#2528 Prognostic value of Glasgow Prognostic Score and its derived scores in patients undergoing hemodialysis

Abstract Background and Aims The Glasgow Prognostic Score (GPS), its modified version (mGPS), and the high-sensitivity mGPS (HS-mGPS) are widely recognized as effective tools for predicting outcomes in patients with cancer. However, their prognostic utility in individuals undergoing hemodialysis remains unclear. This study aimed to evaluate the effectiveness of the GPS, mGPS, and HS-mGPS in predicting the prognosis of patients undergoing hemodialysis. Method A retrospective cohort analysis was performed on 339 patients undergoing hemodialysis. GPS, mGPS, and HS-mGPS scores were determined, and their associations with all-cause mortality were examined using Kaplan–Meier survival analysis and Cox proportional hazards models. Additionally, the predictive performance of GPS for all-cause mortality was assessed by calculating the area under the receiver operating characteristic curve (AUC). Results During the median follow-up period of 1558 days, 221 patients (65.2%) died. Survival rates declined with higher GPS scores (0, 1, and 2), whereas in the mGPS and HS-mGPS scores, similar survival rates were observed between scores of 0 and 1. In the multivariate Cox proportional hazards analysis, GPS scores were significantly associated with higher all-cause mortality risks for patients with scores of 1 (HR 1.76, 95% CI: 1.29–2.42, P = 0.0004) and 2 (HR 2.87, 95% CI: 1.91–4.32, P < 0.0001) when compared to a score of 0. The predictive capability of the GPS for all-cause mortality (AUC, 0.68; 95% CI: 0.63–0.72) was significantly greater than that of the mGPS (AUC: 0.57; 95% CI: 0.54–0.61; P < 0.001). Although GPS showed a better predictive ability than HS-mGPS (AUC: 0.64; 95% CI: 0.59–0.69), this difference was not statistically significant (P = 0.10). Conclusion The findings suggest that GPS provides clearer stratification of hemodialysis patients into risk groups for poor prognosis than mGPS and HS-mGPS.