Atrial natriuretic factor (ANF) release is modulated by several haemodynamic factors, including ventricular and atrial wall stretch. Dipyridamole infusion, which is commonly used as a pharmacological stressor in patients with coronary artery disease, can acutely increase ventricular and atrial pressure via myocardial ischaemia. The aim of this study was to assess whether dipyridamole infusion (up to 0.84 mg kg-1 over 10') can affect ANF release in man. Nineteen patients (13 men, 6 women) with a history of chest pain were studied. Their drug regimen was interrupted and instead they were administered a dipyridamole infusion, combined with two-dimensional echocardiography and 12-lead ECG monitoring. Plasma ANF was measured by RIA while the patients rested, and after dipyridamole infusion. Eight patients had no evidence of myocardial ischaemia, as measured by electrocardiographic and/or echocardiographic criteria, during dipyridamole infusion: among them, ANF values were similar while they were at rest and at peak dipyridamole administration (23.9 +/- 9.5 vs 23.4 +/- 6.9 pg ml-1, P = ns). Eleven patients had dipyridamole-induced transient ischaemia (regional ventricular dyssynergy and/or ST segment depression): among them, ANF values rose significantly at peak dipyridamole administration (31.8 +/- 13.8 vs 65.5 +/- 36.4, P less than 0.01). We conclude that dipyridamole infusion does not increase ANF release in man in the absence of ischaemia. The induction of myocardial ischaemia acutely increases ANF release, probably through a rise in ventricular and atrial wall stress.