目的 探索不同剂量及给药时间的重组人促红细胞生成素（recombinant human erythropoietin，rhEPO）对大鼠肾脏缺血再灌注损伤（ischemia-reperfusion injury，IRI）的保护作用。方法 48只雄性SD大鼠随机分为假手术组（S组）、缺血对照组（IR组）、EPO-1组（再灌注时给药1 000、2 000和3 000 U/kg）和EPO-2组（再灌注前给药1 000、2 000和3 000 U/kg）。阻断左侧肾蒂45 min后切除右侧肾脏，建立IRI模型。检测血液中血肌酐（Scr）、血浆尿素氮（BUN）、超氧化物歧化酶（SOD）、丙二醛（MDA）、IL-6和Caspase-3水平，观察肾脏的病理学改变，采用TUNEL法检测肾小管上皮细胞凋亡。结果 EPO组血清Scr、BUN和肾组织MDA、Caspase-3、IL-6水平明显低于IR组（P<0.05）；EPO组肾组织SOD水平明显高于IR组（P<0.05），并在组织损伤上较IR组轻；TUNEL染色观察到EPO组阳性细胞数明显少于IR组（P<0.05）。再灌注前给药优于再灌注时给药（P<0.05）。结论 rhEPO对肾脏IRI有较好的保护作用，该作用可能通过抗氧自由基、减轻炎症反应、减少肾小管上皮细胞凋亡的协同机制来实现。保护作用与药物剂量呈正相关，再灌前给药优于再灌注时给药。

Objective To investigate protective effect of recombinant human erythropoietin (rhEPO) at different dosages and administration times on renal ischemia-reperfusion injury (IRI) in rats. Methods Forty-eight male Sprague-Dawley rats were divided into 4 groups: Sham, IR, EPO-1 (injection via inferior vena cava 30 min before releasing the clamp at the dosages of 1 000, 2 000 and 3 000 U/kg) and EPO-2 (injection through left renal vein 30 min before releasing the clamp at the dosages of 1 000, 2 000 and 3 000 U/kg). IRI model was made by right nephrectomy 45 min after left renal ischemia with non-traumatic vascular clamps. We detected serum creatinine (Scr), blood urea nitrogen (BUN), MDA, SOD, IL-6 and Caspase-3 levels. Apoptosis of tubular epithelium was observed by TUNEL assay. Results Scr, BUN, MDA, Caspase-3 and IL-6 levels in EPO groups were significantly lower than those in IR group (P<0.05). Compared to IR group, EPO groups had higher SOD levels (P<0.05). It showed that histological damage and TUNEL-positive cells were less in EPO groups than those in IR group (P<0.05). We found that administration before reperfusion at the dosage of 3 000 U/kg had better effect than administration in the moment of reperfusion (P<0.05). Conclusions rhEPO can attenuate renal IRI. The protective effect might be mediated by its inhibition of immflamitation, oxidation and tubular epithelium apoptosis. This effct has dependence on the dosage, and administration before reperfusion has better effect than administration in the moment of reperfusion.