Send to

Choose Destination
Chronic Obstr Pulm Dis. 2019 Nov;6(5):400-413. doi: 10.15326/jcopdf.6.5.2019.0150.

Subtypes of COPD Have Unique Distributions and Differential Risk of Mortality.

Author information

Department of Epidemiology, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora.
Department of Medicine, National Jewish Health, Denver, Colorado.
Division of Pulmonary and Critical Care, University of Michigan, Ann Arbor.
Department of Biostatistics and Informatics, University of Colorado Anschutz Medical Campus, Aurora.
Channing Division of Network Medicine, Brigham and Women's Hospital, Boston, Massachusetts.
Division of Biostatistics and Bioinformatics, Office of Academic Affairs, National Jewish Health, Denver, Colorado.
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore Maryland.
VA Boston Healthcare System, Boston, Massachusetts.
Department of Radiology, National Jewish Health, Denver, Colorado.



Previous attempts to explore the heterogeneity of chronic obstructive pulmonary disease (COPD) clustered individual patients using clinical, demographic, and disease features. We developed continuous multidimensional disease axes based on radiographic and spirometric variables that split into an airway-predominant axis and an emphysema-predominant axis.


The COPD Genetic Epidemiology study (COPDGene®) is a cohort of current and former smokers, > 45 years, with at least 10 pack years of smoking history. Spirometry measures, blood pressure and body mass were directly measured. Mortality was assessed through continuing longitudinal follow-up and cause of death was adjudicated. Among 8157 COPDGene® participants with complete spirometry and computed tomography (CT) measures, the top 2 deciles of the airway-predominant and emphysema-predominant axes previously identified were used to categorize individuals into 3 groups having the highest risk for mortality using Cox proportional hazard ratios. These groups were also assessed for causal mortality. Biomarkers of COPD (fibrinogen, soluble receptor for advanced glycation end products [sRAGE], C-reactive protein [CRP], clara cell secretory protein [CC16], surfactant-D [SP-D]) were compared by group.


High-risk subtype classification was defined for 2638 COPDGene® participants who were in the highest 2 deciles of either the airway-predominant and/or emphysema-predominant axis (32% of the cohort). These high-risk participants fell into 3 groups: airway-predominant disease only (APD-only), emphysema-predominant disease only (EPD-only) and combined APD-EPD. There was 26% mortality for the APD-only group, 21% mortality for the EPD-only group, and 54% mortality for the combined APD-EPD group. The APD-only group (n=1007) was younger, had a lower forced expiratory volume in 1 second (FEV1) percent (%) predicted and a strong association with the preserved ratio-impaired spirometry (PRISm) quadrant. The EPD-only group (n=1006) showed a relatively higher FEV1 % predicted and included largely GOLD stage 0, 1 and 2 partipants. Individuals in each of the 3 high-risk groups were at greater risk for respiratory mortality, while those in the APD-only group were additionally at greater risk for cardiovascular mortality. Biomarker analysis demonstrated a significant association of the APD-only group with CRP, and sRAGE demonstrated greatest significance with both the EPD-only and the combined APD-EPD groups.


Among current and former smokers, individuals in the highest 2 deciles for mortality risk on the airway-predominant axis and the emphysema-predominant axis have unique associations to spirometric patterns, different imaging characteristics, biomarkers and causal mortality.


COPD Genetic Epidemiology; COPDGene; PRISm; airway-predominant disease; biomarkers; chronic obstructive pulmonary disease; copd; emphysema-predominant disease; mortality; preserved ratio-impaired spirometry; subtypes

Free full text

Supplemental Content

Full text links

Icon for COPD Foundation
Loading ...
Support Center