Occupational interstitial lung diseases

Pneumoconioses

Asbestos, silica and coal are the most common causative dusts inducing pneumoconiosis, whereas diseases caused by beryllium and hard metals are much less common. The Global Burden of Disease study identified more than 60,000 incident cases of pneumoconiosis in 2017; 39% of those were silicosis, 25% CWP and 16% asbestosis [10]. Pneumoconioses, though initially described in countries with a high sociodemographic index (SDI), are now seen mostly in developing countries and low SDI regions. In India, over 3 million workers are exposed to silica dust, with 8.5 million more working in construction [19, 20]. Another study from India found that the prevalence of radiological silicosis and progressive massive fibrosis (PMF) was 52% and 7.5%, respectively, among the 529 sandstone mineworkers screened [15]. In a Turkish study, 153 young denim sandblasters reported frequent symptoms, including dyspnea (52%), chest pain (46%) and chronic cough (19%). Based on chest X-rays, 77 (53%) were diagnosed with silicosis, with those diagnosed having a longer exposure duration and lower lung volumes compared to their counterparts [14]. The incidence of CWP has been declining in many regions of the world but is still increasing in some African regions, Southeast Asia and Oceania [10]. Similarly, cases of rapidly progressive CWP continue to occur among coal miners [21]. In the long term, the incidence of asbestosis is also anticipated to decrease in countries where asbestos is no longer used for construction and manufacturing. However, due to the long latency period between exposure and disease, most regions will continue to report a large number of new cases before a drop in incidence can be expected [10].

Over the last two decades, reports from various countries around the world have suggested a concerning increase in silicosis caused by manufacturing and processing of artificial stone, which is largely used for the production of kitchen or bathroom countertops [22]. Artificial stone is composed of more than 90% quartz combined with polymer resins and pigments, which are then heat cured. Cutting of artificial stone produces particularly high levels of respirable crystalline silica dusts. Compared to silicosis developed in traditional silica exposure settings, this form of disease affects younger workers following shorter latency periods – due to either the higher intensity of exposure or increased toxicity of the dust, or both – and rapid progression to advanced lung disease [23-25]. In a recent study from the Bay of Cádiz (southern Spain), one third of patients with artificial stone silicosis progressed from simple silicosis to PMF after a mean of only 4 years of follow-up and despite being removed from continued exposure [26]. The true incidence and prevalence of this new form of silicosis remain unknown but are likely to continue to rise if practices are not changed. In this regard, ventilation systems and wet blade cutting methods are effective measures to reduce dust generation when working on artificial stone [27]. CWP (or ‘coal pneumoconiosis’) is caused by long-term (typically over ≥20 years) inhalation of dust from high-carbon coal (anthracite and bituminous) and rarely graphite. Many cases will arise from co-exposures and radiologically have features of other diseases (e.g. silicosis in coal miners) [30]. Although pure coal pneumoconiosis is increasingly uncommon, severe and rapidly progressive cases of the disease continue to occur. Notably, affected miners are significantly younger than other miners with CWP, implicating recent mining conditions [21].

An estimated 21,500 people died from pneumoconiosis in 2016 [28], and over 12,900 deaths were due to silicosis in 2019 [29]. With its large workforce, China is particularly impacted by pneumoconiosis, accounting for 44% of deaths globally [1]. Reassuringly, there seems to be an overall drop in mortality rates, although stagnant and increasing mortality rates are still reported in women and in low- and middle-income countries [29].

Chronic beryllium disease

Beryllium is utilized in the aerospace, ceramics, electronics and defence industries, and CBD occurs in 2%–16% of exposed workers [31, 32]. Exposure to beryllium remains a major public health concern, with an estimated 800,000 individuals currently at risk for developing beryllium sensitization (BeS) or CBD [31, 33]. In addition to workers directly exposed to beryllium, many more are likely to be exposed indirectly [34]. The disease is characterized by an accumulation in the lung of activated CD4+ T cells that respond to beryllium as an ‘antigen’ in an MHC class II-dependent manner [35-37]. Genetic susceptibility to CBD is strongly linked to HLA-DPB1 alleles possessing a glutamic acid at the 69th position of the β-chain (βGlu69) [38, 39].

Cobalt-induced hard metal disease

Other metals, including aluminium, cobalt, titanium and zirconium, are associated with granulomatous lung disease. Cobalt-induced hard metal disease (also referred to as giant cell interstitial pneumonitis) has a strong association with βGlu69-containing HLA-DP alleles, similar to CBD [40]. The latency period between exposure and clinically overt disease ranges from months to years, with the development of cough and dyspnea heralding the onset of ILD.

Hypersensitivity pneumonitis

Occupational HP is estimated to account for one fifth of all HP cases [2]. The epidemiology of the disease has been investigated most frequently in farmers, with varying incidence being observed depending on agricultural practices and climate conditions. A high prevalence has been reported in mountain areas and countries with a cold and humid climate [41, 42]. However, with the industrialization of many farming practices and application of preventive measures, the incidence may be falling [43]. Modern occupational HP is the most frequently reported in industrial environments, where solvents, chemicals and metalworking fluids are the main inciting agents [44]. Occupational HP may also occur in workplace outbreaks where multiple workers are affected [45, 46]. For instance, industrial machine-manufacturing workers may be exposed to metalworking fluids, which are heavily colonized by bacteria, such as Pseudomonas pseudoalcaligenes and Curtobacterium, although the metalworking fluid bacterial microbiome is likely to be far more complex [47]. The annual incidence of occupational HP in the United Kingdom is estimated at 1–2 cases per million, with metalworking fluids, farming and chemicals being the most frequent causative exposures [6]. Earlier reports from the Czech Republic showed similarly low incidences [7]. A US-based study reported the 1-year HP prevalence ranging from 1.67 to 2.71 per 100,000 persons, with a 1-year cumulative incidence of 1.28–1.94 per 100,000 among an insured population between 2004 and 2013 [48]. When occupational HP is suspected, industrial hygiene assessment or consultation with occupational respiratory specialists should be sought, given the frequent under-recognition of occupational HP by non-occupational specialists [49].

Sarcoidosis

Sarcoidosis is a systemic disorder characterized by the presence of granulomatous inflammation in affected organs. Disease pathogenesis remains poorly understood but is likely to involve a complex interplay between a triggering antigen (or, most likely, antigens) and the host genetic background leading to an exuberant immune response [50]. Although commonly referred to as a disease of ‘unknown cause’, epidemiological studies have revealed significant associations between multiple occupational exposures and increased risk for sarcoidosis [51, 52]. Using data from seven studies, Blanc et al. estimated that the contribution of occupational exposures to the burden of sarcoidosis was 30% (95% CI, 17–45) [2]. The a case control etiologic study of sarcoidosis study that recruited 706 patients newly diagnosed with sarcoidosis and 706 age-, race- and sex-matched controls found positive associations between sarcoidosis and agricultural employment (odds ratio [OR] 1.46), exposures to insecticides at work (OR 1.52) and work environments with mould/mildew (OR 1.61) [53]. Further analysis of the same dataset revealed that workers for suppliers of building materials, hardware and gardening materials were also at higher risk of sarcoidosis [54]. In addition, firefighters exposed to World Trade Center ‘dust’ from construction and furnishing materials were found to be at significantly increased risk of developing sarcoidosis within the first 4 years after exposure [55], although it is difficult to identify the causative compound/s due to heterogeneity of the exposure. More recently, Beijer et al. found immunoreactivity to silica and metals (i.e. aluminium, zirconium and beryllium) in 21.2% of patients with sarcoidosis and 0% of controls, further supporting the hypothesis that these antigens may be pathogenetic in a significant minority of sarcoidosis patients [56].

IPF and other ILD subtypes

Occupational exposures are risk factors for several fibrotic ILDs, including those considered idiopathic such as IPF and other idiopathic interstitial pneumonias [57]. Indeed, although these entities are by definition of unknown aetiology, established mechanisms of lung injury support the hypothesis that inhaled exposures might contribute to disease pathogenesis and progression. However, while inferring causality in patients who develop an acute disease immediately after a high-intensity exposure is usually straightforward, determining the likelihood that any given exposure is causative is difficult when chronic low-level exposures occur over many months to years or if multiple exposures co-occur, particularly in the absence of a histological confirmation of the diagnosis. In such cases, the best approach remains suspecting the possibility of an exposure-related ILD and taking a thorough exposure history (e.g. exposure to agents known to cause ILD, cluster of disease in co-workers or age younger than expected).

Over the last two decades, several case–control studies have investigated the association between occupational exposures and IPF. In the largest meta-analysis to date, pooled findings from 11 cohort and case–control studies reported the following population attributable fractions: silica 3%, wood dusts 4%, metal dusts or fumes 8% and vapours, gases, dusts and fumes 26%, indicating that a quarter of IPF cases are attributable to occupational exposures [2]. A recent Australian study used a population-based control group to estimate the risk of IPF diagnosis by occupational exposures [58]. Respirable (smaller than 10 μm) dust (OR 1.38) and asbestos (OR 1.57) were independently associated with increased risk of IPF. Similarly, a case–control study in Italy found that people with self-reported history of exposure to metal dust, metal fumes or organic dust were at higher risk of developing IPF, with longer exposure time conferring an increased risk of disease [59]. More recently, Park et al. performed a systematic review and meta-analysis of eight case–control studies including 3502 patients [57]. They confirmed that exposure to metal (OR 1.83) and wood dust (OR 1.62), pesticide (OR 2.07) and history of farming or agriculture work (OR 1.88) increased the risk of IPF.

For ILDs overall, registry-based studies show that inhaled exposures, including occupational, home and hobbies, are highly prevalent across ILD subtypes, including IPF, connective tissue disease-associated ILD (CTD-ILD) and unclassifiable ILD. Data from two distinct cohorts identified potentially relevant exposures, including avocational and para-occupational exposures, in 65% and 62% of all ILD cases, with higher exposure prevalence in men compared to women, although the contribution of inhalation exposure to disease development or progression could not be established conclusively in all cases [60-65].

Silica exposure is an established risk factor for autoimmune conditions including systemic sclerosis [63-65] but little is known about its relationship with other CTD-associated ILD [66]. In a longitudinal familial pulmonary fibrosis cohort, development of interstitial lung abnormalities was independently associated with self-reported occupational or environmental exposures to aluminium smelting, lead, birds and mould [67]. Occupational rock dust in this cohort was further identified as a risk factor for radiologic honeycombing [68].

Initial assessment

The possibility of an occupational aetiology should always be considered in the initial assessment of patients with ILD as occupational and idiopathic ILDs may be clinically indistinguishable, and ‘occult’ exogenous causes are easily missed unless an extensive occupational/environmental history is taken. Exposure history should include but is not limited to, job role and details of what activities are undertaken, an estimate of relevant exposures including duration and dose (where possible), latency and disease course. Questionnaires are commonly used in health surveillance programmes and may provide useful information about current or previous exposures. Standardized health questionnaires exist to identify respiratory symptoms suggestive of occupational disease [69, 70], but none is specific to ILD.

A multidisciplinary approach that includes pulmonologists, radiologists, pathologists, occupational medicine specialists and industrial hygienists, where available, should be pursued. Industrial hygienists and occupational physicians have a key role in preventing occupational ILD (and occupational lung diseases in general). As part of a team, these specialists are trained to recognize, evaluate and control environmental health hazards in the workplace. In this regard, the UK Health and Safety Executive's ‘Dustbuster’ is designed to assess compliance with workplace health regulations and the control measures in place to protect workers’ health, with a focus on preventing occupational lung diseases, especially those caused by construction dust and other hazardous substances (https://www.hse.gov.uk/lung-disease/help.htm). Carlier et al. have recently reported that the integration of a specialized occupational consultation in the multidisciplinary evaluation of patients with ILD resulted in a change in diagnosis in 12.8% (18/141) of cases, with 15.6% of patients (22/1141) being eligible to claim compensation for an occupational disease [71]. A diagnosis of occupational ILD carries legal consequences; indeed, once the occupational origin of the disease has been confirmed, patients can claim compensation. This is the case, for instance, of patients initially diagnosed with IPF who are eventually reclassified as asbestosis following multidisciplinary evaluation.

Tools for detecting occupational exposures

Several tools are available for exposure identification in patients with suspected or confirmed occupational ILD, but there is limited evidence guiding their application in clinical practice. A comprehensive patient history supplemented by validated relevant questionnaires to identify potential exposures is an essential part of the clinical assessment. Clinical improvement following antigen avoidance may inform about the causative exposure in HP. Environmental assessment by occupational specialists and/or industrial hygienists may identify otherwise unrecognized exposures and antigens. Occupational environmental sampling may be informative, but these tools are not standardized. In HP, positive serum-specific immunoglobulin testing indicates exposure but does not prove causality, and negative test results do not exclude specific antigens as potential causes. Specific inhalational challenge and lymphocyte proliferation testing provide information on immunologic sensitization but do not establish causality. Overall, a multidisciplinary approach integrating clinical history, questionnaires, environmental assessment and sampling, and in selected cases, immunologic testing, provides the most comprehensive approach to exposure evaluation in patients with suspected occupational ILD [72, 73].

Clinical features

Occupational ILDs affect workers of both genders and all ethnicities, although they are more common in men due to historical working practices. The latency is variable; for many ILDs such as the pneumoconioses, it can be several years but for HP, CBD and hard metal disease, it can be much shorter. Similarly, acute and accelerated silicosis and ILD secondary to inhalation of fumes and toxic gases may have a short latency. Similar to idiopathic ILD, dyspnoea and cough are the most frequent presenting symptoms but wheezing and chest tightness may also occur. Patients with HP may present with chronological symptoms with onset after job or specific task initiation. Symptoms may progress over time and may demonstrate temporal variation with improvement on weekends, holidays, or when working in different areas. Lung auscultation is often normal and generally nonspecific but may reveal end-inspiratory squeaks in patients with bronchiolitis, inspiratory crackles in advanced stages of fibrosis or wheezing in large airways obstruction. Pulmonary function tests are an essential component in the initial assessment of patients with occupational ILD but are similarly nonspecific, as they may reveal a restrictive, obstructive or mixed ventilatory defect.

Diagnostic features

The diagnosis of occupational ILD is generally made based on a combination of compatible occupational exposure history and chest imaging, and the exclusion of other conditions that may mimic occupational ILD (Fig. 1). Despite its low sensitivity for detecting early disease [74], chest radiograph remains a widely used imaging test for screening of occupational ILDs and for making a diagnosis, particularly in low-income countries. Chest CT has higher sensitivity for early disease and greater accuracy in defining disease pattern than chest radiograph (Table 2) and represents a pivotal tool in the diagnosis of occupational ILD [75].

The inhalation of inorganic particles may induce variable degrees of epithelial injury, and fibrogenic and hypersensitivity reactions, leading to pleomorphic and rarely specific CT abnormalities (Table 2). Radiographic features of asbestosis are nonspecific and may be indistinguishable from those of IPF (Fig. 2, Panels A,B); therefore, imaging should always be interpreted according to the exposure history. Asbestosis also shares several histopathological features with IPF [76, 77] (Fig. 2, Panels C,D), with mineralogical studies supporting the concept of asbestosis-IPF misclassification by revealing high fibre burdens in the lung tissue of patients initially diagnosed with ‘IPF’ and revision of the diagnosis to ‘asbestosis’ [78]. The diagnosis of asbestosis requires the presence of interstitial fibrosis coupled with the identification of ≥2 asbestos bodies/cm² counted on 5-μm thick iron-stained sections [79], which correspond to approximately 2000 asbestos bodies/g of dry lung. However, a confirmatory lung biopsy is rarely needed, and claims are generally made based on compatible clinical-radiological presentation and history of asbestos exposure [80-82]. Notably, marked differences exist in the propensity of different asbestos fibre types to cause asbestosis, with commercial amphibole (amosite and crocidolite) being two or three orders of magnitude more potent than chrysotile. It is estimated that, compared to the levels of exposure required to cause pleural plaques, those required for asbestosis to occur are approximately 2.5-fold higher for chrysotile and 10-fold higher for amosite [79, 83].

A high level of exposure may cause alveolar proteinosis, which manifests as ‘crazy paving’ and ground glass opacities, as in acute silicosis (Fig. 3, Panels A,B), or Indium-induced ILD [84, 85]. Non-fibrotic HP may also manifest acutely following exposure to antigens to which the individual is sensitized and hyper-responsive [86]. Long-term, cumulative exposure may be associated with ground glass opacities, subpleural lines and traction bronchiectasis, sometimes with coexisting honeycombing, as in asbestosis or in dust-related diffuse fibrosis (Fig. 4, Panels A,B) [87]. Silica- or silicate-induced diseases can manifest with a nodular pattern, ranging from simple silicosis to PMF, depending on the size of nodules (below or above 1 cm, respectively) (Fig. 3, Panels C,D). Inhalation of ketone butter flavouring diacetyl (as in food-flavouring workers, i.e. ‘popcorn worker's lung’) [88] or sulphur mustard (gas) (as in the US military exposed during the Iran–Iraq War) [89] may result in constrictive bronchiolitis, which is characterized pathologically by subepithelial scarring leading to narrowing of the airway, and radiologically (i.e. CT) by mosaic attenuation/air trapping and centrilobular nodules [90]. Typically, constrictive bronchiolitis causes expiratory flow limitation. Similarly, nylon flock workers’ disease is characterized pathologically by lymphocytic bronchiolitis and, on CT, by patchy ground glass opacities, consolidation and diffuse centrilobular micronodules [91, 92].

Exposure to beryllium can cause BeS, which is diagnosed by the demonstration of a beryllium-induced immune response via the beryllium lymphocyte proliferation test (BeLPT) [93]. BeS is the immunological precursor to CBD, with a rate of progression from BeS to CBD of 6%–8% per year after initial diagnosis [94]. As the BeLPT does not differentiate between CBD and BeS [95], a definite diagnosis of CBD requires evidence of non-caseating granulomatous inflammation in lung tissue (Fig. 5, Panels A–D).

Bronchoalveolar lavage (BAL) plays an important role in the diagnostic workup of patients with ILD, including occupational ILD, particularly when used in conjunction with clinical and CT data, thus reducing the need for more invasive procedures [96]. Electron microscopy (EM) can quantify mineral fibres and particles in BAL (as well as in lung tissue) and is also able to detect detailed elements via X-ray analytical EM [97]. In addition, with regard to the detection of asbestos fibres, EM using BAL has a similar sensitivity as light microscopy using lung tissue [98]. BAL in combination with transbronchial forceps biopsies can also be useful in the evaluation of exposure-related granulomatous lung disease such as CBD, sarcoidosis and non-fibrotic HP. Transbronchial lung cryobiopsy provides pathologists with larger samples of lung tissue compared to conventional forceps biopsies thus serving as an alternative to surgical lung biopsy in certain scenarios [99]. When diagnostic uncertainty persists, surgical lung biopsy may be required to confirm the diagnosis given the potential preventative, therapeutic, prognostic and legal implications of a diagnosis of occupational ILD. In such cases, a thorough review of the lung specimen by an expert pathologist looking for clues to exposures—of which the patient may not be aware, or may have forgotten, given the long latency of some exposure-related disease—by light microscopy, polarized light microscopy or EM, as appropriate, is of outmost importance (Table 3).

Co-morbidities

The risk for tuberculosis is up to 20-fold higher in silica-exposed individuals compared to those not exposed [110]. This combination (often referred to as silico-tuberculosis) may be refractory to standard therapy, thus requiring specific expertise to guide prolonged multi-drug treatment. Silica exposure might further increase the risk for lung cancer [111] and is associated with CTDs [112]. Asbestos exposure is associated with a high incidence of lung cancer and pleural mesothelioma [113], with mesothelioma also potentially originating from peritoneal serosal membranes [114], as well as colon and laryngeal cancer [114]. Given the risk of ILD exacerbation or progression due to infection, appropriate immunization and aggressive treatment of respiratory infections are recommended [75]. CWP might be associated with an increased occurrence of immune-mediated diseases such as rheumatoid arthritis (Caplan syndrome) [75, 115]. Other general principles of ILD management apply also to patients with occupational ILD (e.g. pulmonary rehabilitation, oxygen and lung transplantation) [116].

Economic burden of occupational ILDs

The economic burden of occupational ILDs in terms of disability, workers’ compensation, health expenditures and lost revenue is difficult to estimate, and very few studies have addressed the problem. This is particularly true for occupational ILDs caused by novel/non-conventional exposures. Moreover, physicians unfamiliar with occupational lung disease may fail to elicit a thorough exposure history, thus missing valuable data. Accordingly, the data that do exist (e.g. disability statistics, death certificates and hospitalization admission diagnoses) may not be accurate or complete [117], and the true burden of exposure-related lung disease is likely to be significantly underestimated. In 2004, in the United States only, estimates of cost for occupational diseases, including respiratory illnesses, exceed $26 billion annually, but the true economic burden is likely to be much higher [118].