Predictors of mortality in patients with non-anterior ST-segment elevation myocardial infarction: Analysis from the HORIZONS-AMI trial

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.

The prognostic significance of single-lead ST-segment resolution in ST-segment elevation myocardial infarction patients treated with primary PCI - A substudy of the randomized TOTAL trial

BACKGROUND

ST-segment elevation myocardial infarction (STEMI) is associated with high morbidity and mortality worldwide. Simple electrocardiogram (ECG) tools, including ST-segment resolution (STR) have been developed to identify high-risk STEMI patients after primary percutaneous coronary intervention (PCI).

SUBJECTS AND METHODS

We evaluated the prognostic impact of STR in the ECG lead with maximal baseline ST-segment elevation (STE) 30-60 minutes after primary PCI in 7,654 STEMI patients included in the TOTAL trial. Incomplete or no STR was defined as < 70% STR and complete STR as ≥ 70% STR. The primary outcome was the composite of cardiovascular death, recurrent myocardial infarction (MI), cardiogenic shock, or new or worsening New York Heart Association (NYHA) class IV heart failure at 1-year follow-up.

RESULTS

Of 7,654 patients, 42.9% had incomplete or no STR and 57.1% had complete STR. The primary outcome occurred in 341 patients (10.4%) in the incomplete or no STR group and in 234 patients (5.4%) in the complete STR group. In Cox regression analysis, adjusted hazard ratio for STR < 70% to predict the primary outcome was 1.56 (95% confidence interval 1.32-1.89; P < .001) (model adjusted for all baseline comorbidities, clinical status during hospitalization, angiographic findings, and procedural techniques).

CONCLUSION

In a large international study of STEMI patients, STR < 70% 30-60 minutes post primary PCI in the ECG lead with the greatest STE at admission was associated with an increased rate of the composite of cardiovascular death, recurrent MI, cardiogenic shock, or new or worsening NYHA class IV heart failure at 1-year follow-up. Clinicians should pay attention to this simple ECG finding.

Development and Validation of a Predictive Model for Chronic Kidney Disease After Percutaneous Coronary Intervention in Chinese

AIM

There is no model for predicting the outcomes for coronary heart disease (CHD) patients with chronic kidney disease (CKD) after percutaneous coronary intervention (PCI). To develop and validate a model to predict major adverse cardiovascular events (MACEs) in patients with comorbid CKD and CHD undergoing PCI.

METHODS

We enrolled 1714 consecutive CKD patients who underwent PCI from January 1, 2008 to December 31, 2017. In the development cohort, we used least absolute shrinkage and selection operator regression for data dimension reduction and feature selection. We used multivariable logistic regression analysis to develop the prediction model. Finally, we used an independent cohort to validate the model. The performance of the prediction model was evaluated with respect to discrimination, calibration, and clinical usefulness.

RESULTS

The predictors included a positive family history of CHD, history of revascularization, ST segment changes, anemia, hyponatremia, transradial intervention, the number of diseased vessels, dose of contrast media >200 ml, and coronary collateral circulation. In the validation cohort, the model showed good discrimination (area under the receiver operating characteristic curve, 0.612; 95% confidence interval: 0.560, 0.664) and good calibration (Hosmer-Lemeshow test, P  =  0.444). Decision curve analysis demonstrated that the model was clinically useful.

CONCLUSIONS

We created a nomogram that predicts MACEs after PCI in CHD patients with CKD and may help improve the screening and treatment outcomes.

ST-segment elevation versus non-ST-segment elevation myocardial infarction in current smokers after newer-generation drug-eluting stent implantation

We compared the 2-year major clinical outcomes between ST-segment elevation myocardial infarction (STEMI) and non-STEMI (NSTEMI) in patients who are current smokers who underwent successful percutaneous coronary intervention (PCI) with newer-generation drug-eluting stents (DESs). The availability of data in this regard is limited.A total of 8357 AMI patients were included and divided into 2 groups: the STEMI group (n = 5124) and NSTEMI group (n = 3233). The primary endpoint was the occurrence of major adverse cardiac events (MACE), defined as all-cause death, recurrent myocardial infarction (re-MI), or coronary repeat revascularization. The secondary endpoints were the cumulative incidences of the individual components of MACE and stent thrombosis (definite or probable).After propensity score-matched (PSM) analysis, 2 PSM groups (2250 pairs, C-statistics = 0.795) were generated. In the PSM patients, both for 1 month and at 2 years, the cumulative incidence of MACE (P = .183 and P = .655, respectively), all-cause death, cardiac death, re-MI, all-cause death or MI, any repeat revascularization, and stent thrombosis (P = .998 and P = .341, respectively) was not significantly different between the STEMI and NSTEMI groups. In addition, these results were confirmed using multivariate analysis.In the era of contemporary newer-generation DESs, both during 1 month and at 2 years after index PCI, the major clinical outcomes were not significantly different between the STEMI and NSTEMI groups confined to the patients who are current smokers. However, further research is needed to confirm these results.

Comparison of myocardial tissue reperfusion of inferior wall and a right ventricle among patients after primary angioplasty for an inferior myocardial infarction with right ventricular infarction

BACKGROUND

Impaired myocardial tissue reperfusion affects prognosis of patients with ST-segment elevation myocardial infarction (STEMI) treated with percutaneous coronary intervention (PCI) and can be identified by ST-segment analysis. To date, evaluation of the myocardial tissue reperfusion of the right ventricle (RV) among the patients treated with PCI for inferior STEMI with right ventricular infarction (RVI) has not been made yet.

METHODS

Patients with inferior STEMI were screened for RVI. Tissue reperfusion was evaluated by maximal residual ST-segment deviation post PCI, independently for the RV and for inferior wall. Myocardial injury was assessed by the peak creatine kinase-mb (CK-MB) value.

RESULTS

Among 456 patients with inferior STEMI, concomitant RVI occurred in 153 (33.5%) subjects (59.86±10.35 years old, 71.9% females). Tissue reperfusion of LV was present in 75 (49%), whereas 55 (35.9%) had both successful LV and RV reperfusion. Among 97 (63.4%) with successful tissue reperfusion of RV, 55 (56.7%) had associated successful tissue reperfusion of inferior wall. Adequate LV reperfusion was accompanied by RV in over 73.3% of patients (P=0.006). Mean peak CK-MB was lower in the group with adequate versus impaired RV tissue-perfusion (197±143 vs. 305±199 U/L, P=0.021 respectively).

CONCLUSIONS

Impaired reperfusion of RV is observed in more than one third of inferior STEMIs with RVI and is not strictly associated with impaired reperfusion of inferior wall and clinical or angiographic variables, therefore ST-segment analysis for RV is mandatory.

Detection of H-FABPA by novel SERS combined with magnetic reaction

The purpose of this paper is to establish a method for easy operation, high sensitivity, strong anti-interference ability, and rapid quantitative detection of cardiac fatty acid-binding protein in acute myocardial infarction biomarkers, so that it can be quickly diagnosed at an early stage and provide a basis for further treatment. Based on the SERS principle, the traditional sandwich system generated by the reaction was captured by the streptavidin (SA) magnetic beads through the specific reaction of SA and biotin then enriched by the applied magnetic field. The enriched magnetic beads are subjected to Raman detection to achieve a process of quantitative detection of the antigen. The minimum detection limit of this study was 1.4490 ng ml^-1, the recoveries were 97.36%-98.35%, and the relative standard deviations between batches and batches were less than 15%. There was no crossover between cTnI, D-dimmer and NT-proBNP. In addition to hemoglobin, the common interfering substances in serum and common anticoagulants do not interfere with the test results. Surface-enhanced Raman spectroscopy can quickly and accurately quantify the acute myocardial infarction marker H-FABP, which is easy to operate and strong in anti-interference ability.

MIRKB: a myocardial infarction risk knowledge base

Myocardial infarction (MI) is a common cardiovascular disease and a leading cause of death worldwide. The etiology of MI is complicated and not completely understood. Many risk factors are reported important for the development of MI, including lifestyle factors, environmental factors, psychosocial factors, genetic factors, etc. Identifying individuals with an increased risk of MI is urgent and a major challenge for improving prevention. The MI risk knowledge base (MIRKB) is developed for facilitating MI research and prevention. The goal of MIRKB is to collect risk factors and models related to MI to increase the efficiency of systems biological level understanding of the disease. MIRKB contains 8436 entries collected from 4366 articles in PubMed before 5 July 2019 with 7902 entries for 1847 single factors, 195 entries for 157 combined factors and 339 entries for 174 risk models. The single factors are classified into the following five categories based on their characteristics: molecular factor (2356 entries, 649 factors), imaging (821 entries, 252 factors), physiological factor (1566 entries, 219 factors), clinical factor (2523 entries, 561 factors), environmental factor (46 entries, 26 factors), lifestyle factor (306 entries, 65 factors) and psychosocial factor (284 entries, 75 factors). MIRKB will be helpful to the future systems level unraveling of the complex mechanism of MI genesis and progression.