Impacts of measurement protocols on blood pressure tracking from childhood into adulthood - A metaregression analysis

The best approach for blood pressure (BP) measurement in children remains controversial, specifically regarding the choice of Korotkoff phase 4 versus Korotkoff phase 5 for diastolic BP (DBP) and the use of automated devices. To examine the impacts of different BP measurement protocols on BP tracking from childhood into adulthood, we conducted a meta-analysis of 50 related studies published between 1970 and 2006 identified based on a systematic search of PubMed. These studies provided 617 data points (tracking correlation coefficient, our outcome variable) for systolic BP and 547 data points for DBP for our meta-analysis. The explanatory variables included the use of Korotkoff phase 4/Korotkoff phase 5, BP device, and number of BP measurements per visit. Analyses were adjusted for potential confounders, including sex, baseline age, follow-up length, publication year, and study country. Tracking correlation coefficients for DBP measured using Korotkoff phase 4 was higher than that of Korotkoff phase 5 by 0.035 but not significant. DBP tracking assessed by automated device was higher than that of Korotkoff phase 5 by 0.152 (P = 0.024) and higher than the mercury manometer by 0.223 (P = 0.005). BP tracking was slightly higher with multiple BP measurements per visit, but measurements of >= 3 times did not improve the tracking further compared with 2 measurements. Although policy-making bodies currently recommend the use of Korotkoff phase 5 to assess DBP in children, our metaregression analysis did not support the recommendation. In general, Korotkoff phase 4 seems to be different from Korotkoff phase 5, and automated device is a promising approach for BP assessment in childhood.