Public Health Nutrition: 11(9), 905913 doi:10.

1017/S1368980007001255

Overweight and blood pressure: results from the examination

of a selected group of adolescents in northern Italy
G Turconi1,*, L Maccarini2, R Bazzano1 and C Roggi1
1
Department of Applied Health Sciences, Section of Human Nutrition and Dietetics, University of Pavia,
Via Bassi 21, I-27100 Pavia, Italy: 2Department of Preventive, Occupational and Community Medicine,
Hygiene Institute, Pavia, Italy

Submitted 22 February 2007: Accepted 9 September 2007: First published online 16 November 2007

Abstract
Objective: The aim of the present study was to investigate blood pressure (BP)
levels and their relationship with different indices of body fat in a group
of adolescents, in order to evaluate the prevalence of hypertension and plan
preventive and corrective strategies.
Design: Cross-sectional study in primary care.
Setting: All high schools in the Aosta Valley region, northern Italy.
Subjects: Five hundred and thirty-two adolescents of both sexes, aged 15
4
(standard deviation 0
7) years. The following parameters were measured: body
weight, body height, body mass index (BMI), four skinfold thicknesses, body fat
mass, waist and hip circumferences and BP.
Results: BMI data indicated a high prevalence of overweight subjects in both sexes,
but higher in males, while the prevalence rate of obese adolescents was lower. Of
the total, 11
8 % of subjects suffered from systolic hypertension, while 6
9 % suffered
from diastolic hypertension. In linear correlation analysis, BMI and all adiposity
indices, except waist:hip ratio, were found to be significantly associated (P ranging
between 0
05 and 0
001) with both systolic BP and diastolic BP in both sexes, with
r ranging between 0
152 and 0
359. Multiple regression analysis with the stepwise
method showed BMI and body fat mass to have the strongest association (P , 0
001)
Keywords
with BP, with r ranging between 0
275 and 0
359. Overweight
Conclusion: Unless reversed, these conditions are worrying and predict the Blood pressure
possible development of cardiovascular disease in adulthood. There is a pressing Hypertension
need to develop a comprehensive medical and nutrition plan, together with Body mass index
preventive and corrective strategies, in school programmes. Adiposity indices

The prevalence of overweight and obesity in childhood elevated BP levels at a young age may be predictive of
and adolescence is increasing rapidly worldwide(13). early signs of essential hypertension in adulthood. Careful
Epidemiological data show that prevalence rates are measurements of BP and thorough evaluation of adoles-
increasing not only in industrialised countries but also in cents with sustained high BP should make it possible to
developing countries(4), especially as far as the adolescent identify those who require treatment.
population is concerned. In general, prevalence rates are Even though BP levels are positively correlated with
higher in adolescents than in children(5), and are also height, most authors agree that body mass index (BMI)
higher among males than females. Overweight and contributes significantly to BP variability and is the best
obesity at a young age predicts the likelihood of obesity predictor of BP values(1821). In a sample of 155 Native
in adulthood(6). American youths aged 518 years, Smith and Rinder-
Overweight and obesity at a young age are important risk knecht(22) showed that mean systolic blood pressure
factors for several disorders(7) such as hypertension(8,9), early (SBP) and diastolic blood pressure (DBP) increased with
atherosclerotic damage(10), cardiovascular disease(1113), increasing BMI percentile. Martinez et al.(23) revealed a
metabolic syndrome(14), the onset of non-insulin-dependent 0
83 mmHg increment in BP for each unit increment of
diabetes mellitus(15) as well as morbidity and mortality in BMI in a sample of 2115 secondary-school students aged
later life(16,17). 14
8 (standard deviation, SD 1
6) years.
The importance of measuring blood pressure (BP) In a sample of 2365 healthy schoolchildren aged 816
in childhood has been widely recognised, given that years, Reich et al.(24) showed a significant trend towards a

*Corresponding author: Email [email protected] r The Authors 2007

906 G Turconi et al.
higher prevalence of hypertension in subjects with high The population of the Aosta Valley region is equal
BMI, but the positive correlation of hypertension to BMI to 0
20 % of the Italian population. Five hundred and
was not only caused by overweight subjects. Indeed, thirty-two students, 254 males (47
7 %) and 278 females
when the high-risk group (above the 90th percentile) was (52
3 %), mean age 15
4 (SD 0
7) years (range 14, 17
excluded, the trend remained significant. In addition, in a years), participated as volunteers in the study.
casecontrol study on 1322 Chinese boys and girls aged The study was carried out as part of a wider nutritional
0
16
9 years, He et al.(25) found an increase of one surveillance project that also included an investigation,
BMI unit associated with, on average, an increase of by means of a previously validated questionnaire(29), on
0
56 mmHg and 0
54 mmHg in SBP and DBP, respec- adolescents eating habits and behaviours, nutritional and
tively, for obese children, while in non-obese children, food safety knowledge and dietary beliefs.
the increase in SBP and DBP was 1
22 mmHg and We decided to select only the second-year high-school
1
20 mmHg, respectively. students, as we intend to follow up the adolescents
The significant correlation between BP levels and BMI through their senior year after a preventive and corrective
was also pointed out by Verma et al.(26). Furthermore, healthy education intervention aimed at promoting life-
these authors investigated which was the best determinant style changes, including achievement or maintenance of
of BP among absolute fat mass, body fat percentage, body normal body weight, dietary habits improvement and
fat distribution and BMI, and concluded that BMI was a increase in physical activity. Of the 17 high schools, nine
better predictor of BP than any other measure of body fat were located in Aosta town and eight in the countryside.
in nearly all analyses. In 902 healthy children and adoles- All students were surveyed over the course of four
cents aged 518 years, He et al.(27) found a significant months, from March to the end of June 2002.
positive relationship between SBP and DBP and trunk fat, The study was carried out with the cooperation of
adjusted for total fat, in boys by both dual-energy X-ray the school teachers board and the medical staff of the
absorptiometry and skinfold measurements, while in girls Aosta Valley Regional Public Health Department. Before
trunk fat was not a significant predictor of BP. Lurbe starting the study, a number of meetings were organised
et al.(28) studied a sample of 70 obese and 70 non-obese with teachers and students to explain the aim of the
children and adolescents, aged 616 years, and found that research and to request their participation. Informed
SBP and DBP were significantly higher among obese written consent was obtained from each student and
subjects, while waist:hip ratio (WHR) was independently their parents.
associated with SBP. Moreover, the authors pointed out The research protocol was approved by the Ethics
that obese subjects with a predominantly abdominal fat Committees of both Pavia University Medical School and
distribution showed higher BP values and concluded that the Aosta Valley Regional Public Health Department.
evaluation of body fat distribution in children and
adolescents may help to identify subjects more likely to Measurements
develop hypertension later in life. In addition, using step-
Body fat indices measurements
wise regression analyses, Smith and Rinderknecht(22)
After setting up a classroom in each school to conduct
showed that waist circumference, age and BMI were strong
medical assessments, students were examined by trained
predictors for SBP, while waist circumference and age were
health personnel (four physicians and one dietitian) who
predictors for DBP in the total sample.
had received 12 hours of instruction and were standar-
The aim of the present study was to investigate BP
dised in assessing measurements. A subset of adolescents
levels and their relationship with different body fat indi-
(57 subjects) was measured by all personnel to determine
ces in a selected group of adolescents in northern Italy, in
intra- and inter-operator variability, which resulted to be
order to evaluate the prevalence of hypertension in this
,4 % for all the measurements. The following parameters
group of adolescents and plan preventive and corrective
were measured.
strategies.
> Body weight, measured on subjects wearing only
underwear and without shoes, by means of a steelyard
Methods scale (precision 6100 g).
> Body height, measured on subjects without shoes by
Participants means of a stadiometer (precision 61 mm).
In this cross-sectional study, all students attending the > Four skinfold thicknesses (mid-triceps, mid-biceps,
second year of all high schools (n 5 17) in the Aosta subscapular and suprailiac), measured on subjects
Valley region (a mainly mountainous zone), northern according to standard conditions on the non-dominant
Italy, were informed about the research protocol (number body side using a Harpenden skinfold thickness
of subjects 5 889 individuals, equal to 0
74 % of the entire calliper (resolution 62 mm); three consecutive mea-
population (120 342 individuals) living in Aosta Valley surements were performed and the mean of the three
region on 31 December 2001). values was considered.
Overweight and blood pressure in adolescents 907
> Waist circumference, measured to the nearest mm in students gave to their parents. Familial hypertension
duplicate according to standard conditions, by placing history was considered positive for at least one parent
a flexible tape midway between the lowest rib and the suffering from hypertension.
iliac crest. The tape was snug, but did not squeeze or
compress the skin, and was parallel to the floor. The
Cut-off values
measure was collected on unclothed, relaxed subjects,
In accordance with the International Obesity Task Force
after exhaling.
(IOTF), Cole et al.s cut-off point reference standards for
> Hip circumference, measured to the nearest mm in
BMI(3) were used to identify overweight and obesity in
duplicate according to standard conditions, at the
young age. Coles percentile curves were drawn so that at
maximum extension of the buttocks with a flexible
age 18 years they passed through the widely used cut-off
tape on the subject wearing only underwear.
points of 25 and 30 kg/m2 for overweight and obese
BMI was calculated as the ratio between weight (in adults. The resulting curves were averaged to provide
kilograms) and the square of height (in metres). The sum age- and sex-specific cut-off points from 2 to 18 years(3).
of the four skinfold thicknesses was computed and the The IOTF(3234) suggests that Coles cut-off points are less
body fat percentage was calculated according to the arbitrary and more internationally based than current
Weststrate and Deurenberg equation(30). alternatives (they were developed by measuring 97 876
males and 94 851 females from birth to 18 years of age
living in Brazil, Great Britain, Hong Kong, The Nether-
BP measurements
lands, Singapore and the USA) and will help to provide
SBP and DBP were measured according to standard
internationally comparable prevalence rates of over-
conditions on the right arm, with the subject in a supine
weight and obesity in children and adolescents.
position and after resting for at least 5 min. Measurements
In accordance with the report of the Fourth NHLBI
were made by auscultation with the use of a calibrated
(National Heart, Lung, and Blood Institute) Task Force on
mercury sphygmomanometer and a cuff appropriately
Blood Pressure Control in Children(35), hypertension was
sized for the subjects arm size. The cuff size was chosen
diagnosed when the value of SBP and/or DBP was above
to be as large as possible without having the elbow skin
the 95th reference standard percentile for age and sex,
crease obstruct the stethoscope. SBP was determined by
whereas levels between the 90th and the 95th percentile
the first Korotkoff sound, and DBP by the fifth Korotkoff
were considered high normal (or borderline or at risk).
sound. Three BP measurements were taken at intervals of
Normal percentiles of Menghetti et al.(31), constructed by
35 min and the mean of the three values was considered.
fitting a third-order polynomial model of BP on age and
BP was measured in the supine position because the
height using multiple regression analysis, after BP mea-
normal percentiles of Menghetti et al.(31), which we used
surements in 11 519 healthy subjects of both sexes, aged
as reference standard values, were obtained by measuring
517 years, living in various locations throughout Italy,
BP in the same position.
were used as the Italian reference standards.
Each student was examined and all measurements
were taken during school time; each examination took
about 40 min. Data analysis
Data (mean, SD) were analysed using the Statistical
Smoking and alcoholic drinking habits and Package for the Social Sciences version 10 for PC (SPSS
physical activity level Inc., Chicago, IL, USA); they were normally distributed
Adolescents smoking and alcoholic drinking habits, as and Pearsons correlation coefficients were computed to
well as physical activity level, were also investigated by analyse relationships between BP values and each of the
means of an interview conducted by the trained medical adiposity indices: BMI, body fat mass, waist and hip cir-
staff after the measurements were taken. The interview cumferences and WHR. Multiple regression analyses with
was aimed at investigating both the number of cigarettes the stepwise method were used to verify the association
habitually smoked and the number and kind of alcoholic between BP and the adiposity indices. Students unpaired
drinks habitually drunk per day or per week, as well as t test was used to compare BP values among very active
the number of hours of physical activity practised per day and sedentary subjects.
or per week.
Results
Parental investigation
Parental weight and height, for BMI calculation, educa- BMI and body fat indices
tion and socio-economic levels as well as family history Characteristics of the sample are shown in Table 1. Mean
of hypertension were requested from the parents by BMI was 21
9 (SD 3
4) kg/m2 and 21
0 (SD 2
9) kg/m2
means of a brief self-administered questionnaire that the for males and females, respectively; i.e. in the range of
908 G Turconi et al.
normal-weight subjects. Percentile distribution data indi- P ranging from ,0
05 to ,0
001, except for hip circum-
cated that males BMI values were higher than those of ference and WHR values.
females: 17
4 kg/m2 and 16
4 kg/m2 at the 5th percentile,
21
2 kg/m2 and 20
7 kg/m2 at the 50th percentile, and
28
9 kg/m2 and 26
4 kg/m2 at the 95th percentile for boys BP measurements
and girls, respectively. Table 3 shows the mean, SD and percentile values of SBP
Table 2 lists the mean, SD and 5th, 50th, 95th percentile and DBP for males and females, compared with the
values of body fat indices for males and females. Four reference standard values for the Italian young popula-
skinfold thicknesses, fat mass percentage, waist and hip tion provided by Menghetti et al.(31). Our values were
circumferences and WHR are provided. All the measure- higher than the Italian reference values(31) in both sexes.
ments were statistically different between genders, with The SBP value was between the 50th and 75th percentile

Table 1 Age and anthropometric characteristics of the sample

Males (n 5 254) Females (n 5 278)

Variable Mean SD Mean SD

Age (years) 15
5 0
7 15
4 0
7

Weight (kg) 65
4 11
5 55
7 9
2
Height (m) 1
73 0
07 1
63 0
07
BMI (kg/m2) 21
9 3
4 21
0 2
9
Normal-weight subjects BMI (kg/m2)* 21
2 2
8 20
1 2
5
Overweight subjects BMI (kg/m2)* 25
4 1
4 25
6 1
8
Obese subjects BMI (kg/m2)* 31
2 1
4 30
7 2
5

SD,standard deviation; BMI, body mass index.

*Prevalence of normal weight, overweight and obesity, respectively: 74
0 % (males) and 81
7 % (females), 20
9 % (males) and
14
7 % (females), 4
7 % (males) and 1
1 % (females).

Table 2 Adolescents body fat indices and percentile distribution

Males (n 5 254) Females (n 5 278)

Percentile Percentile

Variable Mean SD 5th 50th 95th Mean SD 5th 50th 95th

Triceps skinfold thickness (mm) 11
6 5
9 5
2 9
8 23
2 17
2 5
4 9
4 16
8 26
2
Biceps skinfold thickness (mm) 6
2 3
8 2
9 4
9 13
4 9
3 4
5 4
2 8
5 17
8
Subscapular skinfold thickness (mm) 10
0 4
3 5
8 8
8 17
7 11
6 4
0 7
0 10
9 19
8
Suprailiac skinfold thickness (mm) 14
4 9
8 5
1 11
5 36
4 17
1 7
8 7
7 16
0 34
4
Body fat mass (%) 19
3 5
6 11
6 18
7 30
5 23
3 4
4 16
1 23
6 30
0
Waist circumference (cm) 75
6 10
7 66
0 74
0 94
2 69
6 7
7 60
5 69
0 84
0
Hip circumference (cm) 90
7 10
5 79
0 90
0 105
2 90
5 9
1 79
0 90
0 104
0
Waist:hip ratio 0
85 0
36 0
76 0
83 0
92 0
79 0
48 0
70 0
77 0
85

SD, standard deviation.

Table 3 Blood pressure measurements obtained in the present study compared with reference standard values of Menghetti et al.(31)

Present study Reference standard values

SBP (mmHg) DBP (mm Hg) SBP (mm Hg) DBP (mm Hg)

M F M F M F M F

Mean 124 118 76 75 117 111 70 66

SD 11 11 7 7 11 11 9 9
Percentile
5th 110 100 60 60 96 94 56 52
10th 110 106 70 65 100 99 58 54
25th 120 110 70 70 108 103 63 60
50th 120 120 80 78 113 112 70 67
75th 130 130 80 80 120 121 74 73
90th 140 130 85 80 130 129 80 80
95th 140 140 90 85 133 134 85 82

SBP, systolic blood pressure; DBP, diastolic blood pressure; M, males; F, females; SD, standard deviation.
Overweight and blood pressure in adolescents 909
Table 4 Prevalence rates of hypertensive subjects by gender and weight status

Systolic hypertensive Diastolic hypertensive

n % n %

Males (n 5 254) 42 16
5 13 5
1

Females (n 5 278) 21 7
6 24 8
6
Total (n 5 534) 63 11
8 37 6
9
Normal-weight males (n 5 188) 19 10
1 9 4
8
Overweight plus obese males (n 5 65) 23 35
4 4 6
1
Normal-weight females (n 5 227) 11 4
8 21 9
2
Overweight plus obese females (n 5 44) 10 22
7 3 6
8

in males and between the 25th and 50th percentile in Smoking and alcoholic drinking habits and
females, while the DBP mean value in both sexes was physical activity level
between the 25th and 50th percentile. Only 26 subjects (4
9 % of the sample) reported being
Based on the Fourth Task Force Report cut-off values current smokers; therefore, no correlation could be
on hypertension(35), in the overall sample 11
8 % of the carried out between BP values and smoking.
adolescents were systolic hypertensive, while 6
9 % were Regarding alcoholic drinking habits, only 45 subjects
diastolic hypertensive (Table 4). In addition, the dis- (8
4 %) reported being habitual consumers of two glasses of
tribution of hypertensive adolescents among normal- wine per day, while the consumption of other alcoholic
weight subjects and overweight plus obese subjects for drinks was occasional. Therefore, the number of drinking
both sexes is provided. In males, the percentage of subjects was too small to carry out statistical analyses.
hypertensive adolescents was higher in the overweight Finally, lower BP values emerged in very active sub-
plus obese subjects both for SBP and DBP compared with jects (practising physical activity for 6 h or more per
normal-weight subjects; in females, the percentage of week, 18
5 % of subjects) compared with those measured
hypertensive subjects was higher in the overweight plus in sedentary subjects (practising no habitual physical
obese subjects for SBP, while for DBP the percentage of activity, 29
7 % of subjects), but without any significant
hypertensive subjects was higher in the normal-weight difference. Nevertheless, a negative trend was observed
subjects than in the overweight plus obese subjects. between both SBP and DBP levels and increasing
physical activity practice.
Relationships between BP levels and
adiposity indices
Discussion
In the linear correlation analysis, BMI and all the adiposity
indices, except WHR, were found to be significantly asso-
BMI data show the high prevalence of overweight sub-
ciated (P ranging from 0
05 to 0
001) with both SBP and
jects in both sexes, although our data are lower than
DBP, with r ranging from 0
152 to 0
359. BMI and body fat
those found in southern Italy(3639), consistent with a
mass had the strongest association with BP. Multiple
higher prevalence of overweight and obesity in that area.
regression analyses showed that BMI was the strongest
As far as body fat mass percentage is concerned, our data
predictor for SBP in males (P , 0
001, r 5 0
275) and
are higher than those reported by Schaefer et al.(40) in
females (P , 0
001, r 5 0
336). For DBP, body fat mass was
a German adolescent population (12 % body fat mass
the strongest predictor only in males (P , 0
001, r 5 0
359).
for males, 21 % for females) and also those reported
by Ellis(41) in a Caucasian male adolescent population
Parental investigation (14
8 %), supporting a greater risk, mainly for males, to
Adolescents BMI correlated positively to the same extent develop obesity, as already revealed by the higher pre-
with both the fathers (P , 0
01) and the mothers valence rates of overweight and obesity in this gender.
(P , 0
01) BMI in males and females, with r values equal Such a high percentage of overweight subjects is a
to 0
23 and 0
24, respectively. While BMI did not show a public health threat; preventive and counteractive strate-
significant correlation with parental socio-economic and gies need to be undertaken in school programmes aimed
educational levels, it nevertheless showed a positive trend at reducing this risk condition.
towards both parental low educational and low socio- With regard to body circumferences, our data are
economic levels. On the other hand, both SBP and DBP higher than those reported by other researchers in same
values showed no correlation or trend with parental BMI, age groups. In British adolescents waist circumference
education or socio-economic level. A positive trend values were 70
8 (SD 7
1) cm in males and 64
9 (SD 4
9) cm
emerged between BP levels (either SBP or DBP) and a in females(42). In Cuba the 50th percentile waist cir-
positive history of familial hypertension (data not shown). cumference values were equal to 68 cm for males and
910 G Turconi et al.
(43)
64 cm for females . On the other hand, our data are to the high prevalence rates of overweight and obesity, as
supported by studies carried out in the USA(42) and well as the high values of adiposity indices, which are
Spain(44) (50th percentile waist circumference values well-known risk factors for hypertension.
equal to 74 cm for males and 69 cm for females), showing The prevalence rates of systolic and diastolic hyper-
that our population is more similar to these ones. tensive adolescents are high, but they are mostly related
The present SBP and DBP values are higher (Table 5) to SBP (11
8 % of subjects). Comparison of our data with
than those reported as Italian reference standards(31) for a others (Table 6) shows higher values than those in ado-
young population aged 15 years. Concerning Menghetti lescent populations of different ages and in different
et al.s data(31), they show that Italian childrens and geographical areas(23,4952).
adolescents BP levels are slightly higher than those of Higher prevalence rates of hypertensive adolescents in
Americans, in contrast with the belief that Italian people overweight plus obese subjects (Table 4), when com-
are protected against cardiovascular risk factors, including pared with normal-weight subjects, support previous
hypertension, more than other populations. findings(1823,53) reporting that high BMI values are a
Our values are higher than those found some years ago predisposing factor in the development of hypertension.
by Maida et al.(45) in Sardinian adolescents aged 1115 Nevertheless, the percentage of hypertensive subjects
years and higher than those reported by Paradis et al.(46) found even in the group of normal-weight subjects,
in Canadian adolescents aged 16 years. On the other consistent with other studies(24,25), shows that hyperten-
hand, they are quite similar to those found in 1526 Bel- sion is an under-recognised clinical entity in children and
gian adolescents(47), aged 1217 years, who had been adolescents, while monitoring of BP levels might help in
selected from a high cardiovascular risk population. Even detecting occult hypertension at a young age and limiting
the 50th percentile values are higher than those reported the magnitude of cardiovascular risk. Indeed, hyperten-
in Brazilian adolescents of the same age(48). sion is the most common form of cardiovascular disease
The SBP and DBP 90th percentile values found in males and it is often associated with overweight and obesity.
in the present study are higher than those in the USA(31) Although it is less common in adolescents than in adults,
(128/79 mmHg) and in northern Europe(31) (137/77 mmHg) hypertension and the associated organ damage can and
in the same age group, while in females they are similar often do begin early in life. Unless reversed, this condi-
(126/80 mmHg in the USA(31) and 130/76 mmHg in northern tion could determine cardiovascular disease outcome in
Europe(31)). In males, the SBP 90th percentile value is adulthood. Consequently, for many with high BP levels,
particularly high (140 mmHg); it is even equal to the value measures directed at the long-term prevention of cardio-
considered as an indicator of first-degree hypertension in vascular morbidity may need to be started in adolescence
the adult population. These values in males may be due to achieve maximal effectiveness.

Table 5 Mean blood pressure values in the present sample compared with those found in other adolescent groups

SBP/DBP (mmHg)

M F Adolescents age (years) Reference

Our sample 124/76 118/75 15 Present study

Italian reference standards 117/70 111/66 15 Menghetti et al. (1999)(31)
Sardinia (Italy) 104/42 109/44 1115 Maida et al. (1982)(45)
Quebec (Canada) 124/61 114/62 16 Paradis et al. (2004)(46)
Luxembourg (Belgium) 125/74 122/74 1217 Paulus et al. (1999)(47)
Rio de Janeiro (Brazil) 117/61* 110/62* 15 Pozzan et al. (1997)(48)

SBP, systolic blood pressure; DBP, diastolic blood pressure; M, males; F, females.
*50th percentile values.

Table 6 Prevalence rates of systolic hypertensive adolescents (males plus females) in the present sample compared with those found in
other young groups

Age (years)

Prevalence rate (%) Mean SD Reference

Our sample 11
8 15
4 0
7 Present study

Rome (Italy) 6
5 1114- Menghetti et al. (2004)(49)
Maceio (Brazil) 9
4 12
4 2
9 Moura et al. (2004)(50)
Corrients City (Spain) 3
0 14
8 1
6 Martinez et al. (2001)(23)
Debrecen (Hungary) 2
5 1518- Pall et al. (2005)(51)
Ludhiana (India) 6
7/2
6* 1117- Mohan et al. (2004)(52)

SD, standard deviation.

*In urban/rural areas.
-Age range (years).
Overweight and blood pressure in adolescents 911
Even though BMI and all the adiposity indices were The high values of BP and hypertension rates, together
found to be significantly associated with BP levels in both with high prevalence of overweight plus obese subjects
sexes, BMI and body fat mass showed the strongest found in our sample, are worrying. There is a pressing
association, most closely related to SBP in both genders need to develop a comprehensive medical and nutrition
and to DBP only in males, in agreement with the literature plan together with preventive and corrective strategies in
data indicating that BMI contributes significantly to BP school programmes to reduce the prevalence of these
variability(1921,24,27,46), even in non-obese subjects(24,25). identified disease states, by empowering teens to increase
The correlation between parental and adolescents BMI physical activity and to improve their own dietary habits
values supports previous findings(18,5456) showing an together with their lifestyle.
influence of both parents BMI on their children. In par-
ticular, the results are in agreement with the data obtained
Acknowledgements
in a previous study carried out in southern Italy(57) in
which childrens BMI correlated with both fathers and
The authors of this paper state that: there are no financial
mothers BMI to the same extent.
or other contractual agreements that might cause conflicts
Although no significant correlation between adoles-
of interest; the research obtained funding from the Italian
cents BMI and parental socio-economic and education
Ministry of Health; and each author has participated
levels emerged, a positive trend was found showing that
actively in the work and has given substantial contri-
these factors may influence weight condition, as reported
bution: G.T. project of the study, data analysis and
by other authors(54,58). In addition, adolescents risk of
supervisor; R.B. supervisor in the assessment of
becoming overweight increased with parental overweight
anthropometric measurements; L.M. statistical analysis;
and obesity, thus showing that familial condition (genet-
C.R. supervisor. The Italian Ministry of Health is thanked
ics, unhealthy dietary habits together with a sedentary
for funding the research.
lifestyle) has to be taken into account to identify risk
groups for preventive measures. It is interesting to point
out that foods such as cheese rich in fats, salami, sau- References
sages, butter and lard are traditional components of local
meals in this region (data not presented). 1. World Health Organization (1998) Obesity: Preventing and
Parental BMI does not seem to influence childrens BP Managing the Global Epidemic. Report of a WHO
levels in our sample; it is more likely that parental history Consultation, Geneva, 35 June 1997. Geneva: WHO.
2. Pi-Sunyer FX, Laferrere B, Aronne LJ & Bray GA (1999)
of hypertension, rather than parental BMI, may influence Therapeutic controversy: obesity a modern day epidemic.
childrens BP levels since a positive trend emerged J Clin Endocrinol Metab 84, 312.
between students BP levels and a positive history of 3. Cole TJ, Bellizzi MC, Flegal KM & Dietz WH (2000)
familial hypertension. Neither parental socio-economic Establishing a standard definition for child overweight
and obesity worldwide: international survey. BMJ 320,
level nor education level influenced childrens BP levels 12401243.
in our study. 4. World Health Organization (2003) Global Strategy on
The data reported by the students about smoking are Diet, Physical Activity and Health. http://www.who.int/
doubtful since a recent investigation(59) showed an dietphysicalactivity/en/ (accessed September 2007).
5. Livingstone B (2000) Epidemiology of childhood obesity in
increase of smokers among adolescents aged 1416 years Europe. Eur J Pediatr 159, Suppl. 1, S14S34.
equal to 33 % in males and 69 % in females during a 6. Parsons TJ, Power C, Logan S & Summerbell CD (1999)
six-year period (19931999). Childhood predictors of adult obesity: a systematic review.
The fact that 45 subjects report being habitual con- Int J Obes Relat Metab Disord 23, Suppl. 8, S1S107.
7. Kiess W, Reich A, Muller G, Meyer K, Galler A, Bennek J &
sumers of two glasses of wine per day is not surprising, Kratzsch J (2001) Clinical aspects of obesity in childhood
since alcohol consumption is widespread in this region, and adolescence diagnosis, treatment and prevention.
while we expected more physical activity practice in a Int J Obes Relat Metab Disord 25, Suppl. 1, S75S79.
mainly mountainous zone. Despite this, a negative trend 8. Burke V, Beilin LJ & Dunbar D (2001) Tracking of blood
pressure in Australian children. J Hypertens 19, 11851192.
was observed between BP levels and increasing physical 9. Chu NF, Wang DG & Shieh SM (2001) Obesity, leptin and
activity practice, confirming the positive role of habitual blood pressure among children in Twain: the Taipei
physical activity in preventing hypertension. childrens heart study. Am J Hypertens 14, 135140.
10. McGill HC Jr, McMahan CA, Malcolm GT, Oalmann MC &
Strong JP (1995) The Pathobiological Determinants of
Atherosclerosis in Youth (PDAY) Research Group. Relation
Conclusion of glycohemoglobin and adiposity to atherosclerosis in
youth. Arterioscler Thromb Vasc Biol 15, 431440.
The limitation of this study arises from the fact that we 11. Srinivasan SR, Meyers L & Berenson GS (2001) Rate of
change in adiposity and its relationship to concomitant
examined subjects of a selected age group and therefore the changes in cardiovascular risk variables among biracial
results may not be extended to a wider range of adoles- (blackwhite children) and young adults: the Bogalusa
cents, even though we explained the reason for our choice. heart study. Metabolism 50, 299305.
912 G Turconi et al.
12. Geiss HC, Parhofer KG & Schwandt P (2001) Parameters of 30. Weststrate JA & Deurenberg P (1989) Body composition in
childhood obesity and their relationship to cardiovascular children: proposal for a method for calculating body fat
risk factors in healthy prepubescent children. Int J Obes percentage from total body density or skinfold thickness
Relat Metab Disord 25, 830837. measurements. Am J Clin Nutr 50, 11041115.
13. Frontini MG, Bao W, Elkasabany A, Srinivasan SR & 31. Menghetti E, Virdis R, Strambi M, Patriarca V, Riccioni MA,
Berenson G (2001) Comparison of weight-for-height Fossali E & Spagnolo A (1999) Blood pressure in childhood
indices as a measure of adiposity and cardiovascular risk and adolescence: the Italian normal standards. Study group
from childhood to young adulthood. J Clin Epidemiol 54, on Hypertension of the Italian Society of Pediatrics.
817822. J Hypertens 17, 13631372.
14. Young-Hyman D, Schlundt DG, Herman L, De Luca F & 32. Bellizzi MC & Dietz WH (1999) Workshop on childhood
Counts D (2001) Evaluation of the insulin resistance obesity: summary of the discussion. Am J Clin Nutr 70,
syndrome in 5- to 10-year-old overweight/obese African- 173S175S.
American children. Diabetes Care 24, 13591364. 33. Dietz WH & Robinson TN (1998) Use of the body mass
15. Pinhas-Hamiel O, Dolan LM, Daniels SR, Standiford D, index (BMI) as a measure of overweight in children and
Khoury PR & Zeitler P (1996) Increased incidence of non- adolescents. J Pediatr 132, 191193.
insulin-dependent diabetes mellitus among adolescents. 34. International Obesity Task Force (2003) Establishing a
J Pediatr 128, 608615. standard definition for child overweight and obesity
16. Power C, Lake JK & Cole TJ (1997) Measurements of long- worldwide: international survey. In Childhood Obesity.
term health risks of child and adolescent fatness. Int J Obes http://www.iotf.org/childhoodobesity.asp (accessed Sep-
Relat Metab Disord 21, 507526. tember 2007).
17. Maffeis C & Tato L (2001) Long-term effects of childhood 35. National High Blood Pressure Education Program
obesity on morbidity and mortality. Horm Res 55, Suppl. 1, Working Group on High Blood Pressure in Children and
4245. Adolescents (2004) The fourth report on the diagnosis,
18. Buonomo E, Pasquarella A & Palombi L (1996) Blood evaluation, and treatment of high blood pressure in
pressure and anthropometry in parents and children of a children and adolescents. Pediatrics 114, 2 Suppl. 4th
southern Italian village. J Hum Hypertens 10, Suppl. 3, Report, 555576.
S77S79. 36. Grispan A, Toselli S, Fabiani M, Bontempo V, Liberatore S,
19. Leccia G, Marotta T, Masella MR, Mottola G, Mitrano G, Gagliardi C & Tarsitani G (1997) Prevalenza dellobesita in
Golia F, Capitanata P, Guida L, Contaldo F & Ferrara LA eta evolutiva Nota 2: studio epidemiologico nelle scuole
(1999) Sex-related influence of body size and sexual medie inferiori di LAquila. Ann Ig 9, 119125.
maturation on blood pressure in adolescents. Eur J Clin 37. De Vito E, La Torre G, Langiano E, Berardi D & Ricciardi G
Nutr 53, 333337. (1999) Overweight and obesity among secondary school
20. Koziel S, Kolodziej H & Ulijaszek S (2001) Body size, fat children in Central Italy. Eur J Epidemiol 15, 649654.
distribution, menarcheal age and blood pressure in 38. Romiti A, Salandri A, Imbrogno N, De Angelis G, Ferraro A
14-year-old girls. Eur J Epidemiol 17, 11111115. & Di Mario U (2000) Prevalence of obesity in adolescence.
21. Uscategui Penuela RM, Perez Giraldo JA, Aristizabal Rivera Clin Ter 151, 335339.
JC & Camacho Perez JA (2003) Excess of weight and their 39. DArgenio P, Citarella A, Palombi EL, Aversano MP, Bassi R,
relationship with high blood pressure in schoolchildren Di Fiore A. (2001) Obesita e sovrappeso tra i pre-
and adolescents of Medellin, Colombia. Arch Latinoam adolescenti. Uno studio dalla provincia di Benevento.
Nutr 53, 376382. BEN-Notiziario ISS 14(1); available at http://www.ben.
22. Smith C & Rinderknecht K (2003) Obesity correlates with iss.it/precedenti/gennaio/2.htm
increased blood pressures in urban native American youth. 40. Schaefer F, Georgi M, Wuhl E & Scharer K (1998) Body
Am J Hum Biol 15, 7890. mass index and percentage fat mass in healthy German
23. Martinez CA, Ibanez JO, Paterno CA, De Roig Bustamante schoolchildren and adolescents. Int J Obes Relat Metab
MS, Itati Heitz M, Kriskovich Jure JO, De Bonis GR & Disord 22, 461469.
Caceres LC (2001) Overweight and obesity in children 41. Ellis KJ (2000) Human body composition: in vivo methods.
and adolescents of Corrients city. Relationship with Physiol Rev 80, 649680.
cardiovascular risk factors. Medicina (B Aires) 61, 42. McCarthy HD, Jarrett KV & Crawley HF (2001) The
308314. development of waist circumference percentiles in British
+
24. Reich A, Muller G, Gelbrich G, Deutscher K, Godicke R & children aged 5.016.9 y. Eur J Clin Nutr 55, 902907.
Kiess W (2003) Obesity and blood pressure results from 43. Martinez E, Devesa M, Bacallao J & Amador M (1994)
the examination of 2365 schoolchildren in Germany. Int J Percentiles of the waisthip ratio in Cuban scholars aged
Obes Relat Metab Disord 27, 14591464. 4.5 to 20.5 y. Int J Obes Relat Metab Disord 18, 557560.
25. He Q, Ding ZY, Fong DY & Karlberg J (2000) Blood 44. Moreno LA, Fleta J, Mur L, Rodriguez G, Sarria A & Bueno
pressure is associated with body mass index in both normal M (1999) Waist circumference values in Spanish children
and obese children. Hypertension 36, 165170. gender-related differences. Eur J Clin Nutr 53, 429433.
26. Verma M, Chhatwal J & George SM (1995) Biophysical 45. Maida A, Mura I, Muresu E & Romano G (1982) Obesity and
profile of blood pressure in school children. Indian Pediatr humoral and pressure indicators in a sample population
32, 749754. between 6 and 20 years of age from the province of Sassari.
27. He Q, Horlick M, Fedun B, Wang J, Pierson RN Jr, Heshka S Nuovi Ann Ig Microbiol 33, 637654.
& Gallagher D (2002) Trunk fat and blood pressure in 46. Paradis G, Lambert M, OLoughlin J, Lavallee C, Aubin J,
children through puberty. Circulation 105, 10931098. Delvin E, Levy E & Hanley JA (2004) Blood pressure and
28. Lurbe E, Alvarez V, Liao Y, Tacons J, Cooper R, Cremades adiposity in children and adolescents. Circulation 110,
B, Torro I & Redon J (1998) The impact of obesity and body 18321838.
fat distribution on ambulatory blood pressure in children 47. Paulus D, Saint-Remy A & Jeanjean M (1999) Blood
and adolescents. Am J Hypertens 11, 418424. pressure during adolescence: a study among Belgian
29. Turconi G, Celsa M, Rezzani C, Biino G, Sartirana MA & adolescents selected from a high cardiovascular risk
Roggi C (2003) Reliability of a dietary questionnaire on population. Eur J Epidemiol 15, 783790.
food habits, eating behaviour and nutritional knowledge of 48. Pozzan R, Brandao AA, Da Silva SL & Brandao AP (1997)
adolescents. Eur J Clin Nutr 57, 753763. Hyperglycemia, hyperinsulinemia, overweight, and high
Overweight and blood pressure in adolescents 913
blood pressure in young adults: the Rio de Janeiro Study. 54. Krassas GE, Tzotzas T, Tsametis C & Konstantinidis T
Hypertension 30, 650653. (2001) Determinants of body mass index in Greek children
49. Menghetti E, DAddesa D, Censi L, Spagnolo A, Martone D, and adolescents. J Pediatr Endocrinol Metab 14, Suppl. 5,
Cellitti R & Sette S (2004) Hypertension in schoolchildren: S1327S1333.
research carried out in a secondary school in Rome and 55. Pietilainen KH, Kaprio J, Rasanen M, Winter T, Rissanen A
observations on dietary patterns. Minerva Pediatr 56, & Rose RJ (2001) Tracking of body size from birth to late
311316. adolescence: contributions of birth length, birth weight,
50. Moura AA, Silva MA, Ferraz MR & Rivera IR (2004) duration of gestation, parents body size, and twinship. Am
Prevalence of high blood pressure in children and J Epidemiol 154, 2129.
adolescents from the city of Maceio, Brazil. J Pediatr (Rio 56. Burke V, Beilin LJ & Dunbar D (2001) Family lifestyle and
J) 80, 3540. parental body mass index as predictors of body mass index
51. Pall D, Katona E, Paragh G, Zrinyi M, Zatik J & in Australian children: a longitudinal study. Int J Obes Relat
Fulesdi B (2005) Epidemiological data of 1518 year Metab Disord 25, 147157.
adolescents and the prevalence of hypertension in 57. Esposito-Del Puente A, Scalfi L, De Filippo E et al. (1994)
Debrecen. The Debrecen hypertension study. Orv Hetil Familial and environmental influences on body composi-
146, 127132. tion and body fat distribution in childhood in Southern
52. Mohan B, Kumar N, Aslam N, Rangbulla A, Kumbkarni S, Italy. Int J Obes Relat Metab Disord 18, 596601.
Sood NK & Wander GS (2004) Prevalence of sustained 58. McMurray RG, Harrell JS, Deng S, Bradley CB, Cox LM &
hypertension and obesity in urban and rural school going Bangdiwala SI (2000) The influence of physical activity,
children in Ludhiana. Indian Heart J 56, 310314. socioeconomic status, and ethnicity on the weight status of
53. Burke V, Beilin LJ, Dunbar D & Kevan M (2004) adolescents. Obes Res 8, 130139.
Associations between blood pressure and overweight 59. Istituto Superiore di Sanita, Italy (2003) I dati sul fumo.
defined by new standards for body mass index in child- http://www.epicentro.iss.it/temi/fumo/dati.asp (accessed
hood. Prev Med 38, 558564. September 2007).