All-cause mortality in patients with COPD over 10 years of follow-up was accurately predicted by a newly developed model based on a point system incorporating imaging-derived variables.
Identifying risk factors is important to develop treatments and preventive strategies, but the role of imaging variables in COPD mortality among smokers has not been well studied, wrote investigator Matthew Strand, PhD, of National Jewish Health in Denver, and colleagues.
An established risk model is the body mass index–airflow Obstruction-Dyspnea-Exercise capacity (BODE) index, developed to predict mortality in COPD patients over a 4-year period. The investigators noted that while models such as BODE provide useful information about predictors of mortality in COPD, they were developed using participants in the Global initiative for obstructive Lung Disease (GOLD) spirometry grades 1-4, and have been largely constructed without quantitative computed tomography (CT) imaging variables until recently.
“The BODE index was created as a simple point scoring system to predict risk of all-cause mortality within 4 years, and is based on FEV1 [forced expiratory volume at 1 second], [6-minute walk test], dyspnea and BMI, a subset of predictors we considered in our model,” the investigators noted. The new model includes data from pulmonary function tests and volumetric CT scans.
In a study published in Chronic Obstructive Pulmonary Diseases, the researchers identified 9,074 current and past smokers in the COPD Genetic Epidemiology study (COPDGene) for whom complete data were available. They developed a point system to determine mortality risk in current and former smokers after controlling for multiple risk factors. The average age of the study population was 60 years. All participants were current or former smokers with a smoking history of at least 10 pack-years.
Assessments of the study participants included a medical history, pre- and post-bronchodilator spirometry, a 6-minute walk distance test, and inspiratory and expiratory CT scans. The researchers analyzed mortality risk in the context of Global Initiative for Obstructive Lung Disease (GOLD) classifications of patients in the sample.
Overall, the average 10-year mortality risk was 18% for women and 25% for men. Performance on the 6-minute walk test (distances less than 500 feet), FEV1 (less than 20), and older age (80 years and older) were the strongest predictors of mortality.
The model showed strong predictive accuracy, with an area under the receiver operating characteristic curve averaging 0.797 that was validated in an external cohort, the researchers said.
The study findings were limited by the observational design that does not allow for estimating the causal effects of such modifiable factors as smoking cessation, that might impact the walking test and FEV1 values, the researchers noted. In addition, the model did not allow for testing the effects of smoking vs. not smoking.
However, the model developed in the study “will allow physicians and patients to better understand factors affecting risk of an adverse event, some of which may be modifiable,” the researchers said. “The risk estimates can be used to target groups of individuals for future clinical trials, including those not currently classified as having COPD based on GOLD criteria,” they said.
The study was supported by the National Heart, Lung, and Blood Institute and by the COPD Foundation through contributions to an industry advisory committee including AstraZeneca, Boehringer-Ingelheim, Genentech, GlaxoSmithKline, Novartis, Pfizer, Siemens, and Sunovion.