Analyzing genetic information can help prevent complications following a heart attack.

Research published in the journal Frontiers in Medicine. Ischemic heart disease (IHD) is a condition where the heart does not receive enough blood and oxygen due to the narrowing or blockage of coronary arteries. This is typically triggered by plaques that form on the walls of blood vessels from fats and cholesterol. IHD can present as angina (chest pain), myocardial infarction (heart attack), or other complications.

According to WHO data, ischemic heart disease is the most common cause of death worldwide, accounting for 13 percent of all deaths. Therefore, it is crucial to prescribe treatment appropriately and reduce the risks of complications and relapses. Researchers from HSE University developed a model capable of predicting the likelihood of complications following a myocardial infarction.

The scientists analyzed data from patients at the Surgut District Center for Diagnostics and Cardiovascular Surgery who were admitted with myocardial infarction from 2015 to 2024. Upon arrival at the emergency department, the research physicians explained the study's provisions to the patients and obtained their consent to participate. Cardiologists then assessed the condition of the coronary arteries supplying the heart and, based on this evaluation, performed procedures to restore blood flow: balloon angioplasty and stenting or coronary artery bypass grafting. Patients were treated pharmacologically with RAAS blockers, beta-blockers, statins, and dual antiplatelet therapy. Data were recorded in inpatient medical histories. For each patient, physicians determined standard clinical indicators: blood pressure, body mass index, cholesterol, and glucose levels.

At the laboratory stage, research physicians isolated DNA from leukocyte rings in the collected blood samples and then froze it at –80 °C for future genetic testing. The genotype was determined based on a specific genetic variation (polymorphism) in the VEGFR-2 gene. The genetic marker VEGFR-2 is a component in the body's signaling system that regulates the growth of new blood vessels. There are three variants of the genotype—C/C, C/T, and T/T—distinguished by variations of the DNA nucleotides cytosine (C) or thymine (T) in this gene segment. This marker has long been known; however, its influence on the prognosis of complications after myocardial infarction was studied for the first time.

The authors of the article examined the impact of 39 factors on the prognosis of risks for cardiac death, recurrent acute coronary syndrome, stroke, and the need for repeated revascularization—a procedure that helps restore blood flow in arteries. To select an effective model, the researchers trained and tested several machine learning algorithms: gradient boosting (CatBoost and LightGBM), random forest, logistic regression, and AutoML approaches.

The CatBoost model demonstrated the best performance—an optimized gradient boosting algorithm designed for categorical data rather than numerical values. It makes predictions by sequentially creating and training "weak" decision trees, where each subsequent tree corrects the errors of the previous ones. When building trees, the algorithm splits the data into two parts: the model is trained on one part, while errors are calculated on the other. This reduces the risk of overfitting, where the model simply memorizes correct answers and helps identify common patterns for predictions in unfamiliar cases.

The influence of features on the model's accuracy was evaluated using the sequential feature addition method, which checks their contribution at each stage. The scientists selected the 9 most significant factors: gender, body mass index, Charlson comorbidity index accounting for serious comorbid conditions, condition of the left ventricular lateral wall, degree of left main coronary artery lesion, number of affected arteries, variant of the VEGFR-2 gene, choice of percutaneous coronary intervention or coronary artery bypass grafting, and statin dosage.

The results showed that the dosage of statins—medications used to lower blood cholesterol levels—is the most important factor influencing the risk of complications. High doses of statins reduce this risk, especially in patients with an unfavorable genotype. The VEGFR-2 polymorphism, particularly the presence of the T allele, ranked fourth in importance.

“Previously, genetic factors were not used in machine learning models, mainly because sequencing or even genotyping of individual nucleotides is not performed in hospitals. However, we had access to data on the polymorphism in the VEGFR-2 gene in addition to standard indicators. This allowed us to compare this measure with others and determine that the risk allele of the VEGFR-2 variant is among the top five most important factors for predicting long-term outcomes in patients with myocardial infarction,” explains one of the article's authors, head of the International Laboratory of Bioinformatics at HSE University, Maria Poptsova.

The researchers emphasize that analyzing genetic data aids in creating more accurate and personalized models for predicting the risks of cardiovascular complications in patients after a myocardial infarction.

“Cardiovascular diseases require resources for diagnosis, treatment, rehabilitation, and prevention and therefore place a significant burden on the healthcare system. Implementing such models in clinical practice will help reduce mortality and the frequency of recurrent heart attacks, optimize treatment, and lessen the burden on physicians,” comments one of the article's authors, research intern at the International Laboratory of Bioinformatics, Alexander Kirdeev.

The study was conducted as part of the HSE University project "Mirror Laboratories".