Early growth and coronary heart disease in later life: longitudinal study

Statistical analyses

We examined trends in hazard ratios with neonatal and childhood measurements using Cox's proportional hazards model.11 As in previous analyses,5 we converted each measurement of height, weight, and body mass index (weight(kg)/(height(m)²)) for each boy to a Z score at each birthday. Z scores represent the difference from the mean value for the whole cohort, expressed in standard deviations.12 We did not assign a Z score at a particular age if the boy had not been measured within two years of that age.

Size at birth

Table 2 shows hazard ratios for coronary heart disease according to size at birth. The hazard ratios fell with increasing birth weight and ponderal index (birth weight(kg)/(length(m)³)). These trends were found in babies born at term or prematurely and therefore reflected low rates of intrauterine growth. Hazard ratios also fell with increasing head circumference (P=0.002, adjusted for gestation), but there were no trends with length at birth or placental weight.

Size at age 1 year

Table 3 shows that hazard ratios for coronary heart disease fell progressively with increasing weight, height, and body mass index at age 1 year. Small body size at this age predicted coronary heart disease independently of size at birth. In a simultaneous analysis the hazard ratios for coronary heart disease associated with a one standard deviation increase were 0.94 (95% confidence interval 0.83 to 1.06) for birth weight and 0.84 (0.75 to 0.94) for weight at 1 year. The corresponding figures for ponderal index at birth and body mass index at 1 year were 0.86 (0.77 to 0.97) and 0.87 (0.78 to 0.96).

Childhood growth

Figure 1 shows the childhood growth, expressed as mean Z scores, of the men who developed coronary heart disease. The Z score for the cohort is set at zero, and a boy maintaining a steady position as large or small in relation to other boys would follow a horizontal path on the figure. Boys who later developed coronary heart disease, however, having been small at birth and during infancy, had accelerated gain in weight and body mass index thereafter. In contrast, their heights remained below average, this difference being significant at each age from 1 to 5 years. We calculated the hazard ratios for coronary heart disease associated with gain in body mass index between the ages of 1 and 12 years. A one unit increase in standard deviation score was associated with a hazard ratio of 1.20 (1.08 to 1.33; P=0.0005).

Fig 1.

Growth of 357 boys who later developed coronary heart disease in a cohort of 4630 boys born in Helsinki

• Download figure
• Open in new tab
• Download powerpoint

In our earlier analyses of an older cohort of men born in Helsinki, whose childhood growth was recorded only from age 7 years onwards, we found that the effects of childhood body mass index on later coronary heart disease depended on ponderal index at birth.5 In the current analysis we therefore examined the simultaneous effects of ponderal index at birth and body mass index at each age. We found significant interactions between the two variables from age 2 years onwards. Figure 2 illustrates this. Boys who had a low ponderal index at birth increased their risk of coronary heart disease if they attained even average body mass index in childhood. In contrast, among boys with a high ponderal index there was no increased risk associated with increase in body mass index. Figure 2 describes body mass index at age 11 years, for consistency with studies of the older cohort, but data for other ages are similar. These are shown in table 4, in which the cohort is divided according to ponderal index at birth, using an index of 26 as the division. This division gives similar numbers of cases of coronary heart disease in the two groups.

Fig 2.

Hazard ratios for coronary heart disease according to ponderal index at birth and body mass index at age 11 years. Arrows indicate average values; lines join points with the same hazard ratios

• Download figure
• Open in new tab
• Download powerpoint

Table 4 shows the hazard ratios for coronary heart disease associated with a standard deviation increase in body mass index at each age from 1 to 12 years. In the low ponderal index group an increase in body mass index at each age from 3 years onwards was associated with increased risk of disease, whereas in the high ponderal index group an increase in body mass index at any age was associated with reduced risk. Table 5 illustrates these contrasting effects of body mass index at the middle of the age range. At this and all other ages the boys who could not be assigned Z scores were similar to the others in their body size at birth and in their later incidence of coronary heart disease.

Limitations of study

Our study was restricted to men who were born in Helsinki University Hospital, where about 60% of all births in the city occurred. This would introduce a bias only if the association between early growth and coronary heart disease differed between those born in the hospital and those born outside it. Our study was further restricted to men who had attended child welfare clinics, which were voluntary. The men may therefore be unrepresentative of all men living in Helsinki. In particular, attendance at child welfare clinics could have been related to socioeconomic circumstances. Poorer families may have made more use of the clinics because their children were less healthy, or better educated mothers may have been more willing to take advice on child rearing. We know that most of the children who attended the clinics did so regularly as each child had an average of 12 recorded measurements of height and weight. Similarly to the older cohort, however, we also have data on the occupations of the fathers of the men, at the time of the birth; 66% of the fathers were labourers. We know that at this time in Helsinki around 60% of men were labourers,11 and the social class distribution of our sample may therefore be similar to that of the city as a whole. Of the boys we identified through birth and clinic records, 84% were still alive and resident in Finland in 1971.

Consistency with other studies

Associations between low birth weight, low ponderal index, and later coronary heart disease (table 2) have been shown in other studies.1^–8 The association with slow growth during infancy is consistent with findings among men in Hertfordshire, where low weight at age 1 year similarly added to the increased risk of coronary heart disease associated with low birth weight. 1 13 The size of the simultaneous effects of birth weight and weight at age 1 in the Hertfordshire study was similar to the size of those in Helsinki. We have replicated the association between coronary heart disease and rapid growth in weight and body mass index after age 6 years, an association first shown in the older Helsinki cohort.5 We have now shown that the adverse effect of this association is evident by age 3 years and is confined to boys who were thin at birth (ponderal index <26). The association we found between low rates of height growth in early childhood (fig 1) and later coronary heart disease is consistent with the known association between the disease and short adult stature in men and short leg length measured in childhood.14^–16

Mechanics

Several possible mechanisms exist by which reduced fetal and infant growth and accelerated weight gain in childhood may lead to coronary heart disease. Babies who are thin at birth lack muscle, a deficiency which will persist, as the critical period for muscle growth is around 30 weeks in utero and there is little cell replication after birth.17 If they develop a high body mass index in childhood they may have a disproportionately high fat mass. This may be associated with the development of insulin resistance, as children and adults who had low birth weight but are currently heavy are insulin resistant. 18 19

Among men with a ponderal index at birth of >26, low body mass index in infancy and childhood was associated with increased risk of coronary heart disease. Findings in Hertfordshire suggest that the process underlying this may be disturbance of lipid metabolism and blood coagulation. Among men in Hertfordshire, low weight at age 1 year was associated with raised serum concentrations of total and low density lipoprotein cholesterol and apolipoprotein B, and also with raised plasma fibrinogen concentrations. 20 21 Circulating low density lipoprotein cholesterol and fibrinogen concentrations are regulated by the liver, and one possibility is that the high concentrations associated with low infant weight gain are persisting responses to impaired liver development during a critical phase in fetal life and infancy.21 During its development the liver remains “plastic” until about age 5 years, and its function may therefore be permanently changed by influences which affect its early growth.22^–24 Support for an important role for liver development in the early pathogenesis of coronary heart disease comes from findings among people born at the Jessop Hospital, Sheffield.25^–27

Conclusion

The interaction between ponderal index at birth and childhood body mass index shows that the pathogenesis of coronary heart disease cannot be understood within a “cumulative” model in which risks associated with adverse influences at different stages of life add to each other.28 Rather, the consequences of some influences, including a high body mass in childhood, depend on events at early critical stages of development. This embodies the concept of developmental “switches” triggered by the environment.29

We have found that gain in weight and body mass index during infancy are associated with reduced risk of coronary heart disease (table 3). This reduction occurs irrespective of body size at birth (table 4). This is important because promoting weight gain in infancy is standard practice. The associations between fetal and infant growth and later coronary heart disease in this study are strong and graded. Boys who at birth had a ponderal index >26 and who at age 1 year were above the cohort average for height (76.2 cm) and body mass index (17.7) had half the risk of coronary heart disease occurring before age 65 years. These are not unrealistic growth targets for public health interventions. There are obvious reservations about quantifying possible benefit in European populations today from observations made in Finland 40 to 50 years ago. Nevertheless these estimates of risk reduction may be compared with the much smaller estimates of the effects of improvements in adult lifestyle.30 Our findings add to the evidence that protection of fetal and infant growth is a key area in strategies for the primary prevention of coronary heart disease. Further benefit will come from preventing rapid weight gain after infancy in boys who were thin at birth. In these boys the adverse effects of rapid weight gain on later coronary heart disease are already apparent at age 3 years.