Reliability of histopathologic diagnosis of fibrotic interstitial lung disease: an international collaborative standardization project

Table 1 Questions used for criteria sets and whole slide images

Fibrosis criteria set	Is the fibrosis severe (yes, no, uncertain)? Is the pattern of fibrosis patchy, diffuse or honeycombing only? If the fibrosis is patchy, is the distribution subpleural/paraseptal, airway centered or uncertain?
Inflammation criteria set	Is there dense inflammation away from scar (yes, no, uncertain)?
Fibroblast foci criteria set	Are fibroblast foci adjacent to scar at the boundary of normal and fibrotic lung (readily identified, rare, none, cannot determine)?
Granuloma criteria set	Does inflammation include granulomas (well formed, poorly formed, scattered giant cells only, none of these, or uncertain)?
Whole slide images, first round only	Is the fibrosis severe (yes, no)? Is the pattern patchy, diffuse or honeycombing only? Is the distribution of fibrosis subpleural/paraseptal, airway centered or uncertain/mixed?
Whole slide images, final round only	Is the fibrosis severe (yes, no, honeycombing only) Is the distribution of fibrosis subpleural/ paraseptal, irregular, airway centered, diffuse or uncertain?
Whole slide images, both rounds	Are fibroblast foci readily identified, rare, none or cannot be determined? Are there non-UIP features present (dense inflammation away from scar, granuloma, organizing pneumonia, smoking related interstitial fibrosis or other)? Is this definite, probable, possible^a or not UIP/IPF?

^aSeveral of the terms we used are not identical to that used in the ATS guidelines. For extent of fibrosis, instead of “is the fibrosis severe”, the 2011 ATS requires “Evidence of marked fibrosis/architectural distortion, +/− honeycombing” while the 2018 ATS requires “Dense fibrosis with architectural distortion (i.e., destructive scarring and/or honeycombing)”. For extent of inflammation, instead of “Is there dense inflammation away from scar” the 2011 ATS requires “marked interstitial inflammatory cell infiltrate away from honeycombing” while the 2018 ATS requires “areas of interstitial inflammation lacking associated fibrosis” (2018). We used the 2011 nomenclature of “possible” UIP instead of the 2018 ATS and Fleishner nomenclature of “indeterminate” for UIP. We do not believe that the meaning of any the phrases we used are substantially different from those of the various guidelines and were understood as such by the participants

ISSN: 1471-2466