摘要Background:Elevated soluble growth stimulation-expressed gene 2(sST2)protein levels are associated with poor prognosis in pa-tients with myocardial infarction or heart failure.However,few studies have investigated the association between sST2 expression and plaque stability.This study aimed to investigate the expression of sST2 in patients with acute myocardial infarction and its predictive value for vulnerable plaque characteristics in culprit lesions.Methods:From October 2022 to December 2024,230 patients with acute myocardial infarction who underwent coronary angiography and optical coherence tomography(OCT)in(emergency)outpatient and inpatient departments of Jining No.1 People's Hospital were se-lected as subjects of this prospective study.Based on the inclusion and exclusion criteria,203 patients were included in this study.Clinical data were analyzed.Based on the characteristics of criminal plaques detected by OCT,123 cases were divided into a thin-cap fibroatheroma(TCFA)group(60.6％)and a non-thin-cap fibroatheroma(NTCFA)group(80 cases).Serum sST2 levels were measured before surgery.Optical coherence tomography was used to observe the nature of the criminal lesions before interventional therapy.The relationship between serum sST2 levels and criminal plaque vulnerability was analyzed using a multivariable logistic regression model.Results:Serum cardiac troponin,C-reactive protein,and sST2 levels were higher in the TCFA group than those in the NTCFA group.OCT observations showed that patients in the TCFA group had many characteristics of vulnerable plaques,including macrophage aggre-gation,larger lipid radians,and thinner minimum fibrous caps.sST2 independently predicted the presence of TCFA in patients with acute myocardial infarction,with enhanced predictive accuracy when combined with C-reactive protein(area under the curve 0.837 vs.0.782 for sST2 alone,P=0.043).Conclusion:sST2 can independently predict the presence of thin cap atherosclerosis in patients with acute myocardial infarction,and when combined with C-reactive protein,its prediction accuracy is higher.