Renal Function at Discharge Among Kidney Recipients Experiencing Delayed Graft Function and Its Associations With Long-term Outcomes.

Abstract

Using Organ Procurement and Transplantation Network data, we examined DGF, and used recipient serum creatinine at discharge as a correlate of renal function and DGF severity (mild: <2.5 mg/dL; severe: ≥2.5 mg/dL). The associations between donor and recipient factors and DGF severity were quantified using logistic regression. We also examined the associations between DGF severity and long-term recipient outcomes, adjusting for potential confounders. A predictive model using donor and recipient factors had a reasonably good ability to discriminate mild (low creatinine) versus severe (high creatinine) DGF (c-statistic of 0.70). In Cox regression, DGF and creatinine at discharge were both independently associated with long-term outcomes, yet their effects differed depending on the outcome (graft function, death-censored graft function, recipient mortality). Our findings suggest that having DGF, but with relatively good renal function (creatinine <2.5) at discharge, may be less deleterious on graft and recipient survival compared with severe, prolonged DGF, which was associated with a decreased median graft survival of ~2.6 y compared with no DGF with low creatinine at discharge. Our novel DGF severity stratification identified unique factors associated with DGF severity, along with DGF's association with long-term graft and patient survival. The adverse cost and outcome implications of severe DGF warrant additional investigation to improve kidney transplantation practice.