HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 2463062200v1 246/3/935    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Best, A. C. Articles by Lynch, D. A. Search for Related Content PubMed PubMed Citation Articles by Best, A. C. Articles by Lynch, D. A.

Idiopathic Pulmonary Fibrosis: Physiologic Tests, Quantitative CT Indexes, and CT Visual Scores as Predictors of Mortality¹

¹ From the Departments of Radiology (A.C.B., D.M.), Obstetrics and Gynecology (A.M.L.), and Preventive Medicine and Biometrics (G.K.G.), University of Colorado Denver, 4200 E 9th Ave, Denver, CO 80262; Pediatric Immunology Research Department, Children's Mercy Hospital, Kansas City, Mo (J.M.); Biogen, Cambridge, Mass (C.M.B.); and Division of Radiology, National Jewish Medical and Research Center, Denver, Colo (D.A.L.). Received December 29, 2006; revision requested March 1, 2007; revision received May 14; final version accepted August 13. Supported by Biogen. Address correspondence to A.C.B. (e-mail: alan.best{at}uchsc.edu).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
Purpose: To retrospectively evaluate quantitative computed tomographic (CT) indexes, pulmonary function test results, and visual CT scoring as predictors of mortality and to describe serial changes in quantitative CT indexes over 12 months in patients with idiopathic pulmonary fibrosis (IPF).

Materials and Methods: Institutional review board approval and informed consent were obtained at all participating institutions. One hundred sixty-seven patients (110 men, 57 women; mean age, 63 years ± 9 [standard deviation]) with IPF were enrolled in a clinical trial. Patients underwent thin-section CT in the supine position at full inspiration at enrollment (baseline) and at 12-month follow-up. After segmentation of the lungs, mean lung attenuation (MLA), skewness, and kurtosis were measured. Extent of ground glass opacity and lung fibrosis were assessed visually. Forced vital capacity (FVC) and total lung capacity (TLC) were measured. Median duration of follow-up for mortality was 1.5 years. Univariate and multivariate survival analyses were used to determine the predictive value of baseline variables for survival.

Results: At univariate analysis, baseline variables predictive of death included TLC, fibrosis, skewness, and kurtosis. At multivariate analysis, FVC (P = .006) and fibrosis (P = .002) were predictors of short-term mortality. In 95 patients who had both baseline and follow-up CT scans, fibrosis (P = .030), MLA (P = .003), skewness (P < .001), and kurtosis (P < .001) all showed change indicating disease progression.

Conclusion: Visually determined disease extent on CT images is a strong independent predictor of mortality in IPF. Serial evaluation of quantitative CT measures can show disease progression in these patients.

© RSNA, 2008

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
Physiologic measurements, particularly forced vital capacity (FVC), have been identified as predictors of mortality in patients with idiopathic pulmonary fibrosis (IPF). More recently, it has been shown that semiquantitative scores of the visual extent of lung fibrosis on computed tomographic (CT) images are a strong independent predictor of mortality (1). While quantitative CT indexes have not been examined as predictors of mortality, predictive information might be helpful to patients and clinicians for determining treatment options and goals and for planning admission to clinical trials.

Quantitative CT indexes have been proposed as a standardized method for scoring the extent of interstitial lung disease. The validity of such indexes was previously examined in a population of patients with IPF by a multi-institutional study (2). That study showed that measurements of skewness, kurtosis, and mean lung attenuation (MLA) (computer-derived values that describe the shape of thin-section CT frequency histograms) correlate with the degree of pulmonary function abnormality. Quantitative CT measurements are attractive for use as indexes of IPF because of their noninvasive and quantitative nature and minimal requirements for user intervention. However, the pattern of change in quantitative CT measurements over time in patients with IPF has not been evaluated.

The two purposes of our study were (a) to determine which baseline variables (CT visual scores, quantitative CT indexes, or pulmonary function test results) are the best predictors of mortality during follow-up and (b) to describe serial changes in quantitative CT indexes over a 12-month period in a population with IPF. We hypothesized that CT visual scores, quantitative CT indexes, and pulmonary function test results could be directly compared for their prognostic value in a population with IPF and that serial quantitative CT measurements could be used to document progression of the disease over time.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
The original Health Insurance Portability and Accountability Act–compliant (for U.S. centers) prospective study received institutional review board approval at all participating institutions, and informed consent was obtained. This retrospective study received exempt status from the institutional review board of the University of Colorado Denver because it consisted of an analysis of preexisting data, and all direct and indirect means of patient identification had been removed from the data. One of the authors (C.M.B.) is employed by Biogen (Cambridge, Mass), and Biogen funded the original clinical study; the authors not employed by Biogen had control of the data and the information submitted for publication.

Patients
Our study included 167 patients (110 men, 57 women; mean age, 63 years ± 9 [standard deviation]) enrolled in a double-blind placebo-controlled clinical trial of interferon β-1a (Avonex; Biogen) for treatment of IPF. Patient enrollment in the original study began in March 1998 and was completed in March 2001. In the study IPF was diagnosed by using the criteria developed by the American Thoracic Society (3). Diagnosis of IPF was confirmed in each patient by a panel of clinicians and was based on a review of clinical history, occupational and environmental exposure, pulmonary function test results, thin-section CT images of the lungs, and, if available, transbronchial or surgical lung biopsy slides. All patients were required to have had progression of IPF, defined as meeting at least one of the following criteria: (a) a greater than 10% relative decrease in total lung capacity (TLC) or FVC or a greater than 15% relative decrease in single-breath diffusing capacity of the lung for carbon monoxide; (b) a greater than 3% decrease in the resting oxygen saturation level, a 3 mm Hg increase in the resting gradient between the partial pressure of oxygen in the artery and that in the alveoli, or a 5% decrease in oxygen saturation with exercise; or (c) radiologic progression of disease (increase in reticular and/or ground glass opacity [GGO]) as assessed on chest radiographs or thin-section CT images.

Patients were excluded if they had end-stage IPF, which was defined as the presence of at least two of the following: (a) TLC less than 45% of the predicted volume, (b) hemoglobin-corrected single-breath diffusing capacity of the lung for carbon monoxide less than 25% of the predicted capacity, (c) resting gradient between the partial pressure of oxygen in the artery and that in the alveoli of more than 40 mm Hg, (d) oxygen saturation of less than 80% with exercise (walking on level ground on room air at own pace for 6 minutes), or (e) New York Heart Association class III or IV heart disease. Other exclusion criteria included (a) environmental or drug exposures likely to cause interstitial lung disease, (b) connective tissue disease, and (c) emphysema occupying more than 50% of the lung.

After the conclusion of this double-blind study, treatment assignment data were revealed. Twenty-nine patients (17.4%) had received a daily placebo, 54 patients (32.3%) had received 15 mg of interferon β-1a daily, 41 patients (24.6%) had received 30 mg of the drug daily, and 38 patients (22.8%) had received 60 mg of the drug daily. Treatment data for five patients (3.0%) were not available for analysis.

Our retrospective study used preexisting data from the prospective clinical trial, including image analyses. Images were not reinterpreted for our study. Pulmonary function test and thin-section CT histogram data were obtained at 30 participating hospitals in the United States and Canada at baseline and at 12-month follow-up. FVC and TLC were measured by using commonly accepted techniques and were expressed as a percentage of predicted performance for each patient, by using the formula developed by Crapo et al (4). Of the 167 enrolled patients, 95 had complete data sets for both baseline and follow-up. Data from these 95 patients were analyzed for assessment of serial change in visual and digital CT variables.

Patients were followed up from the time of enrollment until the end of the entire study to record mortality. Therefore, depending on the date of enrollment for each individual, the duration of follow-up ranged from 0.9 to 2.7 years (median follow-up, 1.5 years).

CT Image Analysis
Inspiratory thin-section (1-mm-thick) CT images were obtained at 2-cm intervals with the patient in the supine position. Images in Digital Imaging and Communications in Medicine, or DICOM, format were sent to a central imaging facility. Visual analysis was performed by two of three thoracic radiologists (including D.A.L., 16 years thoracic CT experience) who independently recorded the extent of pulmonary abnormality. The mean extent of GGO, reticular abnormality, honeycombing, and emphysema was scored to the nearest 5% in three zones in each lung. The upper zone was defined as at or above the aortic arch, the middle zone was defined as between the aortic arch and pulmonary veins, and the lower zone was defined as at or below the pulmonary veins. For the purposes of this study, the mean extent of lung fibrosis was calculated as the mean of the mean extent of reticular abnormality and honeycombing as scored by two observers in each of the six zones. Observers received training sessions prior to the study, and interobserver variability was evaluated and determined to be satisfactory for consistency in visual scoring. Baseline and follow-up images were scored during the trial, and observers were blinded to the temporal sequence of the examinations. Image analysis to determine skewness, kurtosis, and MLA was performed by using semiautomated image segmentation and standard image-analysis techniques (2).

Statistical Analysis
Statistical analyses were carried out by using statistical software (SAS, version 9.1.3; SAS Institute, Cary, NC). Means and standard deviations for measures of pulmonary function (FVC and TLC), quantitative CT indexes (MLA, skewness, and kurtosis), and CT visual scores (fibrosis, GGO, and emphysema) were determined at baseline and at 12 months. The magnitude of change in each of these variables was calculated, as were the number and percentage of patients with increasing or decreasing values. A paired t test was performed to evaluate the change in variables from baseline to 12 months. Mean and standard deviation values for each variable at baseline were also calculated separately for the patients who were alive and for those who were dead at the end of follow-up. The difference between the two groups was assessed with a t test after equal variance was established with the F test.

Univariate and multivariate logistic regression analyses were used to determine the ability of each variable to predict mortality. In the univariate logistic analysis, each variable was examined separately for its association with death. The odds ratio was defined as a ratio of odds in the death group divided by odds in the control group, and 95% confidence intervals, the Wald ² statistic value, and P values were calculated. A P value of less than .05 was considered to indicate a significant difference. The odds ratio reflects an increase (>1) or decrease (<1) in the odds of death within 12 months for each unit of increase in the explanatory variable. Multivariate logistic regression analysis was applied to estimate the odds ratio of each potential predictor of mortality while other possible predictors were taken into account. Time-varying covariates were not used in the multivariate model because we wished to examine the predictive value of the baseline measurements, not of the follow-up measurements.

Survival analysis was performed to assess the value of each variable in predicting mortality while accounting for differences in duration of follow-up. To find the potential predictors we used the Cox proportional hazards model for univariate analysis. Proportional hazard assumptions were tested by using supremum tests and the exact method was used for ties.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
Overall, 35 (21.0%) of the 167 original patients died during the follow-up period. Baseline differences between surviving patients and deceased patients were significant for skewness (P = .036), kurtosis (P = .002), and visual extent of lung fibrosis (P = .002) (Table 1). Other baseline variables were not significantly different between the two groups.

A greater extent of fibrosis at baseline (Table 2) was the strongest independent predictor of death (P = .017). Kurtosis was the second-best predictor of death during follow-up (P = .072). As Table 2 included all available data points in the analysis, the logistic regression analysis was repeated by using only patients (n = 145) who had data for all eight baselines variables, and results were essentially unchanged; visual extent of lung fibrosis was again the most significant predictor (P = .003), followed by kurtosis.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

Because of the documented accuracy of using CT images to establish a confident diagnosis of IPF (12), CT is often used in place of biopsy as an entry criterion in clinical trials (11). There is increasing evidence that the extent of lung disease on CT images is an important determinant of prognosis in patients with IPF (1).

While the role of quantitative CT indexes appears to be quite well established in patients with obstructive lung diseases, particularly emphysema (13–16), quantitative analysis of infiltrative lung diseases is more challenging, largely because it requires methods to identify and quantify the increase in lung attenuation that occurs in these conditions. Although automated analysis of multiple features (17) and fractal analysis (18) have been used, histogram-based methods have been used more extensively (19) and have shown reasonable correlation with pulmonary function test results in patients with asbestosis (20) and IPF (2,20). While the serial change in quantitative CT indexes observed in our study does not by itself confer clinical relevance, this change over 12 months suggests that these techniques may be used to identify progressive fibrosis. The greatest degree of change was found in skewness and kurtosis (P < .001 for each).

The 21.0% mortality rate in our study (median follow-up, 1.5 years) is comparable to published survival figures in patients with IPF (6,7). Our results from univariate logistic regression analysis show that baseline lung fibrosis, kurtosis, skewness, and TLC were statistically significant predictors of short-term mortality. The directionality of the values was as expected: Greater extent of fibrosis and lower TLC, skewness, and kurtosis values predicted mortality during follow-up. Survival analysis demonstrated that these trends did not change when the duration of follow-up was taken into account. These findings support the hypothesis that quantitative CT indexes, specifically kurtosis and skewness, are predictive of short-term mortality. However, multivariate logistic regression analysis showed that the single best predictor of mortality during follow-up was the visual extent of fibrosis on CT images, and there was little additional predictive power to be gained from the more quantitative measures. These results are very similar to those of a recent study by Lynch et al (1) that also found the visual extent of lung fibrosis to be the strongest independent predictor of mortality. That study evaluated physiologic features but did not include quantitative CT measurements. Given the subjectivity of visual assessment of disease extent, it is a little surprising that quantitative CT indexes, which correlate with extent of disease, were not stronger predictors of mortality. Our findings suggest that texture-sensitive image-analysis algorithms with the ability to segment fibrotic lung might be more useful than the relatively crude global quantitative CT indexes used in our study.

Early CT studies of patients defined as having IPF or fibrosing alveolitis suggested that the extent of GGO was a predictor of treatment responsiveness and suggested a better prognosis. However, these studies were published before the wide recognition of the entity of nonspecific interstitial pneumonia, which has a better prognosis than usual interstitial pneumonia and tends to be associated with GGO on CT images. The better prognosis associated with GGO in those studies was likely due to the inclusion of some patients with nonspecific interstitial pneumonia. Our results, like those of Lynch et al (1), indicate that in patients with IPF diagnosed according to current definitions (3), the extent of GGO does not have prognostic value.

Limitations of our study included the lack of spirometric gating in the acquisition of chest CT images. Differing levels of inspiration at CT could lead to substantial differences in quantitative CT indexes, including MLA (21–23). However, spirometric gating or control of CT acquisition remains technically cumbersome and is not widely available on clinical CT scanners. Our findings that quantitative CT can show progression of lung fibrosis even without spirometric control suggest that spirometric standardization may not be necessary for routine use. Another limitation was our specific study group, which included only patients with moderate disease who were eligible for treatment. Further, patient deaths likely had an effect on serial change measurements, as patients who died before follow-up likely had declines in physiologic measurements and progression of CT index values. The relatively short duration of follow-up was a further limitation. The fact that patients were randomized to treatment and placebo groups is a potential concern, but we believe that the results still have implications for an untreated IPF population, since treatment group was not a significant factor in outcome.

We conclude that, in patients with IPF, visual assessment of baseline disease extent on CT images is the strongest available independent predictor of short-term mortality and that serial evaluation of quantitative CT indexes can show disease progression.

	ADVANCES IN KNOWLEDGE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 

• Visual extent of baseline lung fibrosis on CT images is a strong independent predictor of mortality during the observed follow-up in patients with idiopathic pulmonary fibrosis (IPF).
  
• Baseline quantitative CT indexes are also predictive of mortality but do not substantially add to the predictive ability of baseline fibrosis.
  
• Serial measurements of quantitative CT variables demonstrate change, reflecting IPF progression.
  

	IMPLICATIONS FOR PATIENT CARE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 

• Visual extent of baseline fibrosis can be used to predict short-term mortality in patients with IPF.
  
• Serial measurements of quantitative CT indexes can be used to evaluate IPF progression.
  

	FOOTNOTES

 

Abbreviations: FVC = forced vital capacity • GGO = ground glass opacity • IPF = idiopathic pulmonary fibrosis • MLA = mean lung attenuation • TLC = total lung capacity

See Materials and Methods for pertinent disclosures.

Author contributions: Guarantors of integrity of entire study, A.C.B., D.A.L.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, A.C.B., D.M.; clinical studies, A.C.B., A.M.L., C.M.B., D.A.L.; statistical analysis, J.M., G.K.G.; and manuscript editing, A.C.B., J.M., C.M.B., D.M., G.K.G., D.A.L.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 

1. Lynch DA, Godwin JD, Safrin S, et al. High-resolution computed tomography in idiopathic pulmonary fibrosis: diagnosis and prognosis. Am J Respir Crit Care Med 2005;172:488–493.[Abstract/Free Full Text]
2. Best AC, Lynch AM, Bozic CM, Miller D, Grunwald GK, Lynch DA. Quantitative CT indexes in idiopathic pulmonary fibrosis: relationship with physiologic impairment. Radiology 2003;228:407–414.[Abstract/Free Full Text]
3. American Thoracic Society. Idiopathic pulmonary fibrosis: diagnosis and treatment. International consensus statement. American Thoracic Society (ATS), and the European Respiratory Society (ERS). Am J Respir Crit Care Med 2000;161:646–664.[Free Full Text]
4. Crapo RO, Morris AH, Gardner RM. Reference spirometric values using techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 1981;123:659–664.[Medline]
5. American Thoracic Society/European Respiratory Society International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. Am J Respir Crit Care Med 2002;165:277–304.[Free Full Text]
6. Travis WD, Matsui K, Moss J, Ferrans VJ. Idiopathic nonspecific interstitial pneumonia: prognostic significance of cellular and fibrosing patterns—survival comparison with usual interstitial pneumonia and desquamative interstitial pneumonia. Am J Surg Pathol 2000;24:19–33.[CrossRef][Medline]
7. Bjoraker J, Ryu J, Edwin M, et al. Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1998;157:199–203.[Medline]
8. Douglas WW, Ryu JH, Schroeder DR. Idiopathic pulmonary fibrosis: impact of oxygen and colchicine, prednisone, or no therapy on survival. Am J Respir Crit Care Med 2000;161:1172–1178.[Abstract/Free Full Text]
9. Zisman DA, Lynch JP 3rd, Toews GB, Kazerooni EA, Flint A, Martinez FJ. Cyclophosphamide in the treatment of idiopathic pulmonary fibrosis: a prospective study in patients who failed to respond to corticosteroids. Chest 2000;117:1619–1626.[CrossRef][Medline]
10. Crystal RG, Bitterman PB, Mossman B, et al. Future research directions in idiopathic pulmonary fibrosis: summary of a National Heart, Lung, and Blood Institute working group. Am J Respir Crit Care Med 2002;166:236–246.[Abstract/Free Full Text]
11. Raghu G, Brown KK, Bradford WZ, et al. A placebo-controlled trial of interferon gamma-1b in patients with idiopathic pulmonary fibrosis. N Engl J Med 2004;350:125–133.[Abstract/Free Full Text]
12. Hunninghake GW, Lynch DA, Galvin JR, et al. Radiologic findings are strongly associated with a pathologic diagnosis of usual interstitial pneumonia. Chest 2003;124:1215–1223.[CrossRef][Medline]
13. Muller N, Staples C, Miller R, Abboud R. An objective method to quantitate emphysema using computed tomography. Chest 1988;94:782–787.[CrossRef][Medline]
14. Kinsella M, Muller NL, Abboud RT, Morrison NJ, DyBuncio A. Quantitation of emphysema by computed tomography using a "density mask" program and correlation with pulmonary function tests. Chest 1990;97:315–321.[CrossRef][Medline]
15. Uppaluri R, Mitsa T, Sonka M, Hoffman EA, McLennan G. Quantification of pulmonary emphysema from lung computed tomography images. Am J Respir Crit Care Med 1997;156:248–254.[Abstract/Free Full Text]
16. Chabat F, Yang GZ, Hansell DM. Obstructive lung diseases: texture classification for differentiation at CT. Radiology 2003;228:871–877.[Abstract/Free Full Text]
17. Uppaluri R, Hoffman EA, Sonka M, Hunninghake GW, McLennan G. Interstitial lung disease: a quantitative study using the adaptive multiple feature method. Am J Respir Crit Care Med 1999;159:519–525.[Abstract/Free Full Text]
18. Rodriguez LH, Vargas PF, Raff U, et al. Automated discrimination and quantification of idiopathic pulmonary fibrosis from normal lung parenchyma using generalized fractal dimensions in high-resolution computed tomography images. Acad Radiol 1995;2:10–18.[CrossRef][Medline]
19. Beinert T, Kohz P, Seemann M, Egge T, Reiser M, Behr J. Spirometrically controlled high resolution computed tomography: quantitative assessment of density distribution in patients with diffuse fibrosing alveolitis. Eur J Med Res 1996;1:269–272.[Medline]
20. Hartley PG, Galvin JR, Hunninghake GW, et al. High-resolution CT-derived measures of lung density are valid indexes of interstitial lung disease. J Appl Physiol 1994;76:271–277.[Abstract/Free Full Text]
21. Stoel BC, Vrooman HA, Stolk J, Reiber JH. Sources of error in lung densitometry with CT. Invest Radiol 1999;34:303–309.[CrossRef][Medline]
22. Shaker SB, Dirksen A, Laursen LC, Skovgaard LT, Holstein-Rathlou NH. Volume adjustment of lung density by computed tomography scans in patients with emphysema. Acta Radiol 2004;45:417–423.[CrossRef][Medline]
23. Lamers RJ, Kemerink GJ, Drent M, van Engelshoven JM. Reproducibility of spirometrically controlled CT lung densitometry in a clinical setting. Eur Respir J 1998;11:942–945.[Abstract]

This article has been cited by other articles:

				A. M. Groves, T. Win, N. J. Screaton, M. Berovic, R. Endozo, H. Booth, I. Kayani, L. J. Menezes, J. C. Dickson, and P. J. Ell Idiopathic Pulmonary Fibrosis and Diffuse Parenchymal Lung Disease: Implications from Initial Experience with 18F-FDG PET/CT J. Nucl. Med., April 1, 2009; 50(4): 538 - 545. [Abstract] [Full Text] [PDF]

				J. Behr and V. J. Thannickal Update in Diffuse Parenchymal Lung Disease 2008 Am. J. Respir. Crit. Care Med., March 15, 2009; 179(6): 439 - 444. [Full Text] [PDF]

				R. Kim and K. C. Meyer Review: Therapies for interstitial lung disease: past, present and future Therapeutic Advances in Respiratory Disease, October 1, 2008; 2(5): 319 - 338. [Abstract] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 2463062200v1 246/3/935    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Best, A. C. Articles by Lynch, D. A. Search for Related Content PubMed PubMed Citation Articles by Best, A. C. Articles by Lynch, D. A.