Kidneys from donation after circulatory death (DCD) are highly sensitive to ischemia-reperfusion injury and thus require careful reconditioning, such as normothermic regional perfusion (NRP). However, the optimal NRP protocol remains to be characterized. NRP was modeled in a DCD porcine model (30 minutes of cardiac arrest) for 2, 4, or 6 hours compared to a control group (No-NRP); kidneys were machine-preserved and allotransplanted. NRP appeared to permit recovery from warm ischemia, possibly due to an increased expression of HIF1α-dependent survival pathway. At 2 hours, blood levels of ischemic injury biomarkers increased: creatinine, lactate/pyruvate ratio, LDH, AST, NGAL, KIM-1, CD40 ligand, and soluble-tissue-factor. All these markers then decreased with time; however, AST, NGAL, and KIM-1 increased again at 6 hours. Hemoglobin and platelets decreased at 6 hours, after which the procedure became difficult to maintain. Regarding inflammation, active tissue-factor, cleaved PAR-2 and MCP-1 increased by 4-6 hours, but not TNF-α and iNOS. Compared to No-NRP, NRP kidneys showed lower resistance during hypothermic machine perfusion (HMP), likely associated with pe-NRP eNOS activation. Kidneys transplanted after 4 and 6 hours of NRP showed better function and outcome, compared to No-NRP. In conclusion, our results confirm the mechanistic benefits of NRP and highlight 4 hours as its optimal duration, after which injury markers appear.