A nomogram predicts the risk factors for bleeding connected with biliary stent placement therapy

Background : Clinically, the ability to predict intra-and postoperative bleeding during stent placement is limited, particularly in patients with high-risk bleeding factors or coagulation disorders. The aim of this study was to create a nomogram to predict bleeding from stent implantation. Methods : Medical records were collected on March 25, 2024, for 590 cases. Logistic regression was used to construct the nomogram, and variable screening was performed through LASSO regression. The area under the curve (AUC) was calculated to determine the discriminative capacity of each model. A calibration curve was used to evaluate the prediction accuracy of the model, and the consistency index (CI) was used to determine the performance of the model. Clinical effectiveness was assessed through the use of decision curve analysis (DCA). Results： LASSO regression identified four clinical features of patients, which were used to generate a nomogram. The AUC was 0.816 (95% CI, 0.743-0.889) in the training model and 0.804 (95% CI, 0.691-0.916) in the test set model. The proposed nomogram would yield a net gain in forecasting stent insertion bleeding if the threshold likelihood of bleeding fell between 0 and 0.8, as demonstrated by a well-calibrated predictive model and DCA. Conclusions : Four clinical traits of patients who underwent stent placement were combined to create a nomogram predicting risk of bleeding.