Blood Pressure in 57,915 Pediatric Patients Who Are

Overweight or Obese Based on Five Reference Systems

Marion Flechtner-Mors, PhDa,*, Hannelore Neuhauser, MD, MPH, PhDb, Thomas Reinehr, MD, PhDc,
Hans-Peter Roost, PhDd, Susanna Wiegand, MD, PhDe, Wolfgang Siegfried, MDf,
Karl Zwiauer, MD, PhDg, Esther Molza, and Reinhard W. Holl, MD, PhDa, for the
APV initiative and the BMBF Competence Network Obesity

The aim of this study was to determine the prevalence of hypertension in overweight or obese
pediatric subjects using different national or international references, which are based either
on the entire population or on normal weight children only: 188 centers from Germany,
Austria, and Switzerland participated in the Adipositas Patienten Verlaufsbeobachtung
initiative. Data from 57,915 children aged 6 to 18 years who are overweight or obese were used
to determine the prevalence of prehypertension and hypertension based on Second Task
Force, European pooled data, Fourth Report all and Fourth Report nonoverweight, or
German Health Interview and Examination Survey for Children and Adolescents (KiGGS)
references. Three references included overweight children, whereas 2 (Fourth Report non-
overweight and KiGGS) were based on nonoverweight children only. Based on KiGGS,
Fourth Report nonoverweight, Fourth Report all, European pooled data, or Second Task
Force, the prevalence of hypertension was 47%, 42%, 36%, 32%, and 27%, respectively.
Recent references classified more children as hypertensive, whereas fewer children fell into the
prehypertensive group. Only 22% of children were classified as hypertensive by each of the 5
references (8% as prehypertensive). The prevalence of normal blood pressure was independent
of the reference applied. Hypertension as defined by the different reference systems was
significantly correlated, and all methods were significantly associated with impaired glucose
metabolism or dyslipidemia, without significant differences in methods. In conclusion, the
diagnosis of elevated blood pressure depends on the reference population used. A nonover-
weight reference population substantially increases the prevalence of hypertension in children
and adolescents who are overweight or obese. The choice of the reference has significant
implications for risk stratification and treatment decisions. Ó 2015 Elsevier Inc. All rights
reserved. (Am J Cardiol 2015;115:1587e1594)

Pediatric obesity is associated with elevated blood pres-
sure (BP), impaired glucose tolerance, and dyslipidemia,
which are risk factors for cardiovascular disease and type 2
diabetes mellitus.1,2
The aim of our study was to use different national and
international references for the determination of the preva-
lence of increased BP in a large number of children and
adolescents who are overweight or obese from German,
Austrian, and Swiss pediatric treatment facilities. Refer-
ences were the contemporary nonoverweight reference
a
Institute of Epidemiology and Medical Biometry, ZIBMT, University
of Ulm, Germany; bRobert Koch-Institute, Berlin, Germany; cDepartment
of Pediatric Endocrinology, Diabetes and Nutrition Medicine, University
Witten/Herdecke, Datteln, Germany; dInstitute St. Josef Guglera, Giffers,
Switzerland; eCharité Children’s Hospital, Universitätsmedizin Berlin,
Germany; fObesity Center Insula, Bischofswiesen, Germany; and gChild
and Adolescent Medicine—Regional Hospital St. Pölten, St. Pölten,
Austria. Manuscript received November 5, 2014; revised manuscript
received and accepted February 26, 2015.
See page 1593 for disclosure information.
*Corresponding author: Tel: þ49-731-5025353; fax: þ49-731-
5025309.
E-mail address: marion.flechtner-mors@uni-ulm.de (M. Flechtner-
Mors).
population from the same country (Germany, German
Health Interview and Examination Survey for Children and
Adolescents [KiGGS])3 or the Fourth Report database
restricted to nonoverweight children (Fourth Report non-
overweight). These results were compared with current
Fourth Report recommendations (identical to recommen-
dations by European Society for Hypertension (ESH)/Eu-
ropean Society of Cardiology), with the previous ESH/
European Society of Cardiology guidelines, and with the
definition of hypertension by the Second Task Force. These
last 3 reference populations included children who are
overweight/obese. Furthermore, the association of hyper-
tension with impaired glucose and lipid metabolism was
compared based on all 5 references.
Methods
The “Adipositas Patienten Verlaufsbeobachtung” (APV,
obese patients observational study) is a standardized multi-
center database of children and adolescents who are over-
weight or obese (www.a-p-v.de).4 A total of 188 specialized
obesity care centers (161 outpatient programs and 27 inpa-
tient rehabilitation institutions) in Germany, Austria, and
Switzerland participated in the study from January 2000 to
October 2012. The data were anonymized and transmitted

0002-9149/15/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. www.ajconline.org
http://dx.doi.org/10.1016/j.amjcard.2015.02.063
1588 The American Journal of Cardiology (www.ajconline.org)
Figure 1. Flow chart of patient selection in the present study.
for central analysis to the University of Ulm, Germany.
Every 6 months, inconsistent data were reported back to the
centers for correction. Data collection was co-ordinated with
all local data protection committees. Approval for data
analysis was obtained from the Ethics Committee by the
Institutional Review Board of Ulm University.
By October 2012, weight, height, and BP from 60,579
subjects were documented. For the present study, all 58,011
patients of the APV population between 6 and 18 years were
selected. Of these, 80 patients were on antihypertensive
medication. In addition, 16 subjects had secondary hyper-
tension because of renal (n ¼ 5), endocrine (n ¼ 8), or
cardiovascular (n ¼ 3) causes. Therefore, 57,915 subjects
were included in the analysis. Patients’ mean age was 12.7
years (2.5), mean height was 158.9 cm (13.2), mean
weight was 78.9 kg (23.3), body mass index (BMI) 30.5
kg/m2 (5.6), and BMI-Standard Deviation Score (BMI-
SDS) 2.5 (0.5). Nonoverweight was defined as a BMI
<90th percentile; overweight as a BMI 90th but <97th
percentile; obesity as a BMI 97th but <99.5th percentile;
and extreme obesity as a BMI 99.5th percentile, the latter
corresponding to a BMI >40 kg/m2 in adults.5 More
children were obese (n ¼ 25,885, 45%) or extremely obese
(n ¼ 25,064, 43%) than overweight (n ¼ 6,966, 12%).
Slightly more girls (54%) participated in the study. Most
study participants came from Germany (n ¼ 56,420).
Migration background was defined as being born abroad
or having a mother and/or father whose country of birth lies
outside Germany, Austria, or Switzerland. The number of
patients with migration background was 6,594 (11%), with a
mean age of 12.5 years (2.6), a mean height of 157.4 cm
(13.5), a mean weight of 79.2 kg (24.2), BMI 31.2 kg/
m2 (5.8), and BMI-SDS 2.6 (0.5); 50% of children with
migration background were extremely obese (41% were
obese and 9% were overweight). In Germany, the largest
migrant group comes from Turkey, followed by Poland.6 A
flow chart of patient selection is given in Figure 1.
Body height was measured in accordance with the
German working group on obesity in children and adoles-
cents.5 BP was measured according to the guidelines of the
German Hypertension League (www.paritaet.org/RR-Liga).
BP levels were measured after 3 to 5 minutes of rest, the
patient in a seated position, at the upper arm, with proper
cuff size, and the cuff placed at heart level using sphyg-
momanometer or oscillometric measurement devices; 53%
of patients were seen at centers, where mainly auscultatoric
measurement was used; 47% of patients were cared for at
centers, where predominantly oscillometric measurement
was applied. BP measuring devices used at the centers were
validated according to standard protocols and conformed to
the European standard specifications of having a leak rate
within 4 mm Hg/min and a pressure scale accurate to within
 Miscellaneous/Blood Pressure and Childhood Obesity
Figure 2. Prevalence of normal BP, prehypertension, and hypertension in
patients who are overweight and obese, based on Second Task Force, ESH
European pooled data, ESH Fourth Report all, Fourth Report nonover-
weight, and KiGGS reference (95% confidence interval).
3 mm Hg.7,8 BP measurements with oscillometric devices
or mercury sphygmomanometer may differ slightly.9 Ac-
cording to the literature, oscillometric methods may over-
estimate or underestimate systolic and diastolic BP
values.10,11 Therefore, no correction factor was used.
The guidelines define hypertension as BP levels 95th
percentile at 3 measurements on 2 different days. Statistical
analysis was based on median BP documented during the
first 6 weeks after initial presentation of the patient. BP
values were classified as normal (<90th percentile), pre-
hypertension (90th to <95th percentile, 120/80 mm Hg
even if <90th percentile in adolescents), or hypertension
(95th percentile).12
BP was classified based on Second Task Force,13 ESH
European pooled data,14 ESH Fourth Report all children
(Fourth Report all),12 Fourth Report nonoverweight children
(Fourth Report nonoverweight),9 and recent KiGGS.3 To
allow comparison among all 5 references, subjects were
restricted to the age range of 6.0 to 18.0 years.
BMI- and height-SDS were calculated according to cur-
rent German reference data (KiGGS),3 and patients were
divided into 3 groups: <25th, 25th to 75th, and 75th
percentile. To consider the influence of height on hyper-
tension prevalence, the study population was again divided
into 3 groups: <25th, 25th to <75th, and 75th percentile.
Impaired glucose metabolism was defined as a fasting
glucose level of >110 mg/dl (>5.6 mmol/L) or a glucose
concentration of >140 mg/dl (>7.8 mmol/L) 2 hours after an
oral glucose tolerance test (21). Dyslipidemia was defined as
total cholesterol >200 mg/dl (>5.1 mmol/L), low-density li-
poprotein (LDL) cholesterol >130 mg/dl (>3.4 mmol/L),
high-density lipoprotein (HDL) cholesterol <35 mg/dl
(<0.9 mmol/L), or triglycerides >150 mg/dl (>1.7 mmol/L).15
SAS, version 9.4 (SAS Institute, Cary, North Carolina),
was used for statistical analysis. Clinical characteristics of
the study population are given as mean (SD). Association of
hypertension with impaired glucose or lipid metabolism was
1589
estimated by logistic regression adjusted for age, gender,
and BMI category and given as odds ratios (95% confidence
interval [CI]). A random factor with Cholesky covariance
structure was used to model the variability in centers.
Prevalence of hypertension was modeled using hierarchic
logistic regression (SAS proc glimmix, estimation by
restricted partial likelihood, denominator degrees of
freedom according to Kenward-Roger, optimization of it-
erations by the Newton-Raphson method). Patient groups
were compared by adjusted means based on observed
marginal frequencies. Spearman correlation coefficients
with 95% confidence intervals based on Fisher z-trans-
formation were used to compare the classification of hy-
pertension among the 5 different methods. A p value <0.05
was considered to be statistically significant.
Results
In Figure 2, BP in children, aged 6 to 18 years, who were
overweight/obese was evaluated according to Second Task
Force published in 1987, ESH European pooled data (1991),
ESH Fourth Report all (2004), ESH Fourth Report non-
overweight (2008), and KiGGS (2011). Major differences
were observed in the classification of hypertension: Referred
to a solely nonoverweight German reference population
(KiGGS), the prevalence of hypertension was highest with
47.3%, followed by Fourth Report nonoverweight with
42.4%, Fourth Report all (36.3%), European pooled data
(32.0%), and Second Task Force (27.2%). In correspon-
dence to the increase in the prevalence of hypertension, the
prevalence of prehypertension decreased. The percentage of
normal BP was similar for all reference systems.
In addition, the proportion of children consistently cate-
gorized by every method was calculated: 7.9% for pre-
hypertension, 21.8% for hypertension, and 29.4% for
normotension. Therefore, in 40.9% of children classification
differed. To agree with respect to classification as hyperten-
sive was closest for Fourth Report all and Fourth Report
nonoverweight (r ¼ 0.870; 95% CI 0.868 to 0.872) and
slightly lower for KiGGS versus Fourth Report all (r ¼ 0.772;
95% CI 0.769 to 0.776) or Fourth Report nonoverweight (r ¼
0.796; 95% CI 0.793 to 0.799). Hypertension based on Eu-
ropean pooled data agreed better with KIGGS (r ¼ 0.707;
95% CI 0.702 to 0.713) compared with Fourth Report all (r ¼
0.659; 95% CI 0.655 to 0.644). However, hypertension based
on Second Task Force agreed also best with Fourth Report all
(r ¼ 0.759; 95% CI 0.756 to 0.763), whereas it was lower
with KiGGS (r ¼ 0.637; 95% CI 0.633 to 0.642).
The percentage of subjects with hypertension in each
weight status category increased from the older to the more
recent references. An opposite effect was observed for pre-
hypertension. Rates of hypertension related to weight status
category are listed in Table 1. A comparison of children who
are overweight with those who are extremely obese revealed
a greater percentage of hypertension in subjects who are
extremely obese (p <0.001; Table 1). Furthermore, early
pubertal children, aged 11 to <14 years, were hypertensive
and more often compared with the other age groups
(p <0.001; exception European pooled data; Table 1).
Shorter children had hypertension more often than taller
ones (each reference p <0.001), except according to the
1590 The American Journal of Cardiology (www.ajconline.org)

Table 1
Prevalence of hypertension (%) in the study population based on 5 reference systems stratified by weight status category (adjusted for age, sex and height) and
age-group (adjusted for sex, BMI, and height); [95% confidence interval], p-value for difference among categories
Weight status category Age (years)

overweight obese extremely obese P <11 11-<14 14-18 P

Second Task Force 14.4 22.1 35.1 <0.001 27.4 28.8 21.7 <0.001
[13.6-15.2] [21.6-22.6] [34.5-35.7] [26.6-28.1] [28.3-29.4] [21.1-22.4]
European pooled data 19.4 26.8 40.6 <0.001 34.3 32.7 27.4 <0.001
[18.5-20.3] [26.2-27.3] [40.0-41.2] [33.5-35.1] [33.2-33.4] [26.7-28.0]
4th Report all 22.1 30.7 46.0 <0.001 34.2 37.2 35.3 <0.001
[21.1-23.1] [30.2-31.3] [45.4-46.6] [33.9-35.0] [36.5-37.8] [34.5-36.0]
4th Report non-overweight 26.8 36.9 52.3 <0.001 42.0 43.6 40.1 <0.001
[25.8-27.9] [36.0-37.5] [51.7-52.9] [41.1-42.8] [43.0-44.2] [39.4-40.9]
KiGGS 31.6 42.1 56.9 <0.001 47.8 50.3 42.3 <0.001
[30.5-32.7] [41.4-42.7] [56.2-57.5] [47.0-48.7] [49.7-51.0] [41.6-43.1]

Table 2
Prevalence of hypertension (%) in the study population based on 5 reference systems stratified by height-group or sex; [95% confidence interval], p-value for
difference among categories
Height Sex

<25th 25th - <75th 75th P male female P

percentile percentile percentile

Second Task Force 19.9 24.4 29.7 <0.001 29.5 23.3 <0.001
[18.9-20.8] [23.8-24.9] [29.1-30.2] [28.9-30.0] [22.8-23.8]
European pooled data 41.0 32.4 28.0 <0.001 30.8 31.8 < 0.05
[39.8-42.2] [31.8-33.0] [27.4-28.5] [30.2-31.3] [31.2-32.3]
4th Report all 39.1 36.7 34.1 <0.001 36.9 34.8 <0.001
[38.0-40.3] [36.1-37.3] [33.5-34.7] [36.3-37.5] [34.3-35.4]
4th Report non-overweight 44.6 42.1 41.4 <0.001 42.6 41.6 <0.05
[43.4-45.8] [41.5-42.7] [40.7-42.0] [42.0-43.2] [41.0-42.2]
KiGGS 46.9 47.1 47.2 ns 48.1 46.2 <0.001
[45.7-48.1] [46.4-47.7] [46.5-47.8] [47.5-48.7] [45.7-46.8]

Second Task Force and KiGGS (Table 2). Significantly more
boys than girls had hypertension (p <0.05 to p <0.001),
except when European pooled data were the reference. More
girls were hypertensive based on the European pooled data (p
<0.05; Table 2). Prevalence of elevated systolic or diastolic
BP in girls and boys in yearly age intervals is shown in
Figure 3. Different rates were observed depending on the
reference, gender, and age of the child.
Adjusted for age, gender, and BMI category, subjects
with or without migration background showed similar
prevalence rates of normal BP or hypertension according to
each reference. However, the proportion of prehypertension
was slightly lower in children with migration background
based on Second Task Force (p <0.01) and Fourth Report
all (p <0.05; Table 3).
Glucose and lipid values were available from 36,845 or
35,872 children, respectively, excluding children on met-
formin or lipid-lowering drugs. On average, fasting blood
glucose was 84  25 mg/dl. The level for cholesterol was
163  36 mg/dl, for LDL cholesterol 99  32 mg/dl, for
HDL cholesterol 48  12 mg/dl, and for triglycerides 98 
59 mg/dl. Odds ratio estimates for the association of hy-
pertension with impaired glucose or lipid metabolism for all
5 references was significant (p <0.05 e p <0.001)
(Table 4). This was confirmed when a composite end point
(dyslipidemia or impaired glucose metabolism) was used.
According to all references, hypertensive boys had signifi-
cantly more often elevated cholesterol, LDL cholesterol, and
lower HDL cholesterol (p <0.001, each). TG levels were
similar between genders.
Discussion
In the study population, the prevalence of hypertension
depended on the reference used. The 2 most recent reference
systems based on nonoverweight reference populations
(Fourth Report nonoverweight and KiGGS) resulted in the
highest rates of hypertension, whereas the 2 older references
(Second Task Force and European pooled data) returned
lower rates. Weight or BMI is an important predictor of BP in
children: consequently, the highest percentage of abnormal
BP was found when nonoverweight reference populations
were used. In a recent study, BP centiles based on Fourth
Report16 were recalculated including nonoverweight chil-
dren only.17 As expected, the prehypertensive and hyper-
tensive thresholds were lower. Furthermore, Neuhauser et al3
computed the age- and gender-specific 95th percentile from a
cohort of German children that included all weight categories
 Miscellaneous/Blood Pressure and Childhood Obesity 1591

Figure 3. Prevalence of hypertension (systolic BP or diastolic BP) in girls and boys referred to Second Task Force (second TF), ESH European pooled data
(EPD), ESH Fourth Report all (fourth all), ESH Fourth Report nonoverweight (fourth nw), and KiGGS reference.

Table 3
Prevalence (%) of pre-hypertension and hypertension in the study population with (n ¼ 6594) or without (n ¼ 51321) migration background. Mean values
adjusted for age, sex and weight category.
Pre-hypertension P Hypertension P

migration no migration no
migration migration

Second Task Force 34.7 36.5 <0.01 26.4 26.1 ns

[33.6-35.9] [36.1-36.9] [25.4-27.5] [25.8-26.6]
European pooled data 31.5 32.3 ns 31.0 31.5 ns
[30.3-32.6] [31.9-32.8] [29.9-32.1] [31.1-31.9]
4th Report all 25.2 26.5 <0.05 36.4 35.7 ns
[24.2-26.3] [26.1-26.9] [35.3-37.6] [35.3-36.2]
4th Report non-overweight 21.5 22.0 ns 42.4 42.0 ns
[20.5-22.5] [21.6-22.4] [41.2-43.6] [41.6-42.5]
KiGGS 17.7 18.2 ns 46.8 47.1 ns
[16.8-18.6] [17.9-18.6] [45.5-48.0] [46.7-47.6]

P ¼ migration background versus no migration background.

in comparison with nonoverweight children from the same
cohort. The centiles for systolic and diastolic BP were higher
if overweight subjects were included.
Several factors complicate the establishment of definitive
BP percentiles. The diversity of a population and the ethnic
origin are important. The first studies on BP in children
started in the 1970s. Data in the United States were collected
almost exclusively from white children between 5 and
18 years.18 The follow-up report on BP in 1987 was based
on data recorded in white, black, and Mexican-American
children. New data from the 1999 to 2000 US National
Health and Nutrition survey were added in 2004.12,13 In
Europe, BP data were collected in the 1980s in children
from northwestern European countries.14 BP measurements
1592 The American Journal of Cardiology (www.ajconline.org)
Table 4
Odds ratio [95% confidence interval], adjusted for age, sex and weight
category, for the association of hypertension with impaired glucose
metabolism or dyslipidaemia. Five different reference systems are
compared. Blood glucose or lipid values were available from 36,845 or
35,872 children, respectively
Impaired glucose P Dyslipidaemia P data.22 In the future, clinical markers for early cardiovas-
metabolism
Second Task 1.12 [0.99-1.28] <0.001 1.17 [1.12-1.23] <0.001
Force
European 1.17 [1.04-1.33] <0.05 1.20 [1.14 -1.25] <0.001
pooled data
4th Report all 1.14 [1.01-1.29] <0.001 1.21 [1.16-1.27] <0.001
4th Report 1.16 [1.03-1.30] <0.001 1.21 [1.15-1.27] <0.001
non-overweight
KiGGS 1.17 [1.04-1.32] <0.001 1.15 [1.10-1.20] <0.001
of German children were performed recently (2003 to 2006;
KiGGS, 8% migration background). Children in the United
States represent a more heterogeneous group and differ from
European populations.3,13,19 However, European pop-
ulations may also be heterogeneous: for instance, German or
Italian data differ from the pooled European references.3,13
Although in our study, the prevalence of hypertension was
not affected by migration background, ethnic divergence in
BP, not explained by obesity, has been shown.20,21 It is
unknown whether these differences are attributable to ge-
netic or environmental factors.
Different growth curves for children and adolescents
from different populations affect the establishment of
appropriate boundaries for prehypertension and hyperten-
sion. The relation between BP and both age and height is
nonlinear.22 As children who are obese tend to be taller, the
frequency of hypertension is lower for reference systems
including height. Beyond that, over the past decades, the
average weight increased in children worldwide, resulting in
higher values of BP.22,23 For the assessment of BP, the
statistical method to define hypertension cutoffs is relevant.
Rosner et al17 considered 3 types of models for pediatric BP
data, including polynomial regression and restricted cubic
splines, whereas the best fit was achieved with quantile
regression. In addition, the question has to be answered
whether the 95th percentile of a reference population is the
most appropriate to determine hypertension. British guide-
lines, for example, choose the 98th percentile based on a
British reference population.24
Our study revealed that more male and more extremely
obese children had hypertension. The findings are in accor-
dance with other studies.16 However, the rates of hyperten-
sion according to the 5 reference systems were different.
Furthermore, early pubertal children aged 11 to <14 years
had hypertension more often than children at the age of <11
or 14 to 18 years, with the exception of European pooled
data. Moreover, because Second Task Force and European
pooled data do not account for body height, interquartile
range was higher compared with the other references. All
results clearly illustrate the complexity associated with the
choice of a reference population and a definition of hyper-
tension. Potential criteria for the assessment of different
reference systems are summarized in Table 2.
An important issue in pediatric hypertension is the rela-
tion to cardiovascular disease in adulthood. In children and
adolescents, the exact level and duration of BP elevation
that cause target organ damage has not been established.25
Childhood hypertension is defined by the upper 5% of the
distribution because of the lack of long-term outcome
cular disease that include increased carotid intima-media
thickness, arterial stiffness, altered brachial artery flow-
mediated dilatation, retinal vascular narrowing, or
increased cardiovascular biomarkers, such as interleukin-6,
fibrinogen, vascular adhesion molecules, and advanced
oxidation protein products, might help to identify subjects
with increased cardiovascular risk.22,26
Obesity associated with hypertension is a key risk factor
for developing cardiovascular disease.1,2 Using more recent
reference systems, an increasing prevalence of hypertension
together with a decrease in the prevalence of pre-
hypertension was observed. This shift has far-reaching
consequences because more children may require
treatment.27
Hypertension often coincides with impaired glucose
tolerance or dyslipidemia, and it increases the risk for type 2
diabetes or cardiovascular disease.1 The association of hy-
pertension with impaired glucose metabolism or dyslipide-
mia is clearly demonstrated in our study, regardless of the
reference population.
One approach for treatment is weight loss by means of
lifestyle modification, based on dietary counseling and
enhanced physical activity.28 Antihypertensive medication
is a further possibility, if lifestyle intervention is not suc-
cessful.12,16 However, guidelines are not precise on when to
start pharmacologic therapy and which first-line antihyper-
tensive agent to use.12,28 Although some medications are
approved by the Food and Drug Administration and dose
ranges for selected antihypertensive agents in children and
adolescents are recommended, children and adolescents
who are obese rarely receive drug treatment.27 Controlled
clinical studies and long-term follow-up are necessary to
gain knowledge about benefits and disadvantages of BP-
lowering drugs in children and adolescents.12,28
The strength of our study is the large number of over-
weight and obese children and adolescents from obesity care
centers in Germany, Austria, and Switzerland. At all centers,
qualified staff is available, and data collection is standard-
ized and monitored by benchmarking.
A limitation of the study is that preferentially obese and
extremely obese children were examined in the treatment
centers, and the data for overweight subjects might be
under-represented. The selection of patients may be biased
not only by the degree of obesity but also by diagnosed or
presumed metabolic disorders. Furthermore, in this multi-
center approach, there may be some subjectivity and vari-
ability in data collection despite the standardized
procedures.
Acknowledgment: We thank all health professionals of the
APV study group taking care of the children and contrib-
uting to the APV database (list of all participating centers is
provided in the Appendix).
 Miscellaneous/Blood Pressure and Childhood Obesity 1593

Disclosures Homburg CJD, Homburg Universitätskinderklinik, Kassel

Kinderarztpraxis, Kiel städtisches Krankenhaus Förderkids,
The APV initiative is supported by grants from the
Kreischa Klinikum Bavaria Zscheckwitz, Köln
German Federal Ministry of Education and Research
Kinderklinik Amsterdamerstrasse Power Pänz, Köln MeLo
(Competence Network Obesity FKZ 01GI1130), the Euro-
KIDS Schulungsprogramm, Köln Netzwerk Gesundheit,
pean Foundation for the Study of Diabetes (EFSD), and the
Köln - Prävention UniReha GmbH, Köln Sporthochschule,
German Federal Center for Health Education (BZgA). All
Leipzig Sportmedizin, Korbach Ernährungsberatung,
authors have no conflicts of interest to disclose.
Leipzig Universitätskinderklinik, Lindau Forum Adipositas,
Lindenberg/Lindau Adipositasschulung, Lingen Bonifatius-
Appendix Hospital, Lörrach Kinderklinik, Lübeck
Universitätskinderklinik, Magdeburg Städtische
Kinderklinik, Magdeburg Universitätskinderklinik, Mahlow
Participating centers: Amrum Satteldüne Kinder-Reha, Ernährungspraxis, Menden BIG, Mönchengladbach Städt.
Augsburg Bunter Kreis, Bad Bodenteich Moby Dick Kinderklinik, Mühlhausen Präventionspraxis, München
Seeparkklinik, Bad Fallingbostel Gesundheitszentrum, Bad Adieupositas, Münster Arztpraxis, Münster
Frankenhausen Kinder-Reha, Bad Hersfeld Kinderklinik, Ernährungsberatung Moby Dick, Murnau Kinder-Reha,
Bad Kösen Kinder-Reha, Bad Kreuznach Viktoriastift, Bad Nagold Ernährungsberatung, Neumünster
Lippspringe, Bad Mergentheim Kinderklinik, Bad Präventionszentrum, Neunkirchen Kinderklinik, Neuss
Neuenahr DRK Institutsambulanz, Bad Rothenfelde Lukaskrankenhaus, Niederkassel Kinderarztpraxis, Norden
Kinder-Reha, Bad Orb Spessartklinik Kinder-Reha, Bad Klinik Nordendeich, Nürnberg PEP, Nürnberg Kinder-und
Salzungen Reha-Klinik Charlottenhall, Bad Segeberg/ Jugendarztpraxis, Oy-Mittelberg Reha, Oberhausen
Neumünster junior marvelesse, Bensheim Adipositaszentrum, Oberhausen EKO Kinderklinik,
Ernährungspraxis, Berchtesgaden CJD, Berchtesgaden Oberstaufen Ernährungsmedizin, Oldenburg Kids
Klinik Schönsicht Kinder-Reha, Berlin Charite Schulungsprogramm, Oldendorf Ernährungspraxis
Kinderklinik, Berlin DRK Ausbildungszentrum, Berlin KiloKids, Osnabrück Kinderhospital, Overath KIDS
Lichtenberg Kinderklinik, Berlin Vivantes Schulungsprogramm, Paderborn Ernährungspraxis
Behandlungszentrum SPZ, Bischofswiesen/Strub INSULA, Kinderleicht, Passau Kinderklinik, Pforzheim Adipositas
Blaubeuren Ernährungspraxis, Böblingen Kinderarztpraxis, Training, Pocking Kinderarztpraxis, Pönitz FiFaFu
Bonn Ernährungsberatung KIDS Schulung, Braunschweig KIDS-Programm, Poppenricht Ernährungsberatung,
ernährungsmedizinisches Zentrum, Bregenz Potsdam Patienten Trainings Zentrum, Ravensburg
Landeskrankenhaus Kinderklinik, Bremen-Nord Ernährung und Diät, Ravensburg Oberschwabenklinik
Kinderklinik, Bremen Zentralkrankenhaus Kinderklinik, Kinderklinik, Reiskirchen Ernährungspraxis, Rendsburg
Brügge Förderkids, Buchholz Ernährungsberatung, Bühl Villa Schwensen Praxisgemeinschaft, Rosenheim Lufti-
Praxis Ernährungsberatung, Bruchweiler Kinder-Reha, Team, Rottweil Kinder-Leicht, Ronneburg
Darmstadt Kinderklinik, Datteln Vestische Kinderklinik, Ernährungsberatung, Rüsselsheim Gesundheits-und
Düsseldorf Ernährungspraxis “iss gut,” Düsseldorf Pflegezentrum, Saalfeld Kinderklinik, Salzburg
Ernährungsberatung “richtig essen,” Detmold Kinderklinik, Kinderklinik, Salzgitter Kinderklinik, Saarbrücken Moby
Dieburg Ernährungsberatung KIDS Schulung, Dinslaken Dick, Scheidegg Prinzregent Luitpold Reha, Schliengen
Kinderklinik, Dornbirn Kinderklinik, Dorsten Ernährungsberatung, Senden Ernährungsberatung, Seebad
St. Elisabethkrankenhaus, Dresden Moby Dick, Düren Heringsdorf Kinder-Reha, Seebad Kölpinsee Klaus
Gesundheitsamt, Düren sozialpädagogisches Zentrum Störtebecker Kinder-Reha, Siegburg KIDS
Marienhospital, Eppingen Kinderarztpraxis, Erlangen Schulungsprogramm, Siegen DRK Kinderklinik,
Universitätskinderklinik, Eschede Adipositastraining KIDS, Simonswald Klinik Eichhof, Solingen Ernährungsberatung,
Ettenheim Kinderarztpraxis, Feldberg ITZ Caritas-Haus, Sonneberg KIDS Ernährungspraxis, St. Augustin
Feldkirch Landeskrankenhaus Kinderklinik, Flensburg Kinderklinik, St. Gallen Ostschweiz Kinderklinik,
Förderkids, Frankfurt Päd. Endokrinologie, Freiburg Fitoc, St. Pölten Landesklinikum Kinderklinik, Straubing Praxis,
Freiburg Universitätskinderklinik, Friedrichsdorf Tholey SPZ Neunkirchen, Tübingen
Ernährungspraxis, Fürth Kinderklinik, Göttingen KIDS Universitätskinderklinik, Ulm Universitätskinderklinik,
Schulungsprogramm, Göttingen Universitätskinderklinik, Untergruppenbach Ernährungsberatung, Villingen-
Gaissach Fachklinik Deutsche Rentenversicherung Bayern- Schwenningen Kinderarztpraxis, Viersen Kinderklinik
Süd, Garz Fachklinik CJD, Gauting, Kinderarztpraxis, St. Nikolaus, Waldbröl Gemeinschaftspraxis, Waltrop
Gelnhausen Ernährungsberatung, Giffers Ernährungsberatung, Wangen Kinder-Rehaklinik,
Ausbildungszentrum Guglera, Gittelde am Harz Westerland/Sylt Kinder-Reha, Westerland/Sylt Haus
Ernährungsberatung, Gotha Helios Kinderklinik, Quickborn, Wien Universitätskinderklinik, Wiesbaden
Gröbenzell Ernährungsberatung, Hagen Allgemeines DKD Kinderklinik, Wiesmoor KIDS Schulungsprogramm,
Krankenhaus, Hagen Kinderarztpraxis, Hagen Windach Psychosomatische Klinik, Winsen Luhe
Kinderklinik, Hamburg Moby Dick, Hamburg Rallye Ernährungsberatung, Würzburg ambulantes
Energy, Hamburg Wilhelmstift, Hannover BKK Essanelle, Schulungszentrum, Wustrow Ostseebad Fischland, Wyk
Hannover Kinderklinik auf der Bult, Habfurt auf Föhr AOK Kinderkurheim, Zorneding
Adipositasschulung Habberge, Herdecke Kinderklinik, Ernährungsberatung, and Zwickau—Praxis
Herne Praxis Ernährungsmedizin, Hirschberg Praxis, Ernährungsberatung.
1594 The American Journal of Cardiology (www.ajconline.org)
1. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin
BA, Gordon DJ, Krauss RM, Savage PJ, Smith SC Jr, Spertus JA,
Costa F; American Heart Association; National Heart, Lung, and Blood
Institute. Diagnosis and management of the metabolic syndrome: an
American Heart Association/National Heart, Lung, and Blood Institute
Scientific Statement. Circulation 2005;112:2735e2752.
2. Landsberg L, Aronne LJ, Beilin LJ, Burke V, Igel LI, Lloyd-Jones
D, Sowers J. Obesity-related hypertension: pathogenesis, cardiovas-
cular risk, and treatment: a position paper of the Obesity Society and
the American Society of Hypertension. J Clin Hypertens 2013;15:
14e33.
3. Neuhauser HK, Thamm M, Ellert U, Hense HW, Rosario AS. Blood
pressure percentiles by age and height from nonoverweight children
and adolescents in Germany. Pediatrics 2011;127:e978e988.
4. Reinehr T, Wabitsch M, Andler W, Beyer P, Böttner A, Chen-Stute A,
Fromme C, Hampel O, Keller KM, Kilian U, Kolbe H, Lob-Corzilius
T, Marg W, Mayer H, Mohnike K, Oepen J, Povel C, Richter B,
Riedinger N, Schauerte G, Schmahlfeldt G, Siegfried W, Smuda P,
Stachow R, van Egmond-Fröhlich A, Weiten J, Wiegand S, Witte S,
Zindel V, Holl RW; APV Study Group. Medical care of obese children
and adolescents. APV: a standardised multicentre documentation
derived to study initial presentation and cardiovascular risk factors in
patients transferred to specialised treatment institutions. Eur J Pediatr
2004;163:308e312.
5. AGA-Guidelines. Arbeitsgemeinschaft Adipositas im Kindes- und
Jugendalter. S2 Leitlinie. (German Guidelines). Available at: www.a-g-
a.de. Accessed on November 12, 2013.
6. Statistisches Bundesamt Deutschland: Leichter Anstieg der Bevölker-
ung mit Migrationshintergrund. Available at: https://www.destatis.de/
DE/Publikationen/Thematisch/Bevoelkerung/MigrationIntegration/
AuslaendBevoelkerung.html. Accessed January on 30, 2015.
7. European Standard EN 1060-1:1996. Specification for Non-invasive
Sphygmomanometers. Part 1: General Requirements. Rue Stassart,
Brussels: European Commission for Standardisation, 1996.
8. De Greeff A, Lorde I, Wilton A, Seed P, Coleman AJ, Shennan AH.
Calibration accuracy of hospital-based non-invasive blood pressure
measuring devices. J Hum Hypertens 2010;24:58e63.
9. Smulyan H, Safar ME. Blood pressure measurement: retrospective and
prospective views. Am J Hypertens 2011;24:628e634.
10. Ingelfinger JR. Clinical practice. The child or adolescent with elevated
blood pressure. N Engl J Med 2014;370:2316e2325.
11. Neuhauser HK, Ellert U, Thamm M, Adler C. Calibration of blood
pressure data after replacement of the standard mercury sphygmoma-
nometer by an oscillometric device and concurrent change of cuffs.
Blood Press Monit 2015;20:39e42.
12. Lurbe E, Cifkova R, Cruickshank JK, Dillon MJ, Ferreira I, Invitti C,
Kuznetsova T, Laurent S, Mancia G, Morales-Olivas F, Rascher W,
Redon J, Schaefer F, Seeman T, Stergiou G, Wühl E, Zanchetti A;
European Society of Hypertension. Management of high blood pres-
sure in children and adolescents: recommendations of the European
Society of Hypertension. J Hypertens 2009;27:1719e1742.
13. American Academy of Pediatrics, Task Force on Blood Pressure
Control in Children. Report on the second task force on blood pressure
control in children—1987. Pediatrics 1987;79:1e25.
14. de Man SA, André JL, Bachman H, Grobbee DE, Ibsen KK, Laaser U,
Lippert P, Hofman A. Blood pressure in childhood: pooled findings of
six European studies. J Hypertens 1991;9:109e114.
15. Kavey RE, Daniels SR, Lauer RM, Atkins DL, Hayman LL, Taubert K;
American Heart Association. American Heart Association guidelines for
primary prevention of atherosclerotic cardiovascular disease beginning
in childhood. Circulation 2003;107:1562e1566.
16. National High Blood Pressure Education Program Working Group on
High Blood Pressure in Children and Adolescents. The fourth report on
the diagnosis, evaluation, and treatment of high blood pressure in
children and adolescents. National Heart, Lung, and Blood Institute,
Bethesda, Maryland. Pediatrics 2004;114:555e576.
17. Rosner B, Cook N, Portman R, Daniels S, Falkner B. Determination of
blood pressure percentiles in normal-weight children: some methodo-
logical issues. Am J Epidemiol 2008;167:653e666.
18. Blumenthal S, Epps RP, Heavenrich R, Lauer RM, Lieberman E,
Mirkin B, Mitchell SC, Boyar Naito V, O’Hare D, McFate Smith W,
Tarazi RC, Upson D. Report of the task force on blood pressure control
in children. Pediatrics 1977;59(5 2 Suppl); I-II, 797e820.
19. Menghetti E, Virdis R, Stambi M, Patriarca V, Riccioni MA, Fossali E,
Spagnolo A. Blood pressure in childhood and adolescence: the Italian
normal standards. ‘Study group on Hypertension’ of the ‘Italian So-
ciety of Pediatrics’. J Hypertens 1999;17:1363e1372.
20. Rosner B, Cook N, Portman R, Daniels S, Falkner B. Blood pressure
differences by ethnic group among United States children and ado-
lescents. Hypertension 2009;54:502e508.
21. De Hoog MLA, van Eijsden M, Stronks K, Gemke RJBJ, Vrijkotte
TGM. Association between body size and blood pressure in children
from different ethnic origins. Cardiovasc Diabetol 2012;11:136.
22. Sinha MD, Reid CJ. At what level of blood pressure should hyper-
tension be defined in children? Cardiol Young 2009;19:428e430.
23. Lobstein T, Baur L, Uauy R. International Obesity Task Force. Obesity
in children and young people: a crisis in public health. Obes Rev
2004;(Suppl 1):S4eS104.
24. Jackson L, Thalange N, Cole T. Blood pressure percentiles for Great
Britain. Arch Dis Child 2007;92:298e303.
25. Collins RT II, Alpert BS. Pre-hypertension and hypertension in pediatrics:
don’t let the statistics hide the pathology. J Pediatr 2009;155:165e169.
26. Montero D, Walther G, Perez-Martin A, Roche E, Vinet A. Endothelial
dysfunction, inflammation, and oxidative stress in obese children and
adolescents: markers and effect of lifestyle intervention. Obes Rev
2012;13:441e455.
27. Reinehr T, Wiegand S, Siegfried W, Keller KM, Widhalm K, l’Alle-
mand D, Zwiauer K, Holl RW. Comorbidities in overweight children
and adolescents: do we treat them effectively? Int J Obes 2013;37:
493e499.
28. Batisky DL. What is the optimal first-line agent in children requiring
antihypertensive medication. Curr Hypertens Rep 2012;14:603e607.