Perinatal and Childhood Origins of Cardiovascular Disease

Document Actions

Rae-Chi Huang, M.B., B.S., D.C.H., FRACP and Lawrie Beilin, M.B.B.S., M.D., FRCP, FRACP, AO,
School of Medicine and Pharmacology,
Royal Perth Hospital,
University of Western Australia (M570),
50 Murray St.,
Perth, WA 6000, Australia,
Tel : 61 8 9224 0258,
Fax: 61 8 9224 0246,
E-mail: rhuang@meddent.uwa.edu.au

Introduction

Features of the metabolic syndrome comprise a major risk for cardiovascular disease and will increase in prevalence with rising childhood obesity. Birthweight and early life influences affect development of obesity, hypertension, and dyslipidemia in children. Early life influences include maternal smoking during pregnancy, duration of breastfeeding, and postnatal weight gain.

Relationship between Birthweight and the Metabolic Syndrome

The literature in general has shown a linear negative relationship between birthweight and later development of the metabolic syndrome or coronary heart disease. We suggest that these relationships should be reviewed in different contemporary populations. An inverse association between birthweight and cardiovascular risk may be influenced by the inclusion of preterm infants and further modified by the rising prevalence of obesity and gestational diabetes in mothers, and overweight and obesity in children in both developed and developing countries.

These issues have been explored in a recent study of the "Raine" longitudinal childhood cohort in Perth, Western Australia in which a subset of children monitored at age 8 showed a U shaped relationship between birthweight and the risk of exhibiting a cluster of features of the metabolic syndrome (see Figure 1) . A similar U-shaped relationship between birthweight and the risk of diabetes has been observed in two other distinct populations, including Pima Indians and Asians. The results of this Australian birth cohort study need to be taken in the context of an ongoing longitudinal study. These children were analyzed at 8 years old and the larger cohort of 1,800 is being followed through adolescence. As they pass through puberty, changes in the relationships may emerge.

Why should the relation between birthweight and clusters of cardiovascular risk factors differ between populations? Apart from the younger age of the Raine cohort we hypothesized that the shape of the curve for birthweight plotted against subsequent cardiovascular risk can differ according to population nutritional status and obesity rates. By extending the curve further to the right of the birth weight axis there is a greater prominence of higher birthweights in those at risk. In older cohorts the negative slope at the left hand of the curve has been more evident, and reported as a continuous inverse relationship between birthweight and the metabolic syndrome in adults.

We conjecture that a right shift in the birth weight plotted against risk of cardiovascular disease curve may be developing due to increasing caloric excess and decreasing physical activity in Western society over the last 80 years. Undernutrition during pregnancy causing poor fetal growth is now less common; it is far more likely to exist in earlier cohorts affected by the Great Depression, World Wars, and the much-quoted Dutch famine and in contemporary developing countries. Consistent with this, the mean birth weight of the Australian cohort was 3,501 g compared with a mean of 3,166 g and 3,220 g during the Dutch famine and 2,600 g in girls and 2,800 g in boys in contemporary India. The mean male birth weight in the Australian cohort is higher (3,533 g, SD = 458 g) compared with the males in the Helsinki Studies (3,456 g SD = 490) , but similar to those in the Hertfordshire cohort (3,500 g SD 600 g) . Low birth weight is now more likely to be due to preterm birth, placental dysfunction, obstetric complications, maternal disease, poor social environment, or genetic reasons. Important in developmental origins concepts is a gross mismatch between pre- and post-natal environments. This is less likely to exist to a large extent in current Western society, except with obstetric complications or in groups such as Native Americans, Australian Aborigines, and recent immigrants.

Postnatal weight gain

The components of the adult metabolic syndrome are largely driven by obesity. Our Australian childhood cohort showed large BMI and weight differences between children at high and low risk of the cluster of abnormalities including high blood pressure, BMI, and relative dyslipidemia. Postnatal weight gain after 12 months of age was the dominant factor associated with the presence of the metabolic syndrome cluster at age 8, a finding similar to that recently reported for BP and obesity in the same cohort . These results are in keeping with a study of 300 contemporary British 5 year olds , and with analysis of a healthy high school population which showed that postnatal weight gain and current weight were the dominant factors associated with insulin resistance rather than low birth weight.

Weight, BMI, and waist girth are surrogate measures of body fat mass. Studies focusing on body composition and body fat/muscle ratio may provide a better indication of the impact of postnatal growth on risk of cardiovascular disease. We have observed that high risk children are also taller with greater arm circumferences from 3 years of age onwards.

Timing or trajectories of postnatal weight gain are likely to be important. The patterns of postnatal weight gain that evolve into greatest cardiovascular risk have differed in studies. The Australian longitudinal study of 8-year-old children demonstrated that excess postnatal weight gain in the first year or weight at 1 year does not have an independent influence on development of the high risk cluster This differs from findings from a Helsinki Cohort of 8,760 people which demonstrated that small size at birth, low BMI at 2 years of age, and high BMI at 11 years of age were each associated with later coronary events Perhaps the suggestion from the two studies above is that there is a window in early infancy of 1 to 2 years where excess postnatal weight gain does not determine cardiovascular risk. In contradiction to this however, an earlier study showed that infants with early rapid weight gain in the first 4 months of life are at greater risk These findings need to be explored in more detail with maturation of cohorts.

Smoking in pregnancy

Smoking appears to be an important modifier of the relationship between birthweight and cardiovascular risk. Despite maternal smoking being associated with lower birth weight and increased SBP in offspring, a positive relationship was seen between birth weight and SBP in early childhood when mothers were smokers in pregnancy We observed that newborns at greatest risk of developing the metabolic syndrome cluster were those large for gestational age born to mothers who smoked throughout pregnancy. We hypothesize that within the maternal smoking group, well known to cause lower birth weight , those newborns who, not only escape growth retardation, but are in fact large for gestational age, are unusual due to genetic make-up or exposure to maternal hyperglycemia. They are subsequently selected for particular risk of cardiovascular disease. Conversely, for babies born to non-smokers, the greatest risk is in lowest percentage expected birthweight quintile. However, smoking is associated with socioeconomic status and adverse lifestyle which could also confound these relationships.

Breast feeding

Retrospective studies have shown variable breast feeding effects ranging from modest protective to possible adverse effects upon the risk of ischemic cardiovascular disease in adulthood. Further, an association between longer breast feeding and lower risk of overweight between 9-14 years of age has been reported. The Australian Raine cohort showed a lower likelihood of the high risk metabolic syndrome cluster in children breastfed for = 4 months compared to < 4 months and that breast feeding = 4 months is associated with lower weight at age 1 year. However, these observations may be confounded by the fact that shorter duration of breastfeeding is also associated with maternal characteristics such as obesity, smoking, and lower education.

Conclusion

The relationship between birthweight and cardiovascular disease may not be a straight forward negative linear relationship. To be applicable to current Western disease trends, it needs to be studied in contemporary, well-nourished populations of full-term newborns. We found a prominence of higher as well as lowest birthweights in those at risk while investigating such a population. Future health programs should focus on both pre- and post-natal factors (reducing excess childhood weight gain and smoking during pregnancy). Possibly the greatest benefits may arise from targeting the heaviest, as well as lightest newborns, especially with a history of maternal smoking during pregnancy.