Parameters of complete blood count and coronary lesions in predicting inhospital mortality in patients with acute ST-segment elevation myocardial infarction after percutaneous coronary intervention

Abstract

Aim. To evaluate the predictive potential of complete blood count (CBC) and coronary involvement parameters for predicting inhospital mortality in patients with ST-segment elevation myocardial infarction (STEMI) after percutaneous coronary intervention (PCI).Material and methods. This single-center cohort retrospective study was conducted using data from 4677 electronic medical records of patients with STEMI (3203 men and 1474 women). Two groups of people were identified: group 1 — 318 (6,8%) patients who died in hospital; group 2 — 4359 (93,2%) patients with a favorable outcome. The degree of influence of predictors on the end point was determined by the Shapley method.Results. The best quality metrics had a model with structure included, in addition to 7 CBC factors, indicators of three-vessel coronary involvement, left coronary artery remodeling and coronary perfusion after PCI with the TIMI score <2 (AUC — 0,845, sensitivity — 0,78, specificity — 0,786) . The greatest contribution to the implementation of the end point was associated with the level of hemoglobin, white blood cells, immune-inflammatory index and neutrophils >75,4%. Coronary artery parameters had a less noticeable effect on inhospital mortality, and the minimal effect was plateletcrit >0,22%, lymphocytes <13,3% and erythrocyte sedimentation rate.Conclusion. This multivariate logistic regression model based on a combination of CBC indicators and indicators characterizing the coronary system has high predictive accuracy, which determines the prospects for its further use as one of the available risk assessment tools in clinical practice.

Keywords

• myocardial infarction
• percutaneous coronary intervention
• mortality, data mining
• predictor selection
• prediction
• clinical blood analysis.