Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: Systematic review

Abstract

Objective:
To review the literature on studies that estimated the prevalence of high blood pressure (HBP) or systemic arterial hypertension (SAH) in Brazilian adolescents, considering the employed methodological procedures.

Methods:
Bibliographical research of prevalence studies of HBP/SAH in adolescents from 1995 to 2010. The search was conducted in the electronic databases PubMed/Medline, Lilacs, SciELO, and Isi Adolec. The descriptors "hypertension", "BP", "teen", "students", "cross-sectional", "prevalence" and "Brazil" were used in Portuguese and English. Furthermore, a score ranging from 0 to 18 based on Recommendations for Blood Pressure Measurement in Humans and Experimental Animals and the VI Brazilian Guidelines of Hypertension was elaborated, in order to analyze the procedures used to measure BP in studies.

Results:
Twenty-one articles were identified, mostly published in the last 10 years, and 90.5% were performed in school-based and regions of the Southeast, Northeast and South. The prevalence of HBP/SAH ranged from 2.5 to 30.9%. The score of the studies ranged from 0 to 16. A significant negative correlation (rho = -0.504; p = 0.020) was observed between the prevalence of HBP/SAH and the score of BP measurement quality.

Conclusion:
The great variability of PAE/SAH estimates appears to be influenced by methodological procedures used in the studies.

Full text

849 Rev Bras Epidemiol
2013; 16(4): 849-59
Prevalence of high blood
pressure in Brazilian adolescents
and quality of the employed
methodological procedures:
systematic review
Prevalência de pressão arterial elevada
em adolescentes brasileiros e qualidade
dos procedimentos metodológicos
empregados: revisão sistemática
Marina Gabriella Pereira de Andrada MagalhãesI
Luciano Machado Ferreira Tenório de OliveiraII
Diego Giulliano Destro ChristofaroIII
Raphael Mendes Ritti-DiasI,II
IPostgraduation Program of Adolescent Medicine, Universidade de Pernambuco
(UPE) – Recife (PE), Brazil.
IIAssociate Postgraduation Program of Physical Education, Universidade de
Pernambuco and Universidade Federal da Paraíba – Recife (PE), Brazil.
IIIPrograma de Pós-graduação em Fisioterapia, Universidade Estadual Paulista,
Presidente Prudente-SP, Brasil. Programa de Pós-graduação em Ciências da
Motricidade, Rio Claro-SP, Brasil.
Corresponding author: Raphael Mendes Ritti Dias. Rua Arnóbio Marques 310, Santo Amaro,
CEP: 50100-130, Recife, PE, Brasil. E-mail: raphaelritti@gmail.com
Conict of interests: nothing to declare.
Abstract
Objective: To review the literature on stud-
ies that estimated the prevalence of high
blood pressure (HBP) or systemic arterial
hypertension (SAH) in Brazilian adoles-
cents, considering the employed method-
ological procedures. Methods:Bibliographical
research of prevalence studies of HBP/SAH
in adolescents from 1995 to 2010. e search
was conducted in the electronic databases
PubMed/Medline, Lilacs, SciELO, and Isi
Adolec. e descriptors “hypertension”, “BP”,
“teen”, “students”, “cross-sectional”, “preva-
lence” and “Brazil” were used in Portuguese
and English. Furthermore, a score ranging
from 0 to 18 based on Recommendations for
Blood Pressure Measurement in Humans and
Experimental Animals and the VI Brazilian
Guidelines of Hypertension was elaborated, in
order to analyze the procedures used tomea-
sure BP in studies. Results:Twenty-one
articles were identified, mostly published
in the last 10 years, and 90.5% were per-
formed in school-based and regions of the
Southeast,Northeast and South. e preva-
lence of HBP/SAH ranged from 2.5 to 30.9%.
e score of the studies ranged from 0 to 16. A
signicant negative correlation (rho=-0.504;
p = 0.020) was observed between the preva-
lence of HBP/SAH and the score of BP mea-
surement quality. Conclusion:e great vari-
ability of PAE/SAH estimates appears to be
inuenced by methodological procedures
used in the studies.
Keywords: Adolescent. Hypertension. Arterial
pressure. Students. Review. Brazil.

850 Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
Introduction
Cardiovascular diseases are currently
responsible for approximately 40% of the
world mortality1. Systemic arterial hyperten-
sion (SAH) represents the main risk factor for
cardiovascular diseases, and its early diagnosis
has been pointed out as an important public
health strategy.
In Brazil, studies indicate prevalence
of SAH ranging from 22 to 44% in the adult
population2. Even though most of the SAH
diagnoses are established at the adult age,
it is known that this disease can appear dur-
ing childhood3. erefore, measuring blood
pressure (BP) has been recommended by the
VI Brazilian Guidelines of Hypertension4, as
well as by the Second Task Force on Blood
Pressure Control in Children, since 19875, as
an important component of pediatric routine6,
thus enabling its early diagnosis.
Throughout the past years, several stud-
ies have investigated the prevalence of
SAH among children and adolescents in
Brazil; however, the prevalence estimates
provided by the studies present wide varia-
tion. For instance, in the study conducted
by Rezendeetal.7, the prevalence was 2.4%,
while in the analysis by Sakamoto et al.8, the
prevalence was 30.9%. These variable results
concerning the estimates can bearesult
of different methodological procedures,
especially with regard to the procedures
adopted to measure BP. Therefore,it is nec-
essaryto synthetize the studies in order to
better understand the data that are available
in literature, which can subsidize public
policies of prevention and treatment of SAH
among Brazilian adolescents.
us, the objective of this study was to
systematically review the studies that estimate
the prevalence of high BP in Brazilian adoles-
cents, according to sample characteristics and
the employed methodological procedures.
Methods
A bibliographic research of descriptive epi-
demiological studies analyzing the prevalence
Resumo
Objetivo: Revisar a literatura sobre estudos que
estimaram a prevalência de pressão arterial
elevada (PAE) ou hipertensão arterial sistê-
mica (HAS) em adolescentes brasileiros, con-
siderando os procedimentos metodológicos
empregados. Métodos: Pesquisa bibliográ-
ca de estudos de prevalência de PAE/HAS
em adolescentes de 1995 a 2010. A busca
foi realizada nas bases de dados eletrônicos
PubMed/Medline, Lilacs, SciELO, Isi e Adolec.
Foram utilizados os descritores: “hiperten-
são”, “pressão arterial”, “adolescente”, “estu-
dantes”, “estudos transversais”, “prevalência”
e “Brasil”, nas línguas portuguesa e inglesa.
Além disso, foi elaborado um escore, base-
ado nas Recommendations for Blood Pressure
Measurement in Humans and Experimental
Animals e nas VI Diretrizes Brasileiras de
Hipertensão, para a análise dos procedimen-
tos utilizados para medida da PA nos estudos
variando de 0 a 18. Resultados:Foram iden-
ticados 21 artigos, a maioria publicada nos
últimos 10 anos, sendo 90,5% realizados em
base escolar e nas regiões sudeste, nordeste
e sul do país. As prevalências de PAE/HAS
variaram de 2,5 a 30,9%. A pontuação dosestu-
dos variou de 0 a 16. Foi observada uma cor-
relação negativa signicante (rho=-0,504;
p = 0,020) entre a prevalência de PAE/HAS
e o escore da qualidade da medida da PA.
Conclusão: A grande variabilidade das esti-
mativas da PAE/HAS parece ser inuenciada
pelos procedimentos metodológicos utiliza-
dos nos estudos.
Palavras-chave: Adolescente. Hipertensão.
Pressão arterial. Estudantes. Revisão. Brasil.

851
Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
of high BP in adolescents until2010 was
conducted. e research was performed in
the electronic data bases PubMed/Medline
(NationalLibrary of Medicine),Lilacs
(Literatura Latino-Americana e do Caribe
em Ciências da Saúde), SciELO (Scientific
Electronic Library Online), Isi Web of
Knowledge and Adolec. Descriptors were
selected based on consultations to MeSH
(Medical Subject Headings) and DeCS
(Descritores em Ciências da Saúde). e fol-
lowing descriptors were considered, in English
and in Portuguese: “hypertension”, “blood
pressure”, “teen”, “students”, “cross-sectional”,
“prevalence” and “Brazil” (in order to locate
articles with national samples). Besides, the
logical operators “and” and “or” were used to
combine the terms. All of the search, selection
and assessment processes were performed by
pairs, who fully and independently analyzed
al of the studies that met the inclusion criteria.
e results obtained by the evaluators were
compared in order to check for the concor-
dance between pairs. In case of divergences,
the article was revised again by a third member
of the research team.
From the bibliographic research, 479pub-
lications were identified (PubMed=262;
Lilacs=83; SciELO = 13; Isi = 28 and,
Adolec=93). Only studies investigating ado-
lescents were included (10 to 19yearsold)9,
even those which approached other age
groups andthat collected data in Brazil; review
articles, theses, dissertations, and mono-
graphs were not included, as well asstudies
with specic population, such as people with
asthma, obese participants or children of
hypertensive patients.
After reading the titles of the articles,
92remained for the analysis of abstracts.
From this analysis, the articles that met the
inclusion criteria were fully read by the two
evaluators, who fully and independently ana-
lyzed all of the studies that met the inclusion
criteria. e results obtained by the evaluators
were compared in order to verify concordance
between pairs. In the case of divergences, the
article was revisedby a third researcher. It was
also observed if studies had proper sample
size, involving random selection, besides
the mention of ethical aspects. At the end
of these procedures, 19manuscripts were
identied. After reading themanuscripts,
two other studies cited inthe reference list
that had not been identied in the research
were identied10,11, therefore being included
afterwards, accounting for 21 articles, as
presented in Figure 1.
For the analysis of articles, the following
aspects were observed: year and region in
which the study was carried out, age group,
sample size, type of survey, criteria used
to diagnose high BP and/or SAH, criteria
employed to measure BP (placement, previ-
ous rest and calibration), type of used device
(aneroid, mercury column or electronic),
dimension of the sphygmomanometer cu,
number of measurements, the choice of the
value used to determine the prevalence of
high BP and/or SAH by gender and total
prevalence of high BP and/or SAH.
In order to assess the adopted meth-
odology to verify BP in children and ado-
lescents, a score was elaborated based on
the Recommendations for Blood Pressure
Measurement in Humans and Experimental
Animals12 and in the VI Brazilian Guidelinesof
Hypertension4 (Chart 1), whose scores ranged
from one to three points according to the level
of importance of recommendations. Based
on the established criteria, the maximum
possible score was 18 points, when all of the
items of the recommendations were fullled.
For statistical analysis, the software SPSS
for Windows was used, version 10.0. Normality
and homogeneity of data variance were ana-
lyzed, and data did not show normal distribu-
tion. erefore, Spearman’s correlation was
employed to analyze the correlation between
the prevalence found and the obtained scores,
using a 5% signicance level.
Results
Twenty-one articles published between
1998 and 2010 were identied. Table 1 demon-
strates general information about the studies,
the recommendations used to measure BP, the
used terminology and the observed prevalence.
Concerning terminology, 55.1%7,8,13-22 of the

852 Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
articles used the descriptor SAH, 38.1%10-11,23-28
used the term HBP (high blood pressure) and
4.8%29 referred to borderline BPE.Concerning
the criteria used to classify high BP, 4.8%10 did not
provide information as to the adopted criteria,
4.8%13 used the BP criterion 140x90mmHG,
23.8%8,17,24,26,28 adopted the 90 percentile, and
66.7%7,11,14-16,18-23,25,27,29 employed the 95percentile.
All of the articles included male and
female adolescents, however, only 52%
presented high BP prevalence stratified by
gender7,8,13-15,18,20-22,25,28. Out of these, 18.1%14,20
did not present significant differences of high
BP between men and women, 36.4%8,13,15,21
pointed out to the higher prevalence among
male participants and in 45.5%7,18,22,25,28
the female gender had more prevalence
of high BP.
e compiled studies were conducted
in the Southeast (38%)7,13-15,19-21,29, Northeast
(28.5%)11,16,17,22-23,25, South (24%)8,10,24,26,27 and
Center-West regions (9.5%)18,28. No stud-
ies were veried in the North region. e
sample size of the studies ranged from 558
to 3,169 individuals18. Only one analysis was
regional16, and the others were restricted
Stage 1
Stage 2
MEDLINE
262 articles
19 included
articles
21 included
articles
2 articles mentioned in
the references
LILACS
83 articles
ISI
23 articles
ADOLEC
93 articles
SCIELO
13 articles
Unrelated themes (n = 217)
Review articles (n = 7)
Theses (n = 3)
Repeated articles (n = 46)
Specic population (n = 38)
Studies with adults (n = 54)
Studies with children (n = 12)
Did not presente prevalence data (n = 38)
Figure 1 - Selection Flowchart of studies included in the review.
Figura 1 - Fluxograma da seleção dos estudos incluídos na revisão.

853
Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
to municipal areas. Out of the 21 articles,
90% were performed by school-based sur-
veys7,8,10,11,13,14,16-20,22-26,28-30 and 10% by house-
hold survey15,27. In school-based surveys, the
prevalence ranged from 2.57 to 30.9%8; as to
household surveys, prevalence ranging from
12.327 to 17.9%15 was described.
Table 2 presents a summary of the methods
used in the study. e auscultatory method
was used in 61.9%7,10,11,13,16,18-20,22-26 of the stud-
ies; the oscillometric method was used in
14.3%8,14,21, and 23.8% did not report the used
method15,17,27-29. e calibration of the instru-
ment used to measure BP was reported in
19% of the cases7,19,24,26. As to the number of
evaluators to measure BP, 23.8% of the stud-
ies reported a single person7,8,11,22,26 and 4.8%
indicated that measurement was conducted
by 4 evaluators, who were subjected to audio-
metric evaluation.
Out of the 21 studies, 43% did not specify
the use of cus with dierent sizes according
to the circumference of the arm of the partici-
pant8,10,13-15,17,23,28,29. Rest prior to BP measure-
ment was described in 48% of the articles, and
rest lasted from 218 to 30 minutes13.
With regard to the number of BP measure-
ments on the same day, 9.5% of the studies
did not detail the collection logistics10,17, one
study reported one single measurement23,
Chart 1 - Scores to evaluate the method used to measure blood pressure based on the VI Brazilian Guidelines on Hypertension
(2010) and the Recommendations for Blood Pressure Measurement in Humans and Experimental Animals (2005).
Quadro 1 - Pontuação para avaliação do método utilizado na medida da pressão arterial baseada nas na VI Diretrizes Brasileiras
de Hipertensão (2010) e nas Recommendations for Blood Pressure Measurement in Humans and Experimental Animals (2005).
Assessed item Established by the guidelines Score
Conditions prior to
measurements
Not having practiced physical exercise for at least 60 minutes, not having
ingested alcoholic drink, coee, foods and drinks in the 30 minutes prior to
measurement and having an empty bladder during measurement.
1
Placement of the arm during
measurement
The arm should be at the point of the heart, with the hand palm turned up
and elbow slightly exed. 1
Placement of the evaluated
person during measurement
Sitting position, legs not crossed, feet on the oor and back leaning on
the chair. 1
Previous rest At least 5 minutes. 2
Size of the cus Obtain the approximate circumference of the mid-arm.
After measurement, select the adequate cu to the arm circumference. 2
Type of device Mercury column, aneroid or electronic calibrated and validated
phygmomanometers. 1
Number of measurements
According to the Recommendations for Blood Pressure Measurement in Humans
and Experimental Animals, there should be at least two measurements. The VI
Brazilian Guidelines of Hypertension indicate at least three measurements.
2
Interval between measurements At least 1 minute. 1
Used nal value
Mean of the last two measurements should be used to represent BP. If there
is a dierence of 4 mmHg between the rst and the second measurement,
an additional one should be taken and, afterwards, the mean of these
multiple measurements should be used.
2
Diagnostic criteria
Normoteniono: BP < 90 percetile, since it is lower than 120/80 mmHg;
3
Borderline or “pre-hypertension”: between 90 and 95 percntiles or BP
higher than 120/80 mmHg;
High BP or hypertension: BP ≥ 95 percentile.
Diagnostic criteria for SAH BP measurement on dierent days 3
Total score 18

854 Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
Table 1 - Characteristics of articles, referred recommendations, diagnostic criteria and prevalence of high blood pressure
in Brazilian adolescents from 1995 to 2010.
Tabela 1 - Características dos artigos, recomendações referidas, critérios para o diagnóstico e prevalências de pressão arterial
elevada em adolescentes brasileiros no período de 1995 a 2010.
Article Place nAge
group
Recommendations
referred*
Terminology
(Diagnostic criteria)
Prevalence of
SAH or HBP by
gender
Total prevalence
Costa and Sichieri15
Rev Bras Epidemiol, 1998.
Rio de Janeiro
(RJ) 646 12 - 19 1SAH (P ≥ 95) M - 60.9%
F - 39.1% SAH – 17.9%
Oliveira et al.19
J Pediatr (Rio J), 1999.
Belo Horizonte
(MG) 1005 6 - 18 2#Borderline BP (≥ 90 P < 95)
SAH (P ≥ 95) NI SAH – 3.9%
Rezende et al.7
Arq Bras Cardiol, 2003.
Barbacena
(MG) 611 7 - 14 2SAH (P ≥ 95)
3rd measurement:
M - 26.7%
F - 73.3%
SAH
1st measurement:16.6%
2nd measurement : 4.6%
3rd measurement : 2.5%
Almeida et al.13
J Bras Nefrol, 2003. Sorocaba (SP) 633 15 -25 3SAH (≥140x90 mmHg) M - 75.6%
F - 24 .4% SAH - 14.2%
Moura et al.25
J Pediatr (Rio J), 2004. Maceió (AL) 1253 7 - 17 4, 2## SAH (P ≥ 95) M - 44.1%
F - 55.9% HBP - 9.4%
Silva et al.22
Arq Bras Cardiol, 2005. Maceió (AL) 1253 7 - 17 4, 2## SAH (P ≥ 95) M - 48.5%
F - 51.5% SAH - 7.7%
Mendes et al.17
Rev Bras Saúde
Mater Infant, 2006.
Recife (PE) 421 14 - 19 2## SAH (P > 90) NI SAH - 11.4%
da Silva and Rosa27
Arch Latinoam Nutr, 2006.
Porto Alegre
(RS) 706 6 - 17 5, 2## Borderline BP (≥ 90 P < 95)
SAH (P ≥ 95) NI HBP -12.3%
Ribeiro et al.29
Arq Bras Cardiol, 2006.
Belo Horizonte
(MG) 1450 6 -18 5Borderline BP (≥ 90 P < 95)
SAH (P ≥ 95) NI Borderline BP- 12%
SAH - ?
Monego and Jardim18
Arq Bras Cardiol, 2006. Goiânia (GO) 3.169 7- 14 NI Borderline BP (≥ 90 P < 95)
SAH (P ≥ 95)
M - 43.7%
F - 56.3% SAH - 5%
Sakamoto et al.8
Cien Cuid Saude, 2007. Maringá (PR) 55 15 - 17 6SAH
(SAH or DBP P > 90)
M - 94.0%
F - 6.0% SAH - 30.9%
Rosa et al.21
Arq Bras Cardiol, 2007. Niterói (RJ) 456 12 - 17 3SAH (P ≥ 95) M - 57.1%
F - 42.9% SAH - 4.6%
Romanzini et al.26
Cad Saúde Pública, 2008. Londrina (PR) 644 15 - 18 6Pre-hypertension (P ≥ 90 ou
≥ 120/80 mmHg) NI HBP - 18.6%
Leite et al.24
Fisioter Mov, 2009. Curitiba (PR) 270 10 - 16 NI SAH (SBP and/or DBP with
P ≥ 90 or BP ≥ 120/80 mmHg) NI HBP - 10.7%
Freitas et al.11
Rev Bras Enferm, 2010. Fortaleza (CE) 307 12 - 17 NI Borderline BP
(≥ 90 P < 95) high BP (P ≥ 95) NI HBP - 11.7%
Vieira et al.28
Acta Paul Enferm, 2009. Cuiabá (MT) 329 11-14 7High BP (P ≥ 90) M - 48.7%
F - 51.3% HBP -11.2%
Rodrigues et al.20
J Pediatr (Rio J), 2009. Vitória (ES) 380 10 - 14 3Borderline BP (≥ 90 P< 95)
SAH (P ≥ 95)
M - 50%
F - 50% SAH - 3.4%
Cândido et al.14
Eur J Pediatr, 2009.
Ouro Preto
(MG) 780 6 - 14 7Pre-hypertension (≥ 90 P < 95)
SAH (P ≥95)
M - 50%
F - 50% SAH - 2.7%
Gomes and Alves23
Cad Saúde Pública, 2009. Recife (PE) 1878 14 - 20 7
SAH in < 17 years old (P ≥ 95)
or 18 – 20 years old
(≥ 140/90 mmHg)
NI HBP - 17.3%
Cimadon, Geremia and
Pellanda10
Arq Bras Cardiol, 2010.
Bento
Gonçalves
(RS)
590 9 -18 8NI NI HBP - 11.1%
Griz et al.16
Arq Bras Endocrinol
Metabol, 2010.
Região
Metropolitana
de Recife (PE)
1824 14 - 20 9
SAH in < 17 years old (P > 95)
or 18 – 20 years old
(≥ 140/90 mmHg)
NI SAH - 16.9%
*Recomendations referred: 1: Fifth Report of The Joint National Committee; 2: Task Force (#from 1987; ##from 1996); 3: IV Diretrizes Brasileiras de
Hipertensão Arterial; 4: III Consenso Brasileiro de HAS (1998); 5: American Heart Association; 6: National High Blood Pressure Education Program; 7: V
Diretrizes Brasileiras de Hipertensão Arterial; 8: I Diretriz de Prevenção da Aterosclerose na Infância e na Adolescência; 9: Seventh Report of the Joint
National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure.
PA: Blood pressure; HAS: Hypertension; PAE: High blood pressure; NI: Not informed.
*Recomendações referidas: 1: Fifth Report of The Joint National Committee; 2: Task Force (#de 1987; ##de 1996); 3: IV Diretrizes Brasileiras de Hipertensão Arterial; 4:
III Consenso Brasileiro de HAS (1998); 5: American Heart Association; 6: National High Blood Pressure Education Program; 7: V Diretrizes Brasileiras de Hipertensão
Arterial; 8: I Diretriz de Prevenção da Aterosclerose na Infância e na Adolescência; 9: Seventh Report of the Joint National Committee on Prevention, Detection,
Evaluation and Treatment of High Blood Pressure.
BP: Pressão arterial; SAH: Hipertensão arterial sistêmica; HBP: Pressão arterial elevada; NI: Não Informado.

855
Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
Table 2 - Methods used to diagnose high blood pressure among Brazilian adolescents from 1995 to 2010.
Tabela 2 - Métodos utilizados para o diagnóstico de pressão arterial elevada em adolescentes brasileiros no período de 1995 a 2010.
Article Size of the cus Previous
rest Placement during measurement Type of device N. of measurements/Interval
between measurements Used value S core
Costa and
Sichieri15 NI 15 min. Sitting. Without smoking, eating
or drinking for at least 15 min. NI *
2 measurements, and a third one
in case the dierence was higher
than 5 mmHg/10 min. interval.
Mean of 2 meausrements.
The rst measurement was
ignored, in the case of
3 measurements.
10
Oliveira et al.19 25x12 cm and 18x9 cm 10 min. Sitting. Right arm at the height of
the precordium
Calibrated
mercury
column
2 measurements/10 min. interval Mean of 2 measurements 15
Rezende et al.7Dierent sizes 3 to 5
min.
Sitting. Right arm at the height of
the precordium
Calibrated
aneroid
3 measurements on 3 dierent
days/ 2 min. interval Lowest value 16
Almeida et al.13
26x13 cm. When arm
circumference is inferior to
25 cm or superior to 32 cm,
values were corrected by
the Maxwell formula.
30 min. NI Mercury
column
3 measurements/1 min. interval.
When DBP or SBP diered in
6mmHg or more, 3 new measu-
rements were taken
Mean of 3 measurements 10
Moura et al.25 25x12 cm and 18x9 cm NI NI Mercury
column 2 measurements/2 min. interval Measurement presenting
P ≥ 95 11
Silva et al.22 Dierent sizes NI NI Mercury
column 2 measurements Mean of 2 measurements 13
Mendes et al.17 NI NI NI NI NI NI 0
da Silva and
Rosa27 Dierent sizes NI
Sitting. Right arm at the
heart level. Interval of at least
30minutes. After exercise or meal.
NI* 2 measurements/1 min. interval Mean of 2 measurements 13
Ribeiro et al.29 NI NI NI NI 2 measurements Mean of 2 measurements 7
Monego and
Jardim18 Dierent sizes 2 min. Sitting Aneroid 2 meausrements/ 2 min. interval 1st measurement ignored,
2nd measurement used 12
Sakamoto et al.8 NI 5 min. Sitting. Right arm at the
heart level. Electronic 2 measurements NI 6
Rosa et al.21 Dierent sizes NI NI Electronic 3 measurements on 2 dierent
days/1 min. interval Mean of 6 measurements 14
Romanzini et al.26 Dierent sizes 5 min. Sitting Calibrated
aneroid
2 measurements/ 3 min.
interval. There could be the third
measurement in case there was
a 4 mmHg dierence between
measurements.
Mean of 2 or 3 measurements 11
Leite et al.24 Dierent sizes 5 min. Sitting. Right arm at the
heart level.
Calibrated
mercury
column
3 measurements/ 2 min. interval Mean of 2 last measurements 12
Freitas et al.11 12x10 cm and 23x17 cm 5 min.
Sitting, uncrossed legs, feet
on the oor, back leaning on
the chair, empty bladder. Arm
at the heart level, hand palm
turned up and elbow exed.
The non-use of drugs or stimu-
lating food was observed.
Aneroid 3 measurements /1 min. interval Mean between the 2 last
measurements 16
Vieira et al.28 NI NI NI NI 2 measurements/5 min. interval NI 3
Rodrigues et al.20 Adequate size NI NI Mercury
column 3 measurements /2 min. interval Mean of 3 measurements 11
Cândido et al.14 NI NI Sitting. Left arm at the heart level Electronic
3 measurements/10 min.
When pressure was above the
90percentile, it was ratied with
the auscultatory method
Mean of 3 measurements 11
Gomes and
Alves23 NI NI Sitting. Auscultatory
method 1 measurement Measured value 5
Cimadon,
Geremia and
Pellanda10
NI NI NI Aneroid NI NI 1
Griz et al.16 Adequate size NI Sitting Mercury
column
2 measurements/1 to 2 min.
interval NI 10
*These articles report having used equipment brands Sunmark and WANROSS, respectively, but did not describe the type of equipment; BP: Blood pressure;
SAH: Hypertension; HBP: High blood pressure; NI: Not informed.
*Estes artigos referem ter usado aparelhos das marcas SunMark e WANROSS, respectivamente, porém não descrevem o tipo de aparelho; BP: Pressão arterial;
SAH:Hipertensão arterial sistêmica; HBP: Pressão arterial elevada; NI: Não informado.

856 Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
35
30
25
20
15
10
5
0
0 3 6 9 12 15
18
Obtained score
SAH or HBP (%)
r = -0.504
p = 0.020
and the others mentioned two or three mea-
surements. Studies that measured BP con-
secutively observed that the prevalence of
HBP tends to decrease with the increasing
number of measurements7,19,21.
Concerning the choice of the final BP
value, 48% used the mean of the obtained
measurements13-15,19-22,26,27,29, 24% did not
inform the criteria used to choose the nal
value8,10,16,17,28, 14% ignored the rst measure-
ment, therefore using the mean between the
two others11,24 or the second measurement,
when only two were taken18, and 10% used
the mean presenting percentile ≥ 9525 in the
table of BP percentile or the lowest value
found in measurements7.
e scores of the studies ranged from 0 to
16 points. By dividing the articles in tertiles,
with regard to scores, the prevalence found
in the rst tertile (from 0 to 6 points) ranged
from 11.110 to 30.9%8, while in the last tertile
(12 to 18 points), prevalence ranged from 2.57
and 12.3%27. Figure 2 represents the relation
between the aforementioned scores and the
prevalence found in the articles used in this
study. A signicant negative correlation was
observed (r = -0.504; p = 0.020) between the
prevalence of HBP/SAH and the score of BP
measurement quality.
Discussion
With this review, it was possible to iden-
tify that prevalence studies of HBP in ado-
lescents were mostly conducted in the
Southeast, Northeast and South regions
of the country. Only two studies were con-
ducted in the Center-West region, and
none was identified in the North region.
Results showed that the prevalence of HBP
ranged between regions. For instance, the
highest prevalence of HBP identified in the
Center-West region was of 11.7%, while in
the Southeast and Northeast regions they
were of approximately 17%, and in the South
region, 30.9%. Likewise, there were differ-
ences concerning the lowest prevalence
identified in the regions. These results
suggest that the prevalence of HBP can be
influenced by the region where the teenager
lives, therefore it is important that further
studies try to identify the causes of such
inter-regional variability.
The mean values of HBP identified in
Brazil are close to those reported in interna-
tional studies. In a study conducted in China,
prevalence was of approximately 20%31. A study
with adolescents in Mexico also observed
prevalence of HBP in adolescents, of 20.6%32.
Kollias et al.33 studied adolescents aged 12 to
17 years old in Greece and found that 22.9% of
these teenagers had HBP. However, it is worth
to mention that, similarly to national studies,
the comparison with international studies is
also dicult due to the dierent employed
methodological procedures.
The results in this study indicated that
the age group of the samples in the stud-
ies ranged from 6 to 20 years old. Since the
prevalence of HBP tends to increase with
age34, the number of individuals at differ-
ent age groups is a factor that contributes
with the different prevalence described by
the studies34. Out of the analyzed studies,
28.6%14,18,19,23,25,26 demonstrated prevalence
stratified by age group, and most of them
showed that the percentage distribution
Figure 2 - Relationship between the preva-
lence of high blood pressure in Brazilian
adolescents from 1995 to 2010 and the scores
obtained in relation to the used methodolo-
gical processes (r = -0.504; p = 0.020).
Figura 2 - Relação entre as prevalências de
pressão arterial elevada em adolescentes brasi-
leiros no período de 1995 a 2010 e os escores
obtidos em relação aos processos metodológicos
utilizados (r = -0,504; p = 0,020).
SAH: Hypertension; HBP: High blood pressure.
SAH: Hipertensão arterial sistêmica; HBP: Pressão arterial elevada.

857
Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
of HBP as to age group increased with
age14,19,23,25,26. However, in the study by
Monego et al.18, which included children
and adolescents (7 to 14 years old), such a
relationship was not shown, once much of
the HBP prevalence was concentrated in
the age group of children (7 to 10 years old).
A study conducted by Gomes et al.23, who
worked only with the teenage population
(14 to 20 years old), showed the increasing
percentile with age, and found values of
14.9% in the age group of 14 to 15 years old,
to 18.8% in the age group of 18 to 20 years
old. Such relationship, which is similar to
the one found by Romanzini et al.26, who also
worked with teenagers, is possibly caused by
hormone changes and alterations in the body
composition observed in this stage of life.
Most studies opted for the school-basedsur-
veys. This methodology has been adopted
since it provides easy access to this popula-
tion.eresults of the study indicated that in
school-based surveys, the prevalence ranged
from 2.5 to 30.9%; as to household surveys,
prevalence ranged from 12.3 to 17.9%, indicating
more homogeneous data in householdstud-
ies. The wider variabilityin school-based
studies can be caused by theinuence of fac-
tors related to school onthe students. In fact,
factors that are known to have an impact on
blood pressure levels, such as climate, access
to food and spaces available for the practice of
physical activities in the school inuence all
ofthe students. Since the studies on this subject
include dierent schools, such inuence ends
up leading to variable results. On the other
hand, in household surveys, the inuence of
school is minimized, since there are students
from more schools.
With this review, it was possible to identify
that the criteria used to determine the high
levels of blood pressure were dierent between
studies. In fact, while some studies used the
percentile higher than 90 as a cuto point to
dene HBP8,17,24,26,28,29, others used the percen-
tile for people aged more than 18 years old15,19,
as opposed to the recommendations of the VI
Brazilian Guidelines of Hypertension, which
indicates that the pressure of an individual
aged from 1 to 17 years old is considered to
be normal if percentile values are under 90,
and also lower than 120/80 mmHg; between
the 90 and 95 percentiles, it can be consid-
ered as borderline or pre-hypertension; if it
is equal to or higher than 95, it can be con-
sidered as hypertension35, according to the
protocol used in 66.7% of the studies in this
research7,11,14-16,18-23,25,27,29.
According to the VI Brazilian Guidelines
of Hypertension4, in order to diagnose SAH
several procedures must be conducted. e
results in this study indicated a signicantly
negative correlation between the score of BP
measurement quality and the prevalence of HBP
or SAH. ese results suggest thatthe preva-
lence of HBP or SAH observed in thestudies
may have been overestimated because of the
non-realization of the recommended proce-
dures to measure BP.
Among the factors that apparently have
more influence on BP data, the number of
measurements seems to be very important.
In fact, while studies that used three mea-
surements presented their highest preva-
lence as 14.2%13, those that used two or less
measurements identified prevalence of up
to 30.9%8. One of the studies in this review36
observed prevalence of HBP of 8.7% in the
first measurement and 2.3% in the third BP
measurement. Another variable that has
major influence on the estimates of HBP is
the number of days in which the measure-
ment was taken. Rezende et al.7 obtained
prevalence of 16.6; 4.6; and 2;5%for BP
measurements on the first, second and
third days, respectively. Therefore, it is
important to be careful when analyzing
studies that took only onemeasurement
onthe same day, or several measure-
mentson one day, since these two factors
are directly related to the overestimation
of HBP or SAH prevalence.
e results of the studies included in this
review pointed out to the presence of high BP in
childhood and adolescence. In practice, these
results suggest that health managers and pro-
fessionals should elaborate health promotion
actions that can be employed at early ages.

858 Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
References
1. Yusuf S, Reddy S, Ôunpuu S, Anand S. Global burden of
cardiovascular diseases. Circulation 2001; 104(23): 2855-64.
2. Sociedade Brasileira de Cardiologia. IV Diretrizes Brasileiras de
Hipertensão Arterial. Arq Bras Cardiol 2004; 82(Suppl IV): 1-14.
3. Shear CL, Burke GL, Freedman DS, Berenson GS. Value of
childhood blood pressure measurements and family history
in predicting future blood pressure status: results from 8
years of follow-up in the Bogalusa Heart Study. Pediatrics
1986; 77(6): 862-9.
4. Sociedade Brasileira de Cardiologia, Sociedade Brasileira de
Hipertensão, Sociedade Brasileira de Nefrologia. VI Diretrizes
Brasileiras de Hipertensão Arterial. Arq Bras Cardiol 2010. p. 1-51.
5. Horan MJ. Report of the second Task Force on Blood Pressure
Control in Children―1987. Task Force on Blood Pressure
Control in Children. National Heart, Lung, and Blood
Institute, Bethesda, Maryland. Pediatrics 1987; 79(1): 1-25.
6. Sinaiko AR. Hypertension in children. N Engl J Med 1996;
335(26): 1968-73.
7. Rezende DF, Scarpelli RAB, Souza GF, Costa JO, Scarpelli
AM, Scarpelli PA, et al. Prevalence of systemic hypertension
in students aged 7 to 14 years in the municipality of
Barbacena, in the State of Minas Gerais, in 1999. Arq Bras
Cardiol 2003; 81(4): 381-6.
8. Sakamoto FY, Marcon SS, Oliveira AAB, Nardo Junior N.
Relação da hipertensão, sobrepeso e aptidão física em
estudantes do ensino médio, Maringá-PR. Ciên Cuid Saúde
2007; 6(3): 285-90.
9. World Health Organization. Young People´s Health – a
Challenge for Society. Report of a WHO Study Group on
Young People and Health for All. Technical Report Series
731. Geneva: WHO. 1986.
10. Cimadon HMS, Geremia R, Pellanda LC. Hábitos
alimentares e fatores de risco para aterosclerose em
estudantes de Bento Gonçalves (RS). Arq Bras Cardiol 2010;
95(2): 166-72.
11. Freitas RW, Silva AR, Araújo MF, Marinho NB, Damasceno
MM, Oliveira MR. Prática de atividade física por adolescentes
de Fortaleza, CE, Brasil. Rev Bras Enferm 2010; 63(3): 410-5.
12. Pickering TG, Hall JE, Appel LJ, Falkaner BE, Graves J, Hill MN,
et al. Recommendations for blood pressure measurement in
humans and experimental animals: part 1: blood pressure
measurement in humans: a statement for professionals from
the Subcommittee of Professional and Public Education of the
American Heart Association Council on High Blood Pressure
Research. Circulation 2005; 111(5): 697-716.
13. Almeida FA, Yoshizumi AM, Mota AC, Fernandes, APM,
Gushi AC, Nakamoto AYK, et al. Distribuição dos valores
pressóricos e prevalência de hipertensão arterial em jovens
de escolas do ensino médio em Sorocaba, SP. J Bras Nefrol
2003; 25(4): 179-86.
14. Cândido A, Benedetto R, Castro AP, Carmo JS, Nicolato
RL, Nascimento-Neto RM, et al. Cardiovascular risk factors
in children and adolescents living in an urban area of
Southeast of Brazil: Ouro Preto Study. Eur J Pediatr 2009;
168(11): 1373-82.
15. Costa RS, Sichieri R. Relação entre sobrepeso, adiposidade
e distribuição de gordura com a pressão arterial de
adolescentes no município do Rio de Janeiro. Rev Bras
Epidemiol 1998; 1(3): 268-279.
16. Griz LHM, Viégas M, Barros M, Griz AL, Freese E, Bandeira
F. Prevalence of central obesity in a large sample of
adolescents from public schools in Recife, Brazil. Arq Bras
Endocrinol Metabol 2010; 54(7): 607-11.
One limitation of this study was the use
of a score to classify the quality of the stud-
ies, and such a score was not previously vali-
dated. Even though the items used to create
the score were established according to the
protocols to measure BP recommended in
literature, it is possible that, despite being
logical, the weights given to each item may
have presented bias.
Conclusion
Based on this review, it was observed
that the great variation related to the preva-
lence of HBP can be at least partly attrib-
uted to the quality of the procedures used
to measure BP.
Besides, regional divergences should
be minimized, with stimulation to con-
duct studies in the Center-West and North
regions. Another important observation is
that the results should be stratified by age
group, once the evidence demonstrates dif-
ferences in the prevalence of HBP between
children and adolescents. Besides, house-
hold studies should be stimulated in order
to know the risk factors related to children
and adolescents outside of school, thus pro-
viding indications so that possible public
policies and health promotion initiatives
can be elaborated for this population as a
whole; therefore, it could contribute with
the prevention of this cardiovascular risk
factor in adult life.

859
Prevalence of high blood pressure in Brazilian adolescents and quality of the employed methodological procedures: systematic review
Magalhães, M.G.P.A. et al.
Rev Bras Epidemiol
2013; 16(4): 849-59
17. Mendes MJFL, Alves JGB, Alves AV, Siqueira PP, Freire
EFC. Associação de fatores de risco para doenças
cardiovasculares em adolescentes e seus pais. Rev Bras
Saude Mater Infant 2006; 6(Suppl.1): s49-s54.
18. Monego ET, Jardim PCBV. Determinantes de risco para
doenças cardiovasculares em escolares. Arq Bras Cardiol
2006; 87(1): 37-45.
19. Oliveira RG, Lamounier JA, Oliveira AD, Castro MD, Oliveira
JS. Pressão arterial em escolares e adolescentes: O estudo de
Belo Horizonte. J Pediatr (Rio J) 1999; 75(4): 256-66.
20. Rodrigues AN, Perez AJ, Pires JG, Carletti L, Araújo MT,
Moyses MR. Fatores de risco cardiovasculares, suas
associações e presença de síndrome metabólica em
adolescentes. J Pediatr (Rio J) 2009; 85(1): 55-60.
21. Rosa ML, Mesquita ET, da Rocha ER, Fonseca V de M.
Índice de massa corporal e circunferência da cintura como
marcadores de hipertensão arterial em adolescentes. Arq
Bras Cardiol 2007; 88(5): 573-8.
22. Silva MA, Rivera IR, Ferraz MR, Pinheiro AJ, Alves SW, Moura
AA, et al. Prevalência de Fatores de risco cardiovascular
em crianças e adolescentes da rede de ensino da cidade de
Maceió. Arq Bras Cardiol 2005; 84(5): 387-92.
23. Gomes Bda M, Alves JG. Prevalência de hipertensão arterial e
fatores associados em estudantes de Ensino Médio de escolas
públicas da Região Metropolitana do Recife, Pernambuco,
Brasil, 2006. Cad Saude Publica 2009; 25(2): 375-81.
24. Leite N, Moser DC , Góes SM, Cieslak F, Milano GE, Stefanello
JMF. Medidas hipertensivas e excesso de peso em escolares da
rede pública de Curitiba-PR. Fisioter Mov 2009; 22(4): 477-87.
25. Moura AA, Silva MAM, Ferraz MRMT, Rivera IR. Prevalência
de pressão arterial elevada em escolares e adolescentes de
Maceió. J Pediatr (Rio J) 2004; 80(1): 35-40.
26. Romanzini M, Reichert FF, Lopes Ada S, Petroski EL,
de farias Júnior JC. Prevalência de fatores de risco
cardiovascular em adolescentes. Cad Saude Publica 2008;
24(11): 2573-81.
27. da Silva AC, Rosa AA. Blood pressure and obesity of children
and adolescents association with body mass index and waist
circumference. Arch Latinoam Nutr 2006; 56(3): 244-50.
28. Vieira MA, Carmona DPD, Anjos LA, Souza T, Espinosa MM,
Ribeiro RLR. Pressão arterial de crianças e adolescentes de
escolas públicas de Cuiabá, Mato Grosso. Acta Paul Enferm
2009; 22(1): 473-5.
29. Ribeiro RQ, Lotufo PA, Lamounier JA, Oliveira RG, Soares
JF, Botter DA. Fatores adicionais de risco cardiovascular
associados ao excesso de peso em crianças e adolescentes:
o estudo do coração de Belo Horizonte. Arq Bras Cardiol
2006; 86(6): 408-18.
30. Rosa MLG, Fonseca VM, Oigman G, Mesquita ET.
Pré-hipertensão arterial e pressão de pulso aumentada
em adolescentes: prevalência e fatores associados. Arq
Bras Cardiol 2006; 87(1): 46-53.
31. Zhang YX, Zhao JS, Sun GZ, Lin M, Chu ZH. Prevalent
trends in relatively high blood pressure among children
and adolescents in Shandong, China. Ann Hum Biol
2012; 39(3): 259-63.
32. Urrutia-Rojas X, Egbuchunam CU, Bae S, Menchaca
J, Bayona M, Rivers PA, et al. High blood pressure in
school children: prevalence and risk factors. BMC
Pediatr 2006; 6: 32.
33. Kollias A, Antonodimitrakis P, Grammatikos E,
Chatziantonakis N, Grammatikos EE, Stergiou GS. Trends
in high blood pressure prevalence in Greek adolescents. J
Hum Hypertens 2009; 23(6): 385-90.
34. Christofaro DGD, Andrade SM, Fernandes RA, Cabrera
MAS, Dias RMR. Prevalência de pressão arterial elevada em
crianças e adolescentes: revisão sistemática. Rev Bras Saude
Matern Infant 2011; 11(4): 361-7.
35. Sociedade Brasileira de Cardiologia,Sociedade
Brasileira deHipertensão, Sociedade Brasileirade
Nefrologia. V Diretrizes Brasileiras
de Hipertensão Arterial. Arq Bras Cardiol 2006; 1-48.
36. Borges LM, Peres MA, Horta BL. Prevalência de níveis
pressóricos elevados em escolares de Cuiabá, Mato Grosso.
Rev Saude Publica 2007; 41(4): 530-8.
Received on: 12/29/12
Final version presented on: 03/26/13
Accepted on: 06/05/13