OBJECTIVE: To investigate the diurnal blood pressure curve in healthy normotensive children. Thirty-one children were re-examined after a median interval of 123 days in order to study the reproducibility of the diurnal profile. SUBJECTS: Twenty-four-hour ambulatory blood pressure monitoring and conventional blood pressure readings were obtained in 228 normotensive children, whose ages ranged from 6 to 16 years and of whom 116 were boys and 112 girls. RESULTS: The conventional blood pressure averaged 99+/-11/57+/-9 mmHg in boys and 98+/-12/56+/-9 mmHg in girls (means+/-SD); the corresponding 24 h pressures were 111+/-7/66+/-5 mmHg and 109+/-7/65+/-5 mmHg, respectively. Of the children, 83% had a significant diurnal blood pressure rhythm for systolic pressure and 89% for diastolic pressure. The nocturnal blood pressure fall was normally distributed, averaging 12.0+/-6.3 mmHg systolic and 14.2+/-5.9 mmHg diastolic. There was no evidence for a bimodal distribution. The amplitude of the diurnal blood pressure curve, determined by the Fourier approach, was positively skewed with a mean of 12.5+/-4.2 mmHg for systolic and 14.0+/-4.1 mmHg for diastolic blood pressure. The daily blood pressure maximum occurred at 1344+/-4 h 46 min for systolic and 1321+/-4 h 22 min for diastolic blood pressure. For systolic blood pressure the cumulative sum (cusum)-derived circadian alteration magnitude was 1.7+/-6.2 mmHg higher in boys than in girls, whereas the cusum plot height was also 7.3+/-27.0 mmHg x h higher in male subjects. The repeatability coefficient (2SD of the difference between paired recordings, expressed as a percentage of nearly maximal variation) was 80% for the conventional systolic pressure and 40% for the conventional diastolic blood pressure. The repeatability coefficients for the ambulatory blood pressure levels varied from 32 to 45% and for the parameters describing the diurnal blood pressure profile from 42 to 78%. CONCLUSION: A significant diurnal blood pressure rhythm is observed in most normotensive children and adolescents. There is no evidence for a bimodal distribution of the nocturnal blood pressure fall. The reproducibility of the parameters of the diurnal blood pressure curve tended to be less than that of the ambulatory blood pressure level. Thus, one 24 h recording is probably insufficient to characterize a child's diurnal blood pressure profile fully.