Association between blood chloride levels and 30-day mortality in patients admitted to the intensive care unit after coronary artery bypass grafting: analysis of data from MIMIC-IV database

Abstract

Abstract Background: Patients admitted to the intensive care unit (ICU) following coronary artery bypass grafting (CABG) often experience abnormal levels of blood chloride. This study primarily aims to evaluate the association between blood-Cl–-levels and the likelihood of 30-day mortality in this cohort. The authors hypothesized that abnormal blood-Cl–-levels would increase 30-day mortality in these patients. Methods: The dataset for this research consists of patients who were over the age of 18 and underwent CABG procedures at two renowned establishments in the United States. These institutions are Beth Israel Deaconess Medical Center/Harvard Medical School, located in Boston, and Massachusetts Institute of Technology, situated in Cambridge. The data used in this retrospective cohort study spanned from 2008 to 2019 and were derived from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Based on the tertiles of blood-Cl–-levels, the participants were divided into Q1, Q2, and Q3 groups. To investigate the association between blood-Cl–-levels and 30-day mortality, several statistical models were used, including a generalized additive model (GAM), restricted cubic spines (RCS), and a two-piecewise linear regression model. The 30-day mortality rates of the three groups were depicted using Kaplan–Meier (K–M) curves. Additionally, we employed multivariable logistic regression models to account for any potential known confounding factors. Results: Finally, in total, 5224 patients (mean age, 68.8 years; 77% male; 65.9% > 65 years of age) entered the analysis. The rates of 30-day mortality in Q1, Q2, and Q3 groups were 2.8%, 0.9%, and 1.0%, respectively (p < 0.001). After adjusting for all latent known covariates, patients in the lower tertile exhibited an increased 30-day mortality risk (adjusted odds ratio [OR] 2.75 [95% confidence interval (CI) 1.32–5.73]; p = 0.017). The study concluded a nonlinear association between blood-Cl–-level and 30-day mortality, with a threshold at approximately 107.0 mmol/L. The effect sizes and their corresponding confidence intervals (CIs) below and above the threshold were as follows: 0.77 (0.67–0.87) and 1.15 (0.83–1.6), respectively. Conclusion: A nonlinear correlation was observed between the blood-Cl–-levels and the mortality rate within 30 days among patients admitted to the ICU following CABG procedures. Elevated and reduced blood-Cl–-levels were linked to higher rates of 30-day mortality, particularly when Cl– fell below the threshold of 107.0 mmol/L.