Obesity, Body Fat Distribution,

and Ambulatory Blood Pressure

in Children and Adolescents
Empar Lurbe, MD; 1 Vicente Alvarez, MD; 1 Josep Redon, MD 2

Obesity is a common disease with an ever-increas- population, and about 50% of those over 50 years
ing prevalence and usually with late-onset conse- of age, are overweight or obese.1 Although many
quences. If acquired during childhood, it tracks into of the associated medical problems are deferred to
adult life to some extent, and since the relationship adult life, an important proportion of adult obesi-
between obesity and hypertension is well estab- ty has its origins in childhood.1–3 It appears that
lished in adults, obese children appear to be at par- nearly 50% of overweight children become over-
ticularly high risk of becoming hypertensive adults. weight adults. These figures are of great medical
In the authors’study, obese children seemed to have concern, as obesity is a risk factor for a wide vari-
significantly higher casual and ambulatory blood ety of conditions and is associated with myriad co-
pressure than nonobese children, except for night- morbid diseases.
time diastolic blood pressure. The health effects of Substantial clinical and epidemiologic evidence
obesity may depend on the anatomic distribution of supports the influence of obesity on blood pres-
body fat, which in turn may be a better indicator of sure (BP) levels, even early in life.4 The association
endocrinologic imbalance, environmental stress, or between obesity and hypertension has been well
genetic factors than is fatness per se. Subjects with a documented.5 Hypertension is one of the most im-
higher waist-to-hip ratio or a larger waist, as an es- portant obesity-related cardiovascular risk factors
timate of central obesity, tend to have higher blood for two reasons: obesity-induced hypertension is
pressure values even during childhood. Prevention very prevalent, and the lowering of BP levels is the
of the onset of obesity in early life may be impor- most cost-effective way to reduce cardiovascular
tant to reducing the risk of coronary heart disease morbidity and mortality. Factors associated with
in later life. (J Clin Hypertens. 2001;3:362–367). childhood BP may provide important insights into
©2001 Le Jacq Communications, Inc. adult hypertension. Obesity acquired during child-
hood tracks, to some extent, into adult life,6 and
since the relationship between obesity and hyper-

T he prevalence of obesity is high and increas-

ing, and the disorder usually has late-onset
consequences. In westernized societies, it is cur-
tension is well established in adults,7 obese chil-
dren appear to be at particularly high risk of
becoming hypertensive adults.
rently estimated that approximately 35% of the The type of obesity is an important determinant
of the prevalence of hypertension, and subjects
with central or abdominal obesity have the highest
From the Pediatric Nephrology Unit, Hospital General;1
and the Hypertension Clinic, Hospital Clinico,2 risk of developing hypertension, along with other
University of Valencia, Valencia, Spain cardiovascular risk factors. Although the waist/hip
Address for correspondence/reprint requests: (W/H) ratio has been shown to be a useful marker
Empar Lurbe, MD, Pediatric Nephrology Unit, of abdominal obesity in adults,8 waist circumfer-
Department of Pediatrics, Hospital General, ence is increasingly used as the best anthropomet-
University of Valencia, Avda Tres Cruces s/n,
46014, Valencia, Spain ric alternative to the W/H ratio. How body fat
Manuscript received December 27, 2000; distribution patterns develop in childhood, and
accepted April 18, 2001 which index is the more appropriate for measur-

362 THE JOURNAL OF CLINICAL HYPERTENSION VOL. III NO. VI NOVEMBER/DECEMBER 2001
ing the relationship between body fat distribution overestimates systolic BP by at least 5 mm Hg. Con-
and potential cardiovascular risk factors in chil- sequently, it is important to closely follow the guide-
dren, are not clear.9–12 lines when selecting cuff size.18
Measurement of BP is also influenced by the in-
METHODOLOGIC CONSIDERATIONS strument and the individual taking the measure-
To establish the relationship between body size and ment, by the subject’s physical activity and behavior,
BP, we need to bear in mind some caveats pertaining and by the setting in which the BP value is obtained.
to the parameters that are regularly used to estimate Most researchers agree that values obtained from
body size and the relationship between body size and 24-hour monitoring are more accurate in indicating
BP in children and adolescents. Because body weight cardiovascular risk than are those obtained from ca-
and height change during growth and development, sual measurements.19–21 The use of ambulatory BP
the best parameters for defining obesity and for as- monitoring is feasible in a pediatric population, with
sessing the relationship of body size to BP during the application of carefully standardized and special-
childhood are under debate. In our studies, and using ly adapted recording techniques.22 Under these con-
local tables, we have defined childhood obesity in ditions, one feature of ambulatory BP values is their
terms of the body mass index and tricipital and sub- superior reproducibility relative to those obtained
scapular skinfold thickness above the age and sex- with casual measurement.23,24 Moreover, automatic
specific 95th percentile.13 Although there is no ambulatory BP monitoring provides more accurate
established cut-off point for childhood obesity, we BP values and records changes in BP whether sub-
have used the most stringent criteria of those recom- jects are active or asleep.23 It is a useful research
mended in a number of previously published guide- tool, although its use in clinical pediatric practice
lines, and define obesity as higher than the 95th awaits more precise information.
percentile of body mass index.14,15 For a 10-year-old
boy, the threshold for obesity is 21 kg/m2 (46 lbs), BP AND GROWTH
while for a 12-year-old girl, it is 24 kg/m2 (53 lbs). Among all that is known about the levels and distri-
At the time of our first study, the ponderal bution of casual BP measurement in children and
index (kg/m 3 )was used to establish a relation- adolescents, it is well recognized that BP increases
ship between BP and the estimates of body size, during growth and maturation, and that adoles-
because this index has less collinearity with cence is a fast-growth period during which body
height than does the body mass index. Any mass and BP change rapidly. During the last few
three-dimensional structure that increases in size decades, this was the main reason that reference BP
while its configuration and density remain the values were referred to as specific values for age or
same will maintain a constant ratio of weight height and sex in children up to 18 years of age.
to height cubed. The ratio of weight to height Recently, the Task Force for Blood Pressure in Chil-
squared, however, increases proportionally to dren has become aware of the importance of con-
height squared. Similar reasoning applies to the sidering age and height together when defining the
human body; therefore, the ratio of weight to reference values.25 In children of the same age, the
height cubed was preferred.16 upper limit of normal systolic BP for the 5th per-
Knowledge of the factors that affect the measure- centile of height is 8–9 mm Hg lower than the limit
ment of BP is important not only to the epidemiolo- for the 95th percentile (Table).
gist, who is primarily interested in identifying Ambulatory BP in children and adolescents has
precursors of high blood pressure, but also to the not been sufficiently assessed, in part because ambula-
clinician, who relies on epidemiologic data to define tory monitoring has only recently been introduced in
the normal range of BP in children. A major impedi- this population. The distribution of ambulatory BP
ment to the standardization of BP is variation in cuff values in both healthy, normotensive children22 and
size.17 The potential for error in measuring BP may healthy, school-based populations26 was used to ob-
be greater in children, whose wide range of arm tain an approach to reference values. The average sys-
sizes makes selection of the proper cuff critical. Cuff tolic BPs progressively increased from the 50th to the
size is important in children because as the upper 95th percentile during the daytime, nighttime, and
arm grows, cuff artifacts will cause systematic errors 24-hour periods in both boys and girls. The rate of
in measurement that can lead to wrongly labeling a BP increase across age and height ranges, however,
child as hypertensive or falsely reassuring another was lower than that observed in the casual BP read-
whose BP is truly elevated. For an obese child with a ings. In contrast, the average values for diastolic BP
28-cm arm circumference, the 12-cm bladder width did not change with age or height, regardless of the
is the right size; the use of an 8-cm bladder width time period considered.

VOL. III NO. VI NOVEMBER/DECEMBER 2001 THE JOURNAL OF CLINICAL HYPERTENSION 363
 Table. 95th Percentiles of Blood Pressure According to Age, Sex, and Height

BLOOD PRESSURE AGE HEIGHT PERCENTILES HEIGHT PERCENTILES

(YEARS) (BOYS) (GIRLS)

5TH 95TH 5TH 95TH

Systolic 3 104 113 104 110
6 109 117 108 114
10 114 123 116 122
13 121 130 121 128
16 129 138 125 132

Diastolic 3 63 67 65 68
6 72 76 71 75
10 77 82 77 80
13 79 84 80 84
16 83 87 83 86

OBESITY AND BLOOD PRESSURE and nighttime periods. In contrast, diastolic BP did
Our group studied the characteristics of ambulatory not differ during the nighttime period, and only small
BP in obese children and adolescents. Ambulatory BP differences were observed during the daytime (Figure
monitoring was performed to examine the impact of 1, panel B). The physiologic BP fall that normally oc-
obesity and body fat distribution on BP in children curs during sleep was present in equal magnitude in
and adolescents. Although the use of ambulatory BP the two groups, for both systolic and diastolic BP.
monitoring is feasible in a pediatric population,22,26,27 The impact of obesity on BP is independent of
studies involving this technique in obese children have other parameters, as demonstrated in a previous
been scarce. study.4 Adjusting for height and sex had a minimal
Except for nighttime diastolic BP, obese children effect on the systolic BP differences between the
had significantly higher casual and ambulatory BP obese and nonobese children. After further control-
than did nonobese children, regardless of the period ling for the ponderal index, the differences in casual
analyzed. The systolic BP values for the obese and and ambulatory BP between obese and nonobese
nonobese children are shown in Figure 1 (panel A). children were no longer significant. An additional
The obese children had higher BP values over 24 decrease in these differences was seen when tricipi-
hours than did the nonobese subjects, and the magni- tal skinfold thickness, one of the measurements for
tude of the differences was similar during the daytime estimating obesity, was considered.

A 90 Diastolic
B
150 Systolic
Systolic BP
BP Diastolic BP
BP
140
80
130
70
120

110 60

100
50
90

80 40

70
30
Casual
Casual 24-hour
24-hour Daytime
Daytime Nighttime Casual
Casual 24 hour
24-hour Daytime
Daytime Nighttime
Nighttime

Figure 1. Box plot of casual blood pressure (BP) measurement and ambulatory monitoring of 24-hour, daytime, and night-
time systolic BP (panel A) and diastolic BP (panel B) in obese (n=85, shaded boxes) and nonobese (n=88, filled boxes) chil-
dren and adolescents. Casual systolic and diastolic BP were significantly higher in obese than in nonobese subjects (p<0.01
and p<0.05, respectively). Systolic BP over 24 hours was higher in obese than in nonobese children (p<0.01), and the magni-
tude of the differences was similar during the daytime (p<0.01) and nighttime periods (p<0.05). In contrast, diastolic BP did
not significantly differ during the nighttime period, and only small differences were observed during the day (p<0.05).

364 THE JOURNAL OF CLINICAL HYPERTENSION VOL. III NO. VI NOVEMBER/DECEMBER 2001
 Casual and ambulatory systolic BP, day and among studies may be attributable to the morpho-
night, were positively correlated with height, logic changes operating during pubertal develop-
weight, the ponderal index, tricipital and sub- ment,11,12 as well as the studies’ subject inclusion
scapular skinfold thickness, and the W/H ratio. criteria.
These correlations indicated that systolic BP was The impact of body fat distribution, as estimat-
related to more than just body size parameters. All ed from the W/H ratio, on ambulatory BP values
examined indices of obesity were positively related has been studied in children by our group.4 Our
to systolic BP, and a significant relationship to data indicate that subjects with a higher W/H
body fat distribution was also found. In a multiple ratio tended to have higher BP values even during
regression analysis, the W/H ratio, rather than obe- childhood. Taking into account that the redistrib-
sity or body size parameters, was the main deter- ution of body fat from leg to trunk occurs during
minant of ambulatory systolic BP. adolescence and young adulthood, the presence of
a high W/H ratio and high systolic BP during
FAT DISTRIBUTION AND BP childhood can predispose one to develop cardio-
The importance of body fat distribution derives vascular risk later in life.
from the association of obesity with the chronic dis- The use of waist circumference, rather than the
eases of well fed societies. Two individuals of simi- W/H ratio, for estimating central obesity has been
lar body weight, with similar skinfold thickness emphasized in recent years. The waist circumfer-
and/or percent body fat, can have a very different ence can express abdominal fat accumulation bet-
anatomic distribution of subcutaneous fat. It is ter than the W/H ratio does, in part because in
known that the health effects of obesity depend on growing children the hip may reflect changes in
the anatomic distribution of body fat, which in turn bone and muscle more than changes in fat.28 In
may be a better indicator of endocrinologic imbal- our study group of obese and nonobese children
ance, environmental stress, or genetic factors than and adolescents, waist dimension was indepen-
is fatness per se. dently related to the average of 24-hour systolic
The W/H ratio has been shown to be a useful BP when age, sex, height, and weight were taken
marker of abdominal obesity in adults,7 and it is into account. The increased relation of waist to
also informative in postpubertal boys and girls. 24-hour systolic BP, as shown in Figure 2, empha-
However, the use of the W/H in younger children sizes the relationship of waist to cardiovascular
has been questioned. The discrepancies observed risk factors.

140 140

130 130

120 120
Systolic BP (mmHg)

(mmHg)
Systolic BP (mmHg)

(mmHg)

110 110
BP BP
Systolic

100 100
r=0.30
Systolic

r=0.30 r=0.22
90 90
p<0.0001
p<0.0001 p<0.005
p<0.005
80 80
40 50 60 70 80 90 100 110 120 0 20 40 60 80 100 120

WAIST (cm) WEIGHT (kg)

WEIGH(KG)

140 140

130 130
(mmHg)
Systolic BP (mmHg)
Systolic BP (mmHg)

120
(mmHg)

120

110
BPBP

110
Systolic
Systolic

100 100

r=0.27
r=0.27 r=0.20
90 90
p<0.0001 p<0.01
p<0.01
p<0.0001
80 80
10 20 30 40 100 120 140 160 180 200

BMI
BMI (kg/m
(kg/m2)
2) HEIGHT(cm)
HEIGHT (cm)

Figure 2. Regression lines and the 95% confidence interval between the average 24-hour systolic blood pressure (BP) and sev-
eral anthropometric parameters: waist measurement, weight, body mass index (BMI), and height. BPs tend to be higher with
increased weight, waist, and height measurements. r=Pearson correlation coefficient

VOL. III NO. VI NOVEMBER/DECEMBER 2001 THE JOURNAL OF CLINICAL HYPERTENSION 365
 MECHANISMS OF OBESITY-RELATED not the different values observed in the obese children
HIGH BP denote a subset whose risk for developing hyperten-
When obesity-associated hypertension is character- sion may be different is an attractive hypothesis to be
ized principally by increased vascular volume, and explored in further studies.
when peripheral vascular resistance is “abnormally An understanding of the mechanisms responsible
normal” or only slightly elevated, volume expansion for the resetting of pressor natriuresis, inducing a
appears to be a key determinant in the tendency to- permanent expansion of extracellular fluid and
ward higher BP values in obese children and adoles- blood volume, is essential for the elucidation of the
cents. This relationship implies that the primary pathophysiologic link between overweight and high
defect leading to increased vascular volume and car- BP. It is possible that some common mechanisms
diac output, and ultimately to hypertension, consists may affect renal sodium handling and peripheral
of some abnormality in kidney function that is re- vascular resistance, the pathophysiology of the ab-
sponsible for a shift in pressor natriuresis toward normal tubular sodium retention, and the lack of a
higher BP values. Therefore, obesity-induced hyper- normal adaptation of peripheral vascular resistance
tension should be considered a salt-sensitive state.29 to increased cardiac output.32 Several common fac-
Rocchini and colleagues30 observed the shift in tors are involved in the establishment of both sodi-
the pressure-natriuresis curve in obese adolescents. um retention and vascular resistance, and this may
Sixty obese and 18 nonobese adolescents were eval- be, at least in part, critically influenced by the neu-
uated after successive 2-week periods of a high-salt robiologic/genetic mechanisms producing obesity.33
diet of >250 mmol/day of sodium and a low-salt Three mechanisms appear to be especially impor-
diet of <30 mmol/day. The mean arterial pressures tant in initiating the increased sodium reabsorption,
for the two groups were compared as they switched impaired renal pressure natriuresis, and hypertension
from high- to low-sodium diets. Mean pressures un- associated with weight gain. These mechanisms are
derwent a significantly larger decrease in the obese an increase in sympathetic activity, activation of the
than in the nonobese group. In addition, obese ado- renin-angiotensin system, and alteration of intrarenal
lescents had a renal function plot (urinary sodium physical forces due to compression of the kidneys.
excretion as a function of arterial pressure) with a The effect of obesity on the relationship between BP
more gradual slope than that of nonobese adoles- and sodium excretion was evident during sleeping
cents. When the renal function relationship was hours, not during waking hours, and it may reflect
normalized for weight loss, BP was no longer sensi- circadian rhythm alteration by several mechanisms.
tive to dietary sodium intake. Sympathetic nerve activity or the renin-angiotensin
In a previous study,31 our group evaluated system can alter renal sodium handling and/or BP.
whether obesity influences the relationship between Under normal conditions, sympathetic activity de-
BP and urinary sodium excretion by using ambulato- creases during the night and, in the absence of de-
ry BP monitoring and, simultaneously, split urine col- manding situations, basal overactivity can be detected
lections during the waking and sleeping periods. We more easily and may provide early evidence of patho-
compared the age- and sex-adjusted mean BP be- physiologic changes that predispose obese individuals
tween obese and nonobese children within each time- to hypertension.
specific quartile of sodium excretion rate. Obese Obesity, and specifically abdominal obesity, is
children, at the same urinary sodium excretion levels linked to the presence of insulin resistance and hyper-
as nonobese children, had a higher ambulatory sys- insulinemia. The role of insulin in the development of
tolic BP than the nonobese. Weight and sodium ex- obesity-induced hypertension has been greatly debat-
cretion were directly associated with sleep BP and ed. Landsberg32 hypothesized that high calorie intake
24-hour systolic BP, and the effect of sodium excre- increases thermogenesis by activating the sympathetic
tion was modified by weight. The sodium excretion- nervous system to act as a buffer against weight gain.
weight interaction, included in the linear regression However, insulin resistance is another mechanism the
model, allowed us to determine if the effect of obesity obese recruit to stabilize body weight and limit further
was the same at different sodium excretion rates. At gain. The resultant hyperinsulinemia stimulates sym-
the same urinary sodium excretion level, the obese pathetic activity, driving thermogenic mechanisms
children had higher ambulatory systolic BP. At the that increase the metabolic rate.34 Sympathetically
higher levels of sodium excretion, however, differ- mediated vasconstriction and cardiac stimulation and
ences in BP between obese and nonobese children enhanced sodium reabsorption exert a prohyperten-
were smaller. This indicates that obese children may sive effect.35
be heterogeneous in terms of the relationship between The hemodynamic effects of insulin and its implica-
ambulatory BP and sodium excretion. Whether or tion in the structural changes in arterial wall thickness

366 THE JOURNAL OF CLINICAL HYPERTENSION VOL. III NO. VI NOVEMBER/DECEMBER 2001
and/or arterial compliance have been recently re- crecimiento y desarrollo. Madrid, Spain: Fundación F. Or-
begozo, Garsi; 1988.
viewed.36 Insulin’s vasodilator capacity is well estab- 14 Himes J, Dietz W. Guidelines for overweight in adolescent
lished in healthy, normotensive subjects; however, in preventive services: recommendations from an expert com-
insulin-resistant states, such as obesity, insulin-mediat- mittee. Am J Clin Nutr. 1994;59:307–316.
15 Whitaker RC, Wright JA, Pepe MS, et al. Predicting obesi-
ed vasodilatation may be blunted, in part by endothe- ty in young adulthood from childhood and parental obesi-
lial dysfunction and the consequent lower nitric oxide ty. N Engl J Med. 1997;337:869–873.
production or by abnormal glucose uptake, which re- 16 Florey C. The use and interpretation of ponderal index
and other weight-height ratios in epidemiological studies. J
duces the calcium influx to muscle cells. Chronic Dis. 1970;23:93–103.
17 Alpert BS. Cuff width and accuracy of measurement of
CONCLUSIONS blood pressure. Blood Press Monit. 2000;5:151–152.
18 Report of the Second Task Force on Blood Pressure Con-
Substantial clinical and epidemiologic evidence sup- trol in Children. National Heart, Lung and Blood Institute.
ports the influence of obesity in BP levels, even early Pediatrics. 1987;79:1–25.
in life. Ambulatory BP monitoring provides better as- 19 Redon J, Campos C, Narciso ML, et al. Prognostic value of
ambulatory blood pressure monitoring in refractory hyperten-
sessment of BP and has allowed the observation that sion. A prospective study. Hypertension. 1998;31:712–718.
not only obesity but also body weight distribution are 20 Frattola A, Parati G, Cuspidi C, et al. Prognostic value of
independent determinants of BP values. Since obesity 24-hour blood pressure variability. J Hypertens. 1993;11:
1133–1138.
is related to a clustering of risk factor variables in 21 Verdecchia P, Porcellati C, Schillaci G, et al. Ambulatory
children and young adults, the prevention of the onset blood pressure. An independent predictor of prognosis in
of obesity in early life may be important in reducing essential hypertension. Hypertension. 1994;24:793-801.
22 Lurbe E, Redon J, Liao Y, et al. Ambulatory blood pres-
the risk of coronary heart disease in later life. sure monitoring in normotensive children. J Hypertens.
1994;12:1417–1423.
23 Lurbe E, Aguilar F, Gomez A, et al. Reproducibility of am-
REFERENCES bulatory blood pressure monitoring in children. J Hyper-
1 World Health Organization. Obesity: preventing and manag- tens. 1993;11(suppl 5):S288–S289.
ing the global epidemic. Report of a WHO Consultation. Pre- 24 Lurbe E, Thijs L, Redon J, et al. Diurnal blood pressure
sented at: the World Health Organization: June 3–5,1997; curve in children and adolescents. J Hypertens. 1996;14:
Geneva, Switzerland. Publication WHO/NUT/NCD/98.1. 41–46.
2 Tell GS. Cardiovascular disease risk factors related to sex- 25 National High Blood Pressure Education Program Work-
ual maturation: the Oslo youth study. J Chronic Dis. ing Group on Hypertension Control in Children and Ado-
1985;38:633–642. lescents. Update on the 1987 Task Force Report on High
3 Voors AW, Webber LS, Frerichs RR, et al. Body height Blood Pressure in Children and Adolescents: a working
and body mass as determinants of basal blood pressure in group report from the National High Blood Pressure Edu-
children—the Bogalusa Heart Study. Am J Epidemiol. cation Program. Pediatrics. 1996;98:649–658.
1977;106:101–108. 26 Soergel M, Kirschstein M, Busch C, et al. Oscillometric
4 Lurbe E, Alvarez V, Liao Y, et al. The impact of obesity and twenty-four-hour ambulatory blood pressure values in
body fat distribution on ambulatory blood pressure in children healthy children and adolescents: a multicenter trial in-
and adolescents. Am J Hypertens. 1998;11:418–424. cluding 1141 subjects. J Pediatr. 1997;130:178–184.
5 Faloia E, Giacchetti G, Mantero F. Obesity and hyperten- 27 O’Sullivan JJ, Derrick G, Griggs P, et al. Ambulatory
sión. J Endocrinol Invest. 2000;23:54–62. blood pressure in schoolchildren. Arch Dis Child.
6 Kroke A, Bergmann M, Klipstein-Grobusch K, et al. Obe- 1999;80:529–532.
sity, body fat distribution and body build: their relation to 28 Kissebah AH, Krakower GR. Regional adiposity and mor-
blood pressure and prevalence of hypertension. Int J Obes bidity. Physiol Rev. 1994;74:761–811.
Related Metab Disord. 1998;22:1062–1070 29 Hall JE. Mechanisms of abnormal renal sodium handling
7 Okosun IS, Prewitt TE, Cooper RS. Abdominal obesity in in obesity hypertension. Am J Hypertens. 1997;10(5 pt
the United States: prevalence and attributable risk of hy- 2):49S–55S.
pertension. J Hum Hypertens. 1999;13:425–430. 30 Rocchini AP, Key J, Bondie D, et al. The effect of weight
8 Gillum RF. The association of body fat distribution with hy- loss on the sensitivity of blood pressure to sodium in obese
pertension, hypertensive heart disease, coronary heart disease, adolescents. N Engl J Med. 1989;321:580–585.
diabetes, and cardiovascular risk factors in men and women 31 Lurbe E, Alvarez V, Liao Y, et al. Obesity modifies the rela-
aged 18–79 years. J Chronic Dis. 1987;40:421–428. tionship between ambulatory blood pressure and natriuresis in
9 Weststrate JA, Deurenberg P, van Tinteren H. Indices of children. Blood Press Monit. 2000;5(5–6):275–280.
body fat distribution and adiposity in Dutch children from 32 Landsberg L. Diet, obesity and hypertension: an hypothesis in-
birth to 18 years of age. Int J Obes. 1990;14:149–157. volving insulin, the sympathetic nervous system, and adaptive
10 Sangi H, Mueller WH. Which measure of body fat distrib- thermogenesis. Q J Med. 1986;61(236):1081–1090.
ution is best for epidemiologic research among adoles- 33 Mark AL, Correia M, Morgan DA, et al. Obesity-induced
cents? Am J Epidemiol. 1991;133:870–883. hypertension. New concepts from the emerging biology of
11 Sangi H, Mueller WH, Harrist RB, et al. Is body fat distri- obesity. Hypertension. 1999;33(1 pt 2):537–541.
bution associated with cardiovascular risk factor in chil- 34 Biolo G, Toigo G, Ciocchi B, et al. Slower activation of in-
dren? Ann Hum Biol. 1992;19:559–578. sulin action in hypertension associated with obesity. J Hy-
12 Zwiauer KF, Pakosta R, Meller T, et al. Cardiovascular risk pertens. 1998;16(12 pt 1):1783–1788.
factors in obese children in relation to weight and body fat dis- 35 Hall JE. Pathophysiology of obesity hypertension. Curr
tribution. J Am Coll Nutr. 1992;11(suppl):41S–50S. Hypertens Rep. 2000;2:139–147.
13 Hernandez M, Castellet J, Narvaiza JL, et al. Curvas y 36 Mikhail N, Tuck MI. Insulin and the vasculature. Curr
tablas de crecimiento. Instituto de investigaciones sobre Hypertens Rep. 2000;2:148–153.

VOL. III NO. VI NOVEMBER/DECEMBER 2001 THE JOURNAL OF CLINICAL HYPERTENSION 367