| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the National Jewish Medical and Research Center (Drs. Portnoy, Cool, Curran-Everett, Cherniack, and Brown), Denver, CO; University of Colorado Health Sciences Center (Drs. Schwarz and Veraldi), Denver, CO; and University of California at San Francisco (Dr. King), San Francisco, CA.
These authors contributed equally to this article.
Correspondence to: Kevin K. Brown, MD, FCCP, National Jewish Medical and Research Center, 1400 Jackson St, Denver, CO 80206; e-mail: brownk{at}njc.org
Abstract
Background: The clinical and physiologic features of respiratory bronchiolitis (RB)-interstitial lung disease (ILD) have been previously described; however, the natural history and outcome have not been systematically evaluated. The majority of published reports consider RB-ILD to be a nonprogressive ILD that clinically improves with smoking cessation and antiinflammatory treatment. In this study, we sought to determine the outcome of RB-ILD patients with and without smoking cessation and with and without corticosteroid therapy.
Methods: Thirty-two RB-ILD cases confirmed by surgical lung biopsy were identified from a prospectively enrolled cohort of subjects with ILD. Initial and follow-up data on symptoms, physiology, treatment, and outcome were collected and analyzed.
Results: Kaplan-Meier analysis revealed that at least 75% of RB-ILD patients survived > 7 years after diagnosis. Clinical improvement occurred in only 28% of cases, and physiologic improvement occurred in 10.5% of cases. One patient died of progressive ILD, and two patients died of non-small cell lung cancer. While physiologic improvement was limited to those who had ceased smoking, corticosteroids and/or other immunosuppressive therapy had little effect on symptoms or physiology.
Conclusions: This study shows that prolonged survival is common in RB-ILD. However, symptomatic and physiologic improvement occurs in only a minority of patients, and neither smoking cessation nor immunosuppressive therapy is regularly associated with clinically significant benefit.
Key Words: idiopathic interstitial pneumonia • interstitial lung disease • respiratory bronchiolitis • respiratory bronchiolitis-interstitial lung disease • smoking • survival
Respiratory bronchiolitis (RB) was first described as an incidental histopathologic finding at the time of autopsy in the lungs of young cigarette smokers who died from nonpulmonary causes.1 Collections of pigmented macrophages were identified within the bronchiolar lumens and peribronchiolar alveolar spaces associated with chronic inflammation of the bronchiolar walls and peribronchiolar interstitium. It was suggested that RB represented "smoker’s bronchiolitis" and was responsible for the reduced airflows at low lung volumes found in these individuals.2 Some smokers with clinical, physiologic, and radiographic evidence of an interstitial lung disease (ILD) were also noted to have similar histopathologic features on surgical lung biopsy. This clinicopathologic syndrome was termed respiratory bronchiolitis–interstitial lung disease.3
Whether RB and RB-ILD can be distinguished by histopathologic criteria is controversial. Some authors456 have described more extensive peribronchiolar interstitial fibrosis associated with RB-ILD, while others37 consider the histologic features of RB and RB-ILD indistinguishable and separate the two based on the presence of clinical evidence of interstitial lung disease.
Four case series3568 published in the English-language literature over the past 2 decades reported the clinical features of RB-ILD. These studies3568 suggest that RB-ILD is a smoking-related disease and that progressive disease rarely, if ever, occurs regardless of whether smoking is continued. These conclusions, however, are based on limited and irregularly defined outcome measures, and the number of patients in each series is small. In this study, we review our experiences with the clinical course of RB-ILD and evaluate the effects of smoking cessation and corticosteroid therapy on outcome.
Materials and Methods
Case Definitions
A retrospective review was performed of all patients prospectively enrolled in our institutional review board-approved, National Institutes of Health-supported study of the longitudinal investigation of interstitial lung disease at National Jewish Medical and Research Center. Patients with RB-ILD diagnosed between 1982 and 1999 were eligible for inclusion.
The diagnosis of RB-ILD was made based on the presence of respiratory symptoms consistent with ILD, diffuse lung disease on chest radiograph, and the findings of RB on surgical lung biopsy. All biopsies were reviewed by an expert pulmonary pathologist (C.D.C.). Pathologic criteria for the diagnosis of RB-ILD9 included the accumulation of pigmented macrophages within the lumens of respiratory bronchioles and alveolar ducts, accompanied by chronic inflammation of the respiratory bronchiolar walls and both bronchiolar and peribronchiolar alveolar fibrosis causing architectural distortion. These features, at low magnification, were required to be patchy and confined to the peribronchiolar region.
Cases were excluded if RB-ILD coexisted with any other defined histologic form of diffuse lung disease (eg, usual interstitial pneumonia or pulmonary Langerhans cell granulomatosis). Subjects with monotonous contiguous parenchymal collections of pigmented or nonpigmented intraalveolar macrophages with variable amounts of interstitial collagen and the lack of a bronchiolocentric pattern were considered to have desquamative interstitial pneumonia and were excluded from this study.
Pulmonary Function Testing, Pressure Volume Curve, Exercise Testing, and BAL
Pulmonary function testing was performed as previously described.10 Spirometry was performed according to American Thoracic Society guidelines.1112 Lung volumes were determined in a volume displacement, pressure-compensated body plethysmograph. Reference values were those of Goldman and Becklake13 and Rossiter and Weill.14 The single-breath diffusing capacity of the lung for carbon monoxide (DLCO) was measured by the method of Blakemore et al15 and corrected for hemoglobin.16 Results are expressed as a percentage of predicted except in the case of FEV1/FVC, which is expressed as a ratio. The volume-pressure relationship of the lungs was measured in a body plethysmograph as previously described.17 Cardiopulmonary testing was performed according to a previously reported protocol.17 BAL was performed according to a standardized protocol.18
Follow-up Data
Follow-up data were collected by three methods: (1) patients were seen in follow-up or contacted by telephone and asked to determine subjectively, regarding their lung disease, whether they overall felt better, worse, or were unchanged when compared to their initial evaluation for ILD (patient’s "overall subjective assessment"); (2) follow-up questionnaires, the same as those completed at initial evaluation, addressing specific clinical symptoms were mailed to all patients; questions directly queried the type and amount of exertion required to precipitate dyspnea, which was defined as distressful breathing; points (to a maximum of 20) were assigned according to a previously devised dyspnea score10; and (3) follow-up pulmonary function tests were performed and arterial blood gas levels were obtained and compared to initial data. The initial physiologic pattern of abnormality (obstructive, restrictive, or mixed) was determined by standard criteria.19 Follow-up physiology, arterial blood gases, and DLCO measurements were analyzed and classified as better, worse, or unchanged. Significant changes were prospectively defined as a change of 
10% in FVC or FEV1, 15% change in DLCO, and 4 mm Hg in the alveolar-arterial oxygen pressure gradient (P[A-a]O2). Values falling between these ranges were classified as unchanged. Vital status was ascertained from follow-up with the patient, the patient’s family or physician, or by search of the national death registry.
Statistical Analysis
Data analyses were done using statistical software (SAS Release 8.1; SAS Institute; Cary, NC). We compared observations in the RB-ILD cohort using 
2 or Fisher exact tests (categorical data) and using a two-sample t or Wilcoxon rank-sum tests (continuous data). We controlled for multiple comparisons using the false discovery rate procedure.20 We compared Kaplan-Meier estimates of survival from initial presentation through September 1999 using a likelihood ratio statistic.21 In each analysis, we considered p levels < 0.05 to be statistically significant.
Results
Baseline Demographic and Clinical Features
At the time of diagnosis, all RB-ILD patients were current or former smokers, with a median age of 44 years (Table 1
). All were symptomatic, with the insidious onset of dyspnea the most common complaint and subjective wheezing reported by the majority. While hemoptysis has been previously described,22 no patient in our series reported this symptom. Bibasilar, end-inspiratory crackles were common though not universal, and clubbing was infrequently seen.
|
|
High-Resolution CT
High-resolution CT of the chest was available in 21 RB-ILD patients. These findings have been reported previously23 and include, in order of frequency, central bronchial wall thickening (90%), peripheral bronchial wall thickening (86%), centrilobular nodules (71%), and ground-glass opacity (67%). Septal lines and a reticular pattern were noted in 33% of these cases. Honeycombing was found in one case (5%).
Clinical Follow-up
Of the total of 32 subjects, a follow-up overall subjective assessment was available for 25 subjects with a median follow-up of 7.0 years (range, 2.1 to 14.1 years). Of the seven patients for whom follow-up overall subjective assessments were unavailable, four were unavailable for follow-up, one refused follow-up, and two died of causes unrelated to ILD (non-small cell lung cancer). The single patient who died of progressive lung disease was considered worse for this analysis. Follow-up questionnaires regarding individual symptoms were completed by 24 subjects with a median follow-up of 6.5 years (range, 2.1 to 14.1 years). Follow-up physiologic measures were obtained in 19 subjects with a median follow-up of 6.9 years (range, 3.2 to 14.1 years).
From the patients’ overall subjective assessments of their pulmonary status, when compared to their baseline evaluation, 28% were improved, 28% were unchanged, and 44% were worse (Table 3 ). Individual symptoms worsened in up to one fourth of patients. While the subjects described sense of dyspnea remained the same or improved in > 75%, when dyspnea was quantified by the use of the validated dyspnea score the majority (63%) had clearly worsened. Nearly half of the patients had progressive physiologic deterioration as measured by pulmonary function testing (Table 3). Gas exchange deteriorated in 55% when measured by arterial blood gases and in 25% when measured by DLCO.
|
Follow-up data on smoking cessation were available in 27 subjects. Only 18 subjects (67%) were ultimately able to discontinue smoking (Table 4 ). There was no difference in the follow-up subjective assessment between those who continued to smoke and those who had stopped; 44% of both groups noted worsening. There were no significant differences in the four symptom measures between patients who continued to smoke and those who had stopped. Thirty-three percent of patients who stopped smoking reported worsening dyspnea, and when evaluated by the dyspnea score this number rose to 67%. Pulmonary physiology worsened in 50% of both groups. Improvement in physiology and gas exchange occurred infrequently and was limited to those who had stopped smoking. Age of initiation, total years of use, and cumulative pack-years did not correlate with outcome (subjective assessment, spirometry, or gas exchange) in either patient group (data not shown).
|
|
Survival
Survival analysis was performed using the product-limit (Kaplan-Meier) method (Fig 1
). Vital status was confirmed in 28 subjects; 4 subjects were unavailable for follow-up. Three patients died, two of non-small cell lung cancer and one of progressive ILD. Due to the prolonged survival, median survival could not be calculated. The lower limit of the 25% quartile demonstrated that 75% of RB-ILD patients are expected to survive 7 years.
|
RB, also referred to as smoker’s bronchiolitis, is a common histologic feature found in the lung tissue of cigarette smokers. When identified as the sole histopathologic finding in the clinical setting of ILD, the diagnosis of RB-ILD is made. To date, four prior reports3568 totaling 46 subjects described the clinical-physiologic features and outcomes in patients with biopsy-proven RB-ILD. Excluding the four patients in the series of Moon et al5 and the single patient in the series of Yousem et al6 who either had associated autoimmune disease or no tobacco history, a total experience of 41 patients has been reported. Follow-up data were available on 40 of these 41 patients. The present study of 32 patients represents the single largest experience reported.
As in our series, each of the four previously published studies3568 included patients with clinical diagnoses of ILD and findings of RB on surgical lung biopsy. Table 6 summarizes baseline and follow-up data abstracted from these four studies in comparison with our patient group.
|
All of our patients were symptomatic, most commonly with dyspnea. Symptomatic wheezing and the identification of crackles on physical examination were found more frequently than in prior reports.568 The most common pulmonary function pattern in our series was an obstructive ventilatory defect, while both pure restriction and mixed defects were less common. A restrictive defect was the most common abnormality seen in the series published by Ryu et al8 and Moon et al5 (Table 6). While a reduction in the DLCO was common in previous reports, 36% of our patients had normal initial DLCO values, suggesting that this is not as sensitive a marker of clinically significant disease as previously thought. There were no distinguishing features which could differentiate the BAL of RB-ILD patients from that of otherwise normal healthy smokers.
While data on the clinical, physiologic, and pathologic features of patients with RB-ILD are available, there are little confident data regarding the natural history and outcome of this disease. Abstracting the available follow-up data from the four prior series, 35% (14 of 40 subjects) continued to smoke, and clinical improvement or stability was reported in all but 1 subject. Follow-up data were reported for 26 subjects who discontinued smoking, and 18 subjects (69%) improved while only 1 had clinical deterioration.3568 In one study,6 no patient with RB-ILD had progressive lung disease during a mean follow-up of 66.8 months. Based on these data, it has been concluded that RB-ILD not only portends a favorable prognosis but that it is often a reversible disease with smoking cessation.
In contrast to these previous reports, our data do not support RB-ILD as a benign disease. Clinical worsening was common as measured as a self-reported subjective deterioration in overall status or by specific symptoms. The amount of clinical worsening was amplified when the poor correlation between the subjects’ perceived change in dyspnea and the change quantified by the dyspnea score was taken into account. The dyspnea score worsened in the majority of patients, suggesting that patients underestimate the severity or the impact of their breathlessness as it changes over time. Worsening of spirometry and gas exchange was common regardless of smoking status, corticosteroids or other therapies. The significant decline in pulmonary function in the corticosteroid-treated vs untreated group may, however, merely reflect the initiation of treatment in a more advanced or clinically progressive group. Less than a third of patients who discontinued smoking reported subjective improvement, similar to those who continued to smoke. While uncommon, spirometry and gas exchange occasionally showed objective improvement, and this was limited to those who discontinued smoking. Furthermore, while our data confirm that patients with RB-ILD are generally heavy cigarette smokers, specific features of their smoking history (age of initiation, total years of use, and cumulative pack-years) were not predictive of disease severity or outcome (data not shown).
The current study confirms that prolonged survival can be expected in most patients with RB-ILD and that death secondary to progressive ILD is rare. Median survival time could not be calculated for RB-ILD patients, as only three deaths occurred during the follow-up period, but our data suggest that at least 75% of the patients would be expected to live at least 7 years.
Our study has several limitations. It is possible that referral bias resulted in our patient population having generally more advanced or progressive disease than in prior reports. The retrospective nature of our study limits the strength of the conclusions that can be drawn regarding any outcome other than survival. Subjective, physiologic, and gas exchange data were not simultaneously acquired, limiting the power of any correlations. In the absence of prospectively defined therapy, any assumptions about the usefulness of the therapeutic approaches described should be made with caution.
We have minimized the impact of these limitations on our study in a variety of ways. Using the same standardized symptom questionnaire in follow-up as was used in the initial evaluation, and directly interviewing subjects has allowed us to make useful statements about an individual subject’s overall sense of the activity of his or her disease as well as changes in particular symptoms. Confidence in the changes described in lung function and gas exchange is assisted by the use of strict, prospectively defined criteria for improvement and deterioration. While we cannot comment on the relative benefits of smoking cessation and oral corticosteroid usage for patients with RB-ILD, it is clear that neither of these therapeutic interventions regularly results in symptomatic or physiologic improvement and that deterioration commonly occurs regardless of the intervention used.
In conclusion, a large percentage of patients suffer both clinical and physiologic decline regardless of current therapeutic interventions, although prolonged survival can be expected. In contrast to previous reports, this study suggests that RB-ILD is not a benign disease, symptoms and physiology improve in only a minority of patients, and neither smoking cessation nor oral corticosteroid therapy will regularly provide clinically significant benefit.
Footnotes
Abbreviations: DLCO = diffusing capacity of the lung for carbon monoxide; ILD = interstitial lung disease; P(A-a)O2 = alveolar-arterial oxygen pressure gradient; RB = respiratory bronchiolitis
This work was performed at National Jewish Medical and Research Center, Denver, CO, and was supported by National Institutes of Health grants HL-56556 and HL-67671–02.
The authors have no conflicts of interest to disclose.
Received for publication July 30, 2006. Accepted for publication October 12, 2006.
References
Related Editorial
This article has been cited by other articles:
|
|
 |
|
  H. R. Collard On Beyond Zebra: Understanding Rare Diseases Chest, March 1, 2007; 131(3): 644 - 646. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
