Fetal and Childhood Growth and Hypertension in Adult Life
Abstract—The association between low birth weight and raised blood pressure has been extensively replicated. Little is known about the way childhood growth modifies the effects of low birth weight. We report on the fetal and childhood growth of 1958 men and women who received treatment for hypertension and belong to a cohort of 7086 people born in Helsinki, Finland, during 1924–1933. As expected, the men and women who developed hypertension had low birth weight (P=0.002). They were also shorter in body length at birth (P=0.02). After birth they experienced accelerated growth, so that by 7 years their heights and weights were approximately average. In a simultaneous regression, both birth length and tall height had statistically significant although opposing effects on hypertension (P=0.003 for birth length and 0.009 for height at 7 years). Accelerated postnatal growth was associated with better childhood living conditions. Children who later developed both hypertension and type 2 diabetes, rather than hypertension alone, had small placental size as well as small body size at birth, and their accelerated postnatal growth continued beyond 7 years. We suggest that hypertension may originate through retarded growth in utero followed by accelerated postnatal growth as a result of good living conditions. Retarded fetal growth leads to permanently reduced cell numbers in the kidney and other tissues, and subsequent accelerated growth may lead to excessive metabolic demand on this limited cell mass.
The association between low birth weight and raised blood pressure in later life has now been reported in more than 50 published studies of men, women, and children.1 2 It has been shown to result from retarded fetal growth rather than premature birth. The “fetal origins” hypothesis proposes that the association reflects permanent resetting of blood pressure by undernutrition in utero.3 Birth weight is a summary measure of fetal growth; associations between more precise indices of fetal growth retardation and raised blood pressure have been described.4 In different studies, babies who are short or thin, have small placentas, or are disproportionately small in relation to the size of their placentas have all been shown to have raised blood pressure.5 6 7
Fetal undernutrition may occur if the mother is malnourished, which will be reflected in her body composition, nutritional stores, or diet during pregnancy or may result from failure of the fetal supply line, which includes the placenta. Experiments in animals show that fetal malnutrition induced in different ways and at different stages of gestation has different short-term effects on body proportions and different long-term effects on the offspring’s physiology and metabolism.8 9 The associations between raised blood pressure and indices of fetal growth retardation that differ between one study and another may therefore reflect different causes of fetal undernutrition.
There is preliminary evidence that the effects of retarded fetal growth on later cardiovascular disease are modified by postnatal growth. In a study of men born in Uppsala, Sweden, the highest blood pressures were found in those who had low birth weight but were tall as adults.10 In a study of men born in Helsinki, Finland, whose birth size and childhood growth were recorded, the highest rates of coronary heart disease were found in those who were thin at birth but whose weight caught up through accelerated weight gain in early childhood.11 We report here on 7086 men and women from the Helsinki cohort, of whom 1958 were being treated for hypertension.
The study cohort consists of 7086 individuals who were born during 1924–1933 at the Helsinki University Central Hospital, who went to school in Helsinki, were resident in Finland in 1971, and remained in Finland thereafter. Each member of the study cohort had both a detailed birth record and a school health record. Details of the birth records and the school health records have been previously described.11 12 Data on the newborn babies include birth weight, length, head circumference, and placental weight. Data on the mothers include age, parity, and date of the last menstrual period. The school health records include an average of 10 (SD 4) measurements of height and weight between the ages of 6 and 16 years. They also include the number of people living in the child’s home, recorded at the time of first examination, and the number of rooms. The maternal, neonatal, and childhood characteristics of the 7086 men and women in the study cohort have already been published.11 12
In 1971 every Finnish citizen was assigned a unique personal identifier. We used this number to identify 1958 people with hypertension and 471 people with type 2 diabetes in the cohort. Information was collected from the nationwide database of the Social Insurance Institution’s Register of people on medication for chronic diseases. In Finland the costs of antihypertensive drugs are partly reimbursed by the state, subject to the approval by a physician who reviews each case history, and antidiabetic drugs prescribed by a physician are free of charge. In our study we therefore defined hypertension and type 2 diabetes by the use of medication for these disorders.
Using the fathers’ occupation, which was recorded on the birth records, we grouped the men and women according to a social classification used by the Central Statistical Office. Overall, 78% of the fathers were laborers and were classified as lower social class. Fourteen percent were middle class or self-employed and were classified as higher social class. The social class of 8% of the cohort was unclassified. Through the Office of Statistics, the personal identification numbers were used to identify data on adult social class obtained at the 1970/1971 census. Sixty-one percent of the men and women were classified as higher officials, lower officials, or self-employed, which we grouped as upper social class. The remainder were laborers or in other employment and were grouped as lower social class. There were no social class data for 234 subjects.
Tests for trends were based on multivariate logistic regression using continuous variables, which included year of birth to adjust for the effects of age. As previously described, we converted each height, weight, and body mass index (BMI) (weight/height2) measurement for each individual to a Z score using the method of Royston.13 Interpolation between successive Z scores with a piecewise linear function was performed, and a Z score was obtained at each birthday from age 7 to age 15 years. These Z scores were back transformed to obtain the corresponding height, weight, and BMI at these ages.
We found that 1958 people, 975 men and 983 women, received reimbursement for antihypertensive medication. The cumulative incidence of hypertension in the study cohort of 7086 people was therefore 27.6% (26.8% among men and 28.5% among women). As we have previously reported, 471 of the cohort (286 men and 185 women) were on medication for type 2 diabetes,14 of whom 250 were also on medication for hypertension.
Size at Birth
Table 1⇓ shows that the cumulative incidence of hypertension among men and women fell with increasing birth weight, length, and ponderal index (birth weight/length3). There was no statistically significant trend with either placental weight or the ratio of placental weight to birth weight. There was no trend with length of gestation, and the trends in Table 1⇓ were only slightly changed by adjustment for gestation. There were no trends with parity or maternal age.
Figure 1⇓ shows the childhood growth of the men and women with hypertension with mean Z scores for height, weight, and BMI at each age from 7 to 15 years. The mean values for all 7086 subjects are zero; the SDs are 1. At the age of 7 years, the mean heights and weights of boys and girls who later developed hypertension were around or above the average. Their body size had therefore caught up since birth. Table 1⇑ shows that the effects of birth weight and length on the risk of hypertension are strengthened by adjusting for weights and heights at 7 years. In a simultaneous regression with birth weight and weight at 7 years, both had statistically significant, although opposing, effects on hypertension (P=0.0003 for birth weight and 0.02 for weight at 7 years). In a simultaneous regression with birth length and height at 7 years, the probability values were 0.003 for length and 0.009 for height.
Boys and girls who later developed hypertension became taller and heavier than the other boys and girls between 7 and 15 years of age (in boys, P=0.002 for height at 15 years and P=0.005 for weight; in girls, P=0.17 for height at 15 years and P=0.03 for weight). The girls who later developed hypertension were also on average heavier than other girls (P=0.04 for BMI at age 15 years).
Body Size of Mother
Hypertension was not related to the mother’s height or weight, but the incidence rose with increasing maternal BMI (P=0.04). This association was strengthened in a simultaneous analysis with birth weight (P=0.002 for maternal BMI and P=0.0002 for birth weight). The incidence of hypertension was unrelated to parity but rose with increasing maternal age (P=0.03). This association with age became nonsignificant in a simultaneous regression analysis with mother’s BMI.
Socioeconomic Status in Childhood
Hypertension was not related to social class at birth. The average number of rooms in the houses where the men and women grew up was 2 (range, 1 to 14). Forty-seven percent lived in homes with only 1 room. The number of rooms in the houses was not related to the later development of hypertension. There were on average 4 other persons in the home during the childhood of these men and women. The cumulative incidence of hypertension fell with increasing numbers of inhabitants in the home (P=0.005). This was independent of size at birth. Children in houses with more inhabitants tended to be shorter and lighter (correlation coefficients at 7 years=−0.17 for height and −0.15 for weight).
Socioeconomic Status in Adult Life
The cumulative incidence of hypertension was greater in the upper social class. Among men, the incidence was 39.6% in the upper social class and 24.3% in the lower (P<0.001). The corresponding figures for women were 29.7% and 24.9% (P=0.008). These associations remained statistically significant in simultaneous analyses with birth size and childhood growth.
Hypertension and Type 2 Diabetes
We divided the subjects with hypertension into 2 groups according to the presence or absence of type 2 diabetes. The cumulative incidence of both hypertension alone and hypertension with type 2 diabetes fell with increasing birth weight, length, and ponderal index (P=0.009 and P=0.02 for birth weight; P=0.05 and P=0.10 for birth length; P=0.06 and P=0.12 for ponderal index). Findings were similar in men and women. Among subjects with hypertension alone, there was no trend with placental weight, whereas among men and women with both hypertension and type 2 diabetes, the cumulative incidence fell with increasing placental weight. In contrast, among subjects with hypertension alone, the cumulative incidence rose as the ratio of placental weight to birth weight increased, whereas there was no similar trend among men and women with both hypertension and type 2 diabetes. Since the ratio falls during gestation, we adjusted for length of gestation in a simultaneous regression, which strengthened the association (Table 2⇓).
Figure 2⇓ shows the childhood growth of the 2 groups. Both experienced accelerated growth in body size between birth and 7 years. Thereafter, boys and girls who later developed hypertension alone were similar to all other boys and girls from 7 to 15 years. In contrast, boys and girls who later developed both hypertension and type 2 diabetes had above-average rates of growth (in boys, P=0.03 for growth in height, P=0.06 for growth in weight; in girls, P<0.001 for growth in both height and weight). They became taller and heavier than the other boys and girls (in boys, P=0.001 for height at 15 years and P=0.008 for weight; in girls, P=0.001 for height at 15 years and P<0.001 for weight). The girls were also heavier than other girls (P<0.001 at age 15 years).
We studied the birth size and childhood growth of 1958 men and women being treated for hypertension within a cohort of 7086 people who were born in Helsinki, Finland. The cohort was defined by the availability of birth and school records, and we were able to trace 91% of the subjects. This study of hypertension was made possible because the costs of antihypertensive medication are reimbursed to the patients in Finland, and a register of all people receiving such medication is maintained. The cumulative incidence of 27.6% reflects the age of members of our study cohort, who were born during 1924–1933. No previous study has been able to address the association between adult hypertension and both birth size and childhood growth. Consistent with other studies of the associations between birth weight and blood pressure, we found that hypertension was associated with low birth weight in relation to gestational age.1 2 It was also associated with 2 more specific markers of fetal growth retardation, short body length and thinness at birth, associations that have also been recorded previously.4 5 6 7 By the age of 7 years, the weights and heights of the boys and girls who later developed hypertension had increased until they reached the average.
Short body length and thinness at birth may reflect the effects of fetal undernutrition, possibly acting through different processes at different times in gestation.4 Because of the range of mechanisms that control blood pressure, one may speculate that undernutrition at different times in gestation may program hypertension through different mechanisms, but little is known about this. In this cohort, catch-up growth between birth and 7 years has been shown to be associated with the later development of coronary heart disease.11 12 We have now shown that it is also associated with hypertension and that the risk of hypertension associated with small size at birth is amplified by rapid growth between birth and 7 years. Previous studies have shown that rapid growth in height and weight is associated with higher blood pressure in childhood and adult life.1 10
Rapid growth in childhood has been implicated in the development of essential hypertension.15 Children who are growing more rapidly have high blood pressure for their age. Because blood pressure “tracks” through childhood, this may lead to higher blood pressure in adult life. One hypothesis suggests that essential hypertension is determined by 2 separate mechanisms, a growth-promoting process in childhood and a self-perpetuating mechanism in adult life.15 One possible link between accelerated childhood growth and hypertension is that retarded fetal growth leads to permanently reduced cell numbers in tissues such as the kidney, in which there is no further cell replication after birth.16 17 Subsequent accelerated growth may lead to excessive metabolic demand on this limited cell mass. Brenner and Chertow16 have suggested that reduction in the number of nephrons leads to hyperperfusion of each nephron and resulting glomerular sclerosis, further nephron death, and a cycle of increasing blood pressure and nephron death.
Our findings show that catch-up growth is related to lower levels of crowding in the home. Men and women who grew up in houses with fewer people in them had more rapid growth between birth and 7 years. This could be a result of lower rates of infection.18 Fewer people in the house predicted hypertension independently of size at birth. Although social class at birth did not predict hypertension, there was an association with higher social class in adult life that was independent of birth size and childhood growth. The highest incidences of hypertension were therefore associated with small size at birth, growing up in less crowded homes, catch-up in size in childhood, and high social class in adult life.
We examined the variables that were associated with hypertension occurring together with type 2 diabetes. Although placental size did not predict hypertension in the entire cohort, when we subdivided the hypertensive subjects according to the presence or absence of type 2 diabetes we found opposing effects. Hypertension alone was associated with large placental size in relation to birth weight, whereas hypertension and type 2 diabetes were associated with small placental size. Both these associations have been described before.5 19 Their coexistent and statistically opposite effects in the same cohort may explain why some studies have failed to find associations between placental weight and later blood pressure.20 Large placental size in relation to birth weight may indicate a fetal adaptation whereby the fetus enlarges the placenta to extract more nutrients from the mother. Small placental size may indicate that inadequate placental development is the proximate cause of fetal undernutrition.
Although the boys and girls who later developed hypertension and type 2 diabetes were similar to those who developed hypertension alone in body size at birth and catch-up growth before 7 years, their childhood growth differed in that they had rapid growth in height and weight and attained above-average body size between 7 and 15 years. We do not know their adult body size, but this is closely predicted by body size at 15 years. This pattern of growth shown by people with both hypertension and type 2 diabetes is similar to the pattern of growth shown by all subjects with type 2 diabetes in the cohort.14
We conclude that the development of hypertension in this cohort was associated with reduced fetal growth and catch-up growth in early childhood. This catch-up growth occurred because the children had better living conditions and a less crowded home than other children. People who developed both hypertension and type 2 diabetes had a pattern of fetal and childhood growth similar to those who developed hypertension alone, but the origins of their reduced fetal growth may lie in small placental size, and their accelerated postnatal growth continued beyond the age of 7 years.
This study was supported by the British Heart Foundation, Jahnsson Foundation, and Finska Läkaresällskapet. We thank Terttu Nopanen, Tiina Saarinen, Hillevi Öfverström-Anttila, Liisa Toivanen, and Hanna Pehkonen for abstracting the data from the records. Sigrid Rosten was responsible for data management.
- Received March 30, 2000.
- Revision received April 19, 2000.
- Accepted May 30, 2000.
Barker DJP. Fetal origins of coronary heart disease. BMJ. 1995;311:171–174.
Barker DJP, Bull AR, Osmond C, Simmonds SJ. Fetal and placental size and risk of hypertension in adult life. BMJ. 1990;301:259–262.
Moore VM, Miller AG, Boulton TJC, Cockington RA, Hamilton Craig I, Magarey AM, Robinson JS. Placental weight, birth measurements, and blood pressure at age 8 years. Arch Dis Child. 1996;74:538–541.
Taylor SJC, Whincup PH, Cook DG, Papacosta O, Walker M. Size at birth and blood pressure; cross sectional study in 8–11 year old children. BMJ. 1997;314:475–480.
Barker DJP. Mothers, Babies and Health in Later Life. 2nd ed. Edinburgh, Scotland: Churchill Livingstone; 1998.
Leon DA, Koupilova I, Lithell HO, Berglund L, Mohsen R, Vagero D, Lithell U-B, McKeigue PM. Failure to realise growth potential in utero and adult obesity in relation to blood pressure in 50 year old Swedish men. BMJ. 1996;312:401–406.
Eriksson JG, Forsen T, Tuomilehto J, Winter PD, Osmond C, Barker DJP. Catch-up growth in childhood and death from coronary heart disease: longitudinal study. BMJ. 1999;318:427–431.
Forsen T, Eriksson JG, Tuomilehto J, Osmond C, Barker DJP. Growth in utero and during childhood among women who develop coronary heart disease: longitudinal study. BMJ. 1999;319:1403–1407.
Merlet-Benichou C, Leroy B, Gilbert T, Lelievre-Pegorier M. Retard de croissance intra-uterin et deficit en nephrons [Intrauterine growth retardation and inborn nephron deficit]. Medecine/Sciences. 1993;9:777–780.
Tanner JM. Foetus into Man. 2nd ed. Castlemead, UK: Ware; 1989.
Whincup P, Cook D, Papacosta O, Walker M. Birth weight and blood pressure: cross sectional and longitudinal relations in childhood. BMJ. 1995;311:773–776.