Machine Learning for Early Prediction of Major Adverse Cardiovascular Events After First Percutaneous Coronary Intervention in Patients With Acute Myocardial Infarction: Retrospective Cohort Study

BACKGROUND

The incidence of major adverse cardiovascular events (MACEs) remains high in patients with acute myocardial infarction (AMI) who undergo percutaneous coronary intervention (PCI), and early prediction models to guide their clinical management are lacking.

OBJECTIVE

This study aimed to develop machine learning-based early prediction models for MACEs in patients with newly diagnosed AMI who underwent PCI.

METHODS

A total of 1531 patients with AMI who underwent PCI from January 2018 to December 2019 were enrolled in this consecutive cohort. The data comprised demographic characteristics, clinical investigations, laboratory tests, and disease-related events. Four machine learning models-artificial neural network (ANN), k-nearest neighbors, support vector machine, and random forest-were developed and compared with the logistic regression model. Our primary outcome was the model performance that predicted the MACEs, which was determined by accuracy, area under the receiver operating characteristic curve, and F1-score.

RESULTS

In total, 1362 patients were successfully followed up. With a median follow-up of 25.9 months, the incidence of MACEs was 18.5% (252/1362). The area under the receiver operating characteristic curve of the ANN, random forest, k-nearest neighbors, support vector machine, and logistic regression models were 80.49%, 72.67%, 79.80%, 77.20%, and 71.77%, respectively. The top 5 predictors in the ANN model were left ventricular ejection fraction, the number of implanted stents, age, diabetes, and the number of vessels with coronary artery disease.

CONCLUSIONS

The ANN model showed good MACE prediction after PCI for patients with AMI. The use of machine learning-based prediction models may improve patient management and outcomes in clinical practice.

Feasibility of off-pump coronary artery grafting for patients with impaired left ventricular ejection fraction: A retrospective cohort study from a single institutional database

OBJECTIVE

Despite substantial advances in surgical practice, the management of patients with impaired left ventricular ejection fraction (LVEF) remains challenging. Furthermore, evidence on the outcomes of off-pump coronary artery bypass (OPCAB) surgery in this population is inconsistent. We conducted the present study to compare the short- and long-term outcomes of OPCAB in patients with different ejection fractions.

METHODS

This retrospective cohort study used data from the Hua-Shan Cardiac Surgery and included consecutive patients aged ≥ 18 years who underwent OPCAB procedures during 2016-2019. The patients included in the study were followed up until death or the end of data collection. Patients with different ejection fractions were matched 1:2 using propensity score matching. Factors associated with short-term outcomes were determined using logistic regression, and Kaplan-Meier survival analyses for the differences in all-cause death were generated.

RESULTS

The two propensity score matched groups consisted of 40 left ventricular dysfunction (LVD) and 80 normal left ventricular function (NLVF) patients. No significant intergroup differences were observed in the postoperative outcomes for the occurrence of left heart failure (22.5% in LVD vs. 5.0% in NLVF, p = .009). Age (odds ratio = 1.11, 95% confidence interval = 1.04-1.18) but not the preoperative LVEF was shown to be a strong predictor of short-term events in logistic regression analyses. Kaplan-Meier curves displayed similar freedom from all-cause death (p = .119) or cardio-death (p = .092) between groups.

CONCLUSION

The immediate postoperative outcomes and long-term outcomes were similar between the groups, indicating that OPCAB is a safe and effective choice for patients with LVD.

Prognostic role of carotid intima-media thickness in off-pump coronary artery bypass surgery

Carotid intima-media thickness (IMT) is a well-known predictor of adverse outcomes in the ischemic heart disease patients; however, evidence is lacking in patients undergoing off-pump coronary artery bypass surgery (OPCAB). Data from 407 patients who underwent OPCAB between April 2013 and August 2016 were retrospectively reviewed. A composite of cardiovascular morbidity endpoints was defined as the presence of stroke, acute myocardial infarction, new cardiac arrhythmia (newly developed atrial fibrillation, atrial flutter, or atrioventricular block), cardiovascular death, or cerebrovascular death within 30 days after surgery. Increased carotid IMT was defined as ≥0.9 mm on one or both sides. The incidence of a composite of cardiovascular morbidity endpoints was 24.0% in the normal IMT group (n = 221) and 34.4% in the increased IMT group (n = 186) (p = 0.021). Multivariable analysis revealed increased IMT (odds ratio 1.719, 95% confidence interval 1.108 to 2.666, p = 0.016) and preoperative renal replacement therapy (odds ratio 4.264, 95% confidence interval 1.679 to 10.829, p = 0.002) as independent predictors of a composite of cardiovascular morbidity endpoints. In patients undergoing OPCAB, preoperative assessment of carotid IMT may help predicting the development of a postoperative composite of cardiovascular morbidity endpoints.

Circumferential strain of carotid arteries does not differ between patients with advanced coronary artery disease and group without coronary stenoses

PURPOSE

Speckle tracking echocardiography is widely used for the analysis of myocardial function. Recently, circumferential strain (CS) of carotid arteries was postulated as novel indicator of vascular function. Our aim was to characterize and compare CS of carotid arteries in patients with advanced coronary artery disease and controls without significant coronary stenoses.

PATIENTS/METHODS

We compared CS of both common carotid arteries (CCA) in the 25 patients with three-vessel coronary artery disease (3VD) (mean age 69±9 years, 9 male) and in 16 age-matched subjects without significant coronary lesions (C) (69±8 years, 7 male). Additionally in 11 patients we estimated pulse wave velocity (PWV) and assessed the correlation between PWV and CS. Short-axis images of arteries were acquired for strain analysis with linear probe of echocardiograph. The assessment of CS was performed off-line by two observers.

RESULTS

The intraobserver variability for the CS (coefficient of variation) were 4.9 and 5.4% for left and right CCA and interobserver variability were 11.7% and 12.5%, respectively. The mean CS for left and right CCA did not differ between compared groups. We did not find correlation between CS strain and PWV. The only difference was related to the more prevalent plaque presence and thicker intima-media complex (IMT) in 3VD (p=0.0039 for IMT of left CCA and p=0.016 for IMT of right CCA).

CONCLUSIONS

The global CS of CCA, contrary to IMT, did not allow for differentiation between 3VD and C subjects. Despite good feasibility and concordance of CS measurements its clinical significance remains to be established.