Size at birth, gestational age and cortisol secretion in adult life: foetal programming of both hyper- and hypocortisolism?

Clin Endocrinol (Oxf). 2002 Nov;57(5):635-41. doi: 10.1046/j.1365-2265.2002.01659.x.


Objective: Recent studies have suggested that lifelong programming of the hypothalamic-pituitary-adrenal (HPA) axis in utero is an important mechanism in explaining the link between small size at birth and adult cardiovascular disease. However, direct evidence from human birth cohorts has so far been contradictory. We set out to study reasons for this discrepancy by examining the relationship between adult HPA axis function and birthweight and body proportions at birth in a group of elderly subjects with detailed birth records.

Design: Birth cohort study.

Subjects: Four hundred and twenty-one men and women (mean age 69.5 years, range 65.1-75.8 years) born at term in Helsinki, Finland, during 1924-33, with body size and gestational age at birth recorded.

Measurements: Fasting serum cortisol and cortisol-binding globulin concentrations. The concentration of free cortisol was estimated by their ratio.

Results: There was no significant correlation between fasting cortisol concentrations and birthweight in either men or women. However, there was a weak inverse association between fasting cortisol and length at birth in women but not in men. There was also a significant positive association between cortisol and ponderal index in both genders. We found that the association between foetal growth on fasting total and free cortisol concentrations differed in subjects born at different gestational ages. In subjects born before 39 weeks of gestation, both total and free cortisol showed inverse correlations with birthweight (P = 0.02 and P = 0.09, respectively) and length at birth (P = 0.001 and P = 0.02), whereas in subjects born after 40 weeks of gestation there were positive correlations with birthweight (P = 0.06 and P = 0.002) and ponderal index at birth (P = 0.003 and P = 0.003). The interactions between birthweight and gestational age were statistically significant (P = 0.01 for total and P = 0.003 for free cortisol).

Conclusions: These data suggest that the relationship between size at birth and cortisol concentrations in adult life is different in subjects born at different gestational ages: both hyper- and hypocortisolism may arise as a consequence of foetal programming of the hypothalamic-pituitary-adrenal axis during intrauterine life.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Adult
  • Aged
  • Birth Weight
  • Body Constitution*
  • Cardiovascular Diseases / blood
  • Cardiovascular Diseases / embryology
  • Cardiovascular Diseases / physiopathology*
  • Carrier Proteins / analysis
  • Cohort Studies
  • Female
  • Gestational Age
  • Humans
  • Hydrocortisone / blood
  • Hydrocortisone / metabolism*
  • Hypothalamo-Hypophyseal System / embryology
  • Infant, Newborn
  • Male
  • Pituitary-Adrenal System / embryology
  • Pregnancy
  • Prenatal Exposure Delayed Effects


  • Carrier Proteins
  • cortisol binding globulin
  • Hydrocortisone