Predictive Models for the Functional Recovery of Transplanted Kidney

Abstract

Background: Renal transplantation is the gold standard treatment for end-stage renal disease, however, in 20% of cases, the graft develops a delayed graft function (DGF) that is associated with both early and late worsening of the outcome. The aim of this study was to examine and validate in a population of transplanted patients the appropriateness of the predictive score systems of DGF available to identify patients who might take advantage of a tailored immunosuppressive therapy.

Materials and methods: We conducted a systematic review of the literature to identify articles concerning scoring systems predicting DGF to identify those applicable to the study population and subsequently comparing their appropriateness for defining the most accurate one.

Results: From an analysis of the scientific literature, we found 7 scoring systems predicting DGF. Of these, 3 can be calculated for the study population. We enrolled 247 renal transplants in the study. DGF was recorded in 41 cases (15.95%). The Irish score recognized 25 of 41 cases (60.98%), the Jeldres score 41 of 41 cases (100%), and the Chapal score only 7 of 41 (17.07%). Although the Irish score did not identify all cases of DGF, the analysis of data revealed that it is the most accurate, with area under the receiver operating characteristic almost overlapping.

Conclusions: The study resulted in some interesting and promising conclusions about the predictability of DGF, defining the Irish score as the most reliable. This result can be considered the fundamental requirement to develop a custom therapeutic algorithm to be applied to all recipients with higher probability of developing DGF.