Diffusing capacity of the lung for carbon monoxide and total lung capacity may be associated with interstitial lung disease severity in patients with connective tissue diseases, according to a study published in RMD Open.
Investigators analyzed 76 patients with connective tissue diseases (CTD) and interstitial lung disease (ILD) (mean age = 59 years; 71% women) who underwent pulmonary function testing and high-resolution computed tomography (HRCT) between 2010 and 2022. They used artificial intelligence (AI)-based quantification of pulmonary HRCT (AIqpHRCT) through the SATORI platform to quantify ILD features such as ground-glass opacities, reticulations, high-attenuation lung volume, emphysema, and overall extent of ILD.
Multiple linear regression analysis adjusted for demographics, behavioral health factors, and substance use revealed consistent associations. Male gender, total lung capacity (TLC), and diffusing capacity of the lung for carbon monoxide (DLCO) demonstrated significant negative correlations with all ILD severity measures.
For high-attenuation lung volume, male gender showed a beta coefficient of −0.65, TLC showed −0.47, and DLCO showed −0.36. Similar patterns emerged for ground-glass opacities (beta coefficients = −0.86, −0.53, and −0.34, respectively), reticulations (beta coefficients = −0.49, −0.53, and −0.39), and overall extent of ILD (beta coefficients = −0.75, −0.58, and −0.37).
The study population comprised patients with the CTDs systemic sclerosis (40.8%), systemic lupus erythematosus (165.8%), Anti-Jo1 syndrome (14.5%), dermatomyositis (10.5%), Sjögren's disease (7.9%), mixed CTD (6.6%), and polymyositis (3.9%). Most of the patients (64.5%) received concurrent initial diagnoses of CTD and CTD-ILD.
Mean pulmonary function values were 78.1% ± 17.7% predicted for forced expiratory volume in 1 second, 75.5% ± 17.5% for forced vital capacity, and 77.4% ± 14.8% for TLC. DLCO showed the most significant reduction at 51.6% ± 16.9%. AIqpHRCT demonstrated mean values of 8.91% ± 9.66% for ground-glass opacities, 4.06% ± 7.33% for reticulations, 12.70% ± 9.37% for high-attenuation lung volume, and 13.60% ± 16.20% for overall extent of ILD.
C-reactive protein showed positive associations with high-attenuation lung volume (beta = 0.22), ground-glass opacities (beta = 0.22), reticulations (beta = 0.21), and overall extent of ILD (beta = 0.22). Pulmonary symptoms, including dyspnea, cough, and bibasilar inspiratory crackles didn't demonstrate statistically significant associations with ILD severity in multivariate models.
Although commonly used as a primary endpoint in therapeutic trials for CTD-ILD, forced vital capacity showed no statistically significant association with any ILD severity measure in multivariate analysis.
Study limitations included the monocentric retrospective design, limited sample size of 76 patients, and potential selection bias. The investigators also couldn't rule out that patients with poorer pulmonary function were more likely to receive imaging.
"DLCO and TLC proved to be important parameters that determine the extent of ILD in [patients with] CTD," wrote lead study author Tobias Hoffmann, MD, MPH, of the Department of Internal Medicine III at the Friedrich Schiller University Jena, and colleagues.
The study was partially funded by the Bundesministerium für Bildung und Forschung, Germany, as part of Netzwerk Universitätsmedizin. The authors reported no competing interests.
Source: RMD Open
