A total of 184 children aged, 13 months to 11 years, suffering from their first attack of steroid-responsive nephrotic syndrome were included in a randomized study. They were treated according to three treatment protocols. All children received 1-2 mg of prednisone/kg body weight/day (up to 80 mg daily) for 4 weeks, and thereafter 1 mg/kg body weight/48 h for the next 4 weeks. Treatment was discontinued at this point in 44 children (protocol A); in 68 (protocol B) the dose was reduced by 25% each week, tapering off to 0 at the end of the third month, while in 72 children (protocol C), after the first 2 months of initial treatment the dose was reduced by 25% each month and tapered off to 0 by the end of the sixth month. All patients completed a 2-year follow-up period after withdrawal of prednisone. Treatment results were expressed as: percentage of children relapse-free within the first 6 months and 2 years after withdrawal of treatment, and average number of relapses per patient per year. The best results were obtained in children who had been treated for 6 months; 65.3% of them remained relapse-free within the first 6 months and 50% over the entire 2-year follow-up period; the number of relapses per patient per year in this group was 0.49. The respective values for children treated 2 and 3 months were: 36.4% and 32.4% for the 6-month period; 27.3% and 20.6% for the 2-year period; the numbers of relapses per patient per year were 0.79 and 0.77, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)