ABSTRACT: INTRODUCTION: Endothelial progenitor cells (EPC) capable of initiating or augmenting vascular growth were recently identified within the small population of CD34-expressing cells that circulate in human peripheral blood and which are considered hematopoietic progenitor cells (HPC). Soon thereafter human HPC began to be used in clinical trials as putative sources of EPC for therapeutic vascular regeneration, especially in myocardial and critical limb ischemias. However, unlike HPC where hematopoietic efficacy is related quantitatively to CD34+ cell numbers implanted, there has been no consensus on how to measure EPC or how to assess cellular graft potency for vascular regeneration. We employed an animal model of spontaneous neovascularization to simultaneously determine whether human cells incorporate into new vessels and to quantify the effect of different putative angiogenic cells on vascularization in terms of number of vessels generated. We systematically compared competence for therapeutic angiogenesis in different sources of human cells with putative angiogenic potential, to begin to provide some rationale for optimising cell procurement for this therapy. METHODS: Human cells employed were mononuclear cells from normal peripheral blood and HPC-rich cell sources (umbilical cord blood, mobilized peripheral blood, bone marrow), CD34+ enriched or depleted subsets of these, and outgrowth cell populations from these. An established sponge implant angiogenesis model was adapted to determine the effects of different human cells on vascularization of implants in immunodeficient mice. Angiogenesis was quantified by vessel density and species of origin by immunohistochemistry. RESULTS: CD34+ cells from mobilized peripheral blood or umbilical cord blood HPC were the only cells to promote new vessel growth, but did not incorporate into vessels. Only endothelial outgrowth cells (EOC) incorporated into vessels, but these did not promote vessel growth. CONCLUSIONS: These studies indicate that, since EPC are very rare, any benefit seen in clinical trials of HPC in therapeutic vascular regeneration is predominantly mediated by indirect proangiogenic effects rather than through direct incorporation of any rare EPC contained within these sources. It should be possible to produce autologous EOC for therapeutic use, and evaluate the effect of EPC distinct from, or in synergy with, the proangiogenic effects of HPC therapies.