Acute mental stress can result in transient endothelial dysfunction, but the prognostic relevance of this phenomenon is unknown. To determine the association between mental stress-induced impairment in endothelium-dependent relaxation as assessed by brachial artery flow-mediated vasodilation and adverse cardiovascular outcomes among individuals with stable coronary artery disease. This cohort study was conducted at a university-affiliated hospital network between June 2011 and August 2014. A cohort of individuals with stable coronary artery disease were included. Data analysis took place from November 2018 to May 2019. Study participants were subjected to a laboratory mental stress task (public speaking). Flow-mediated vasodilation was measured before and 30 minutes after a public-speaking mental stress task. We examined the association of the rest (prestress), poststress, and δ flow-mediated vasodilation (poststress minus prestress levels) with an adjudicated composite end point of adverse events, including cardiovascular death, myocardial infarction, unstable angina leading to revascularization, and heart failure hospitalization, after adjusting for sociodemographic factors, medical history, and depression. A total of 569 patients were included (mean [SD] age, 62.6 [9.3] years; 420 men [73.8%]). Flow-mediated vasodilation decreased from a mean (SD) of 4.8% (3.7%) before mental stress to 3.9% (3.6%) after mental stress (a 23% reduction; P < .001), and 360 participants (63.3%) developed transient endothelial dysfunction (a decrease in flow-mediated vasodilation). During a median (interquartile range) follow-up period of 3.0 (2.9-3.1) years, 74 patients experienced a major adverse cardiovascular event. The presence of transient endothelial dysfunction with mental stress was associated with a 78% increase (subdistribution hazard ratio [sHR], 1.78 [95% CI, 1.15-2.76]) in the incidence of major adverse cardiovascular event. Both the δ flow-mediated vasodilation (sHR, 1.15 [95% CI, 1.03-1.27] for each 1% decline) and poststress flow-mediated vasodilation (sHR, 1.14 [95% CI, 1.04-1.24] for each 1% decline) were associated with major adverse cardiovascular event. Risk discrimination statistics demonstrated a significant model improvement after addition of either poststress flow-mediated vasodilation (change in the area under the curve, 0.05 [95% CI, 0.01-0.09]) or prestress plus δ flow-mediated vasodilation (change in the area under the curve, 0.04 [95% CI, 0.00-0.08]) compared with conventional risk factors. In this study, transient endothelial dysfunction with mental stress was associated with adverse cardiovascular outcomes in patients with coronary artery disease. Endothelial responses to stress represent a possible mechanism through which psychological stress may affect outcomes in patients with coronary artery disease.