More importantly, the role of tobacco smoking is better identi ed as a risk factor for the wide spectrum that comprises IPF, CPFE, and chronic obstructive pulmonary disease with emphysema [9], altogether belonging to smoking-induced lung diseases [10]. Lung cancer [11] and PH may also be part of the spectrum. Importantly, it has become clear that CPFE encompasses a number of radiological and pathological patterns; such heterogeneity and lack of consensus de - nition and terminology until recently have hampered comparisons between clinical series from different groups and clinical research.

It is still unclear whether CPFE corresponds to the co-­ incidental association of IPF and emphysema by chance or to a genuinely unique entity with causal links between the two pathologies. Much remains to be learned about the pathophysiology of this syndrome, diagnostic boundaries with IPF, and management. However, CPFE clearly presents with a characteristic clinical and functional pro le, and an increased risk of precapillary PH, lung cancer, and mortality, not to mention signi cant relevance for clinical care, approach to monitoring, and trial design. Therefore, CPFE is considered a clinical syndrome [12].

Epidemiology and Etiologies

Tobacco Smoking and Male Sex

Emphysema has been reported in 21–33% of patients with IPF (up to 67% in some series) [13–23], most of whom are current or past tobacco smokers. A lower prevalence (8%) was found in a series from the USA [13], likely accounting for the lesser use of tobacco in this population. In addition to variations in smoking, differences in the prevalence of emphysema between studies and countries are likely due to de nitions used for CPFE and to underlying genetic susceptibility. Tobacco smoking is by far the predominant etiology of the CPFE syndrome, with a smoking history present in 98% of patients [24], and a mean tobacco history of about 40 pack-years [8, 11, 25, 26]. Interestingly, exposure to tobacco smoke may cause both brosis and emphysema in an animal model [27]. A comprehensive review of common and distinct mechanisms of disease pathogenesis in IPF and chronic obstructive pulmonary disease can be found elsewhere [28, 29].

CPFE associated with tobacco smoking in the absence of any other potential etiologic factor has a 9:1 male:female ratio [24]. Both IPF and emphysema also separately predominate in males [30–33], but the male predominance of CPFE is particularly important. It is not unknown whether the male predominance of the CPFE syndrome is related solely to the higher prevalence of tobacco smoking in men, and to the link between tobacco smoking and both emphysema and

IPF. Smoking rates in men versus women generally tend to become similar over time as the prevalence of smoking falls in western countries, and likely do not explain the totality of the gender difference in CPFE. Emphysema generally precedes brotic ILD. A large epidemiologic study would be required to study the respective role of gender and tobacco smoking in predisposing to CPFE.

Patients with a smoking history as the only etiologic or risk factor for developing CPFE will hereafter be referred to as “CPFE” to distinguish them from those with more clear-­ cut etiologic factors for ILD. CPFE in the absence of tobacco smoking should suggest the presence of connective tissue disease (which is probably by itself a risk factor for CPFE) [17, 34–36] or hypersensitivity pneumonitis [37–40].

Genetic Predisposition

It has been postulated [12] that a particular genetic background may predispose a subset of smokers to develop the typical syndrome of CPFE, in much the same was as the risk of emphysema or IPF in smokers is through to depend on largely unknown genetic factors that likely impact lung aging and cell senescence [28, 41–43]. The potential genetic basis of CPFE is only beginning to be explored, with a few cases of mutations conferring a Mendelian risk of CPFE or IPF appearing in the literature (Table 33.1). However, it is likely that CPFE may result from both a genetic predisposition and a second hit with tobacco smoking or exposure to other aerocontaminants or risk factors. Identifying genetic mutations that render individuals vulnerable to CPFE would provide great pathophysiological insights into the disorder, with the caveat that epigenetic alterations almost certainly also play a major role [62].

The CPFE syndrome was reported in a patient with familial ILD carrying a mutation in the surfactant protein C gene [49]. Typical CPFE was found in a 41-year-old non-smoker

Table 33.1  Genetic variants and polymorphisms that are associated with CPFE

patient with a pathological mutation in the ABCA3 gene [50]. Some “emphysema-like” lesions were further observed in patients with familial (genetic) IPF with mutations in the surfactant protein C gene [51, 52] or in the telomerase complex [44, 45]. As telomere length is reduced both in COPD and IPF patients, telomeres would be expected to be shorter than normal in patients with CPFE [12, 24], a concept that requires further study [63]. Germline mutations were identi ed in telomerase as a Mendelian risk factor for COPD susceptibility that clusters in families with autosomal dominant telomere-­mediated disease including pulmonary brosis [46]. For example, a CPFE syndrome was present in a family with TERT mutation, with one individual presenting nonspeci c interstitial pneumonia (NSIP) and emphysema with scattered ill-de ned granulomas, another with usual interstitial pneumonia (UIP) and emphysema, and one (exposed to wood dust) who had emphysema with mild pulmonary brosis of the lower lobes [47]. As observed in familial pulmonary brosis (without emphysema), mutations in genes associated with surfactant or telomerase, when present in a family, can be associated with a diversity of patterns, especially UIP and NSIP. Incidentally, scattered cystic lesions and ILD with mostly ground-glass attenuation can be observed in patients with neuro bromatosis, albeit with an imaging pattern that is not typical of that of (tobacco-related) CPFE syndrome [64]. Gene mutations and genetic modi ers that confer a speci c predisposition to CPFE have not been explored.

Systemic Diseases

The CPFE syndrome may occur in virtually any of the connective tissue diseases, with no clear differences in presentation according to the individual systemic diseases or autoantibody type [65]. The prevalence of coexistent emphysema has been formally studied in only a few cohorts of patients with connective tissue diseases and associated ILD. In one study, the prevalence of emphysema at HRCT was 23% in patients with ILD associated to connective tissue diseases and 44% in patients with IPF (p = 0.05) [66]. The emphysema score at HRCT was signi cantly lower in patients with connective tissue diseases and a pathological UIP pattern than in those with IPF in a retrospective study, however, a difference in smoking history could have accounted for this difference [67].

In a study of 34 patients with typical CPFE occurring in connective tissue diseases [26], the predominant underlying diseases were rheumatoid arthritis (Fig. 33.1) and systemic sclerosis (Fig. 33.2) (limited or diffuse cutaneous variant), possibly owing to the relative frequency of these diseases in the general population. The diagnosis of CPFE followed that

of connective tissue disease by a median of 4 months in two thirds of the patients, whereas both diagnoses were simultaneous in the other cases [26]. The onset of a CPFE syndrome before the occurrence of the connective tissue disease is very rare [26].

In 150 consecutive patients with rheumatoid arthritis [68], 19% had ILD, 15% had the so-called emphysematous bullae, and 8% (12 out of 150) had both ILD and emphysema. In 116 never smokers with rheumatoid arthritis-asso- ciated ILD, emphysema was present on HRCT in 27% [34]. Emphysema was observed in 24% of 63 patients with rheumatoid arthritis and ILD [69], and was signi cantly more frequent among patients with an HRCT pattern of UIP (38%) than in other ILD patterns; patients with UIP also had a higher prevalence of smoking and a greater smoking history than those with other patterns. Antoniou et al. reported the presence of emphysema on chest HRCT in 48% of ever-smoker patients with rheumatoid arthritis and ILD, and in 35% of ever-­smoker patients with IPF, despite median smoking histories of less than 25 pack-years in both cohorts [17]. These data suggest that subjects with rheumatoid arthritis who smoke may be particularly vulnerable to emphysema. In addition, patients with rheumatoid arthritis and a history of smoking had a “coarser”brosis at imaging than never smokers. In rheumatoid arthritis, interaction of genetic background (e.g., the human leukocyte antigen DRB1 shared epitope) with environmental exposures (especially tobacco smoking) and autoimmunity (e.g., anti-cyclic citrullinated peptide antibodies) is well established [70]. Briefy, tobacco smoking is responsible for infammation and citrullination of proteins within the lung, a post-translational modi cation that converts l-arginine residues into l-citrulline residues through the activity of peptidyl arginine deiminase. This modi cation is thought to alter protein folding and charge, and enhance degradation by proteases and exposure of cryptic epitopes, which in turn increases the risk of developing anti-cyclic citrullinated peptide autoantibodies produced in the lung. In this model, the end result is an autoimmune response that leads rheumatoid arthritis [71], in which antibodies developed against modi ed self-proteins in the lungs leads to emphysema in the lungs and infammation of the joints [72]. Tobacco smoking increases the incidence and severity of rheumatoid arthritis [73], including an enhanced risk of developing extra-articular complications [74], possibly including ILD [75, 76]. Acute exacerbation of ILD is known to occur in patients with CPFE and rheumatoid arthritis, occasionally upon institution of drug therapy [77]. Whether emphysema in rheumatoid arthritis is also a consequence of autoimmune processes promoted by cigarette smoking is unknown. Of note, anti-elastin antibodies are not found in patients with CPFE [78].

Fig. 33.1  CPFE syndrome at chest HRCT in a patient with rheumatoid arthritis (male, smoker). (a) upper lobes showing centrilobular and paraseptal emphysema; (b) mid regions of the lungs, showing predomi-

nantly paraseptal emphysema, with thickening of the interlobular septa; (c) lower zones showing usual interstitial pneumonia pattern with reticulation, honeycombing, and traction bronchiectasis

Fig. 33.2  CPFE syndrome at chest HRCT in a 28-year-old female patient with severe systemic sclerosis and high-titer anti-U1-RNP autoantibodies, with a smoking history of less than 5 pack-years. (a) upper

lobes demonstrating centrilobular and paraseptal emphysema; (b) lower zones showing nonspeci c interstitial pneumonia pattern with reticulation, ground-glass opacities, and mild traction bronchiectasis