Computed tomography has the potential to inform COPD prognosis. We sought to determine associations of emphysema phenotype with clinical parameters including lung function, inflammatory markers, and quality of life.

Participants of this single-center observational cohort (n = 83) were 40–80 years old, had ≥10 pack-year smoking, and a diagnosis of COPD confirmed by spirometry. All participants had available historic chest CT scans which were systematically reviewed by a single expert radiologist and scored for emphysema subtype, extent, and distribution. Associations between radiographic findings and clinical parameters were determined.

Median age of participants was 72 years, median smoking 40 pack-years, and median FEV₁ 59% predicted. 84% of the participants had radiographic emphysema. Of those, 26% had panlobular emphysema (PLE), 68% centrilobular emphysema (CLE), and 6% paraseptal emphysema (PSE). As compared to the participants with no radiographic emphysema, the presence of PLE-dominant emphysema was associated with a lower BMI (P = 0.012) and greater extent of emphysema (P = 0.014). After adjusting for age, sex, and pack-years smoking history, PLE was associated with greater airflow obstruction by FEV₁% (48% vs 71%, P = 0.005), greater symptom burden by CAT score (18 vs 9, P = 0.015), worse quality of life by SGRQ score (43 vs 22, P = 0.025), and more systemic inflammation by erythrocyte sedimentation rate (P = 0.001). CLE- or PSE-dominant emphysema were not similarly associated with clinical features or symptom burden.

The presence of PLE-dominant emphysema was associated with greater extent of emphysema, greater airflow obstruction, increased respiratory symptoms, worse quality of life, and systemic inflammation. Further investigation is indicated to explore the pathogenesis of the PLE phenotype and the prognostic and treatment implications of PLE.