Multifactorial Prediction of the Risk of Hospital Mortality in Patients With Acute Coronary Syndrome

Aim      To identify predictors for the risk of in-hospital death and to develop a prognostic scale for individual risk of death in patients with acute coronary syndrome (ACS) at the hospital stage of treatment.Material and methods  A sequential retrospective analysis was conducted, including 225 patients with ACS (n=101, main group of patients who died in hospital; n=124, control group) hospitalized in the Regional Vascular Center #2 of the Semashko Nizhny Novgorod Regional Clinical Hospital from January, 2021 through July, 2022. Clinical, demographic, laboratory and instrumental characteristics of patients were studied. Statistical analysis was performed using the Statistica version 10.0 and MedCalc version 20.0 software. The cutoff threshold for quantitative variables was determined by ROC analysis. Potential outcome predictors were identified by a univariate logical regression analysis followed by the construction of a multivariate model for predicting in-hospital mortality using the stepwise analysis with backward inclusion. The prognostic degree of a predictor was expressed as an odds ratio (OR) with a 95% confidence interval (CI). Differences were considered statistically significant at p<0.05. For each variable of the multivariate regression model, an individual score was calculated using a linear transformation of the beta coefficients of each variable.Results Seven independent predictors of hospital death were identified in patients with ACS: Killip class II or higher acute heart failure (AHF) (OR 5.96; 95% CI 1.82-19.48; p=0.0031), low hemoglobin ≤127 g/l (OR 3.75; 95% CI 1.39-10.07; p=0.0087), elevated blood glucose on admission ≥9.7 mmol/l (OR 4.86; 95% CI 1.55-15.21; p=0.0065), high body mass index (BMI) ≥32 kg/m2 (OR 7.18; 95% CI 2.65-19.42; p=0.0001), high pulmonary artery systolic pressure (PASP) ≥38 mmHg (OR 3.95; 95% CI 1.48-10.51; p=0.0059), reduced left ventricular ejection fraction (LVEF) according to Simpson (%) ≤42% (OR 5.80; 95% CI 2.15-15.68; p=0.0005), reduced glomerular filtration rate (GFR) according to CKD-EPI ≤55 ml/min (OR 5.75; 95% CI 2.16-15.28; p=0.0005). An individual score was calculated for each predictor. The total score of all predictors formed a scale that was ranged from score 0 to 43 with a cutoff threshold of 14, where a result >14 indicated a high probability of in-hospital death. This scale has a high prognostic potential with the sensitivity 93.07%, specificity 86.29%, and the area under the curve (AUC) 0.957.Conclusion      Based on the obtained multifactorial model that included 7 major predictors, a scale (scoring system) was developed for predicting the risk of death for ACS patients at the hospital stage of treatment. This scale allows fast identification of patients with a high risk of in-hospital death with a high prognostic accuracy in real clinical practice.