A comparison of paediatric hypertension clinical practice guidelines and their ability to predict adult hypertension in an African birth cohort

Abstract

It remains unclear which paediatric hypertension clinical practice guideline (CPG) should be applied in an African population. We, therefore, aimed to compare commonly used CPG (2017 AAP, 2016 ESH, 2004 Fourth Report) developed in high-income countries for use in South African children at four paediatric ages (children: 5 years, 8 years; adolescents: 13 years, 17 years) to determine which best predicts elevated blood pressure (BP) in adulthood (22 years, 28 years). Moreover, the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for each specific paediatric CPG was calculated across the age points. The 2017 AAP definition identified more children and adolescents with hypertension when compared to the 2004 Fourth Report and 2016 ESH guidelines. In computed hazards ratios, ages 8 years to 17 years, all three paediatric CPG significantly predicted the risk of elevated BP in young adulthood (p ≤ 0.032). However, sensitivity to predict elevated BP at age 22 years for all CPG was generally low (17.0%–33.0%) with higher specificity (87.4%–93.1%). Sensitivity increased at age 28 years (51.4%–70.1%), while specificity decreased (52.8%–65.1%). Both PPV and NPV at both adult age points varied widely (17.9%–79.9% and 29.3%–92.5% respectively). The performance of these paediatric CPG in terms of AUC were not optimal at both adult age points, however, the 2017 AAP definition at age 17 years met an acceptable level of performance (AUC = 0.71). Our results, therefore, highlight the need for more research to examine if an African-specific CPG would better identify high-risk children to minimise their trajectory towards adult hypertension.