Fig-1-Production of immunoreactiveendothelin-l (ir ET-1) in the umbilical artery: electrolytic solution (I);rH uE PO solvent(II);30U of rHuEPO (III).Dear Sir, Induced hypertension due to treatments with recombinant erythropoietin (rHuEPO) is with no doubt the most frequent and impor tant side effect of this hormone therapy, ca pable of conditioning the therapeutic success in subjects affected by anemia from renal failure. Although different factors, among which the increase in hematic viscosity, are recognized as possibly responsible for the pressure rise which often goes along with rHuEPO treatment, up to now a reliable ex planation of the pathogenetic mechanisms of hypertension induced by erythropoietin [1] is still missing. In previous investigations, through the evaluation of the changes brought about by the intravenous administra tion of rHuEPO on the prétibial flux by means of the Xenon 133 clearance, we showed the existence of a direct vasoconstric tive action exerted by erythropoietin on the compliance vasa [2], The presence of endothe lial receptors forthe erythropoietin could give account of such action [3]. Furthermore, in our previous investigations as well as in other investigations [4], it was shown how the in travenous chronic treatment was able to change the plasma levels of endothelin-l, a substance produced by the endothelium, able to induce a high vasoconstriction. Yet, the plasma dosage of endothelin does not seem to correspond to the real activity of the vasal endothelium. Therefore, we wanted to use an experimental model based on the umbilical arteries, where the receptors for endothelin are to be found [5]. The arteries were quickly isolated from umbilical cords from neonates of nonhypertensive mothers with the purpose of evaluating the rHuEPO action on the pro duction of endothelin-1. Our method con sisted in taking, soon after a natural delivery, 10 cm of the median part of the umbilical cord from which we immediately isolated the um bilical artery by means of a stereo microscope which was divided into 3 2-cm long segments. The 1st segment (I) was put in I ml of cold buffered electrolyte solution (4°C), the 2nd (II) was kept in a medium with buffered elec trolyte solution plus 20 pi of rHuEPO solvent (human serum albumin 2,5 m g/m l; sodium 160 mM; chlorides 100 m M: citrate 20 mM), the 3rd (III) was put in I ml cold buffered electrolyte solution and 30 U of rHuEPO with 20 pi of solvent. Then, the 3 umbilical artery segments were kept for 10 min at 37°C, were weighed and the corresponding supernatant frozen at -30°C, the endothelin-1 was dosed with the RIA method (Amersham, UK). The results obtained from 25 umbilical arteries showed the presence of a significant increase in the concentration of endothelin-1 from the medium to which rHuEPO was added ( fig. 1). Results are expressed as means ± S E and Student's test was performed for all compari sons. Therefore, our study seems to confirm the possibility that the vasoconstrictive action played by erythropoietin might be caused by the stimu...