Validated nomogram may Predict upper GI bleeding in peritoneal dialysis Patients: Study

Researchers have found in a new study that Calcium and PPI use, history of upper gastrointestinal bleeding (UGIB), hemoglobin levels, and uric acid were identified as significant predictors of UGIB in peritoneal dialysis (PD) patients. Therefore a validated nomogram based on these factors shows promise as a clinical tool for predicting UGIB risk in this population. The study was published in BMC Nephrology by Sijia S. and colleagues.

This retrospective study involved information from 2,107 PD patients who were treated from February 1, 2007, to November 15, 2021. The patients were split into training and validation groups randomly in a 1:1 ratio. Researchers used the least absolute shrinkage and selection operator (LASSO) regression method to determine candidate UGIB predictors. The predictors were then utilized to create a nomogram model.

To test the performance of the nomogram, the accuracy and clinical value of the model were both checked with a variety of metrics, ranging from receiver operating characteristic (ROC) curves and calibration curves to decision curve analysis (DCA). This strict testing guaranteed that the model was not only statistically valid but also useful in practice.

Key Findings

Of the 2,107 patients examined, 114 had UGIB. Five predictors were found to be predictive of UGIB and were included in the nomogram:

  • Calcium supplement use

  • Proton pump inhibitor (PPI) use

  • History of UGIB

  • Hemoglobin

  • Uric acid

  • The nomogram had excellent predictive performance.

  • The C-index, a measure of model discrimination, was 0.859 (95% CI: 0.810–0.909) in the training set and 0.874 (95% CI: 0.829–0.919) in the validation set.

  • These results reflect high accuracy in separating patients at risk for UGIB.

The analysis conclusively demonstrated that the application of calcium and PPIs, history of UGIB, hemoglobin level, and uric acid were the key predictors of UGIB in peritoneal dialysis patients. The reliable nomogram derived from these factors provides a convenient method of estimating the risk of UGIB. It has the potential to be used as a clinical decision-support tool to enhance patient care and outcomes in this high-risk group.

Reference:

Shang, S., Zhang, X., Zhan, X. et al. Development and validation of a predictive model for upper gastrointestinal bleeding in peritoneal dialysis patients: a retrospective, multicenter cohort study. BMC Nephrol 26, 383 (2025). https://doi.org/10.1186/s12882-025-04304-y

Facebook Comments