〔Abstract〕 Objective To explore the diagnostic value of a column chart constructed based on composite inflammatory indicators for cardiac valve calcification(CVC) in peritoneal dialysis(PD) patients. Methods A retrospe-ctive analysis was conducted on the data of 117 PD patients admitted in the past 5 years, and the patients were divided into a CVC group and a non CVC group based on whether they had formed CVC. The general clinical data of the two groups were compared, and univariate and multivariate binary logistic regression analyses were used to determine the predictive variables and construct a predictive model. The two predictive models, which only included traditional factors and included composite inflammation indicators at the same time, were evaluated from the aspects of discrimination, calibration, and clinical practicality. Reclassification analysis was used to evaluate the improvement of the column chart model in identifying CVC formation in PD patients. Results A prediction model was established by incorporating six predictive variables: age, pan immune inflammatory value, blood calcium, C-reactive protein/albumin, low-density lipoprotein cholesterol, and parathyroid hormone. The area under the curve of the traditional prediction model without composite inflammatory indicators and the column chart prediction model with composite inflammatory indicators were 0.909 4(95%CI: 0.858 0-0.960 7) and 0.972 7(95%CI: 0.947 7-0.997 7), respectively, indicating good discriminability of the column chart prediction model. The calibration curve showed that the calibration curves before and after calibration were close to the fitting line, indicating that the calibration degree of the column chart prediction model was high. The decision curve showed that the column chart prediction model had a high net benefit. By calculating the net reclassification index and comprehensive discriminant improvement index. It was found that the column chart model had a significant improvement in identifying the risk of CVC formation in PD patients, indicating its good clinical effectiveness. Conclusion The column chart prediction model constructed with age, pan immune inflammatory value, blood calcium, C-reactive protein/albumin, low-density lipoprotein cholesterol, and parathyroid hormone can help identify the risk of CVC in PD patients and provide guidance for clinical diagnosis and treatment.