Pulmonary Emphysema: Effect of Lung Volume on Objective Quantification at Thin-Section CT

• Published in: Radiology Vol. 257; no. 1; pp. 260 - 268
• Main Authors: MADANI, Afarine, VAN MUYLEM, Alain, ALAIN GEVENOIS, Pierre
• Format: Journal Article
• Language: English
• Published: Oak Brook, IL Radiological Society of North America 01.10.2010
• Subjects: Adult; Biological and medical sciences; Case-Control Studies; Chronic obstructive pulmonary disease, asthma; Diving; Female; Humans; Investigative techniques of respiratory function; Investigative techniques, diagnostic techniques (general aspects); Linear Models; Lung Volume Measurements; Male; Medical sciences; Middle Aged; Pneumology; Prospective Studies; Pulmonary Emphysema - diagnostic imaging; Pulmonary Emphysema - pathology; Respiratory Function Tests; Statistics, Nonparametric; Tomography, X-Ray Computed - methods; Vital Capacity; Lung volume; Lung disease; Radiodiagnosis; Respiratory disease; Lung; Quantization; Medical imagery; Bronchus disease; Computerized axial tomography; Obstructive pulmonary disease; Respiratory system; Emphysema
• ISSN: 0033-8419; 1527-1315
• DOI: 10.1148/radiol.10091446

To prospectively investigate the effect of submaximal inspiration on computed tomographic (CT) indexes used to quantify emphysema and to discriminate between effects of lung tissue loss and increase in total lung capacity (TLC) on these indexes. In this ethical committee-approved study, 20 control subjects and 16 patients with chronic obstructive pulmonary disease (COPD) who provided written informed consent were included. Three 1-mm-thick sections were obtained from each participant at 100%, 90%, 80%, 70%, and 50% of vital capacity (VC). At each volume, eight percentiles of attenuation distribution, as well as relative area (RA) of lung occupied by attenuation coefficients lower than nine thresholds, were calculated. A linear regression line between TLC and each CT index was plotted for control subjects. Mean distance from data points measured in patients with COPD to the normal regression line was used to reflect the effect of lung tissue loss, regardless of TLC. The RA of lung occupied by attenuation coefficients lower than -950 HU (RA(950)) at any percentage of VC lower than 100% decreased significantly from that at 100% VC (P ≤ .002) in control subjects and patients with COPD; however, between 100% VC and 90% VC, the average difference in RA(950) was only 3% and 2% in control subjects and patients with COPD, respectively. The 1st percentile at any percentage of VC lower than 100% increased from that at 100% VC (P < .001) in control subjects. This percentile did not significantly differ from 100% VC at 90% VC or 80% VC (P = .176 and P = 0.077, respectively), but it did significantly differ from 100% VC at 70% VC and 50% VC (P ≤ .002 for both) in patients with COPD. Slope and mean distance were different from zero for all RAs and percentiles except for mean distance for RAs between RA(900) and RA(920). Submaximal inspiration induces underestimation of pulmonary emphysema. Both lung tissue loss and TLC influence CT indexes, suggesting the need to establish normal CT values.