Cholesterol-to-Lymphocyte Ratio as a Predictor of 1-Year Unplanned Readmission in Patients with Coronary Artery Disease and Type 2 Diabetes Mellitus: A Retrospective Cohort Study

Yumin Wu, Chaozhong Luo, Juan Du, Changjiang Zhang Department of Cardiology, Minda Hospital of Hubei Minzu University, Enshi, Hubei, People’s Republic of ChinaCorrespondence: Changjiang Zhang, Email zcj2008@163.comBackground: Patients with coronary artery disease (CAD) and type 2 diabetes mellitus (T2DM) face a markedly increased risk of unplanned readmission due to metabolic disturbances and chronic inflammation. The cholesterol-to-lymphocyte ratio (CLR), reflecting both dyslipidemia and inflammatory status, may serve as a useful marker for risk stratification. This study aimed to evaluate the predictive value of CLR for 1-year unplanned readmission in CAD patients with T2DM.Methods: We conducted a single-center retrospective cohort study of patients hospitalized from January to December 2023 who were diagnosed with CAD and T2DM by coronary angiography. Demographic characteristics, comorbidities, medication use, and laboratory findings were collected, and CLR was calculated. Least absolute shrinkage and selection operator (LASSO) regression was applied to identify key predictors. Logistic regression, multilayer perceptron (MLP), and extreme gradient boosting (XGBoost) models were developed to assess predictive performance. Model discrimination and calibration were evaluated using AUC, accuracy, sensitivity, specificity, and calibration curves. Variable importance was further explored with SHAP value analysis.Results: A total of 973 patients were included, comprising 769 without readmission (NRG) and 204 with readmission (RG). CLR was significantly higher in the RG group (3.63 vs 3.07, P < 0.001). Six predictors were selected by LASSO: sex, age, atrial fibrillation, β-blocker use, oral anticoagulant use, and CLR. Among the models, XGBoost achieved the best performance in the test set (AUC = 0.81, accuracy = 0.72) with good calibration. SHAP analysis identified CLR as the most influential variable (mean absolute SHAP = 0.42), contributing most strongly to readmission risk. Regular outpatient use of ...