Main

It is well recognized that hypercortisolemia predisposes to a dysmetabolic phenotype, with accumulation of visceral fat, hyperglycemia, insulin resistance, decreased HDL (C), increased triglycerides, and high blood pressure (BP) (1).

Glucocorticoids have long been recognized to be essential for maintenance of normal BP and, when in excess, either generally or locally, produce hypertension (2). An association between cortisol and both systolic BP (SBP) and diastolic BP (DBP) has been reported by different authors (1,3,4). Discussion exists as to whether prenatal growth may mediate these associations (3,5).

There is considerable interest in the notion that cortisol may play a role in some forms of essential hypertension, and it has been suggested that a higher cortisol production may contribute to 30% of all cases of hypertension (68). Recent data indicate that hypercortisolemia may induce hypertension by an up-regulation of oxidative stress and impairment of NO availability (9).

Several studies suggested that cortisol may play a role in the development of atherosclerosis (10,11). Smith et al. (12) found an association between cortisol and incident ischemic heart disease. Eller et al. (13) described associations between salivary cortisol and progression in carotid intima-media thickness (IMT), a preclinical marker of atherosclerosis, in women. In adolescents with insulin-dependent diabetes mellitus, carotid IMT was shown to be increased and to be related to higher urinary free cortisol (14).

However, the contribution of physiological variations in serum cortisol to vascular risk markers in children is largely unknown. Because childhood risk factors predict adult risk for cardiovascular disease (15), it is important to identify these risk factors early in life. We posited that morning serum cortisol is an early risk factor for cardiovascular disease. In this study, we examined the cross-sectional associations between morning serum cortisol and vascular risk markers: BP and carotid IMT, in a sample of healthy prepubertal children.

SUBJECTS AND METHODS

Population and ethics.

The study group consisted of 223 school-aged healthy Caucasian children (106 boys and 117 girls; age, 6.8 ± 0.1 y). Subjects were included in a study of cardiovascular risk factors in children and were consecutively recruited among those seen for well-child check-up visits by their primary care pediatricians in Alt Empordà, a region in Northern Spain. Participation ranged from 50 to 70% among the different clinics. All children were apparently healthy and their morning serum cortisol levels were within normal values (16).

Inclusion criteria were 1) age between 5 and 9 y and 2) absence of puberty, as based on the standards by Marshall and Tanner. Exclusion criteria were as follows: 1) major congenital abnormalities; 2) abnormal blood counts, liver or kidney or thyroid functions; 3) evidence of chronic illness or chronic use of medication; and 4) acute illness or medication use within the previous month of inclusion.

The research was approved by the Institutional Review Board of Dr. Josep Trueta Hospital. Signed consent was obtained from the parents.

Clinical assessments.

Clinical examination was performed in the morning, in the fasting state followed by venous blood sampling. A local anesthetic cream was used to minimize the discomfort of venipuncture. Weight was measured wearing light clothes with a calibrated scale, and height was measured with a Harpenden stadiometer. BMI was calculated as weight divided by the square of height in meters. Age- and sex-adjusted SD scores (SDS) for BMI were calculated using regional normative data. Waist circumference was measured in the supine position at the umbilical level.

BP and heart rate (HR) were measured before venous blood sampling in the supine position on the right arm after a 10-min rest; an electronic sphygmomanometer (Dinamap Pro 100; GE Healthcare, Chalfont St. Giles, United Kingdom) with cuff size appropriate for arm circumference was used. Two to three measurements were taken in each individual, and results were averaged.

Body composition was assessed by bioelectric impedance (Hydra Bioimpedance Analyzer 4200; Xitron Technologies, San Diego, CA). Fat mass was calculated as the difference between weight and lean mass.

Intima-media thickness.

In a subset of subjects (n = 91) who agreed to participate further in the study and whose clinical characteristics did not diverge from the whole group (Table 1), carotid IMT was measured by high-resolution ultrasonography (MyLab25; Esaote, Florence, Italy) using a linear 12-MHz transducer. Diastolic images were obtained on the right side at the level of the distal common carotid artery, 1 cm away from its bifurcation. Averages of five IMT measurements on the far wall of the artery were used in the study. All IMT measurements were taken on a separate visit in all children and were performed by the same observer who was blinded to the clinical and laboratory characteristics of the subjects. Intrasubject coefficient of variation for IMT measurements was <6%.

Table 1 Clinical and laboratory variables in subjects with carotid IMT measurements (n = 91)

Laboratory variables.

All serum samples were obtained between 0800 and 0900 h under fasting conditions. Serum glucose was analyzed by the hexokinase method. Serum immunoreactive insulin was measured by immunochemiluminiscence (IMMULITE 2000; Diagnostic Products, Los Angeles, CA). Lower detection limit was 0.4 mIU/L and intra- and interassay CVs were <10%. Insulin resistance was calculated using the homeostasis model assessment for insulin resistance [HOMA-IR = (fasting insulin in mU/L) × (fasting glucose in mM)/22.5]. Total serum triglycerides were measured by monitoring the reaction of glycerol-phosphate-oxidase and peroxidase. HDL (C) was quantified by means of the homogenous-accelerator selective detergent method. Cortisol was measured by a fluorescence polarization immunoassay (FPIA; AxSYM, ABBOTT, Abbott Park, IL). Lower detection limit was 1.1 μg/dL, and intra- and interassay CVs were <10%.

Statistics.

Statistical analyses were performed using SPSS version 12.0 (SPSS Inc, Chicago, IL). Results are expressed as mean ± SEM. Nonparametric variables were mathematically transformed to improve symmetry. The relation between variables was analyzed by simple correlation followed by multiple regression analysis in a stepwise manner. Significance level was set at p < 0.05. The study had an 80% power to detect a significant association between serum cortisol and IMT with a Pearson correlation coefficient of 0.26 in the studied subjects.

RESULTS

Clinical and laboratory characteristics of the studied subjects are summarized in Table 2. Serum cortisol was related to BMI SDS (r = −0.23, p < 0.001), both SBP (r = 0.21, p < 0.005; Fig. 1) and DBP (r = 0.14, p < 0.05), HR (r = 0.14, p < 0.05), and carotid IMT (r = 0.38, p < 0.001; Fig. 1) in the studied subjects. No gender or BMI interactions were apparent in the associations with SBP and carotid IMT. Serum cortisol was unrelated to age, body fat, waist, insulin resistance, or serum lipids.

Table 2 Clinical and laboratory variables in the studied subjects
Figure 1
figure 1

Correlation graphs of morning serum cortisol with both (A) SBP (n = 223, r = 0.189, p < 0.005) and (B) carotid IMT (n = 91, r = 0.361, p < 0.0005) in healthy prepubertal children. •: boys; : girls. The r and p values (Pearson analyses) are for all children as a group.

In multiple regression analyses (Table 3), serum cortisol together with HOMA-IR, HR, and waist contributed to 23% of SBP variance, and serum cortisol and BMI contributed to 19% of carotid IMT variance in the studied subjects. The independent associations with SBP and carotid IMT were evident in both genders.

Table 3 Multivariate linear models of carotid IMT, SBP, and DBP as dependent variables

DISCUSSION

Physiological concentrations of serum cortisol were related to both SBP and carotid IMT in prepubertal boys and girls. Although the association between serum cortisol and visceral fat is well accepted (1720), the impact of total body adiposity on serum cortisol is less clear, with some reports showing positive (5,21) and others negative associations between obesity measures and serum, salivary, or urine cortisol (3,22). These differences may be explained, in part, by the selection criteria and clinical characteristics of the studied subjects (i.e. obese versus nonobese subjects), because increased body weight is known to be associated with not only increased cortisol production but also increased cortisol clearance rate (23). In our study, higher BMI was independently associated with lower serum cortisol concentrations in unselected healthy prepubertal children.

A number of studies have reported on the association between cortisol and BP (3,4,8,24,25). However, little is known about the effects of physiological variations in serum cortisol concentrations on BP in children. Our findings show that higher cortisol concentrations within the normal range are associated with higher SBP. These associations were independent from age, BMI, body fat, waist, insulin resistance, serum lipids, and HR. Although a possible role of a stress response mediating these associations cannot be excluded, the fact that they persisted after adjusting for HR points to additional mechanisms in the relationship between serum cortisol and vascular risk markers. It has been proposed that prenatal growth restriction may program the adrenal axis for an exaggerated stress response, and we cannot exclude the possibility that the observations herein described are related to prenatal programming of the adrenal axis (3). In support of this hypothesis, a recent report from a birth cohort study has shown that cord-blood cortisol predicts a higher BP at age of 3 y (26).

Besides BP, carotid IMT is a well-recognized vascular risk marker in both adults (27) and children (28). To the best of our knowledge, ours is the first study to show an independent association between serum cortisol and IMT in healthy children, thereby suggesting that serum cortisol is not only associated with functional but also with anatomical changes in the vasculature.

No gender interactions were apparent in these associations, and therefore, our data are consistent with studies showing that the sexual dimorphism in cortisol secretion starts after the age of 10 y in healthy children (29). However, the vascular risk markers studied were differentially associated with obesity measures and insulin resistance in our study. Differences in the pathogenesis of vascular dysfunction may therefore exist in males and females (30).

The mechanisms by which higher cortisol can contribute to cardiovascular disease may include both systemic and local effects of the hormone (31). It is well recognized that hypercortisolemia predisposes to a dysmetabolic phenotype by impairing liver, adipose tissue, pancreas, and muscle function (32). In addition, glucocorticoids have direct effects on vascular smooth muscle cells and indirect effects promoting an increase in vascular tone in response to various vasoactive hormones (3335). Finally, increased local cortisol levels in the blood vessels may promote perivascular inflammation (31,34).

Serum cortisol changes rapidly after awakening and shows a clear diurnal variation (36). In our study, all children were assessed beyond 30 min of awakening, and blood was sampled in a short time frame in the morning in all subjects.

In conclusion, physiological concentrations of serum cortisol are associated with vascular risk markers as early as in childhood. Because childhood risk factors predict adult risk for cardiovascular disease, these observations may have implications in the prevention of cardiovascular disease early in life.