Accuracy of individual variables in the monitoring of long-term change in pulmonary sarcoidosis as judged by serial high-resolution CT scan data.

BACKGROUND: In pulmonary sarcoidosis, the optimal means of quantifying change is uncertain. The comparative usefulness of simple lung function trends and chest radiography remains unclear. We aimed to explore and contrast the disease-monitoring strategies of serial pulmonary function tests (PFTs) and chest radiography compared against morphologic change on high-resolution CT (HRCT) scan.

METHODS: Seventy-three patients with sarcoidosis were identified who had two HRCT scans with concurrent chest radiography and PFTs. Chest radiography and HRCT scans were assessed by two radiologists for change in disease extent. Concordance between the scoring systems, as well as agreement between PFT trends (% change from baseline in FEV, FVC, and diffusing capacity of the lung for carbon monoxide [Dlco]), chest radiography, and chest HRCT scan change, were examined using the weighted κ coefficient of variation (Kw).

RESULTS: There was fair agreement between change in extent of disease on chest radiograph and significant PFT trends (Kw = 0.35, P < .001) and moderate agreement between change in extent of disease on serial HRCT scan and significant PFT trends (Kw = 0.64, P < .0001). The integration of Dlco trends did not improve concordance between change on HRCT scan and PFT change. Change in gas transfer coefficient (ie, Dlco/alveolar volume) displayed no overall linkage with change in disease extent on chest radiograph (Kw = 0.07, P = .27) and only poor agreement with change in disease extent on HRCT scan (Kw = 0.17, P = .07).

CONCLUSIONS: Significant PFT trends correlate better with morphologic change as defined by serial HRCT scan than extent of disease on radiograph. Isolated change in gas transfer coefficient is more frequently discordant with change in disease extent on chest radiograph and HRCT scan and may suggest a pulmonary vascular component.