Introduction

The term”work-related asthma” encompasses both occupational asthma (OA) and work-exacerbated asthma (WEA) [1]. OA is characterized by variable airflow limitation and/or hyperresponsiveness and/or inflammation due to causes and conditions attributable to a particular occupational environment and not to stimuli encountered outside the workplace [2], while WEA is defined as the aggravation of pre-existing or coincident (adult new-onset) asthma due to workplace environmental exposure [3]. OA is subdivided into immunological or non-immunological forms, with the reactive airway dysfunction syndrome (RADS) being the most characteristic example of the latter presentation [4]. A recent evidence-based review of the literature identified 372 causative agents of immunological asthma and 184 different causes of irritant or non-immunological OA [5].

Immunological OA, caused by workplace sensitizers, is characterized by the appearance of work-related asthma symptoms after a latency period. The causative agent may be either a high or a low molecular weight agent (HMW and LMW, respectively). HMW agents are protein-derived antigens and are generally considered to cause sensitization through an IgE-mediated mechanism and so allergy skin-prick test and measurements of allergen-specific antibodies can aid diagnosis [6]. Although specific IgE antibodies have also been detected in OA induced by some LMW agents [7], and several studies have suggested that immunologic mechanisms are involved in LMW-related OA [8–9], the exact mechanisms have not yet been fully characterized. In fact, the few studies carried out to date have demonstrated variable patterns of response to HMW and LMW agents; in the main, HMW agents seem to induce early or dual asthmatic reactions, while LMW agents produce delayed reactions [10–11].

It is not known whether differences in the pathogenesis of immunological OA also lead to differences in clinical presentation. Early studies in the 1990s suggested that the natural history of the onset of OA varies according to the sensitizing agent, and that factors such as age, gender, atopy, rhinitis, and smoking habit might influence the development of the condition [10]. However, recent studies suggest that the intensity of exposure may affect the risk of development of symptoms and sensitization more than host markers or the type of agent [12]. Whether or not the severity of asthma is related to the type of causal agent is also unclear.

The objective of this study was to analyse the differences in relation to possible risk factors, asthma presentation and severity, and response to various diagnostic tests in patients with OA caused by either HMW or LMW agents.

Patients and Methods

Results

Of the 78 patients with final diagnosis of OA, 23 responded to HMW agents and 55 to LMW agents. The various sensitizing agents responsible for OA are shown in Table 1. In the group sensitized to HMW agents, flour was the more prevalent (48% of cases). For LMW agents, isocyanates and persulphates were the most prevalent (36% and 24% of cases respectively). Subjects’ demographic data are summarized in Table 2. Although the percentage of patients with atopy was similar in the two groups, patients with OA caused by HMW more frequently presented rhinitis and conjunctivitis than patients with OA caused by LMW.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Sensitizing agents responsible for OA in the study population.

https://doi.org/10.1371/journal.pone.0156141.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Demographic characteristics of the study population.

https://doi.org/10.1371/journal.pone.0156141.t002

Table 3 presents data on the occupational exposure of patients, severity of asthma and the treatment they were receiving at the time of diagnosis. Patients with OA caused by LMW agents seemed to have more severe asthma than those with OA caused by HMW agents and a greater use of long-acting beta-antagonists (LABA), probably related to the greater severity. No differences were seen between the groups in other variables. A logistic regression analysis confirms that patients with OA caused by LMW agents had a significantly higher risk of asthma severity according to the GINA classification after adjusting for potential confounders (OR = 7.16, 95% CI: 1.13–15,20; p = 0.036) (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Data on occupational exposure, asthma severity and treatment received by patients at the time of diagnosis.

https://doi.org/10.1371/journal.pone.0156141.t003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Logistic regression analysis with patients exposed to HMW agents as independent variable comparing patients with intermittent or mild asthma and those with moderate or severe asthma.

https://doi.org/10.1371/journal.pone.0156141.t004

Finally, Table 5 shows the results in the SIC, suggesting that HMW agents typically induced an early reaction, whereas LMW typically induced a delayed reaction (p = 0.0001). In patients with OA caused by HMW agents SIC was positive with a shorter exposure time (p = 0.025) and these patients required more rescue medication during the SIC (p = 0.003) than those with OA caused by LMW agents. However, patients with OA caused by LMW agents presented a greater degree of bronchial hyperresponsiveness after the SIC, measured as the difference in the values of methacholine DRR (p = 0.024). No differences were found in the fall in FEV1, regardless of the type of response (early, late or dual) or type of agent. Considering the population as a whole, patients requiring treatment with IC alone had a greater decrease in FEV1 after SIC than those taking IC + LABA; median (range): 27.5 (16. 61) and 21 (15–48), respectively, p = 0.005. These differences are not observed when the population is divided into LMW and HMW groups.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Results of specific inhalation challenge.

https://doi.org/10.1371/journal.pone.0156141.t005

Discussion

To our knowledge, this is the first study to show that OA produced by LMW agents may be more severe than that produced by HMW agents. In fact, few studies have assessed the severity of OA at the time of diagnosis, even though it is acknowledged that OA in general may be a particularly severe form of the disease, for three main reasons: 1) the asthma persists in all patients with OA who remain in contact with the causal agent and worsens in 50% in spite of treatment; likewise, it may persist in 50% of patients even though they avoid exposure, and may worsen in 10–40% [18–21]; 2) there have been reports of patients who died after developing acute asthma after occupational exposure to agents to which they were sensitized [22–23]; and 3) patients with OA and WEA consume ten times more medical resources than patients with non-work-related asthma [24]. Other reasons that may explain why OA may be more severe than non-OA are the difficulty of diagnosis, the difficulty of management, generally higher levels of exposure (peaks) in the workplace than in other environments, co-exposures to irritants, and so on.

Interestingly, in the Epidemiological Study on the Genetics and Environment of Asthma (EGEA study) of patients with severe asthma, Le Moual et al [25] found that up to 30% may be exposed to OA-causing agents and that the condition may be more severe in those who are exposed simultaneously to both HMW and LMW agents. The recent observation that persistent occupational exposure to asthmagens (either HMW or LMW) is associated with uncontrolled adult-onset asthma [26] or more severe forms of the disease [25, 27] supports this hypothesis. Furthermore, a longer latency between onset of symptoms and diagnosis of OA carries a worse prognosis for persistence and severity of asthma, whether or not exposure to the causative agent is avoided [28–29]. However, in our study it does not seem that either the total exposure time, or the latency between onset of symptoms and diagnosis were associated with the greater severity recorded in LMW-related asthma. In fact, in agreement with Dufour et al [11], we did not find any differences in the latency period before the onset of symptoms or in the duration of exposure between HMW and LMW agents. In this regard, several authors have pointed out that the length of time necessary for sensitization may depend, among other factors such as genetics or the concentration of inhaled agents [30], upon the nature of each agent rather than on the molecular weight of the agent alone. In this sense, in contrast to the results found in our study, Descatha et al [31], comparing the characteristics of patients with OA to HMW and LMW agents, found that the severity of the disease at the time of diagnosis does not appear to be influenced by the molecular weight of the causal agent.

In the present study, we did not find any association that might explain why OA is more severe when caused by LMW agents. A plausible hypothesis is that agents with different mechanisms of action may trigger different responses in terms of both inflammation and bronchial hyperresponsiveness, thus altering the degree of severity. In line with other authors [10–11, 17] we found that in the context of the challenge test, HMW agents tend to present an early airway response, while in the case of LMW the response is usually late, although early reactions can occur with LMW agents and late reactions with HMW agents. It is generally accepted that HMW agents cause asthma through an IgE-mediated mechanism, that is, via a Th2 response, and generate a clearly eosinophilic airway inflammation; they are also associated with a higher proportion of patients with rhinitis and conjunctivitis [6], as we observed in the present study. In this regard, Malo et al [32] found that the prevalence of symptoms did not differ for HMW and LMW agents, although rhinitis was more intense for HMW than for LMW. Likewise, it is recognized that eosinophilic asthma generally responds well to treatment with IC [13].

The situation is the reverse in the case of patients with OA caused by LMW agents. In fact, the pathogenesis of OA caused by LMW agents remains largely unclear. The data available suggest that the T-cell subsets and cytokine profiles involved in LMW-induced OA may differ from those operating in atopic asthma. Although some of them induce IgE-mediated responses [6], most induce asthma through a non-IgE related mechanism [33] in which non adaptative immune responses might play a role [34]. The possible role of non-immunological mechanisms such as epithelial injury, remodeling of the airway wall, oxidative stress or neurogenic inflammation are under debate [35]. This means that although the inflammation is eosinophilic in some patients, in many others it is neutrophilic or mixed, and in these cases the response to IC treatment is lower [35], these patients may require more treatment and their condition may therefore be classified as more severe [13].

Finally, another interesting result of this study is the observation that individuals with OA caused by LMW agents present greater bronchial hyperresponsiveness 24 hours after the SIC. In this sense, Vandenplas et al [36] demonstrated that SIC to LMW agents is the principal risk factor for the occurrence of asthmatic reactions requiring administration of short-acting beta agonists with or without oral or intravenous corticosteroids. We cannot rule out the possibility that different intrinsic mechanisms may be at work in thepathogenesis of OA caused by HMW or LMW agents. The IgE-mediated response characteristic of HMW agents causes a histamine release which in turn leads to a fall in FEV1 and also, since it is an isolated exposure, a return to baseline levels within a short period of time, which may mean that the degree of bronchial hyperresponsiveness remains unchanged. In OA caused by LMW agents, on the other hand, in addition to possible inflammatory mechanisms, the bronchial hyperresponsiveness may depend on a neuroimmune interaction involving both mast cell activation and the transient receptor potential ankyrin (TRPA)1-dependent stimulation of sensory neurons [37].

Without doubt, the main limitation of this study is its retrospective nature. We do not have objective measures of the degree of asthma control in our patients and so we cannot be sure that the classification of asthma severity at the time of diagnosis was correct. However, the data (including the treatment required by patients) were recorded at the time the SIC was conducted. It is essential that asthma is controlled before performing SIC, because otherwise the results may be misinterpreted and false positives may be obtained [17]. Prior to the SIC, asthma control is usually established by checking that there are no clinical changes or changes in pulmonary function after administration of a placebo [38]. None of our patients presented any such alterations and all underwent the SIC, so it can probably be assumed that their disease was controlled and that the severity was correctly classified. Another limitation is the small number of participants. We can not rule out the possibility that other variables might have reached statistically significant values with a larger number of observations. Finally, some authors have suggested that the outcome of OA varies according to geographical location [29]. This study was conducted in a European country, in which IC are widely used in the treatment of asthma–a practice which may alter the natural history of the disease [39].

In conclusion, this study demonstrates that OA caused by LMW agents may be more severe than that caused by HMW agents. However, the characteristics of the study do not allow us to draw any conclusions about the prognosis of the disease, especially since no differences were found in the baseline lung function, emergency visits or hospitalizations between patients with OA exposed to HMW or LMW agents.The confirmation that most LMW agents induce a delayed response and HMW agents an early response in the SIC, and the differences in the degree of bronchial hyperresponsiveness after the challenge, suggest that the two types of agent have different mechanisms of action. Equally, the absence of any variables associated with the increased severity caused by LMW agents in the present study suggests that these different mechanisms of action are also responsible for the severity of OA. Future studies with larger study populations are necessary to confirm these findings.

Acknowledgments

MJC is a researcher supported by the Miguel Servet programme from Instituto de Salud Carlos III (CP12/03101). The funders had no role in the study design, the data collection or analysis, the decision to publish, or the preparation of the manuscript.