The purpose of this study was to evaluate the long-term effectiveness of basic fibroblast growth factor (bFGF) in achieving neovascularization following ischemia from arterial ligation and to determine an optimal dosage level. We used an Ameroid constrictor to produce progressive occlusion of the left femoral artery of rabbits. At 2 weeks, the rabbits were randomized to receive intravenous injection of vehicle (group A, n = 15); 3 microg/kg/day bFGF (group B, n = 12); 10 microg/kg/day bFGF (group C, n = 12); or 16 microg/kg/day bFGF (group D, n = 15) for 3 days. At 1 to 37 days after surgery, we assessed limb neovascularization by transcutaneous oximetry (TCPO(2)), angiography, heart rate, arterial pressure, peripheral vascular resistance (PRU), and muscle blood flow (MBF) during steady-state intra-arterial infusion of saline (basal), acetylcholine, papaverine, or serotonin under anesthesia and capillary density (cap/mm(2)) and capillary per muscle fiber ratio (cap/F). Groups B and C showed significantly greater change in TCPO(2) over time than groups A and D (P < 0.0001). Group D showed the lowest TCPO(2) values from days 14 to 37 and group C the highest. Groups B and C showed a higher number of vessels filled with contrast agent than groups A and D (P < 0.0001). Calf cap/mm(2) and cap/F were significantly higher in groups B and C than groups A and D (P < 0.0001). Calf basal MBF values were higher in groups B and C than in groups A and D, but were not statistically significant. Group D showed the highest level in basal PRU. There were no significant differences in heart rate or blood pressure among the groups. These results show (1) treatment with bFGF has no adverse hemodynamic effects, (2) bFGF enhances angiogenesis and circulation at moderate doses, and these effects persist at least several weeks, and (3) high doses of bFGF may inhibit angiogenesis and collateral circulation.