Multi-marker network in ST-elevation myocardial infarction (STEMI) patients undergoing primary percutaneous coronary intervention : when and what to measure

Background: A multi-marker strategy including biomarkers contributing independent information is expected to improve the prognostic evaluation of STEMI patients. We sought to explore the relationship between cardiac troponin I (cTnI), Creactive protein (CRP), B-type natriuretic peptide (BNP) and chromogranin A (CgA), accounting for biomarker profiles detected within 48 h from successful primary percutaneous coronary intervention (PPCI). Methods: We evaluated cTnI, CRP, BNP and CgA profiles based on measurements performed before PPCI and 6, 24 and 48 h later in 73 STEMI patients. DIRECT STATIS and DUAL STATIS based on Principal Components Analysis (PCA) were employed to assess similarities between average profiles of patients and the relationship between markers measured at different times of sampling. Results: In evaluating ‘similarities between patients’, measurements at 24 h and 48 h contributed most variability to subject profiles (score =0.942 and 0.915, respectively) as highly associated with the first principal component (accounting for 77.4% of explained variance). Patients’ profiles at these times resulted highly correlated (correlation coefficient, 0.91). In evaluating ‘similarities between patterns of markers’, the sampling time mainly contributing to average biomarker profiles was 24 h. Concerning different biomarkers, BNP and cTnI were highly correlated and mainly explained the first component (accounting for 40.1% of explained variance). The second component (accounting for 26.3% of variance) was mainly explained by CgA, contributing independent information to previous markers, and partially by CRP. CRP as inflammatory marker partially overlapped BNP and cTnI and seemed to have correlated but opposite effects with respect to CgA. Conclusions: The sampling time contributing most information from biomarker measurements to the definition of patients’ profile was 24 h after PPCI. BNP and cTnI resulted interchangeable in a multi-marker panel for reporting about the extent of necrosis, whereas independent information derived from CgA. This marker may reflect a pathophysiologic mechanism opposite to inflammatory response and orthogonal to the one explained by BNP, which resulted the most informative marker.