A multicenter cohort study proposed a new multidimensional approach to diagnosing chronic obstructive pulmonary disease incorporating computed tomography imaging and respiratory symptoms to identify patients at risk for poor outcomes. The findings were validated in an independent cohort.
The proposed diagnostic schema included additional patients with high mortality and respiratory morbidity and excluded those with airflow obstruction on spirometry who lacked symptoms or evidence of structural lung disease.
Key Findings
Investigators analyzed 9,416 participants from the COPDGene cohort and validated their findings in 1,341 participants from the CanCOLD study. In COPDGene, 15.4% (n = 811/5,250) of the patients without airflow obstruction were newly classified as having chronic obstructive pulmonary disease (COPD) by minor diagnostic category, and 6.8% (n = 282/4,166) of the patients with airflow obstruction were reclassified as not having COPD.
Reclassified patients who received a new COPD diagnosis had greater all-cause mortality (adjusted hazard ratio [HR] = 1.98, 95% confidence interval [CI] = 1.67–2.35, P < .001) and respiratory-specific mortality (HR = 3.58, 95% CI = 1.56–8.20, P = .003), more exacerbations (adjusted incidence rate ratio [IRR] = 2.09, 95% CI = 1.79–2.44, P < .001), and more rapid decline in forced expiratory volume in the first second of expiration (FEV₁) (adjusted beta = −7.7 mL/y, 95% CI = −13.2 to −2.3, P = .006) compared with individuals classified as not having COPD.
New Diagnostic Framework
The diagnostic schema included a major criterion and five minor criteria. The major criterion was airflow obstruction, defined by a postbronchodilator FEV₁/forced vital capacity ratio less than 0.70 for the primary analyses and below the lower limit of normal in sensitivity analyses. The five minor criteria included two imaging findings (emphysema and bronchial wall thickening on visual chest computed tomography [CT] analysis) and three symptom-based criteria (dyspnea, poor respiratory quality of life, and chronic bronchitis).
COPD can be diagnosed in 2 ways: by meeting the major diagnostic category (major criterion plus ≥ 1 minor criterion) or the minor diagnostic category (≥ 3 of 5 minor criteria). When symptoms may be attributed equally or more to other comorbidities, such as coronary artery disease or heart failure, clinicians are advised to confirm both imaging criteria are present among the 3 required minor criteria.
Clinical Outcomes Validate the Schema
During a median 10.5-year follow-up in COPDGene, patients diagnosed by the minor diagnostic category had a significantly higher risk of poor outcomes compared with those without COPD. Respiratory-specific mortality rates were 0.5 (no COPD), 18.9 (COPD by new schema overall), 1.5 (COPD by minor category), and 22.3 (COPD by major category) per 1,000 person-years.
The participants classified as not having COPD based on the new schema but who met traditional spirometric airflow obstruction criteria had clinical outcomes similar to those without airflow obstruction.
Study Population and Methods
The COPDGene cohort enrolled individuals aged 45 to 80 years who currently or formerly smoked. Following exclusions, 9,416 participants remained (mean [SD] age = 59.6 [9.0] years, 53.5% male, 32.6% Black, and 67.4% White). The CanCOLD cohort included 1,341 participants following exclusions.
Outcomes assessed included all-cause mortality, respiratory-specific mortality, COPD exacerbations, and annualized change in FEV₁. Adjusted models accounted for age, sex, race, body mass index, smoking status, and pack-years.
Rationale for New Criteria
Current guidelines do not incorporate imaging findings, even though up to half of smokers without airflow obstruction may have evidence of emphysema or airway wall thickening on CT. These patients are at significantly increased risk of developing spirometric COPD over 5 years. The new schema integrates these imaging findings with symptom measures to provide a more comprehensive diagnostic approach.
Limitations
The investigators noted limitations including reliance on CT imaging, possible interobserver variability in image interpretation, limited generalizability beyond Black and White populations, and potential misclassification of asthma as COPD. Although CT imaging may not be universally available, it's more commonly acquired worldwide than spirometry.
Funding and Disclosures
The study was funded by the National Heart, Lung, and Blood Institute and supported by the COPD Foundation and Canadian Respiratory Research Network. Conflicts of interest are available in the published study.
Source: JAMA
