Idiopathic pulmonary fibrosis and occupational risk factors.

Idiopathic pulmonary fibrosis (IPF) is a rare lung disease of unknown origin that rapidly leads to death. However, the rate of disease progression varies from one individual to another and is still difficult to predict. The prognosis of IPF is poor, with a median survival of three to five years after diagnosis, without curative therapies other than lung transplantation. The factors leading to disease onset and progression are not yet completely known. The current disease paradigm is that sustained alveolar epithelial micro-injury caused by environmental triggers (e.g., cigarette smoke, microaspiration of gastric content, particulate dust, viral infections or lung microbial composition) leads to alveolar damage resulting in fibrosis in genetically susceptible individuals. Numerous epidemiological studies and case reports have shown that occupational factors contribute to the risk of developing IPF. In this perspective, we briefly review the current understanding of the pathophysiology of IPF and the importance of occupational factors in the pathogenesis and prognosis of the disease. Prompt identification and elimination of occult exposure may represent a novel treatment approach in patients with IPF.

Introduction

Idiopathic Interstitial Pneumonias (IIPs) belong to the broader group of Interstitial Lung Diseases (ILD) (figure 1). IIPs have unknown aetiology and share a number of clinical and radiological features, being distinguished primarily by their histopathologic pattern on lung biopsy. Diagnosis is based on clinical history, physical examination, high-resolution computed tomography (HRCT) imaging, pulmonary function tests, and lung biopsy. IIPs, which are classified into 8 histological subtypes (table 1) (89), are characterised by different degrees of inflammation and fibrosis. Treatment and prognosis vary according to subtype and range from excellent to nearly always fatal. Of the various IIPs, Idiopathic Pulmonary Fibrosis (IPF) (also known as Cryptogenic Fibrosing Alveolitis, CFA) is the most common distinct entity and will be discussed below.

Figure 1

Classification of Interstitial Lung Diseases

Figura 1 - Classificazione delle patologie interstiziali polmonari

Table 1

Classification of Idiopathic Interstitial Pneumonias

Tabella 1 - Classificazione delle Interstiziopatie Polmonari Idiopatiche

- Idiopathic Pulmonary Fibrosis (IPF), also called Cryptogenic Fibrosing Alveolitis (CFA) (identified histologically as Usual Interstitial Pneumonia, UIP) - Nonspecific Interstitial Pneumonia (NSIP) - Respiratory Bronchiolitis-associated Interstitial Lung Disease (RB-ILD) - Desquamative Interstitial Pneumonia (DIP) - Cryptogenic Organizing Pneumonia (COP) - Acute Interstitial Pneumonia (AIP) - Lymphocytic Interstitial Pneumonia (LIP) - Idiopathic Pleuro-Parenchymal FibroElastosis (IPPFE)

Modified from/Modificata da: Travis et al (89)

Epidemiology

Reported prevalence and incidence data for IPF vary and depend on ascertainment methods, and on the age and geographic location of the patient population. Both prevalence and incidence increase with age, with onset commonly occurring in the sixth and seventh decades; IPF is rarely seen in patients under 50 years of age; both prevalence and incidence are higher in men than in women (75). Having close relatives affected by IPF seems to be the strongest risk factor for developing the disease, as reported by a Mexican study by García-Sancho et al., in which a parent or sibling with IPF was associated with an adjusted Odds Ratio (OR) for IPF of 6.1 (95% Confidence Interval, CI, 2.3-15.9), p<0.0001 (29).

In a subsequent systematic review, the reported prevalence of IPF ranged from 0.5 to 27.9/100,000 and incidence ranged from 0.22 to 8.8/100,000 person-years (40). In the United States, the estimated overall incidence of IPF ranges from 7 to 16 cases per 100,000 (overall) person-years (74). In Europe, IPF prevalence ranges from 1.25 to 23.4 cases per 100,000 person-years and the annual incidence between 0.22 and 7.4 per 100,000 person-years (66).

Pathogenesis

Understanding the pathogenesis of IPF remains a challenge. Current consensus is that it is influenced by both genetic and environmental factors. Several pathogenetic and observational studies suggest that the definition “idiopathic” could actually conceal an intricate relationship between individual genetic and epigenetic factors on the one hand, and external factors on the other. A recent literature review by Sgalla et al. explains that although this interaction between “internal” and “external” factors, and their burden on the pathogenesis of IPF, has yet to be fully investigated, some evidence regarding IPF risk factors is already available and can be summarised as follows (figure 2) (84).

Figure 2

Risk factors involved in IPF

Figura 2 - Fattori di rischio per lo sviluppo di IPF

- Environment:

• smoking history remains the strongest risk factor after gene mutations (which account for the familial forms, 5% of total cases) (46), and also affects prognosis, which is poorer in smokers than in non-smokers: Taskar and Coultas found a summary estimated OR for IPF of 1.58 (95% CI 1.27-1.97) for ever-smokers compared to never-smokers (88). A recent Australian study by Abramson et al. showed that passive smoking in the workplace also increases the risk of developing IPF, with OR found to be 2.04 (95% CI, 1.16-3.60) (1). Paolocci et al. reached the same conclusion in their assessment of the risk of developing UIP amongst workers exposed to environmental tobacco smoke (OR 2.2; 95% CI, 1.2-4.0) (71).

• Taskar and Coultas’ review of available literature identified also significant correlations between IPF and several occupational exposures, such as farming (summary estimated OR 1.65; 95% CI 1.20-2.26), livestock (summary estimated OR 2.17; 95% CI 1.28-3.68), wood dust (summary estimated OR 1.94; 95% CI 1.34-2.81), metal dust (summary estimated OR 2.44; 95% CI 1.74-3.40), stone, sand, and silica (summary estimated OR 1.97; 95% CI 1.09-3.55) (88);

• studies carried out to assess the role of microbial agents have shown that they play a very significant part in both exacerbations and the induction of pulmonary fibrosis in animal models. A recent literature review states that despite the very limited data regarding the induction of fibrosis in human beings, the connections between microbial agents and lung fibrosis are becoming clearer. Viruses have been widely investigated, albeit with a number of contradictions in the results: the majority of studies assessed the presence of human herpes viruses (HHVs) in lung specimens or in bronchoalveolar lavage (BAL) fluids, with some evidence regarding Epstein-Barr virus (EBV), cytomegalovirus (CMV), type 1 herpes simplex virus (HSV-1), type 6, 7 and 8 human herpes virus (HHV type 6, 7, 8) (17). Some studies involved hepatitis C virus (HCV), which can cause alveolitis and consequent fibrosis, and adenoviruses, which up-regulate TGF-beta in epithelial cells and induce them to express mesenchymal markers (17). Current evidence therefore suggests that viruses may play a role in progression and acute exacerbations rather than in the induction of lung fibrosis (17). Bacteria may also contribute to the development of the disease; BAL fluid analysis in IPF patients showed an increased bacterial burden, with the presence of many different species such as M. Catharralis, P. Aeruginosa, S. Pneumoniae, P. Mirabilis, Neisseria spp and Veillonella spp (17, 61). S. Pneumoniae could act as a trigger for the progression of fibrosis, since treatment against this infection has been seen to halt the progression of fibrosis in mice models (17). Bacteria may also benefit from gastro-oesophageal reflux, a typical comorbidity in IPF patients; however, it is still unclear whether bacterial infection represents a primary or secondary cause of IPF, or merely an exacerbating agent (60). As far as fungi are concerned, a few studies focus on P. Brasiliensis and Aspergillus infections (17). The available data regarding antimicrobial therapy in IPF patients show that prognosis is more favourable when the treatment is administered together with antifibrotic therapies. This effect could also be due to the immunomodulatory and anti-inflammatory properties typical of certain antimicrobial classes (17). Moreover, many types of environmental exposure may cause changes in the microbiome, another possible key factor in the pathogenesis of IPF, as discussed later in this paper.

- Genes: the gene variants implicated in Familial Interstitial Pneumonia and in sporadic cases associated with gene mutations involve the telomere length maintenance genes (TERT, TERC, PARN, RTEL) (6, 84), and surfactant dysfunction genes (SFTPC, SFTPA2) (84). The risk factors for disease development are mutations involving telomere biology (TERT, TERC, OBFC1) (6, 21, 84), host defence genes (MUC5B, ATP11A, TOLLIP, TLR3) (27, 84) and cellular barrier genes (DSP, DPP9) (84). The most important of these mutations would seem to be a gain-of-function variant in the MUC5B promoter region, which gives carriers a better IPF prognosis (72).

- Epigenetic alterations: DNA methylation, histone modifications and microRNA dysregulation (induced by genetic or external factors) have an increasing relevance in the pathogenesis of IPF (84). Several studies identify cigarette smoke as a strong effector of epigenetic modifications and ageing also plays an important role in these regulatory mechanisms (92).

- Ageing: as pointed out by Sgalla et al., many age-related changes may be involved in IPF (26, 84). Ageing correlates with an increased fibrotic response to injury and apoptosis resistance in myofibroblasts (52).

Taskar and Coultas classified the biological mechanisms underlying IPF into four categories: 1) delivery and persistence of an inhaled agent, 2) endogenous biochemical response, 3) immune response, and 4) fibrotic response (Figure 3) (88).

Figure 3

Mechanisms of lung response to inhaled agents and factors leading to IPF

Figura 3 - Meccanismi della risposta polmonare verso agenti inalati e fattori influenzanti lo sviluppo di IPF

1. Delivery and persistence of an inhaled agent: animal models exposed to mineral particles suggest that inhaled doses, physical features and lung clearance mechanisms are involved on the epithelial cell uptake-lung retention-injury pathway triggered by inhaled particles (18). Indeed, overwhelmed clearance mechanisms are the starting point for the deposition of inhaled particles in the interstitium, where they cause inflammation, with the release of macrophage growth factors that stimulate mesenchymal proliferation and extracellular matrix deposition (88). Individual anatomical and physiological features may favour particle penetration and deposition; increased ventilation such as that generated by physical exercise may increase particle delivery to the lung.

2. Endogenous biochemical response: inhaled external agents that cause oxidative stress in the lungs have been observed when analysing antioxidant glutathione-dependent enzymes in patients with coal worker’s pneumoconiosis and Hypersensitivity Pneumonitis (HP), and low glutathione levels have been observed in BALs from IPF patients (14). This evidence supports a relationship between IPF and oxidative stress induced by environmental agents.

3. Immune response: fibrosis secondary to injury may arise from an abnormal immune system response pattern. Macrophages are among the most important effectors involved in the development of pulmonary fibrosis. Impairment in the M1/M2 macrophage equilibrium, shifting from the M1 to the M2 pro-fibrogenic phenotype, causes aberrant injury response and fibrotic tissue deposition (23). The relationship between immune cells and lung fibrosis is an area of active research, since the presence of a high number of these cells may contribute to the creation of a pro-fibrotic microenvironment; moreover, as discussed previously, immune response to microorganisms may alter and affect the pathogenesis of IPF.

4. Fibrotic response: some studies based on animal models have shown that the fibrotic response to inhaled environmental agents depends on the production of specific cytokine patterns: Mangum et al. studied osteopontin expression in an experimentally-induced lung disease in rats inhaling titanium dioxide, and observed an increased level of this pro-adhesive and chemoattractant cytokine in BAL fluid samples prior to the development of parenchymal lesions (56). Richeldi et al. pointed out that a bad response to chronic lung injuries, leading to fibrotic tissue deposition, may be linked with Alveolar type II Epithelial Cell (AEC2) dysfunction and premature ageing (78). Indeed, as observed by Naikawadi et al., AEC2 telomere dysfunction leads to age-dependent lung remodelling and fibrosis in mice models (65). Another study by Liang et al. found that the AEC2 from IPF patients presented an impaired renewal capacity (51).

According to Taskar and Coultas, the connection between exposure and IPF matches biological plausibility, as the disease develops with an initial epithelial injury followed by abnormal repair mechanism activation. Moreover, animal models showed that exogenous mineral particles are taken up by epithelial cells with possible injuries. The same authors also state that genotypical variants may affect particle uptake, cell injury and inflammatory and fibrotic response, thereby creating a variable susceptibility among individuals exposed to fibrosing agents. They also point out that animal models of drug-induced pulmonary fibrosis support a link between IPF and inhaled harmful agents (88). Indeed, in mice, inhaling silica particles represents a trigger to storing them in the interstitium and activating inflammation through oxidative stress, with T cell and macrophage intervention (3); welding fumes show reparation genes up-regulation as a chronic exposure effect (86); the analysis of transcriptional gene patterns in chronic exposure to welding fumes in rats showed many common aspects with lung injury and fibrosis induced by other factors, such as bleomycin (68).

Pathobiology

Although in the past IPF was defined as an inflammatory disease, as pointed out by Sgalla et al., it is currently defined as an epithelium-driven disease, since many sources of evidence show that a dysfunctional, ageing lung epithelium, undergoing multiple chronic microinjuries, attempts to regenerate, but local conditions may switch to excessive mesenchymal activity, thereby commuting tissue repair into chronic fibrotic tissue deposition (78, 84). Our understanding of what happens before diagnosis is still incomplete: the pathological features found at diagnosis are the final result of a pathway including the persistence of one or more harmful factors, repair mechanism alteration, inflammation (with increased levels of IL-1 and TNF-α) and a pro-fibrotic microenvironment that induces chronic regeneration and tissue remodelling (15). As mentioned previously, lung epithelium in IPF patients may become dysfunctional due to many causes; this dysfunctional behaviour leads to repair impairment (78).

Alveolar type I epithelial cells (AEC1) can be damaged by many factors, such as gastro-oesophageal reflux, smoke, infections, inhaled vapours, gases, dusts, fumes (VGDF), and so on. AEC2 proliferate and differentiate to restore lung integrity after an injury causing AEC1 loss; if the AEC2 are dysfunctional, epithelial restoration may be compromised. This could be the starting point of IPF pathogenesis (20). As stated previously, many sources of evidence have shown that, in IPF patients, AEC2s present activation of various fibrosing signal pathways, such as the TGF-β1, epithelial-mesenchymal transition (EMT) and unfolded protein response (UPR) pathways; however, exactly how these pathways are activated is still unclear, although infections (HSV) and inhaled VGDF have been considered as a possible cause (84). Moreover, endothelial and epithelial damage induces the coagulation cascade and a pro-coagulative state through the production of tissue factor (TF), plasminogen activation-inhibitors 1 and 2 (PAI 1 and 2) and protein C inhibitors (84). This microenvironment reduces the degradation of extracellular matrix and promotes fibroblast differentiation. It is also responsible for circulating fibrocyte and fibroblast recruitment. Maharaj et al. noticed that the percentage of fibrocytes increases during acute exacerbations of IPF, and decreases when the hyperacute phase ends, suggesting them as a possible marker for the acute phase of fibrotic tissue deposition (54).

The physiological behaviour of the airways also changes during the pathogenesis of IPF. Airway basal cells regulate their phenotype to suit chronic injuries, thereby initiating an aberrant proliferation that leads to the “bronchiolisation of the alveolar space” a process typically observed in the usual interstitial pneumonia (UIP) pattern (78).

Another typical trait of the UIP pattern is represented by Fibroblastic Foci (FF), small clusters of active fibroblasts and myofibroblasts that develop next to hyperplastic AEC2s. This proximity enhances the effects of TGF-β1 and of the profibrotic microenvironment in a lung evolving towards IPF (84).

Inflammation secondary to tissue damage may also contribute to IPF pathobiology through macrophage production of cytokines and the induction of a pro-fibrotic environment, as they recruit fibroblasts, epithelial and endothelial cells. If damage persists, inflammation may induce a dysfunctional wound-healing response due to chronic reactive oxygen species (ROS) production, which worsens the epithelial damage and induces an imbalance between pro-oxidant and anti-oxidant factors. The role played by lymphocytes is still unclear: some of their cytokines have a pro-fibrotic role, and Th2 and Th17 subsets have been linked with the pathogenesis of IPF (15). The connection between inflammation and IPF warrants further investigation.

Clinical manifestations

IPF should be considered in all adult patients with unexplained chronic exertional dyspnoea, cough, bibasilar inspiratory crackles, and/or digital clubbing that occur without constitutional or other symptoms that suggest multisystem disease. IPF usually occurs in patients aged 60 years or over (26).Patients often report a gradual onset of dyspnoea on exertion and non-productive cough over several months. Fatigue, fever, myalgia, and joint pain are rarely reported.

The rate of clinical decline observed in patients with IPF is very variable. Many patients with IPF will experience a slow but progressive decline in their clinical condition, whereas a small proportion will experience rapid deterioration, as part of an accelerated clinical course punctuated by acute exacerbations; the rate of acute exacerbations in patients with IPF has been estimated to be between 10 and 20% of patients per year (26).

On the physical examination, bibasilar crackles are usually audible, but in rare cases they may be absent or only heard unilaterally in the early stages of the disease. Patients with more advanced disease may have end-inspiratory “squeaks” due to traction bronchiectasis. Although early reports describe finger clubbing in 45 to 75 percent of patients, it is a typical manifestation of advanced IPF.

Diagnosis

IPF is diagnosed by the identification of a UIP pattern on the basis of radiological or histological criteria in patients without evidence of an alternative cause, since a UIP pattern can be detected in other ILDs, including occupational diseases (table 2) (91). This approach is endorsed in consensus guidelines worldwide and has helped to standardise the diagnosis of IPF (58). A 2002 statement by the American Thoracic Society (ATS) and European Respiratory Society (ERS), updated in 2013, advocated a multidisciplinary approach to the diagnosis of idiopathic interstitial pneumonias, involving a review of all the clinical, radiological, and pathological information (when biopsy material is available) from the patient (89). IPF is a diagnosis of exclusion, formulated by the multidisciplinary team (MDT), in patients with evidence of interstitial lung disease in the absence of identifiable causes.

Table 2

Non-IPF interstitial lung diseases possibly showing UIP pattern

Tabella 2 - Patologie interstiziali polmonari non-IPF con possibile pattern UIP

- Chronic hypersensitivity pneumonitis (C-HP) - Pleuroparenchymal fibroelastosis (PPFE, often coexisting with IPF) - Connective tissue diseases (particularly Rheumatoid Arthritis and Sjögren’s Syndrome) - Antisynthetase Syndrome (anti-KS, anti-PL7, anti-EJ) - Asbestosis - Chronic sarcoidosis - Drug toxicity (e.g. Amiodarone) - Familial interstitial lung disease - IgG4-related disease

Sources/Fonti: Lynch et al (53), Wuyts et al (91)

The key radiological features required for the diagnosis of IPF, as identified by the latest guidelines, include a UIP pattern on high-resolution computed tomography (HRCT), characterised by “honeycombing with or without peripheral traction bronchiectasis or bronchiolectasis” with a predominantly subpleural and basal distribution. The same guidelines recommend performing a surgical lung biopsy to confirm or refute a diagnosis of IPF in patients with ‘probable UIP’, ‘indeterminate UIP’ or ‘inconsistent with UIP’ patterns on HRCT (53, 76).

Although IPF is defined as an idiopathic disease, epidemiological studies have identified several risk factors (including different kinds of environmental and occupational exposure) associated with diagnosis. These different types of exposure can considerably affect the disease’s phenotype: one Korean national survey showed that dust exposure correlates positively with an earlier age of IPF onset, longer symptom duration before diagnosis, and a less favourable prognosis (47). On the contrary, a Belgian retrospective comparison between the survival rates of chronic hypersensitivity pneumonitis (CHP) patients, IPF patients exposed to organic dusts such as mould/birds, and non-exposed IPF patients found that exposed IPF patients’ outcome was better than non-exposed IPF patients, and worse than that of CHP patients (22). The relationship between IPF and external exposure undoubtedly warrants further investigation. Therefore, before formulating a diagnosis of IPF, it is important to carefully consider the potential alternatives, on account of the many clinical and radiological similarities between IPF and other interstitial lung diseases (2).

Due to the radiological similarities on HRCT, many patients presumed to have IPF on the basis of the initial clinical evaluation may actually have CHP (63). Patients may forget having been exposed to occupational and environmental antigens, important factors that may favour a diagnosis of CHP, in the past. Numerous studies have also shown that environmental risk factors, including occupational exposure, are associated with disease pathogenesis (table 3).

Table 3

Occupations and exposures assessed by literature as possible risk factors for IPF

Tabella 3 - Professioni ed esposizioni considerate dalla letteratura come possibili fattori di rischio per lo sviluppo di IPF

- Welding fumes - Farming/agriculture - Hairdressing - Dentists/dental technicians - Metal dust - Wood dust/paper mill factory workers - Livestock, particularly birds - Nuclear waste/radiation hazards - Chemicals vapors - Aluminium, Corion® - Stone cutting/sand/granite/silica - Talc - Asbestos

Modified from/Modificato da: Sack C, Raghu G. Idiopathic pulmonary fibrosis: unmasking cryptogenic environmental factors. Eur Respir J 2019; 53: 1801699

Treatment

The treatment of IPF has recently focused on targeting the fibrotic pathway, an aftermath of the onset of the disease pathway. The two available anti-fibrotic agents, Pirfenidone and Nintedanib, have only a modest effect on slowing the decline of forced vital capacity (FVC), a marker of disease progression, and on overall mortality (30).

The restricted approval by regulatory agencies of Pirfenidone and Nintedanib as therapeutic anti-fibrotic agents for IPF may prompt off-label use for other fibrotic lung diseases; in addition, physicians may mislabel a diagnosis of IPF in an attempt to get prescription coverage for these drugs for non-IPF fibrotic lung diseases.

Occupational and Environmental Exposure in patients diagnosed with IPF

Many patients labelled as having IPF may instead have a form of interstitial lung disease that has actually been triggered by occupational or environmental exposure (35, 63). This would, of course, imply that the patients’ clinical condition is not idiopathic; this occupational population attributable fraction of IPF was recently estimated as being 26% (16). Metal dust and fumes, organic dust, mineral dust, vapours, gases, non-metal fumes, and environmental tobacco smoke have been all associated with the pathogenesis and, in some cases, the prognosis of IPF. Occupational dust may also be an aggravating factor associated with a poor prognosis in IPF, and patients exposed to dust may present an earlier onset of IPF, relatively longer symptom duration, and worse prognosis than those in unexposed groups (47).

Studies conducted in the early 1990s reported that UIP mortality was higher in men, in the elderly and in the central parts of England and Wales, suggesting that occupational exposure to certain factors might account for the concentration of the disease in industrialised areas (38). Ley and Collard confirmed these findings, suggesting that if some inhaled agents such as asbestos and silica can cause lung fibrosis, the same could be true of less-well characterised types of exposure (49).

The 2006 review by Taskar and Coultas focused attention upon working exposures and development of IPF; findings were obtained about farming, livestock, metals, wood dust, metal dust, stone, sand and silica (88).

A recent study performed by Sack et al. examined occupational exposure to vapours, gases, dusts and fumes in community-dwelling adults, finding qualitative and quantitative HRCT alterations related to subclinical ILD (79).

A retrospective study performed in 2018 in Switzerland showed that nearly half of all IPF patients originating from the area reported occupational exposure (farming, welding, asbestos, birds and dust) and the majority were former smokers (31).

A very recent study by Li et al. assessed the level of dust exposure and the risk of developing pulmonary fibrosis amongst World Trade Center responders, who inhaled large amounts of toxic dusts including heavy metals, such as titanium, silica, asbestos fibres and wood dusts: the data revealed that the risk of pulmonary fibrosis was higher in subjects with intermediate and very high levels of exposure, than amongst those with lower-level exposure, with Adjusted Hazard Ratios (AHR) of 2.5 (95% CI 1.1-5.8) and 4.5 (95% CI 2.0-10.4), respectively (50).

Another possible form of exposure in the workplace is passive smoking; Paolocci et al. examined the relationship between environmental tobacco smoke (ETS) at work and the risk of developing a UIP pattern, which doubled in presence of ETS: the OR, adjusted for age, gender and active smoking history, was 2.2 (95% CI 1.2-4.0) (71).

A commentary sent as a letter to the editor by Marques et al. underlines how often clinicians may find environmental or occupational exposure by performing a self-reported clinical history with patients presenting probable UIP patterns: in their experience, up to 76.5% of these patients reported at least one identifiable exposure (birds, moulds, drugs, occupational factors such as silica, metal dust, agriculture) (57). The challenge lies in establishing whether they are clinically relevant.

Metals

The first study to formally examine the role of occupational exposure in the aetiology of IPF was a case-control study carried out in Nottingham, UK. The study recruited 40 prevalent cases of IPF and 106 age-, sex- and community-matched controls and used a postal questionnaire to collect details of lifetime occupational history and dust exposure. Cases were marginally more likely to report exposure to any occupational dust than controls (OR 1.32, 95% CI 0.84-2.04); however, amongst the reported dust exposures there was a marked increase in metal dust exposure (OR 10.97, 95% CI 2.30-52.4) (82).

A more extensive case-control study of occupational exposure to metal dust was carried out as a follow-up to this pilot study. A total of 218 cases of UIP and 569 community-matched controls were recruited from the East Midlands area of the UK. Information on occupational dust exposure was collected by both self-completed questionnaires and telephone interviews. Evidence of increased exposure to metal dust amongst cases, compared to controls, was found. There was evidence of a dose-response relationship between disease and exposure to metal dust in terms of number of hours of exposure per day and number of years of exposure. The specific types of metal exposure that were increased in patients with IPF included brass, lead and steel (34).

In 1994, IPF risk-related factors were epidemiologically investigated on the basis of 1,311 Japanese IPF autopsy cases selected from annual autopsy data records in Japan over a 12-year period. The IPF rate was more than double amongst individuals who had been exposed to solvents or organic dusts. A live case-control study was subsequently performed, together with a post-mortem case-control study; a significantly higher OR was noted amongst metalworkers and miners than healthy and hospital control subjects (1.37 and 1.34, respectively). Both studies provided evidence of an increased risk of death from CFA among metalworkers. Increased risks were also reported for various other types of occupational exposure, including working with wood, with organic solvents, and various dusts. Cadmium, chromium and lead production work were specifically implicated (35).

In the United States, Baumgartner et al. conducted a multicentre case-control study that included 248 patients with IPF and 491 age-, sex-, and geography-matched control subjects recruited using random-digit dialling. Metal dust was significantly associated with IPF (OR, 2.0; 95% CI, 1.0-4.0). Moreover, risk increased with the duration of exposure to metal dust, suggesting a dose- response relationship (13).

A multicentre hospital-based case-control study was performed in 2001, including 102 cases aged 40 years or over who had been diagnosed in accordance with the most recent criteria within the previous 2 years. Multiple logistic regression analysis was used to estimate the adjusted ORs and 95% CIs of IPF for single factors, with adjustment for age, sex and geographical region. Exposure to metal dust was significantly associated with an increased risk of IPF (OR 9.55; 95% CI 1.68-181.12) (59).

In an American study including 22 deceased and living uranium miners with evidence of diffuse IPF on a chest radiograph there was evidence of a UIP pattern of fibrosis with honeycombing in each of the five selected subjects whose lung tissue pathology slides (autopsies or wedge biopsies) were reviewed (5).

In 2007, a Swedish article described a case-control study to investigate which types of occupational exposure are associated with the development of severe pulmonary fibrosis and, in particular, IPF. The Authors did not find any increased risk to be associated with metal dust exposure (OR 0.9; 95% CI 0.51-1.59) (32). These results are in line with the study published by Harris et al., using data from death certificates in England and Wales, in which the authors found no evidence of an increased risk amongst individuals in occupations potentially exposed to metal dusts (33).

Further autopsy studies conducted in Japan revealed a higher level of inorganic particulate, such as silicon and aluminium, in the hilar lymph nodes of IPF patients than in those of controls (42).

Pinheiro and colleagues examined Proportional Mortality Ratios (PMR) of IPF patients coming from different work exposures; among them, metal miners scored a PMR 2.4 (95% CI 1.3-4.0), and manufacturers of fabricated structural metal products scored a PMR 1.9 (95% CI 1.1-3.1) (73).

In 2017, a multicentre hospital-based case-control study was conducted in Korea. With simple logistic regression analysis, exposure to metal dust and any exposure for >1 year in an occupational setting were significantly correlated with IPF (metal dust OR 4.00, 95% CI 1.34-11.97; any exposure OR 3.67, 95%CI 1.02-13.14). After adjustment for environmental and military exposure and smoking history, the OR for metal dust exposure was 4.97 (95% CI 1.36-18.17) on multiple logistic regression analysis (43).

In 2018, an Italian case-control study showed that working history in a metal- or a steel industry (OR 4.8; 95% CI 1.5-15.3) increased the risk of UIP (71).

Asbestos

The first study considering a possible relationship between asbestos and IPF is represented by a case-control study by Mullen et al. in 1998. They examined ILD cases recruited in two medical practices in Connecticut, USA, and orthopedic patients seen in the same practices as controls. A questionnaire was sent to cases and controls; after exclusions, a total amount of 15 cases and 30 controls were retained in the analysis. Two cases and one control reported occupational exposure to asbestos (OR 6.77; 95% CI 0.57-80.7). However, they included a small number of subjects, with a low response rate, and exposure information was self-reported (64).

In their case-control previously mentioned, Baumgartner and coworkers found an OR for asbestos equal to 1.1 (95% CI 0.6-1.9), based upon 26 exposed cases of IPF and 45 exposed controls (13).

A link between asbestos exposure and IPF was shown by Barber et al., who reported a relationship between mortality due to IPF in both sexes and asbestos imports, and this relationship was similar to that observed for mesothelioma mortality (10). This observation raises the question as to whether a proportion of IPF mortality is actually due to unrecognised asbestos exposure. Asbestosis and IPF can be clinically and radiologically indistinguishable, and thus physicians have to rely mainly on the exposure history provided by the patient in order to make a distinction between them (4, 9, 53). The Authors also suggest that there will be a better understanding of a potential relationship between asbestos exposure and IPF when the death rate for asbestos-related conditions has reached its peak in 2020 (9).

The same authors found a significant relationship between mesothelioma mortality and IPF mortality (but not with other common forms of interstitial lung disease) in 31 European countries; although limited by the available data, they also observed a significant association between asbestos use and mortality due to IPF for each country (11). According to the Authors, as recently stated in a letter-to-the-editor, these data are unexpected, since IPF is defined as a disease with no recognisable cause, and various sources of evidence support a hypothetical link between IPF and asbestos exposure, thereby satisfying biological plausibility criteria (12).

In 2007 the case-control study by Gustafson and colleagues found and OR for asbestos exposure equal to 0.8 (95% CI 0.44-1.47), considering 15 exposed cases and 72 exposed controls (32).

A survey performed by Schoenheit et al. in 2011 assessing quality of life in patients diagnosed with IPF and originating from different European countries included a specific evaluation of self-reported exposure. The results show that the patients reporting a history of asbestos exposure accounted for 20% of the total population considered (81).

Recently, Attanoos et al. examined UIP fibrosis pattern findings in lung samples of asbestos-exposed cohorts: they inferred that these findings are not to be labelled as atypical asbestosis, rather they are more likely to be cases of IPF (7).

These findings are consistent with the hypothesis that a proportion of IPF cases is likely to be caused by unknown exposure to asbestos. More research is needed in this area, particularly addressing patients known to have asbestos exposure, who are not currently considered candidates for new treatments for IPF.

Agriculture and organic dust

The large epidemiological study published in 2000 by Baumgartner and colleagues provided strong evidence correlating IPF with occupational exposure to farming and organic dust. Various types of occupational exposure were investigated in a multicentre case-control study of clinically and histologically diagnosed IPF. The results are based on 248 cases, aged 20-75 years, diagnosed at 16 referral centres between January 1989 and July 1993. There were 491 controls ascertained by random digit dialling and matched to cases for sex, age, and geographic region. Several occupational factors, adjusted for age and smoking in conditional multivariate logistic regression analyses, were significantly associated with IPF: farming (OR=1.6, 95% CI 1.0-2.5); livestock (OR=2.7, 95% CI 1.3-5.5); raising birds (OR=4.7, 95% CI: 1.6, 14.1); and vegetable dust/animal dust (OR=4.7, 95% CI: 2.1, 10.4). An interaction was detected between smoking and exposure to livestock (p=0.06) and farming (p=0.08) (13).

Of 181 patients with severe pulmonary fibrosis and respiratory failure reported to the Swedish Oxygen Register by completing an extensive postal questionnaire including 30 specific items regarding occupational exposure, 140 were considered as having IPF. The questionnaire was also completed by 757 control subjects. The data were stratified for age, sex and smoking history and the authors found an increased risk for IPF in men exposed to birch dust and to hardwood dust (see later in text) with only a slightly increasing risk for men raising birds (32).

In an Egyptian multicentre case-control study, Awadalla et al. showed that the risk of IPF was higher in females working in farming, raising birds and with occupational exposures to animal feeds, products and dusts and pesticides (8).

In 2018, Paolocci et al. observed that farmers, veterinarians and gardeners showed an increased risk of developing IPF (OR 2.73, 95% CI 1.47-5.10); in these patients, the risk of end-stage chronic hypersensitivity pneumonitis (CHP) cannot completely be ruled out, even when serum precipitins are negative, because of their known exposure to organic dust such as feed grain, bedding and materials of bovine origin (71). In a previous study, this group observed an increased IPF risk for bird fanciers (OR=2.14; 95%CI=1.13-4.05) (70).

A possible link between IPF and organic dusts is further supported by the fact that some patients diagnosed with IPF might have chronic hypersensitivity pneumonitis (CHP) instead, and exposure to organic antigens might have been occult. In a prospective observational study of consecutive patients with a diagnosis of IPF matching the current criteria, Morell et al. reported that a third of patients, on further history elicitation and through specific questionnaires, presented occult exposure to feather duvets or pillows and two-thirds of exposed patients were then diagnosed with CHP (63).

A possible misclassification of IPF rather than CHP was also noted by Ohtani et al., who found that patients with chronic bird fancier’s lung had UIP-like lesions with a temporally heterogeneous appearance at low magnification and alternating areas of normal lung structure, limited interstitial inflammation, fibrosis, and honeycomb changes, as well as fibroblastic foci. These patients did not experience any acute episodes and most of them had no specific antibodies or lymphocytosis in their BAL fluid, although the antigen-induced lymphocyte proliferation tests and laboratory-controlled inhalation provocation tests were positive. A significant percentage of patients had actually been diagnosed with IPF before visiting a specialised hospital (69).

An interesting association between IPF and organic dust exposure may derive from the microbiota factor, which, through its interaction with the host immune system, may contribute to the sequence of events that result in fibrosis (25). Recent evidence suggests that environmental exposure changes the microbiome in adult humans; in occupational literature, studies show that the upper respiratory tract of livestock workers is colonised by strain-specific Methicillin-Resistant Staphylococcus Aureus (MRSA) present in their working environment (28). Moreover, working with animals in a dairy environment appears to be associated with increased microbial diversity, as seen in the nasal microbiome of adult dairy farmers (85). Comorbidities such as gastro-oesophageal reflux could also be involved in inducing changes in the resident microbiome, or in contributing to bacteria transfer from the digestive system to the airways (23).

Wood

Since the early 1990s, observational studies have suggested that inhaled dust, especially wood dust (WD), may contribute to up as much as 12% of all cases of IPF. In 1990, a British case-control study with lifetime occupational data, obtained through a postal questionnaire, identified a substantially but not significantly higher risk for IPF in subjects exposed to WD (OR=2.94, 95% CI: 0.87-9.9) (82).

A subsequent, larger follow-up study showed an increased risk of IPF among workers exposed to WD (OR=1.71, 95% CI 1.01-2.92, p=0.048, when exposure was explored through a questionnaire; OR 2.22, 95% CI 1.26-3.91 for interview data), with a significant exposure-response correlation (OR per work-year of exposure=1.12; 95% CI 1.02-1.24, p=0.020) (34).

As previously said, the meta-analysis published by Taskar and Coultas identified the odds ratio for IPF in WD exposure as being 1.94 (95% CI 1.32-2.81) (88); a more recent mortality study by Pinheiro and colleagues identified the OR in WD exposure equal to 5.3 (95% CI 1.2-23.8) with a Proportional Mortality Ratio (PMR) of 4.5 (95% CI 1.2-11.6) for workers employed in “wood buildings and mobile homes” (73).

Several case reports have illustrated situations in which patients received a diagnosis of IPF despite the fact that their clinical history reported an ascertained and relevant exposure to wood dust, suggesting the presence of wood dust-related lung fibrosis instead (77).

The Egyptian case-control by Awadalla and colleagues found that exposure to wood dusts and wood preservatives was associated with an OR of 2.71 (95% CI 1.01-7.37) in men and 4.32 (95% CI 0.84-22.12) in women, respectively (8).

In their case-control study, Paolocci et al. did not find a significant correlation between wood dust exposure and risk of IPF (71); these data were in contrast with the findings of an older study by Gustafson et al., in which exposure to birch dust or hardwood dust was associated with an increased risk of IPF, with an OR of 2.7 (95% CI 1.30-5.65) for birch dust and OR of 2.7 (95% CI 1.14-6.52) for hardwood dust (32). According to Paolocci et al., this difference could be explained by the different geographical context of the two populations, and the different productive settings examined (71).

Dental personnel

The American Centres for Disease Control and Prevention (CDC) reported a cluster of IPF among dental professionals at a specialty clinic in Virginia. More specifically, between September 1996 and June 2017, nine (1%) out of 894 patients treated for IPF at a single tertiary-care centre in Virginia were identified as dental personnel (67).

In the past, a query of the National Occupational Respiratory Mortality System for 4 separate years (1999, 2003, 2004, and 2007) for “other interstitial pulmonary diseases with fibrosis” (which would include IPF) listed as the underlying or contributing cause of death, revealed 35 decedents who were classified as having worked in a “dental practice” and 19 classified as having the occupation “dentist,” (Respiratory Health Division, CDC, unpublished data, 2017).

Dentists and other dental personnel experience unique occupational exposures, including exposure to infectious organisms, dusts, gases, and fumes. It is possible that occupational exposure contributed to this cluster (48).

Sand and silica

The first evidence of an association between sand/silica exposure and IPF dates back to 1990s, when Monso et al. conducted a mineralogical analysis of lung tissue on 25 samples from patients who had been diagnosed as having IPF; it revealed that these patients presented an excessive level of silica (62).

Hubbard et al. conducted a larger case-control study on 218 patients and 569 community-matched control subjects from the Trent region of the United Kingdom. Data on occupational dust exposure were collected using a postal questionnaire and a telephone interview. The results confirmed a significant, independent, and dose-response increase in the risk of IPF associated with stone/sand dust (34).

Baumgartner and colleagues performed a multicenter case-control study in which they assessed, among other working exposures, the activity of stone cutting and polishing. It showed an OR for IPF equal to 3.9 (95% CI 1.2-12.7), supporting the hypothesis of an existing link between these dusts and lung fibrosis development (13).

Taskar and Coultas conducted a meta-analysis and demonstrated a significant association between the increased risk of IPF and sand/stone/silica exposure (OR, 1.97; 95% CI, 1.09-3.55) (88).

Another multicenter case-control study performed in Egypt in 2012 did not find an increase in OR for IPF among workers exposed to stone dust (9).

A Korean study showed that occupational and environmental exposure to stone/sand/silica alone was significantly associated with chronic fibrosing IIP (OR 4.98; 95% CI 1.05-23.63) adjusted for age and smoking (41).

Solvents

In 2000, Baumgartner and colleagues found an OR 1.4 (95% CI 0.9-2.2) for IPF in patients exposed to diesel exhausts (13).

In 2014 Paolocci et al. noticed an increased OR for IPF (9.65; 95% CI 1.32-70.34) in workers exposed to cutting oils (70).

Experimental studies on animals have shown that exposure to solvents produces changes in the lung similar to those found in CFA. Occupational exposure to a large number of solvents has been associated with the development of systemic sclerosis. Pulmonary involvement in systemic sclerosis is morphologically indistinguishable from CFA, and a close relationship between the two diseases would seem to exist. This close relationship and the experimental work linking solvent exposure to CFA suggest that solvent exposure may be one cause of its misunderstood causes.

IPF and environmental exposure

The best practice for an individual patient with IPF should now include education regarding awareness and avoidance of excess ambient air pollution. Indeed, published data clearly show the impact of air pollution on both IPF incidence and prognosis.

The previously cited work by Schoenheit et al. also considered self-reported exposure to air pollution in European IPF patients, finding that 38% of these patients reported this kind of background (81).

An Italian study investigated the association between chronic exposure to NO2, O3 and particulate matter with an aerodynamic diameter <10 μm (PM10) and IPF incidence in Northern Italy between 2005 and 2010, reporting the incidence of IPF as positively associated with NO2 concentrations (19).

Ambient air pollution, particularly increased exposure to O3 and NO2 primarily from motor vehicle exhaust fumes, increases the risk of acute exacerbation of IPF. More specifically, acute exacerbations were significantly associated with antecedent 6-week increases in mean level, maximum level and number of exceedances above accepted O3 standards (Hazard Ratio (HR) 1.57, 95% CI 1.09-2.24; HR 1.42, 95% CI 1.11-1.82; and HR 1.51, 95% CI 1.17-1.94, respectively) and NO2 (36).

The same Authors reported that lower mean predicted FVC % was consistently associated with increased mean exposure to PM10 in the 2 to 5 weeks preceding the clinical measurements; lower mean predicted FVC % over the study period was inversely related to the mean levels of NO2, PM2.5, and PM10 (37).

Another part of the previously cited MESA study by Sack et al. reported that environmental pollution (fine particulate matter (PM2.5), nitrogen oxides (NOx), nitrogen dioxide (NO2) and ozone (O3)) assessed at the home of community-dwelling participants correlated with qualitative (ILA) and quantitative (HAA) HRCT abnormalities suggesting the presence of subclinical ILD (80).

In 2018, Winterbottom et al. assessed the relationship between average PM10 levels and rate of functional decline in 135 IPF patients from 2007 to 2013. They found a significant correlation between average PM10 levels and increased rate of FVC decline in these subjects during the study period, with each μg/m3 increase in PM10 corresponding with an additional 46 cc/y decline in FVC (P=0.008) (90).

In a recent French study, IPF patients were selected from the French cohort COhorte FIbrose for a follow-up study assessing factors associated with disease progression, Acute Exacerbations (AE) and death. AE events were significantly associated with a higher mean concentration of O3 during the exposure period, with an HR of 1.47 (95% CI 1.13 to 1.92) per 10 µg/m3 (p=0.005). No association was observed between AE and NO2, PM10 and PM2.5. Disease progression was not associated with increased cumulative concentrations of NO2, O3, PM10 or PM2.5. Mortality was significantly associated with increased cumulative exposure to PM10 and to PM2.5 (83).

A Finnish study assessing the presence of coal dust pigment and inorganic particulate matter in 73 lung tissue samples from the national IPF registry found that all of them contained varying amounts of coal dust pigment and inorganic particulate matter. Interestingly, samples from southern Finland, which is more densely populated and presents higher levels of fine particulate matter in the air, more frequently showed a higher dust content than samples coming from northern Finland. The highest particulate content scores were observed in samples from patients with known exposure to inorganic dust (55).

IPF and smoking

Cigarette smoking seems to be the most strongly associated risk factor in both sporadic IPF and familial pulmonary fibrosis, as stated by Steele and colleagues in 2005: their work assessed an OR for familial IPF equal to 3.6 (95% CI 1.3-9.8) among ever-smokers (87). Current or former smokers have consistently been overrepresented in IPF.

In 2014, a Swedish study assessed the effects of smoking, gender and occupational exposure on the risk of developing severe pulmonary fibrosis (PF). The results showed that the effects of smoking were amplified by male gender and occupational exposure, OR 4.6 (95% CI 2.1-10.3) for PF, and OR 3.0 (95% CI 1.3-6.5) for IPF, compared to non-exposed women. Higher cumulative smoking exposure was linearly associated with increased risks. Smoking also had a dose-related association with the risk of severe PF: compared with smoking less than 10 pack-years, smoking ≥20 pack-years was associated with increased risk of PF and IPF, with an OR of 2.6 (95% CI 1.4-4.9) and of 2.5 (95% CI 1.3-5.0), respectively. Men with a history of smoking and occupational exposure constituted a particularly high-risk group (24).

A Finnish study examined 45 non-smokers, 66 ex-smokers and 17 current smokers with IPF. Current smokers were younger at baseline compared to non-smokers and ex-smokers. The median survival of non-smokers and current smokers was longer (55.0 and 52.0 months, respectively) than that of ex-smokers (36.0 months) (p=0.028 and 0.034, respectively) (39).

An Australian study reported on 503 cases with IPF and 902 controls. Current or past tobacco smoking was associated with an increased risk of IPF with an OR=2.20 (95% CI 1.74-2.79). More specifically, occupational exposure to passive smoke (OR=2.04; 95% CI 1.16-3.60) and respirable dust (OR=1.41; 95% CI 1.07-1.88) were associated with an increased risk, whereas other forms of occupational exposure to specific organic, mineral or metal dusts were not associated with IPF (1).

Discussion

Determining exposure in IPF is always challenging, since this disease is usually diagnosed after a lifetime of chronic and often mixed exposure, making it difficult to identify potentially relevant causes. Furthermore, in clinical settings, appointment time constraints, patient unawareness and complex exposure histories are all barriers to obtain a comprehensive environmental and occupational exposure history. Simple standardised questionnaires could help physicians to identify those patients who deserve second-level assessment by an occupational physician. In addition to detailed questionnaires, there is a need for better objective measurements of exposure. Having a trained professional, e.g. an industrial hygienist, assessing the home and work environment, could better identify and quantify exposure. Individual circulating or other biomarkers of exposure could help determine the internalised dose and sensitisation. As many of these forms of exposure are ubiquitous in the environment, it would be important to determine whether there is individualised susceptibility to certain types of exposure.

Better questionnaires and objective assessments of environmental exposure may also help to distinguish individuals with fibrotic lung diseases other than IPF. In particular, chronic exposure to occult environmental factors can cause chronic hypersensitivity pneumonitis with UIP features.

As recently stressed by Lee et al. when dealing with the occupational burden of non-malignant respiratory diseases, the occupational burden is greater than previously acknowledged, and could affect all types of ILD (16, 45). The Authors insist in particular on the importance of occupational medicine training for all pulmonary and clinical practitioners performing routine exposure assessment of ILD patients; they also suggest the importance of future investigations regarding environmental and occupational exposure, by collecting data in multicentre studies and national registries. These initiatives, conducted through structured interviews by trained professionals, should highlight potentially modifiable ILD-predisposing factors, such as environmental antigens, and assess the impact of on-going exposure on the clinical course of ILD (44).

The benefits of eliminating exposure need to be explored as a therapeutic approach. For example, treatment of GERD may slow progression and perhaps improve survival in IPF (78); however, there is a paucity of data investigating similar outcomes for environmental exposure. Measurements performed during home and workplace assessments by a trained professional could lead to environmental remediation and potential improvements in health. The efficacy of such interventions in reducing respiratory symptoms has been demonstrated for asthma, but no studies have been performed in individuals with IPF.

Conclusions

Several sources of evidence, therefore, including investigations of pathogenesis and observational studies, support the hypothesis that environmental agents may have an aetiological role in IPF, which probably represents the end-stage of antecedent acute or subacute disease provoked by a variety of causes. Moreover, current sources of evidence firmly support the need for a multidisciplinary assessment of IPF patients, also involving occupational physicians to complete the clinical history with a study on environmental and occupational exposure. This approach may also help scientific research to collect epidemiological data useful for better understanding this challenging disease.

No potential conflict of interest relevant to this article was reported by the authors