Non-selective and cyclo-oxygenase-2-specific non-steroidal anti-inflammatory drugs impair the hyperaemic response of skin to brief axillary artery occlusion

Background. Cyclo-oxygenase-2 (COX-2)-specific inhibitors are marketed as safer analgesics than non-selective non-steroidal anti-inflammatory drugs. However, there has been conflicting evidence concerning endothelial function and cardiovascular risk after COX-2 inhibitor use. We investigated forearm skin vascular reactivity to brief axillary artery occlusion in healthy volunteers after a single dose of the non-steroidal anti-inflammatory drugs ibuprofen (non-selective) and rofecoxib (COX-2 specific). Methods. Ten healthy male volunteers were studied. Forearm skin blood flow was measured using laser Doppler flowmetry. Two non-invasive probes were placed on the volar surface of the forearm. The magnitude of hyperaemic response to brief (20 s) and prolonged (5 min) occlusion of the axillary artery was measured before and 2-3 h after administration of ibuprofen 800 mg or rofecoxib 25 mg. The transient hyperaemic response ratio (THRR), defined as the net peak hyperaemic flow-flux divided by the net baseline flow-flux, was calculated. Each volunteer received both drugs in random order at least 1 week apart. Results. Ibuprofen and rofecoxib increased net baseline blood flow (median (range): ibuprofen, from 23.3 (12.1-40.8) to 31.5 (17.4-77.3); rofecoxib, from 22.0 (14.6-41.0) to 29.9 (19.5-112.0); P<0.01). The net hyperaemic peak flow-flux after brief and prolonged occlusion was unchanged after both drugs. THRR was reduced by both drugs (ibuprofen, from 2.92 (2.38-3.86) to 2.09 (1.45-3.54); rofecoxib, from 3.36 (2.06-5.16) to 2.67 (1.15 -3.84); P<0.01). Conclusions. Both COX-2 and non-selective non-steroidal anti-inflammatory drugs, when given to healthy volunteers as single therapeutic doses, decrease skin microvascular tone but do not impair maximal vasodilatory ability.