The anticoagulant effect of heparin during percutaneous transluminal coronary angioplasty was monitored by measurements of the activated clotting time in two studies that compared the effects of a single bolus of heparin with those of a bolus of heparin combined with a continuous infusion of the drug. In a preliminary study 40 patients received a single heparin bolus of 10,000 units (protocol I) and a further 40 patients received both a heparin bolus of 10,000 and a continuous infusion of heparin at a rate of 2000 units per hour (protocol II). During the first 45 minutes, nine patients (23%) in protocol I but only two patients (5%) in protocol II were found to be inadequately anticoagulated. For 24 hours after angioplasty both groups received an infusion of heparin at the rate of 2000 units per hour which led to consistent anticoagulation in 73 (91%) of patients. In a subsequent randomised study, 40 patients received heparin according to either protocol I or II. Protocol II was again found to lead to a higher rate of adequate anticoagulation. During the first 60 minutes 11 patients (55%) in protocol I but only three patients (15%) in protocol II were inadequately anticoagulated. In addition, the activated clotting time of arterial blood in the first 30 minutes was significantly higher than that of venous blood in 70% of the patients. A bolus of heparin (10,000 units) together with an infusion of 2000 units per hour should be routinely given during coronary angioplasty. The effects of heparin, which can vary considerably from patient to patient, should be monitored by the measurement of the activated clotting time of arterial blood.