Lack of Superiority for Soluble ST2 over High Sensitive C-Reactive Protein in Predicting High Risk Coronary Artery Calcium Score in a Community Cohort

PURPOSE: Soluble ST2 (sST2) is an emerging prognostic biomarker in patients with cardiovascular disease (CVD). A recent study showed that sST2 predicted incident hypertension. High sensitive C-reactive protein (hsCRP) has been a widely-used biomarker for risk-stratifying in CVD. We compared the abilities of sST2 and hsCRP to predict high risk coronary artery calcium score (CACS). MATERIALS AND METHODS: The CACS was assessed by cardiac computed tomography, and sST2 was measured in 456 subjects enrolled in the Mapo-gu community cohort. In accordance with the 2013 ACC/AHA guidelines, we defined the high risk CACS group as individuals with a CACS ≥300 Agatston units (AU). RESULTS: There were 99 (21.7%) subjects with a CACS ≥300 AU. There was a strong correlation between log sST2 and log hsCRP (r=0.128, p=0.006), and both log sST2 and log hsCRP showed significant associations with CACS (r=0.101, p=0.031 for sST2, r=0.101, p=0.032 for hsCRP). In net reclassification improvement (NRI) analysis, the NRI for hsCRP over sST2 was significant [continuous NRI 0.238, 95% confidence interval (CI) 0.001–0.474, integrated discrimination index (IDI) 0.022, p=0.035], while the NRI for sST2 over hsCRP was not significant (continuous NRI 0.212, 95% CI -0.255–0.453, IDI 0.002, p=0.269). CONCLUSION: sST2 does not improve net reclassification for predicting a high risk CACS. Using hsCRP provides superior discrimination and risk reclassification for coronary atherosclerosis, compared with sST2.