The use of balloon dilatation to treat native coarctation of the aorta is gaining acceptance among interventional pediatric cardiologists, but is still controversial. We describe our experience with this procedure in 21 children, mean age 5.6 years and mean weight 21.1 kg. Most had an additional congenital heart defect, most commonly a bicuspid aortic valve. 17 were asymptomatic, 3 had tachypnea and 1 infant had severe congestive heart failure and was ventilated. The mean systolic blood pressure was 129.7 mm Hg. Balloon dilatation was successful in 90% (19), decreasing the mean maximal systolic gradient from 35.3 to 9 mm Hg (p < 0.001), and increasing the narrowest area from 3.9 to 8.2 mm (p < 0.001), with a mean balloon-to-coarctation width-ratio of 2.8. There were no complications. Of 15 who underwent repeat cardiac catheterization at a mean interval of 10.6 months, 2 had a maximal systolic gradient of more than 20 mm Hg. 1 of these underwent successful repeat angioplasty and the other, who also had a small aneurysm, underwent surgical repair successfully. 2 others had small aneurysms and they are being followed clinically. All patients were seen again after a mean interval of 31 months. The mean systolic blood pressure was 104 mm Hg, significantly lower than before intervention (p < 0.002). 1 had an increased pressure gradient between right arm and leg of 35 mm Hg at later follow-up, and repeat cardiac catheterization demonstrated a good result 13 months after the initial procedure. She is awaiting a third catheterization. Overall, 90% had good mid-term results. Based on our experience and recent reports, balloon angioplasty is safe and effective in most children older than 7 months and should be considered a viable alternative to operation for discrete aortic coarctation. Further long-term evaluation is needed.