Prevalence of high blood pressure and hypertension among children and adolescents in Europe: a systematic review and meta-analysis

Abstract

Objective

This systematic review and meta-analysis aims to estimate the prevalence of high blood pressure (BP) and hypertension among European children and adolescents (1990–present).

Methods

We included population-based cross-sectional and cohort studies reporting BP in children (≤18 years) from European countries, excluding studies from electronic health records or specialist care referrals. PubMed was searched on 27 January 2025. Risk of bias was assessed with the Hoy et al tool, and data were pooled using random-effects meta-analysis. The main outcomes were pooled prevalence of high BP and hypertension.

Results

We analysed 56 studies (n=179 279, mean age, 12 years) from 18 European countries. The prevalence of high BP was 8% (95% CI 7% to 10%), higher in boys (7%, 95% CI 6% to 9%) than in girls (5%, 95% CI 4% to 7%). Among children with obesity, prevalence was 23% (95% CI 18% to 29%) vs 6% (95% CI 4% to 9%) in children without obesity. Prevalence was highest in 10–13 years old (8% (95% CI 5% to 13%) compared with 6% (95% CI 4% to 8%) and 6% (95% CI 3% to 9%) in 3–9 years old and 14–19 years old, respectively. The prevalence of hypertension was 4% (95% CI 3% to 5%). Risk of bias was judged to be moderate or high for most studies, with variability in BP measurement methods.

Conclusions

The prevalence of high BP and hypertension among European children and adolescents is considerable, particularly among those with obesity, where one-fourth has high BP. It is more common in boys and peaks during puberty. These findings should be interpreted with caution due to methodological limitations of the included studies.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Hypertension in childhood and adolescence is associated with cardiovascular disease later in life.

• The prevalence of childhood hypertension in Europe is unknown.

WHAT THIS STUDY ADDS

• High blood pressure (BP) and hypertension among European children and adolescents is substantial, especially among children with obesity, where one out of four has high BP.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Early intervention targeting childhood hypertension, with particular attention to children with obesity, may help prevent vascular ageing and yield long-term positive health outcomes.

• Additionally, these findings support the need for the development of targeted prevention programmes and sustainable healthcare systems for the future adult population.

Introduction

The prevalence of hypertension among children and adolescents is increasing globally, most likely related to rising body mass index (BMI) and obesity.^{1 2} Childhood hypertension has been associated with increased blood pressure (BP) levels later in life, as well as with markers of target organ damage.³ Recently, an international collaboration published 35-year follow-up data of 38 589 children and adolescents, showing for the first time an association between childhood cardiovascular risk factors, including hypertension, and adult cardiovascular events.⁴ Additionally, we previously reported a 10 percentage point risk difference between adolescents with normal BP (systolic BP <120 mm Hg and diastolic BP <80 mm Hg) and combined hypertension (systolic BP ≥140 mm Hg and diastolic BP ≥90 mm Hg) after 50 years of follow-up.⁵ It is reasonable to assume that the increasing prevalence of hypertension in children and adolescents today will significantly contribute to the global cardiovascular disease pandemic of tomorrow.⁶

HyperChildNET was established to bridge the knowledge gap regarding prevention, diagnosis and treatment of high BP in children.⁷ It is an interdisciplinary network emphasising the need for pan-European action focused on hypertension in youth. Comprehensive prevalence estimates are needed to determine the potential impact of hypertension in children and adolescents on long-term cardiovascular health in Europe. No cross-European studies assessing the prevalence of hypertension in children and adolescents are currently available. The diagnostic criterion for hypertension has remained consistent; however, the underlying reference populations and tables have changed across eras and settings,68,11 making comparisons challenging.

The aim of this study was to systematically review the prevalence of high BP and hypertension among children and adolescents in Europe from 1990 onwards in accordance with objective 2.1 of HyperChildNet.¹²

Methods

We performed a systematic review and meta-analysis guided by the recommendations of the Cochrane Collaboration.¹³ A protocol (online supplemental eMethod 1), including eligibility criteria, search strategy, data extraction plan and preliminary analyses, was established before commencing the work. This review was not registered a priori. This report was written according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 guidelines (online supplemental eMethod 2).¹⁴

Eligibility criteria

To be included in this systematic review, studies needed to have a cross-sectional or cohort design with a random or population-based selection of participants from a European country or subregion within a European country. BP recordings had to be performed from 1990 onwards. Studies were limited to participants ≤18 years of age, although a slightly higher age limit (19 years) was allowed if the majority of participants were ≤18 years old or if data for participants ≤18 years old were presented separately. Studies based on electronic health records from usual care or referrals to specialist care were excluded because of the potential risk of selection bias. We used the largest sample sizes available if multiple publications presented data from the same study.

Study selection and searches

A literature search was conducted on 27 January 2025 and restricted to PubMed, as the added value of other databases, including broader coverage of pharmacology and other geographical areas, was not considered relevant to the scope of the review. A combination of search terms was used, related to hypertension, childhood, prevalence and Europe, without language restrictions consistent with recent guidance on minimising language bias in systematic reviews.¹⁵ Additionally, a separate search was conducted to identify previous systematic reviews by using the systematic review filter. The retrieved reviews were manually browsed to identify relevant papers. Full search strategies are presented in online supplemental eMethod 3.

The search results were merged, and duplicates were removed using EndNote V.20 reference management software. Titles and abstracts were screened by GS and HR under the supervision of MB to remove irrelevant studies. Full-text articles were then examined by GS, MB and HR separately against the above eligibility criteria, with the final decision on inclusion based on consensus, including possible re-examination of the original articles.

Data extraction

Data were independently extracted by GS and HR using a specially designed Excel spreadsheet. The following variables were extracted: author name, journal, year of publication, year of examination, study location, sampling method, age, sex, BMI, obesity prevalence, mean systolic and diastolic BP, total number of participants, total number of participants with hypertension, hypertension definition, BP measurement procedure, BP measurement device and obesity definition. The definition of obesity and choice of reference material differed between the studies and are provided in the legend of table 1. The prevalence of high BP and hypertension was estimated as the number of participants meeting the respective criteria divided by the total number of participants in each study. Data were extracted separately for boys and girls, different age categories and non-obese/obese participants, when such data were available. The extracted data were then compared, and any discrepancies were resolved by re-examining the original paper, through discussion and involvement of a third investigator (MB). If relevant information was lacking in the retrieved studies, other publications from the same study were sought.

Table 1. Baseline characteristics of included studies.

Reference	Country	Year of examination	Sample size (n)	Girls n (%)	Hypertension definition	Mean age, years (range)	Obesity prevalence (%)
Aguirrea et al 2012	Spain		157	92 (58.6)	4th report*	(6–11)
Ahammed et al 2021	Albania	2017–2018	2355		JNC7	(15–19)
Antal et al 2004	Hungary	1997–2000	6303	3261 (51.7)	Age 15: ≥135/≥85 mm Hg Age 16–18: ≥140/≥ 90 mm Hg	16.4 (15–18)
Barba et al 2006	Italy	2003–2004	3923	1937 (49.4)	4th report*	8.72 (6–11)	20.9†
Blancas et al 2022	Spain	2018–2019	228	108 (47.4)	ESH 2016*	(6–16)	22.3‡
Botton et al 2007	France	1999	452	217 (48.0)	NHBPEP*	13.6 (8–17)	1.8†
Buonomo et al 1996	Italy	1991	166	80 (48.2)	The second task force 1987§	(3–15)
Cairella et al 2007	Italy	2003–2005	487	234 (48.0)	4th report*	12.7 (11–14)	10.3†
Chiolero et al 2007	Switzerland	2005–2006	5207	2586 (49.7)	4th report*	12.3 (10–15)	1.7†
Chirita-Emandi et al 2013	Romania	2010–2011	3626	2062 (56.9)	4th report*	12.54 (7–18)	11.8¶
Cinteza et al 2013	Romania	2006–2008	4866	2459 (50.5)	4th report*	10.9 (3–17)	6.4¶
Dinç et al 2009	Turkey		1346	580 (43.1)	4th report*	16.1 (15–18)	3.2¶
Drago et al 1994	Italy	1992–1993	861	441 (51.2)	4th report**	10.4 (6–13)	15.0††
Dulskiene et al 2014	Lithuania	2010–2012	7457	3963 (53.1)	4th report*	13.0 (12–15)	2.4†
Furdela et al 2022	Ukraine	2019	540	263 (48.7)	AAP 2017/ESH 2016*	13.83 (10–17)	4.4‡
Genovesi et al 2005	Italy		2416	1210 (50.1)	NHBPEP *	(6–11)
Hughes 2024	United Kingdom	2009–2011	3850	2146 (55.7)	ESC/ESH*	17.7
Jakab et al 2020	Hungary	2005–2018	8624	3905 (45.3)	4th report*	(3–18)	9.2†
Karatzi et al 2017	Greece	2007–2009	2571	1285 (50.0)	4th report*	11.17 (9–13)
Kardas et al 2005	Poland		637	349 (54.8)	NHBPEP*	11.9 (9–14)
Katona et al 2011	Hungary	1999	10 213	5031 (49.3)	4th report**	16.6 (15–18)
Kollias et al 2009	Greece	2004, 2007	1004	524 (52.2)	4th report*	14.2 (12–17)	10.2†
Kollias et al 2012	Greece	1990	166	77 (46.4)	4th report*	(5–12)	2.4†
Kolpa et al 2016	Poland	2014	1093	556 (50.9)	4th report*	11.04 (10–12)	5.0†
Krzywińska-Wiewiorowska et al 2017	Poland	2009–2010	4941	2490 (50.4)	4th report	13.98 (10–18)
Manzoli et al.2005	Italy	2001–2002	15 612	7531 (48.2)	4th report*	9.9 (6–14)	19.5¶
Marcovecchio et al 2016	Italy		564	311 (55.1)	4th report*	8.8 (6–12)	10.0‡‡
Marras et al 2009	Italy	2005	839	441 (52.6)	4th report*	(11–14)
Martin-Espinosa et al 2017	Spain	2013	1604	816 (50.9)	4th report*	5.34 (4–6)	8.4†
Martin-Espinosa et al 2022 (MOVI-da 10)	Spain	2017	362	183 (50.6)	AAP 2017§§	4.81 (4–6)
Martin-Espinosa et al 2022 (MOVI-2)	Spain	2010	1158	571 (49.3)	AAP 2017§§	9.5 (8–11)
Martinovic et al 2017	Montenegro	2012–2013	3254	1616 (49.7)	4th report*	10.0 (7–13)
Mavrakanas et al 2009	Greece	2007	572	285 (49.8)	4th report**	7.3 (4–10)	14.0†
Menghetti et al 2015	Italy		2007	1009 (50.3)	4th report*	(6–17)	6.6†
Mesas 2022	Spain	2021–2022	1261		ESH 2016*	7.1 (6-9)
Morales-Suárez-Varela 2023	Spain		1142	599 (52.5)	AAP 2017§§	7.2 (6–9)
Murgan et al 2013	Germany		121	53 (43.8)	4th report*	16.2 (13–19)
Musil et al 2012	Croatia	2003–2004	965	470 (48.7)	4th report**	14.7
Neuhauser et al 2009	Germany	2003–2006	14 730	7203 (48.9)	JNC7	(3–17)
Nika et al 2019	Greece	2013–2016	2832	1419 (50.1)	ESH 2016* (age 6–15 years) >15år ESH adult cut-off.	11.22(6–18)	8.5†
Nowakowska et al 2005	Poland		2823	1273 (45.1)	4th report**	14.6 (7–19)
Nur et al 2008	Turkey	2004	1020	427 (41.9)	4th report*	15.9 (14–18)	0.2¶
Ostrowska-Nawarycz et al 2007	Poland	2005–2006	25 309	12 669 (50.1)	4th report**	12.9 (7–18)	3.7†
Pecin et al 2013	Croatia		756	381 (50.4)	4th report*	15.9 (15–18)	3.54¶
Perez-Gimeno 2023	10 European cities	2006–2007	868	459 (52.9)	AAP 2017	14.8 (12.5–17.5)
Pileggi et al 2005	Italy	2002–2003	603	319 (52.9)	4th report**	12.7 (6–18)	11.1¶
Pileggi et al 2021	Italy		556	278 (50.0)	4th report**	8.1 (8)	12.4¶¶
Rabbia et al 1994	Italy	1991	1413	698 (49.4)	The second task force 1987§	13.5 (12–15)	21.5***
Ramos et al 2005	Portugal	2003–2004	2 023	1043 (51.6)	4th report*	13 (13)	10.2¶
Rocha et al 2014	Portugal	2009	853	459 (53.8)	>140 and/or >90 mm Hg	16.3 (15–18)	4.9†††
Rodrigues et al 2018	Portugal	2009–2010	1555	785 (50.5)	4th report*	7.58 (6–9)	7.5†
Schwandt et al. 2015	Germany	1993–2008	22 051	10 723 (48.6)	4th report*	9.5 (3–18)	5.5†
Stankute	Lithuania	2019–2020	3668	1740 (47.4)	AAP§§	10.8 (7-17)	5.8†
Steinthorsdottir et al 2011	Iceland	2009	970	489 (50.4)	4th report*	9.6 (9–10)	7.2¶
Stergiou et al 2009	Greece	2004–2005	765	407 (53.2)	4th report*	12.3 (6–18)
Uçar et al 2000	Turkey		4026	2065 (51.3)	The second task force 1987§	12.7 (7–18)	1.5‡‡‡
Wieniawski et al 2021	Poland		690	366 (53.0)	4th report*	15.7 (15–17)	11.3¶¶

^*Stratum 1 in stratified analysis. Definition according to the 4th report, European Society of Hypertension (ESH) 2016, The US National High Blood Pressure Education Programme (NHBPEP) applying international reference tables.

^†Definition according to International Obesity TaskForce.

^‡Definition according to World Health Organisation.

^§Stratum 4 in stratified analysis. Definition according to the 2nd Taskforce of 1987.

^¶Definition according to the Centers for Disease Control.

^**Stratum 2 in stratified analysis. Definition according to the 4th report, European Society of Hypertension (ESH) 2016, The US HBPEPapplying national or modified reference tables.

^††Definition according to Waterlow method.

^‡‡Definition according to Cacciari.

^§§Stratum 3 in stratified analysis. Definition according to the American Academy of Pediatrics (AAP).

^¶¶Definition according to WHO with national reference values.

^***Definition according to Cole- index.

^†††Definition not stated.

^‡‡‡Definition according to the National Health and Nutrition Examination Survey (NHANES).

Risk of bias

The risk of bias was evaluated for each study individually using the risk of bias tool developed by Hoy et al,¹⁶ which was specifically designed to assess the risk of bias in prevalence studies. The details of the tool are described in online supplemental eMethod 4. The risk of bias assessment was performed independently by GS, HR and MB, and the results were compared. Discrepancies were resolved through discussions.

Data synthesis and analysis

Prevalence estimates from individual studies were pooled using the meta esize command in Stata V.18. Before pooling the extracted estimates, the individual variances of each study’s prevalence were stabilised using the Freeman-Tukey double arcsine transformation.¹⁷ The estimates were then transformed back for presentation. To assess the heterogeneity among the included studies, we used the Cochran Q test and I².¹⁸

Owing to the high clinical and methodological diversity in the included studies, a random-effects model was used as the main analysis to calculate the pooled estimates, including 95% CIs. Because random-effects models are susceptible to bias from small study effects, we performed a sensitivity analysis using a fixed-effects model.¹⁹ Publication bias was explored using a funnel plot and Egger’s test.

In descriptive epidemiology, a single measurement is commonly accepted to define hypertension.²⁰ Therefore, our main analysis reports the overall prevalence of BP above guideline-defined limits, based on measurements from one or more occasions. Paediatric guidelines define hypertension as high BP measured across three separate occasions. Accordingly, we use the term high BP for the main analysis and provide a separate analysis reflecting the clinical diagnosis of hypertension.¹⁰

The prevalence of high BP was assessed across different population strata: obesity (obese vs non-obese), sex (boys vs girls), number of BP measurement occasions and cardiovascular disease (CVD) risk regions (low-risk, moderate-risk, high-risk or very high-risk European regions according to Systematic COronary Risk Evaluation 2 (SCORE2)).²¹ Age-specific prevalence was evaluated in the following categories: 3–9, 10–13 and 14–19 years. To be allocated to a category, age had to be reported per year or according to the specified category in the individual study. These categories were chosen based on the available data. Hypertension prevalence was assessed across the same strata, with the exception of CVD risk regions and age-specific categories, due to small subgroup sizes.

To examine the impact of different hypertension definitions on high BP prevalence, separate analyses were performed. The first included studies using the 4th report,⁸ European Society of Hypertension (ESH),⁶ and US National High Blood Pressure Education Programme⁹ definitions with international reference tables for age, sex and height. The second applied the same definitions but with modified reference tables. The third used the American Academy of Pediatrics (AAP) definition,¹⁰ and the fourth applied the 2nd Task Force of the 1987 definition.¹¹ These analyses were not performed for hypertension prevalence due to small subgroup sizes.

To assess potential sources of variation in high BP, separate sensitivity analyses were performed by BP device used (oscillometric vs auscultatory) and risk of bias.

Time trends were assessed using meta-regression with year of BP measurement as the independent variable. To assess whether the individual studies had an excessive impact, a leave-one-out meta-analysis was performed.

Results

A total of 1964 potentially eligible studies were identified through the search strategies (Figure 1). After screening titles and abstracts, 132 full-text studies were assessed. Of these, 76 were excluded, primarily because of insufficient data, non-random sampling or multiple publications from the same study, resulting in 56 studies (online supplemental file 1, e-references)comprising 179 279 participants from 18 countries (online supplemental file 1, e-Fig. 1), analysed. Italy accounted for the highest number of studies (12), encompassing 29 447 participants. The mean age of the participants was 12 years (range 3–19 years).

Figure 1. Flow chart of the inclusion process.

Most studies (33/56; 59%) were published from 2010 onwards. BP was primarily measured in schools, with measurement techniques evenly divided between oscillometric and auscultatory methods. Baseline characteristics of the included studies are presented in table 1. The risk of bias was judged to be moderate to high in most studies (online supplemental eTables 1 and 2).

Pooled and stratified prevalence of high BP and childhood hypertension

The pooled overall prevalence of high BP and hypertension among European children aged 3–19 years was 8% (95% CI 7% to 10%) and 4% (95% CI 3% to 5%), respectively, although the analyses were subject to substantial heterogeneity, as outlined in Figure 2A,B.

Figure 2. (A) Pooled overall prevalence of high blood pressure. (B) Pooled overall prevalence of hypertension.

Of the 56 studies, 41 (73%) reported separate data for boys and girls. The pooled prevalence of high BP was 5% (95% CI 4% to 7%) in girls and 7% (95% CI 6% to 9%) in boys (Figure 3A and online supplemental eFigures 2 and 3). The overall prevalence was higher due to some studies not disaggregating data by sex. The overall prevalence of hypertension was 3% (95% CI %2 to 5%) and 5% (95% CI 3% to 6%) in girls and boys, respectively (Figure 3A and online supplemental eFigures 4 and 5).

Figure 3. (A) Pooled overall prevalence of high blood pressure by subgroup. (B) Pooled overall prevalence of hypertension by subgroup.

High BP stratified by obesity was reported in 21 studies, while hypertension stratified by obesity was reported in eight studies. The pooled prevalence of high BP was 23% (95% CI 18% to 29%) among children with obesity and 6% (95% CI 4% to 9%) among children without obesity (Figure 3A and online supplemental eFigures 6 and 7). The corresponding analysis for hypertension showed a pooled prevalence of 18% (95% CI 12% to 24%) and 3% (95% CI 2% to 5%) for children with and without obesity, respectively (Figure 3B and online supplemental eFigures 8 and 9).

The pooled prevalence of high BP was 6% (95% CI 4% to 8%) in 3–9 years old (13 studies included), 8% (95% CI 5% to 13%) in 10–13 years old (11 studies included) and 6% (95% CI 3% to 9%) in 14–19 years old (14 studies included) (Figure 3A and online supplemental eFigures 10-12).

The pooled prevalence of high BP in children decreased with repeated measurements: 10% (95% CI 8% to 13%) on one occasion, 7% (95% CI 3% to 14%) on two occasions and 4% (95% CI 3% to 5%) on three or more occasions (Figure 3A and online supplemental eFigures 13-15). The study used 19 different BP measurement procedures, most commonly: two readings on one occasion (mean value), three readings on one occasion (mean value) or repeated measurements after 2–3 weeks if initial readings were elevated (online supplemental eTable 1).

The prevalence of high BP varied by definition: the 4th Report, ESH and the US National High Blood Pressure Education Programme with international reference tables yielded a pooled prevalence of 9% (95% CI 7% to 11%). Applying the same definitions but with national reference tables resulted in a pooled prevalence of 5% (95% CI 3% to 6%). The AAP and the 1987 2nd Task Force definitions showed pooled prevalences of 8% (95% CI 4% to 12%) and 10% (95% CI 3% to 20%), respectively (Figure 3A and online supplemental eFigures 16-19).

When stratified by SCORE2 risk regions, the pooled prevalence of high BP was 6% (95% CI 4% to 10%) in low-risk countries, 10% (95% CI 7% to 13%) in moderate-risk countries, 5% (95% CI 4% to 7%) in high-risk countries and 10% (95% CI 5% to 16%) in very high-risk countries (Figure 3A and online supplemental eFigures 20-23). By device, prevalence was 8% (95% CI 6% to 10%) with the auscultatory method and 9% (95% CI 6% to 12%) with the oscillometric method (Figure 3A and online supplemental eFigures 24 and 25).

We found no association between year of measurement and prevalence of high BP (online supplemental eFigure 26). The leave-one-out sensitivity analysis for the main analysis showed that no single study had an excessive effect on the pooled prevalence of high BP (online supplemental eFigure 27). Sensitivity analyses using a fixed-effects model showed a pooled prevalence of high BP of 7% (95% CI 7% to 7%) and 8% (95% CI 6% to 11%) in studies with low to moderate risk of bias (online supplemental eFigure 28 and 29).

No obvious publication bias was observed on visual inspection of the funnel plot (online supplemental eFigure 30) or by Egger’s test (p=0.0843), although these results should be interpreted with caution due to substantial heterogeneity.

Discussion

In this systematic review and meta-analysis of 56 studies, including 179 279 children, the pooled prevalence of high BP was 8% among European children, affecting one in four children with obesity.

Consistent with previous research,22,25 obesity was strongly associated with high BP in children, highlighting the need to measure BP in children with overweight or obesity, and supporting current guidelines that target obesity to manage hypertension.^{6 10} Early interventions addressing hypertension, especially among obese children, may prevent vascular ageing, thereby improving long-term cardiovascular outcomes.² School-based, community-based and parent-based interventions targeting lifestyle modifications exist, with combined approaches appearing most effective.^{26 27} Additionally, these findings can inform future cardiovascular prevention programmes and sustainable healthcare planning for the future adult population.

The prevalence of hypertension in European children in our study aligns with the global meta-analysis by Song et al.¹ The contrast between high BP at a single time point and hypertension defined across three time points clearly illustrates how the number of measurements affects the estimated prevalence.²⁸ Repeated measurements thus seem important to avoid overdiagnosis. However, we recently demonstrated that even a single hypertensive BP measurement at age 18 years is associated with an increased risk of future cardiovascular event, highlighting the prognostic importance of single elevated recordings.⁵

The large sample size of the present study allowed assessment of age-specific and sex-specific prevalences of high BP in European children. Prevalence peaked among 10–13 years old, after which a decline was observed, consistent with previous observations of a hypertension peak at puberty.²⁹ High BP and hypertension were more common in boys than in girls, aligning with prior literature.^{24 29 30} A secular trend towards earlier puberty in girls has been documented, and a similar, but less certain, trend in boys is observed. These shifts are multifactorial, with nutritional status and adiposity strongly influencing pubertal timing via mechanisms including elevated serum leptin levels.^{31 32} Consequently, it remains unclear whether the noted increase in prevalence is ‘true’ or a mismatch between puberty onset in contemporary children and the outdated reference standards, emphasising the need for updated normative data. While the physiological links between puberty and BP are not fully understood, sex hormones and growth hormone are thought to play key roles.^{29 33}

In line with previous systematic reviews of childhood hypertension in different parts of the world, we found substantial heterogeneity.34,37 This reflects the diversity in hypertension definitions, number of BP measurement occasions and procedures of BP measurement used across studies. Overall, 19 different diagnostic procedures were identified with variability in the use of highest, lowest or mean reading even when three measurements were obtained at the same occasion. Such methodological heterogeneity may introduce detection bias, underscoring the need for standardisation and adherence to current guidelines. Nevertheless, sensitivity analysis limited to studies with low to moderate risk of bias yielded a similar prevalence estimate, although with a slightly wider CI, reflecting the reduced sample size.

The choice of reference tables for sex, height and weight also influences hypertension prevalence. Using the definition of the 4th Report, ESH and the US National High Blood Pressure Education Programme, applying international reference tables, yielded a higher prevalence of high BP compared with the group using the same definition but national or regional reference tables. This is in line with previous findings³⁸ and is likely due to international reference tables not fully accounting for ethnicity and geographic variability.³⁹ The AAP definition had the second lowest prevalence, which is unexpected given that the AAP reference population is non-obese.¹⁰ This finding is likely due to the small number of trials in this stratum and should be interpreted with caution. The absence of globally unified BP reference tables is a hindrance to international comparisons.⁴⁰

The analysis stratified by risk region did not reveal a clear pattern. This may reflect differences in the number of included studies and participants, as well as inherent characteristics of the studies themselves. Additionally, SCORE2 includes variables such as smoking, which are uncommon in children, and may therefore not accurately capture demographic differences in this population.

This systematic review and meta-analysis is the first to address the prevalence of high BP and hypertension in European children using random or population-based samples. The extensive coverage across Europe and the inclusion of studies in any language enhances the generalisability of the results. Some limitations of the study should be noted, however. No association was found between year of measurement and the prevalence of hypertension, which may be influenced by ecological bias due to the comparison of studies with different intrinsic characteristics. Additionally, the predominance of recent studies limits the power to detect temporal differences. Some studies were excluded from the stratified analyses because they did not report stratified data, which should be considered when interpreting these findings. Lastly, factors that may contribute to childhood hypertension, such as a sedentary lifestyle, elevated dietary sodium intake and low physical activity, were not addressed due to limited reporting in the included literature. Future research would benefit from including these variables to broaden our understanding of childhood hypertension.

Conclusion

This systematic review and meta-analysis demonstrates that high BP and hypertension among European children is considerable. Prevalence is higher among boys than girls and peaks during puberty. One-fourth of children with obesity have high BP, highlighting the need to measure BP in this group. Given the high risk of bias and diverse methodology among included studies, further high-quality epidemiological research using harmonised definitions and standardised BP measurement protocols is needed.

Data availability statement

Data are available upon reasonable request.