Coronavirus disease 2019 (COVID-19) global pandemic has affected more than 600 million people up to date. The symptomatology and severity of COVID-19 are very broad, and there are still concerns about the long-term sequelae that it can have on discharged patients. The development of pulmonary fibrotic sequelae after this infection is especially worrying. Our aim was to determine if there was a metabolomic signature that could predict the development of pulmonary fibrotic sequelae. It is a multicenter prospective observation subcohort based on the COVID-FIBROTIC study. A metabolomic analysis was performed by nuclear magnetic resonance (NMR) on serum samples from patients admitted with bilateral COVID-19 pneumonia collected 2 months after hospital discharge. One year after admission, clinical, functional and radiological data were collected from these same patients. Finally, 109 patients (mean age 57.68 [DS14.03], 65.13% male) were available. Fibrotic sequelae 1 year after COVID-19 disease were found in 33% of them. Based on the NMR analysis of the serum samples, it was possible to distinguish with 80.82% of sensitivity, 72.22% of specificity and 0.83 of an area under the curve (AUC) value which patients would have radiological signs of pulmonary fibrotic pattern 1 year after sample collection. According to the metabolites participating in the discriminative model and the univariate statistics, glucose, valine, and fatty acids (═CH-CH2-CH═) were suggested as potential biomarkers of the development of pulmonary fibrotic sequelae after COVID-19. TRIAL REGISTRATION NUMBER: clinicaltrials.gov NCT04409275 (June 1, 2020).