| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Medical Research Council Environmental Epidemiology Unit, University of Southampton, Southampton General Hospital, Southampton, United Kingdom SO16 6YD; and Cobbold Laboratories, Middlesex Hospital, London, United Kingdom W1N 8AA
Address all correspondence and requests for reprints to: Dr. Caroline Fall, Medical Research Council Environmental Epidemiology Unit, Southampton General Hospital, Southampton, United Kingdom SO16 6YD. E-Mail: . chdf{at}mrc.soton.ac.uk
Abstract
Low birth weight and weight in infancy are associated with adult insulin resistance and type 2 diabetes. A proposed mechanism is programming of the hypothalamic-pituitary-adrenal axis by intrauterine undernutrition, leading to persistently elevated cortisol concentrations. We examined 24-h serum cortisol profiles (samples every 20 min) in 83 healthy elderly men and women whose birth weight and infant weight were recorded. Variables derived from these profiles included trough, peak, and area under the curve concentrations; the time of onset, rate of rise, duration, and peak of the early morning cortisol rise; postprandial secretion; and regularity of secretion (approximate entropy). None of these parameters was related to birth weight, weight at 1 yr, or change in weight SD score between birth and 1 yr. Consistent with other studies, 07300900 h cortisol concentrations were higher in men and women of lower birth weight, although this was not statistically significant (P = 0.08). Our findings do not support the hypothesis that reduced intrauterine and infant growth are associated with continuously raised cortisol concentrations in old age. Programmed effects on the hypothalamic-pituitary-adrenal axis may influence reactivity rather than resting secretion.
EPIDEMIOLOGICAL studies have shown that low birth weight is associated with an increased risk of adult cardiovascular disease, hypertension, type 2 diabetes, and the insulin resistance syndrome (1, 2, 3, 4). The fetal programming hypothesis proposes that these arise from persistence of structural, hormonal, and metabolic adaptations made by the developing fetus in response to undernutrition (1).
A disturbance of the hypothalamic-pituitary-adrenal axis (HPAA) may be one such mechanism (5). In animals, stressful intrauterine events, including maternal undernutrition, lead to increased fetal cortisol concentrations (6, 7) and increased basal and/or stress-induced glucocorticoid secretion postnatally (8, 9, 10, 11, 12). In human populations, associations have been shown between low birth weight and increased postnatal HPAA activity, including elevated morning cortisol concentrations (5, 13) and an increased cortisol response to ACTH (14).
In a metabolic ward setting, we have now studied 24-h cortisol profiles in men and women whose birth weight and weight at 1 yr were recorded. We hypothesized that men and women of lower birth weight and infant weight have higher mean cortisol concentrations and a greater and more sustained elevation of the physiological rise in cortisol in the early morning.
Subjects and Methods
We studied 83 healthy adults, aged 6172 yr (45 men and 38 women), who were born and still lived in the United Kingdom county of Hertfordshire and whose weights at birth and 1 yr were recorded by health visitors. They were selected to represent a wide range of birth weight and infant weight gains (Fig. 1
) from a larger group (n = 765) who took part in earlier studies of cardiovascular risk factors (2, 15). We previously described their 24-h GH profiles (16). None was receiving medication known to influence cortisol secretion or hormone replacement therapy. Subjects known to have coronary heart disease and type 2 diabetes were excluded.
|
Serum cortisol was measured in all 72 samples using solid phase RIA (Coat-a-Count, Diagnostic Products, Los Angeles, CA). Within-assay coefficients of variation were 5.7%, 3.1%, and 2.6% at 28, 96, and 552 nmol/liter, and between-assay coefficients of variation were 6.3% and 4.5% at 138 and 276 nmol/liter, respectively. The sensitivity of the assay was 25 nmol/liter. Cortisol-binding globulin (CBG) was assayed in the first sample of the 24-h series using a commercial assay (Medgenics Diagnostics, Fleurus, Belgium) (17).
A number of variables were derived from the cortisol profiles to describe basal secretion and physiological peaks in the early morning and after meals. The trough and peak concentrations were those below which 5% and 95%, respectively, of all values lay during the 24-h period (Fig. 2
). The mean of all 72 values was used as an estimate of total secretion over 24 h (area under the curve). For comparison with the earlier study in this cohort (5), the mean morning cortisol concentration (07300900 h; the mean of samples 15) was calculated. The free cortisol concentration was derived from cortisol and CBG concentrations in the first sample, using a standard formula (17).
|
Finally, cortisol approximate entropy (ApEn), a measure of disorderliness in secretion, was analyzed. ApEn is a regularity statistic (18, 19, 20, 21, 22) that evaluates patterns in hormone secretion time series data and detects changes in underlying episodic behavior not reflected in peak occurrences or amplitudes. Larger ApEn values correspond to greater randomness, and smaller values to more instances of recognizable patterns in the data.
Statistical methods
Cortisol variables were examined in relation to birth weight and weight at 1 yr using multiple regression, with all variables as continuous, and with and without adjustment for age, sex, and body mass index. They were also examined in relation to catch-up or catch-down growth in infancy, measured as the change in SD scores for weight between birth and 1 yr. Mean cortisol concentrations, rate of rise in morning concentrations, and postprandial cortisol concentrations had skewed distributions and were log-transformed for the analysis.
Results
The mean (±SD) birth weight was 3.58 kg (±0.61) for men and 3.37 kg (±0.59) for women, and corresponding weights at 1 yr were 10.50 kg (±1.50) and 9.22 kg (±1.17). Two subjects had a birth weight below 2.5 kg (Fig. 1
). Mean age and body mass index at the time of the study were 69.3 yr and 26.7 kg/m2 in men and 69.7 yr and 27.6 kg/m2 in women. Five men and four women were hypertensive (defined as having a systolic blood pressure
160 mm Hg, a diastolic pressure
90 mm Hg, or a history of high blood pressure diagnosed by a doctor for which the subject was taking medication).
Effects of age, sex, and body mass index
Men had higher trough, peak, and mean cortisol concentrations (P < 0.001 for all), lower CBG concentrations (P = 0.02), and higher free cortisol concentrations (P < 0.001) than women (Table 1
). There were no significant differences between the sexes in the time of onset of the morning rise in cortisol, the rate of rise, or the degree to which levels were sustained. For most variables, there was no significant effect of age across this narrow age range; however, peak cortisol concentrations fell with age in women (P = 0.03). Mean cortisol and free cortisol concentrations were lower in men with a higher body mass index (P = 0.02 for both), but there was no effect of body mass index in women.
|
None of the variables examined showed a statistically significant relationship (either linear or quadratic) with birth weight, weight at 1 yr, or change in SD score for weight between birth and 1 yr, with or without adjustment for age and body mass index. Twenty-four-hour profiles were similar at all levels of birth weight and infant weight (Fig. 3
). As in the larger sample, of which these men and women were a subset, mean morning (07300900 h) serum cortisol concentrations were higher in men and women of lower birth weight, although this was not statistically significant (P = 0.3 unadjusted; P = 0.08, adjusted for age, sex, and body mass index). Concentrations fell by 28.1 nmol/liter (95% confidence intervals, -59.7 to +3.5 nmol/liter) per kg increase in birth weight. There was no similar trend with weight at 1 yr.
|
This was a large study of 24-h cortisol profiles in healthy elderly men and women. As cortisol secretion shows marked circadian variation and a pulsatile response to food intake, frequent sampling over 24 h gives a more complete picture of the physiological pattern of secretion than isolated blood samples. Profiles were similar at all levels of birth weight and infant weight. There was no evidence that trough, peak, or mean cortisol concentrations or early morning and postprandial cortisol surges are increased in individuals with reduced prenatal or infant growth, nor was there evidence that their cortisol secretion was more disorderly. Early morning (07300900 h) cortisol concentrations were higher in men and women of lower birth weight. This was of borderline statistical significance, but consistent with other studies.
The subjects in our study were born in Hertfordshire between 19201930, still lived in the county, took part in earlier research studies, and were willing to travel to London and have their hormone profiles measured. Their mean birth weight, weight at 1 yr, age, height, and body mass index were similar to those of the larger group from which they came. Studies carried out 24 yr earlier in this cohort showed cardiovascular risk factor profiles similar to those found in population studies of United Kingdom men and women of this age (body mass index, blood pressure, prevalence of impaired glucose tolerance, and serum lipid concentrations) (2, 3, 15). The cortisol measurements were made during a carefully controlled in-patient admission, and attempts were made to minimize the stress of the procedure by allowing an overnight settling-in period. Methods for characterizing cortisol profiles are not standardized, and the variables we derived were chosen for their mathematical simplicity to assess basal secretion as well as the size and duration of physiological peaks in the early morning and after meals.
In view of the old age of our subjects, one possible explanation of our mainly negative findings is that age-related changes in the HPAA may overshadow effects of early life programming. Compared with young adults, elderly men and women have higher mean and nadir cortisol concentrations, a reduced nocturnal quiescent period, flattening of the circadian amplitude, subtle differences in pulsatility, and reduced sensitivity to dexamethasone suppression (23, 24, 25, 26, 27). However, they maintain circadian rhythmicity and stress responses (23, 28). It seems unlikely that the old age of our subjects is an explanation for our findings. A potential disadvantage of the Hertfordshire cohort is that gestational age at birth was not recorded, making it impossible to distinguish low birth weight due to intrauterine growth retardation from that due to prematurity. Despite this, strong relationships have been shown in earlier studies between birth weight and adult disease outcomes such as cardiovascular disease and type 2 diabetes (1, 2, 3). The sample of men and women in this study was chosen to represent a wide range of birth weight and infant growth, because a striking feature of the earlier studies was the continuous relationship between birth weight/infant weight and cardiovascular disease, hypertension, diabetes, and morning cortisol concentrations (1, 2, 3, 5). However, this resulted in few subjects at the extremes of birth weight, which, in a smaller study like this, may have led to insufficient power to detect associations with birth/infant weight.
Only one other study, in Swedish children with short stature, has examined unstimulated glucocorticoid concentrations over a 24-h period in relation to size at birth. Consistent with our findings, it showed no difference in peak, nadir, or area under the curve cortisol concentrations in children born small from those born appropriate for gestational age (29). A population study of healthy United Kingdom children using 24-h glucocorticoid metabolite excretion showed a U-shaped relationship with birth weight, with higher excretion in children of both low and high birth weights (30). There was no evidence of this in our study. However, our finding of higher 07300900 h cortisol concentrations in men and women of lower birth weight, although of borderline statistical significance, was consistent with our earlier Hertfordshire study (which included the subset described here) and with findings in middle-aged adults in Preston, United Kingdom (13), and young adults in Australia (13) and South Africa (31). The effect size, a fall of 2030 nmol/liter·kg increase in birth weight, was similar in these studies. The main brake on early morning ACTH secretion is negative feedback exerted by glucocorticoids, acting via hippocampal GR. In experimental animals, fetal growth restriction associated with maternal food restriction or exposure to dexamethasone in late gestation reduces fetal hippocampal GR expression, resulting in increased ACTH and cortisol responses to CRH postnatally (7, 32). High morning cortisol concentrations in our study could reflect increased early morning ACTH secretion. However, concentrations 4, 5, and 6 h after the early morning rise provided no other evidence for a greater or more prolonged morning cortisol response in subjects of lower birth weight.
It is possible that programmed effects on the HPAA are more evident in a stressed or stimulated situation than at rest. In another Hertfordshire study (14) and in the South African children (31), the cortisol response to iv ACTH was increased in lower birth weight individuals. The high early morning cortisol concentration found in our and other studies may be a mild stress response to either fasting or the onset of blood sampling or may be related to the rise in cortisol seen after waking from sleep (33). We tried to minimize stress during the study; hence, the intensity of any such stress was unmeasured and would have varied widely between subjects. The exact time of waking was not recorded. In experimental animals, programmed alterations in HPAA activity resulting from a variety of intrauterine stimuli (8, 9, 10, 11, 12, 32, 34, 35) are more evident in stress-induced responses than in basal hormone concentrations. For example, compared with controls, adult rats born to mothers subjected to stress during the last week of pregnancy (8), lambs whose mothers were treated with corticosteroids in late pregnancy (32), and lambs whose mothers were subjected to nutrient restriction in early pregnancy (11) had normal basal cortisol concentrations, but increased ACTH and cortisol responses to a CRH challenge. Another recent study, however, showed no evidence of altered plasma corticosterone profiles, either basally or after stress, in adult rats who had experienced intrauterine growth retardation due to maternal protein deprivation during pregnancy (36). Apart from the Synacthen test data from Hertfordshire (14), there are few data in humans relating early events to stimulated glucocorticoid responses. Salivary cortisol concentrations during psychological and physical stress situations were not related to birth weight in Jamaican children, but numbers were small and cortisol concentrations did not rise above baseline during the period of stress (37). Salivary cortisol concentrations after the stress of vaccination, but not basal cortisol values, were higher at the age of 8 wk in babies born by assisted delivery (most stressful mode of delivery) than in those born by elective cesarean section (least stressful) (38).
In conclusion, resting 24-h cortisol profiles in healthy elderly men and women were similar across the range of birth weight and weight at 1 yr. Higher early morning cortisol concentrations in men and women of lower birth weight may indicate a more reactive HPAA. Future research into the early life programming of the HPAA in humans should focus on reactivity and responses to stress.
Acknowledgments
We thank the men and women who participated in this study, the nursing staff who coordinated the study and carried out the blood sampling, laboratory staff at University College, London, for measuring the cortisol concentrations, and Dr. Steve Pincus and his colleagues for ApEn analysis.
Footnotes
Abbreviations: ApEn, Approximate entropy; CBG, cortisol-binding globulin; HPAA, hypothalamic-pituitary-adrenal axis.
Received June 28, 2001.
Accepted January 31, 2002.
References
This article has been cited by other articles:
![]() |
J. W. Honour, R. Jones, S. Leary, J. Golding, K. K. Ong, and D. B. Dunger Relationships of Urinary Adrenal Steroids at Age 8 Years with Birth Weight, Postnatal Growth, Blood Pressure, and Glucose Metabolism J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4340 - 4345. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Kajantie, K. Feldt, K. Raikkonen, D. I. W. Phillips, C. Osmond, K. Heinonen, A.-K. Pesonen, S. Andersson, D. J. P. Barker, and J. G. Eriksson Body Size at Birth Predicts Hypothalamic-Pituitary-Adrenal Axis Response to Psychosocial Stress at Age 60 to 70 Years J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4094 - 4100. [Abstract] [Full Text] [PDF] |
||||
![]() |
D I. Phillips External influences on the fetus and their long-term consequences Lupus, November 1, 2006; 15(11): 794 - 800. [Abstract] [PDF] |
||||
![]() |
C. Power, L. Li, and C. Hertzman Associations of Early Growth and Adult Adiposity with Patterns of Salivary Cortisol in Adulthood J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4264 - 4270. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Jones, K. M. Godfrey, P. Wood, C. Osmond, P. Goulden, and D. I. W. Phillips Fetal Growth and the Adrenocortical Response to Psychological Stress J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1868 - 1871. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. I. W. Phillips and A. Jones Fetal programming of autonomic and HPA function: do people who were small babies have enhanced stress responses? J. Physiol., April 1, 2006; 572(1): 45 - 50. [Abstract] [Full Text] [PDF] |
||||
![]() |
N van Montfoort, M J J Finken, S le Cessie, F W Dekker, and J M Wit Could cortisol explain the association between birth weight and cardiovascular disease in later life? A meta-analysis Eur. J. Endocrinol., December 1, 2005; 153(6): 811 - 817. [Abstract] [Full Text] [PDF] |
||||
![]() |
T.-M. Hng, N. W. Cheung, and M. McLean Growth Hormone and Cortisol Dynamic Function in Relation to Birth Weight: A Study in Adult Twins J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2781 - 2786. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. C. Mcmillen and J. S. Robinson Developmental Origins of the Metabolic Syndrome: Prediction, Plasticity, and Programming Physiol Rev, April 1, 2005; 85(2): 571 - 633. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. M Reynolds, B. R Walker, H. E Syddall, R. Andrew, P. J Wood, and D. I W Phillips Is there a gender difference in the associations of birthweight and adult hypothalamic-pituitary-adrenal axis activity? Eur. J. Endocrinol., February 1, 2005; 152(2): 249 - 253. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. V. Ward, H. E. Syddall, P. J. Wood, G. P. Chrousos, and D. I. W. Phillips Fetal Programming of the Hypothalamic-Pituitary-Adrenal (HPA) Axis: Low Birth Weight and Central HPA Regulation J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1227 - 1233. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Q. Purnell, D. D. Brandon, L. M. Isabelle, D. L. Loriaux, and M. H. Samuels Association of 24-Hour Cortisol Production Rates, Cortisol-Binding Globulin, and Plasma-Free Cortisol Levels with Body Composition, Leptin Levels, and Aging in Adult Men and Women J. Clin. Endocrinol. Metab., January 1, 2004; 89(1): 281 - 287. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |