Background: Chronic Obstructive Pulmonary Disease (COPD) is a heterogeneous disease with a significant public health burden. Currently there is no biomarker that identifies those at risk of developing COPD, progression of disease or disease phenotypes. We performed metabolomic profiling of bronchoalveolar lavage fluid (BALF) from COPD patients to determine if metabolites correlated with clinical measurements such as lung function, functional status and degree of emphysema. Methods: Metabolomic components of BALF from 59 subjects with COPD and 20 healthy controls were separated by reversed-phase UPLC and analyzed by ESI-ToF mass spectrometry. We used univariate analysis and multiple regression models to investigate associations between metabolomic features and various clinical variables, such as lung function, functional status as measured by the St. George Respiratory Quotient Score and emphysema as measured by the CT density mask score. Results: We identified over 3900 features by mass spectrometry, many consistent with peptides. Subjects with severe COPD had increased concentration of peptides compared to controls (p < 9.526e-05). The peptide concentration correlated with spirometry, specifically pulmonary function tests associated with airflow obstruction. There was no correlation with CT density, i.e. emphysema, or functional status. Conclusions: Metabolomic profiling of BALF in COPD patients demonstrated a significant increase in peptides compared to healthy controls that associated strongly to lung function, but not emphysema or functional status.