Chest Radiograph-Derived Age Acceleration as an Early Marker of Pulmonary Dysfunction in Middle-Aged Asian Adults

Area of Science:

• Pulmonary Medicine
• Radiology
• Artificial Intelligence

Background:

• Deep learning models can estimate chronological age from chest X-rays (CXRs), potentially reflecting subclinical lung changes.
• Spirometric impairment, including preserved ratio impaired spirometry (PRISm) and obstructive lung disease (OLD), significantly impacts respiratory health.

Purpose of the Study:

• To investigate the association between accelerated aging on CXRs and the prevalence and incidence of PRISm and OLD.
• To determine if CXR-derived aging is a predictor of future lung function decline.

Main Methods:

• Retrospective analysis of 231,278 Korean adults who underwent CXRs and spirometry.
• Participants categorized into decelerated, reference, or accelerated aging groups based on CXR-Lung-Risk scores.
• Multivariable logistic and Cox proportional hazards models used to assess associations with prevalent and incident PRISm and OLD.

Main Results:

• Accelerated CXR aging was associated with higher odds of prevalent PRISm (aOR, 1.37) and OLD (aOR, 1.58).
• Accelerated CXR aging predicted higher risks of incident PRISm (aHR, 1.37) and OLD (aHR, 1.28) in longitudinal analysis.
• Associations remained consistent across various demographic and lifestyle subgroups.

Conclusions:

• CXR-derived accelerated aging is a significant indicator for both current and future PRISm and OLD.
• Routine CXRs may serve as a valuable tool for opportunistic screening of pulmonary dysfunction.