A novel risk stratification system "Angiographic GRACE Score" for predicting in-hospital mortality of patients with acute myocardial infarction: Data from the K-ACTIVE Registry

Prediction of in-hospital mortality risk for patients with acute ST-elevation myocardial infarction after primary PCI based on predictors selected by GRACE score and two feature selection methods

Introduction

Accurate in-hospital mortality prediction following percutaneous coronary intervention (PCI) is crucial for clinical decision-making. Machine Learning (ML) and Data Mining methods have shown promise in improving medical prognosis accuracy.

Methods

We analyzed a dataset of 4,677 patients from the Regional Vascular Center of Primorsky Regional Clinical Hospital No. 1 in Vladivostok, collected between 2015 and 2021. We utilized Extreme Gradient Boosting, Histogram Gradient Boosting, Light Gradient Boosting, and Stochastic Gradient Boosting for mortality risk prediction after primary PCI in patients with acute ST-elevation myocardial infarction. Model selection was performed using Monte Carlo Cross-validation. Feature selection was enhanced through Recursive Feature Elimination (RFE) and Shapley Additive Explanations (SHAP). We further developed hybrid models using Augmented Grey Wolf Optimizer (AGWO), Bald Eagle Search Optimization (BES), Golden Jackal Optimizer (GJO), and Puma Optimizer (PO), integrating features selected by these methods with the traditional GRACE score.

Results

The hybrid models demonstrated superior prediction accuracy. In scenario (1), utilizing GRACE scale features, the Light Gradient Boosting Machine (LGBM) and Extreme Gradient Boosting (XGB) models optimized with BES achieved Recall values of 0.944 and 0.954, respectively. In scenarios (2) and (3), employing SHAP and RFE-selected features, the LGB models attained Recall values of 0.963 and 0.977, while the XGB models achieved 0.978 and 0.99.

Discussion

The study indicates that ML models, particularly the XGB optimized with BES, can outperform the conventional GRACE score in predicting in-hospital mortality. The hybrid models' enhanced accuracy presents a significant step forward in risk assessment for patients post-PCI, offering a potential alternative to existing clinical tools. These findings underscore the potential of ML in optimizing patient care and outcomes in cardiovascular medicine.

Machine learning prediction of one-year mortality after percutaneous coronary intervention in acute coronary syndrome patients

BACKGROUND

Machine learning (ML) models have the potential to accurately predict outcomes and offer novel insights into inter-variable correlations. In this study, we aimed to design ML models for the prediction of 1-year mortality after percutaneous coronary intervention (PCI) in patients with acute coronary syndrome.

METHODS

This study was performed on 13,682 patients at Tehran Heart Center from 2015 to 2021. Patients were split into 70:30 for testing and training. Four ML models were designed: a traditional Logistic Regression (LR) model, Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Ada Boost models. The importance of features was calculated using the RF feature selector and SHAP based on the XGBoost model. The Area Under the Receiver Operating Characteristic Curve (AUC-ROC) for the prediction on the testing dataset was the main measure of the model's performance.

RESULTS

From a total of 9,073 patients with >1-year follow-up, 340 participants died. Higher age and higher rates of comorbidities were observed in these patients. Body mass index and lipid profile demonstrated a U-shaped correlation with the outcome. Among the models, RF had the best discrimination (AUC 0.866), while the highest sensitivity (80.9%) and specificity (88.3%) were for LR and XGBoost models, respectively. All models had AUCs of >0.8.

CONCLUSION

ML models can predict 1-year mortality after PCI with high performance. A classic LR statistical approach showed comparable results with other ML models. The individual-level assessment of inter-variable correlations provided new insights into the non-linear contribution of risk factors to post-PCI mortality.

Relation of GRACE Risk Score to Coronary Lipid Core Plaques in Patients with Acute Coronary Syndrome

The GRACE risk score is established to predict thrombotic events in patients with acute coronary syndrome (ACS). Although thrombotic events including myocardial infarction after ACS are mainly attributable to vulnerable plaque formation, whether the GRACE score correlates with coronary lipid-rich plaque is unclear. A total of 54 patients with ACS undergoing primary percutaneous coronary intervention under near-infrared spectroscopy intravascular ultrasound (NIRS-IVUS) guidance were included in a prospective manner. Patients were divided into two groups according to the median of the GRACE risk score. Coronary lipid plaques in the target vessel were assessed by NIRS-IVUS with lipid core burden index (LCBI) and a maximum LCBI in 4 mm (maxLCBI4mm). The receiver operating characteristics (ROC) curve analysis was performed based on the major adverse cardiovascular events as an exploratory analysis. The GRACE risk score was significantly and positively correlated with LCBI (r = 0.31, p = 0.03) and maxLCBI4mm (r = 0.38, p = 0.006). LCBI (111.7 ± 85.7 vs. 169.0 ± 83.5, p = 0.02) and maxLCBI4mm (428.5 ± 227.1 vs. 600.6 ± 227.7, p = 0.009) in the target vessel were significantly higher in the high GRACE risk score group than their counterpart. In the ROC curve analysis, LCBI and maxLCBI4mm were predictive for clinical events. In conclusion, the higher GRACE risk score may serve as a discriminator of risk comprising more lipid-rich plaques as an underlying mechanism of an increased risk of thrombotic events after ACS. In patients with ACS, the higher GRACE risk score was significantly and modestly associated with greater coronary lipid plaques in the target vessel.

Prognostic Value of Machine Learning in Patients with Acute Myocardial Infarction

(1) Background: Patients with acute myocardial infarction (AMI) still experience many major adverse cardiovascular events (MACEs), including myocardial infarction, heart failure, kidney failure, coronary events, cerebrovascular events, and death. This retrospective study aims to assess the prognostic value of machine learning (ML) for the prediction of MACEs. (2) Methods: Five-hundred patients diagnosed with AMI and who had undergone successful percutaneous coronary intervention were included in the study. Logistic regression (LR) analysis was used to assess the relevance of MACEs and 24 selected clinical variables. Six ML models were developed with five-fold cross-validation in the training dataset and their ability to predict MACEs was compared to LR with the testing dataset. (3) Results: The MACE rate was calculated as 30.6% after a mean follow-up of 1.42 years. Killip classification (Killip IV vs. I class, odds ratio 4.386, 95% confidence interval 1.943–9.904), drug compliance (irregular vs. regular compliance, 3.06, 1.721–5.438), age (per year, 1.025, 1.006–1.044), and creatinine (1 µmol/L, 1.007, 1.002–1.012) and cholesterol levels (1 mmol/L, 0.708, 0.556–0.903) were independent predictors of MACEs. In the training dataset, the best performing model was the random forest (RDF) model with an area under the curve of (0.749, 0.644–0.853) and accuracy of (0.734, 0.647–0.820). In the testing dataset, the RDF showed the most significant survival difference (log-rank p = 0.017) in distinguishing patients with and without MACEs. (4) Conclusions: The RDF model has been identified as superior to other models for MACE prediction in this study. ML methods can be promising for improving optimal predictor selection and clinical outcomes in patients with AMI.

Value of GRACE and SYNTAX scores for predicting the prognosis of patients with non-ST elevation acute coronary syndrome

BACKGROUND

GRACE and SYNTAX scores are important tools to assess prognosis in non-ST-elevation acute coronary syndrome (NSTE-ACS). However, there have been few studies on their value in patients receiving different types of therapies.

AIM

To explore the value of GRACE and SYNTAX scores in predicting the prognosis of patients with NSTE-ACS receiving different types of therapies.

METHODS

The data of 386 patients with NSTE-ACS were retrospectively analyzed and categorized into different groups. A total of 195 patients who received agents alone comprised the medication group, 156 who received medical therapy combined with stents comprised the stent group, and 35 patients who were given agents and underwent coronary artery bypass grafting (CABG) comprised the CABG group. General information was compared among the three groups. GRACE and SYNTAX scores were calculated. The association between the relationship between GRACE and SYNTAX scores and the occurrence of major adverse cardiovascular events (MACEs) was analyzed. Pearson’s correlation analysis was used to determine the factors influencing prognosis in patients with NSTE-ACS. Univariate and multivariate analyses were conducted to analyze the predictive value of GRACE and SYNTAX scores for predicting prognosis in patients with NSTE-ACS using the Cox proportional-hazards model.

RESULTS

The incidence of MACE increased with the elevation of GRACE and SYNTAX scores (all P < 0.05). The incidence of MACE was 18.5%, 36.5%, and 42.9% in the medication group, stent group, and CABG group, respectively. By comparison, the incidence of MACE was significantly lower in the medication group than in the stent and CABG groups (all P < 0.05). The incidence of MACE was 6.2%, 28.0% and 40.0% in patients with a low GRACE score in the medication group, stent group, and CABG group, respectively (P < 0.05). The incidence of MACE was 31.0%, 30.3% and 42.9% in patients with a medium GRACE score in the medication group, stent group, and CABG group, respectively (P > 0.05). The incidence of MACE was 16.9%, 46.2%, and 43.8% in patients with a high GRACE score in the medication group, stent group, and CABG group, respectively (P < 0.05). The incidence of MACE was 16.2%, 35.4% and 60.0% in patients with a low SYNTAX score in the medication group, stent group, and CABG group, respectively (P < 0.05). The incidence of MACE was 37.5%, 40.9%, and 41.7% in patients with a medium SYNTAX score in the medication group, stent group, and CABG group, respectively (P > 0.05). MACE incidence was 50.0%, 75.0%, and 25.0% in patients with a high SYNTAX score in the medication group, stent group, and CABG group, respectively (P < 0.05). Univariate Cox regression analyses showed that both GRACE score (hazard ratio [HR] = 1.212, 95% confidence interval [CI]: 1.083 to 1.176; P < 0.05) and SYNTAX score (HR = 1.160, 95%CI: 1.104 to 1.192; P < 0.05) were factors influencing MACE (all P < 0.05). Multivariate Cox regression analyses showed that GRACE (HR = 1.091, 95%CI: 1.015 to 1.037; P < 0.05) and SYNTAX scores (HR = 1.031, 95%CI: 1.076 to 1.143; P < 0.05) were independent predictors of MACE (all P < 0.05).

CONCLUSION

GRACE and SYNTAX scores are of great value for evaluating the prognosis of NSTE-ACS patients, and prevention and early intervention strategies should be used in clinical practice targeting different risk scores.

Impact of CADILLAC and GRACE risk scores on short- and long-term clinical outcomes in patients with acute myocardial infarction

BACKGROUND

Recent guidelines recommend risk stratification using objective scoring systems in patients with acute coronary syndrome. In this context, the CADILLAC (Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications) and GRACE (Global Registry of Acute Coronary Events) risk scores were both originally established to predict short-term mortality. However, their impact on short- and long-term clinical outcomes in a contemporary cohort of patients with acute myocardial infarction (MI) is unclear.

METHODS

This bi-center registry included 809 patients with acute MI undergoing primary percutaneous coronary intervention. Patients were divided into three groups according to the pre-defined thresholds and tertiles of the CADILLAC and GRACE scores. The study endpoints included all-cause death and major adverse cardiovascular events (MACE) during the index hospitalization and after discharge.

RESULTS

Of 809 patients, 323 (39.9%) and 255 (31.5%) had high CADILLAC and GRACE risk scores. During the index hospitalization, 61 (7.5%) patients died and 262 (32.4%) had MACE. Both CADILLAC and GRACE risk scores were associated with in-hospital mortality and MACE rates. After discharge, out of 683 patients with available follow-up information who survived to discharge, 42 (6.1%) died and 123 (18.0%) had MACE during the median follow-up period of 632 days. Significantly higher incidence of MACE in higher CADILLAC and GRACE risk scores was observed in a stepwise manner.

CONCLUSION

Both CADILLAC and GRACE risk scores were predictive for short- and long-term mortality and MACE rates in a contemporary cohort of acute MI patients undergoing primary percutaneous coronary intervention.