Development and Validation of a Pre-COPD Score to Predict Incident COPD

The ability to predict progression from a pre-disease state in chronic obstructive pulmonary disease (COPD), pre-COPD, to overt COPD remains limited.

To test the hypothesis that development of COPD can be predicted by characterizing pre-COPD and creating a simple pre-COPD score.

We identified individuals without airflow obstruction, defined as pre-bronchodilator (BD) forced expiratory volume in 1 second (FEV1)/forced vital capacity (FVC)≥0.70 and ≥lower limit of normal, from the Copenhagen General Population Study (CGPS)(N = 14,506), Copenhagen City Heart Study (CCHS)(N = 8673), and Genetic Epidemiology for COPD (COPDGene) Study(N = 2982). A pre-COPD score was developed in CGPS by assigning 1 point to each of smoking history≥10 pack-years, active smoking, any respiratory symptom, asthma, and pre-BD FEV1/FVC 0.70-0.75, and 4 points for pre-BD FEV1<80% predicted, indicating presence of Preserved Ratio Impaired Spirometry (PRISm), with external validation in CCHS and COPDGene. Participants were followed for 5-10 years for incident COPD, defined as pre-BD FEV1/FVC<0.70 and FEV1<80% predicted.

During follow-up, COPD developed in 669 individuals (4.6%) in CGPS, 685 (7.9%) in CCHS, and 367 (12%) in COPDGene. In CGPS, 10-year incidence was 1.5% in the low-risk, 9.3% in the intermediate-risk, 24% in the high-risk, and 42% in the very high-risk category. Corresponding 5-year incidences were 3.7%, 7.3%, 12%, and 25% in CCHS, and 4.1%, 5.3%, 13%, and 31% in COPDGene, respectively. The pre-COPD score demonstrated good discrimination with areas under curve of 0.88 in CGPS, 0.76 in CCHS, and 0.81 in COPDGene. The pre-COPD score was also associated with exacerbations.

The pre-COPD score is useful in predicting the onset of COPD with clinically relevant discrimination in three cohorts with different risk profiles.