Pulmonary disease assessment in cystic fibrosis: comparison of CT scoring systems and value of bronchial and arterial dimension measurements.

Abstract

To retrospectively compare thin-section computed tomographic (CT) scores obtained with five scoring systems for assessment of pulmonary disease in children with cystic fibrosis and to determine additional value of bronchial and arterial dimension measurements. Scores obtained with five thin-section CT scoring systems were compared. A score of 0 indicated the absence of abnormalities; a higher score meant that more structural abnormalities were seen. Three observers assigned scores and then reassigned scores after intervals varying from 1-2 weeks to 1-2 months at review of thin-section CT scans obtained in 25 children with cystic fibrosis. Interobserver and intraobserver reliability was calculated with intraclass correlation coefficients. Quantitative measurements of bronchial and arterial dimensions were obtained. Thin-section CT scores were correlated (Spearman correlation) with bronchial and arterial dimensions and with results of pulmonary function tests (PFTs), such as forced expiratory volume in 1 second (FEV(1)). Scores with all five scoring systems were reproducible, with intraclass correlation coefficients of 0.74 and higher (P <.05), and showed significant correlations with FEV(1) (R = -0.73 to -0.69, P <.01). Ratio of bronchial diameter to accompanying pulmonary arterial diameter was correlated with thin-section CT scores but not with FEV(1). Ratio of bronchial wall thickness to accompanying pulmonary arterial diameter was not correlated with thin-section CT scores or PFT results. Thin-section CT scores were reproducible and were correlated with PFT results. Measurements of bronchial dimensions were not significantly related to scores or PFT results.