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

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted 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 Google Scholar Google Scholar Articles by Nieber, M. Articles by Mascalchi, M. Search for Related Content PubMed PubMed Citation Articles by Nieber, M. Articles by Mascalchi, M.

Prediction of Pulmonary Function in COPD on the Basis of CT Measurements of Bronchial Wall Thickness

Department of Radiology, Image Processing, Leiden University Medical Center Albinusdreef 2,
2333 ZA Leiden,
the Netherlands e-mail: m.nieber{at}lumc.nl

Editor:

With the latest technical advances in computed tomography (CT), in vivo measurements of human airway dimensions, such as lumen area and wall thickness, have now become feasible even for small bronchi. Authors of recent studies have explored the correlations between these bronchial dimensions and pulmonary function. In this respect, we read with interest the article by Dr Orlandi and colleagues (1), in the February 2005 issue of Radiology, on the associations between wall thickening and lung function in chronic obstructive pulmonary disease (COPD). They assessed wall thickening by measuring percentage wall area (PWA) and wall thickness-to-diameter ratio (TDR) and subsequently normalized PWA for body weight. In a multiple regression analysis they found that a combination of PWA, TDR, and PWA per kilogram of body weight could significantly predict gas exchange (diffusing capacity of the lung for carbon monoxide [DLCO]) and airway obstruction (forced expiratory volume in 1 second–vital capacity ratio [FEV₁/VC]) (both expressed as percentage of predicted value). However, it is unclear why these three parameters were included in the analysis, as they basically measure the same feature (ie, wall thickness relative to lumen size). Moreover, the inclusion of highly correlated predictors such as PWA and TDR is undesirable, since it causes unstable regression coefficient estimates (2). Stable estimates of regression coefficients and their confidence intervals are essential for the interpretation of a model, but these were not provided by the authors.

However, our main concern is that the authors probably have introduced a confounder. The normalization of PWA to body weight suggests that weight was associated with airway dimensions (the authors did not elucidate this matter). By entering both original PWA and weight-normalized PWA as predictor variables, it is likely that a confounding factor has been introduced, as an association between body weight and DLCO (or FEV₁/VC) could explain the association found by means of the multiple regression. A relationship between body weight and DLCO and FEV₁/VC is conceivable, since weight loss is related to the progression of COPD (3). We suggest that the authors test whether the combination of PWA, PWA normalized to body weight, and weight significantly improves the prediction compared with a model that includes only weight.

References

1. Orlandi I, Moroni C, Camiciottoli G, et al. Chronic obstructive pulmonary disease: thin-section CT measurement of airway wall thickness and lung attenuation. Radiology 2005;234(2):604–610.[Abstract/Free Full Text]
2. Harrell FE. Regression modeling strategies. New York, NY: Springer, 2001.
3. Wilson DO, Rogers RM, Wright EC, Anthonisen NR. Body weight in chronic obstructive pulmonary disease. The National Institutes of Health Intermittent Positive-Pressure Breathing Trial. Am Rev Respir Dis 1989;139(6):1435–1438.[Medline]

Dr Camiciottoli and colleagues respond:

Gianna Camiciottoli, MD,*, Ilaria Orlandi, MD,, Stefano Diciotti, PhD, and Mario Mascalchi, MD, PhD

Respiratory Medicine Unit, Department of Critical Care,* Radiodiagnostic Section, Department of Clinical Physiopathology, Department of Electronics and Telecommunications, University of Florence,
Viale Morgagni 85, 50134 Florence, Italy
e-mail: m.mascalchi@dfc.unifi.it

We thank Dr Nieber and colleagues for their interest in the article we recently published in Radiology (1), which demonstrated that use of a combination of PWA, TDR, and PWA normalized for body weight predicted airway obstruction, as measured with FEV₁/VC, and gas exchange impairment, as measured with DLCO, in 42 patients with COPD and that this prediction was stronger in the subgroup of 20 patients with clinical features of chronic bronchitis.

They wondered why we included in the analysis three parameters—namely, PWA, TDR, and PWA per kilogram of body weight—that basically measure the same feature (wall thickness relative to lumen size). Admittedly, despite several precautions in the measurements of the airways on thin-section CT images, including selection of round bronchi with an external diameter of greater than 2 mm and a maximum diameter–to–minimum diameter ratio less than 1.5, neither PWA nor TDR stands alone as a perfect measurement, because bronchial wall thickness is probably overestimated on CT scans owing to inaccuracies in boundary detection and inclusion of adjacent peribronchial interstitium (2). As a matter of fact, like in other studies (2,3), we measured both.

However, the main concern Dr Nieber and colleagues had was the introduction of a confounder, namely body weight, in the analysis we performed. In particular, they speculated that the association between body weight loss and functional parameters as DLCO or FEV₁/VC in patients with COPD could explain the association found at multiple regression analysis. Actually, the correlations between FEV₁/VC and body weight (r = 0.30, P = .051) and between D_LCO and body weight (r = 0.30, P = .049) were weak in the 42 patients with COPD with moderate functional impairment we examined. No correlation was observed in the two subgroups of patients with or those without chronic bronchitis.

At multivariate analysis the combination of PWA, PWA normalized to body weight, and weight did not substantially improve the prediction of airway obstruction and gas exchange in the 42 patients with COPD (R values range, 0.32–0.49) and notably in the subgroup of 20 patients with chronic bronchitis (R values range, 0.30–0.51) compared with that obtained with the model including PWA, TDR, and PWA per kilogram of body weight (1).

References

1. Orlandi I, Moroni C, Camiciottoli G, et al. Chronic obstructive pulmonary disease: thin-section CT measurement of airway wall thickness and lung attenuation. Radiology 2005;234:604–610.[Abstract/Free Full Text]
2. Awadh N, Muller N, Park CS, et al. Airway wall thickness in patients with near fatal asthma and control groups: assessment with high resolution computed tomography scan. Thorax 1998;53:248–253.[Abstract/Free Full Text]
3. Nakano Y, Muro S, Sakai H, et al. Computed tomographic measurements of airway dimensions and emphysema in smokers. Am J Respir Crit Care Med 2000;162:1102–1108.[Abstract/Free Full Text]

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted 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 Google Scholar Google Scholar Articles by Nieber, M. Articles by Mascalchi, M. Search for Related Content PubMed PubMed Citation Articles by Nieber, M. Articles by Mascalchi, M.