We studied the effect of intravenous, polyethyleneglycol-treated, human immunoglobulin, administered at 200 mg/kg per day (group A: n = 147; male 86, female 61; age < 1 year, 50) or 400 mg/kg per day (group B: n = 152; male 87, female 65; age < 1 year, 52) for five consecutive days and compared it with freeze-dried, sulfonated human immunoglobulin [group C: n = 152; male 87, female 65; age < 1 year, 51), administered at 200 mg/kg per day for five consecutive days, on the prevention of coronary artery abnormalities in Kawasaki disease. Echocardiograms were interpreted blindly and independently. Proportions of 87.1%, 95.4%, and 82.3% in groups A, B, and C, respectively, had no coronary artery abnormalities. The confidence limits of difference between the proportions of groups A and C, groups B and C, and groups B and A were -4.4% and 10.4%, 7.8% and 15.9%, and 4.0% and 10.8%, respectively. Duration of fever and serum immunoglobulin G (IgG) levels were correlated with the prevalence of coronary artery abnormalities. We concluded that intravenous, polyethyleneglycol-treated, human immunoglobulin and freeze-dried, sulfonated human immunoglobulin had clinically equivalent effects on coronary artery abnormalities, and that five daily doses of 400 mg/kg of intravenous, polyethyleneglycol-treated, human immunoglobulin is more effective than that of 200 mg/kg gamma globulin.