Development and Validation of a Nomogram for Predicting All-Cause Mortality in Patients with Hemodialysis Having Pulmonary Hypertension

Abstract

<b><i>Introduction:</i></b> Patients with end-stage renal disease receiving hemodialysis (HD) have a high morbidity and mortality rate associated with pulmonary hypertension (PH). A nomogram was developed to predict all-cause mortality in HD patients with PH. In this study, we aimed to validate the usefulness of this nomogram. <b><i>Methods:</i></b> A total of 274 HD patients with PH were hospitalized at the Affiliated Hospital of Xuzhou Medical University between January 2014 and June 2019 and followed up for 3 years. Echocardiography detected PH when the peak tricuspid regurgitation velocity (TRV) was more than 2.8 m/s. To evaluate the all-cause mortality for long-term HD patients with PH, Cox regression analysis was performed to determine the factors of mortality that were included in the prediction model. Next, the area under the receiver-operating characteristic curve (AUC-ROC) was used to assess the predictive power of the model. Calibration plots and decision curve analysis (DCA) were used to assess the accuracy of the prediction results and the clinical utility of the model. <b><i>Results:</i></b> The all-cause mortality rate was 29.20% throughout the follow-up period. The nomogram comprised six commonly available predictors: age, diabetes mellitus, cardiovascular disease, hemoglobin, left ventricular ejection fraction, and TRV. The 1-year, 2-year, and 3-year AUC-ROC values were 0.842, 0.800, and 0.781, respectively. The calibration curves revealed excellent agreement with the nomogram, while the DCA demonstrated favorable clinical practicability. <b><i>Conclusion:</i></b> The first developed nomogram for predicting all-cause mortality in HD patients with PH could guide clinical decision-making and intervention planning.