* Deceased.
PEDIATRICS (ISSN 0031 4005). Copyright © 1987 by the
American Academy of Pediatrics.
Blood Pressure
Gregory L. Burke,
MD,
MS
Antonie
W. Voors,
MD, DrPh,*
Charles L.
Shear,
DrPH, Larry S. Webber,
PhD, Carey
G. Smoak,
MS
James
L Cresanta,
MD, MS
Hyg, and Gerald
S. Berenson,
MD
From the Departments of Medicine, Public Health and PreventWe Medicine, Biometry and Genetics, and Pediatrics, Louisiana State University Medical Center, New Orleans
ABSTRACT.
BP was measured in 440 children followedlongitudinally from birth to 7 years of age in Bogalusa,
LA. Levels, trends, and determinants of BP were
evalu-ated in this newborn cohort. Both systolic and diastolic BP levels remained relatively constant between the ages
of 6 months and 7 years. BP levels varied between the different instruments, and differences were also noted
between measures obtained using the same instrument
before and after venipuncture. White children were noted
to have slightly higher levels of systolic and diastolic BP pressure at 6 months and 1 year of age, even after adjustment for body size. Significant prediction of year 7 BP rank occurred as early as 6 months of age for systolic and at 1 year of age for diastolic BP levels. Body size was
inconsistently related to BP levels from ages 6 months
through 4 years, but the relationship was stronger and more consistent with changes in body size. Of interest is the relatively constant levels of indirect BP during this
period ofrapid growth, as measured by currently available
instruments. These data emphasize the importance of
cardiovascular risk factor measurement during early life and of the need to improve methods of indirect BP measurement in infancy. Pediatrics
1987;80(suppl):784-788; blood pressure, cardiovascular risk factor.
BP levels in early life may provide important clues in the search for determinants of adult hyper-tension. Systolic and diastolic BP levels have been shown to increase approximately 1.5 mm Hg and 1 mm Hg per year of age, respectively,. in children ages 5 to 17 years.’ Starting at levels much lower than found in adulthood, school-aged children tend to persist within BP ranks into young adult-hood.7 However, the specific levels and trends of BP in young children are less clear. Accurate, reli-able measurements of BP in infants and young children are difficult to obtain and require careful attention to measurement technique.’6
BP levels and trends in a cohort of children followed longitudinally from 6 months to 7 years of age are reported in this paper. The influence of race, sex, and body size on BP levels in early childhood was also evaluated.
MATERIALS AND METhODS
Population
The population consisted of all infants born from Jan 1, 1974, to June 30, 1975, of residents in ward 4 of Washington Parish, Louisiana, mainly includ-ing the city of Bogalusa. Surveys of these children have been conducted at birth and subsequently at ages 6 months and 1, 2, 3, 4, and 7 years. A total of 440 infants (98% of those born during period) com-posed the initial newborn group. See Webber et al’3 for a more complete description of the group.
General Study Design
Cardiovascular disease risk factors were ascer-tamed on eligible infants according to detailed pro-tocols.2”3’14 Measurement of BP, height (or length), weight, skinfold thickness, serum lipid and lipopro-tein concentrations, dietary recalls, and physical examinations were performed on all participants from 6 months through 7 years of age as described earlier.’3
BP Measurement
. .
.--- --.
0- #{149}-0
ARTERIOSONDE INFRASONDE MERCURY
PHYSIOMETRICS
PRE-VENIPUNCTURE ARTERIOSONDE
a
I
E E
w
Cl)
CO
Lu
a.
0
0 0
-a
100
90-
80-
70-60
-50
-40
-Systolic
i:
2Diastolic (4th Ph...)
-..-- . -0
2T
.-_#{243}.s i 2 3 4 7
TABLE 1. Blood Pressure Instruments Used at Each Age, Newborn Cohort-Bogalusa Heart Study*
Instrument Type Mechanism Ages Measured
6mo lyr 2yr 3yr 4yr 7yr
Arteriosonde (model Semiautomatic Doppler-ultrasonic X X X X X
1010) transducers
Infrasonde (model 3000) Semiautomatic Low-frequency
sen-sing device
X . X X X X
Mercury (Baumanome- Manual Human detection of X X X
ten) Konotkoff sounds
Physiometnics (model Automatic Low-frequency sen- X
USM 105) sing device
*
x
indicates use.sounds. A mercury sphygmomanometer (W. A. Baum Co, Inc, Copiague, NY) instrument allowing for human detection of low-frequency Korotkoff sounds was used in participants 3, 4, and 7 years of age. A Physiometrics USM 105 (Sphygmetrics Inc, Woodland Hills, CA) automatic instrument, which detected low-frequency sounds, was used to mea-sure BP in the 7-year-old children from this cohort. Measurement error for the different BP instru-ments has been described previously.13 Briefly, the measurement error was noted to increase with age regardless of the instrument used, and the Arter-iosonde instrument had a consistently lower meas-urement error than the Infrasonde instrument from ages 6 months to 4 years of age, especially for diastolic BP levels.
Right arm length and circumferences were mea-sured to ensure proper cuff size, and BP levels were determined three times at each station by well-trained observers.2”3 Systolic and diastolic BPs were recorded as the first and fourth Korotkoff phases, respectively. A subjective appraisal of the participant’s mood was determined at each BP sta-tion on a scale of 1 to 9, ranging from sleeping to crying. BP levels were routinely measured during the last stages of screening, however, additional Arteriosonde BP measurements were taken prior to venipuncture on a sample of infants at 6 months of age and all participants at ages 1, 2, 3, and 4 years.
Statistical Analysis
Descriptive statistics were used to illustrate BP trends from 6 months through 7 years of age. BP levels obtained using the Arteniosonde instrument (prior to age 7 years) were combined with measures obtained with the mercury sphygmomanometer at 7 years of age for analyses that require BP infor-mation throughout the entire follow-up period (ie, tracking and prior predictors of BP levels in early childhood). Spearman correlation coefficients were used to estimate tracking or persistence in peer ranks of BP throughout time.
AGE (years)
Fig 1. BP levels in young children followed longitudi-nally from 6 months to 7 years of age, 1973 to 1982. Note
similar trends present across all four measurement in-struments.
RESULTS
DISCUSSION
As noted by other investigators1#{176}’2”’8 only
sub-r
oo-E E 4,
80-S
a 60 0
0
40-SystolIc
0 :-.._.
.-. - . .
g
-Diastolic (4th Phass)
G-.0 WHITE
.-. BLACK
$ p<O.O5
7
trends were seen with diastolic BP-increasing slightly between 6 months to 1 year of age and then decreasing or remaining unchanged through 7 years of age. Arteriosonde diastolic BP measurements were approximately 5 to 10 mm Hg higher than those taken with the Infrasonde instrument. Mer-cury sphygmomanometer measurements of dia-stolic BP remained stable from 3 to 7 years of age. The levels seen for BPs measured with the Ar-teriosonde instrument prior to venipuncture showed little change between 6 months and 7 years of age. In comparison with the postvenipuncture Arteriosonde measurements, these prevenipuncture levels were lower in infants less than 2 years of age but were slightly higher after age 3 years. Adjust-ment for the mood of the child caused BP levels to decrease in the younger children (less than 3 years of age), remaining only slightly higher than the prevenipuncture levels.
Racial differences in BP are shown in Fig 2. White children had slightly higher (P < .05) systolic and diastolic BP levels at 6 months and 1 year of age. No racial differences were noted in children 2 through 7 years. The racial differences were found with all BP instruments.
The relationship between anthropometric varia-bles and BP in early childhood is shown in Table 2. For this analysis, we used the 7-year-old BP values and correlated them with both single prior measurements of height and weight, as well as changes in body size from earlier years of measure-ments. A significant relationship between prior sin-gle measures of body size and BP was not seen until 2 to 3 years of age. Correlation coefficients for prior height ranged from .03 (at 6 months of age) to .21 (at 7 years of age). Correlations were somewhat higher for prior weight (ranging from .09 to .35). However, when year 7 BP was correlated with
120
‘1
o. . .
0 .5 1 2 3 4
AGE (years)
Fig 2.
BP
levels in young children followed longitudi-nally from 6 months to 7 years by race, 1973 to 1982. White children have significantly higher levels of BP at 6 months and 1 year ofage. No racial differences are seen in older children.change in height and weight, a more constant pat-tern of correlation existed with age. Moreover, the correlations were much higher than that found for the single prior measurements for both change in height (ranging from .18 to .25 for systolic BP) and weight (ranging from .35 to .36 for systolic BP).
The persistence of subsequent BP levels between the ages of 6 months to 7 years is shown in Table 3. No race or sex differences in tracking were seen; therefore, results are presented for all children. Statistically significant prediction of subsequent systolic BP levels occurred as early as 6 months of age (r = 24 for correlation with year 1); however,
a consistent prediction of subsequent diastolic BP levels did not occur until 1 year of age (r = .14 for
correlation with year 2). The prediction of year 7 BP levels from measurements taken at ages 6 months to 4 years is shown in Fig 3. As previously noted, tracking of systolic BP levels was seen as early as 6 months of age, and the closer the initial measure was to age 7, the higher the prediction (r = .50 between 4 and 7 years of age), as might be
expected. Diastolic BP tracking was present from 1 year of age on, but the increased prediction of year 7 levels with age was not seen (r = .20 between
4 and 7 years).
TABLE 2. Correlation Between Blood Pressure at 7
Years of Age and Measurements of Body Size, Newborn Cohort-Bogalusa Heart Study*
Body Size Systolic BP Diastolic BP
Measurement at7yr at7yr
andAgeat ofAge(r) ofAge(r)
Measurement
Body Change Body Change
Size at in Body Size at in Body
Age Size Age Size
Specified From Specified From
Age Age
Specified Specified
atAge toAge
7yr 7yr
Height
6mo .03 .21 .07 .13
lyr .05 .22#{176} .10 .12
2yr .05 15b .12 .12
3 yr .09 #{149}24b .17#{176} .06
4 yr .15 .18#{176} .19” .03
7yr .21#{176} .15
Wt
6mo .09 35C .13 .22”
lyr .12 .36c .10 .22”
2yr .16 .36c .18” .19”
3yr .17 .36c .18#{176} .20”
4 yr 23b .36c .22” .18”
7 yr 35C .22”
* All correlations are Pearson correlations with N = 134.
0
N 220 206 183 171 178 220 206 183 171 178
TABLE 3. Persistenc
Childhood, Newborn C
e of Blood Pressure ohort-Bogalusa Heart
in Early Study*
BP and
Age
lyr
Correlation Coefficients
2yr 3yr 4yr 7yr
Systolic
Gmo .24 .16 .17 .07” .13
1 yr .26 .21 .22 .19
2 yr .41 .20 .35
3 yr .36 .37
4yr .49
Diastolic
6 mo .12#{176} 00b 04b 04b 04b
1 yr .14#{176}.17 .19#{176} .15a
2yr .20 .17#{176} .23
3yr .18 .25
4yr .23
* Significance: P < .01 bp>
.os.
z w 0
Ui
0
0 z
0
Ui
0
0
SYSTOLIC
unless otherwise noted, “ P < .05,
DIASTOLIC
AGE (years) .5 1 2 3 4 .5 1 2 3 4
Fig 3. Prediction of year 7 BP levels based on prior measures, 1973 to 1982. Tracking occurs as early as 6 months for systolic and 1 year of age for diastolic BP
levels.
tle changes in BP levels were seen from 6 months
to 7 years of age, and the level found was highly dependent on the instrument used and the mood of the child. The variations in absolute levels of BP noted between the different instruments produced
inconsistent trends from ages 6 months through 7
years for both systolic and diastolic BP levels. In our earlier studies of preschool children (ages 2#{189} to 5#{189}years), we noted slight increases in systolic BP levels with the Infrasonde instrument and the
Physiometnics unit in participants ages 3 to 8 years of age.’6 Schachter et al’#{176}noted a large increase in BP levels from birth to 6 months of age with little difference in levels from 6 months through 5 years,
whereas in a study of infants in Boston only a slight
increase in BP was found from 1 month to 2 years of age.’2 De Swiet and colleagues9 noted a slight increase in systolic BP from 6 months through 4 years of age and concluded that no significant dif-ferences exist in BP levels from 6 months through
6 years of age.’8
The reason for the lack of an increase in BP
levels as detected by certain instruments in these
children is not clear. It is possible that BP meas-urement techniques and instruments used in in-fancy are inadequate to detect a child’s “true” BP and that the lack of an increase is due to an artifact of measurement. Alternatively, because a tremen-dous amount of growth takes place during early childhood (height increased approximately 165% and weight increased 320% from 6 months to 7
years of age), a BP increase would have been antic-ipated. A clue to the lack of increase in BP during this period of growth may come from our nutritional studies of this cohort. We have previously noted a peak in sodium intake per kilogram of body weight at approximately 2 years of age, with a subsequent decrease in intake in older children.19 This high
sodium intake in early life with a decrease there-after may offset what would otherwise be increasing levels of BP between 2 and 7 years of age. This hypothesis is supported by Hofman et al2#{176}who reported the relationship between sodium intake and BP in infants.
Schachter et al” found a relationship between mood and BP in infants. The difference we noted between pre- and postvenipuncture BP measure-ments may have occurred because young children
(6 months to 2 years of age) misconstrue the appli-cation of the BP cuff as another attempt at veni-puncture. The higher prevenipuncture BP levels
seen in slightly older children may represent a
training effect similar to that seen in schoolchildren and adults.’5 The importance of anxiety in
chil-dren’s BP is thus shown by the uniformity of the
BP distribution from 6 months to 7 years of age after adjustment for the mood of the child.
No significant sex differences in BP levels were seen from 6 months to 7 years of age, and these findings agree with those of other investiga-tons.’#{176}”21’22 Although other investigators have not seen significant racial differences in BP levels in infants and children,’#{176}”6’21’22 we noted that white children younger than 1 year of age had slightly higher levels ofboth systolic and diastolic BPs. The subtle differences in BP levels across race groups may be related to slightly larger body size in white children at birth and 6 months of age.23
Although prior height and weight measurements were related to both systolic and diastolic BP levels at age 7 years, a strong and consistent relationship between BP and body size was seen only by using
changes in body size as predictors of future levels. These results indicate that monitoring of body size, beginning at an early age, may help predict future
high-risk children.
seen as early as 6 months for systolic and 1 year of age for diastolic BPs. The relationship grew stronger with increasing age. Significant persist-ence of BP levels during infancy has been reported in other cohorts of infants.9”#{176}”7’226 Levine et al and Zinner et al noted significant tracking of BP from as early as 6 months of age. In addition, greater persistence of BP levels has been shown to occur for school-aged children.’5’27 This persistence of BP levels in young children also emphasizes the need to monitor BP levels in childhood so that early identification and intervention on the natural his-tory of essential hypertension can be achieved.
IMPLICATIONS
It is important to study the early natural history of hypertensive disease. BP levels can be measured in early childhood with almost the same degree of reproducibility as is present in older children and adults. The relatively similar levels of BP from 2 through 7 years of age is puzzling, given the dra-matic increase in body size in these rapidly growing children, but may, in part, be related to the high density of sodium intake in these infants when compared with older children,19 to the anxiety with cuff pressure, especially up to 2 years of age, and to the recording instruments. BP levels track from infancy into early childhood, pointing to important influences of future BP levels which appear to be present even early in life.
REFERENCES
1. National Heart, Lung, and Blood Institute’s Task Force on Blood Pressure Control in Children: Report on the task force on blood pressure control in children. Pediatrics
1977;59(suppl):797-82O
2. Berenson GS, McMahan C, Voors AW, et a!: Cardiovascular Risk Factors in Children: The Early Natural History of Atherosclerosis and Essential Hypertension. New York,
Ox-ford University Presa, 1980
3. Harlan WE, Cornoni-Huntley J, Leaverton PE: Blood
pres-sure in childhooth The National Health Examination Sur-vey. Hypertension 1979;1:559-565
4. Cornoni-Huntley J, Harlan WR, Leaverton PE: Blood pres-sure in adolescence: The United States Health Examination
Survey. Hypertension 1979;1:566-.571
5. Webber LS, Cresanta JL, Voors AW, et a!: Tracking of
cardiovascular disease risk factor variables in school-age
children. J Chronic Dis 1983;36:647-660
6. Lauer RM, Clarke WR, Beagleho!e R: Level, trend and variability of blood pressure during childhood. The Muses-tine study. Circulation 1984;69:242-249
7. Halt HI, Lemeshow 5, Rosenman lCD: A longitudinal study
of blood pressure in a national survey of children. Am J Public Health 1982;72:1285-1287
8. Webber LS, Voors AW, Foster TA, et a!: A study of instru-ments in preparation for a blood pressure survey in children. Circulation 1977;56:651-656
9. de Swiet M, Fayers P, Shinebourne EA: Value of repeated blood pressure measurements in children-The Brompton
study. Br Med J 1980;280:1567-1569
10. Schachter J, Kuller LH, Perfetti C: Blood pressure during
the first five years of life: Relation to ethnic group (black or white) and to parental hypertension. Am J Epidemiol 1984;119:541-553
11. Schachter J, Kuller LH, Perfetti C: Blood pressure during
the first two years of life. Am J Epidemiol 1982;116:29-41
12. Zinner SH, Kass EH: Epidemiology of blood pressure in infants and children, In Loggie JMH, Horan MJ, Gurskin AB, et a! (eds): NHLBI Workshop in Juvenile Hypertension, proceedings from symposium. New York, Biomedical
Infor-mation Carp, 1984, pp 93-105
13. Webber LS, Frank GC, Smoak CS, et a!: Cardiovascular disease risk factors from birth to 7 years of age: Bogalusa
heart study: I. Design and participation. Pediatrics
1987;80(supp!):767-778
14. Berenson GS, Blonde CV, Farris RP, et a!: Cardiovascular risk factor variables during the first year ofhfe: Heart study.
Am J Dis Child 1976;133:1049-1057
15. Souchek J, Stamler J, Dyer AR, et a!: The value of two or
three versus a single reading ofblood pressure at afirst visit. J Chronic Dis 1979;30:197-210
16. Voors AW, Webber LS, Berenson GS: Blood pressure of children, ages 2-#{189}-5-#{189}years, in a total community-The Bogalusa heart study. Am J Epidemiol 1978;107:403-411 17. Higgins M, Keller J, Moore F, et a!:Studies ofblood pressure
in young people and its relationship to personal and familial
characteristic and complications of pregnancy in mothers. Am J Epidemiol 1980,111:142-153
18. De Sweit M, Fayers P, Shinebourne EA: Systolic blood pressure apopulation of infants in the first year of life: The Brompton study. Pediatrics 1980;65:1028-1035
19. Frank GC, Nicklas TA, Webber LS, et a!: Na, K, Ca4,
and P intakes ofinfants and children: Bogalusa heart study,
abstracted. Presented at the American Dietetic Association
Meeting, New Orleans, October 1985
20. Hofman A, Hazeibroek A, Valkenburg HA: A randomized trial ofsodium intake andb!oodpressure in newborn infants JAMA 1983;250370-373
21. Kuller LA, Schachter J, Adams L, et a!: Epidemiology of
blood pressure in the young, in Loggie JMH, Horan MJ, Gurskin AB, et a! (eds): NHLBI Workshop on Juvenile Hypertension, proceedings from a symposium. New York,
Biomedical Information Corp, 1984, pp 73-92
22. Londe 5, Gollub SW, Goldring D: BlOOd pressure in black and in white children. J Pediatr 1977;90:93-95
23. Harsha DW, Smoak CG, Nicklas TA, et a!: Cardiovascular disease risk factors from birth to 7 years of age: The
Boga-lusa heart study: II. Tracking ofbody composition variables. Pediatrics 1987;80(suppl):779-783
24. Zinner SH, Rosner B, Oh W, et a!: Significance of blood
pressure in infancy: Familial aggregation and predictive effect on later blood pressure. Hypertension 1985;7:411-416 25. Zinner SH, Margolis HS, Rosner B, et a!: Stability of blood
pressure rank and urinary kallikrein concentration in child-hood. An eight-year follow-up. Circulation 1978;58:908-915 26. Levine RS, Hennekens CH, Klein B, et a!: Tracking
corre-lations of blood pressure levels in infancy. Pediatrics 1978;61:121-125
27. Voors AW, Foster TA, Frerichs RR, et a!: Studies of blood
pressure in children ages 5-14 in a total biracial
commu-nity-The Bogalusa heart study. Circulation
