There is a strong association between delayed graft function (DGF) and reduced graft survival (GS) of cadaveric renal transplants. This study was performed to identify donor characteristics that might predict adverse outcomes.We reviewed the folders of 509 consecutive organ donors for 586 renal transplant recipients receiving grafts between 1990 and 1995. A uniform immunosuppression protocol was employed.The factors that did not alter the rate of DGF were procurement year, local versus shared organs, donor gender, race, hypotension, serum creatinine level and trend, blood transfusions, and vasopressor use and dose. The factors that did alter the frequency of DGF were cause of death (P=0.0053), donor age (P=0.0017), cold ischemic time (P=0.0009), anastomotic time (P=0.0012), combined cold ischemic time and anastomotic time (P=0.00018), and body mass index (P=0.009). All of the factors with the exception of body mass index were of comparable import when analyzed by multiple logistic regression. One-year GS of patients without DGF was 93.2%, and the GS of those with DGF was 76.6% (P < 0.0001). However, none of the donor factors correlated with 1-year GS. Seventy-seven donors were the source of paired transplants performed by our program. Sixty percent were concordant for immediate function, 32% were discordant for DGF with equal numbers affecting the first or second graft, and in only 8% did DGF affect both grafts.Donor factors associated with DGF were increased ischemia, donor age, and cause of death. Although there is a close association between DGF and reduced GS, there is no association between these donor factors and GS. This seeming paradox suggests that unknown variables contribute heavily to early graft outcome.