Optimal scoring of serial change on chest radiography in sarcoidosis

Abstract

Background: The optimal means of quantifying change on chest radiography in sarcoidosis is uncertain. In current guidelines, the role of serial measurement of carbon-monoxide diffusing capacity (DLCO) remains undefined and the prevalence of discordance between serial chest radiographic change and pulmonary function tends is unknown. Objective: To identify and explore key uncertainties in the monitoring of sarcoidosis by serial pulmonary function tests and chest radiography. Design: 354 patients with sarcoidosis and concurrent tests (chest radiography and PFTs within three months at baseline, two years and/or four years) were studied. Chest radiographs were assessed by two radiologists for changes in stage and disease extent. Radiographic change and pulmonary function trends were quantified and compared. Results: Change in radiographic extent of lung disease was always more frequent than change in stage (p<0.0001) and there was poor agreement between change in stage and change in radiographic extent (Kw = 0.21 at two years; Kw = 0.23 at four years). Change in disease extent on chest radiography was linked to PFT trends on analysis of variance (p<0.0005 for FEV1, FVC, DLCO), whereas change in radiographic stage was not. Changes in gas transfer were often isolated or discordant with other serial data. Discordance between pulmonary function data and chest radiographic data was observed in 50% of cases. Conclusions: Change in radiographic extent is more applicable to routine monitoring in sarcoidosis than change in radiographic stage. In future guidelines, the role of serial gas transfer estimation and reconciliation of divergent chest radiographic and functional trends might usefully be addressed.