Jonathan G. Goldin, Abstract Co-Author: Nothing to Disclose

David Augustine Lynch MD, Presenter: Nothing to Disclose

Diane C Strollo MD, Abstract Co-Author: Nothing to Disclose

Sarinnapha Vasunilashorn BS, Abstract Co-Author: Nothing to Disclose

Philip J Clements MD, Abstract Co-Author: Nothing to Disclose

Daniel E Furst MD, Abstract Co-Author: Nothing to Disclose

Robert M Elashoff PhD, Abstract Co-Author: Nothing to Disclose

Donald P Tashkin MD, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

Radiographic detection and characterization of Scleroderma (SSc) Lung Disease (SLD) remain poorly defined and challenging. We now report an inter- reader study to detect active alveolitis in SSc patients undergoing screening for participation in a clinical therapeutic study.

As part of the study all subjects with SSc (defined by clinical criteria) with a forced vital capacity (FVC), 45-85% of predicted underwent prone high resolution computed tomography (HRCT) and bronchoalveolar lavage (BAL) to determine presence of alveolitis. This was defined by either any ground glass attenuation on CT (with or without evidence of fibrosis) (AGG) and/or ≥3% polymorphonuclear leukocytes (PMNs) and/or ≥2% eosinophilson BAL. The scans were then independently read by 2 experienced thoracic radiologists, who were blinded to the BAL or initial CT results, for the presence or absence of pure ground glass attenuation (without any CT evidence for fibrosis) (PGG). Consensus was then reached at a joint session with a 3rd tie-break reader. Kappa analysis (K) was used to assess inter-reader variation, the relationship between the 2 definitions of GG, and their relationship with BAL cellularity.

Of 162 CT scans read at randomization 144 (89%) met the AGG definition (CT+). BAL was positive (+) in 99(61%) of which 84(85%) were also AGG CT+ and only 15(15%) were AGG CT- (K 0.2). Of 148 cases read by consensus only 72 were CT+ by both definitions of GG (K 0.14). There was very good inter-reader agreement for PGG between the 2 core readers (K 0.73,p<0.0001). Of 106 cases only 53 (50%) were PGG CT+. BAL was + in 70 of these 106 cases (66%), and of the 70 BAL+ cases, 39 (55%) were also PGG CT+ and 31(45%) were PGG CT-. The correlation between BAL + and PGG+ was again poor (K 0.15).

Ground glass attenuation with or without CT evidence of fibrosis does not correlate well with abnormality on BAL. The CT definition of “active” lung disease in scleroderma remains unclear. Follow-up imaging of the patients in this study should be helpful in determining whether CT features are predictive of treatment responsiveness or progression.Funded by NHLBI grant 5R01-HL-HL60587.

Goldin, J, Lynch, D, Strollo, D, Vasunilashorn, S, Clements, P, Furst, D, Elashoff, R, Tashkin, D, et al, , HRCT in Scleroderma Lung Disease: The Ground Glass Dilemma.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4415134.html