A randomized placebo controlled trial found no benefit from testosterone replacement or physical exercise on vascular function. The abstract says:
"We sought to determine whether T supplementation improved markers of vascular aging in men with low-normal T, and whether T supplementation prevented arterial stiffness with resistance exercise. We studied 160 community dwelling older men (66 ± 5 years) with low-normal baseline total T levels (200-350 ng/dL). Participants were randomized to transdermal T gel targeting either a lower [400-550 ng/dL] or higher [600-1000 ng/dL] T range or to placebo gel, and to either progressive resistance training (PRT) or to no exercise for 12 months. Carotid artery stiffness (arterial compliance) and carotid IMT were measured at baseline, 6, and 12 months. Endothelial function (brachial artery flow-mediated dilation) was measured in a subset (N=86). Changes in carotid artery compliance, IMT and endothelial function with either the lower or higher range of T supplementation were not different from placebo at 6 or 12 months. There were no differences between PRT and no PRT groups, alone or with T supplementation, in changes in any of the vascular measures at either time point. Supplementation of T and PRT in older men with low-normal levels do not appear to improve or harm vascular function."
"We sought to determine whether T supplementation improved markers of vascular aging in men with low-normal T, and whether T supplementation prevented arterial stiffness with resistance exercise. We studied 160 community dwelling older men (66 ± 5 years) with low-normal baseline total T levels (200-350 ng/dL). Participants were randomized to transdermal T gel targeting either a lower [400-550 ng/dL] or higher [600-1000 ng/dL] T range or to placebo gel, and to either progressive resistance training (PRT) or to no exercise for 12 months. Carotid artery stiffness (arterial compliance) and carotid IMT were measured at baseline, 6, and 12 months. Endothelial function (brachial artery flow-mediated dilation) was measured in a subset (N=86). Changes in carotid artery compliance, IMT and endothelial function with either the lower or higher range of T supplementation were not different from placebo at 6 or 12 months. There were no differences between PRT and no PRT groups, alone or with T supplementation, in changes in any of the vascular measures at either time point. Supplementation of T and PRT in older men with low-normal levels do not appear to improve or harm vascular function."