Nomogram effective for Early detection of peritoneal dialysis-associated peritonitis, reveals study
The proposed nomogram demonstrated strong predictive accuracy and clinical usefulness, offering a valuable tool for early identification and intervention in peritoneal dialysis-associated peritonitis. However, further external validation and prospective studies are advised to confirm its reliability. A recent study was published in BMC Nephrology by Yuehong W. and colleagues. The model, developed based on clinical and laboratory factors, performed well in both the training and validation sets. Early detection of patients at increased risk for PDAP may allow timely interventions to affect improved clinical outcomes in dialysis management.
Peritoneal dialysis (PD) is lifesaving renal replacement therapy, but PDAP is a serious complication that can result in catheter loss, hospitalization, or even death. Individualized risk prediction tools for PDAP are lacking, yet it is such a valuable one. This research sought to address this lack by developing and validating a predictive model with actual clinical data.
Research was carried out at Nanhai District People's Hospital in Foshan City, Guangdong Province, on clinical data of 376 patients who underwent PD between December 2017 and December 2024. The patients were divided randomly into two groups: a training set of 244 patients and a validation set of 132 patients.
To determine risk factors, researchers utilized Least Absolute Shrinkage and Selection Operator (LASSO) regression, followed by multivariate logistic regression analysis. A nomogram was built based on the variables identified using R software (version 4.1.3) for visual prediction of the likelihood of developing PDAP.
Model performance was assessed using a wide range of tools:
• Receiver Operating Characteristic (ROC) curves
• Hosmer-Lemeshow goodness-of-fit test
• Decision Curve Analysis (DCA)
• Clinical Impact Curves (CICs)
Key Findings
Eight predictors were identified by LASSO regression as the most important predictors. Multivariate logistic regression also found that the following independent factors were associated with higher PDAP risk (P = 0.001):
• Age
• Duration of dialysis
• Albumin
• Hemoglobin
• β2-microglobulin
• Potassium
• Lymphocyte count
The nomogram's predictive ability was found to be excellent:
• The area under the curve (AUC) was 0.929 in the training set (95% CI: 0.896–0.962)
• The AUC was 0.905 in the validation set (95% CI: 0.855–0.955)
• The Hosmer-Lemeshow test revealed good model calibration:
• Training set χ² = 13.181, P = 0.106
• Validation set χ² = 8.264, P = 0.408
Decision Curve Analysis (DCA) and Clinical Impact Curves (CICs) established that the model yielded true clinical benefit over a broad spectrum of risk thresholds. That is, the nomogram could potentially be utilized to inform clinical decisions about monitoring and prophylactic measures.
This research successfully established and validated a nomogram that shows great accuracy and clinical use in the prediction of peritoneal dialysis-associated peritonitis risk. The instrument provides significant assistance in the early risk evaluation and specific intervention in PD patients.
Reference:
Wang, Y., Wu, Z., Huang, L. et al. A nomogram for predicting the risk of peritoneal dialysis-associated peritonitis in patients with end-stage renal disease undergoing peritoneal dialysis: model development and validation study. BMC Nephrol 26, 248 (2025). https://doi.org/10.1186/s12882-025-04165-5
