BACKGROUND Although cardiac troponin is recommended as the biomarker of choice to evaluate myocardial injury, inappropriate low-value ordering practice is frequent, particularly routine ordering of creatine kinase–myocardial band tests where troponin is available. The aim of this study was to evaluate the impact of an educational intervention for rational request of cardiac biomarkers in the intensive care unit. METHODS We conducted a quasi-experimental, pre-post implementation study of an educational program (expository-dialogue presentation and disclosure of a decision algorithm) for rational cardiac biomarker testing in adult critically ill patients. The study was divided into two twelve-month periods: pre-intervention (September 2017 to August 2018) and post-intervention (October 2018 to September 2019). An interrupted time series with a segmented regression model was applied to analyze variation over time in creatine kinase–myocardial band and troponin testing. RESULTS We included 4429 patients: 2181 patients in the pre-intervention period and 2248 patients in the post-intervention period. A reduction in the concomitance of creatine kinase–myocardial band and troponin testing was observed (concomitance in 1415 tests in the pre-intervention period vs 348 tests in the post-intervention period). The interrupted time series analysis demonstrated a noticeable immediate reduction in the concomitance of creatine kinase–myocardial band with troponin after the intervention (-0.13 tests per patient, p=0.0016), but not in the secular trend for the concomitance. The proportion of patients with acute coronary syndrome as a discharge diagnosis was not different between the pre and pos-intervention period. CONCLUSIONS Our pre-post interventional study demonstrated a significant decrease on the concomitance of creatine kinase–myocardial band and troponin tests. A rational high-value ordering practice of cardiac biomarkers is possible in critically ill patients and might be suitable for educational interventions. Trial registration : Brazilian Registry of Clinical Trials – ReBEC – number RBR-5kk5j2 before data analysis (October 14th, 2019).