Login / Signup

A risk score model to predict in-hospital mortality of patients with end-stage renal disease and acute myocardial infarction.

Yuan FuHao SunZongsheng GuoLi XuXinchun YangLefeng WangKuibao LiMulei ChenYuanfeng Gao
Published in: Internal and emergency medicine (2020)
Chronic kidney disease (CKD) significantly increases the rate of adverse cardiovascular events in patients with coronary artery disease. In this study, we aimed to establish a risk score (RS) model to predict in-hospital mortality risk in patients with end-stage renal disease (ESRD) and acute myocardial infarction (AMI). A total of 113 consecutive patients with ESRD and AMI were retrospectively enrolled between January 1, 2015 and December 31, 2019. All patients received regular hemodialysis and were divided into two groups according to the prognosis during hospitalization. Univariable and multivariable logistic regression analyses were used to identify the risk factors of in-hospital mortality. A RS model was developed based on multiple regression analysis and was internally validated using 1000 bootstrap analysis. The receiver operating characteristic (ROC) curve was performed, and the area under curve (AUC) was analyzed to evaluate the performance of the RS model. AUCs were compared using the Z test. Thirty-three patients died during hospitalization, resulting in in-hospital mortality rate of 29.2%. After multivariate logistic regression, an RS model (0-8) was established based on five independent factors that were assigned with different points according to relative coefficients (coefficient of the index risk factor divided by the lowest coefficient among these five risk factors; rounded to closest integer): 1 for C-reactive protein (CRP) ≥ 14.2 mg/L and left ventricular ejection fraction (LVEF) ≤ V3%; 2 for age ≥ 65 years old, heart rate (HR) at admission ≥ 86 beats per minute (bpm) and D-dimer ≥ 2.4 mg/L FEU. The present RS model had a sensitivity of 85.7%, the specificity of 84%, and an accuracy of 78.1%. In ROC curve analysis, the model demonstrated a good discriminate power in predicting in-hospital mortality (AUC = 0.895, 95% CI 0.814-0.96; P < 0.001), which was significantly better than the predictive power of the Global Registry of Acute Coronary Events risk score (GRACE RS) (AUC = 0.754, 95% CI 0.641-0.868; P < 0.001 after Z test). A novel RS model, which was established to help predict in-hospital mortality of patients with ESRD and AMI, was easy to use and had higher accuracy than the GRACE RS.
Keyphrases