OBJECTIVE: The objective of this study was to analyze whether erythropoietin (EPO) protects from necrosis of critically perfused musculocutaneous tissue and the mechanisms by which this protection is achieved. BACKGROUND: EPO is the regulator of erythropoiesis and is used to treat patients with anemia of different causes. Recent studies suggest that EPO has also other tissue-protective effects, irrespective of its erythropoietic properties. MATERIAL AND METHODS: C57BL/6-mice were treated with 3 doses of EPO at 500 IU/kg intraperitoneally. EPO was given either before (preconditioning, n = 7), before and after (overlapping treatment, n = 7), or after (treatment, n = 7) surgery. Animals receiving only saline served as controls (CON). Acute persistent ischemia was induced by elevating a randomly perfused flap in the back of the animals. This critically perfused tissue demonstrates an initial microvascular failure of approximately 40%, resulting in approximately 50% tissue necrosis if kept untreated. Repetitive fluorescence microscopy was performed over 10 days, assessing angiogenesis, functional capillary density, inflammatory leukocyte-endothelial cell interaction, apoptotic cell death, and tissue necrosis. Additional molecular tissue analyses included the determination of inducible nitric oxide synthase, erythropoietin receptor (EPO-R), and vascular endothelial growth factor (VEGF). RESULTS: EPO preconditioning did not affect hematocrit and EPO-R expression, but increased inducible nitric oxide synthase in the critically perfused tissue. This correlated with a significant arteriolar dilation, which resulted in a maintained functional capillary density (CON: 0 +/- 0 cm/cm(2); preconditioning: 37 +/- 21 cm/cm(2); overlapping treatment: 72 +/- 26 cm/cm(2); P < 0.05). EPO pretreatment further significantly reduced microvascular leukocyte adhesion and apoptotic cell death. Moreover, EPO pretreatment induced an early VEGF upregulation, which resulted in new capillary network formation (CON: 0 +/- 0 cm/cm(2); preconditioning: 40 +/- 3 cm/cm(2); overlapping treatment: 33 +/- 3 cm/cm(2); P < 0.05). Accordingly, EPO pretreatment significantly reduced tissue necrosis (CON: 48% +/- 2%; preconditioning: 26% +/- 3%; overlapping treatment: 20% +/- 3%; P < 0.05). Of interest, EPO treatment was only able to alleviate ischemia-induced inflammation but could not improve microvascular perfusion and tissue survival. CONCLUSIONS: EPO pretreatment improves survival of critically perfused tissue by nitric oxide -mediated arteriolar dilation, protection of capillary perfusion, and VEGF-initiated new blood vessel formation.