David Augustine Lynch MB, Presenter: Consultant, Actelion Ltd Research support, Siemens AG Consultant, Gilead Sciences, Inc Consultant, Novartis AG Scientific Advisor, Perceptive Informatics, Inc

Francine L. Jacobson MD, MPH, Abstract Co-Author: Research support, Toshiba Corporation

James R. Murphy, Abstract Co-Author: Nothing to Disclose

Carla G. Wilson, Abstract Co-Author: Nothing to Disclose

John D. Newell MD, Abstract Co-Author: Research Consultant, Siemens AG Research grant, Siemens AG Consultant, WebMD Health Corp (WebMD, Inc) Author, The Humana Press

Philippe A. Grenier MD, Abstract Co-Author: Nothing to Disclose

Hans-Ulrich Kauczor MD, Abstract Co-Author: Research grant, Siemens AG Research grant, Boehringer Ingelheim GmbH

James D. Crapo MD, Abstract Co-Author: Nothing to Disclose

The COPDGene™ research program convened a qualitative scoring workshop with the goal of defining key visual CT characteristics that may define specific COPD subtypes.

395 CT scans were reviewed by 58 radiologists and pulmonologists, to identify qualitative features that can be used to define CT phenosubtypes of COPD. Consensus standards to be used for evaluation were developed using expert presentations and group discussion. Each observer was asked to score 80 cases, resulting in 9-11 readings for each case. Data recorded included emphysema presence, type (paraseptal, bulla, centrilobular, panlobular), zonal distribution and severity, bronchial wall thickening, cylindrical bronchial dilatation, bronchiectasis, centrilobular nodules, mosaic attenuation, gas trapping, tracheobronchial disease, ground-glass, honeycombing and dominant CT phenotype (normal, large airway, small airway, emphysema, and mixed).

No significant differences in agreement scores were found between radiologists and pulmonologists. Emphysema globally and by specific type had the highest inter-observer agreement (Kappa .43-.67). A 2-way, unsupervised cluster analysis was run which identified 4 disease clusters. The clusters and the variables associated with them were (1) tracheal disease with saber sheath, tracheobronchomalacia, airway outpouching and presence of mucoid material in central airways; (2) small airway disease with mosaic attenuation, ground-glass, honeycombing and centrilobular opacities (3) centrilobular and panlobular emphysema with gas trapping, and airway wall thickening and (4) bronchial disease with cylindrical bronchial dilatation and bronchiectasis. Visual assessment of airway wall thickening appeared to be a better logistic discriminator among disease groups categorized by GOLD criteria than similar data obtained from automated airway analysis.

Qualitative evaluation of CT features appears to identify specific COPD subtypes.

Visual subtypes derived from CT images may facilitate genetic analysis and individualized therapy

Lynch, D, Jacobson, F, Murphy, J, Wilson, C, Newell, J, Grenier, P, Kauczor, H, Crapo, J, Visual CT Subtypes of COPD: Preliminary Observations from the COPDGene Trial‚Presented on Behalf of the COPDGene Qualitative CT Workshop Participants.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9013165.html