asthmagen or a dummy ‘placebo’. In practice, such tests require sophisticated equipment, are very time-consuming, pose potential risks, and are inevitably restricted to a few centers. They involve a fraction of 1% of all cases, but are particularly valuable (arguably indispensable) when hitherto unrecognized occupational asthmagens are first investigated. 

Return-to-Work Studies

A useful, and practical, compromise is the ‘return-towork’ challenge test. For this, the test subject is kept from work (or at least from exposure to the suspected asthmagen) for a period of 2–3 weeks, during which time any asthmatic medication is reduced to a minimum (ideally discontinued). In true occupational asthma, some improvement is likely (or there is no deterioration with treatment reduction), and can be demonstrated by serial measurements of spirometry and airway responsiveness. Hourly spirometric monitoring over the 3 days prior to the return-to-work generates confidence limits, and so allows the detection of any statistically significant deterioration subsequently (usually within a few days only if the asthma is occupational). If airway responsiveness too increases significantly, OA in the individual is reasonably ‘confirmed’. The method does not confirm the identity of the asthmagenic agent, and is only suitable if the test subject is still employed and the employer cooperates fully. In many cases the possibility of an occupational cause does not arise until after the affected individual has ceased employment or the work environment has changed.  

Animal Models

Several animal species have provided invaluable insight as to how asthma arises following exposure to occupational agents. ‘Sensitization’ has been achieved by inhaled, dermal, and/or peritoneal routes for many occupational asthmagens (notably acid anhydrides, colophony, di-isocyanates, latex, plicatic acid), and both immediate and late asthmatic reactions have been provoked by subsequent inhalation challenge. The airways are then characterized by inflammation, eosinophil infiltration, mucus hypersecretion, and hyperresponsiveness. Hypersensitivity mechanisms have been confirmed by transferring lymphocytes or serum from affected to unaffected animals, which then respond to inhaled challenge in a similar fashion to the donor animals. Specific IgE antibodies to the inducing agents (or hapten conjugates) have been evident commonly, but not invariably. Occasionally specific IgG antibodies are reported. There is involvement of both CD4þ and CD8þ T cells, with a dominant T-helper-2 cells (Th2) response. The mechanisms include deposition of excess extracellular matrix (and increased activity of matrix metalloprotease, MMP) and activation of the vascular endothelium associated with the release of vascular endothelial growth factor (VEGF). Experiments with inhibitors of MMP and VEGF have shown substantial reductions of the markers of asthmatic activity, possibly pointing a way to novel strategies for future management.

Animal models have additionally confirmed that airway inflammation induced by exposure to toxic levels of certain reactive chemicals may also induce airway hyperresponsiveness. This simulates the RADS pathway. Ozone and chlorine exert their effects through oxidative stress, which is associated with increased expression of inducible nitric oxide synthase and an increase in airway responsiveness. Ozone has also been shown to suppress Th1 responses (possibly thereby enhancing Th2 responses), and acute exposure to ozone or nitrogen dioxide amplifies asthmatic responses to allergenic agents in already-sensitized animals.

Animal models have also been used to investigate the exposure threshold levels at which airway inflammation and hyperresponsiveness first develop. The importance of this for occupationally induced asthma in humans is readily evident, though extrapolation from animals is necessarily fraught with uncertainty. The threshold levels of exposure that trigger meaningful responses once sensitization has occurred may differ critically from those that are responsible for initial sensitization, and are likely to differ appreciably from individual to individual. 

Management and Current Therapy

Drug and ancillary therapies for occupational asthma are those of asthma of any cause, but occupational asthma arising by the hypersensitivity route offers an additional and critical means of management. If the relevance of an occupational sensitizer is recognized within 6–24 months of symptom onset, and if exposure then ceases, there is a meaningful probability of the asthmatic state resolving entirely (i.e., the level of airway responsiveness falls into the nonasthmatic range). This is almost unknown for adults with nonoccupational asthma, unless it is drug induced (beta-blockers, nonsteroidal anti-inflammatory drugs (NSAID). There is inevitably much variability from individual to individual, but if exposure ceases within as short a period as 6 months, the probability of complete resolution may exceed 50%. If the period exceeds 24 months, it is more likely that active asthma will continue, even in the absence of ongoing exposure.

Most individuals developing occupational asthma have unskilled or semiskilled jobs. Their susceptibility to develop asthma with exposure to sensitizing agents will inevitably put them at a disadvantage in the labor market, and this will be enhanced if active asthma persists. It is unfortunate that most lose their current jobs and never become gainfully employed again. This poses several management challenges. A correct diagnosis is the cornerstone, since mild asthma rarely leads to job loss of itself. A false-positive diagnosis may have devastating but unnecessary economic consequences for both the affected individual and the employer. A dilemma arises when asthma is occupational, but is mild, has been active for several years, and does not improve much after an experimental period away from the workplace. If the affected individual cannot find (or does not wish for) alternative employment, the risk of continuing exposure needs to be assessed. It will certainly diminish the long-term probability of regaining a nonasthmatic level of airway responsiveness, but this may be minimal anyway. More importantly it may cause a further and permanent increase in airway responsiveness, and hence worsening asthma with the possibility of remodeling and fixed airway obstruction. A period of close surveillance, with objective serial measurements of spirometry and airway responsiveness, may help the affected individual and his medical advisors to judge whether the obvious benefits justify the uncertain risks. There is anecdotal evidence that tolerance develops in some affected individuals and that they have little to lose by continued exposure.

Complete cessation of exposure offers the best outcome, especially if this can be achieved by using an alternative, nonsensitizing agent for the particular industrial process. If this is not possible or practical, it may be that exposure levels can be reduced by modifications to job plans, task sharing/exchanging, improved industrial hygiene (ventilation/extraction), or the use of respiratory protection equipment.


Prevention offers the best means of control. When a risk of occupational asthma is identified, surveillance programs may identify emergent cases at a point when exposure cessation is likely to produce cure. This assumes such programs are efficient. Some employers rely primarily on environment measurements, and reassure themselves (inappropriately) that if they have complied with regulatory limits, any emergent cases of asthma must be nonoccupational. When a clear risk of OA is recognized, the onus should lie with disproving the diagnosis when new cases of asthma arise, rather than the converse. It is the very nature of ‘allergy’ that some affected individuals become sensitized at exposure levels that are harmless to the majority, and that lie within regulatory limits.


Compensation is inevitably an issue for affected workers, particularly those who become disabled and unemployed. Equally inevitably, compensation systems differ from country to country. Regretfully, few include provision for retraining and re-placement. The physician managing occupational asthma needs to be familiar with local procedures in order to offer proper advice. At a time of increasing litigation, the physician too may find him- or herself the focus of a legal (negligence) suit if he or she has failed to provide correct advice.


The respiratory effects of dust from cotton and other biologic fibers became prominent in the nineteenth and early/mid-twentieth centuries. They appeared initially to be specific to the cloth manufacturing industry and the appellation, byssinosis, arose to identify the disorder characterized by them. This has had the misleading consequence of segregating cotton dust asthma from the many other types of occupational asthma that were identified later, and of disguising the other respiratory disorders that may result from cotton dust. A very characteristic feature of reported byssinosis has been a prominence of symptoms on the first work day following periods without exposure. These tend to diminish or even cease on subsequent days, at least in some individuals, despite similar levels of ongoing exposure. While chest tightness and breathlessness may be a consequence of asthma, some affected individuals describe an influenza-like reaction with fever and malaise (‘Monday fever’). The latter may occur on the very first employment day, that is, before any possibility of sensitization. Such symptoms have also been generated in experimental conditions when naı¨ve volunteers are exposed. The same febrile reaction may occur in subjects heavily exposed to metal fume or the fume arising from certain polymers. It is closely related to neutrophil inflammation and the release into plasma of pyrogenic interleukin- 6 (IL-6).

These non-specific systemic symptoms are features of ‘inhalation fevers’ generally. They occur widely with exposure to a variety of organic dusts, and in these contexts are now known more commonly by the term ‘organic dust toxic syndrome (ODTS)’. Products from microbial overgrowth appear to be responsible, and high airborne concentrations of respirable spores and endotoxins are characteristically seen. This is consistent with the recognition that byssinosis is most prevalent among those who harvest, transport, store, sort, process, and clean cotton before its ‘decontaminated’ fibers are woven into cloth. The matter is complex since heavy exposure of naı¨ve subjects to organic dust may also provoke cough, chest tightness, and bronchoconstriction. These symptoms are rarely severe, and rarely last for more than a day or two. Nevertheless, this response simulates RADS, and no clear sensitizer has been incriminated in the pathogenesis of byssinosis.

Some individuals affected by other forms of occupational asthma may also have more prominent symptoms on Mondays that peter out as the working week progresses. A further ‘characteristic’ of byssinosis has also proved to be non-specific, namely the development of chronic obstructive pulmonary disease (COPD). Early studies of byssinosis were necessarily cross-sectional, and it was assumed that subjects with chronic obstructive pulmonary disease (COPD) had initially experienced the typical acute and reversible features (i.e., asthma and/or inhalation fever), which had progressed to produce ‘chronic byssinosis’. It seems more likely now that COPD and asthma (and inhalation fever/ODTS) are independent disorders arising from cotton dust exposure, which result from different levels of susceptibility within exposed workers and different patterns of cellular response. Even so, it is to be expected that COPD might develop because of asthma as well as independently of it, and there is convincing evidence to support this from recent longitudinal studies. 

Further Reading

Beach JR, Young CL, Avery AJ, et al. (1993) Measurement of airway responsiveness to methacholine: relative importance of the precision of drug delivery and the method of assessing response. Thorax 48: 239–243.

Brooks SM, Weiss MA, and Bernstein IL (1985) Reactive airways dysfunction syndrome (RADS). Persistent asthma syndrome after high level irritant exposures. Chest 88: 376–384.

Burge PS, Pantin CFA, Newton DT, et al. (1999) Development of an expert system for the interpretation of serial peak expiratory flow measurements in the diagnosis of occupational asthma. Occupational and Environmental Medicine 56: 758–764.

Cote J, Kennedy S, and Chan-Yeung M (1990) Outcome of patients with cedar asthma with continuous exposure. American Review of Respiratory Diseases 141: 373–376.

Glindmeyer HW, Lefante JJ, Jones RN, et al. (1994) Cotton dust and across-shift change in FEV1 as predictors of annual change in FEV1. American Journal of Respiratory and Critical Care Medicine 149: 584–590.

Hendrick DJ and Burge PS (2002) Occupational asthma. In: Hendrick DJ, Burge PS, Beckett WS, and Churg A (eds.) Occupational Disorders of the Lung – Recognition, Management, and Prevention, pp. 33–76. London: Saunders.

Herrick CA, Xu L, Wisnewski AV, et al. (2002) A novel mouse model of diisocyanate-induced asthma showing allergic-type inflammation in the lung after inhaled antigen challenge. Journal of Allergy and Clinical Immunology 109: 873–878.

Lee KS, Jin SM, Kim SS, and Lee YC (2004) Doxycycline reduces airway inflammation and hyperresponsiveness in a murine model of toluene diisocyanate-induced asthma. Journal of Allergy and Clinical Immunology 113: 902–909.

Lee YC, Kwak YG, and Song CH (2002) Contribution of vascular endothelial growth factor to airway hyperresponsiveness and inflammation in a murine model of toluene diisocyanate-induced asthma. Journal of Immunology 168: 595–600.

Meyer JD, Holt DL, Cherry NM, and McDonald JC (1999) SWORD ’98: surveillance of work-related and occupational respiratory disease in the UK. Occupational Medicine 47: 485–489.

Newman Taylor A (2000) Asthma. In: McDonald JC (ed.) Epidemiology of Work Related Diseases, 2nd edn., ch. 8, pp. 149– 174. London: BMJ Books.

Schilling RSF, Hughes JPW, Dingwall-Fordyce I, et al. (1995) An epidemiological study of byssinosis among Lancashire cotton workers. British Journal of Industrial Medicine 12: 217–227.

Stenton SC, Avery AJ, Walters EH, and Hendrick DJ (1994) Technical note: statistical approaches to the identification of late asthmatic reactions. European Respiratory Journal 7: 806–812.

Stenton SC, Dennis JH, Walters EH, and Hendrick DJ (1990) The asthmagenic properties of a newly developed detergent ingredient – sodium iso-nonanoyl oxybenzene sulphonate. British Journal of Industrial Medicine 47: 405–410.

Venables KM, Topping MD, Howe W, et al. (1985) Interaction of smoking and atopy in producing specific IgE antibody against a hapten protein conjugate. British Medical Journal 290: 201–204.