Determination of blood pressure curves in healhty children and comparison of the different curve models

Aim: Hypertension is one of the most important public health problem in worldwide. Hypertension in childhood and adolescence is a significant risk factor for cardiovascular disease in the adult life and it also causes early development of atherosclerosis and end-organ damage in childhood. The aim of this study is to determine the systolic and diastolic BP percentiles using different models aged 2 to18 years in Turkish children and to evaluate the performances of the models. Methods: A cross-sectional study was performed and the probability sampling method was applied. The total sampling size was 3456 children. Of the total 5,417 children at 10 selected schools (nursery, pre-elementary school, primary school, and high school), 4,984 (the reached rate 92%) were evaluated. Weight, height and systolic and diastolic blood pressure of these children were measured by suitable methods. The estimations of systolic and diastolic blood pressure percentiles in children using polynomial regression and quantile regression models were done. Data were analyzed by R program, version 3.0.2 (2013-09-25) and version 2.15. Results: The mean values of systolic and diastolic blood pressure increased with age and hight at both genders. For both genders, polynomial regression analyzes were performed in which the age-10, height Z score, and weight Z score variables were included in the fourth-degree regression model. It was found that age and height variables affected blood pressure and models were statistically significant. It was observed that the estimated values of systolic and diastolic blood pressures for boys were slightly higher than the predicted values for girls. Conclusion: The blood pressure normal values of Turkish children should be determined with samples of wider geographical representation and at certain intervals. Different statistical methods and models should be used to evaluate the results.