Plasma levels and vascular effects of vasopressin in patients undergoing coronary artery bypass grafting☆

Abstract

Recent studies have suggested that endogenous vasopressin (AVP) acts as a spasmogen during coronary artery bypass grafting (CABG). Given that AVP could induce vasospasm in the grafted vessel, we assessed the release of this peptide during and after CABG, and explored ways of counteracting its contractile effect on the internal mammary artery (IMA). Plasma levels of AVP were determined by radioimmunoassay in 16 patients before, during and after CABG. Using isometric force recording techniques, we also investigated the mechanisms involved in the contractile effect of AVP in ring preparations of IMA specimens taken from 95 patients. Plasma AVP levels peaked after the start of cardiopulmonary bypass (CPB) and correlated well with serum osmolality (Pearson's r=0.9490; P<0.0001; n=16). An inverse correlation was observed between plasma AVP levels recorded at this stage and the maximal contraction induced in vitro by AVP in vascular rings from the same patients (Pearson's r=-0.6968; P<0.01; n=16). No change in the AVP response was produced by endothelium removal, exposure to the NO precursor (3 x 10(-4)M L-arginine), inhibition of nitric oxide (NO) synthase (3 x 10(-5) M L-NAME) or soluble guanylate cyclase (3 x 10(-6) M 1H-[1,2,4]oxadiazol [4,3,-alpha]quinoxalin-1-one (ODQ)), removal of the superoxide anion (100 U/ml superoxide dismutase (SOD) plus 1200 U/ml catalase) or hydroxyl radical (10(-4) M deferoxamine), or specific alpha1 - (10(-6) M prazosin) or endothelin (10(-5) M bosentan) receptor antagonism. In contrast, adenylate cyclase activation (3 x 10(-8) M forskolin) reduced the contractile response to AVP, while prostanoid synthesis (3 x 10(-6) M indomethacin) inhibition and blockade of Ca2+ -activated potassium channels (KCa) (10(-3) M tetraethylammonium (TEA)) enhanced AVP contraction. Age, gender and smoking also modified the AVP response. Our findings suggest a role for AVP as a modulator of vascular tone in human IMA. The effect of AVP is dependent on prostanoids and Ca2+ -activated K+ channels, so its dysfunction in pathophysiological cardiovascular processes could mean that AVP, among other factors, produces vasospasm in IMA grafts.