Granulocytc-macrophage colony-stimulating factor stimulates arteriogenesis in a pig model of peripheral artery disease using clinically applicable infusion pumps

Background: A growing number of patients suffer from peripheral artery disease (PAD). Current therapies are often limited by the extent of vascular pathology and the occurrence of restenosis after angioplasty. The stimulatory effect of growth factor administration on collateral vessel formation (arteriogenesis) has evolved as a potential new treatment for this patient group. Granulocyte-macrophage colony-stimulating factor (GM-CSF) was shown to stimulate arteriogenesis in small-animal models and in a pilot study in patients with coronary artery disease. Although a recent clinical study demonstrated disappointing results after subcutaneous GM-CSF application in patients with PAD, we hypothesized that intra-arterial cytokine application using implantable infusion pumps might well stimulate arteriogenesis in a large-species model of peripheral vascular disease. We also aimed to compare continuous and intermittent infusion regimens and to validate experimental and clinically available measurements of collateral artery growth. Methods: Twenty-four pigs underwent unilateral occlusion of the right femoral artery and received either GM-CSF continuously, GM-CSF intermittently, or phosphate-buffered saline (PBS). After 1 week, collateral conductance was determined under maximal vasodilatation with adenosine and by using a pump-driven extracorporal shunt system. Results. Conductance showed a significant stimulatory effect of GM-CSF on arteriogenesis (collateral conductance [mL/min/ mm Hg]: PBS, 37.7 +/- 5.4; GM-CSF continuous, 69.2 +/- 12.5; GM-CSF intermittent, 71.5 +/- 11.1). Flow measurements under reactive hyperemia were consistent with these results (flow occluded/non-occluded hind limb: PBS, 40.5% +/- 9.1%; GM-CSF continuous, 48.9% +/- 3.9%; GM-CSF intermittent, 48.7% +/- 4.4%). Measurements of ankle/brachial indices were not sensitive enough to detect the differences in collateral growth between the three groups. Conclusion: These results demonstrate the proarteriogenic properties of GM-CSF in larger animal species, revealing comparable efficacy of continuous and intermittent intra-arterial infusion. Furthermore, we provide evidence that implantable pumps offer a possible means for the intra-arterial application of growth factors. Intra-arterial application of GM-CSF might be a future treatment option for vascular occlusive disease. Finally, we show that in the peripheral circulation, pressure measurements alone have a low sensitivity to determine the effects of proarteriogenic therapy compared with flow or combined flow-pressure measurements.