Assessment of blood pressure (BP) in children and adolescents is based on percentiles according to age, sex and height, with the difference between the 90th (pre-hypertension) and 95th (hypertension) percentiles being only 3–4 mmHg. Although BP is known to be highly variable in children, which may influence the accuracy of BP assessment, the magnitude and direction of BP variability (BPV) over consecutive measurements has never been investigated in this population. We aimed to determine BPV, and factors influencing BPV, over consecutive measurements in a large representative population of Australian children and adolescents.Design and Method:
Two consecutive BP measurements were recorded by oscillometry among 3,047 children (aged 12 [95CI; 12, 13] years, males 52%). A third BP measurement was taken if the absolute difference between the first and second systolic BP (SBP) readings (|ΔSBP|) was ≥10 mmHg. BPV was calculated as the coefficient of variation (CV; [SD/mean SBP] × 100).Results:
From the first to second measurement, SBP decreased in 58%; did not change in 10% and increased in 32% of the population. SBP CV and |ΔSBP| were significantly greater among children with elevated BP compared to children with normal BP (SBP CV; 10.5 mmHg [95CI; 8.4, 12.3] vs 7.8 mmHg [95CI; 7.5, 8.3], p = 0.005 and |ΔSBP|; 11.7 mmHg [95CI; 8.4, 15.0] vs 6.4 mmHg [95CI; 6.0, 6.7]; p = 0.002). The association between |ΔSBP| and initial SBP level was significantly modified by age for both female and male participants (p < 0.0001 and p = 0.0002, respectively; figure), indicating a smaller change in SBP over consecutive measurements for adolescents compared with younger children for any given SBP level.Conclusions:
The magnitude of BPV is greater among children with elevated BP and it is also greater with younger age. These factors may influence the accuracy of BP assessment and imply that the optimal number of BP measurements in children and adolescents still needs to be determined.