SEARCH

SEARCH BY CITATION

Keywords:

  • glucose use;
  • glucose production;
  • lipolysis;
  • glucose tolerance;
  • glucocorticoids

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Objective: Insulin resistance is observed in individuals with normal glucose tolerance. This indicates that increased insulin secretion can compensate for insulin resistance and that additional defects are involved in impaired glucose tolerance or type 2 diabetes. The objective of this study was to evaluate a procedure aimed at assessing the compensatory mechanisms to insulin resistance.

Research Methods and Procedures: Eight healthy nonobese female patients were studied on two occasions, before and after administration of 2 mg/d dexamethasone for 2 days during a two-step hyperglycemic clamp. Insulin secretion was assessed from plasma insulin concentrations. Insulin sensitivity was assessed from the ratio of whole-body glucose use (6, 6 2H2 glucose) to plasma insulin concentrations. This procedure is known to induce a reversible impairment of glucose tolerance and insulin resistance.

Results: In all subjects, dexamethasone induced a decrease in insulin sensitivity and a proportionate increase in first-phase insulin secretion and in insulin concentrations at both steps of glycemia. The resulting hyperinsulinemia allowed the restoration of normal whole-body glucose uptake and the suppression of plasma free fatty acids and triglycerides. In contrast, the suppression of endogenous glucose production was impaired after dexamethasone (p < 0.01).

Discussion: Increased insulin secretion fully compensates dexamethasone-induced insulin resistance in skeletal muscle and adipose tissue but not in the liver. This suggests that failure to overcome hepatic insulin resistance can impair glucose tolerance. The compensatory insulin secretion in response to insulin resistance can be assessed by means of a hyperglycemic clamp after a dexamethasone challenge.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Hypoinsulinemia and extrahepatic and hepatic insulin resistance can all contribute to the development of hyperglycemia (1). A low insulin secretion and decreased insulin action on extrahepatic, insulin-sensitive tissues will increase blood glucose concentration by decreasing whole-body glucose use, especially after carbohydrate loading (2). An increase in endogenous glucose production secondary to hepatic insulin resistance, in contrast, contributes to increased glycemia through an elevated fasting and postprandial endogenous glucose production (3, 4). Although it is generally accepted that the liver is responsible for the essential portion of endogenous glucose production, several recent observations indicate that the kidney (5), and possibly the gut (6), might also be involved. However, how these factors interact to eventually trigger the development of impaired glucose tolerance or type 2 diabetes remains under debate.

Insulin resistance, as defined by a low insulin-mediated glucose disposal, is highly prevalent in Westernized societies (7, 8). It is associated with obesity, low physical activity, and genetic factors. Many obese and lean insulin-resistant individuals, however, retain a normal glucose tolerance. Maintenance of normal blood glucose concentrations in the presence of reduced insulin sensitivity can be attained through an increased insulin secretion (8, 9), thus restoring a normal glucose disposition index (the latter being defined as the product of insulin sensitivity and insulin secretion indexes). According to this scheme, plotting insulin secretion vs. insulin sensitivity indexes in a group of individuals with normal glucose tolerance results in a hyperbolic curve (9). Also according to this scheme, impaired glucose tolerance may be the result of an inadequate compensation of insulin secretion in response to insulin resistance. This model, however, takes into consideration only whole-body insulin-mediated glucose disposal (which corresponds essentially to insulin-mediated glucose metabolism in skeletal muscle) (10) and insulin secretion, with no reference to hepatic glucose metabolism. Given the role of the liver in glucose homeostasis (11, 12, 13), it can be hypothesized that glucose intolerance will also develop in conditions where increased insulin secretion fails to adequately inhibit endogenous glucose production.

The purpose of our study was to evaluate a procedure aimed at assessing the compensatory responses that occur in response to insulin resistance. We, therefore, studied a group of healthy individuals both before and after a 2-day administration of dexamethasone. It has been widely documented that short-term administration of dexamethasone leads to reversible insulin resistance, hyperinsulinemia, and impaired glucose tolerance in healthy individuals (14, 15). It has even been proposed that a low increase in insulin secretion after a dexamethasone challenge may allow identification of individuals at risk of subsequently developing type 2 diabetes (16). Dexamethasone-induced impaired glucose tolerance is also known to involve both extrahepatic insulin resistance (14) and increased endogenous glucose production (17). In our protocol, glucose-induced insulin secretion, insulin-mediated glucose disposal, and glucose production were measured during a two-step hyperglycemic clamp procedure. This allowed assessment of the extent to which hyperinsulinemia secondary to dexamethasone treatment was effective to counterbalance extrahepatic and hepatic insulin resistance.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Ten healthy female subjects with normal body weight and no family history of type 2 diabetes were recruited to take part in this study. All participants were in good health, had no identified disease, and were not currently taking any medications. They had a mean age of 23.6 ± 0.6 years (range, 21 to 26 years) and a mean BMI of 22.0 ± 0.8 kg/m2 (range, 19.0 to 25.7 kg/m2). The experimental protocol was approved by the ethical committee of Lausanne University Medical School, and all participants provided informed written consent. Each subject was studied with a two-step hyperglycemic clamp on two occasions, separated by 25 to 35 days. Each test was scheduled during the first 7 days of a menstrual cycle. On one occasion, the participants received four doses of 0.5 g dexamethasone at 8:00 am, 12:00 pm, 6:00 pm, and 10 pm for the 2 preceding days. On the other occasion, they did not receive any medication. The tests were performed in an unblinded fashion, and the order with which the dexamethasone and no medication tests were done was randomized.

For each test, subjects reported to the metabolic investigation laboratory at 7:00 am after an overnight fast. One indwelling venous cannula was inserted into an antecubital vein of the left arm for infusion of a 20% dextrose solution labeled with 1.25% 6, 6 2H2 glucose. This glucose infusion was adjusted to maintain a plasma glucose concentration 2.5 mM above basal values for the first hour and then a concentration 5 mM above basal values for the second hour (18, 19). A second indwelling venous cannula was inserted into a wrist vein on the right side and was used for blood collections at 0, 2, 4, 6, 8, 10, 30, 60, 90, and 120 minutes. This hand was maintained in a thermostabilized box heated at 56 °C to achieve arterialization of venous blood. A 60-minute rest period was allowed between the insertion of the venous cannula and the beginning of the glucose infusion. An infusion of 6, 6 2H2 glucose (prime dose, 2 mg/kg; continuous infusion, 20 μg/kg per minute) was administered during this time. Throughout the basal and clamp study period, an indirect calorimetry was performed to calculate net glucose and lipid oxidation rates (20).

Plasma glucose concentration was determined using a Beckman glucose analyzer II (Beckman Instruments, Palo Alto, CA). Plasma insulin and glucagon concentrations were measured by radioimmunoassays (Linco, St. Charles, MO). Plasma free fatty acid concentrations were measured colorimetrically, using a kit from Wako (Freiburg, Germany). Plasma triglyceride concentrations were measured using a kit from Biomérieux (Geneva, Switzerland).

Plasma 6, 6 2H2 glucose was measured by gas chromatography—mass spectrometry (21). Whole-body glucose turnover was calculated by plasma 6, 6 2H2 glucose dilution analysis [hot infusion model (22)].

Several parameters were calculated to assess insulin secretion. The first-phase insulin secretion was evaluated from the peak insulin concentration observed during the first 10 minutes of glucose infusion. The second-phase insulin secretion was evaluated at each step of hyperglycemia from the average of the values of plasma insulin obtained during the second half hour of each plateau of glycemia. Net whole-body glucose and lipid oxidation rates were calculated from respiratory gas exchanges and urinary nitrogen excretion with the equations of Livesey and Elia (23). Insulin sensitivity was evaluated by dividing whole-body glucose turnover by plasma insulin concentrations (19). The slope of the curve linking plasma glucose to plasma insulin concentration was calculated for each subject by plotting the average plasma glucose concentrations vs. the average plasma insulin concentrations at the two plateaus of glycemia. Endogenous glucose production was calculated during the last half hour of each plateau of glycemia as the difference between whole-body glucose turnover and the exogenous glucose infusion rate.

All results in text, tables, and figures are expressed as mean ± SEM. Comparison between the studies with and without dexamethasone was done by means of paired Student's t tests, with Bonferroni's adjustment.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Dexamethasone increased fasting plasma insulin concentrations by 89% but otherwise did not significantly alter basal hormone and substrate levels or glucose and lipid oxidation rates (Figure 1).

image

Figure 1. Plasma glucose, insulin, and free fatty acid concentrations. Lines with squares indicate healthy volunteers without dexamethasone administration; lines with diamonds indicate the same volunteers after a 2-day administration of dexamethasone.

Download figure to PowerPoint

During the two-step hyperglycemic clamp, plasma glucose concentrations of 7.5 and 10 mM were effectively attained and maintained throughout the first and second hour of the hyperglycemic clamp, respectively (Figure 1). There was no difference in plasma glucose concentrations between the tests performed without and with dexamethasone at any time-point. First-phase insulin secretion and plasma insulin concentrations at each plateau were, however, increased by 80% to 120% after dexamethasone administration (Figure 1 and Table 1). The slope of the glucose-insulin curve, calculated from the data obtained at the two plateaus of glycemia, was also increased by 130%.

Table 1.  Indexes of insulin secretion and insulin sensitivity
FastingGlucose infusion (μmol/kg/min)Glucose rate of disappearance (GRd) (μmol/kg/min)Endogenous glucose production (μmol/kg/min)Insulin (pM)Insulin sensitivity (GRd/I)*
  • *

    GRd/I, glucose rate of disappearance/plasma insulin concentrations.

  • p < 0.01 vs. controls.

ControlsDexamethasoneControlsDexamethasoneControlsDexamethasoneControlsDexamethasoneControlsDexamethasone
       55 ± 34109 ± 12  
First phase      204 ± 96472 ± 101  
First plateau13.6 ± 1.212.7 ± 1.715.8 ± 1.217.3 ± 0.91.90 ± 0.302.88 ± 0.27130 ± 19234 ± 320.123 ± 0.0210.062 ± 0.011
Second plateau27.2 ± 2.127.7 ± 1.925.8 ± 1.627.0 ± 1.70.97 ± 0.361.84 ± 0.38246 ± 31501 ± 700.130 ± 0.0250.064 ± 0.010

Plasma glucagon, free fatty acid, and triglyceride concentrations, measured in basal conditions and at each plateau of glycemia, were similar with and without dexamethasone (Figures 2 and 3).

image

Figure 2. Plots of plasma free fatty acid, glucagon, and triglyceride concentrations as a function of the plasma insulin concentration attained at both levels of glycemia. Lines with squares indicate healthy volunteers without dexamethasone administration; lines with diamonds indicate the same volunteers after a 2-day administration of dexamethasone.

Download figure to PowerPoint

image

Figure 3. Plots of endogenous glucose production and glucose rate of disappearance as a function of the plasma insulin concentration attained at both levels of glycemia. p < 0.01 vs. without dexamethasone. Lines with squares indicate healthy volunteers without dexamethasone administration; lines with diamonds indicate the same volunteers after a 2-day administration of dexamethasone.

Download figure to PowerPoint

Table 2 shows whole-body glucose use rates and net substrate oxidations in basal conditions and at each step of hyperglycemia. There were, however, differences between the measurements obtained after dexamethasone and without dexamethasone administration. In contrast, insulin sensitivity was markedly reduced after dexamethasone (p < 0.01; Table 1). Endogenous glucose production rates were increased after dexamethasone by 52% (p < 0.05) and 83% (p < 0.05) during the first and second plateaus, respectively (Table 1). Figures 2 and 3 show the relationship existing between plasma insulin concentrations and the various parameters related to insulin sensitivity at each step of hyperglycemia. It is apparent that dexamethasone administration resulted in higher insulin concentrations, but similar concentrations of plasma free fatty acid, triglycerides, and glucagon, as well as similar rates of whole-body glucose turnover at the two plateaus of hyperglycemia, were found. Endogenous glucose production was different because it remained increased after dexamethasone despite hyperinsulinemia at both plateaus of glycemia (Figure 3).

Table 2.  Glucose and fat oxidation
ControlsFat oxidation (mg/kg/min)Glucose oxidation (μmol/kg/min)
ControlsDexamethasoneControlsDexamethasone
Basal0.46 ± 0.180.36 ± 0.2510.0 ± 1.18.9 ± 3.9
First plateau0.30 ± 0.170.10 ± 0.1712.8 ± 1.712.2 ± 3.3
Second plateau0.16 ± 0.17−0.09 ± 0.1115.0 ± 1.715.6 ± 2.8

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Consistent with many previous studies (14), short-term administration of dexamethasone drastically reduced insulin sensitivity in this group of healthy nonobese females. This was documented by a 51% decrease in whole-body glucose use normalized for insulin concentrations. Simultaneously, dexamethasone administration led to a marked enhancement of insulin secretion. A stimulation of both first-phase and second-phase insulin secretion was observed. In addition, the slope of the curve linking plasma glucose to insulin concentration (i.e., the increase in plasma insulin concentration elicited by a 1-mM increase in plasma glucose) could be evaluated through the two-step hyperglycemic clamp procedure used in these experiments. It was more than doubled after dexamethasone treatment.

The mechanism responsible for an increase in insulin secretion after dexamethasone treatment remains unknown. Glucocorticoids suppress insulin secretion both in vivo (24) and in vitro (25). The present data and numerous previous reports, however, indicate that dexamethasone administration in vivo results in an enhanced glucose-induced insulin secretion (14, 15, 26, 27). This may be attributed to a delayed effect of dexamethasone because islet cells exposed to this agent for several days in vitro increased their intracellular cyclic adenosine monophosphate concentrations, as well as enhanced their insulin output (28). This increase in cyclic adenosine monophosphate may, in turn, enhance glucose-induced insulin secretion. This mechanism is indeed operative in the potentiation of insulin secretion by gluco-incretins (29). Alternatively, insulin secretion might be increased by yet unspecified neural mechanisms.

It has been proposed that maintenance of a normal glucose homeostasis requires that the product of glucose-induced insulin secretion and insulin sensitivity remains constant. In other terms, glucose tolerance will remain within normal limits when insulin resistance occurs, provided that insulin secretion increases appropriately. Such an appropriate stimulation of insulin secretion was indeed documented in this group of healthy female volunteers when insulin sensitivity was induced by dexamethasone. At both steps of hyperglycemia, hyperinsulinemia was able to fully overcome the dexamethasone-induced inhibition of glucose use. The same conclusion was reached regarding net glucose oxidation. Glucose oxidation increased at each step of hyperglycemia, clearly as a result of hyperinsulinemia. Stimulation of glucose oxidation, much like stimulation of whole-body glucose use, was identical before and after dexamethasone treatment. Because skeletal muscle plays a prominent role in these two processes (30, 31, 32), these results indicate that muscle insulin resistance was fully compensated by hyperinsulinemia.

Insulin also exerts antilipolytic effects on adipose tissue, resulting in the suppression of plasma free fatty acid concentrations. This suppression of plasma free fatty acid concentration may, in turn, contribute to the regulation of whole-body glucose metabolism because increased plasma free fatty acids are known to inhibit muscle glucose uptake (33, 34, 35) and enhance endogenous glucose output (36, 37). The data displayed in Figure 3 indicate that the antilipolytic effects of insulin, like those exerted on muscle, were also fully compensated by hyperinsulinemia.

Hypertriglyceridemia is frequently encountered in patients with insulin resistance (38) and can be attributed to an enhanced hepatic triglyceride secretion and/or decreased extrahepatic clearance of triglyceride-rich lipoprotein particles (39). We, therefore, assessed whether dexamethasone altered the regulation of plasma triglyceride concentrations during hyperglycemic clamp. In studies performed without dexamethasone, plasma triglyceride concentrations were suppressed during glucose infusion. This effect might be attributed to inhibition of adipose tissue lipolysis, leading to a decrease of hepatic free fatty acid re-esterification. In addition, insulin activates lipoprotein lipase (40) and is thus expected to increase triglyceride clearance. Interestingly, this suppression of triglyceride concentrations during hyperglycemia was similar with and without dexamethasone. This indicates that any resistance to the effect of insulin on plasma triglyceride production and use was also fully compensated by increased insulin secretion.

In view of these adequate compensations of metabolic defects by hyperinsulinemia, one would predict that dexamethasone has little effect on glucose tolerance. Although it was not directly measured in this study because of the impracticability of performing sequential hyperglycemic clamps and oral glucose tolerance tests, our previous data (15) and studies by other investigators (41) clearly indicate that dexamethasone invariably leads to the development of impaired glucose tolerance within a 2-day period. The present data suggest that it might be related to alterations of endogenous glucose production. In contrast to all other effects of insulin monitored in this study, endogenous glucose production was significantly higher after dexamethasone at the two plateaus of glycemia tested. This indicated that increased insulin secretion was unable to provide full compensation for the defect in hepatic glucose metabolism induced by dexamethasone. It has indeed been observed that impaired suppression of endogenous glucose production was present in both obese and lean patients with impaired glucose tolerance, but not in those with oral tolerance (12). Our present data suggest that an increased endogenous glucose production is likely to be a key factor in the pathogenesis of glucocorticoid-induced impaired glucose tolerance. However, we cannot completely discard the hypothesis that this higher endogenous glucose production was caused by inhibition of insulin secretion by dexamethasone (24, 25).

In conclusion, our results indicate that short-term administration of dexamethasone blunts the stimulatory effects of insulin on glucose use, as well as the inhibitory effects of insulin on lipolysis and endogenous glucose production, in healthy nonobese females. Increased insulin secretion, however, allowed the restoration of the full effects of insulin, except with regard to inhibition of glucose production, which remained somewhat higher after dexamethasone treatment. Whether these observations are valid in other populations (males, elderly, or obese individuals) remains to be investigated. The enhanced glucose-induced insulin secretion and the compensatory effects of endogenous hyperinsulinemia that occur in response to insulin resistance can conveniently be assessed by performing a two-step hyperglycemic clamp without and with prior administration of dexamethasone. Prospective studies will be required to evaluate whether such an evaluation allows identification of individuals who would be at particularly high risk for metabolic complications if their insulin sensitivity declined.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

This study was supported by Swiss National Science Foundation Grant 32-67, 787.02 and Eli Lilly (Switzerland).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References