The use of fourth phase Korotkoff sound (K4) versus fifth phase Korotkoff sound (K5) for the determination of diastolic blood pressure (DBP) has been a subject of controversy since the indirect method of determining arterial blood pressure was described. Using data from the Bogalusa Heart Study, we evaluated the differences between K4 and K5 (K4-K5) from 4633 subjects 5 to 30 years of age examined between 1987 and 1991. The overall mean difference between K4 and K5 was 9.9 +/- 5.6 mm Hg (mean +/- SD). The average difference was highest in 5 to 8 year olds, where it measured 12.3 +/- 5.5 mm Hg. The average K4-K5 difference fell with increasing age and reached a value of 6.3 +/- 2.6 mm Hg by 25 years of age. For all race/sex groups, the youngest two age groups differed statistically from the oldest age groups in K4-K5 difference (P < .006). There were significant differences between blacks and whites (P < .015) and between men and women (P < .001) for subjects between 13 and 17 years of age. Additional analyses were performed with individuals having K5 = 0 added to yield an expanded population of 5117 persons. Overall, 9.5% had at least one of six measurements of K5 = 0 and most were young subjects: 27% of children 5 to 8 years and 13% of children 9 to 12 years. We conclude that by age 15 there may be no relevant clinical difference in K4-K5. However, in children, K4 and K5 should be recorded and K4 is a more reproducible measure of diastolic blood pressure.