Predicting 5-year survival after kidney transplantation in Colombia using the survival benefit estimator tool

• Published in: PloS one 2023-08, Vol.18 (8), p.e0290162-e0290162
• Main Authors: Nino-Torres, Laura, García-Lopez, Andrea, Patino-Jaramillo, Nasly, Giron-Luque, Fernando, Nino-Murcia, Alejandro
• Format: Article
• Language: English
• Subjects: Age; Biology and Life Sciences; Creatinine; Diabetes; Ethics; Evaluation; Health risks; Hemodialysis; Hypertension; Kidney transplantation; Kidney transplants; Kidneys; Medical research; Medicine and Health Sciences; Medicine, Experimental; Mortality; Observations; Outcome and process assessment (Health Care); Patient monitoring; Patients; People and places; Performance evaluation; Performance prediction; Postoperative period; Prospective payment systems (Medical care); Regression analysis; Regression models; Risk factors; Social Sciences; Spain; Subgroups; Survival; Transplantation
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0290162

Introduction A complex relationship between donor and recipient characteristics influences kidney transplant (KT) success. A tool developed by Bae S. et al. (Survival Benefit Estimator, SBE) helps estimate post-KT survival. We aim to evaluate the predictive performance of the SBE tool in terms of 5-year patient survival after a kidney transplant. Methods A retrospective cohort study of all deceased-donor KT recipients between January 2009 to December 2021. A descriptive analysis of clinical and sociodemographic characteristics was performed. The SBE online tool was used to calculate the predicted patient survival (PPS) and the survival benefit at five years post-KT. Comparisons between predictive vs. actual patient survival were made using quintile subgroups. Three Cox regression models were built using PPS, EPTS, and KDPI. Results A total of 1145 recipients were evaluated. Mortality occurred in 157 patients. Patient survival was 86.2%. Predictive survival for patients if they remained on the waiting list was 70.6%. The PPS was 89.3%, which results in a survival benefit (SB) of 18.7% for our population. Actual survival rates were lower than the predicted ones across all the quintiles. In unadjusted analysis, PPS was a significant protective factor for mortality (HR 0.66), whereas EPTS (HR 8.9) and KDPI (HR 3.25) scores were significant risk factors. The discrimination of KDPI, PPS, and EPTS scores models were 0.59, 0.65, and 0.66, respectively. Conclusion SBE score overestimated actual survival rates in our sample. The discrimination power of the score was moderate, although the utility of this tool may be limited in this specific population.