Sensitivity and specificity of various lung function parameters recommended for phenotyping COPD in clinical practice

Chronic obstructive pulmonary disease (COPD) is characterized by irreversible airflow limitation and heterogeneous structural impairments that affect both the airways and lung parenchyma. These variations result in different phenotypes with distinct clinical progressions and therapeutic responses. Despite several pulmonary function tests, identifying the underlying pathogenetic determinants of COPD in clinical practice remains challenging.

to assess and compare the diagnostic accuracy, sensitivity and specificity of function parameters in predicting and estimating emphysema burden detected by CT imaging.

Stable COPD patients aged ≥40 years without comorbidities influencing lung function were included. Lung function variables assessed included: BMI, FEV₁, FEV₁/FVC, RV, TLC, RV/TLC, Emphysema Severity Index (ESI), DL_CO, simultaneous single-breath sDL_NO/sDL_CO, and lung capillary blood volume (Vc). Emphysema was quantified by CT scans and categorized as "high emphysema burden" if CT-emphysema was >10 % of lung volume and "low emphysema burden" otherwise. ROC curve analysis evaluated the predictive power of lung function parameters alone or in combination.

The study involved 78 patients (34 with low and 44 with high emphysema burden). Significant differences in BMI and lung function were observed between subgroups. Key findings showed that Vc values had the highest correlation to CT-emphysema (R² = 0.8226), followed by sDL_CO (R² = 0.7445). The RV + Vc model achieved the highest predictivity (pseudo R² = 0.667).

Integrating novel lung function technologies with traditional methods can enhance COPD phenotyping. Specific parameters like Vc and sDL_NO/sDL_CO provide higher sensitivity and specificity in identifying the destructive components of COPD, supporting a multi-parametric approach in clinical practice.