High-Frequency Biomarker Measurements of Troponin, NT-proBNP, and C-Reactive Protein for Prediction of New Coronary Events After Acute Coronary Syndrome.

Circulation. 2019;139:134–136. DOI: 10.1161/CIRCULATIONAHA.118.036349 134 The BIOMArCS study (Biomarker Study to Identify the Acute Risk of a Coronary Syndrome) was designed to study the relation between temporal changes in cardiovascular biomarkers and ischemic cardiovascular events in patients discharged after acute coronary syndrome admission (The Netherlands Trial Register NTR1698).1 Eight hundred forty-four patients with acute coronary syndrome were enrolled in 18 hospitals in the Netherlands. Venipuncture was scheduled at 19 regular intervals during a year. Forty-five patients (cases) reached the study end point, defined as the first event of the composite of cardiac death (n=8), myocardial infarction (n=24), or unstable angina requiring urgent coronary revascularization (n=8) within 1 year. BIOMArCS was approved by the institutional review committees of the participating hospitals. All patients gave informed consent. We used a case-cohort approach for biomarker determination and analysis.2 The case-cohort study comprises a random subcohort from the full cohort, together with all cases. The main advantage of the case-cohort design over a cohort study is that full covariate data (in our situation, biomarker data) are needed only for the cases and subcohort individuals, not all the original cohort.3 Thus, the advantages of a cohort study are combined with the efficiency of a nested case-control study.3 We randomly selected a subcohort of 150 individuals (18%), including 8 cases. Our case-cohort therefore consisted of (all) 45 cases and 142 noncases. Four established cardiovascular biomarkers were then measured (in 1478 blood samples), reflecting different components of cardiovascular pathophysiology: troponin, which was assessed with high-sensitivity cardiac troponin I and T assays (hs-cTnI [Abbott] and hs-cTnT [Roche]); NT-proBNP (N-terminal pro-B-type natriuretic peptide, validated in-house sandwich ELISA); and high-sensitivity C-reactive protein (hs-CRP [Beckman Coulter]).1 Biomarker measurements were performed in a single batch; personnel were blinded to any patient data. Patient-specific longitudinal biomarker trajectories were analyzed by linear mixed-effect models, with adjustment for GRACE (Global Registry of Acute Coronary Events) risk score (including age), sex, clinical risk factors (recorded at inclusion), and creatinine value (recorded at each sampling time point). The relationships between biomarker levels (based on the linear mixed-effect models) and the end point were analyzed by Cox proportional hazard models. Unadjusted hazard ratio estimates for each biomarker were obtained, as well as estimates adjusted for GRACE risk score and multiple biomarkers. We applied bayesian semiparametric joint modeling, enabling simultaneous estimation of the linear mixed-effect and Cox model parameters.4 Median age was 62.5 years; 77.9% were male; and 51.7% presented with STsegment elevation. Measured biomarkers were elevated during the index acute coronary syndrome but subsequently decreased and stabilized within 30 days. Canadian Cardiac Society angina class was ≤1 at 95.5% of the post 30-day visits, Rohit M. Oemrawsingh, MD, MSc et al