Phenotyping Chronic Obstructive Pulmonary Disease Through Principal Component Analysis: Identification of Clinical Clusters

Introduction Chronic obstructive pulmonary disease (COPD) is a heterogeneous condition with varied clinical presentations and prognoses. Identifying patient phenotypes is essential for developing personalized treatment strategies. Principal component analysis (PCA) is a statistical method that can be employed to uncover clinical clusters and gain insight into the relationships among different disease characteristics. This study aims to analyze COPD patient phenotypes using PCA and to identify the key clinical features influencing their distribution. Materials and methods This was a prospective, observational outpatient study involving 96 patients diagnosed with COPD. Data collected included demographic, clinical, spirometric, echocardiographic, laboratory, and functional parameters. PCA was applied to reduce data dimensionality and to identify the principal components underlying phenotype structure. Results The first two principal components accounted for 62% of the total variance, underscoring the clinical heterogeneity of COPD. Visualization of the PCA revealed four distinct clusters that align with recognized COPD phenotypes: chronic bronchitis, emphysema, COPD with asthmatic features (previously referred to as asthma-COPD overlap), and the non-exacerbator type. Each cluster was associated with specific clinical characteristics. Conclusions PCA enabled the identification of four distinct clinical clusters among COPD patients: bronchitis, emphysema, COPD with asthmatic features, and non-exacerbator. This approach helps clarify the relationship between clinical characteristics and supports a more personalized approach to treatment.