A discrepancy exists worldwide between the number of suitable liver donors and the increasing demand for transplantation. Thus many centers have considered widening their liver donor acceptance criteria and this may increase the incidence of primary dysfunction (PD) with negative effect on the results of transplantation. In order to reduce the incidence of PD and improve patient and graft survival it becomes important to identify those risk factors associated with its occurrence. In a retrospective univariate and multivariate analysis we evaluated several donor, preservation and recipient parameters and their correlation with PD. In our Department 282 orthotopic liver transplantations (OLT) were performed on 256 adult patients over a 10-year period. Excluded were 15 cases with early vascular problems and 4 intraoperative deaths. A complete series of donor, recipient and procedure-related data were analyzed. About 30% of donors showed abnormal values. In 70 cases of PD (26%) there was a 61.4% graft failure rate compared with 15% in the group with immediate function (P < 0.05). Univariate analysis showed donor age, steatosis, ischemia time, amines, oliguria, hypotension and ICU stay to be significantly associated with PD. Multivariate analysis showed steatosis, ischemia time and amine dosage to be independent risk factors for the development of primary non function. In conclusion, the acceptance of marginal donors worsened the results of transplantation, but the rejection of these donors would reduce by about 30% our transplant activity resulting in increased mortality in the waiting list. Combinations of risk factors when possible should be avoided, and ischemia time, as the only variable that can be controlled, should be kept as short as possible.