The use of arterial grafts, such as the radial artery (RA) and internal mammary artery (IMA), in coronary artery bypass grafting (CABG) is often preferred over saphenous vein (SV) because of the improved long-term patency. However, arterial grafts, particularly the RA, are prone to spasm. The rise in vasoconstrictor levels following CABG has been implicated in the pathogenesis of spasm (Downing & Edmunds, 1992), but whilst the properties of most platelet-derived vasoconstrictors have been extensively studied in human conduits, there is limited information on the diadenosine polyphosphates (DPPs; Ap_nA, where n is 3-7). Despite this, Ap₄A and Ap₅A have been proposed as hypotensive agents (Kikuta et al. 1999; van Ginneken et al. 2002). We therefore investigated the effect of DPPs on conduits used in CABG.

Sections of human RA, SV and IMA were obtained surplus to CABG, with ethical committee approval. Arterial rings in Krebs saline at 37°C, gassed with 95 % O₂-5 % CO₂, were set at a pretension equivalent to 100 mmHg for 1 h followed by 30 min at 20-30 mmHg. SV was pretensioned at 35 mN for 1 h followed by 30 min at 10 mN. Data are presented as means ± S.E.M. from n patients.

RA gave maximal responses to KCl (90 mM) of 68.3 ± 4.0 mN (n = 8) and IMA 25.5 ± 1.6 mN (n = 4). RA responded to the P2X receptor agonist α,β-methyleneATP with concentration-dependent increases in tension, giving EC₅₀ values of 1.9 ± 0.3 µM (n = 3) and maximal responses 74.0 ± 8.1 % KCl. Ap₅A also increased tension in a concentration-dependent manner, giving EC₅₀ values of 9.9 ± 2.7 µM and a maximal response 92.9 ± 3.0 % KCl. Prior application of 100 µM α,β-methylene-ATP to desensitize P2X receptors reduced the response to 10 µM Ap₅A to 16.2 ± 5.1 % of control values (n = 5). However, in the presence of 30 µM pyridoxalphosphate-6-azophenyl-2â,4â-disulphonic acid (PPADS) the response to 10 µM Ap₅A was enhanced to 151.7 ± 11.8 % (n = 4) of controls. RA also responded to Ap₄A and Ap₆A giving EC₅₀ values of 20.2 ± 7.9 µM (n = 3) for Ap₄A and 12.1 ± 2.1 µM (n = 3) for Ap₆A and maximal responses 79.1 ± 9.4 % and 88.6 ± 17.9 % KCl for Ap₄A and Ap₆A, respectively. In contrast, sections of IMA gave only weak responses to 100 µM α,β-methylene-ATP (10.0 ± 5.3 % KCl; n = 3) and negligible responses to 100 µM Ap₄A or 50 µM Ap₅A (n = 3). SV showed robust concentration-dependent contraction to Ap₄A, Ap₅A and Ap₆A (n = 3).

In conclusion, the DPPs are potential mediators of spasm in the RA but not the IMA and their use as hypotensive agents should be treated with caution.