Pulmonary arterial volume and mortality in smokers without chronic obstructive pulmonary disease.

BACKGROUND

Pulmonary vascular remodelling is a recognised consequence of chronic tobacco exposure and may have prognostic associations even without airway obstruction. The association between the extraparenchymal pulmonary artery (ePA) and all-cause mortality in current and ex-smokers without COPD has not been demonstrated and replicated in large cohort studies.

STUDY DESIGN AND

METHODS

Retrospective analysis of prospective data from the COPDGene cohort, with validation in the SUMMIT lung cancer screening trial. ePA volume and lung volume were measured on chest CT using automated deep learning-based segmentation. All-cause mortality risk was estimated using Cox proportional hazards models, adjusted for age, gender, body mass index, smoking status, pack-years, emphysema extent, Charlson Comorbidity Index, ILA presence, and FEV1.

RESULTS

4,719 participants without COPD were included in the discovery COPDGene cohort (median age 56 [51-63] years, 50% male) and 3,391 in the SUMMIT validation cohort (median age 64 [59-69] years, 61% male). There were 600 deaths (13%; 39,531 person-years) in COPDGene and 214 deaths (6.3%; 16,629 person-years) in SUMMIT. Each one-standard-deviation (∼25mL) increase in ePA volume was independently associated with all-cause mortality in COPDGene (aHR 1.42 [1.25-1.60], p<0.001) and SUMMIT (aHR 1.23 [1.07-1.42], p=0.003). A parsimonious, ridge-regularised multivariable model developed in COPDGene showed good discrimination (C-index 0.70 [0.65-0.74]) and calibration (integrated calibration index 2.1% [1.3-6.4]) when validated in SUMMIT.

INTERPRETATION: Increasing ePA volume in current and ex-smokers without COPD is associated with all-cause mortality and could represent a candidate imaging marker for cohort enrichment in studies of pulmonary vasculopathy, although this requires prospective validation.

TRIAL REGISTRATION: N/A.