Development and Validation of a Risk Prediction Model for Interdialytic Hyperkalemia in Patients Undergoing Maintenance Hemodialysis.

Background: Interdialytic hyperkalemia is linked to a heightened risk of adverse clinical outcomes among patients undergoing maintenance hemodialysis (MHD). This study aimed to develop and validate a nomogram to assess the risk of interdialytic hyperkalemia in this patient population. Methods: A total of 312 patients undergoing MHD were retrospectively enrolled from two hemodialysis centers between January 2024 and December 2024. The group was randomly divided into a training set and a validation set in a 7:3 ratio. Least absolute shrinkage and selection operator regression was applied to identify independent predictors of hyperkalemia, which were subsequently incorporated into a multivariate logistic regression model. Model performance was assessed using receiver operating characteristic curve analysis, area under the curve (AUC), calibration plots, and decision curve analysis (DCA) to assess both discrimination and clinical utility. Results: The overall incidence of interdialytic hyperkalemia was 28.2%. The final nomogram included seven predictors: dietary potassium intake, dietary phosphorus intake, serum albumin concentration, pre-dialysis blood glucose level, interdialytic weight gain rate, time interval since the last dialysis session, and a history of hyperkalemia within the preceding three months. The model demonstrated strong discriminatory ability with an AUC of 0.905 (95% Confidence Interval (CI): 0.883-0.931) in the training set and 0.782 (95% CI: 0.756-0.819) in the validation set. Calibration plots indicated good agreement between predicted and observed outcomes. DCA confirmed the clinical applicability of the model by demonstrating a net benefit across a range of threshold probabilities. Conclusion: A nomogram-based risk prediction model for interdialytic hyperkalemia in patients undergoing MHD was developed and externally validated. The model demonstrated robust predictive performance and may assist clinicians in early identification of patients who are at high-risk, thereby supporting timely interventions.