Quantifying the extent of emphysema: factors associated with radiologists' estimations and quantitative indices of emphysema severity using the ECLIPSE cohort.

TitleQuantifying the extent of emphysema: factors associated with radiologists' estimations and quantitative indices of emphysema severity using the ECLIPSE cohort.
Publication TypeJournal Article
Year of Publication2011
AuthorsGietema, HA, Müller, NL, Fauerbach, PVNasute, Sharma, S, Edwards, LD, Camp, PG, Coxson, HO
Corporate AuthorsEvaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) investigators
JournalAcad Radiol
Volume18
Issue6
Pagination661-71
Date Published2011 Jun
ISSN1878-4046
KeywordsAdult, Aged, Algorithms, Analysis of Variance, Biological Markers, Case-Control Studies, Cluster Analysis, Disease Progression, Female, Forced Expiratory Volume, Humans, Longitudinal Studies, Male, Middle Aged, Pulmonary Emphysema, Radiographic Image Interpretation, Computer-Assisted, Regression Analysis, Severity of Illness Index, Spirometry, Tomography, X-Ray Computed
Abstract

RATIONALE AND OBJECTIVES: This study investigated what factors radiologists take into account when estimating emphysema severity and assessed quantitative computed tomography (CT) measurements of low attenuation areas.MATERIALS AND METHODS: CT scans and spirometry were obtained on 1519 chronic obstructive pulmonary disease (COPD) subjects, 269 smoker controls, and 184 nonsmoker controls from the Evaluation of COPD Longitudinally to Indentify Surrogate Endpoints (ECLIPSE) study. CT scans were analyzed using the threshold technique (%<-950HU) and a low attenuation cluster analysis. Two radiologists scored emphysema severity (0 to 5 scale), described the predominant type and distribution of emphysema, and the presence of suspected small airways disease.RESULTS: The percent low attenuation area (%LAA) and visual scores of emphysema severity correlated well (r = 0.77, P < .001). %LAA, low attenuation cluster analysis, and absence of radiologist described gas trapping, distribution, and predominant type of emphysema were predictors of visual scores of emphysema severity (all P < .001). CT scans scored as showing regions of gas trapping had smaller lesions for a similar %LAA than those without (P < .001).CONCLUSIONS: Visual estimates of emphysema are not only determined by the extent of LAA, but also by lesion size, predominant type, and distribution of emphysema and presence/absence of areas of small airways disease. A computer analysis of low attenuation cluster size helps quantitative algorithms discriminate low attenuation areas from gas trapping, image noise, and emphysema.

DOI10.1016/j.acra.2011.01.011
Alternate JournalAcad Radiol
PubMed ID21393027
Grant List1P50 HL084948 / HL / NHLBI NIH HHS / United States
R01 HL085096 / HL / NHLBI NIH HHS / United States