Objective  To explore the influencing factors of coronary artery calcification (CAC) in patients with chronic kidney disease (CKD) stages G3～5 and to develop and validate a nomogram prediction model. Methods  This single-center retrospective case-control study included 1,376 patients admitted to the Second Affiliated Hospital of Dalian Medical University between January 2017 and December 2022. Clinical data were collected, and the Agatston score combined with dedicated CAC quantification software was used to quantify the coronary artery calcification score (CACS). Patients were categorized into the non-severe calcification (0≤CACS≤400) and severe calcification (CACS＞400).  Results  overall, 963 patients were assigned to the training cohort and the remaining 413 to the validation cohort. Multivariate logistic regression analysis identified CKD stage G5 (OR=2.872, 95% CI：1.400～6.555, P=0.007), age (OR=1.036, 95% CI：1.018～1.055, P＜0.001), log-transformed B-type natriuretic peptide (LogBNP) (OR=1.419, 95% CI：1.236～1.638, P＜0.001), low-density lipoprotein cholesterol (OR=1.915, 95% CI：1.550～2.381, P＜0.001), intact parathyroid hormone (OR=1.000, 95% CI：1.000～1.001, P=0.039), calcium–phosphorus product (OR=1.321, 95% CI：1.117～1.563, P=0.001), use of acid-suppressive drugs (OR=1.756, 95% CI：1.134～2.739, P=0.012), and diabetes mellitus (OR=2.121, 95% CI： 1.347～3.335, P=0.001) as independent risk factors for severe CAC. Thyroxine(T4) (OR=0.920, 95% CI：0.857～0.986, P=0.021), serum magnesium (OR=0.028, 95% CI：0.007～0.109, P＜0.001), serum sodium (OR=0.923, 95% CI：0.878～0.970, P=0.002), and use of lipid-lowering drugs (OR=0.290, 95% CI：0.180～0.460, P＜0.001) were protective factors. The area under the curve (AUC) for the training and validation cohorts was 0.848 (95% CI：0.814～0.881) and 0.836 (95% CI：0.784～0.887), respectively, indicating good discriminative ability. The Hosmer–Lemeshow test and calibration curves demonstrated good agreement between predicted and observed outcomes in both cohorts (mean absolute error: 0.008 and 0.036; P=0.478 and 0.215). Decision curve analysis (DCA) showed favorable clinical net benefit across threshold probabilities of 0.01～0.78 in the training cohort and 0.04～0.52 in the validation cohort. Conclusion: The established nomogram effectively predicts the risk of severe CAC in patients with CKD stages G3～5, showing good discrimination and clinical utility, and provides quantitative evidence for individualized clinical decision-making.