Voluntary apnea during dynamic exercise evokes marked bradycardia, peripheral vasoconstriction, and pressor responses. However, the mechanism(s) underlying the cardiovascular responses seen during apnea in exercising humans is unknown. We therefore tested the hypothesis that the muscle metaboreflex contributes to the apnea-induced pressor response during dynamic exercise. Thirteen healthy subjects participated in apnea and control trials. In both trials, subjects performed a two-legged dynamic knee extension exercise at a workload that elicited heart rates at ~100 beats/min. In the apnea trial, after reaching a steady state, subjects began voluntary apnea. Immediately after cessation of the apnea, arterial occlusion was initiated at both thighs and the subjects stopped exercising. The occlusion was sustained for 3 min in the postexercise period. In the control trial, the occlusion was started without subjects performing the apnea. The apnea induced marked bradycardia, pressor responses, and decreases in arterial O2 saturation, cardiac output, and total vascular conductance. In addition, arterial blood pressure was significantly higher and total vascular conductance was significantly lower in the apnea trials than the control trials throughout the occlusion period. In separate sessions, we measured apnea-induced changes in exercising leg blood flow in the same subjects. Leg blood flow was significantly reduced by apnea and reached the resting level at the peak of the apnea response. We conclude that the muscle metaboreflex is activated by the decrease in O2 delivery to the working muscle during apnea in exercising humans and contributes to the large pressor response. NEW & NOTEWORTHY We demonstrated that apnea during dynamic exercise activates the muscle metaboreflex in humans. This result indicates that a reduction in O2 delivery to working muscle triggers the muscle metaboreflex during apnea. Activation of the muscle metaboreflex is one of the mechanisms underlying the marked apnea-induced pressor response.