The prevalence of elevated blood pressure in a sample of slovene children and adolescents: a pilot study

Elevated blood pressure (BP) in children and adolescents is becoming a public health concern. The increasing prevalence of arterial hypertension (HTN) is observed worldwide (1, 2) and relates to the epidemic of obesity (3, 4), among other factors.

According to the 2016 European Society of Hypertension guidelines (5), unlike the previously used guidelines (6), high BP is defined differently in children and adolescents aged ≥16 years. Normal BP in children is defined as a systolic and diastolic BP, which is less than the 90th percentile, adjusted for gender, age, and height. Children with an average systolic blood pressure (SBP) or diastolic blood pressure (DBP) in the 90th or higher percentile, but less than the 95th percentile, are prehypertensive. HTN in children is defined as an average SBP and/or DBP, which is in the 95th or higher percentile, measured on at least three different occasions and confirmed by auscultatory measurement. The first stage of HTN is defined as the 95th to 99th percentile BP plus 5 mmHg and the second stage of hypertension as BP greater than 99th percentile plus 5 mmHg. Isolated systolic hypertension (ISH) is defined by SBP ≥95th percentile and DBP <90th percentile (5).

In adolescents aged ≥16 years, normotension is defined as a BP <130/85 mmHg. With repeated measures, a BP of 130-139/85-89 mmHg defines prehypertension, and a BP value ≥140/90 mmHg defines HTN. The first stage of HTN in older adolescents is defined as BP 140-159/90-99 mmHg, and second stage as BP 160-179/100-109 mmHg. ISH is defined as BP ≥140/<90 mmHg (5).

Most studies report the prevalence of prehypertension around 9–12% (4). On single BP measurement, an American study (7) estimated the prevalence of HTN (here described as ‘elevated BP’ if <3 measurements were carried out with readings in the hypertensive range) in adolescents at approximately 10%, the same applies to singlemeasurement prevalence of prehypertensive BP. After three BP measurements, the prevalence of HTN decreased to 3.2%, while the prevalence of prehypertension increased to 15.7%, as some of the previous hypertensive individuals dropped to the prehypertensive group. In a Swiss study (8), the prevalence of prehypertension was 13.3% and elevated BP 11.4% at the first measurement, and after the second, 9.5% and 9.4%, respectively. At the third measurement, the prevalence of prehypertension was 2.2% and HTN 1.7%. Those results are similar to a Slovene paediatric study (1978–1990), which estimated the prevalence of HTN at 1.2% (9, 10).

Since then, there has been a lack of research on this topic in Slovenia. With this pilot study, we want to contribute new data on the estimated prevalence of prehypertensive BP and elevated BP. Additionally, we also search for some anthropometric features that correlate with elevated BP, such as increased body mass index (BMI).

The only published study on the prevalence of HTN in children in Slovenia between 1978 and 1990 estimated the prevalence at 1.2% (9, 10). In foreign literature, the prevalence of elevated BP ranges from 9%–19% on a single measurement, and the prevalence of prehypertension 10%–30%, which is higher than our results, with 3.9% prevalence of prehypertensive BP and 7.1% of elevated BP in hypertensive range (7, 8, 19).

If we compare the results of the prevalence from regular check-ups (12.0%) and the results obtained with the auscultatory method (7.1%), we can see that they differ to a high extent. This could be explained with findings from research by Park et al. (20), where the values of BPs measured with automatic-oscillometric device were 10 mmHg higher in systolic BPs and 5 mmHg higher in diastolic BPs, compared to the auscultatory method. Differences between settings of out-patient clinics in oscillometric measurements and primary school cabinets in auscultatory measurements could have led to some differences in BP values, further potentiated by the white-coat effect in the clinical setting. We also observed relatively large proportions of elevated BP readings in the young child group, aged 2–6 years, which were seen by primary care paediatricians. The reason might be a positive selection of patients, in which a BP measurement was indicated and not measured only as part of a screening protocol.

We believe that our estimate of HTN prevalence of about 7% obtained with auscultatory measurement at one occasion is the most accurate. Based on these findings and the assumption that the prevalence of HTN is at least halved after the second and third measurement (7, 21), we presume the actual prevalence of HTN in Slovenian paediatric population is around 3–4%, concordant with results from recent foreign literature (20).

In the second part of our analysis, we have seen that BMI is positively correlated to BP height, which is consistent with findings from several other studies (3, 4, 21). Surprisingly, the relative risk for HTN was not statistically increased with an increase in waist-to-hip ratio, as it would be expected (16). Foreign studies (16, 17, 22) also report that an increased waist-to-height ratio strongly correlates with a high BP, which was not observed in our study, due to unknown reasons, and should be further elucidated.

A major methodological issue was the difference between the auscultatory and oscillometric cohorts, which prevented us from analysing the two groups together. The most important limitation of our study were measurements of BP on a single occasion, often with a single measurement if below 90th percentile, and not at three, as it is expected (5). Single measurement studies (7, 8, 19) overestimate the prevalence of HTN. In at least two studies, the prevalence of arterial hypertension was more than halved after completion of the second and third measurements (7, 21). A limitation of our sample was also the absence of measurements of certain parameters which could correlate with HTN, such as family history of hypertension, race, birth weight, quality of diet, drug therapy and physical activity.

Our methodology risked a selection bias due to opportunistic sampling of children in primary schools. This was reduced by supplementing it with data from regular check-ups. Observer bias was reduced by double-checking the measured elevated BP by another examiner.

The estimated prevalence of elevated BP in hypertensive range in Slovenian paediatric population, measured at a single occasion, is around 7.1%, and prehypertensive BP approximately 3.9%, which is consistent with recent foreign studies. Moreover, a positive correlation between blood pressure and body mass index was shown. In other words, normal blood pressure appeared less often in overweight and obese children and adolescents. These findings of a pilot study give a reason to conduct a representative, larger-scale study on the prevalence of elevated BP and obesity in Slovene paediatric population, and, thereby, gather insight into the burden of a disease gaining epidemic proportions worldwide.