Unclassifiable-interstitial lung disease: Outcome prediction using CT and functional indices

Area of Science:

• Pulmonology
• Radiology
• Medical Diagnostics

Background:

• Unclassifiable-interstitial lung disease (uILD) is a complex group of lung diseases lacking definitive diagnostic criteria.
• Identifying reliable predictors of disease progression and mortality in uILD is crucial for patient management.
• This study investigates both functional and computed tomography (CT) imaging variables for outcome prediction in uILD.

Purpose of the Study:

• To identify baseline and longitudinal predictors of outcome in patients with unclassifiable-interstitial lung disease (uILD).
• To evaluate the predictive power of functional tests and CT imaging, including quantitative analysis (CALIPER), for mortality and disease progression.
• To determine the independent contribution of various clinical and imaging parameters to patient outcomes in uILD.

Main Methods:

• A cohort of 95 uILD patients underwent comprehensive baseline assessments, including pulmonary function tests (FVC, DLco, CPI) and CT scans (visual and CALIPER analysis).
• Univariate and multivariate Cox regression models were used to analyze baseline variables for outcome prediction.
• Functional and CT variables were also analyzed longitudinally in a subset of 37 patients to identify dynamic outcome indicators.

Main Results:

• The composite physiologic index (CPI) was the strongest functional predictor of mortality (p < 0.0001).
• CT findings, including traction bronchiectasis, pulmonary artery diameter, and honeycombing, along with quantitative CALIPER pulmonary vessel volume (PVV), were significant outcome predictors.
• Multivariate analysis confirmed traction bronchiectasis, PA diameter, and CPI as independent predictors of mortality. Longitudinal analysis revealed increasing CALIPER fibrosis extent as the strongest outcome predictor, even with minimal FVC decline.

Conclusions:

• Composite physiologic index (CPI), traction bronchiectasis severity, and pulmonary artery diameter are independent baseline predictors of outcome in uILD.
• Quantitative CT-derived fibrosis extent (CALIPER) is a powerful predictor of outcome, irrespective of baseline disease severity.
• CALIPER fibrosis extent effectively predicts outcomes, particularly in patients experiencing marginal declines in FVC.