Assessment of the Performance of Obesity Measures in the Screening of Hypertension in a South African Adolescent Population

Abstract

Obesity is known to be one of the most significant risk factors for essential hypertension in childhood. However, whether obesity cut-offs may predict hypertension screening in adolescents remains controversial. This study investigated the performance of obesity cut-off values for the screening of hypertension in a South African adolescent population. In this cross-sectional study, 1144 adolescents aged between 11 and 17 years were recruited from the Eastern Cape Province of South Africa. Anthropometric and blood pressure (BP) parameters including diastolic blood pressure (DBP), systolic blood pressure (SBP), and heart rate (HR) were measured. Assessment of the sensitivity and specific of obesity cut-off values in predicting hypertension was performed using receiver operating characteristic (ROC) analysis. Obesity was positively associated (p < 0.01) with hypertension. Obese individuals, based on the predicted obesity using BMI percentile cut-off (pBMI85.2%), were more likely to develop hypertension (odds ratio: 2.070; p < 0.001) than their counterparts based on the observed obesity cut-off (pBMI95%) (Odd ratio: 1.748 p = 0.004). The area under the curve (AUC) of BMI percentile and WHtR for screening SBP percentile, and DBP percentile and HR as per ROC analysis, was low (<0.65). Equally, the sensitivity and specificity were low (<0.6) for all BP measures (SBP, DBP, and HR). Furthermore, the cut-off values for blood pressure measures, as established by ROC analysis using anthropometric measures, were far below the recommended cut-off values for hypertension screening. The obesity cut-offs for BMI percentile and WHtR established in this populations showed poor performance in diagnosing hypertension even though they were strong predictors of hypertension.