Imaging Diagnosis of Interstitial Pneumonia with Emphysema (Combined Pulmonary Fibrosis and Emphysema)

Based on clinical and radiological findings, Cottin defined combined pulmonary fibrosis and emphysema (CPFE) as pulmonary emphysema in the upper lungs and interstitial pneumonia in the lower lungs with various radiological patterns. Pathologic findings of CPFE probably corresponded with diffuse interstitial pneumonia with pulmonary emphysema, emphysema with fibrosis, and the combination of both. We described reported radiological findings of CPFE.


Introduction
Interstitial pneumonia (IP) with pulmonary emphysema has long been a topic of controversy because we have not decided if the disease is simple coincidence of IP and emphysema or IP and emphysema may be caused by common etiology. In 1990, Wiggins et al. [1] presented 8 cases in the first report in the English literature. In 2005, Cottin et al. [2] described combined pulmonary fibrosis and emphysema (CPFE) based on clinical and radiological findings and noted some characteristic clinical features. Description of the pathological mechanisms in such cases is very limited.
Defined by clinical and imaging findings, CPFE is one of clinicoradiological syndromes and probably includes several kinds of pulmonary fibrosis, so its clinical course and prognosis various [3]. It is unclear how CPFE should be categorized under interstitial pneumonia to ensure the integrity of idiopathic interstitial pneumonia (IIP) classification. We herein review and summarize the imaging features of CPFE.
Coexistence of pathologic findings of LCH, DIP, or RBILD in the same patient with smoking habit has been reported, and these three disease entities were sometimes categorized as smoking-related interstitial lung disease (SRILD) in narrow definition [7,8]. However, in a broad sense of term, SRILD may include pulmonary emphysema, IPF, and CPFE.
Smoking and/or dust inhalation are also important factors in the development of idiopathic interstitial pneumonias [9], and most patients with idiopathic pulmonary fibrosis (IPF) have a history of smoking [10]. Overt pulmonary emphysema complicates IPF in approximately 30% of patients [11]. Iwai et al. calculated an odds ratio of 3.4 relating IPF to smoking or dust inhalation [12]. A greater prevalence of nonspecific interstitial pneumonia (NSIP) is also reported in smokers [13], with DIP and RBILD generally confined to current smokers [8,[14][15][16].
Smoking has been closely related to DIP, RBILD, and LCH [8,17], and features of all three are reported in pathologic specimens of one individual [16,17]. These 3 diseases and IPF are also generalized as smoking-related interstitial lung disease (SRILD) [4,5,18].
Based on clinical and imaging findings, Cottin et al. also rediscussed CPFE as pulmonary emphysema, defined as lowattenuation areas surrounded by normal lung with thin or no wall and/or multiple bullae predominantly in upper lungs, and interstitial pneumonia in the lower lungs with imaging patterns of various IPs, including usual interstitial pneumonia (UIP), NSIP, RBILD, cryptogenic organizing pneumonia (OP), and others [8].
Cottin's article reported characteristic clinical features including poor prognosis, frequently complicated by pulmonary hypertension [19,20], preserved forced expiratory volume in one second (FEV 1.0), and severely impaired diffusion capacity [2,21]. Mejía et al. also reported decreased survival of patients with IPF and emphysema with pulmonary hypertension [10]. However, we believe that the prognosis of interstitial pneumonia is better than that of IPF/UIP [22], despite a high prevalence of lung cancer reported in cases of CPFE [23,24] and even IPF/UIP [25].
In a mass screening of smokers in the United States, the COPD Gene Study Group reported interstitial shadow or radiologically overt IP on high-resolution computed tomography (HRCT) in approximately 8% of 2500 subjects with chronic obstructive pulmonary disease (COPD) [26,27]. They categorized interstitial changes on HRCT into following four types: centrilobular, subpleural, mixed abnormal opacity, or radiologically overt IP. On pulmonary function tests, total lung capacity was increased in subjects with centrilobular or mixed opacity but not in those with subpleural abnormal opacity and overt IP. These results suggested a correlation between interstitial lung change and smoking. Pulmonary injury may cause pulmonary destruction in emphysema or abnormal repair processes in fibrosis. Smoking can cause simultaneous pulmonary emphysema and fibrosis. Bates et al. [28,29] described changes in the lung matrix as a common pathway of destructive change and fibrosis. However, CPFE may not be a single disease and may include heterogeneous diseases [3].

Pathologic Features of CPFE
Pathologic features of SRILD in narrow definition (DIP, RBILD, and LCH) have been described [7,[30][31][32]. Desquamative interstitial pneumonia (DIP) shows accumulation of alveolar macrophages within the alveolar lumen and mild fibrotic thickening of the alveolar septa, but advanced fibrosis is uncommon. RBILD shows accumulation of alveolar macrophages laden with brown pigment within the respiratory bronchioles. Langerhans cell histiocytosis results when Langerhans histiocytes and eosinophils form granulomas in the upper lung, cavitary nodules, and cysts by the necrosis of granulomas.
Although CPFE is closely related to smoking, its pathologic features differ. In noncancerous portions of lung specimens obtained by lobectomy for bronchogenic carcinoma, Kawabata et al. [33] so frequently observed relatively thickwalled cysts that they designated the finding as air space enlargement with fibrosis (AEF). They depicted AEF related to smoking index and possibly induced by smoking. Katzenstein et al. [34] found similar hyalinizing collagenous fibrotic change of the alveolar septa along the emphysematous walls and respiratory bronchioles in the lungs of smokers; change that they designated occult fibrosis of smokers or smokingrelated interstitial fibrosis.
Wright et al. [35] described 2 types of smoking-related interstitial lung disease-diffuse form, which includes overlapping pulmonary emphysema and IP, such as UIP or NSIP, and localized form, which includes emphysema with fibrosis, AEF, or occult fibrosis of smokers. Although pulmonary emphysema is defined as air space without fibrosis, Wright described frequently varying degrees of pulmonary fibrosis in the alveolar septa around the emphysema wall. Radiological evaluation aids differentiation of diffuse and localized forms. Fibrotic change is also identified along bronchioles in addition to alveolar septa around pulmonary emphysema [31].
However, pathologic features of CPFE have not been fully studied; thus further investigation is mandatory to show pathologic characteristics of CPFE.

Imaging Features of CPFE
Imaging features of CPFE have not been fully described. Radiological findings include centrilobular and/or paraseptal emphysema evident in the upper lungs and IP in the lower lungs [30,36]. On CXR, volume of lung maintains although there is reticular opacity in bilateral lower lung fields ( Figures  1(a), 2(a), and 7(a)). CT patterns of interstitial pneumonia mimic those of IIP in some patients and are unclassifiable in others. Cottin et al. noted that the IP pattern may mimic NSIP (Figure 1), UIP (Figures 2 and 3), OP, RBILD, and other such entities.
When pulmonary emphysema is present, the radiological pattern of interstitial pneumonia does not necessarily show characteristic one. Akira  physema and noted that the pattern of pathologically proven NSIP with emphysema might mimic the pattern of UIP on HRCT [37]. Radiological features of CPFE include large relatively thick-walled cysts in addition to pulmonary emphysema ( Figure 4). The cysts may be related to large paraseptal emphysema in the upper lungs just beneath the chest wall. Some large cysts may grow within the areas with interstitial pneumonia. These thick-walled cysts probably correspond to emphysema with fibrosis described by Wright et al. [35] or air space enlargement with fibrosis by Kawabata et al. [33], but pathological confirmation as well as determination of the operation of other mechanisms of cyst formation is needed.
As above mentioned, Wright et al. classified smokingrelated interstitial changes into two types: emphysema with fibrosis (localized form) and apparent interstitial pneumonia (diffuse form), and HRCT seems to be useful in differential diagnosis of the two forms [35]. However, mixture of the two forms is sometimes observed; apparent interstitial pneumonia pattern with pulmonary emphysema mixed with large thick-wall cysts is sometimes observed (Figures 1, 4, and 6) on HRCT images. Clinical, radiological differences between two forms classified by Wright et al. must be investigated in further study. Other findings include centrilobular relatively thickwalled cyst accompanying interstitial change or centrilobular nodules/subpleural curvilinear opacity. Centrilobular lesions correspond with airway-centered fibrosis, that is, fibrotic changes along the respiratory bronchioles ( Figure 5).
CPFE may show very high prevalence of lung cancer; the possibility of lung cancer must be remained in the radiological differential diagnosis of nodules/mass coinciding with CPFE ( Figure 4).
Cottin et al. frequently observed pulmonary hypertension with poor prognosis, but our clinical experience suggests less frequent pulmonary hypertension and relatively good prognosis of interstitial pneumonia [23]. The causes of the difference are unclear but may represent difference in race or disease severity.
CT features of pulmonary hypertension are dilatation of the central pulmonary arteries, enlargement of the rightsided heart chambers, reduced number of peripheral pulmonary artery branches, and mosaic attenuation of the lung parenchyma [38] (Figure 6). However, changes in peripheral pulmonary vasculature and parenchyma are unclear when destructive and fibrotic changes in the pulmonary parenchyma are prominent.
Acute exacerbation is less frequent for CPFE than idiopathic pulmonary fibrosis, but its precise frequency has not been determined (Figure 7). Neither is the precise risk of drug-induced lung injury known. Because preexisting IP is a risk factor for drug-induced lung injury, CPFE may affect indication for anticancer drug treatment [39]. In patients with acute lung injury immediately following surgery for lung Pulmonary Medicine cancer, acute exacerbation of occult minimal IP might be suspected because this minimal fibrosis is frequently identified in postoperative patients [40]. The radiological features of CPFE correspond to pathologic features of destructive pulmonary changes in smokers, such as emphysema with fibrosis and overt interstitial pneumonia with emphysema. Further investigation is needed to clarify clinical characteristics, complications, and prognosis of CPFE.