Multi-biomarker approach for predicting cardiac magnetic resonance parameters at 30 days after ST-segment elevation myocardial infarction

Abstract Background Prior data showed associations of circulating Proprotein Convertase Subtilisin / Kexin type 9 (PCSK9) and inflammatory/anti-fibrotic Cellular Communication Network factor 1 (CCN1) at index ST-segment elevation acute myocardial infarction (STEMI) with left ventricular ejection fraction at 6 and 12 months, respectively and for CCN1 also with infarct size. Purpose PCSK-9, CCN1 and reference biomarkers high-sensitivity Troponin T (hsTNT) and N-terminal pro-brain natriuretic peptide (NTproBNP) were measured in patients from the CLEVER-ACS trial (clinicaltrials.gov NCT01529554) at presentation with STEMI and correlated with clinically relevant cardiac magnetic resonance (CMR) parameters at 30 days. Methods Associations between baseline biomarkers (PCSK-9, CCN1, hsTNT, NTproBNP) and CMR parameters at 30 days were evaluated using Spearman correlation and linear regression analysis (dichotomised at their median). Receiver operating characteristic (ROC) and area under the curve (AUC) were determined for significant values based on correlation analysis; AUC > 0,7 considered relevant. CMR parameters comprised left ventricular structural and functional parameters including scar mass and microvascular obstruction. Results We identified 56 STEMI patients (mean age 61.7 ± 11.2 (SD) years) who completed 30 days follow-up with biomarker data. Among biomarkers, significant associations with CMR parameters were only found for hsTNT, and NTproBNP. The strongest associations were found for left ventricular ejection fraction (LVEF; hsTNT RSp =-0.66; NTproBNP RSp =-0.57), left ventricular scar (LVSCAR; hsTNT RSp =0.58; NTproBNP RSp =0.56) and scar mass (SM; hsTNT RSp =0.65; NTproBNP RSp =0.55). Conversely, left ventricular end-diastolic volume index (LVEDVI; hsTNT RSp =0.47; NTproBNP RSp =0.32), left ventricular end-systolic volume index (LVESVI; hsTNT RSp =0.51; NTproBNP RSp =0.39) and microvascular obstruction (MVO; hsTNT RSp =0.34; NTproBNP RSp =0.34) were significantly, albeit only modestly associated. ROC and AUC analysis yielded relevant associations of biomarkers with CMR parameters LVEF (hsTNT=0.79; NTproBNP=0.81), LVSCAR (hsTNT=0.84; NTproBNP=0.76) and scar mass (hsTNT=0.83; NTproBNP=0.78). Furthermore, relevant associations of biomarkers with CMR were found for LVEDVI (hsTNT=0.79; NTproBNP=0.72), LVESVI (hsTNT=0.84; NTproBNP=0.72) and MVO (hsTNT=0.72; NTproBNP=0.71). Conclusion Baseline levels of hsTNT and NT-proBNP are strong predictors for functional and dimensional CMR parameters of the left ventricle including clinically relevant imaging surrogates of myocardial injury shortly after STEMI.ROC LVEF