Allocation rules should be able to guarantee an adequate balance between efficiency and equity. In our country, one of the two kidneys procured will be allocated locally to the recipient with the highest score from the transplant center were the kidney is harvested, if it has a transplant center, or to the center that performs the retrieval if the donor comes from a non transplant hospital For that purpose there is a rotating schedule among transplant centers (local kidney). The other kidney is nationally shared to the recipient with the highest score in the whole country (shared kidney). The allocation rule gives 60% of the score for HLA match, 20% for time on the waiting list (TWL) and 20% for PRA. Pediatric recipients get extra points. Locally, Blood Group 0 kidneys are given to any compatible recipient. Shared Blood Group 0 kidneys are given to Group 0, B and AB except for emergencies and after to years on the waiting list for Group A. The aim of this paper is to report the outcomes of this policy on TWL as a measure of equity.
Patients and Methods: Between January 2000 and December 2009, 3824 patients entered the kidney waiting list. Male 56%, pediatric 8%, 94% first transplant, 63% blood group 0, 27% Group A, 8% Group B.  Patients HLA A,B and DR antigens, were divided in four quartiles according to their relative frequency in the whole study population. TWL was calculated for each relevant factor. A Cox proportional hazards analysis was performed. Results are presented as Hazard Ratios (HR).
Results: A total of 1826 (48%) patients received a transplant during the study period. Mean TWL for those transplanted was 568 ± 522 days, 1251 ± 928 for those still waiting (weighed average 891 ± 817 days). Male patients  (HR: 1,18), pediatric recipients  (HR:2,6), Group A (HR:2,4) Group B (HR2,2), and Group AB (HR 4,2) had significantly higher chance of receiving a transplant. On the contrary, waiting for a second transplant  (HR: 0,36), PRA higher than 60% (HR:0,37) or an infrequent HLA DR  (HR:0,84) had a significantly lower chance of being transplanted. When the analysis was restricted to patients who had received a transplant we noticed that local kidney recipients had a significantly lower TWL (563  ± 532   vs 583  ± 543   p <0,05)   and a higher chance of receiving a transplant (HR: 1.11). In this subgroup the higher chance of receiving a transplant for males was not significant, suggesting that the difference TWL for males was related to inequities related to locally shared kidneys.
Conclusions: The policy of keeping one, sharing one was implemented to try to increase organ donation by incentivizing transplant teams. However this policy introduces biases favoring certain groups (males) and disfavoring Group 0 patients. Notably donation rates in our country have not been increased by this rule. This policy should be revised and probably changed to a national sharing of all kidneys and the blood group distribution should also be changed in order to compensate blood Group 0 patients.