The aim of this study was to determine the impact of catheter sheath insertion, a model of endothelium disruption in humans, on the conventional FMD (flow-mediated dilatation) response in vivo. Seventeen subjects undergoing transradial catheterization were recruited and assessed prior to, the day after, and 3-4 months postcatheterization. The catheter sheath's external diameter was 2.7 mm, and the average preprocedure internal radial artery diameter was 2.8 mm, indicating a high likelihood of endothelial denudation as a consequence of sheath placement. Radial artery flow-mediated and endothelium-derived NO (nitric oxide)-dependent function (FMD) was assessed within the region of sheath placement (sheath site) and also above the sheath (catheter site). GTN (glyceryl trinitrate) endothelium-independent NO-mediated function was also assessed distally. Measurements were made in both arms at all time points; the non-catheterized arm provided an internal control. Neither sheath (4.5+/-0.9%) nor catheter (4.4+/-0.9%) insertion abolished FMD, although both significantly decreased FMD from preintervention levels (9.0+/-0.8% sheath segment; 8.4+/-0.8% catheter segment; P<0.05). The impact of sheath and catheter placement on FMD was no longer evident after approximately 3 months recovery (8.0+/-1.5 and 8.1+/-1.7%, sheath and catheter, respectively). GTN responses also decreased from 14.8+/-1.7 to 7.9+/-1.0% (P<0.05) as a result of sheath placement, but values returned to baseline at approximately 3 months (13.0+/-1.8%). These results suggest that the presence of an intact, functional endothelial layer and consequent NO release may not be obligatory for some component of the FMD response. This raises the possibility of an endothelium-independent contribution to the flow-induced vasodilatation in humans.