Mortality and Predictive Factors for Death Following the Diagnosis of Interstitial Lung Disease in Patients with...

Objective: The aim of this study was to determine mortality and predictive factors for death in patients with rheumatoid arthritis (RA) diagnosed with and without interstitial lung disease (ILD). Methods: We retrospectively performed a long-term follow-up study of patients diagnosed with RA at our medical center between April 2001 and June 2023. The diagnosis and classification of ILD were made based on pulmonary high-resolution computed tomography (HRCT), taken at RA diagnosis and during follow-up. Results: Among 781 patients with RA, 78 were diagnosed with ILD; all cases except one were subclinical. The most common HRCT pattern was definite usual interstitial pneumonia (UIP) followed by nonspecific interstitial pneumonia (NSIP)/UIP, probable UIP, NSIP, and early UIP. During follow-up (mean of 10.0 years), the crude incidence rate of death (95% confidence interval [CI]) was 7.1 (5.2-10.0) and 1.5 (1.0-1.9) per 100 person-years in RA patients with and without ILD. Poor control of RA activity was associated with increased incidence rates of death. The standardized mortality ratio (95% CI) compared with the general population was 1.32 (1.11-1.53) for all RA patients, 2.09 (1.45-2.73) for RA-ILD patients, and 1.16 (0.95-1.38) for non-ILD RA patients. Lung cancer and respiratory failure were the most common causes of death in RA-ILD patients. The Multivariable Fine-Gray regression analysis revealed that ILD (adjusted hazard ratio [HR] 2.97 [95% CI 1.95-4.53]), advanced age (1.08 per additional year [1.05-1.10]), and low body mass index (3.07 [2.10-4.49]) were strong predictive factors for mortality in RA patients. HRCT patterns did not affect the risk of death in RA-ILD patients. Conclusions: Regardless of HRCT pattern, RA-ILD contributes to the increased mortality risk in patients with RA.