Serum metabolomics analysis of patients with chronic obstructive pulmonary disease and 'frequent exacerbator' phenotype.

Chronic obstructive pulmonary disease (COPD) exacerbations accelerate loss of lung function and increased mortality. The complex nature of COPD presents challenges in accurately predicting and understanding frequent exacerbations. The present study aimed to assess the metabolic characteristics of the frequent exacerbation of COPD (COPD‑FE) phenotype, identify potential metabolic biomarkers associated with COPD‑FE risk and evaluate the underlying pathogenic mechanisms. An internal cohort of 30 stable patients with COPD was recruited. A widely targeted metabolomics approach was used to detect and compare serum metabolite expression profiles between patients with COPD‑FE and patients with non‑frequent exacerbation of COPD (COPD‑NE). Bioinformatics analysis was used for pathway enrichment analysis of the identified metabolites. Spearman's correlation analysis assessed the associations between metabolites and clinical indicators, while receiver operating characteristic (ROC) analysis evaluated the ability of metabolites to distinguish between two groups. An external cohort of 20 patients with COPD validated findings from the internal cohort. Out of the 484 detected metabolites, 25 exhibited significant differences between COPD‑FE and COPD‑NE. Metabolomic analysis revealed differences in lipid, energy, amino acid and immunity pathways. Spearman's correlation analysis demonstrated associations between metabolites and clinical indicators of acute exacerbation risk. ROC analysis demonstrated that the area under the curve (AUC) values for D‑fructose 1,6‑bisphosphate (AUC=0.871), arginine (AUC=0.836), L‑2‑hydroxyglutarate (L‑2HG; AUC=0.849), diacylglycerol (DG) (16:0/20:5) (AUC=0.827), DG (16:0/20:4) (AUC=0.818) and carnitine‑C18:2 (AUC=0.804) were >0.8, highlighting their discriminative capacity between the two groups. External validation results demonstrated that DG (16:0/20:5), DG (16:0/20:4), carnitine‑C18:2 and L‑2HG were significantly different between patients with COPD‑FE and those with COPD‑NE. In conclusion, the present study offers insights into early identification, mechanistic understanding and personalized management of the COPD‑FE phenotype.