Screening metabolic features of stroke in atrial fibrillation

Abstract

Background: Atrial fibrillation (AF) is an independent risk factor for stroke, increasing the likelihood of stroke by three to five times. AF-related stroke (AFS) has a poor prognosis, making early prediction crucial. However, the underlying mechanisms have only been studied in limited samples. This lack of understanding of pathogenic pathways hampers the development of targeted therapies and patient risk stratification.

Results: Our study increased the patient count to 90 individuals with AF and 66 with AFS. We collected plasma samples from these patients, detected 671 metabolites, and performed Metabolite Set Enrichment Analysis (MSEA), which revealed differences between the two groups. The results showed statistically significant variations in 12 metabolic pathways between the two groups. In plasma samples from AFS patients, we observed 25 up-regulated and 26 down-regulated differentially expressed metabolites (DEMs), according to the statistical significance criteria of fold change >1.2 or < 0.83, p < 0.05, and VIP score >1. The DEMs demonstrated enrichment in pathways involved in arachidonic acid, taurine, and hypotaurine metabolism, as well as other related processes. Furthermore, we established a predictive model based on 12 metabolites to determine their potential as biomarkers for predicting stroke in patients with AF using Lasso regression. The model attained a remarkable area under the curve of 0.96. Finally, we utilized Weighted Gene Co-expression Network Analysis to examine the clinical comorbidities in individuals with AF and AFS.

Significance: This study provides an in-depth exploration of the basic pathogenic mechanisms involved in how AF leads to stroke in a large sample. In addition, it identifies new biomarkers that can be used to assess the likelihood of stroke and enhance outcomes in individuals with AF. These findings hold great promise for the development of more effective preventive and therapeutic strategies for stroke in patients with AF in the near future.

Keywords: Atrial fibrillation; Biomarkers; Comorbidities; Metabolomics; Stroke.