Addition of routinely measured blood biomarkers significantly improves GRACE risk stratification in patients with myocardial infarction

Abstract

Aim: To investigate whether blood biomarkers measured routinely at hospital admission in myocardial infarction (MI) patients can improve the admission GRACE score for the composite endpoint of all-cause mortality and non-fatal MI at 6 months.

Methods: 2055 patients treated for MI in the Northwest clinics, the Netherlands, between 2013 and 2016 were examined. As part of the prevailing MI treatment protocol, 19 biomarkers were measured and the GRACE score was ascertained. Information on the composite endpoint was derived from municipal registries and electronic medical records. We applied elastic net logistic regression (LR) analysis to select biomarkers that had statistically significant additive prognostic value on top of the GRACE score. We then studied the prognostic performance of the LR model containing the GRACE score and the selected biomarkers.

Results: At six months follow-up 143 (6.96%) reached the composite endpoint. Nine variables were included in the final LR model: GRACE score, urea, sodium, potassium, alkaline phosphatase, LDL cholesterol, glucose, hemoglobin and C-reactive protein. This extended GRACE score model showed improved discrimination (C-statistic 0.76 vs 0.70, p = <0.001) and classification (continuous net reclassification index 0.49, p < 0.001) compared with the GRACE score only.

Conclusion: The ability of the GRACE score for detecting MI patients at high risk for mortality or MI within 6 months, was significantly improved by adding several biomarkers measured routinely at admission.