Introduction

Asthma is the 14th most important disease in the world and the most prevalent in childhood [1]; thus, it should be a real concern in our society. Its prevalence has been estimated around 14% in children [2] and 8.6% in adults [1]. The two main problems in the determination of this prevalence are the variation in the criteria used for asthma diagnosis and the inadequate access to medical care in many countries. Nowadays prevalence estimation varies from the 235 million asthmatics given by the WHO [1] and the 335 million stablished by the Global Asthma Report 2014 [3].

Although asthma has a significant prevalence worldwide, its mortality rate is not very high with 250,000 annual attributable deaths [4], out of which only 50,000 occur in developed countries. The highest mortality rates are found among subjects over 70 years old [3]. However, these subjects also have comorbidities that represent confounding factors in the identification of the true cause of death, leading to an overdiagnosis that ranges from 39 to 80% [5,6,7,8].

On the other hand, fatal asthma (FA) and near-fatal asthma (NFA) usually affects youngsters and adults below 50 years of age [9•, 10]. However, despite its relevance in terms of epidemiology and social impact, FA is not yet fully understood and effectively prevented. In fact, on clinical grounds, it includes several phenotypes, characterized by variable causes and risk factors. Thus, asthma mortality still represents a critical issue in the management of the disease.

Occupational triggers of asthma are not taken into account in most asthma studies, despite the fact that OA is the second most common occupational pulmonary disease in underdeveloped countries and the most common in the developed world [11]. Moreover, work-related asthma (WRA) is estimated to affect between 15 and 25% of asthmatic adults [12] and it mainly affects the under 50-year-old group who also may experience FA and NFA events, showing that a deeper research is needed in this field.

In this review, we performed an in-depth analysis of occupational FA and NFA. In order to do so, we used Medline, through the portals PubMed http://www.ncbi.nlm.nih.gov/pubmed/ and SciELO (http://www.scielo.br) as bibliographic database, using “severe asthma,” “severe occupational asthma,” “severe work related asthma,” “near fatal asthma,” “fatal asthma,” “asthma mortality,” and “asthma deaths” as keywords. Case reports and books were also considered. All near-fatal and fatal events directly related to work exposure were selected and analyzed.

Occupational asthma

WRA is a clinical entity that includes both OA and work-exacerbated asthma (WEA). OA can be induced by a sensitizing (sensitizer-induced OA or allergic OA) or irritant agent found in the workplace (irritant-induced OA), with reactive airway dysfunction syndrome (RADS) being the most characteristic form of irritant-induced OA (Fig. 1). WEA implies a pre-existing or concurrent diagnosis of asthma that is exacerbated, or whose severity is increased, by an agent found in the workplace [13•, 14] (Fig. 1).

Fig. 1
figure 1

Modified from Malo JL and Vandenplas O. Definitions and Classification of Work-Related Asthma. Immunol Allergy Clin North Am 2011 [13•].

Full size image

In allergic OA, the causative agent may either be a high- (HMW) or low-molecular weight (LMW) agent. Sensitization to HMW agents is considered to be due to an IgE-mediated mechanism. Although specific IgE antibodies are less often detected in OA induced by LMW than HMW agents, some studies have suggested the implication of immunological mechanisms in LMW-OA, as well as non-IgE-mediated responses resembling delayed hypersensitivity reactions.

OA is the most important work-related pulmonary disease in developed countries and second most important in underdeveloped countries [11]. However, underreporting and underdiagnosis of OA are two major obstacles to the accurate determination of its prevalence. Workplace exposures are only documented in 7 to 15% of evaluations of adults with new-onset asthma [15], and larger asthma population studies do not always take workplace exposures into account. As diagnostic tools are not widely available, less than half of subjects with OA go through the proper evaluation for diagnosing OA [16]. In fact, even though the specific inhalation challenge (SIC) is the reference standard for OA diagnosis, it is not available in all tertiary centers. Hence, OA diagnosis relies on the clinician’s ability to relate the disease’s development and exacerbations to a workplace exposure. Nevertheless, studies have reported OA prevalence rates of 4–58% with a median of 15% [17•, 18]. The prevalence of WEA is even more uncertain than that of OA, but is considered to be similar [19].

Severity

The ERS/ATS and GINA guidelines define severe asthma as a subtype of asthma which requires treatment with high-dose inhaled corticosteroids plus a second controller and/or systemic corticosteroids in order to prevent it from becoming uncontrolled, or that remains uncontrolled despite optimal therapy [20, 21]. Accordingly to the Global Asthma Report 2014, 10% of asthma cases can be considered as severe [3], but according to GINA, severe asthma represents 20% of cases, 50% of which is uncontrolled [20]. Due to the discrepancies between the prevalence of severe asthma found in different studies, together with OA underdiagnosis, it is very difficult to establish the prevalence of severe OA.

Nevertheless, it should be pointed out that most NFA and FA events happen among youngsters and subjects under 50 years old and it is among this population that occupational exposures occur. Therefore, although OA is not usually considered as an individual cluster in asthma severity studies and is markedly underdiagnosed, it might be an important eliciting factor for severe asthma. Thus, future studies should look into this specific subgroup more deeply and evaluate its real prevalence and relevance.

Since OA asthma etiology and severity are related to the type and persistence of exposure to the causative agent, this exposure should be carefully analyzed. Early cessation of allergen exposure is associated with a better prognosis and milder OA severity [22••], finding which was also highlighted in a systematic review by Vandenplas and Lemière [23] on the cessation and reduction of exposure in OA. Meca et al. [24] suggested that OA induced by LMW agents is more severe than OA caused by HMW agents. These authors retrospectively studied 78 patients diagnosed with OA, where HMW agents were responsible for 23 (mostly cereal flour 48%) and LMW for 55 cases (mainly isocyanates 36% and persulfates 24%). After analyzing asthma severity according to GINA classification, they found a significantly higher risk for asthma severity and increased bronchial hyperresponsiveness for LMW agents. Of note, baseline lung function, hospitalizations, and emergency visits did not differ between the two groups of patients with OA.

A study undertaken by Le Moual et al. [25] also examined the relationship between asthma severity and exposure to occupational agents. In total, 148 subjects with OA and 228 controls with non-occupational asthma were categorized according to asthma severity using a clinical severity score combining the frequency of asthma attacks, symptoms between attacks, and hospitalizations. A significant association was found between exposure to any sensitizing agent (LMW agents, HMW agents, or mixed environments) and severe adult-onset asthma with an overall odds ratio (OR) of 4 (LMW 4.4 OR/HMW 3.7 OR), but not between exposure to irritants and asthma severity. Although the criteria used to define severity in this study relate to asthma control more than to treatment required to achieve control as per GINA guidelines, the results remained significant when analyzed using a variety of alternative severity definitions. Moreover, in order to avoid biased estimates of exposition, the study did not rely on self-reported exposure, but combined a job-exposure matrix and a standardized review step to give reliable exposure estimates.

This relation between exposure to LMW/HMW and asthma severity was also studied in The European Community Respiratory Health Survey (ECRHS) [26], where random samples of workers aged 22–44 years completed respiratory health-related questionnaires and underwent investigations including lung function testing, non-specific bronchial hyperresponsiveness testing, and specific IgE determination. Previous 12-month and 10-year occupational exposures were determined according to an asthma-specific job-exposure matrix and expert re-evaluation. Significant associations were found between uncontrolled asthma and past 12-month exposure to any HMW or LMW agent or to cleaning agents. The association was stronger when using the past 10-year exposures than 12-month exposures and no correlation was observed with irritants.

Occupational exposures seem to be an important cause of severe and/or uncontrolled asthma, and OA tends to be more severe than other types of asthma [27]. A longer latency between symptom onset and OA diagnosis is also related to a worse prognosis in terms of persistence and severity of OA [23]. Neutrophilic inflammation was also identified as a risk factor for refractory asthma in an Australian study performed in asthmatic workers with persistent occupational exposures [28]. Other OA biomarkers, such as fractional exhaled nitric oxide, have not been shown to correlate with disease severity [29]. We should keep in mind that OA and occupational rhinitis frequently occur simultaneously and their severity is related [30].

Severe exacerbations

Exacerbations are the main cause of fatal and near-fatal events and therefore should be closely analyzed in order to implement preventive measurements and treatments to avoid them. The agents associated with exacerbations include work-specific agents such as gases, fumes, and spills, but also non-specific agents such as common aeroallergens, stress, exercise, and extreme temperature [31].

Henneberger et al. [32] performed different studies analyzing occupational risk factors associated with severe asthma exacerbations. They performed a study in 2010 with 998 patients where they found a higher risk for severe exacerbations among bakers and blue-collar workers (drivers, sculptors, painters, and art-related works). Cumulative incidence of severe asthma exacerbations was elevated for high exposure to any dust, gas, or fume in the workplace. The authors determined that 14.7%, or one in seven severe exacerbations, was induced by occupational exposures and could possibly be prevented by eliminating high dust, gas, and fume exposures. Later, another study [33] was carried out in 557 workers, of which 29% had presented a severe exacerbation in the previous year. Seven types of exposures were associated with severe exacerbations: LMW agents such as epoxy resins (prevalence ratio [PR] 2.50, p = 0.046), other highly reactive agents (PR 1.93, p < 0.03), and isocyanates (PR 3.11, p = 0.001) and irritant agents such as inorganic dusts (PR 3.61, p < 0.001), metalworking fluids (PR 2.84, p = 0.005), combustion particles (PR 1.52, p = 0.07), and environmental tobacco smoke (PR 1.88, p < 0.001). Although the study focused on the severity of the exacerbations, not the severity of asthma as such, when the analysis was limited to subjects whose asthma onset was prior to the start of the job, the results remained unchanged. This also suggests that continuous exposure affects the severity of exacerbations, which is also supported by the results of Lemière’s prospective cohort study of 154 subjects described earlier [34]. In this study, the only predictor of severe asthma exacerbations in the 2 years of follow-up was the persistence of work exposure in subjects with WEA and OA (OR 0.2, 95% CI 0.05–0.6).

Unlike other types of asthma, OA has specific workplace triggers that induce all the symptoms. Thus, avoiding the causative allergen and early diagnosis are the cornerstones of OA management [35]. OA outcomes are best when the diagnosis is established early, the exposure is stopped, and the asthma is not yet severe [22••]. Besides avoidance and early diagnosis, OA treatment does not differ from other asthma phenotypes. Inhaled corticosteroids are the mainstay of treatment, but new therapeutic approaches such as immunotherapy and biological treatments can also be used [36].

Fatal and near-fatal asthma reports

Proper diagnosis and management of OA are of crucial importance, as the lack of an early diagnosis or proper disease management can lead to severe asthma and FA or NFA events. These events are not limited to specific triggers and have happened with a remarkable variety of allergens and among different industries (Table 1).

Table 1 Occupational asthma related death reports (modified and updated from Ortega et al. [37])
Full size table

As it is shown in Table 1, fatal events are widespread among different jobs, but in all cases, persistence of exposure to the offending allergens and a long duration of asthma are observed. Moreover, most cases had presented previous exacerbations related to the same exposure. Patient ages range from 26 to 75 years old and the delay between the onset of OA symptoms and death ranges from 6 months to 20 years.

Autopsy was performed in most fatal events to confirm the cause of death, but allergic tests such as specific bronchial challenge, specific IgE measurements, or skin prick tests were not so consistently undertaken prior to the fatal event, leading to a lack of a proper diagnosis to prevent the event.

Reported cases

Among these FA and NFA events, isocyanates were the trigger in four cases. These events occurred in car spray painting [41] or with truck bedliner application [43]. In addition, a FA event with diphenylmethane diisocyanates occurred in a foundry worker, in a plant where this substance was used to produce metal belts and tank pieces [42].

While isocyanates seem to be the most common trigger in the reported FA and NFA events, other substances induce severe exacerbations as well. Two deaths were reported in food industries, where green coffee dust was considered as the responsible allergen [44••] and three cases occurred in laboratories; two cases were caused by bicycloheptadine dibromide in pharmacology laboratories [39] and a third case was caused by papain powder [40]. Bakers are at high risk of OA and a fatal event has also been reported in a baker [45]. Milling and printing industries have also been affected by fatal events, with shark cartilage [46••] and gum arabic [38] as the respective triggers.

NFA events have also been described in the literature. Hairdressers are one of the most frequently affected collectives. Severe events have been described among hairdressers, but also in ordinary dye exposure with no occupational relation [48]. Finally, NFA events have also been described in dairy processors; one event caused by locust bean gum was recently reported [37] and in a further past, a FA event while cleaning a bulk milk tank was documented [47]. In the latter, the occupational trigger was not studied prior to the fatal event, but the report suggests that the chloramines released by the combination of ammonia and acid was the probable cause of the fatal event. While it is unclear whether the patient had a previous asthma diagnosis (WEA) or started presenting symptoms after repeated occupational exposure (OA), we should not forget that exposure to gas, dust, and other unspecific irritants is also responsible for FA or NFA events.

Conclusions

OA is a recognized as an asthma phenotype that should not be underestimated by clinicians. This type of asthma is of particular importance among adult late-onset severe asthma. Early diagnosis, recognition of specific triggers, and allergen avoidance are the cornerstones of management. However, diagnosis is not always easy to perform and every patient with WRA should be studied by a clinician with experience in OA. Identifying this asthma phenotype is of utmost importance in the differential diagnosis of adult-onset asthma.

WRA is divided into OA and WEA that should be differentiated because its management differs. FA or NFA events can occur in both WRA subtypes, and both conditions should be properly studied and treated to avoid asthma worsening and recurrent exacerbations. Late diagnosis and untreated or uncontrolled asthma are the main risks for FA and NFA events in any subjects with asthma. However, in OA, not achieving complete allergen avoidance, prolonged exposure to the allergen, and not preventing the development of severe asthma seem to be the main risk factors for FA and NFA. Despite some studies suggesting that LMW agents induce more severe asthma than other substances, others have shown that severe OA can be induced by LMW and HMW agents as well as mixed environments.

In conclusion, OA is an asthma phenotype that is susceptible of causing FA and NFA events regardless of the offending allergen involved. FA and NFA events can be prevented with a correct and timely diagnosis, which implies the accurate identification of the culprit agent and early cessation of exposure.