Prediction of kidney transplant outcome based on different DGF definitions in Chinese deceased donation

Background Delayed graft function (DGF) is an important complication of kidney transplantation and can be diagnosed according to different definitions. DGF has been suggested to be associated with the long-term outcome of kidney transplantation surgery. However, the best DGF definition for predicting renal transplant outcomes in Chinese donations after cardiac death (DCDs) remains to be determined. Method A total of 372 DCD kidney transplant recipients from June 2013 to July 2017 in the First Affiliated Hospital of Xi'an Jiaotong University were included in this retrospective study to compare 6 different DGF definitions. The relationships of the DGF definitions with transplant outcome were analyzed, including graft loss (GL) and death-censored graft loss (death-censored GL). Renal function indicators, including one-year estimated glomerular filtration rate (eGFR) and three-year eGFR, and were compared between different DGF groups. Results The incidence of DGF varied from 4.19 to 35.22% according to the different DGF diagnoses. All DGF definitions were significantly associated with three-year GL as well as death-censored GL. DGF based on requirement of hemodialysis within the first week had the best predictive value for GL (AUC 0.77), and DGF based on sCr variation during the first 3 days post-transplant had the best predictive value for three-year death-censored GL (AUC 0.79). Combination of the 48-h sCr reduction ratio and classical DGF can improve the AUC for GL (AUC 0.85) as well as the predictive accuracy for death-censored GL (83.3%). Conclusion DGF was an independent risk factor for poor transplant outcome. The combination of need for hemodialysis within the first week and the 48-h serum creatinine reduction rate has a better predictive value for patient and poor graft outcome.