Occupational asthma, IgE and IgG

Of the 300 or so agents encountered in the workplace which have been demonstrated as capable of inducing asthma, some dozen are responsible for at least half of the reported cases of occupational asthma in industrialized nations such as the United Kingdom [ 1]. These agents are traditionally and conveniently categorized as being of either high or low molecular weights. Important among the former, generally proteins of biological origin, include proteins excreted by laboratory animals, enzymes used in the detergent or food industries, a variety of cereal proteins found in bakeries, and natural rubber latex proteins to which health care staff are often exposed. Low molecular weight or ‘chemical’ agents include the diisocyanates, consistently the most frequently reported cause of occupational asthma in the United Kingdom, colophony fume liberated from cored solder in the electronics industry, complex platinum salts and the family of acid anyhdrides used in the manufacture of resins. In most instances occupational asthma induced by high molecular weight agents is accompanied by the production of allergen-specific IgE antibodies; once sensitization has occurred, subsequently inhaled allergen cross-links with specific IgE bound to mast and other cells provoking the familiar, inflammatory cascade which gives rise to the asthmatic response. Low molecular weight agents probably act as ‘incomplete’ allergens or haptens, becoming allergenic only after binding with one or more serum proteins; in many cases (but not all — see below) specific IgE to the hapten-protein (frequently human serum albumen) conjugate is detectable. In these ways the mechanisms responsible for the clinical manifestations of occupational asthma mimic those responsible for (‘extrinsic’) asthma associated with common inhalant allergens encountered outside the workplace; indeed there are good reasons to assume that much occupational asthma provides a useful model for non-occupational or ‘environmental’ asthma, including the large part of this which is induced in childhood [ 2]. If this is the case then the relationship between occupational asthma and specific IgE production —generally termed ‘sensitization’ in the occupational setting and ‘atopy’ outside — bears further scrutiny. In the clinical setting, where an employee in a job with exposure to a recognized allergen presents with asthmatic symptoms, the detection of allergen-specific IgE antibodies is considered a wholly sensitive diagnostic test; that is, all patients in this setting with occupational asthma attributable to that agent will have measurable levels of specific IgE antibody or, to put it another way, there are no ‘false negative’ tests. In this light the detection of IgE antibodies becomes very useful, their absence making the diagnosis of occupational asthma virtually untenable. Such an approach is often taken with symptomatic workers exposed to high molecular weight agents, including those listed above but it is not, of course, universally applicable and for some agents in this class may be misleading; occupational asthma attributable to the inhalation of antibiotics, for example, and demonstrable by specific bronchial provocation is frequently unaccompanied by detectable levels of specific IgE antibodies [ 3]. Non-asthmatic, allergic symptoms such as rhinitis or urticaria may more frequently be unaccompanied by evidence of sensitization [ 4]. Faced with an asthmatic patient exposed at work to a low molecular weight agent the clinician often needs to more cautious in the interpretation of specific IgE antibody results. For some agents, a high sensitivity may usefully be inferred; most cases of occupational asthma induced by trimellitic [ 5] and tetrahydrophthalic [ 6] anhydrides, for example, are accompanied by detectable levels of specific IgE antibody; curiously the same does not appear to hold for maleic anhydride [ 7]. The detection of specific IgE antibodies to complex (unconjugated), halide platinum salts, either by skin-prick test or by radioimmunoassay in serum, is also considered a highly sensitive diagnostic aid; interestingly patients with occupational asthma from these agents, confirmed by provocation testing, in whom antibodies are not detectable at the time of diagnosis generally develop them soon afterwards [ 8]. Thus the time sequence of sensitization and asthma may not always be straightforward. Unfortunately the diagnostic role of specific IgE antibodies in the two most important causes of low molecular weight occupational asthma is less clear. No reliable method of antibody detection in colophony-fume asthma has been established [ 9] and in these cases the test does not have a useful diagnostic role. For asthma induced by diisocyanates the presence of specific IgE antibodies to an isocyanate-HSA conjugate is relatively insensitive being found in less than half of clinically confirmed cases [ 10]; thus there are many ‘false negative’ results and the failure to detect such antibodies is not useful in reaching a diagnosis. The apparent presence of specific antibodies in only a proportion of cases of diisocyanate asthma is puzzling. Given the di-valent nature of these compounds it would not be surprising if they formed a variety of hapten-protein conjugates with a variety of serum proteins; in this case the search for IgE antibodies to a specific conjugate type would inevitably be insensitive — although it is fair also to point out that detection of antibodies to other conjugates has improved sensitivity only marginally [ 11]. In some cases, where diisocyanate exposure has ceased prior to testing, the level of specific antibodies may have fallen to a level below detection; thus the sensitivity of the test may be dependent on the interval since last exposure. Finally, the absence of detectable IgE antibodies in many cases has led several authors to postulate alternative, non-immunological mechanisms for diisocyanate-induced asthma [ 12]. It is difficult, in this setting, to distinguish those pathological features which characterize induction from those which result from persistent disease provoked by continuing exposure. The clinical picture of diisocyanate asthma — as a disease whose symptoms only begin after a latent, asymptomatic period and thereafter may be provoked by very small exposures — and, crudely, its epidemiology — whereby only a fraction of an exposed workforce develops asthma — suggest strongly that this is primarily an immunologically mediated disease. The price of high sensitivity is, frequently, a low specificity. Not surprisingly therefore a proportion of employees with detectable IgE antibodies to high molecular weight agents such as laboratory animal proteins [ 13] or detergent enzymes [ 14] will have no symptoms. Thus there is a variable — often high —frequency of ‘false positive’ results, a fact which is important both in clinical practice and in the interpretation and dissemination of the results of workforce surveys. Curiously the specificity of IgE antibodies to diisocyanate-protein conjugates may be very much higher [ 10]. It is, of course, difficult to determine accurately the relationship between sensitization and disease in a clinical setting to which subjects have presented by virtue of their symptoms; this requires careful epidemiological studies of exposed workforces. Most of these latter are cross-sectional in design and open to survival biases whereby symptomatic employees leave at a faster rate than their asymptomatic colleagues (the ‘healthy worker effect’); this is unlikely to be the case for those who are sensitised without evidence of asthma, rendering it difficult to interpret any observed relationship between sensitization and disease. Nonetheless the few published studies of occupational cohorts indicate the presence of a population of employees who develop antibodies without evidence of asthma. Among those exposed to high molecular weight allergens, the factors which independently increase the risk of sensitization — with or without evidence of asthma — include high intensity of allergen exposure [ 13, 15], ‘atopy’ and — perhaps inconsistently — cigarette smoking [ 13, 16]; this last also appears to be important in some types of low molecular weight sensitization [ 17] whereas the role of atopy, as generally assessed, is far less clear. Recently published studies suggest an important role for specific HLA genotypes in sensitization to both low [ 18] and high molecular weight allergens [ 19]; these require confirmation. The factors by which some exposed workers develop both IgE antibodies and asthma whereas others become sensitised without evidence of disease are both intriguing and unclear; and are an issue which is equally topical in non-occupational asthma. In the epidemiology of this subject to date, sensitization and asthma have generally been treated as equivalent outcomes. In the examination of risk factors, the former (which is more common and therefore allows more powerful analyses) is often considered a surrogate for the latter, particularly since it is difficult on this scale to confirm diagnoses of occupational asthma. As a result it is, at present, impossible to disentangle the epidemiologies of sensitization and disease. Factors which determine one outcome rather than the other may include genotype, differing intensities, durations or patterns of allergen exposure and concomitant exposure to respiratory irritants. In this issue of the journal, Yokota et al. [ 20] report on a cross-sectional survey of 148 employees exposed to methyltetrahydrophthalic anhydride (MTHPA) in two electronics factories. In spite of relatively low current exposures to MTHPA a high proportion of those surveyed had detectable levels of specific IgE antibodies, a finding attributable perhaps to higher exposures in the past. No clear relationships between sensitization and crude estimates of current exposure were demonstrated but sensitization was more common among those with work-related rhinitis. No cases of occupational asthma were identified. These results typify many of the difficulties inherent in surveys of this design; although a high proportion of eligible employees took part, no information on turnover of the workforce is provided making it difficult to assess the extent of any survival bias. In these cases estimates of current exposure may not be suitable for the examination of exposure–response relationships. The focus of the paper however, is the role of a different class of immunoglobulins. MTHPA-specific IgG4 antibodies were detected in a third of employees, were significantly and independently related to continuous (rather than intermittent) exposure but were not associated with the presence of work-related symptoms. The authors conclude that the development of IgG4 antibodies reflects exposure alone and is immunologically dissociated from the development of allergic symptoms. In this respect they confirm the findings of earlier surveys of IgG development in workforces with similar exposures: for example to MTHPA [ 21] and to phthalic anhydride [ 22] — although the latter suggested a pathogenic role for IgG4. A small number of studies have examined IgG antibodies among employees with high molecular weight exposures and provided broadly similar results. Among laboratory workers, specific IgG antibodies to rat allergens were correlated with an estimate of current exposure intensity and were found in all those with IgE sensitization; and were not significantly related to the reporting of allergic symptoms [ 23]; this pattern has also been reported in bakers exposed to wheat flour allegens [ 24]. Taken as a whole — but mindful of their methodological shortcomings — these reports indicate that the development of occupational respiratory allergies is unrelated to the development of IgG antibodies; nor is there convincing evidence, in the form of inverse relationships between IgE sensitization or disease and the presence of IgG antibodies, that these latter act in a ‘blocking manner’. Their role therefore remains mysterious; for the time being it seems appropriate to consider them as bystanders.