The 24-Hour Average Concentration of Cortisol is Elevated in Obese African-American Youth with Type 2 Diabetes

Abstract

Introduction:

24-hour average (IC) plasma concentrations of cortisol and growth hormone are lower in obese youth and adults without Type 2 diabetes (T2D) compared to lean subjects. Here we examined IC-cortisol and IC-growth hormone levels in obese youth with and without T2D.

Methods:

We pooled ½-hourly samples from 20–24-hour sampling to create an IC for cortisol, cortisone, C-peptide, insulin, growth hormone and cortisol-binding-globulin in obese African-American youth with (n=8) and without T2D (N=9). Analytes were assayed by standard methods.

Results.

The groups were similar in age and sex, all participants had BMI% ≥94. T2D patients had slightly lower BMI z-score (2.25±0.36 versus 2.58 ± 0.16, p=0.0429). IC-cortisol (5.70 ±1.8 μg/dl vs 4.18± 1.07 μg/dl, p=0.0481) was higher and IC-C-peptide (2.33± 0.89 ng/ml vs 4.36±1.12 ng/ml, p=0.001) lower in T2D. There were no differences in cortisone/cortisol or for other analytes between groups. IC-cortisol was correlated with IC-cortisone (r=0.46, p=0.0471) but not with ICs of insulin, C-peptide, cortisol-binding-globulin, or growth hormone.

Conclusions:

IC-cortisol levels are higher and IC-C-peptide lower in obese African-American youth with T2D. Higher levels of IC-cortisol in obese youth with T2D may indicate a change in hypothalamic-pituitary-adrenal regulation which may exacerbate hyperglycemia and other metabolic complications of obesity.

INTRODUCTION:

The endocrine system can often adjust in response to metabolic perturbations associated with obesity in order to preserve metabolic homeostasis. Changes in circulating insulin level with increasing body mass are a well-known example. Increased body fat leads to insulin resistance, higher circulating insulin levels are needed in obese individuals without diabetes compared to lean individuals in order to overcome insulin resistance and maintain euglycemia (1, 2). Obese individuals unable to sufficiently increase their circulating insulin levels to cope with insulin resistance develop dysglycemia and diabetes.

A number of hormones: glucagon, cortisol, growth hormone, epinephrine, counteract insulin action and act to raise blood glucose levels (3). These “insulin counter-regulatory” hormones increase insulin resistance. We previously demonstrated that the 24-hour average plasma concentrations (IC) of two counter-regulatory hormones, growth hormone (GH) and cortisol (F) were markedly lower in obese youth compared to lean children (4, 5). Many of the normal stature but obese youth had IC-GH and IC-F levels which were in the range typical of short, poorly-growing children with hypopituitarism (6). IC-cortisol has also been found to be lower in some populations of obese adults without diabetes (7–9) compared to lean individuals. Potentially, lower circulating levels of cortisol and growth hormone would be less antagonistic to insulin, thus reducing insulin resistance which might help to prevent dysglycemia (10, 11). Reduced circulating concentrations of GH and cortisol in obese children without diabetes is potentially another adaptive hormonal response which may complement the increased insulin secretion in order to help preserve physiologic levels of glucose.

In the progression from simple obesity to type 2 diabetes, the patient loses the ability to overcome insulin resistance with further increases in circulating insulin. The role of IC-cortisol and GH in obese youth with diabetes is unclear. As cortisol and GH both have pulsatile secretion and diurnal variation, measuring the average or integrated concentrations (IC) of hormones over the course of a day has been recommended as the most accurate estimate of tissue exposure (12–15). Besides secretory changes in cortisol, its concentration may be influenced by the activity of a pair of 11βOH hydroxysteroid dehydrogenase (HSD) enzymes (16). Type 2 11βOH-HSD inactivates cortisol by converting it to cortisone, while type 1 11βOH-HSD regenerates cortisol from cortisone. The relationship between circulating IC-cortisol and IC-cortisone might indicate the balance of inter-conversion of cortisol and cortisone by the two 11βOH HSD isoforms in patients with pediatric diabetes (17). To determine if changes in cortisol and GH level are associated with pediatric T2D we measured IC hormone levels in obese youth with and without T2D.

METHODS:

Patients with diabetes were recruited for the study from the diabetes clinics at the Children’s Hospital of New Orleans. Obese patients without diabetes were recruited from individuals participating in a life style management program to control obesity through the Children’s Hospital of New Orleans and had had no prior reported symptoms or laboratory evidence of hyperglycemia. All participants self-identified as AA. The study exclusively recruited self-identified AA participants, as pediatric T2D is predominantly a disease of AA youth who have obesity in our patient population in the greater New Orleans area.

A participant was diagnosed as having T2D after the occurrence of typical diabetes symptoms, with persistent fasting (≥126 mg/dL) and random (≥200 mg/dL) hyperglycemia, BMI≥90% for age and sex, before glucose lowering therapy. All T2D patients were negative for pancreatic autoantibodies from a panel of GAD65, IA-2, insulin autoantibody and islet cell antibody. Duration of diabetes at the time of the study was 6 months or greater since initial diagnosis. During that time patients with T2D continued to require medication to control hyperglycemia. Patients receiving drugs other than metformin or basal insulin at the time of participation in the study were excluded. This study was reviewed and approved by the IRBs of the Louisiana State University Health Sciences Center, New Orleans, Louisiana and the Children’s Hospital of New Orleans, Louisiana. Informed consent and assent for participation was obtained from all participants.

In preparation for the study, patients on metformin did not take this medication the night prior and during the 24 hours stay for hormone sampling. Patients were allowed to take their basal insulin during the study. All 24-hour studies were performed in a specialized two chambered unit at the Children’s Hospital of New Orleans. One chamber contained equipment for processing and storage of blood samples and the adjoining larger room was the participant living space with bed, chairs, TV, bathroom and sink.

Patients had indwelling catheters placed and blood samples for the study were not obtained until ½ hour after placement of the catheter had elapsed. Samples were then drawn at half-hourly for 24 hours. If indwelling catheter required restart the next study sample was not obtained until 30 minutes post insertion. A study was not considered complete without a total of 40 or more ½-hour samples. Participants were able to make meal choices and pick snacks from the hospital in-patient menu. Participants were allowed walk about within their room, sit up in a chair or be in bed as their choice during the day. Participants wore their own clothes. They were free to use the bathroom as needed. Lights were turned off at 11 pm.

In the adjoining small room, withdrawn blood samples were centrifuged and plasma separated. A 24-hour IC was made by pooling an equal sized aliquot from each half-hourly sample. Hormone assays were performed in commercial reference laboratories. Cortisol was assayed by high-pressure liquid chromatography/tandem mass spectrometry (HPLC/MS-MS). Cortisone was measured by liquid chromatography/tandem mass spectrometry (LC-MS/ MS). CBG, growth hormone, insulin and C-peptide were measured by immunoassay. A single sample for glucose was obtained after overnight fast before the participant’s breakfast.

Statistical Analyses:

Between group (obese subjects with and without T2D) comparisons of the patient characteristic and hormone levels were performed by t-test (2-tailed) if the variable was normally distributed or by Wilcoxon rank sum test (Mann Whitney U test) when not normally distributed. Pearson correlation between variables was assessed using PROC CORR. Comparison of the number of patients with IC-cortisol over 2SD of the mean of a historical reference population was performed using Fisher’s Exact Test. The ratio of insulin to C-peptide was performed after conversion to SI units. Statistics were performed using SAS version 9.4. Aggregate results in tables are reported as mean±1SD. Statistical significance was accepted by conventional at P=0.05 level or less.

RESULTS:

Patient Characteristics:

A total of 17 obese subjects completed full studies, 8 patients with diabetes and 9 patients without diabetes. Please refer to Table 1 for a comparison of participant characteristics between the two groups. The groups were similar in age, height and sex distribution. All subjects had BMI percent ≥ 94%, (BMIz similarly elevated) and would be considered extremely obese (18). Subjects without T2D had somewhat higher BMI metrics than the T2D group. No participants reported hypoglycemic symptoms during the study.

Table 1:

Characteristics of Participants, mean(1SD)

Characteristics	Without T2D (n=9)	With T2D (n=8)	P-Value
Age, years	14.8(2.1)	15.0(3.1)	0.8304
Sex, F/M	7/2	5/3	0.6199
Weight, kg	116.4 (18)	98.4 (14.6)	0.0751
Height, cm	166.3 (12.7)	166.0 (7.7)	0.9514
BMI (kg/m2)	41.9 (3.9)	35.9 (5.6)	0.0189
BMI z-score	2.6 (0.2)	2.3 (0.4)	0.0429
Duration Diabetes (yrs)		3.3(3.5)
HbA1c(%)	-	7.7(2.8)
Pre-Breakfast Glucose (mg/dL)	90(11)	132(63)	0.052

Hormone Levels:

Table 2 contains comparison of the IC hormone and IC-CBG levels measured in this study for each group. IC-F was higher in patients with T2D than individuals without diabetes (p=0.0481). IC-F and IC-GH levels for obese participants in the current study without diabetes were very similar to levels we previously reported using continuous withdrawal technology (4, 5, 19). Figure 1 depicts the range of IC-cortisol for the individual participants. None of the participants without diabetes had an IC-F over the mean IC-F=5.7 μg/dL of the patients with T2D, (or over +2SD of the reported mean of the historic reference group of obese individuals without diabetes (4, 5, 19)). Five of the participants with T2D had IC-Fs over the upper range of the obese youth without diabetes in this study.

Table 2:

Blood metabolites of obese subjects with and without Type 2 diabetes, mean(1SD) and [min, max]

Integrated Concentrations	Without T2D (n=9)	With T2D (n=8)	P-Value
IC-Cortisol (μg/dL)	4.2 (1.1) [2.6, 5.1]	5.7 (1.8) [2.8, 7.7]	0.0481
IC-Cortisone (μg/dL)	0.9 (0.4) [0.3, 1.6]	1.2 (0.3) [0.7, 1.7]	0.1311
IC-CBG (mg/dL)	2.6 (0.5) [2.1, 3.6]	2.7 (0.4) [2.2, 3.2]	0.6486
IC-cortisone/IC-Cortisol ratio	0.2 (0.2) [0.07, 0.62]	0.2 (0.1) [0.14, 0.29]	0.7363
IC-GH (ng/mL)	1.5 (0.8) [0.4, 2.8]	1.8 (0.8) [0.9, 3.5]	0.4606
IC-Insulin (μU/mL)	41.1 (17.1) [13.0, 73.0]	26.4 (25.7) n=6 [0.29, 68.0]	0.2019
IC-C-peptide (ng/mL)	4.7 (1.1) [2.6, 6.5]	2.3 (0.9) [0.6, 3.4]	0.0010
IC-Insulin/IC-C-peptide	0.2(0.1) [0.1, 0.3]	0.2(0.2) n=6 [0.003, 0.4]	0.7940

Figure 1.

IC-Cortisol (IC-F) of study participants with and without diabetes. IC-Cortisol was higher in patients with T2D, p=0.0481. The majority of patients with T2D have IC-cortisol over the range of individuals without diabetes. Horizontal black line is mean for the group.

IC-E, IC-CBG, the ratio of IC-E/IC-F, and IC-GH were not statistically different between the groups. IC-F was correlated with IC-E (r=0.46, p=0.0471) but not with IC-CBG, IC-GH, or IC-CP. IC-CP was higher in patients without diabetes (p=0.001) compared with the T2D group. There were two fewer samples for IC-insulin (IC-I) in the T2D group (n=6), and although mean IC-I was also higher in individuals without T2D this relationship did not achieve statistical significance. The ratio of IC-insulin/IC-C-peptide was similar between the groups.

DISCSUSSION:

To our knowledge this is the first study to document higher IC-F levels and no difference in IC-E/IC-F ratio or IC-GH in obese children with T2D compared to obese youth without diabetes. Participants were from a young obese African-American population which is highly susceptible to develop type 2 diabetes. Although challenging to perform and complete, especially in obese individuals, IC plasma studies are often considered the gold standard for quantitative assessment of hormones with pulsatile secretion, ultradian and circadian variation (12, 13, 20). Plasma collections also have advantages over urine and salivary estimates of cortisol and reflect the circulating hormone levels to which the tissues are exposed (15, 21–23).

We previously found that IC-F was lower in otherwise healthy obese youth compared to healthy lean age matched subjects (4). Many of the obese subjects had IC-F levels in the range of short patients with hypopituitarism (4, 6). We have surmised that lower IC-cortisol in obese children may help reduce insulin resistance and prevent dysglycemia (11.). Higher IC-cortisol in obese youth with diabetes may indicate a loss in the ability of the HPA axis to maintain reduced cortisol levels in the face of obesity and further exacerbate insulin resistance and other metabolic derangements of obesity (10). In our group of patients higher cortisol was combined with impaired ability to secrete insulin/C-peptide in the T2D.

Some investigators have suggested that obesity and sex may influence the balance cortisone and cortisol levels through differential activity of the 11βOH HSD enzyme system (24). We previously found that reduced IC-F in obese children was associated with proportionately reduced IC-cortisone (IC-E) levels (25). Thus, the lower IC-Fs in children with uncomplicated obesity were not associated with relatively higher circulating cortisone levels as might be expected to occur with increased enzymatic inactivation of cortisol to cortisone by 11βOH HSD2(17, 25). In the current study, IC-E levels and the ratio of IC-E/IC-F were not different between the T2D and non-T2D groups and similar to the ratio previously noted in lean youth (25). Therefore, the higher systemic circulating IC-F of patients with T2D does not appear to be from altered differential conversion of cortisone and cortisol by 11βOH HSD enzymes.

Similar findings to ours of lower cortisol levels in adult obese individuals have been reported from 24-hour episodic sampling studies in some populations (7–9) but not all (7, 26). Potentially age, sex, ethnicity and other factors may all play a role in IC-cortisol levels in adult patients (7, 26). Age appears to play a role in explaining some differences among obese adults, as we previously found in a cross-sectional study that IC-F levels increase in obese patients from the lower levels in youth to levels similar to lean individuals in adulthood (19).

In addition to IC-F, we also previously reported that another counter-regulatory hormone IC-GH was lower in obese youth without diabetes while IC-insulin was higher compared to lean individuals (5). The combination of lower IC-F and IC-GH along with increased IC-insulin are potentially adaptive changes that would help maintain glucose tolerance in the face of insulin resistance due to obesity (11). In this follow up study mean IC-GH was similarly reduced in both obese groups compared to the higher levels found in lean individuals (14). Lower C-peptide in our T2D patients indicates inadequate endogenous insulin secretion needed to maintain normal glucose tolerance. The observed combination of lower insulin/C-peptide secretion and higher antagonistic cortisol levels would put obese T2D patients at additional risk for abnormal glucose tolerance and potentially other metabolic changes. However, whether increased IC-F and/or associated lower insulin levels are a contributing cause or consequence of T2D is unclear (27).

Strengths of this study were the use of a comprehensive method to measure average plasma 20–24-hour hormone concentrations in obese young subjects with and without diabetes. Participants were drawn from an African-American population which is extremely vulnerable to obesity, T2D and other metabolic derangement related to diabetes.

Some limitations of the study should be pointed out. Maintaining IV access for frequent sampling in very obese patients throughout a 20- 24-hour period is challenging and expensive. The number of participants who were able to complete a full 20-24 sampling study in each group was small. This study included only African-American participants and its generalizability to obese youth with T2D of other ethnicities and ages will need to be evaluated. Metformin was held one day prior to the collection studies but it is not clear whether all its influence would be entirely abated. Metformin can influence 11βHSD activity however there was no evidence of such an effect as the cortisol/cortisone ratio was similar between groups. Basal insulin was not withheld for T2D patients and potentially exogenous insulin may have influenced measured levels of insulin and C-peptide levels. However, the ratio of IC-insulin to IC-C-peptide was similar between the groups suggesting that exogenous insulin may not have had a marked influence. Although we speculate that higher IC-F youth with T2D may exacerbate insulin resistance, metrics of insulin sensitivity were not performed in this study.

In conclusion, we found that IC-cortisol levels in obese youth with T2D were higher than obese youth without diabetes. Prior studies indicate that obese adults and youth without diabetes have lower IC-cortisol than lean individuals. Higher IC-F in T2D might indicate a loss of the ability of the HPA axis to lower cortisol levels in obese youth in the face of hyperglycemia. The loss of this potential metabolic adaptation might exacerbate insulin resistance in obese individuals with T2D. A longitudinal study would be required to deduce the relationship of cortisol level, insulin secretion and resistance over time in obese youth predisposed to developing diabetes.

Highlights.

• Pathogenic factors that trigger Type 2 Diabetes (T2D) and metabolic complications are poorly understood.

• 24-hour-average plasma cortisol is higher in obese youth with T2D and may play a role in its pathogenesis and other metabolic complications.

• Despite higher cortisol, the cortisol/cortisone ratio was similar in obese youth with and without T2D.

ABBREVIATIONS USED:

AA

African-American

BMI

Body Mass Index

CBG

Cortisol- Binding Globulin

CP

C-peptide

E

Cortisone

F

cortisol

GH

growth hormone

I

Insulin

IC

Average or Integrated Concentration

HSD

hydroxysteroid dehydrogenase

T2D

Type 2 Diabetes