The important metabolic features of acute pulmonary embolism (APE) risk stratification and their underlying biological basis remain elusive. Our study aims to develop early diagnostic models and classification models by analyzing the plasma metabolic profile of patients with APE.
Serum samples were collected from 68 subjects, including 19 patients with confirmed APE, 35 patients with confirmed NSTEMI, and 14 healthy individuals. A comprehensive metabolic assessment was performed using ultra-performance liquid chromatography-mass spectrometry based on an untargeted metabolomics approach. In addition, an integrated machine learning strategy based on LASSO and logistic regression was used for feature selection and model building.
The metabolic profiles of patients with acute pulmonary embolism and NSTEMI is significantly altered relative to that of healthy individuals. KEGG pathway enrichment analysis revealed differential metabolites between acute pulmonary embolism and healthy individuals mainly involving glycerophosphate shuttle, riboflavin metabolism, and glycerolipid metabolism. A panel of biomarkers was defined to distinguish acute pulmonary embolism, NSTEMI, and healthy individuals with an area under the receiver operating characteristic curve exceeding 0.9 and higher than that of D-dimers.
This study contributes to a better understanding of the pathogenesis of APE and facilitates the discovery of new therapeutic targets. The metabolite panel can be used as a potential non-invasive diagnostic and risk stratification tool for APE.