1. ¹Department of Internal Medicine III, Kurume University School of Medicine, Kurume, Fukuoka, Japan
2. ²Cardiovascular Research Institute, Kurume University School of Medicine, Kurume, Fukuoka, Japan

Correspondence: Hidehiro Matsuoka, MD, PhD, Department of Internal Medicine III, Kurume University School of Medicine, 67 Asahi-machi, Kurume, Fukuoka 830-0011, Japan E-mail: matsuoka@med.kurume-u.ac.jp

^*This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture, Tokyo, by a research grant from Kimura Memorial Heart Foundation, Kurume, and by a research grant from the Renal Anemia Foundation, Tokyo.

Received 20 April 1998; Accepted 7 February 1999.

Abstract

Hypertension is a major complication of recombinant human erythropoietin therapy in patients with end-stage renal disease. Although the mechanisms for this pressor effect remain unknown, in vitro and ex vivo experiments suggest that erythropoietin stimulates release of cyclooxygenase-dependent endothelium-derived contracting factors (EDCFs) and attenuates endothelium-dependent vasorelaxation. To investigate the effect of erythropoietin on human endothelial function, we measured forearm blood flow by strain gauge plethysmography before and after intraarterial incremental administration of erythropoietin (10 to 300 IU/min, total 3000 U), while infusing acetylcholine (4 to 24 g/min) and sodium nitroprusside (0.2 to 1.2 g/min) in healthy male volunteers (n = 12). To examine a possible role of EDCFs, we repeated the same protocol under the pretreatment with oral indomethacin (50 mg), a cyclooxygenase inhibitor (n = 8). Infusion of erythropoietin into the brachial artery increased the erythropoietin blood concentration of venous effluents significantly (P < .0001) without changes in blood pressure and basal forearm blood flow. Acetylcholine and sodium nitroprusside increased forearm blood flow dose dependently before and after erythropoietin administration. However, acetylcholine-induced vasodilation (endothelium dependent) was significantly attenuated (P < .001) after erythropoietin administration, whereas vasodilation to sodium nitroprusside (endothelium independent) was unchanged. After indomethacin pretreatment, erythropoietin no longer attenuated endothelium-dependent vasorelaxation. Our results indicate that erythropoietin may impair endothelial function acutely, probably through cyclooxygenase-dependent EDCFs in humans. Long-term effects of erythropoietin on endothelial function in uremic patients remain to be elucidated.