Early Identification of Serum Biomarkers and Pathways of Sepsis Through GC-MS-Based Metabolomics Analysis.

BACKGROUND

Early identification of sepsis improves the survival rate; however, it is one of the most challenging tasks for physicians, especially since symptoms are easily confused with those of systemic inflammatory response syndrome (SIRS). Our aim was to explore biomarkers for early identification of sepsis that would aid in its differential diagnosis.

METHODS

Eight patients with SIRS, eight with sepsis, and eight healthy controls were included in this study. Metabolites were screened using gas chromatography-mass spectrometry (GC-MS). Metabolism profiles were analyzed using the untargeted database of GC-MS from Lumingbio (LUG) database, and metabolic pathways were enriched based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. The S-plot was used to screen the potential biomarkers distinguishing between patients with SIRS, sepsis, and healthy controls. Receiver operating characteristic (ROC) curve analysis was used to evaluate potential biomarkers between SIRS and sepsis patients. Correlation analysis was used to measure the degree of correlation between differential metabolites. Correlation analysis between 2-deoxy-d-erythro-pentofuranose-5-phosphate and clinical indicators was performed.

RESULTS

There were 51 metabolites that were distributed in the SIRS group, and they were enriched with 18 metabolic pathways compared with healthy controls. Moreover, 63 metabolites in the sepsis group were significantly distinguishable compared to the healthy controls, and were associated with 21 metabolic pathways. Methyl 3-o-acetyl-d-galactopyranoside and N-acetylputrescine were found to be candidate biomarkers for distinguishing between SIRS, sepsis, and healthy controls using the S-plot model. Only four differential metabolites, including 2-deoxy-d-erythro-pentofuranose-5-phosphate, terbutaline, allantoic acid, and homovanillic acid (HVA), were enriched in the dopaminergic synapse and tyrosine metabolism pathways when sepsis patients were compared with SIRS patients. The Area Under Curve (AUC) of 2-deoxy-d-erythro-pentofuranose-5-phosphate was 0.9297, indicating a strong diagnostic ability for sepsis. A significant negative correlation was identified between 2-deoxy-d-erythro-pentofuranose-5-phosphate and lactate (r = -0.8756, = 0.0044).

CONCLUSIONS

Methyl 3-o-acetyl-d-galactopyranoside and N-acetylputrescine may be used as candidate biomarkers to distinguish SIRS and sepsis patients from healthy controls using GC-MS. 2-deoxy-d-erythro-pentofuranose-5-phosphate may be the candidate biomarker to distinguish sepsis from SIRS. Our study explored candidate biomarkers for the early identification of sepsis, which is vital for improving its prognosis.