Propensity score analysis of red cell distribution width to serum calcium ratio in acute myocardial infarction as a predictor of in-hospital mortality.

Abstract

Red cell distribution width (RDW) and serum calcium (Ca) levels are predictors of in-hospital mortality in acute myocardial infarction (AMI) patients. However, their sensitivity and specificity are limited. Therefore, this study aimed to determine whether the RDW to Ca ratio (RCR) acquired on admission can be used to predict the in-hospital mortality of AMI patients. This retrospective cohort study extracted clinical information from the Medical Information Market for Intensive IV (MIMIC-IV) database on 2,910 AMI patients enrolled via propensity score matching (PSM). Prognostic values were assessed using a multivariate logistic model and three PSM approaches. Analysis was performed based on stratified variables and interactions among sex, age, ethnicity, anemia, renal disease, percutaneous transluminal coronary intervention (PCI), coronary artery bypass grafting (CABG), atrial fibrillation, congestive heart failure, dementia, diabetes, paraplegia, hypertension, cerebrovascular disease, and Sequential Organ Failure Assessment (SOFA) score. A total of 4,105 ICU-admitted AMI patients were analyzed. The optimal cut-off value of the RCR for in-hospital mortality was 1.685. The PSM was performed to identify 1,455 pairs (2,910) of score-matched patients, with balanced differences exhibited for nearly all variables.The patients' median age was 72 years (range, 63-82 years) and 60.9% were male. The risk of in-hospital mortality incidence increased with increasing RCR levels. After adjusting for confounders, the risk ratio for the incidence of in-hospital mortality for high RCR was 1.75 [95% confidence interval (CI): 1.60-1.94, P = 0.0113] compared to that associated with low RCR in the PSM cohort. High RCR was also substantially implicated in in-hospital mortality incidence in the weighted cohorts [odds ratio (OR) = 1.76, 95% CI: 1.62-1.94, P = 0.0129]. Assessment of RCR in three groups showed that patients with high RCR also had a higher risk of in-hospital mortality (OR = 3.04; 95% CI, 2.22-4.16; P < 0.0001) than in patients with RCR in the adjusted model. In the sensitivity analysis, both the original and weighted groups showed similar results. The RCR at admission may be useful for predicting in-hospital mortality in ICU-admitted AMI patients.