The Significance of Blood Pressure Measurements in Childhood

Symposium on Hypertension

The Significance of Blood Pressure Measurements in Childhood Patricia A. Van Behren, M.D.,* and Ranald M. Lauer, M.D. **

The observations that hypertension in adults is extremely common, that hypertension is predictive and causative of stroke and coronary heart disease, and that a reduction in blood pressure improves the prognosis of adults with hypertension31 have resulted in consideration of the significance of blood pressure measurements in childhood. Measurements of blood pressure in infants and children have shown that there is a continuous rise in pressures throughout the school age years. 17 In early childhood, blood pressure levels are much less than adult levels and the level of blood pressure that is predictive of disease in adults is seldom reached. Few children whose blood pressures were measured in early childhood have been longitudinally followed to late adult life to correlate the significance of early pressure levels with the later onset of coronary heart disease or stroke. This paper will review, in part, what is known about blood pressure levels in children and suggest a clinical strategy that may be used for physicians dealing with individual children of school age. Measurement of Blood Pressure Blood pressure can be accurately and consistently measured in children over 3 years of age and should be recorded as a part of a routine physical examination. The child should be in a seated or supine position and blood pressure should be measured with a blood pressure cuff with a bladder which is wide enough to cover at least two thirds of the upper arm, and long enough to encircle most of the arm without overlapping. 14 The pressure should be estimated with a mercury-filled sphygmomanometer or a recently calibrated anaeroid sphygmomanometer. The first, fourth, and fifth phases of the Korotkoff sounds should be *Fellow, Division of Pediatric Cardiology, University of Iowa Hospitals and Clinics, Iowa City, Iowa **Professor of Pediatrics, Director of Division of Pediatric Cardiology, University of Iowa Hospitals and Clinics, Iowa City, Iowa The investigations cited from the Muscatine Study are supported by a SpeCialized Center for Research Grant in Atherosclerosis from the National Heart, Lung and Blood Institute, HL14230. Dr. Von Behren is supported by an individual training grant from the National Heart, Lung and Blood Institute, HL05212.

Medical Clinics of North America-Vol. 61, No. 3, May 1977

487

488

PATRICIA

A.

VON BEHREN AND RONALD

M.

LAUER

recorded. In children, the Korotkoff sounds frequently can be heard when the cuffis completely deflated. For this reason, the fourth phase is suggested as a better index of diastolic pressure in children. The relationship of the Korotkoff sounds to intravascular pressures have been studied by Moss and Adams 23 and by Holland.!l There is a high correlation with systolic pressures and a lower relationship with diastolic pressures. Yet, despite their less than perfect accuracy, pressures measured by the usual indirect techniques have wide application and, in adults, have predictive value for future coronary heart disease, stroke, and renal disease. Systolic blood pressure can now be accurately measured in infants by the Doppler method. Hernandez and Goldring 8 have shown good correlation between pressures obtained intra-arterially and those recorded by the Doppler method. Large screening surveys have indicated that this is a reproducible and reliable method of blood pressure measurement in infants.9

Normal Values Normal values of seated blood pressure in relation to age and sex have been established in children in Muscatine, Iowa,17 Rochester, Minnesota,29 and Bogalusa, Louisiana. 32 Figure 1 shows selected percentiles of blood pressures in seated white school age children in Muscatine, Iowa. Londe and Goldring 19 have established normal values for children 3 to 15 years of age with the patient in a supine position. All these crosssectional surveys have shown a continuous rise in blood pressure with age from infancy onward, with an acceleration of the rise in pressure during the adolescent years. Repeated serial de terminations of blood pressure are necessary in order to document persistent elevations of blood pressure. In screening the blood pressures in more than 300 high school students, Kilcoyne 13

160

160

160

150

150

150

FEMALES

160 N ~ 2460

SYSlOUC ME. VAWES ~th.5OttI

150

and 95th %

:::~::~~. ::

140 130

110

~

~

110

110

':~':~': 100

~

~

~

:

m

m

60

60

60

-----

50

50

50

---------

AGE

80

100

80

~Lr-r-r-r-r-r-r-r-r-r-,-,-,~40 6 7 8 9 10 11 12 13 14 15 16 17 18

=

80

:~

_______

70

70

6

7

8

9

10

60

~t~~~~::'~~UJE 11

12

13

14

15

16

50 17

18

AGE

Figure 1. Selected percentiles of blood pressures in seated white schoolage children in Muscatine, Iowa.

BLOOD PRESSURE MEASUREMENTS IN CHILDHOOD

489

found that the overall prevalence of systolic hypertension (> 140 mm Hg) decreased from 5.4 to 1.2 per cent and diastolic hypertension (> 90 mm Hg) decreased from 7.8 to 2.4 per cent on repeat screening. In Muscatine, Iowa, Rames et al.,27 after screening 6600 predominantly white school children, identified those whose systolic and/or diastolic blood pressures were greater than the 95th percentile or were greater than 140 mm Hg systolic, 90 diastolic. After repeated examinations of these subjects, less than 1 per cent maintained pressures at these levels. Thus fixed hypertension in the school age population is not a common finding. Tracking of Blood Pressure An important epidemiologic consideration is whether blood pressures taken in childhood are predictive of blood pressure levels in adulthood. If this were so, then future hypertensives could be identified in childhood. In Evans County, Georgia, in 1961,1° 11 per cent of 435 adolescents were found to be hypertensive (>140 mm Hg systolic and/or >90 mm Hg diastolic) with a single blood pressure measurement. Seven years later 30 of this group were re-evaluated. Sustained hypertension developed in 5 people and 6 had sustained hypertension with vascular complications. Two of the black females in this group died from cerebral hemorrhage both verified by autopsy. Zinn er et al. 34 followed 609 children 6 to 18 years of age over a four year period. Of 88 children with initial systolic pressures greater than 1 standard deviation unit above the mean, 65 per cent had positive scores at follow up and 70 per cent of children with initial systolic scores less than 1 standard deviation unit below the mean had negative scores at follow-up. These data suggest that stratification of blood pressure within peer groups begins and is detectable in childhood. Buck2 followed children from ages 5 to 12 years and found that those whose systolic and diastolic blood pressures had been greater than the 90th percentile at age 5 years had a statistically significant elevation of both systolic and diastolic pressures at age 12 years when compared to age-matched controls. Over a 5 year period in Muscatine, Iowa, 1953 children have had their blood pressures measured on three occasions. The overall correlation coefficient for systolic blood pressure from the first to the second reading was 0.36 and from the first to the third reading was 0.36. The correlation coefficient for diastolic blood pressure from the first to the second reading was 0.17 and from the first to the third reading was 0.21.3 These observations show that although there is a relationship between the blood pressures obtained in early childhood and those obtained five years later, there is marked variability in blood pressure so that it is difficult to predict who will be a future hypertensive. They indicate the need for repeated measurements of children's blood pressures to establish those that have consistent elevations of blood pressure. In the Thousand Aviator Study,24 young men with a mean age of24 years were followed for 24 years. Of those men whose systolic blood pressure was in the upper quintile at age 24, 40 per cent remained at that

490

PATRICIA

A.

VON BEHREN AND RONALD

M.

LAUER

level over the ensuing 24 years. Prediction of future difficulties appears to be better in later life than in childhood. Familial Aggregation of Blood Pressure Hypertension has been observed in several members of a family. Ayman 1 has reported elevated blood pressure in 28 per cent of children in families with one hypertensive parent and in 41 per cent of children with two hypertensive parents as well as in 65 per cent of adult siblings of hypertensive patients. Mial12 1.22 has shown that there is a familial resemblance of blood pressure in adults compared to blood pressure in first degree relatives. In a study of 350 siblings of 178 patients with severe hypertension, Platt26 found a greater incidence of hypertension in the siblings as compared to the general population. Johnson et al.1 2 in a study of children in Tecumseh, Michigan, found distinct resemblances in the levels of blood pressure between parents and children over the entire distribution of blood pressure. The correlation coefficient between parents and children, although significant, was of a low order. Between siblings they found a similar low correlation. Zinner and Kass 33 found a clustering of blood pressure among children 2 to 14 years of age compared to the blood pressures of their siblings and of their mothers (correlation coefficients for systolic blood pressure of sibling-sibling and mother-child were 0.34 and 0.16 respectively). The clustering effect was evident at all levels of blood pressure. In the Evans County Prevalence Survey10 similarities for first degree relatives, as measured by correlation coefficients, were of the same order of magnitude in blacks and whites and of the same order of magnitude as reported in other populations. A similar degree of familial aggregation between a child and his siblings and his parents was shown to exist as early as 2 months of age by Klein. 15 The degree of these relationships suggest that blood pressure is affected by environmental or genetic factors, but the low order of relationship is not strong enough to allow the identification of hypertensive families by the examination of one family member. Electrolytes and Hypertension Epidemiologic studies in Polynesia, Micronesia, Africa, and South America have suggested that populations who eat small quantities of salt are relatively free of hypertension. 5.6.16.25.30 It is well established that these populations add no salt to their food but eat only that which is present naturally. Analysis of the diets among such groups indicate daily intake of not more than 5 gm with some eating less than 1 gm. 4 By using 24 hour urine sodium measurements, it has been found . that Eskimos consume an average of less than 4 gm of salt per day, Marshall Islanders about 7 gm, white male Americans about 10 gm, Southern Japanese farmers about 14 gm, and Northern Japanese farmers about 26 gm.20 Meneely fed high-salt diets to rats and found a median survival of 8 months less than in control rats. If one were to translate this figure to man, this would be equivalent to a difference in the median duration of life of about 24 years. When potassium chloride was added to the diet there was no change in the hypertension resulting from the extra

BLOOD PRESSURE MEASUREMENTS IN CHILDHOOD

491

sodium chloride, but survival time increased. The median duration of life was increased by 7 months, although blood pressure persisted at intermediate levels similar to animals who received no extra potassium chloride. In South American Indians who excrete less than 30 mg of sodium in 24 hours, none had a systolic pressure greater than 120 mm Hg25 while in the Northern Japanese farmers who excrete more than 26 gm of salt in 24 hours, 84 per cent had systolic pressures greater than 140 mm Hg.28 Lauer, Filer, et al.1 8 studied the salt preferences and taste threshold for salt in 33 children whose mean blood pressures were less than the fifth percentile, 33 whose pressures were between the 47.5 and 52.5 percentiles, and 33 whose pressures were greater than the 95th percentile. There were no significant differences in preference and threshold between the three groups. These data suggest that if salt intake is a determining factor in the level of children's blood pressures, it is not related to their preference or ability to taste salt. The question whether children with higher pressures consume more salt for reasons unrelated to preference or threshold is yet unsettled. Evidence presented by Dabl and Meneely20 indicates that hypertension is common among animals and human beings who consume large amounts of salt. Interestingly, some populations who consume a low salt diet in the adolescent years fail to show a further rise in pressure, while peoples with salt freely available have increasing pressures with increasing age. 25 Populations who move from a low salt environment to a high salt environment show an increase of blood pressure with age similar to those populations who have always lived in a high salt culture. 7 It is suggested that the chronic intake of salt in amounts well in excess of requirements may play a primary role in the pathogenesis of hypertension in man. Since salt appetite is unrelated to salt requirements and is perhaps a learned phenomenon, one wonders if salt should be restricted from infancy in an effort to decrease the incidence of hypertension in young adulthood. It may be easier to restrict salt intake in the young rather than reverse salt appetite at a later age. A Working Definition of Hypertension Because blood pressure levels in childhood have not been studied for their predictive value in adult life, hypertension in childhood is most difficult to define. The Task Force on Pediatric Hypertension 29 has suggested that systolic or diastolic blood pressures greater than the 95th percentile on at least two separate occasions several months apart are an indication for clinical action. If the blood pressure continues to remain elevated to these levels, the patient should be evaluated to rule out secondary causes of hypertension. In the Muscatine Study,27 causes for secondary hypertension were detected early in the screening process. The initial examinations detected obesity, urinary symptoms, use of oral contraceptives, and coarctation of the aorta. Only 0.6 per cent of the total population studied had secondary hypertension; 0.3 per cent were obese with hypertension and 0.2 per cent were lean without identifiable causes for hypertension.

492

PATRICIA

A.

VON BEHREN AND RONALD

M.

LAUER

Treatment

There are few data to indicate which specific level of blood pressure should be treated in children. A Task Force on Pediatric Hypertension is presently addressing this problem. While some studies imply that preadolescents with pressures greater than 140 mm Hg systolic, 90 diastolic are at risk, these data are insufficient to support the view that antihypertensive agents should be initiated. It is clear that blood pressures in children are extremely labile and once elevated pressures do not persist in the majority of cases. Only patients with elevated blood pressures documented by multiple observations should be considered for therapy and then only after all causes for secondary hypertension have been eliminated. Associated mechanisms in the development of hypertension must also be considered. Diseases of the renal, cardiovascular, neurologic, or endocrine systems should be excluded. Drugs, toxins, and chemicals may cause hypertension in previously normotensive children. Chronic use of steroids may also result in hypertension. Oral contraceptives are often associated with elevations of blood pressure. Many toxic agents have been considered as causes of hypertension but only lead is recognized as a possible cause. For subjects with unexplained hypertension, salt intake, obesity, and cigarette smoking may be altered by changes in lifestyle. Although there is evidence to suggest that excess sodium intake may play a primary role in the genesis of hypertension, available evidence does not warrant general recommendations regarding dietary habits in the population at large. However, there is general agreement that individuals with hypertension should reduce salt intake. Drug therapy itself is not without potential hazards. While serious adverse pharmacologic effects have not be described at present, there has never been an adequate longitudinal evaluation of children receiving antihypertensive agents for prolonged periods. Thus, only subjects with extreme hypertension (diastolic pressures greater than 90 mm Hg in children under 6 years of age or greater than 100 mm Hg for children over 7 years of age) should be considered for drug therapy.

REFERENCES 1. Ayman. D.: Heredity in arteriolar (essential) hypertension: A clinical study of the blood pressure of 1,524 members of 277 families. Arch. Intern. Med., 53 :792-802, 1972. 2. Buck, C. W.: Persistence of elevated blood pressure first observed at age 5. J. Chron. Dis., 26:101-104, 1973. 3. Clarke, W. R, Woolson, R., Schrott, H., et al.: Tracking of blood pressure, serum lipids and obesity in children: The Muscatine Study. Abstract of 49th Scientific Session, American Heart Association, November 1976. 4. Dahl, K. L.: Medical progress. Salt intake and salt need. New Eng. J. Med., 258:1152, 1958. 5. Dahl, K. L., and Love, R. A.: Etiological role of sodium chloride intake in essential hypertension in humans. J.A.M.A., 164 :391-400, 1957. 6. Dahl, K. L., and Love, R A.: Evidence for relationship between sodium (chloride) intake and human essential hypertension. Arch. Intern. Med., 94 :525-531, 1954.

BLOOD PRESSURE MEASUREMENTS IN CHILDHOOD

493

7. Freis, K D.: Age, race, sex and other indices of risk in hypertension. Amer. J. Med., 55:275-280, 1973. 8. Hernandez, A., Goldring, D., and Hartman, A. F.: Measurement of blood pressures in infants and children by the Doppler ultrasonic technique. Abstract. Midwest Soc. Ped. Res., November 1970. 9. Hernandez, A., Meyer, D. A., and Goldring, D.: Blood pressure in neonates. Contemporary OB/GYN, 5 :34-37, 1975. 10. Heyden, S., Bartel, A., Hames, C., et al.: Elevated blood pressure levels in adolescents, Evans County, Georgia, seven year follow-up of 30 patients and 30 controls. J.A.M.A., 209:1683-1689, 1969. 11. Holland, W. W., and Humerfelt, S.: Measurement of blood pressure: Comparison of intra-arterial and cuff values. Brit. Med. J., 2: 1241-1243, 1964. 12. Johnson, B. C., Epstein, F. H., and Kjelsberg, M. 0.: Distributions and familial studies of blood pressure and serum cholesterol levels in a total community-Tecumseh, Michigan. J. Chron. Dis., 18:147-160,1965. 13. Kilcoyne, M. M., Richter, R. W., and Alsup, P.A.: Adolescent hypertension. I. Detection and prevalence. Circulation, 50:750-764, 1974. 14. Kirkendall, W. M., Burton, A. C., Epstein, F. H., et al.: Recommendations for human blood pressure determinations by sphygmomanometers. Circulation, 36:980, 1967. 15. Klein, B. K, Hennechens, C. H., Jesse, M. J., et al.: Longitudinal studies of blood pressure in offspring of hypertensive mothers. In Paul, 0., ed.: Epidemiology and Control of Hypertension. New York, Stratton Medical Books, 1975, p. 387. 16. Kohlstaedt, K. C., Moser, M., Francis, T., et al.: Panel discussion on genetic and environmental factors in human hypertension. Circulation, 17:728-742,1958. 17. Lauer, R M., Connor, W. K, Leaverton, P. K, et al.: Coronary heart disease risk factors in school children: The Muscatine Study. J. Pediat., 86 :697-706, 1975. 18. Lauer, R. M., Filer, L. J., Reiter, M. A., et al.: Blood pressure, salt preference, salt threshold and relative weight. Amer. J. Dis. Child., 130:493-497, 1976. 19. Londe, S., and Goldring, D.: Blood pressure standards for normal children as determined under office conditions. Clin. Pediat., 7 :400-403, 1968. 20 Meneely, G. R., and Dahl, K. L.: Electrolytes in hypertension: The effects of sodium chloride. The evidence from animal and human studies. MED. CLIN. N. AMER., 45 :271-283, 1961. 21. Miall, W. R, Heneage, P., Khosla, T., et al.: Factors influencing the degree of resemblance in arterial pressure of close relatives. Clin. ScL, 33 :271-283, 1967. 22. Miall, W. R, Kass, K H., Ling, J., et al.: Factors influencing arterial pressure in the general population in Jamaica. Brit. Med. J., 2 :497-506, 1962. 23. Moss, A. J., and Adam, F. H., eds.: Heart Disease in Infants, Children and Adolescents. Baltimore, Williams and Wilkins Co., 1968, p. 60. 24. Oberman, A., Lane, N. K, Harlan, W. R, et al.: Trends in systolic blood pressure in thousand aviator cohort over a twenty-four year period. Circulation, 36 :812-822, 1967. 25. Oliver, W. J., Cohen, K L., and Neel, J. V.: Blood pressure, sodium intake and sodium related hormones in the Yanomano Indian, a "no-salt" culture. Circulation, 52:146151, 1975. 26. Platt, R: Heredity in hypertension. Lancet, 1 :899-904, 1963. 27. Rames, L., Lauer, R. M., Clarke, W. R, et al.: Unpublished data. 28. Tahahosi, E., Sashi, N., Takeda, J., et al.: The geographic distribution of cerebral hemorrhage and hypertension in Japan. Hum. BioI., 29:139-166, 1957. 29. Task Force in Pediatric Hypertension: National Heart, Lung and Blood Institute and National High Blood Pressure Education Program. Pediatrics, in press. 30. Trusell, A. S., Kennelly, B. M., Hansen, J. D. L., et al.: Blood pressure ofKung bushmen in Northern Botswana. Amer. Heart J., 84 :5-12, 1972. 31. Veterans Administration Cooperation Study Group on Antihypertensive Agents: Effects of treatment on morbidity in hypertension. n. Results in patients with diastolic blood pressure averaging 90 through 114 mm Hg. J.A.M.A., 213 :1143-1152, 1970. 32. Voors, A. W., Foster, T. A., Frerichs, R R, et al.: Studies of blood pressures in children, ages 5-14 years, in a total biracial community. The Bogalusa Heart Study. Circulation, 54 :319-327, 1976. 33. Zinner, S. H., Levy, P. S., and Koass, K H.: Familial aggregation of blood pressure in childhood. New Eng. J. Med., 284 :401-404, 1971. 34. Zinner, S. H., Martin, L. F., Sachs, R, et al.: A longitudinal study of blood pressure in childhood. Amer. J. Epidemiol., 100:437-442,1974. Division of Pediatric Cardiology Department of Pediatrics University of Iowa Hospitals and Clinics . Iowa City, Iowa 52242 (Dr. Lauer)