OBJECTIVE: In the light of the reported inconsistent anti-ischaemic and antianginal effects of transdermal glyceryl trinitrate, its efficacy and influence on the effects of intracoronary glyceryl trinitrate were examined during coronary angioplasty, which provides a model of controlled, reversible ischaemia. DESIGN: Double blind, randomised study of the effect of transdermal and intracoronary glyceryl trinitrate on ischaemia during coronary angioplasty. PATIENTS: 40 patients with isolated severe stenosis of the left anterior descending coronary artery. INTERVENTIONS: Patients were randomised (double blind) to transdermal glyceryl trinitrate (10 mg per day) and placebo, starting four to six hours before angioplasty. After 4 one-minute balloon inflations intracoronary glyceryl trinitrate was injected (0.2 mg) and then 4 further one-minute inflations were performed. MAIN OUTCOME MEASURES: The time to angina and the time to > 0.2 mV ST shift on surface electrocardiogram (ECG) or intracoronary ECG during the individual inflations. RESULTS: These times did not significantly differ during initial inflations between transdermal glyceryl trinitrate (27 (11), 25 (9), and 19 (9) s, respectively) and placebo (34 (11), 30 (8), and 21 (7) s. After intracoronary glyceryl trinitrate, they were significantly prolonged compared with the initial values, without differences between patients with transdermal glyceryl trinitrate (37 (10), 30 (8), and 23 (8) s, respectively) or placebo (39 (15), 36 (11), and 28 (12) s). Ischaemic preconditioning was not seen. CONCLUSIONS: Transdermal glyceryl trinitrate (10 mg per day), unlike intracoronary glyceryl trinitrate, did not alleviate the myocardial ischaemia produced by balloon inflation during coronary angioplasty.