Pediatric Cardiology

Reference Values of Brachial-Ankle

Pulse Wave Velocity in Children

Chin-Fu Wang,1 Yu-Tsun Su1 and Kai-Sheng Hsieh2

Background: Arterial wall properties are important parameters reflecting the physical characteristics of the arteries.

They bear relationship to cardiac-arterial interactions of the various arterial wall properties. Values of pulse wave

velocity (PWV), especially the brachial-ankle PWV (baPWV) are a practical predictor of the severity of atherosclerosis

in adults. However, changes of value of baPWV in healthy children (8-19 years old) have rarely been assessed. We

therefore investigated the changes in values of baPWV in healthy Taiwanese children.

Method: Between Oct. 2004 and Dec. 2005, in the Pediatric Cardiovascular Outpatient Department of Kaohsiung

Veterans General Hospital, we measured values of baPWV and other vital measurements in 123 healthy (systolic

blood pressure (SBP) � 120 mmHg, body mass index (BMI) � 24 and no underlying heart disease) children (51

boys and 72 girls). The relationship between baPWV and vital measurements was analyzed by linear regression and

step-wise multiple linear regression using SPSS 10.0 software.

Result: The value of baPWV could be predicted by ankle diastolic pressure (aDBP) and body height (Ht) (baPWV =

279.78 + 5.02 * aDBP + 2.67 * Ht, p < 0.05). It also increased with age, body weight (Wt), brachial systolic blood

pressure (bSBP), and brachial diastolic blood pressure (bDBP), independently. It was not statistically correlated

with BMI, heart rate (HR) or ankle systolic blood pressure (aSBP).

Conclusion: Based on our results, it is worthwhile to use values of baPWV as indicators of changes in arterial wall

properties in healthy Taiwanese children.

Key Words: Pulse wave velocity � Blood pressure � Body mass index

INTRODUCTION

Arterial wall properties are important parameters re-

flecting the physical characteristics of the arteries. They

bear relationship to cardiac-arterial interactions of the

various arterial wall properties. Measuring the value of

baPWV is recognized as a simple and non-invasive met-

hod of screening for arterial stiffness in a large popula-

tion.1 It is believed that the value of baPWV increases

with age and is a practical predictor of the severity of

atherosclerosis in adults.

Although atherosclerosis is rare in childhood, it is

important to emphasize the ability of reference values of

baPWV to reflect changes in arterial walls in children. In

the past, there have been few studies2-5 of developing

children in other countries, or of adults in Taiwan.6

Therefore, we investigated values of baPWV and an-

alyze correlations between baPWV and vital measure-

ments in healthy Taiwanese children.

METHODS

Between Oct. 2004 and Dec. 2005, we routinely

measured values of baPWV, blood pressures of four

35 Acta Cardiol Sin 2008;24:35�8

Original Articles Acta Cardiol Sin 2008;24:35�8

Received: May 8, 2007 Accepted: December 24, 20071Department of Pediatrics, E-Da Hospital/I-Shou University;2Department of Pediatrics, Veterans General Hospital, Kaohsiung,

Taiwan.

Address correspondence and reprint requests to: Dr. Kai-Sheng

Hsieh, Department of Pediatrics, Veterans General Hospital, No. 386,

Ta-Chung 1st Rd, Kaohsiung 81364, Taiwan. Tel: 886-7-342-2121

ext. 8202; E-mail: kshsieh@isca.vghks.gov.tw

limbs, heart rate, body height, and body weight in 123

healthy children (8-19 years old, 51 boys and 72 girls) in

the Pediatric Cardiovascular Outpatient Department of

Kaohsiung Veterans General Hospital.

Patients were examined in supine position, and

sphygmomanometer cuffs were wrapped on the brachia

and ankles. The cuffs inflated and deflated automati-

cally, and pulse wave contours from the four extremities

were recorded simultaneously. Pulse transit time be-

tween the brachial and ankle regions was computed from

the pulse volume recording. Brachial-ankle PWV was

determined from the pulse transit time and the distance

between these two segments. The distance of each seg-

ment was calculated automatically, based on the height

of the subject.5 All measurements were taken by the

same technician using the volume plethysmographic ma-

chine, Colin VP-1000 (Made in Japan).7

The inclusion criteria for subjects in our study were

(1) normal age-related SBP (� 120 mmHg), (2) normal

age-related BMI (� 24), (3) absence of underlying car-

diovascular disease, and (4) absence of lipid disorder as

reported by the family. The reference values of age-re-

lated SBP and BMI were published by the Department of

Heath in Taiwan, 2002.

Values of baPWV and blood pressures of the right

arm and ankle were included in the analysis. We divided

the subjects into groups on the basis of gender and pu-

bescent stage. The values of baPWV, age, Ht, Wt, blood

pressure, HR and BMI were analyzed by linear regres-

sion and step-wise multiple linear regression, using

SPSS 10.0 software.

RESULTS

On the basis of gender, the characteristics of the two

groups are summarized in Table 1. Systolic blood pres-

sure and body weight were higher in the male group than

in the female group. Although there was no statistically

significant difference between baPWV in the two groups,

female subjects had smaller values of baPWV than ma-

les. Correlations between baPWV and each factor are

shown in Table 2. In females, age, Ht, Wt, bSBP, bDBP,

and aDBP correlated positively with baPWV. This was

different from males.

On the other hand, values of baPWV in groups of

both prepubescent and postpubescent females were low-

er than those in the corresponding male groups (Table

3). In order to clarify the interactions among baPWV,

age and gender, two-way ANOVA was performed. The

results are shown in Table 4. The linear regression analy-

sis of the values of baPWV and each correlated factor is

shown in Figure 1.

According to the results of step-wise multiple linear

Acta Cardiol Sin 2008;24:35�8 36

Chin-Fu Wang et al.

Table 1. Characteristics of subjects

n = 123 Male (51) Female (72) P

Age (years) 12.1 � 2.7 12.7 � 3.0

Wt (kg) 046.6 � 11.7 42.4 � 8.3 < 0.05

Ht (cm) 154.5 � 15.1 150.6 � 10.8

BMI (kg/m2) 19.2 � 2.3 18.6 � 2.3

HR (bpm) 072 � 11 071 � 11

rbaPWV (cm/s) 1011 � 159 0979 � 159

lbaPWV (cm/s) 0996 � 142 0959 � 153

rbSBP (mmHg) 108 � 90 105 � 90 < 0.05

rbDBP (mmHg) 57 � 5 57 � 7

lbSBP (mmHg) 109 � 90 105 � 90 < 0.05

lbDBP (mmHg) 58 � 6 57 � 6

raSBP (mmHg) 125 � 21 114 � 15 < 0.05

raDBP (mmHg) 61 � 8 61 � 9

laSBP (mmHg) 121 � 17 112 � 13 < 0.05

laDBP (mmHg) 62 � 8 60 � 8

Wt: weight; Ht: height; BMI: body mass index; HR: heart

rate; rbaPWV: right brachial-ankle pulse wave velocity;

lbaPWV: left brachial-ankle pulse wave velocity; rbSBP: right

brachial systolic blood pressure; rbDBP: right brachial diastolic

blood pressure; lbSBP: left brachial systolic blood pressure;

lbDBP: right brachial diastolic blood pressure; raSBP: right

ankle systolic blood pressure; raDBP: right ankle diastolic

blood pressure; laSBP: left ankle systolic blood pressure;

laDBP: left ankle diastolic blood pressure.

Table 2. Pearson’s correlations of rbaPWV with each factor

Male (51) Female (72) All (123)

Age 0.218 *0.352** 0.287**

Wt 0.203 0.276* 0.251**

Ht 0.234 0.283* 0.266**

BMI -0.021- 0.117* 0.073**

HR 0.104 -0.037-* 0.021**

rbSBP 0.176 *0.331** 0.281**

rbDBP *0.310* 0.300* 0.300**

raSBP 0.023 0.208* 0.138**

raDBP **0.419** 0.248* 0.316**

*P < 0.05; **P < 0.01

regression analysis with a 95% confidence interval, the

value of baPWV was dependent on both aDBP and Ht

(baPWV = 279.78 + 5.02 * aDBP + 2.67 * Ht, p < 0.05).

DISCUSSIONS

According to previous studies in children5 and adu-

lts,8,9 females have lower values of baPWV than males.

Other studies8-10 reported that values of baPWV in pre-

pubescent females were higher than those in males and

then became lower after puberty. It was believed that fe-

male hormones affect the changes of arterial stiffening.

However, in our study, prepubescent females had smaller

values of baPWV than males (Table 3). We consider that

precocious puberty may result those differences from

other reports.5,8,9 On the other hand, age was positively

correlated with baPWV in all subjects and females but

not in males. We found the interaction of age and gender

did not statistically affect the values of baPWV (Table

4). This result was also different from that in other re-

ports. Nevertheless, the females had lower values of

baPWV than the males.

The structures of vessels change when an increase in

37 Acta Cardiol Sin 2008;24:35�8

Brachial-Ankle Pulse Wave Velocity in Children

Table 3. rbaPWV (Mean � SD, cm/s) of subjects divided on the

basis of pubescent stage

Male Female P

n = 22 n = 27Prepuberty (8~11 y/o)

960 � 179 929 � 129> 0.05

n = 29 n = 45Postpuberty (12~19 y/o)

1050 � 1320 1009 � 1680> 0.05

Table 4. Two-way ANOVA of rbaPWV for age and gender

Variable F P

Sex 1.015

Age 1.909 < 0.05

Sex * Age 1.465

Figure 1. Relationships between rbaPWV and vital measurements (95% confidence interval).

blood pressure causes an augmentation of vascular ten-

sion.11-13 Increased cardiac output associated with age also

strongly affects increases in blood pressure, and age-re-

lated increases in values of baPWV. In our study, blood

pressure correlated positively with baPWV but heart rate

did not. These results imply that heart rate can not statisti-

cally represent cardiac output, although cardiac output is

the product of heart rate and stroke volume (CO = HR �

SV). Unfortunately, we did not measure the stroke vo-

lume or ejection fraction in children by echocardio-

graphy. The values of cardiac output were not available.

The value of BMI is usually considered to be a pre-

dictor of obesity, but it is not proportional to serum lipid

levels. Therefore, higher BMI values are not correlated

with hyperlipidemia, which predisposes to atherosclero-

sis. Several previous studies14-17 reported that baPWV

was positively correlated with serum lipid levels but not

with BMI. Although BMI is proportional to body weight,

these measurements have different influences on the

values of baPWV. Our results show that body weight

correlated positively with baPWV but BMI did not. In

addition, we did not check the serum lipid levels in sub-

jects, so the relationship between baPWV and serum li-

pid levels was not clear.

Although no subjects suffered cardiovascular dis-

ease, higher values of baPWV were measured in many

patients. According to their family’s description, no lipid

disorder had previously been found. Therefore, in order

to know the exact metabolic conditions in our subjects,

examinations of lipid profiles should be considered.

According to our results, it is worthwhile to use va-

lues of baPWV as indicators of changes in arterial walls

in healthy Taiwanese children. The values provide infor-

mation regarding the health of developing children. In

Taiwan, baPWV in adults has been studied.6 The results

were consistent with those in previous studies elsewhere.

More studies are necessary to determine and elaborate

on the clinical significance of values of baPWV in chil-

dren’s development.

REFERENCES

1. Yamashina A, Tomiyama H, Takeda K, et al. Validity, repro-

ducibility, and clinical significance of noninvasive brachial-ankle

wave velocity measurement. Hypertens Res 2002;25:359-64.

2. Avolio AP, Chen SG, Wang RP, et al. Effects of aging on changing

arterial compliance and left ventricular load in a northern Chinese

urban community. Circulation 1983;68:50-8.

3. Cheung YF, Brogan PA, Pilla CB, et al. Arterial distensibility in

children and teenagers: normal evolution and the effect of child-

hood vasculitis. Arch Dis Child 2002;87:348-51.

4. Senzaki H, Akagi M, Hishi T, et al. Age-associated changes in ar-

terial elastic properties in children. Eur J Pediatr 2002;161:

547-51.

5. Niboshi A, Hamaoka K, Sakata K, et al. Characteristics of

brachial-ankle pulse wave velocity in Japanese children. Eur J

Pediatr 2006;165(9):625-9.

6. Chuang SY, Chen CH, Cheng CM, Chou P. Combined use of

brachial-ankle pulse wave velocity and ankle-brachial index for

fast assessment of arteriosclerosis and atherosclerosis in a com-

munity. Int J Cardiol 2005;98(1):99-105.

7. Munakata M, Ito N, Nunokawa T, et al. Utility of automated

brachial ankle pulse wave velocity measurements in hypertensive

patients. Am J Hypertens 2003;16:653-7.

8. London GM, Gurein AP, Pannier B, et al. Influence of sex on arte-

rial hemodynamics and blood pressure: role of body height. Hy-

pertension 1995;37:1374-80.

9. Tomiyama H, Yamashina A, Arai T, et al. Influences of age and

gender on results of noninvasive brachial-ankle pulse wave ve-

locity measurement: a survey of 12517 subjects. Atherosclerosis

2003;30:303-9.

10. Ahimastos AA, Formosa M, Dart AM, et al. Gender differences in

large artery stiffness pre- and postpuberty. J Clin Endocrinol

Metab 2003;88:5375-80.

11. Yang L, Gao YJ, Lee RM. Quinapril effects on resistance artery

structure and function in hypertension. Naunyn Schmiedebergs

Arch Pharmacol 2004;370:444-51.

12. Athanase B, Chris A, Jeanne MB, et al. Determinants of acceler-

ated progression of arterial stiffness in normotensive subjects and

in treated hypertensive subjects over a 6-year period. Circulation

2002;105:1202-7.

13. Boutouyrie P, Tropeano AI, Asmar R, et al. Aortic stiffness is an

independent predictor of primary coronary events in hypertensive

patients: a longitudinal study. Hypertension 2002;39:10-5.

14. Shiotani A, Motoyama M, Matsuda T, et al. Brachial-ankle pulse

wave velocity in Japanese university students. Intern Med 2005;

44(7):696-701.

15. Nakamura U, Iwase M, Nohara S, et al. Usefulness of brachial-

ankle pulse wave velocity measurement: correlation with abdo-

minal aortic calcification. Hypertens Res 2003;26:163-7.

16. Okamura T, Moriyama Y, Kadowaki T, et al. Noninvasive mea-

surement of brachial-ankle pulse wave velocity is associated with

serum C-reactive protein but not with alpha-tocopherol in Japa-

nese middle-aged male workers. Hypertens Res 2004; 27:173-80.

17. Giral P, Atger V, Amar J, et al. A relationship between aortic stiff-

ness and serum HDL3 cholesterol concentrations in hypercho-

lesterolaemic, symptom-free men. J Cardiovasc Risk 1994;1:

53-8.

Acta Cardiol Sin 2008;24:35�8 38

Chin-Fu Wang et al.