Gut Microbiome-Based Diagnostic Model to Predict Coronary Artery Disease.

In the present study, we aimed to characterize gut microbiome and develop a gut microbiome-based diagnostic model in patients with coronary artery disease (CAD). Prospectively, we collected 309 fecal samples from Central China and Northwest China and carried out the sequencing of the V3-V4 regions of the 16S rRNA gene. The gut microbiome was characterized, and microbial biomarkers were identified in 152 CAD patients and 105 healthy controls (Xinjiang cohort, = 257). Using the biomarkers, we constructed a diagnostic model and validated it externally in 34 CAD patients and 18 healthy controls (Zhengzhou cohort, = 52). Fecal microbial diversity was increased in CAD patients compared to that in healthy controls ( = 0.021). Phylum was increased in CAD patients versus healthy controls ( = 0.001). Correspondingly, 48 microbial markers were identified through a 10-fold cross-validation on a random forest model, and an area under the curve (AUC) of 87.7% (95% CI: 0.832 to 0.916, < 0.001) was achieved in the Xinjiang cohort (development cohort, = 257). Notably, an AUC of 90.4% (95% CI: 0.848 to 0.928, < 0.001) was achieved using combined analysis of gut microbial markers and clinical variables. This model provided a robust tool for the prediction of CAD. It could be widely employed to complement the clinical assessment and prevention of CAD.