Cigarette smoking has been recognized to be a risk factor for developing interstitial lung disease as well as chronic obstructive pulmonary disease. Weiss and others found that the presence of small irregular opacities on chest roentgenograms from cigarette smokers correlated with the duration and degree of cigarette smoking.1–4 Respiratory bronchiolitis (RB) was defined in 1974 as a distinct histopathologic disease entity that contributes to small airway dysfunction in cigarette smokers.5 This is a mild inflammatory reaction commonly noted in asymptomatic cigarette smokers. The pathologic findings of RB are nonspecific, and similar abnormalities have been demonstrated in patients with other conditions. A clinicopathologic syndrome called respiratory bronchiolitis-associated interstitial lung disease (RB-ILD) has been described among current or former cigarette smokers, mainly in the fourth or fifth decade of life, with an average Brinkman index of >600 and has features consistent with the diagnosis of an interstitial lung disease.1,6–8 The lesions observed in RB-ILD are more pronounced with regard to the clusters of brown-pigmented macrophages found in respiratory and membranous bronchioles than those in healthy cigarette smokers.
RB-ILD and desquamative interstitial pneumonitis (DIP) are thought to be smoking-induced interstitial lung diseases.9,10 However, the clinical presentation of RB-ILD and DIP seems to be nonspecific, although well-defined pathologic features have been described. The radiographic findings might not distinguish RB-ILD from DIP.1,7,11,12 Many of the reported cases of RB-ILD have also been initially misdiagnosed as DIP.7 Therefore, in the present study, the clinical characteristics, including bronchoalveolar lavage fluid (BALF) findings, of patients with RB-ILD were compared with those of patients with DIP.
MATERIALS AND METHODS
Five male patients histologically diagnosed as having RB-ILD (age 55.3 ± 2.7 years, mean ± standard error [SE]) and 7 patients with histologically confirmed DIP (5 males and 2 females, age 53.6 ± 2.7 years, mean ± SE) were evaluated in this study. Diagnosis was confirmed based on histologic specimens obtained by video-assisted thoracoscopic surgery. They were referred to our hospital between 1998 and 2001 because of an abnormal shadow on chest radiographs or respiratory symptoms. Present and past histories regarding occupation, symptoms, smoking, and medication use were obtained. A physical examination, blood examination, pulmonary function test, arterial blood gas analysis, chest radiography, chest high-resolution computed tomography (HRCT), gallium (Ga) scintigraphy, and bronchoalveolar lavage were conducted. None had a history of significant occupational or environmental exposure or of collagen vascular disease. BALF was obtained from all patients.
Data are presented as mean ± SE. Laboratory findings were compared with the Mann-Whitney U test. P < 0.05 was considered statistically significant.
RB-ILD showed the patchy bronchiolocentric distribution (Fig. 1). Respiratory bronchioles, alveolar ducts, and peribronchiolar alveolar spaces were irregularly filled by pigmented macrophages. Alveolar septa were thickened as a result of mild peribronchiolar fibrosis. Alveoli away from bronchioles were normal or slightly hyperinflated.
DIP showed large numbers of pigmented macrophages in the alveoli (Fig. 2). Alveolar septal thickening was mild to moderate and tended to be diffuse and uniform from field to field, and the airway epithelium showed goblet cell hyperplasia and/or a metaplastic cuboidal epithelium. The minimal or mild associated fibrosis without honeycomb was found. DIP could be distinguished from RB-ILD by diffuse and uniform involvement rather than a patchy, bronchiolocentric distribution.
The clinical characteristics and main laboratory findings of the 5 patients with RB-ILD and the 7 patients with DIP are listed in Table 1. The mean age was 55.3 in RB-ILD and 53.6 in DIP, and both groups were male-dominated. All patients with RB-ILD and DIP were current cigarette smokers except 1 with DIP.
The clinical features are compared in Table 2. Digit clubbing was not found in patients with RB-ILD, whereas it was observed in 3 patients with DIP. Abnormal respiratory sounds were audible in 1 patient with RB-ILD and all with DIP. As for subjective symptoms, productive cough was noticed more in RB-ILD than in DIP patients. All patients with DIP experienced dyspnea on exertion, whereas 1 patient with RB-ILD experienced it. None of the patients fulfilled the criteria of connective tissue disorders, which includes rheumatoid arthritis, systemic sclerosis, polymyositis/dermatomyositis, systemic lupus erythematosus, Sjögren’s syndrome, and mixed connective tissue disorders.
Regarding laboratory findings, C-reactive protein values were elevated (>0.4) in 6 of 7 patients with DIP but in 2 of 5 patients with RB-ILD, whereas leukocytosis (WBC count >8000) was more observed in patients with RB-ILD than in patients with DIP. Lactic dehydrogenase, which is one of the general indicators of active inflammation, was not different between the groups. It is noteworthy that antinucleotide antigen (ANA) was detected in 6 of 7 patients with DIP and in 3 of 5 patients with RB-ILD, although the degree of the increase was mild in patients with RB-ILD. The average PaO2 and PaCO2 values were not statistically different (72.7 ± 4.9 and 42.9 ± 4.3 in RB-ILD; 76.4 ± 5.5 and 39.6 ± 2.4 in DIP, respectively). In pulmonary function tests, %VC was higher in the RB-ILD group, whereas FEV1.0% was higher in the DIP group (P < 0.05).
Radiologic findings are compared in Table 2. Diffuse, bilateral, and basilar small linear and nodular opacities with minimal patchy ground-glass opacities were observed in both patients with RB-ILD and those with DIP on chest HRCT.
In RB-ILD, HRCT showed diffuse centrilobular, nodular or patchy, ground-glass opacities (Fig. 3). No appreciable honeycombing was observed. Peripherally distributed, irregular linear opacities, ground-glass opacities, bronchial wall thickening, and centrilobular emphysema were observed.
In DIP, chest CT revealed diffuse, bilateral panlobular ground-glass opacities. Centrilobular emphysema was also observed, whereas neither irregular linear opacities nor honeycombing was observed (Fig. 4).
Ga scintigraphy showed mild, diffuse uptake in the same lung areas that abnormal CT findings were recognized, both in RB-ILD and DIP.
The total cell number of BALF increased in both groups. Eosinophils were observed only in DIP. The proportion of macrophages was greater in RB-ILD, whereas that of neutrophils was greater in DIP. Cell differentiation of BALF cells was different between DIP and RB-ILD (Fig. 5).
All patients were required to quit smoking. All patients were followed in at least 3 months after the initial diagnosis without steroid therapy because their clinical symptoms were minimal. Clinical symptoms, leukocytosis, and increased KL-6 values improved in 6 months after quitting smoking, and ground-glass opacities on chest HRCT disappeared shortly after quitting smoking in all patients with RB-ILD. No clinical recurrence was observed without steroid therapy at least 2 years after the initial diagnosis. On the other hand, all cases with DIP were required systemic steroid (5–20 mg/d), because clinical symptoms and radiologic findings were not improved 3 months after quitting smoking. Steroid therapy was eventually difficult to withdraw in 2 years and the maintenance therapy (systemic steroid, 5–10 mg/d) was required, because chest radiologic findings were worsened by withdrawal of steroids.
Cigarette smoking is the main causative factor of respiratory bronchiolitis, ie, the inflammation of respiratory and terminal bronchioles. Recently, RB-ILD, an uncommon condition in current or exsmokers, has been reported as an inflammatory lung disorder associated with cigarette smoking. RB-ILD can be distinguished from DIP, which is also recognized to be a smoking-related interstitial lung disease, in terms of its good short-term prognosis.
This study suggests that BALF findings could be one of the useful means to distinguish RB-ILD from other interstitial lung diseases, especially DIP. It has been suggested that the noteworthy feature of BALF findings in RB-ILD is a lack of an increase of lymphocytes, neutrophils, and eosinophils, whereas that in DIP is the increase in eosinophils.13 In the present study, eosinophils were not observed and the proportion of neutrophils was significantly lower in BALF in patients with RB-ILD. The number of neutrophils in BALF of RB-ILD seems to be less than reported in idiopathic pulmonary fibrosis, in which 10% or greater neutrophils was observed.14 The proportion of neutrophils was 0.1% in cell differentiation of BALF in this study. In RB-ILD, only a small number of leukocytes was observed, as it is the case in smokers with simple RB. However, only a part of smokers develop RB-ILD, whereas simple RB is observed in many smokers. In our opinion, RB-ILD and DIP are separate clinicopathologic entities in regard to BALF cell differentiation, although the pathogenesis of interstitial pneumonitis in RB-ILD has not been determined. In addition, we cannot rule out that RB-ILD is a subset of DIP that is mild and relatively nonprogressive.7
In this study, the clinical severity in RB-ILD was mild in regard to subjective symptoms, although productive cough was found in 80% of the patients. As for laboratory findings, leukocytosis without an increase in C-reactive protein values was found in patients with RB-ILD. The mechanism of leukocytosis has not been well-defined, but white blood cell count could increase in proportion to the number of cigarette smoking. Serum KL-6 increased and increased KL-6 values improved 6 months after quitting smoking in most patients with RB-ILD, suggesting that KL-6 values appear to be one of the parameters of disease activity in RB-ILD. It is noteworthy that serum ANA was detected both in RB-ILD and DIP, although none of the patients fulfilled the criteria of connective tissue disorders. Both diseases could occur as a result of any unknown immunologic reaction in addition to smoking, although the levels of ANA in RB-ILD were milder than those in DIP. For example, 1 patient with DIP who had not smoked showed a remarkably high level of ANA.
Regarding chest HRCT findings, ground-glass opacities were remarkable in patients with RB-ILD compared with normal smokers and patients with simple RB, in whom reticulonodular shadows are observed. Patients with RB-ILD could be overlooked because their clinical symptoms are mild, and abnormal findings are sometimes so subtle that are usually missed in routine chest roentgenogram examination. In addition, chest roentgenogram findings of RB-ILD appear to be quite similar to those of DIP. However, HRCT findings showed unique features with diffuse bronchocentric patchy ground-glass opacities in RB-ILD. This patchiness and centrilobular distribution could be useful to distinguish RB-ILD from DIP, whose distribution is more extensive and diffuse.12
Smoking cessation has been reported to be enough for remission of the disease in most patients with RB-ILD.1,7 In the present study, clinical symptoms as well as abnormal radiologic and laboratory findings improved within 6 months after quitting smoking in all patients, especially ground-glass opacities on the chest roentgenogram disappeared early after they quit smoking. On the other hand, all patients with DIP required systemic steroid therapy, because quitting smoking was not sufficient to improve clinical symptoms as well as abnormal radiologic and laboratory findings, and steroid therapy was difficult to withdraw 2 years after the initial diagnosis. Both RB-ILD and DIP are smoking-related interstitial lung diseases and are assumed to occur through different immunologic reactions. However, the prognosis of RB-ILD seems to be good, and many patients might go on with their lives without stopping smoking. It may be important to recognize this disease and to distinguish it clinically and pathologically from other types of interstitial pneumonitis, because its prognosis is favorable. Lung biopsy is required to confirm the diagnosis and to distinguish RB-ILD from more serious causes of diffuse parenchymal lung disease, especially idiopathic pulmonary fibrosis. However, it could be suitable to treat patients with RB-ILD just by advising them to stop smoking when they are suspected to have RB-ILD from clinical and radiologic features. If quitting smoking is not enough to obtain remission, surgical biopsy should be performed.
1. Myers JL, Veal CF Jr, Shin MS, et al. Respiratory bronchiolitis
causing interstitial lung disease: A clinicopathologic study of six cases. Am Rev Respir Dis
2. Weiss W. Cigarette smoking and diffuse pulmonary fibrosis. Am Rev Respir Dis
3. Weiss W. Smoking and pulmonary fibrosis. J Occup Med
4. Weiss W. Cigarette smoking and diffuse pulmonary fibrosis: a preliminary report. Arch Environ Health
5. Niewoehner D, Kleinerman J, Rice D. Pathologic changes in the peripheral airways of young cigarette smokers. N Engl J Med
6. Guerry-Force ML, Mueller NL, Wright JL, et al. A comparison of bronchiolitis obliterans with organizing pneumonia, usual interstitial pneumonia, and small airways disease. Am Rev Respir Dis.
7. Yousem SA, Colby TV, Gaensler EA. Respiratory bronchiolitis
-associated interstitial lung disease and its relationship to desquamative interstitial pneumonia. Mayo Clin Proc.
8. Moon J, du Bois RM, Colby TV. Clinical significance of respiratory bronchiolitis
on open lung biopsy and its relationship to smoking related interstitial lung disease. Thorax
9. Nagai S, Hoshino Y, Hayashi M, et al. Smoking-related interstitial lung diseases. Curr Opin Pulm Med
10. Aubry MC, Wright JL, Myers JL. The pathology of smoking-related lung diseases. Clin Chest Med
11. Myers J. Respiratory bronchiolitis
with interstitial lung disease. Semin Respir Med
12. Holt RM, Schmidt RA, Godwin D, et al. High resolution CT in respiratory bronchiolitis
-associated interstitial lung disease. J Comput Assist Tomogr
13. Kubo M, Koshino T, Shimizu H, et al. A case of desquamative interstitial pneumonia with increased numbers of eosinophils in bronchoalveolar lavage fluid [English abstract]. J Jap Resp Society
14. Daniele R, Elias J, Epstein P, et al. Bronchoalveolar lavage: role in the pathogenesis, diagnosis and management of interstitial lung disease. Ann Intern Med