Real-world screening for interstitial lung disease in rheumatoid arthritis: the value of spirometry, DLCO, and clinical risk factors

OBJECTIVE: This study aimed to characterize pulmonary function patterns in rheumatoid arthritis (RA), explore their clinical and serologic correlates, and evaluate the ability of spirometry and DLCO together with routine clinical factors to identify patients with high-resolution computed tomography (hrCT)-confirmed interstitial lung disease (ILD). METHODS: In this cross-sectional observational study, consecutive adults with RA underwent pulmonary function testing (PFT) and chest radiography irrespective of respiratory symptoms. Patients with unexplained dyspnea, abnormal PFT, or abnormal chest X-ray were referred for hrCT. RESULTS: Among the 106 included patients (81.1% women; mean age 65.3±9.7 years), DLCO correlated negatively with age and inflammatory markers, while FVC and FEV1 showed associations with serologic status, treatment exposure, and radiographic abnormalities. Forty-seven patients (44.3%) underwent hrCT, of whom 24 (51.1%) had ILD, corresponding to an overall prevalence of 22.6%. In the hrCT subgroup, swollen joint count at RA diagnosis was higher in ILD cases, whereas radiographic emphysema occurred only in non-ILD patients. In multivariable analysis, age, smoking history, and DAS28-CRP were not independently associated with ILD (AUC=0.634; 95%CI=0.451-0.807). ROC analyses demonstrated poor discrimination of ILD by PFT z-scores: DLCO AUC=0.431, FVC AUC=0.562, and FEV1 AUC=0.444, with high sensitivity but low specificity at optimal thresholds. CONCLUSIONS: In this real-world RA cohort, spirometry/DLCO showed limited ability to discriminate hrCT-confirmed ILD once patients were clinically selected for imaging. These findings support an integrated screening strategy in which PFT results are interpreted alongside clinical and radiographic risk factors to guide hrCT referral rather than used as standalone screening tools.