Birthweight, body-mass index in middle age, and incident coronary heart disease

Several studies have shown a relation between fetal development, as shown by birthweight, and later coronary heart disease. This study investigated whether this relation is predominantly the consequence of early life exposures, or can best be explained in terms of an interaction between influences in early life and in adulthood.

This prospective study in Caerphilly, South Wales, included 1258 men, aged 45-59 at initial screening, who were able to provide birthweight data. These men are from an initial cohort of 2512 men, from whom information has been obtained in a series of examinations since 1979 on health-related behaviours, incidence of coronary heart disease, and risk factors. The main outcome measure was fatal and non-fatal coronary heart disease during 10 years of follow-up.

Higher birthweight was related to lower risk of coronary heart disease during the follow-up period: coronary heart disease occurred in 46 (11.6%) men in the lowest birthweight tertile, 44 (12.0%) of those in the middle tertile, and 38 (9.1%) of those in the highest tertile (p = 0.03). Stratification of the cohort by body-mass index (BMI) revealed a significant interaction such that the inverse association between birthweight and risk of coronary heart disease was restricted to men in the top tertile of BMI (interaction test p = 0.048 adjusted for age, and p = 0.012 fully adjusted). Within the top BMI tertile, coronary heart disease occurred in 19 (16.4%) of men in the lowest birthweight tertile, 13 (12.6%) of those in the middle tertile, and 13 (7.5%) of those in the highest tertile (p = 0.0005). These associations were not changed substantially by adjustment for age, father's social class, own social class, marital status, fibrinogen and cholesterol concentrations, systolic blood pressure, and smoking history.

The association between birthweight and risk of coronary heart disease cannot be explained by associations with childhood or adulthood socioeconomic status. Nor do conventional risk factors for coronary heart disease in adulthood account for the association. However, there is an important interaction between birthweight and BMI such that the increased risk of coronary heart disease associated with low birthweight is restricted to people who have high BMI in adulthood. Risk of coronary heart disease seems to be defined by the combined effect of early-life and later-life exposures.