Editorial referring to the article: Validation of the Grace Risk Score to Predict In-Hospital and 6-Month Post-Discharge Mortality in Patients with Acute Coronary Syndrome

Acute Coronary Syndrome (ACS) is the most common cause of death worldwide.¹ Nevertheless, ACS represents a heterogenous group of diseases, encompassing since low-risk unstable angina (30-day mortality below 1%), until patients with ST-elevation myocardial infarction (STEMI) and cardiogenic shock (30-day mortality around 50%). Multivariable prediction models have been developed to classify short-term and long-term risk of these patients (Table 1). For patients with the diagnosis of ACS, the TIMI risk score and the Global Registry of Acute Coronary Events (GRACE) score have been largely used in clinical practice; the latter, despite being more complex, has shown better performance as a prognostic tool, including prognostic information not only about the acute phase but also about the risk within six months after the cardiac event.²

Neves et al.,³ analyzed the performance of GRACE score in 160 patients admitted for ACS in a single center in Brazil. The results corroborate the good discrimination and calibration of GRACE score for in-hospital mortality in the Brazilian population and added information regarding its performance for six-month mortality.³ Despite the limited number of events and wide confidence intervals, the consistency of good discrimination and calibration of this score in different populations reinforces this model as an appropriate tool to estimate the risk of patients with ACS.^2,3

Once a model has shown good performance in estimating risk, it is important to determine if this information may change the clinical practice. Patients with STEMI usually receive a standard level of care and changes in the approach are made more due to complications (e.g. cardiogenic shock) than to risk scores. However, patients with non-ST elevation ACS are more heterogenous, and risk stratification models have exerted greater influence on decision making (Table 2).^4–7 High-risk patients (GRACE score > 140) represent a group of patients who benefit from an invasive approach in the first 24 hours,^6,7 meanwhile patients at low risk may be considered for treatment outside the intensive care unit and early discharge.⁸ Other risk models have been developed, as the one developed by the National Cardiovascular Data Registry (NCDR®), whose accuracy can be further improved by the continuous use of a very large and diverse database.^8,9 In addition to the individual risk estimation of patients with ACS, these models can also be used to adjust the risk of mortality in quality-improvement registries using the observed/expected ratio, which is of great value for epidemiological analysis.⁹

Finally, the use of artificial intelligence as machine learning and the technique of deep learning may represent the next step in risk prediction of patients with ACS with the potential of integrating this information into the decision-making process of diagnosis and treatment.¹⁰ Until then, we should consider the traditional prediction models as a support in situations where they could provide useful information to the medical team and the patient about the risk of mortality. Finally, the result of a risk score should never be used apart from medical judgment and the combination of both represents the current good clinical practice.