This study tested whether the soluble (s)ST2 is a superb biomarker predictive of moderate to severe cerebral–cardiac syndrome (CCS) (defined as coexisting National Institute of Health Stroke Scale (NIHSS) >8 and left-ventricular ejection fraction (LVEF) <60%) in patients after acute ischemic stroke (IS). Between November 2015 and October 2017, a total of 99 IS patients were prospectively enrolled and categorized into three groups based on NIHSS, i.e., group 1 (NIHSS ≤ 8, n = 66), group 2 (NIHSS = 9-15, n = 14) and group 3 (NIHSS ≥ 16, n = 19), respectively. Blood samples were collected immediately after hospitalization, followed by transthoracic echocardiographic examination. The results showed that the flow cytometric analysis for assessment of inflammatory biomarkers of TLR2+/CD14+cells, TLR4+/CD14+cells, Ly6g+/CD14+cells, and MPO+/CD14+cells, and ELISA assessment for circulatory level of sST2 were significantly higher in groups 2/3 than in group 1 (all p < 0.01). However, these parameters did not show significant differences between groups 2 and 3 (all p > 0.05). The LVEF was significantly lower in group 3 than in group 1 (p < 0.001), but it displayed no difference between groups 1/2 or between groups 2/3. These inflammatory biomarkers ((TLR2+/CD14+cells// TLR4+/CD14+cells// MPO+/CD14+cells) and sST2)) were significantly positively correlated to NIHSS and strongly negatively correlated to LVEF (all p < 0.05). Multivariate analysis demonstrated that both MPO/CD14+cells >20% (p = 0.027) and sST2 ≥ 17,600 (p = 0.004) were significantly and independently predictive of moderate-severe CCS after acute IS. Receiver operating characteristic curve analysis demonstrated that sST2 was the most powerful predictor of CCS with a sensitivity of 0.929 and a specificity of 0.731 (p < 0.001). In conclusion, sST2 is a useful biomarker for prediction of CCS severity in patients after acute IS.