GRACE score - is it good for all NSTEMI patients

Type of funding sources: None.

Patients (P) with a non-ST elevation myocardial infarction (NSTEMI) have a heterogenous prognosis and early risk stratification at admission is essential. GRACE score is validated to estimate in-hospital outcomes, as well as long term prognosis.

Evaluation of discriminatory capacity of GRACE score in P <65 years versus ≥65 presenting with NSTEMI.

Based on a single-center retrospective study, data collected from admissions between 1/01/2016 and 11/12/2019. Patients with cardiac arrest or hemodynamically unstable patients at admission were excluded. GRACE score was calculated using Killip-Kimball classification as a surrogate of heart failure. P were divided in 2 groups (G): G1 if <65 years old, G2 if ≥65 years. We evaluated the discriminatory capacity of GRACE score in predicting in-hospital MACE in both groups through ROC-curve analysis.

We identified 405 patients with NSTEMI, 62.7% were males with an average age of 68.4±12.3 years old. Mean GRACE in global population was 118±28.3 points, lower in G1 P (G1 92.6±22.6, G2 130±21.2, p<0.001). At univariate analysis, GRACE score was predictor of MACE in G1 (p=0.05, OR 1.05) and G2 p=0.026, OR 1.03). Its discriminatory capacity was acceptable: for G1 the area under the curve (AUC) was 0.687 and in G2 AUC 0.785. Given the lower discriminatory capacity in G2, the authors determined other predictor factors (excluding GRACE) to accurate risk stratification in this G. At univariate analysis, predictor factors for MACE were diabetes (p=0.006, OR 5.4), left ventricle branch block (p=0.011, OR 2.2) and haemoglobin at admission (p=0.003, OR 0.63). At multivariate analysis, both diabetes (p=0.002) and haemoglobin (p=0.031) were independent predictor factors for MACE.

In our center, GRACE score was a better predictor for younger patients, while for other patients other factors, such as diabetes and haemoglobin at admission must be taken into account to assess risk stratification.