Restenosis is a major clinical problem following successful percutaneous transluminal coronary angioplasty. Since magnesium has vasodilator and antithrombotic effects, this study was designed to evaluate its potential to decrease the rate of restenosis. In an open-labelled, randomized controlled study, 148 patients underwent successful coronary angioplasty. Ninety-eight patients were treated with 46-52 mmol/18-20 h intravenous magnesium sulphate (groups M1 and M2), and 49 of them continued with oral supplements of magnesium hydroxide 600 mg.day-1 (group M2). The other 50 patients served as controls (group C). Coronary angiography was performed before, immediately after and at 6 months follow-up or earlier if clinically indicated. Clinical, laboratory, ergometric and radionuclide evaluations were also carried out. One hundred and thirty-nine patients (94%) with 163 dilated segments completed the study. Intravenous magnesium was well tolerated. The cross-sectional area at the site of angioplasty increased by 3.55 +/- 2.01 mm2 in groups M1 and M2 compared with an increase of 2.90 +/- 1.63 mm2 in the control group, (P = 0.03). A trend towards a lower rate of restenosis (> 50% reduction in luminal diameter) was noticed in the magnesium groups (28/110, 25%) compared with the control group (20/53, 38%) P = 0.10. Oral administration of magnesium was well tolerated, did not have an additive effect on restenosis, but an improved clinical course was noted. It is concluded that intravenous administration of magnesium in patients undergoing coronary angioplasty is feasible and safe and that the beneficial trend of magnesium to prevent acute recoil and late (within 6 months) restenosis is encouraging and should promote further investigation in a larger patient population.