| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Psychiatry (U.S., W.G., S.R., M.P., T.M., S.W., J.H., E.F., K.G.K., K.M.O., F.H.) and Internal Medicine (A.P.), Luebeck University Medical School, Luebeck, Germany.
Address correspondence and reprint requests to Ulrich Schweiger, Department of Psychiatry, University of Luebeck, Ratzeburger Allee 160, D-23538 Lübeck, Germany. E-mail: ulrich.schweiger{at}psychiatrie.uk-sh.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: We studied 19 patients with typical major depressive disorder (MDD), 7 patients with atypical major depression, and 30 men and women of a healthy comparator group using a stepwise glucose clamp procedure. Glucose disposal rates were assessed and concentrations of hormones involved in glucose allocation were measured.
Results: Glucose disposal rates were lower by 19% in patients with typical MDD and 30% in patients with atypical MDD than in the group of healthy controls (3.2 ± 0.8 and 2.8 ± 0.7 versus 4.0 ± 1.0 mmol h–1 kg–1). C-peptide concentrations were 26% higher in patients with atypical MDD and similar in patients with typical MDD and healthy controls. Vascular endothelial growth factor concentrations were 30% higher in typical MDD and similar in atypical MDD and the control group.
Conclusions: Whole-body glucose disposal is reduced in patients with typical and atypical depression. The observed neuroendocrine responses suggest a hyperactive allocation system in typical depression and a hypoactive allocation system in atypical depression.
Key Words: depressive disorder • insulin • glucose clamp technique • cortisol • norepinephrine • vascular endothelial growth factor
Abbreviations: ACTH = adrenocorticotropic hormone; ANOVA = analysis of variance; BMI = body mass index; CV = coefficient of variation; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, version IV; GH = growth hormone; GLUT1 = glucose transporter Type 1; GLUT4 = glucose transporter Type 4; HDL = high-density lipoproteins; HPA = hypothalamic-pituitary-adrenal; MDD = major depressive disorder; SCID = Structured Clinical Interview for DSM-IV; VEGF = vascular endothelial growth factor.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Glucose disposal is an aspect of glucose metabolism in depressive disorders that remains to be studied. Glucose disposal is controlled by the brain allocation system (9) that primarily uses the sympathetic nervous system and the hypothalamic-pituitary-adrenal (HPA) system as effectors. Whole-body glucose disposal is essentially effected by the sum of glucose transport by two glucose transporters: GLUT 1, located primarily at the blood-brain barrier, and the insulin-sensitive GLUT 4, located on the surface of muscle and fat cells. Because glucose transport into the brain via GLUT 1 transporters is independent of insulin, the allocation system can maintain glucose supply to the brain in situations of stress or decreased food intake through alterations in sympathetic outflow. Through stimulation of adrenergic inhibition of pancreatic insulin secretion, and thus glucose transport through the GLUT 4 transporter (10,11), the allocation system can restrain glucose uptake into the muscle and fat cells, providing increased glucose for the brain.
The glucose clamp technique is considered to be the "gold standard" for the assessment of whole-body glucose disposal (9,10). The glucose disposal rate (GDR) assessed in this way is a strong correlate for whole-body glycolysis (11). Thresholds for counterregulatory hormone secretion have been studied using stepwise glucose clamp procedures (12). To our knowledge, the glucose clamp technique has not yet been applied to patients with depressive disorders.
When assessing glucose disposal in MDDs, it has to be considered that these patients are heterogeneous with respect to the direction of alterations in their eating behavior (13). In typical or melancholic depression, appetite is decreased and weight loss occurs. In atypical depression, appetite and sleep are increased so that body weight increases. Findings on the direction of weight change are reproducible across depressive episodes (14). These observations suggest a differential regulation of the allocation system in typical and atypical depressive disorders. It has been proposed that, in typical depression, the HPA system and the sympathetic nervous system are stimulated as their activity is attenuated in atypical depression (15).
At rest, approximately 50% of the whole-body glucose consumption occurs in the brain (16). Invasive experiments have shown that the brain's energy needs increase with psychological challenge and decrease during sleep (17,18). However, studies using the glucose clamp technique to assess insulin sensitivity have implicitly assumed that the glucose need of the brain is negligible, time-constant, or at least very similar among the tested subjects. Brain glucose utilization is redistributed in subjects with depressive disorders. In particular, such patients show a low glucose metabolism in the frontal parts of the brain (19). This may correspond to a global reduction of cerebral metabolic rates for glucose observed in mood disorders (20). Thus, the assumptions that are commonly made to interpret glucose clamp experiments do not necessarily hold true for patients with psychiatric disorders.
Our primary aim in this study was to assess the whole-body glucose disposal in patients with both typical and atypical depression. The secondary goal was to characterize the neuroendocrine responses during a hyper-, eu-, hypoglycemic stepwise clamp experiment in patients with both subtypes of major depression.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
17 years. Five of the 19 patients with MDD had received treatment with selective serotonin reuptake inhibitors before the study, but no other psychotropic medication. Fourteen healthy women and 16 men served as the comparison group. A standardized psychiatric interview gave no evidence of either an individual history of major psychiatric disorders or family history in their first-degree relatives in any subject of this group. None of the comparison subjects suffered from an acute or chronic illness or took any medication. The study was approved by the local ethics committee and all subjects gave their written informed consent.
|
Stepwise Glucose Clamp Procedures
We used a stepwise glucose clamp with hyperglycemic, euglycemic, and hypoglycemic plateaus. Participants were asked to rest and abstain from food and drink starting at 11 PM on the day before the study. Experiments took place in a sound-attenuated room with the subjects sitting with their trunk in a semireclined position (approximately 60°) and their legs in a horizontal position on a bed. A cannula was inserted into a vein on the back of the hand, which was placed into a heated box (about 50°C) to obtain arterialized blood. A second cannula was inserted into an antecubital vein of the contralateral arm. Blood sampling and control of the dextrose infusion took place in the adjacent room. Subjects were given no information about blood glucose concentrations. After a 30-minute baseline period starting at 8 AM, insulin (Insuman Rapid, Aventis Pharma Deutschland GmbH, Germany) was infused at a constant rate of 1.5 mU/min/kg. The 20% dextrose solution was applied at variable rates to adjust plasma glucose plateaus. During the hyperglycemic period, plasma glucose concentrations were held stable at 8.88 mmol/l. Subsequently, plasma glucose concentrations were lowered in a stepwise manner to achieve respective plateaus of 6.83, 4.83, and 2.78 mmol/l. Each plateau was maintained for a 30-minute period, and the next lower plateau was induced gradually over 30 minutes. Plasma glucose concentrations were measured every 5 minutes using a blood glucose analyzer (Hemocue, Grossostheim, Germany). After 240 minutes, insulin infusion was stopped and plasma glucose concentrations were returned to normal by dextrose infusion and food intake.
Blood samples were collected every 15 minutes for determination of serum insulin, c-peptide, cortisol, growth hormone (GH), and vascular endothelial growth factor (VEGF) and every 30 minutes for determination of adrenocorticotropic hormone (ACTH) and norepinephrine. GDR was calculated as average glucose infusion rate during previous 5 minutes divided by body weight (mg/min per kg).
Assays
All blood samples were immediately centrifuged and the supernatants stored at –80°C until analysis. An immunometric assay (IMMULITE 1000, DPC, Nauheim, Germany) was used for determination of serum insulin (inter-assay coefficient of variation (CV) <8.0%; intra-assay CV <6.4%), c-peptide (inter-assay CV <14.4%, intra-assay <10.3%), ACTH (inter-assay CV <9.4%, intra-assay CV <9.6%), cortisol (inter-assay CV <10.0%, intra-assay CV <8.8%), and GH (inter-assay CV <6.2%, intra-assay CV <6.5%). VEGF was determined using ELISA kits (Quantikine, R&D Systems, Wiesbaden, Germany) with inter-assay CV <8.8% and intra-assay CV <6.7%. High-performance liquid chromatography was used to determine plasma norepinephrine (inter-assay CV <6.1%, intra-assay CV <5.8%).
Data Analysis
Study groups (typical MDD versus atypical MDD versus control group) were compared using repeated-measures analysis of variance (ANOVA). One-way ANOVA and post hoc pairwise analysis (least significant difference) was performed for each glycemic plateau and selected single values for exploratory analysis. Data are presented as mean ± standard deviation. A two-sided p < .05 was considered significant. Data were analyzed using Statistical Package for the Social Sciences (SPSS, version 13.0; SPSS Inc., Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
|
Glucose concentrations averaged over the whole period of the stepwise glucose clamp were similar in the three groups (5.9 ± 2.2, 6.0 ± 2.2, and 6.0 ± 2.2; df = 2, F = 1.2, p = .29). During the hyperglycemic and the two euglycemic plateaus, the concentrations were equally similar. Only during the hypoglycemic plateau, the glucose concentrations were slightly lower in the healthy control group than in the two MDD groups (2.7 ± 0.2 versus 2.9 ± 0.4 and 2.9 ± 0.3; df = 2, F = 3.6, p = .034).
Mean c-peptide concentrations (Figure 2.) tended to be higher among patients with atypical MDD than in patients with typical MDD or healthy controls (4.7 ± 1.2 versus 3.7 ± 1.2 and 3.5 ± 1.4 ng/ml; group factor df = 2, F = 2.99, p = .059, ANOVA for repeated-measures). Higher concentrations of c-peptide were observed in the group with atypical MDD both before and during the first hyperglycemic plateau (6.0 ± 1.2 versus 4.9 ± 1.5 and 4.7 ± 1.7 ng/ml; one-way ANOVA: df = 2, F = 2.0, p = .15; post hoc analysis: atypical MDD versus control group, p = .053) as well as during the hypoglycemic plateau (3.2 ± 2.2 versus 2.2 ± 1.5 and 1.8 ± 0.9 ng/ml; one-way ANOVA: df = 2, F = 3.3, p = .043; post hoc analysis: atypical MDD versus control group p = .013). During the euglycemic plateaus, concentrations of c-peptide were similar in all groups.
|
During the clamp procedure, insulin was infused at a constant rate of 1.5 mU/ min/kg in all groups. Measured concentrations therefore represent both infused insulin and endogenous secretion. Patients with atypical MDD had 29% higher insulin concentrations than patients with typical MDD or healthy controls (184 ± 76 versus 142 ± 61 and 142 ± 70 µU/ml; group factor: df = 2, F = 2.7, p = .078; ANOVA for repeated-measurements). Higher concentrations of insulin in the group with atypical MDD were observed both during the second euglycemic plateau (234 ± 76 versus 168 ± 67 and 182 ± 52 µU/ml; one-way ANOVA: df = 2, F = 3.1, p = .053; post hoc analysis: atypical MDD versus control group, p = .044; atypical MDD versus typical MDD, p = .016) and during the hypoglycemic plateau (224 ± 81 versus 164 ± 52 and 144 ± 50 µU/ml; one-way ANOVA: df = 2, F = 9.9, p = .005; post hoc analysis: atypical MDD versus control group p = .001; atypical MDD versus typical MDD, p = .017). During the other plateaus, concentrations of insulin were similar in all groups.
ACTH, Cortisol, and Norepinephrine
Mean ACTH levels were similar in the three groups (typical MDD, 15.4 ± 2.4 pg/ml; atypical MDD, 17.6 ± 2.8 pg/ml; control group, 18.7 ± 4.0 pg/ml). Higher ACTH concentrations were observed during the first euglycemic plateau in patients with atypical MDD than in patients with typical MDD and healthy controls (20.7 ± 10.9 versus 12.6 ± 5.4 pg/ml and 15.2 ± 4.5 pg/ml; one-way ANOVA: df = 2, F = 4.9, p = .011; post hoc analysis: atypical MDD versus control group, p=.031; atypical MDD versus typical MDD, p = .003). At the end of the hypoglycemic plateau, ACTH concentrations increased in the healthy control group but not in the MDD groups (27.8 ± 21.5 versus 17.3 ± 9.2 pg/ml and 17.6 ± 3.7 pg/ml; one-way ANOVA: df = 2, F = 2.7, p = .078; post hoc analysis: typical MDD versus control group, p = .038). Mean cortisol and norepinephrine concentrations were similar in all groups. Patients with atypical MDD tended toward lower cortisol concentrations at baseline than patients with typical MDD and the control group (10.5 ± .3 versus 14.3 ± 5.8 µg/dl and 13.0 ± 3.6 µg/dl; one-way ANOVA: df = 2, F = 2.0, p = .15; post hoc analysis: typical MDD versus atypical MDD, p = .055). Similarly, patients with atypical MDD showed a trend toward lower norepinephrine concentrations at baseline than patients with typical MDD or control individuals (145 ± 43 versus 245 ± 126 pg/ml and 203 ± 87 pg/ml; one-way ANOVA: df = 2, F = 2.8, p = .072; post hoc analysis: typical MDD versus atypical MDD, p = .026).
VEGF and Growth Hormone
VEGF concentrations (Figure 3) were approximately 30% higher in typical MDD than in atypical MDD or control individuals (67.1 ± 9.0 versus 42.3 ± 5.7 pg/ml and 43.2 ± 2.5 pg/ml; group factor df = 2, F = 5.8, p = .006; ANOVA for repeated-measurements). One-way ANOVA revealed significant differences between the groups also at all plateaus.
|
GH concentrations showed a significant interaction between effects of time and group (Wilks Lamb = 0.297; df = 2,36, F = 1.67; p = .033). Higher GH concentrations were observed during the first euglycemic plateau in patients with typical MDD than in patients with atypical MDD or healthy individuals (1.3 ± 1.9 versus 1.1 ± 1.3 and 0.5 ± 0.9 pg/ml; one-way ANOVA: df = 2, F = 2.1, p = .131; post hoc analysis: typical MDD versus control group p = .050).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
In patients with both typical and atypical depression, we found that whole-body glucose disposal was decreased compared with healthy subjects. Such a low glucose disposal may result from reduced brain glucose uptake or impaired glucose uptake into muscle and adipose tissue or a combination of both. An impairment of cerebral glucose metabolism has been hypothesized to constitute a common denominator of neuroendocrine and brain alterations in depressive disorders (21). Here we emphasized that measuring disposal rates using the glucose clamp technique does not allow us to distinguish whether energy fluxes are decreased to the brain or the periphery.
Low glucose disposal as measured by euglycemic clamp has been established as a risk factor for coronary heart disease and heart failure (22,23). It is particularly noteworthy that low GDRs in the group with typical depression occurred in the absence of overweight or any other signs of metabolic syndrome, such as high blood pressure or dyslipidemia, and in the absence of significant alterations in lifestyle concerning exercise, alcohol consumption, and smoking. In contrast, the group with atypical depression had increased body weight and waist girth and reported lower physical activity. GDR values are corrected for body weight, but in the absence of an obese comparison group, we cannot exclude that the results may be confounded by the differences in body composition. The findings at least in the group with typical depression suggest that low glucose disposal may be an early step in the interaction between depression and the metabolic syndrome (24), and may be a risk factor contributing to the increased mortality and morbidity seen in depression (25).
By measuring neuroendocrine responses during the stepwise glucose clamp procedure, information might be gained about the directions of energy fluxes within the organism. In the patients with atypical depression, the sympathoadrenal responses (norepinephrine, ACTH, cortisol) were partially lower and serum c-peptide and insulin concentrations were higher than they were in healthy controls. Norepinephrine concentrations were found to be lower in atypical depression during the baseline part of the clamp. Cortisol concentrations also tended to be lower during baseline and at high glucose concentrations, whereas ACTH concentrations tended to be lower during the hypoglycemic period of the clamp procedure. The latter finding has to be interpreted with caution due to the slightly higher glucose concentrations during the hypoglycemic plateau in both depressed groups. We tentatively argue that the blunted sympathoadrenal responses may indicate a weak reaction of the allocation system in atypical depression. In line with these findings, insulin and c-peptide concentrations were found to be increased in atypical MDD throughout the clamp procedures. Hyperinsulinemia also indicates a brain allocation failure in the atypical subgroup (26). Previously, hyperinsulinemia has been confirmed as a risk factor for long-term weight gain, where this effect is manifested particularly in insulin-sensitive subjects (27–29). Thus, the hyperinsulinemia observed here in patients with atypical depression mirrors the fact that body mass has already increased excessively in such subjects.
In patients with typical major depression, a marked elevation of VEGF was observed. Among other hormones, VEGF concentrations increase in response to neuroglucopenia (30). VEGF is known to regulate the permeability of the blood-brain barrier to glucose. The VEGF increase in the subgroup with typical depression may be regarded as a neuroprotective mechanism that enhances the cerebral glucose supply (31).
The data on sympathoadrenal responses in the subgroup with atypical depression point toward a weaker activity of the allocation system. We suggest that, in this special context, overeating may be a behavioral tool to compensate for weak allocation and to maintain a sufficient supply of glucose to the brain. In contrast, the higher activity of the allocation system and the higher permeability of the blood-brain barrier achieved by the action of VEGF in the subgroup with typical depression may offset the low appetite and calorific intake seen in these patients. This implies that the alterations of eating behavior in depression may be secondary to a disturbance of brain glucose metabolism and glucose allocation.
This study has several limitations. The assessment of glucoregulatory hormones within the context of a glucose clamping study must always take into account the presence of insulin and glucose infusion during this diagnostic procedure and cannot be compared with naturalistic states. We tried to minimize the confounding effects of comorbidity, treatment, and pretreatment by excluding subjects with disorders as well as those undergoing treatments with agents, such as tricyclic antidepressants that are known to dampen the allocation system. We decided not to exclude patients taking selective serotonin reuptake inhibitors, a substance group that has been shown to improve insulin sensitivity mildly (32,33). Although this might have introduced a Type II error, this group was not different with regard to any parameter estimated in our study.
In summary, we have been able to show that whole-body glucose disposal is reduced in patients with typical and atypical depression. The observed neuroendocrine responses suggest a hyperactive allocation system in typical depression and a hypoactive allocation system in atypical depression. The findings imply that low glucose disposal may be an early indicator of metabolic diseases that develop with depressive disorders.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Received for publication May 21, 2007; revision received August 2, 2007.
DOI:10.1097/PSY.0b013e318164231d
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Disturbances of Glucose Metabolism in Depression Journal Watch Psychiatry, April 7, 2008; 2008(407): 3 - 3. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
