Prognostic value of angiographic and inflammatory parameters in non ST-segment elevation acute coronary syndrome based on real-world data

Background . The relevance of real-world data is promising for prognosis, as it represents the entire spectrum of patients with their individual characteristics, including comorbidities, which are often not included in clinical studies. The application of machine learning methods has the potential to enhance the prognostic value; the availability of a substantial amount of data allows to perform cross-validation and confirm results. Objective . To identify clinically relevant laboratory and angiographic markers that are associated with an unfavourable out-come in patients with non-ST-segment elevation acute coronary syndrome (NSTE-ACS). Design and methods . A total of 2348 medical records of patients diagnosed with acute coronary syndrome were analyzed. Factors evaluated included the disease course, risk factors and comorbidity, as well as laboratory and instrumental investigations. A logistic regression model was developed using a cross-validation approach. Results . The analysis revealed a number of risk factors for unfavourable course of NSTE-ACS, including angiographic factors (such as left main coronary artery stenosis or chronic coronary artery occlusion) and laboratory factors (haemoglobin level, MPV, monocyte count and SII index). A prognostic model was developed to assess the risk of in-hospital mortality, demonstrating optimal accuracy, sensitivity and specificity. Conclusion . The data obtained support the prognostic value of indicators derived from routine clinical examination in prediction of in-hospital mortality in patients with NSTE-ACS.