Flow-induced dilation (FID) is a physiologically important mechanism contributing to the local regulation of the coronary circulation. Animal studies demonstrate that FID is typically mediated by nitric oxide (NO), prostacyclin, or both. However in coronary arterioles (HCA) from patients with CAD, FID is dependent largely on endothelium-derived hyperpolarizing factor (EDHF) with little role for nitric oxide synthase (NOS) or cyclooxygenase (COX) products. Therefore we hypothesized that both NOS and COX contribute to FID in human arterioles from patients without CAD. HCA (∼150 μm ID) obtained at surgery from patients with or without CAD were prepared for videomicroscopy. After constriction with endothelin-1, vasomotor responses to flow were evaluated in the presence, absence or combination of the COX inhibitor Indomethacin (Indo) 10⁻⁵ M and the NOS inhibitor L-nitro-arginine methyl ester (L-NAME) 10⁻⁴ M. We examined the following age groups: children 0–18 years; adults 19–55 years; older adults >55 years without CAD, and adults with CAD (mean age: 64±8.2 years). FID was abolished by Indo in children (%max dilation: 3±9.6% vs. 79±1.5%; n=3; p<0.05) and reduced in adults (%max dilation: 50±3.7% vs. 78±2.9%; n=7; p<0.05). In contrast, Indo had no effect on FID in older adults (%max dilation: 58±4% vs. 63±4%; n=5) or adults with CAD (%max dilation: 65±4.2% vs. 65±4.1%; n=6). In children L-NAME + Indo eliminated FID, while L-NAME alone did not have this effect (%max dilation: −5.1±5.2% vs. 49±2.9%; n=3; p<0.05). In adults L-NAME and Indo tended to further reduce FID vs. L-NAME alone (%max dilation: 16±6.5% vs. 32.7±7.5%; n=5, p=NS). In conclusion, FID in children is mediated mostly by COX; whereas, during the aging process or in the presence of CAD its contribution is diminished. These novel findings may have clinical implications for the use of COX inhibitors in children and young adults.