Assessing and comparing rival definitions of delayed renal allograft function for predicting subsequent graft failure

Background. The traditional definition of delayed graft function (DGF) rests on dialysis requirement during the first postoperative week. Subsequently, a more objective and "functional" definition of DGF (fDGF) has been proposed as an alternative to this dialysis-based definition of DGF (dDGF) and defined as a failure of the serum creatinine to decrease by at least 10% daily on 3 successive days during the first week posttransplantation, irrespective of dialysis requirement. However, an association between fDGF and long-term graft failure has not been fully established, and it is unknown whether fDGF is a better marker of subsequent outcomes than dDGF. Methods. We studied 750 adult deceased donor kidney transplant recipients (1996-2006) and analyzed the association between these two DGF definitions and long-term graft outcome. Results. Univariable associations with death-censored graft failure were seen for both dDGF and fDGF (hazard ratio [HR] 1.59; 95% confidence interval [CI] 1.16-2.18; P=0.004 and HR 1.72; 95% CI 1.26-2.36; P=0.001, respectively). On bivariable analysis (dDGF vs. fDGF), dDGF lost significance, whereas the effect of fDGF persisted (HR 1.52; 95%CI 1.03-2.25; P=0.04). This was also the case in a multivariable model, where fDGF but not dDGF was significantly associated with graft failure (HR 1.47; 95%CI 1.06-2.03; P=0.02). Results were similar for overall graft failure. Conclusions. This study confirms the utility of fDGF as an early marker of subsequent inferior allograft outcomes, suggesting superiority over the traditional (often subjective) dialysis-based definition. Wider adoption of the fDGF definition should be considered, both as a risk-stratification tool in clinical practice and a clinical trial endpoint.