| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Clinical Research Centers |
Division of Endocrinology, Metabolism, and Nutrition, Mayo Clinic (R.B., A.B., P.S., R.A.R.), Rochester, Minnesota 55905; Aarhus Kommunehospital (M.N.), Aarhus, Denmark
Address all correspondence and requests for reprints to: Dr. Robert A. Rizza, 200 First Street SW, 5194 Joseph, Rochester, Minnesota 55905. E-mail: rizza.robert{at}mayo.edu
| Abstract |
|---|
|
|
|---|
5 mmol/L), or glucose was
permitted to remain at ambient hyperglycemic levels (
10 mmol/L)
until the following morning when euglycemia was achieved with a
variable insulin infusion. A prandial glucose infusion (containing
35 g glucose) was started at 1000 h, and the variable insulin
infusion was replaced by a constant infusion of insulin (0.25
mU/kg·min), somatostatin (60 ng/kg·min), glucagon (0.65
ng/kg·min), and GH (3 ng/kg·min) to maintain hormone concentrations
at constant basal levels. Although nocturnal glucose concentrations
were (by design) higher (P < 0.01) on the
hyperglycemic than on the euglycemic study day (10.1 ± 0.2
vs. 5.4 ± 0.1 mmol/L), glucose concentrations did not
differ either before (4.9 ± 0.1 vs. 4.9 ±
0.1 mmol/L) or during the prandial glucose infusion (peak, 11.1 ±
0.5 vs. 11.3 ± 0.5 mmol/L; incremental area,
1390 ± 254 vs. 1409 ± 196 mmol/L·6 h).
Furthermore, glucose-induced stimulation of glucose disappearance
(2068 ± 218 vs. 1957 ± 244 µmol/kg·6 h)
and suppression of glucose production (-2253 ± 378
vs. -2124 ± 257 µmol/kg·6 h) did not differ.
Thus, restoration of euglycemia by means of an overnight insulin
infusion does not alter glucose effectiveness in people with type 2
diabetes. | Introduction |
|---|
|
|
|---|
The present experiments were, therefore, undertaken to determine whether short term restoration of euglycemia by means of an overnight insulin infusion improves the ability of glucose to stimulate its own uptake and to suppress its own release in subjects with type 2 diabetes mellitus. To do so, we measured glucose effectiveness in diabetic individuals after either overnight hyperglycemia (i.e. glucose concentrations permitted to remain at ambient levels) or overnight euglycemia. On the hyperglycemic study day, the glucose concentration was acutely lowered the following morning by means of a variable insulin infusion so that glucose effectiveness was assessed in all subjects in the presence of comparable baseline glucose and insulin concentrations. We report that in contrast to its effects on insulin action (19), overnight restoration of euglycemia does not alter glucose effectiveness in subjects with type 2 diabetes.
| Subjects and Methods |
|---|
|
|
|---|
After obtaining approval from the Mayo Institutional review board, 10 subjects (6 men and 4 women) with type 2 diabetes (age, 55 ± 3 yr; height, 1.7 ± 0.04 m; weight, 84.2 ± 2.7 kg; body mass index, 30.2 ± 2.6 kg/m2; lean body mass, 52.3 ± 2.6 kg) gave informed written consent to participate in the study. Glycosylated hemoglobin concentration (GlycAffin, Isolab, Akron, OH) averaged 9.1 ± 2.7% (normal range, 47%), and the fasting glucose concentration was 9.3 ± 0.7 mmol/L. The average duration of diabetes was 9 ± 2.5 yr. All subjects were in good health and had no evidence of significant diabetic complications. Seven of the 10 volunteers were treated with oral hypoglycemic agents, which were discontinued 3 weeks before the study.
Experimental design
Each subject was studied on two occasions separated by at least 1 week. The order of study was random. On each occasion subjects were admitted to the General Clinical Research Center at 1800 h the evening before the study. After the ingestion of a standard 10 Cal/kg meal (55% carbohydrates, 30% fat, and 15% protein), subjects remained fasting (with the exception of occasional sips of water) until the end of the study. After the meal, an 18-gauge cannula was inserted into each forearm. On one occasion, human insulin (0.1 U/mL Humulin R, Eli Lilly & Co., Indianapolis, IN) was infused in amounts sufficient to maintain euglycemia throughout the night (referred to as the euglycemic study day), whereas on the other occasion, saline was infused, thereby permitting glucose to remain at ambient levels (referred to as the hyperglycemic study day) until the following morning when a variable insulin infusion was started and adjusted so as to achieve euglycemia within 23 h. On both occasions, glucose concentrations were measured at 30- to 60-min intervals, and the results were used to adjust the insulin infusion rate. The morning variable insulin infusion was started at 0800 h in the first subject, at 0700 h in the next two subjects, and at 0600 h in the remaining seven subjects to insure there were at least 60 min of stable glucose and insulin concentrations before the start of the prandial glucose infusion at 1000 h (15). Hypoglycemia did not occur either during night or on the morning of the study in any subject.
At 0530 h, an 18-gauge cannula was inserted retrogradely into a
dorsal hand vein. The hand was then placed in a heated box (
55 C) to
obtain arterialized venous blood samples. A primed continuous infusion
of [3-3H]glucose (12 µCi prime; 0.12 µCi/min
continuous; New England Nuclear, Boston, MA) was started at 0700 h
and continued until the end of the study at 1600 h. On both
occasions, the variable insulin infusion was replaced at 1000 h by
a constant infusion of insulin (0.25 mU/kg·min), somatostatin (60
ng/kg·min), glucagon (0.65 ng/kg·min), and GH (3 ng/kg·min). A
prandial infusion of glucose (containing 35 g glucose) also was
started at 1000 h as previously described (22). All infused
glucose contained [3-3H]glucose in amounts equal to the
calculated basal glucose specific activity. In addition, the rate of
infusion of [3-3H]glucose was altered so as to
approximate the anticipated pattern of fall in glucose production
(-180 to 0 min, 100%; 015 min, 70%; 1575 min, 30%; 75120 min,
45%; 120360 min, 50%). This resulted in a less than a 10% change
in plasma [3-3H]glucose specific activity during the
prandial glucose infusion. Arterialized venous blood samples and
expired air were collected at regular intervals as previously described
(23).
Analytical techniques
Plasma samples were placed on ice, centrifuged at 4 C, separated, and stored at -20 C until assay. Glucose and lactate concentrations were measured using a glucose oxidase method (YSI, Inc., Yellow Springs, OH). Plasma insulin, cortisol, and GH were measured using a chemiluminescence assay with reagents obtained from Beckman Coulter, Inc. (Access Assay, Beckman Coulter, Inc., Chaska, MN). Plasma glucagon and C peptide were measured by RIA using reagents supplied by Linco Research, Inc.. Free fatty acid concentrations were measured using a colorimetric assay (Wako Pure Chemical Industries, Osaka, Japan). Body composition was measured using dual energy x-ray absorptiometry (DPX scanner, Hologic, Inc., Waltham, MA). Plasma glucose specific activity was measured using liquid scintillation counting.
Calculations
Glucose specific activity was smoothed using the method of Bradley et al. (24). Glucose appearance and disappearance were calculated using the equations of Steele (25). The volume of distribution of glucose was assumed to equal 200 mL/kg, and the pool correction factor was assumed to equal 0.65. Endogenous glucose production was determined by subtracting the glucose infusion rate from the tracer-determined rate of glucose appearance (1).
Statistical analysis
All data are expressed as the mean ± SEM. Rates of glucose turnover are expressed per kg lean body mass. Values during the 30 min before the start of the prandial glucose infusion (i.e. -30 to 0 min) were meaned and considered as the baseline. Statistical analysis was performed using Students two-tailed paired t test. P < 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
Glucose concentrations were higher (P < 0.01) during the night on the hyperglycemic than on the euglycemic study days (10.1 ± 0.2 vs. 5.4 ± 0.1 mmol/L). After initiation of the variable insulin infusion on the morning of the hyperglycemic study day, plasma insulin concentrations increased, and plasma glucose concentrations began to fall. The variable insulin infusion was then tapered so that the insulin infusion rate from -60 min onward no longer differed on the hyperglycemic and euglycemic study days. This resulted in stable and comparable baseline glucose (4.9 ± 0.1 vs. 4.9 ± 0.1 mmol/L) and insulin concentrations (90 ± 24 vs. 72 ± 12 pmol/L) on the 2 study days.
|
|
C Peptide and glucagon concentrations (Fig. 3
)
C Peptide concentrations from 23000600 h were higher (P < 0.01) on the hyperglycemic than on the euglycemic study days (0.6 ± 0.1 vs. 0.4 ± 0.1 pmol/L). After initiation of the morning variable insulin infusion, C peptide concentrations on the hyperglycemic study day fell in parallel with the fall in glucose concentrations to values no longer different from those observed on the euglycemic study day. Somatostatin (starting at time zero along with the prandial glucose infusion) resulted in comparable suppression of C peptide on both occasions, indicating equivalent inhibition of endogenous insulin secretion. Plasma glucagon remained constant and equal on both study days. Cortisol and GH concentrations also did not differ on the 2 study days (data not shown).
|
Rates of glucose disappearance (15.9 ± 0.8 vs. 14.8 ± 0.8 µmol/kg·min) and glucose production (16.2 ± 1.4 vs. 14.3 ± 0.9 µmol/kg·min) during the 30 min before the start of the prandial glucose infusion did not differ on the hyperglycemic and euglycemic study days. The prandial glucose infusion resulted in a comparable increase in glucose disappearance (2068 ± 218 vs. 1957 ± 244 µmol/kg·6 h) and a comparable decrease in glucose production (-2253 ± 379 vs. -2124 ± 257 µmol/kg·6 h) on both the hyperglycemic and euglycemic study days.
|
Free fatty acid and lactate concentrations were comparable both before and during the prandial glucose infusion on both study days.
|
| Discussion |
|---|
|
|
|---|
There is ample evidence that chronic hyperglycemia can impair insulin action (a phenomenon commonly referred to as glucose toxicity), and that this impairment can at least in part be reversed by an improvement in glycemic control (8, 16, 17, 18, 28, 29, 30). However, we are unaware of any previous studies examining whether chronic hyperglycemia also has the same deleterious effect on the ability of glucose to regulate its own metabolism. Lack of concordance between the effect of improved glycemic control on insulin action and glucose effectiveness is not particularly surprising, as insulin and glucose use different mechanisms to modulate glucose production and utilization (3, 29, 31, 32, 33, 34). Whereas hyperglycemia and hyperinsulinemia both acutely increase glucose uptake and translocation of GLUT4 transporters to the plasma membrane, the former does so via a calcium dependent pathway (i.e. it can be inhibited by dantrolene), whereas the latter does so via the phosphoinositol kinase pathway (i.e. it can be inhibited by wortmanin) (35). Similarly, whereas an increase in either glucose or insulin concentration can suppress net hepatic glucose production, hyperglycemia appears to do so in the presence of basal insulin concentrations primarily by inhibiting glycogen breakdown, whereas hyperinsulinemia in the presence of euglycemia appears to do so primarily by stimulating glycogen synthesis (31). The lack of improvement in glucose effectiveness observed in the present experiments implies that either chronic hyperglycemia does not alter the ability of glucose to modulate its own metabolism (i.e. glucose toxicity is limited to insulin action) and therefore cannot be reversed by lowering glucose concentration or that a longer period of euglycemia is necessary for an effect to be seen. In either case, in contrast to insulin action, where an enhanced ability of insulin to suppress glucose production may help explain the clinical observation that beginning the day with a normal glucose concentration is typically associated with better daytime glycemic control, an improvement in glucose effectiveness does not appear to be a factor.
Proving no difference is impossible. The best that can be done is to minimize experimental variability so that a difference, if present, could be detected with confidence. Toward this end, the same subjects were studied on both occasions so as to avoid intersubject variation. A standard meal was ingested on both occasions on the evening before the study. All subjects received exactly the same basal insulin infusion rate to insure that insulin concentrations during the prandial glucose infusion were the same on the 2 days. This resulted in a virtually identical increase in glucose concentration above basal on the hyperglycemic and euglycemic study days. Perhaps more importantly, the increment above basal was higher in six and lower in four subjects on the hyperglycemic than on the euglycemic study days, strongly arguing against the possibility that a meaningful difference would be detected by studying a larger number of subjects.
The present experiments only examined the effects of a short term improvement in glycemic control on glucose effectiveness. On the euglycemic study day, the variable insulin infusion was adjusted so as to achieve normoglycemia by approximately 2300 h, whereas on the hyperglycemic study day, the variable insulin infusion was not started until 06000800 h on the following morning resulting in a 7- to 9-h difference in glycemic control. The time of initiation of the morning variable insulin infusion was moved from 0800 to 0700 h after the first study and to 0600 h after the second two studies to insure that the lack of difference observed in those experiments between the hyperglycemic and euglycemic study days was not due to any residual effects of the morning variable insulin infusion or to a subtle instability of the baseline glucose concentrations. Both of these possibilities seem unlikely, as the results were virtually identical in the next seven patients when the variable insulin infusion was started at 0600 h, permitting earlier achievement of stable basal glucose and insulin concentrations. The comparable free fatty acid concentrations on the two occasions further argues against different degrees of insulinization on the hyperglycemic and euglycemic study days, as lipolysis is exquisitely sensitive to small differences in insulin concentration (36). Nevertheless, we cannot totally exclude the possibility that the acute increase in insulin on the morning of the hyperglycemic study day had an impact (either positive or negative) on glucose effectiveness measured later that same morning.
It remains possible that glucose effectiveness would eventually improve after a longer period of normoglycemia. However, the present pattern of glycemic control would probably reflect the difference between that observed in an individual who received an injection of intermediate or long acting insulin at supper vs. that observed in an individual whose glucose concentrations were permitted to remain elevated throughout the night (37). Glucose concentrations were lowered on the morning of the hyperglycemic study day so that they would be same as those on the euglycemic study day at the time of initiation of the prandial glucose infusion. If this were not done, the protocol not only would have assessed the effects of a difference in antecedent glycemic control, but also the effects of a difference in ambient glucose concentration at the time of measurement of glucose effectiveness. As a number of investigators have shown that glucose clearance falls as glucose concentrations rise (38, 39, 40), interpretation of putative differences (or lack of differences) in glucose effectiveness measured in the presence of differing ambient glucose concentrations would be problematic.
A potential limitation of the present study is the fact that approximately 2 h were required to lower the glucose concentrations on the morning of the hyperglycemic study day. Furthermore, insulin was used to lower glucose. We used this approach because we have previously demonstrated (19) that a gradual lowering of glucose over 2 h does not alter counterregulatory hormone concentrations (i.e. epinephrine, norepinephrine, GH, and glucagon), which in themselves could alter glucose effectiveness. We cannot exclude the possibility that any adverse effect of chronic hyperglycemia on glucose effectiveness was fully reversed during this time. Ideally, glucose concentrations would have been instantly lowered by a means that did not require insulin and did not evoke a counterregulatory response. This, of course, is not possible in humans.
Glucose effectiveness was only assessed in the presence of a basal insulin concentration. We chose this insulin concentration because it corresponds to the level at which glucose effectiveness is measured with the so-called minimal model (41). Furthermore, it is also the insulin concentration at which we and other have shown glucose effectiveness to be impaired in subjects with type 2 diabetes mellitus (11, 12, 13, 14, 15, 26, 27). As we (15, 26) as well as other investigators (11, 12, 13, 14, 27) have consistently shown that glucose effectiveness in the presence of basal insulin concentrations is impaired in subjects with type 2 diabetes, we did not include a nondiabetic control group studied at individually determined basal insulin concentrations. Therefore, it is possible that glucose effectiveness was not impaired in all of the diabetic volunteers. In addition, as we did not measure insulin action in these subjects, we do not know whether they were also insulin resistant. However, neither of these limitations detracts from our conclusion that glucose effectiveness in people with type 2 diabetes after overnight hyperglycemia does not differ from that observed after overnight euglycemia. Insulin was infused into the peripheral rather than the portal venous system. Although the importance of portal vs. peripheral insulin infusion continues to be debated (42, 43), this is unlikely to be an issue in the present studies, as somatostatin resulted in inhibition of endogenous insulin secretion, thereby insuring that all subjects had comparable portal and peripheral insulin concentrations on the hyperglycemic and euglycemic study days. This approach also resulted in comparable portal glucagon concentrations on the 2 study days.
Thus, in summary, short term restoration of euglycemia by means of an overnight insulin infusion does not alter glucose effectiveness. Further studies will be required to determine whether a longer term improvement in glycemic control and/or whether treatment with modalities other than insulin improve glucose effectiveness in subjects with type 2 diabetes mellitus.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received January 7, 1999.
Revised April 1, 1999.
Accepted April 7, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Kishore, J. Tonelli, S. Koppaka, C. Fratila, A. Bose, D.-E. Lee, K. Reddy, and M. Hawkins Time-Dependent Effects of Free Fatty Acids on Glucose Effectiveness in Type 2 Diabetes Diabetes, June 1, 2006; 55(6): 1761 - 1768. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Hawkins, J. Tonelli, P. Kishore, D. Stein, E. Ragucci, A. Gitig, and K. Reddy Contribution of Elevated Free Fatty Acid Levels to the Lack of Glucose Effectiveness in Type 2 Diabetes Diabetes, November 1, 2003; 52(11): 2748 - 2758. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Hawkins, I. Gabriely, R. Wozniak, K. Reddy, L. Rossetti, and H. Shamoon Glycemic Control Determines Hepatic and Peripheral Glucose Effectiveness in Type 2 Diabetic Subjects Diabetes, July 1, 2002; 51(7): 2179 - 2189. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. Greenbaum, R. L. Prigeon, and D. A. D'Alessio Impaired {beta}-Cell Function, Incretin Effect, and Glucagon Suppression in Patients With Type 1 Diabetes Who Have Normal Fasting Glucose Diabetes, April 1, 2002; 51(4): 951 - 957. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Hawkins, I. Gabriely, R. Wozniak, C. Vilcu, H. Shamoon, and L. Rossetti Fructose Improves the Ability of Hyperglycemia Per Se to Regulate Glucose Production in Type 2 Diabetes Diabetes, March 1, 2002; 51(3): 606 - 614. [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 |