ObjectiveTo assess reference values of ambulatory blood pressure in normotensive children.Subjects and designTwenty-four-hour non-invasive ambulatory blood pressure monitoring (ABPM) was carried out in 241 healthy normotensive children aged from 6 to 16 years (126 boys, mean ± SD age 11.2 ± 2.7 years; 115 girls, mean ± SD age 10.9 ± 2.9 years). The subjects were subdivided into three age-sex groups: 6–9, 10–12 and 13–16 years.SettingPrimary care.Main outcome measuresABPM was performed using an oscillometric device (SpaceLabs model 90207) and appropriate cuff size during a regular school day. Blood pressure was measured every 20 min from 0600 to 2400 h, and thereafter every 30min. At each monitoring session the following parameters were calculated for both systolic (SBP) and diastolic blood pressure (DBP): means and centiles for 24-h, daytime (0800–2200 h) and night-time (2400–0600 h); circadian variability, estimated as the blood pressure fall between the day and the night periods and the day: night ratio; and load, as the percentage of measurements above the age- and sex-specific 95th centile (P95).ResultsThe upper limits of 'normality' for the mean of 24-h SBP and DBP estimated as the P95in each age subgroup were 121/71 and 119/71 mmHg, 123/78 and 120/74 mmHg, and 124/78 and 125/75 mmHg, for boys and girls, respectively. A progressive increase in SBP with age was observed in both sexes, in contrast, DBP was similar throughout the age range. A nocturnal blood pressure fall of approximately 11 mmHg was observed for both SBP and DBP in all subgroups. The day: night ratio was 1.12 and 1.22 for SBP and DBP, respectively. The upper limit of blood pressure load, estimated as the P95in all children, was 39% for SBP and 26% for DBP. A significant positive correlation was observed between casual blood pressure and 24-h ambulatory blood pressure (SBP: r = 0.61, P < 0.0001; DBP: r = 0.31, P < 0.0001). In general, mean ambulatory blood pressure, during the 24-h or the daytime period, was higher than casual blood pressure for both SBP and DBP.ConclusionABPM is feasible in children, and the values obtained are useful as a departure point in establishing reference values.