The Role of Gut Bacteria and Plasma Metabolites in Predicting Post-Stroke Depression in Patients with Acute Ischemic Stroke.

Introduction: Early diagnosis of Post-Stroke Depression (PSD) is challenging. This study aimed to identify possible biomarkers in gut microbiota and plasma metabolites within 72 hours after Acute Ischemic Stroke (AIS) to predict PSD occurring 2 weeks later.

Method: In this study, 86 patients with AIS were observed within 3 days of stroke onset and followed up for 2 weeks. We collected the feces and plasma within 72 hours of AIS onset for 16S rRNA sequencing and liquid chromatography-mass spectrometry analysis, respectively.

Results: At the genus level, PSD patients at 2 weeks following a stroke had a higher relative abundance of Blautia, Eubacterium_hallii_group, Tyzzerella, and a lower abundance of Ellin6067, Massilia, Luedemannella, and Gemmataceae_others within 3 days of AIS onset. Meanwhile, when all metabolites in plasma collected within 72 hours after AIS onset were used to predict 2-week PSD, 31 altered metabolites were identified, of which 28 metabolites increased and 3 decreased, belonging predominantly to steroid and steroid derivatives, glycerophospholipids, fatty acyls, and prenol lipids. The Area Under the Curve (AUC) values for the clinical data, metabolic profiles, gut microbiota, and combined dataset were 0.664 (0.549,0.779), 0.739 (0.621, 0.857), 0.870 (0.781,0.960), and 0.955 (0.888,1), respectively.

Discussion: Our study identified potential biomarkers from clinical data, gut bacteria, and plasma metabolites that contribute to PSD. Within 72 hours after AIS, combining these biomarkers from all three sources showed preliminary ability to predict PSD at 2 weeks. Metabolites had the highest contribution, followed by gut bacteria and clinical data.

Conclusion: A biomarker panel including metabolites, gut microbiota, and clinical data within 72 hours after AIS onset could preliminarily predict PSD 2 weeks later.