Optimized Prognosis Assessment in ST-Segment–Elevation Myocardial Infarction Using a Cardiac Magnetic Resonance Imaging Risk Score

Background— Cardiac magnetic resonance (CMR) demonstrated great potential for the prediction of major adverse cardiac events (MACE) in ST-segment–elevation myocardial infarction. The aim of this study was to develop and validate a CMR-based risk score for ST-segment–elevation myocardial infarction patients. Methods and Results— The scoring model was developed and validated on ST-segment–elevation myocardial infarction cohorts from 2 independent randomized controlled trials (n=738 and n=458 patients, respectively) and included left ventricular (LV) ejection fraction, infarct size, and microvascular obstruction. Primary end point was the 12-month MACE rate consisting of death, reinfarction, and new congestive heart failure. In the derivation cohort, LV ejection fraction ≤47%, infarct size ≥19%LV, and microvascular obstruction ≥1.4%LV were identified as the best cutoff values for MACE prediction. According to the hazard ratios in multivariable regression analysis, the CMR risk score was created by attributing 1 point for LV ejection fraction ≤47%, 1 point for infarct size ≥19%LV, and 2 points for microvascular obstruction ≥1.4%LV. In the validation cohort, the score showed a good prediction of MACE (area under the curve: 0.76). Stratification into a low (0/1 point) and high-risk group (≥2 points) resulted in significantly higher MACE rates in high-risk patients (9.0% versus 2.2%; P =0.001). Inclusion of the CMR score in addition to a model of clinical risk factors led to a significant increase of C statistics from 0.74 to 0.83 ( P =0.037), a net reclassification improvement of 0.18 ( P =0.009), and an integrated discriminative improvement of 0.04 ( P =0.010). Conclusions— Our approach integrates the prognostic information of CMR imaging into a simple risk score that showed incremental prognostic value over clinical risk factors in ST-segment–elevation myocardial infarction patients. Clinical Trial Registration— URL: http://www.clinicaltrials.gov . Unique identifiers: NCT00712101 and NCT02158468.