We tested the hypothesis that the oral alpha1-adrenergic agonist, midodrine, would limit the fall in arterial pressure observed during exercise in patients with pure autonomic failure (PAF). Fourteen subjects with PAF underwent a stand test, incremental supine cycling exercise (25, 50, and 75 W), and ischemic calf exercise, before (control) and 1 h after ingesting 10 mg midodrine. Heart rate (ECG), beat-to-beat blood pressure (MAP, arterial catheter), cardiac output (Q, open-circuit acetylene breathing), forearm blood flow (FBF, Doppler ultrasound), and calf blood flow (CBF, venous occlusion plethysmography) were measured. The fall in MAP after standing for 2 min was similar ( approximately 60 mmHg; P = 0.62). Supine MAP immediately before cycling was greater after midodrine (124 +/- 6 vs 117 +/- 6 mmHg; P < 0.03), but cycling caused a workload-dependent hypotension (P < 0.001), whereas increases in Q were modest but similar. Midodrine increased MAP and total peripheral resistance (TPR) during exercise (P < 0.04), but the exercise-induced fall in MAP and TPR were similar during control and midodrine (P = 0.27 and 0.14). FBF during cycling was not significantly reduced by midodrine (P > 0.2). By contrast, recovery of MAP after cycling was faster (P < 0.04) after midodrine ( approximately 25 mmHg higher after 5 min). Ischemic calf exercise evoked similar peak CBF in both trials, but midodrine reduced the hyperemic response over 5 min of recovery (P < 0.02). We conclude midodrine improves blood pressure and TPR during exercise and dramatically improves the recovery of MAP after exercise.