| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Medicine/Endocrinology, University of New Mexico Health Sciences Center, Albuquerque, New Mexico 87131
Address all correspondence and requests for reprints to: Mark R. Burge, M.D., Assistant Professor of Medicine, University of New Mexico Health Sciences Center, Department of Medicine/Endocrinology-5ACC, Albuquerque, New Mexico 87131. E-mail: MBurge{at}Salud.unm.edu
Abstract
Hypoglycemia is the principal barrier to achieving target glucose goals in type 2 diabetes. The effect of short-term improvement in glycemic control on plasma glucose thresholds for symptomatic and hormonal responses to hypoglycemia in type 2 diabetes is not known. We hypothesized that the thresholds for these events would be increased by 1 wk of improved glycemic control in elderly patients with type 2 diabetes. Ten elderly patients with type 2 diabetes were admitted for an 8-d inpatient protocol. All subjects underwent insulin-induced hypoglycemia on days 2 (preglucose control) and 8 (postglucose control). Between days 2 and 8, subjects received intensive diabetes management to improve their glycemic control. Timed blood glucose profiles were obtained daily during the week before and during admission. Plasma glucose, counterregulatory hormones, and hypoglycemic symptoms were assessed at baseline and every 10 min during the hypoglycemic studies. Mean blood glucose concentrations were significantly reduced by intensive diabetes management from 9.8 ± 3.7 mmol/liter to 7.7 ± 3.3 mmol/liter (P < 0.001). The plasma glucose threshold for epinephrine release during insulin-induced hypoglycemia was significantly increased by intensive management from a glucose concentration of 3.7 ± 0.5 mmol/liter at baseline to 3.1 ± 0.3 mmol/liter after intensive management (P < 0.05). The plasma glucose threshold for hypoglycemic symptoms was also increased by intensive therapy from a glucose concentration of 5.3 ± 1.2 to 3.3 ± 0.6 mmol/liter (P = 0.003). These rapid changes may increase the risk for severe hypoglycemia in type 2 diabetes and limit the ability of physicians to rapidly correct hyperglycemia in elderly type 2 diabetes patients.
ALTHOUGH THE SYMPTOMATIC and hormonal responses to hypoglycemia have been extensively studied in healthy subjects and in patients with type 1 diabetes, less is known about how patients with type 2 diabetes respond to a hypoglycemic stress (1, 2, 3, 4). Recent studies have established that improvement in glycemic control alters the clinical and hormonal responses to hypoglycemia in type 2 diabetes. Specifically, Levy et al. (5) have demonstrated that poorly controlled patients with type 2 diabetes exhibit counterregulatory hormone responses that are initiated at higher plasma glucose concentrations than do healthy patients or patients with type 1 diabetes during a stepped hypoglycemic clamp. Additionally, Korzon-Burakowska et al. (6) have shown that improvement in glycemic control over a period of several months alters the clinical and hormonal response to hypoglycemia in type 2 diabetes by normalizing the plasma glucose thresholds for counterregulatory hormone responses relative to nondiabetic subjects and by decreasing the severity of hypoglycemic symptoms at any given plasma glucose concentration relative to poorly controlled subjects. Finally, our group has previously demonstrated that elderly, poorly controlled, sulfonylurea-treated patients with type 2 diabetes secrete epinephrine as the primary hormonal response to a 23-h fast and, furthermore, that epinephrine is secreted at elevated concentrations of glucose in such patients (7).
Despite these recent advances, however, the effect of short-term improvement in glycemic control on the clinical and hormonal response to hypoglycemia in type 2 diabetes is unknown. On the basis of our previous findings and published data in patients with type 1 diabetes, we hypothesized that elderly patients with type 2 diabetes would demonstrate physiological alterations in their response to hypoglycemia within 1 wk of initiating improved glycemic control. Specifically, we hypothesized that counterregulatory hormone release and hypoglycemic symptoms would occur at reduced plasma glucose concentrations after 1 wk of intensive diabetes therapy, compared with the response before the institution of intensive diabetes management. An elderly study population was chosen because these subjects may be particularly vulnerable to hypoglycemia and its consequences (8, 9). Our results suggest that such alterations do occur and that the effect of improved glycemic control is more pronounced with epinephrine than with the other counterregulatory hormones.
Subjects and Methods
Study subjects
Ten elderly patients with type 2 diabetes were admitted to the
University of New Mexico General Clinical Research Center (GCRC) for an
8-d inpatient protocol. Volunteers were excluded from study by the
presence of severe cardiovascular, gastrointestinal, renal, or hepatic
disease; use of concurrent medications that altered glucose
homeostasis; malignancy; or a history of substance abuse. All subjects
were aged 60 yr or greater, carried a diagnosis of type 2 diabetes for
at least 1 yr, were treated solely with oral agents for at least 6
months (e.g. sulfonylureas as monotherapy or in
combination with metformin), were poorly controlled with hemoglobin
A1C levels of 8% or greater, and were free of
advanced complications of diabetes. None of the subjects had a history
of clinically significant autonomic neuropathy or its sequelae.
Specifically, none of the study subjects had a history of hypoglycemia
unawareness, gastroenteropathy, lower extremity ulceration, or
orthostatic hypotension, although three of six men studied reported
some degree of sexual dysfunction. Finally, no subject had a corrected
QT interval greater than 440 ms at screening, a value that has been
associated with autonomic neuropathy and increased mortality in
patients with type 1 diabetes (Table 1
)
(10). One subject was excluded from study because of a
history of hypoglycemia unawareness and a recent episode of severe
hypoglycemia. Before study, all subjects provided written informed
consent as approved by the University of New Mexico Human Research
Review Committee. Baseline descriptive characteristics of the study
participants are summarized in Table 1.
|
All subjects received glyburide 20 mg/d for at least 1 wk before study enrollment and during their admission for the 8-d inpatient protocol. Subjects were admitted to the GCRC at 0800 h on study day 1, and they subsequently ingested three standard meals according to the recommendations of the American Diabetes Association (50% carbohydrate, 20% protein, and 30% fat) (11). In the evening, forearm venous catheters were placed in each arm for the administration of insulin (one arm) and for the repeated sampling of arterialized blood (contralateral arm). All study subjects subsequently underwent an insulin-induced hypoglycemia study in the fasting state at 0900 h on the morning of day 2 to reflect the clinical and hormonal response to hypoglycemia under baseline, nonintensified diabetes therapy. During the insulin-induced hypoglycemia studies, all subjects received a continuous infusion of regular human insulin (Eli Lilly & Co., Indianapolis, IN) mixed to a strength of 0.1 U/ml in heparinized 0.45 M saline at a rate of 0.2 u/kg per hr. Blood samples were collected in duplicate before initiating the insulin infusion for baseline determination and then every 10 min for the remainder of the study for determination of plasma glucose, total serum insulin, C peptide, glucagon, epinephrine, norepinephrine, cortisol, and GH concentrations. Eleven hypoglycemic symptoms were also assessed every 15 min and were scored on a scale of 1 (no symptoms) to 3 (severe symptoms). These symptoms included sweating, nervousness, shaking, palpitation, feeling hot, difficulty speaking, confusion, dizziness, irritability, drowsiness, and tingling.
The insulin infusion was continued until plasma glucose concentrations fell below 3.3 mmol/liter (60 mg/dl) with typical hypoglycemic symptoms or below 2.8 mmol/liter (50 mg/dl) in the absence of symptoms. Following the first insulin-induced hypoglycemia study, study subjects received intensive diabetes management with modest caloric restriction (23 kcal/kg per day) and sc insulin injections in addition to their glyburide. Insulin injections consisted of human 70/30 (Eli Lilly & Co.) initiated at 0.25 U/kg per day in divided doses. These injections were supplemented with premeal injections of insulin lispro (Humalog, Eli Lilly & Co.) to achieve target glycemic goals of 6.7 mmol/liter (120 mg/dl) before meals and 9.9 mmol/liter (180 mg/dl) 2 h after meals.
To assess the efficacy of intensified diabetes management, daily timed capillary blood glucose profiles were obtained during the week before admission (to reflect baseline glycemic control) and during admission (to reflect intensive diabetes management) using the HemoCue blood glucose analyzer (HemoCue, Mission Viejo, CA) (12). The blood glucose profiles consisted of daily determination of blood glucose concentrations immediately before each meal and 2 h after each meal. Additionally, blood glucose concentrations were determined at 0300 h during the week of intensive diabetes management. Care was taken to avoid hypoglycemia during the week of intensive diabetes management. A second insulin-induced hypoglycemia study was performed on study day 8 that was identical to the first insulin-induced hypoglycemia study. This study was undertaken to reflect the clinical and hormonal response to hypoglycemia among elderly patients with type 2 diabetes after a week of intensified diabetes management.
Sample analyses
Plasma glucose concentrations during the insulin-induced hypoglycemia studies were determined with a glucose analyzer II (Beckman Coulter Instruments, Inc., Fullerton, CA). Plasma was separated from blood elements by centrifugation immediately after sampling and frozen at -20 C for later determination unless capillary blood glucose values were less than 4.4 mmol/liter, at which time plasma glucose levels were determined immediately. Total serum insulin concentrations were determined using the Coat-a-Count insulin RIA kit (Diagnostic Products Corp., Los Angeles, CA), and C-peptide concentrations were also determined using RIA (INCSTAR Corp., Steelwater, MN). Samples for plasma epinephrine and norepinephrine were placed on ice immediately after sampling and stored at -70 C until being assayed radioenzymatically (13). Concentrations of cortisol were determined by RIA (Coat-a-Count, Diagnostic Products), as were concentrations of GH (Diagnostic Products) and glucagon (Linco Research, Inc., St. Louis, MO).
Statistical analysis. All substrate and hormonal variables were compared for an effect of treatment group (preglucose control vs. postglucose control) with a repeated-measures ANOVA and post hoc pair-wise comparison, when appropriate. For clarity of presentation, data for all variables were reduced to four summary parameters for subsequent statistical analysis: (a) baseline concentration obtained before starting insulin infusion, (b) peak concentration during the insulin-induced hypoglycemia study, (c) nadir concentration during the insulin-induced hypoglycemia study, and (d) mean concentration during the insulin-induced hypoglycemia study. Results are expressed as the mean ± SD in the text and tables and as mean ± SE in the figures.
Plasma glucose thresholds for composite symptom scores and counterregulatory hormone release were determined using a two-segment nonlinear model that employed an intercept, slope, and glucose threshold at which the slope changed using SAS (SAS Institute, Inc., Cary, NC), as previously described (7). Variables were analyzed in conjunction with concomitantly obtained plasma glucose concentrations, and the glucose concentration at which the slope changed from zero was designated the plasma glucose threshold for counterregulatory hormone release (14). Glucose thresholds for counterregulatory hormone release and hypoglycemic symptoms were compared using a paired t test. Finally, in an attempt to clarify the terminology used in this report, plasma glucose thresholds for a biological event (such as hormone release) are said to increase if a greater degree of hypoglycemia (i.e. a lower plasma glucose concentration) is required to trigger the event.
Results
Glycemic control
Repeated-measures ANOVA demonstrated a significant reduction in
the mean blood glucose concentration during intensive diabetes
management, compared with preglucose control (7.7 ± 3.3
vs. 9.8 ± 3.7 mmol/liter; P < 0.001)
(Fig. 1, A and B). Hypoglycemia was
minimized during the week of intensive therapy. Specifically, 11
episodes of capillary blood glucose < 3.3 mmol/liter of 378
determinations were recorded during the week of intensive therapy, and
none of these occurred within 48 h of the second (intensive
therapy) insulin-induced hypoglycemia study.
|
Baseline concentrations were not significantly different between
the two study conditions for any of the counterregulatory
hormones. Results of paired analyses for all of the study summary
variables are shown in Table 2.
|
). As shown in Table 2
, mean glucagon
concentrations during insulin-induced hypoglycemia were 60 ± 20
ng/liter preglucose control and 60 ± 20 ng/liter postglucose
control (P = 0.68). Similarly, peak glucagon
concentrations during insulin-induced hypoglycemia were 80 ± 30
ng/liter preglucose control and 80 ± 40 ng/liter postglucose
control (P = 0.86). Plasma glucose thresholds for
glucagon release could not be determined because glucagon
concentrations failed to significantly increase during the observation
period.
|
).
Plasma glucose thresholds for epinephrine release were determined for 9
of the 10 subjects. Thresholds could not be determined in one patient
using the statistical model employed owing to an insufficient rise in
epinephrine concentrations. Plasma glucose thresholds for epinephrine
release derived by piece-wise linear regression were increased after 1
wk of intensive diabetes therapy, compared with results obtained after
1 wk of baseline therapy: preglucose control = 3.7 ± 0.5
mmol/liter vs. postglucose control = 3.1 ± 0.3
mmol/liter (P < 0.05). Pooled results for plasma
epinephrine as a function of plasma glucose are shown in Fig. 3.
|
and Fig. 2C). Similarly, peak
norepinephrine concentrations did not differ significantly during
insulin-induced after 1 wk of intensive therapy, compared with results
obtained after 1 wk of baseline therapy: preglucose control =
3.52 ± 1.34 nmol/liter vs. 3.51 ± 2.15
nmol/liter (P = 1.00).
Serum GH concentrations were not different between the two study days
(Fig. 2D). Specifically, the mean GH concentration during
insulin-induced hypoglycemia after 1 wk of baseline therapy was
3.7 ± 2.6 µg/liter, compared with 2.3 ± 1.1 µg/liter
after 1 wk of intensive therapy (P = 0.08). Similarly,
peak GH concentration did not differ significantly during
insulin-induced hypoglycemia after 1 wk of intensive therapy:
preglucose control = 13.6 ± 8.8 µg/liter vs.
postglucose control = 8.7 ± 4.7 µg/liter
(P = 0.13). Plasma glucose thresholds for GH release
were determined for 9 of the 10 subjects. A threshold could not be
determined in the postglucose control study in one patient using the
statistical model employed owing to an insufficient rise in GH
concentrations. The plasma glucose threshold for GH release did not
differ after 1 wk of intensive diabetes therapy, compared with results
obtained after 1 wk of baseline therapy: preglucose control =
4.7 ± 2.1 mmol/liter vs. postglucose control =
3.5 ± 0.9 mmol/liter (P = 0.28).
As shown in Fig. 2E, mean serum cortisol concentrations did not differ
statistically between the two insulin-induced hypoglycemia studies.
Preglucose control = 410 ± 100 nmol/liter vs.
postglucose control = 400 ± 120 nmol/liter
(P = 0.70). Similarly, mean peak cortisol concentration
did not differ significantly during insulin-induced after 1 wk of
intensive therapy, compared with results obtained after 1 wk of
baseline therapy: preglucose control = 650 ± 170 nmol/liter
vs. postglucose control = 570 ± 160 nmol/liter
(P = 0.27). Plasma glucose thresholds for cortisol
release were determined for 7 of the 10 subjects. Thresholds could not
be determined in three patients using the statistical model employed
owing to an insufficient rise in cortisol concentrations. The plasma
glucose threshold for cortisol release did not differ after 1 wk of
intensive diabetes therapy, compared with results obtained after 1 wk
of baseline therapy: preglucose control = 6.6 ± 3.2
mmol/liter vs. postglucose control = 4.0 ± 3.3
mmol/liter (P = 0.24).
Hypoglycemia symptoms
Analysis of the hypoglycemia symptom questionnaire data revealed
statistically significant reductions in mean (14 ± 1
vs. 12 ± 1, P = 0.01) and peak
(21 ± 2 vs. 17 ± 3, P = 0.04)
composite symptom scores after 1 wk of intensive diabetes therapy
(Table 2). Additionally, there was a statistically significant increase
in the plasma glucose threshold for development of hypoglycemic
symptoms during the insulin-induced hypoglycemia study after a week of
intensive diabetes therapy, compared with the response that occurred
following baseline therapy: preglucose control = 5.3 ± 1.2
mmol/liter vs. postglucose control = 3.3 ± 0.6
mmol/liter (P = 0.003). Pooled results for composite
hypoglycemia symptom scores as a function of plasma glucose are shown
in Fig. 4.
|
The Diabetes Control and Complications Trial demonstrated that although near-normal glycemia is critical to the avoidance of the secondary complications of type 1 diabetes, hypoglycemia is the limiting factor preventing achievement of target glycemic goals in most patients (15). Intensive diabetes management has been also recommended for subjects with type 2 diabetes to decrease the incidence and progression of microvascular complications of diabetes (16, 17). However, the determinants of hypoglycemia have been characterized to a much lesser extent in this condition. In particular, the relationship between glycemic control and glucose counterregulation in type 2 diabetes remains uncertain (18, 19, 20, 21, 22). Nevertheless, emerging evidence suggests that counterregulatory mechanisms are as subject to change in type 2 diabetes as they are in type 1 diabetes. Korzon-Burakowska et al. (6) have recently shown that more than 3 months of intensive glycemic control with insulin therapy in type 2 diabetes normalizes the hormonal responses to hypoglycemia and reduces the plasma glucose thresholds for hypoglycemic symptoms. The effect of short-term improvement in glycemic control on plasma glucose thresholds for hormonal and symptomatic responses in type 2 diabetes has not been previously reported.
Our results demonstrate that the epinephrine response to hypoglycemia is initiated at elevated plasma glucose concentrations in poorly controlled subjects with type 2 diabetes (i.e. 3.7 ± 0.5 mmol/liter). These findings are consistent with other studies in poorly controlled type 2 diabetes patients (7, 20, 22). Our study further shows, however, that the plasma glucose concentration at which epinephrine is released during hypoglycemia is significantly reduced after only 1 wk of intensive blood glucose control using multidose insulin in the context of modest caloric restriction and sulfonylurea therapy. Spyer et al. (23) have recently published a report comparing the counterregulatory and symptomatic responses of seven well-controlled type 2 diabetes patients with those of seven healthy controls during a stepped hyperinsulinemic, hypoglycemic clamp study. They found that the type 2 diabetes patients experienced hypoglycemic symptoms and released counterregulatory hormones at higher plasma glucose concentrations than did the nondiabetic subjects. Specifically, patients with type 2 diabetes developed hypoglycemic symptoms at a blood glucose concentration of 3.7 mmol/liter, compared with 2.6 mmol/liter for the control subjects. Thresholds for counterregulatory hormone release also occurred at significantly higher glucose concentrations for epinephrine (3.8 vs. 2.9 mmol/liter), glucagon (3.1 vs. 2.6 mmol/liter), GH (3.7 vs. 3.1 mmol/liter), and cortisol (4.5 vs. 3.1 mmol/liter). These findings suggest that clinically significant alterations exist in the response to hypoglycemia among well-controlled subjects with type 2 diabetes.
Compared with the current study, Spyer et al. (23) studied subjects with lower hemoglobin A1C concentrations (7.4% vs. 9.0%) and a shorter duration of diabetes (less than 7 yr vs. an average of 9 yr). The lack of a glucagon response to hypoglycemia in the current study contrasts with the study of Spyer et al. (23), suggesting that an acquired loss of the glucagon response to hypoglycemia may occur with a longer duration of diabetes, as is the case in type 1 diabetes (24). Aside from these differences, the basal and peak hormone concentrations reported are comparable between these two studies. It is noteworthy, however, that hypoglycemic symptoms and hormone release during the baseline hypoglycemia study reported in our study occur at higher plasma glucose concentrations than those reported in the Spyer study but that the thresholds reported during the postglucose control study are generally comparable with those of Spyer et al. (23).
There are currently many experimental approaches that can be employed to characterize and/or quantify the biological response to hypoglycemia. One commonly used method involves the serial assessment of hormonal and symptomatic responses following a step-wise, incremental reduction of plasma glucose using a stepped hyperinsulinemic clamp (5, 6, 23). This method provides the advantage of careful experimental control over a number of variables but has the disadvantage of artificially prolonging an event that occurs very briefly (i.e. over a few minutes) in the clinical setting (25). Another method is to rapidly reduce plasma glucose from euglycemic levels to a predetermined target level of hypoglycemia (22). This approach duplicates the rapidity with which hypoglycemia occurs in the clinical setting and allows comparison of the magnitude of a given hormonal response at a given level of blood glucose, but it does not allow assessment of the precise threshold at which events occur. A third approach, first employed by Clutter et al. (26), is to duplicate the insulin excess and relatively rapid fall in plasma glucose that occurs in hypoglycemic diabetic subjects and to characterize what is happening during this period of change by sampling frequently and using statistical models to decipher when different variables have changed from baseline. Although none of these methods has been demonstrated to be superior to another, we have (in consultation with our statistician) chosen the latter approach for use in the current study as being the most representative of clinical hypoglycemia. To the extent that the statistical model accurately describes the data, this strategy provides a reliable, objective, and reproducible method for describing events that are occurring very rapidly in vivo under nonsteady state conditions.
A common counterregulatory defect observed in patients with type 1
diabetes is the absence of a glucagon response to hypoglycemia
(24, 27). In type 2 diabetes, glucagon responses to
hypoglycemia have been variably reported as being either similar or
reduced in magnitude to those observed in nondiabetic subjects
(18, 22, 28). No significant glucagon response to
hypoglycemia was observed in the patients who participated in the
current study, and thresholds for its release could not be derived.
This may be attributable to the suppression of glucagon secretion by
hyperinsulinemia, as previously reported (29).
Specifically, glucagon secretion has been shown to be suppressed by
hyperinsulinemia in both nondiabetic and type 2 diabetic subjects
(30, 31). Mean serum insulin concentrations were quite
high in the current study (see Table 2), so an effect of
hyperinsulinemia on glucagon release cannot be excluded.
Korzon-Burakowska et al. (6) have shown a
significant increase in the plasma glucose thresholds for glucagon, GH,
and cortisol release after 3 months of improved glycemic control with
insulin in poorly controlled type 2 diabetic patients. The differences
in glucagon responses and plasma glucose thresholds observed in their
study, compared with ours could potentially be explained by the
duration of the glycemic control or by the severity and/or duration of
diabetes in the subjects who participated in these two studies.
Concentrations of GH and cortisol did not differ significantly between the two study conditions in this study, but this is not surprising given that these hormones are not involved in the acute response to hypoglycemia in humans (32, 33). However, the fact that concentrations of each of these hormones increased similarly in both the baseline and the postglucose control insulin-induced hypoglycemia studies suggests that these responses are unaffected by short-term improvements in glycemic control. Additionally, it is possible that the sample size of this study may be too small to detect a significant difference in GH concentrations.
The main clinical concern associated with reduced counterregulatory responses to a given hypoglycemic stimulus involves the risk of hypoglycemia unawareness and the severe hypoglycemia associated with this phenomenon. Previous studies suggest that the hypoglycemic symptoms that occur during elevated plasma glucose concentrations in poorly controlled diabetic patients constitute a significant barrier to improving glycemic control (6). As such, the reduction in symptoms associated with successful intervention to improve glycemia in type 2 diabetes raises concerns over patient safety and the appropriateness of target glycemic goals in such patients. Elderly patients with type 2 diabetes may be at particular risk for developing severe asymptomatic hypoglycemia with intensive diabetes therapy because normal aging is associated with blunted symptoms and cognitive impairment during hypoglycemia, compared with healthy young control subjects (34, 35). Our results demonstrate that the plasma glucose threshold for development of hypoglycemic symptoms occurs at plasma glucose levels that are in the normoglycemic range in poorly controlled type 2 diabetes receiving sulfonylurea therapy (i.e. 5.3 ± 1.2 mmol/liter).
One potential limitation of this study is the fact that the counterregulatory response to hypoglycemia has been shown to be blunted by antecedent episodes of hypoglycemia in nondiabetic and type 1 diabetic humans (1, 3, 4, 37, 38, 39). This phenomenon is thought to be short lived, however, lasting no more than 48 h following a 2-h episode of hypoglycemia in patients with type 1 diabetes (40). Some investigators, however, have demonstrated a more prolonged effect of antecedent hypoglycemia resulting in a blunted epinephrine response to subsequent hypoglycemia lasting up to 5–6 d in normal healthy subjects (41, 42). Therefore, it remains possible that the results of this study are a manifestation of pathophysiology precipitated by previous hypoglycemia that occurred during the week of intensive diabetes management. This qualification does not, however, diminish the main findings of this study that patients with type 2 diabetes experience hypoglycemic unawareness and so are at increased risk for severe hypoglycemia, soon after initiating intensified diabetes management.
It is also possible that the lower baseline plasma glucose concentrations after the week of intensive therapy, compared with the preglucose control study (e.g. 6.6 vs. 10.2 mmol/liter) may have influenced the results of the current study. This would require, however, that either rate of fall or magnitude of fall of plasma glucose determines the metabolic response to a hypoglycemic stimulus. This has been found not to be the case in type 1 diabetes (43). The aim of our study was to duplicate as closely as possible the situation that occurs in hyperinsulinemic type 2 diabetes patients shortly after the institution of intensive therapy. In such a scenario, it would be expected that prevailing glucose concentrations would be reduced shortly after the addition of intensive therapy. Thus, although plasma glucose levels are lower during day 8 testing than during day 2 testing and although it is possible that this may have affected the thresholds for counterregulatory hormone release, this is an expected clinical result of intensified diabetes management. Finally, although the inclusion of a standard treatment control group would have been advantageous for determination of the effects of intensive diabetes management on thresholds for counterregulatory hormone release, it was determined that it was impractical to hospitalize subjects for a week without providing them the benefit of improved diabetes control.
In conclusion, our study shows that plasma glucose thresholds for epinephrine release and hypoglycemic symptoms are increased by 1 wk of intensive glycemic control. This physiologic occurrence increases the risk for hypoglycemic unawareness and may increase the risk for severe hypoglycemia shortly after instituting intensified management in such patients. Finally, these data suggest that, similar to the situation in type 1 diabetes, epinephrine assumes a primary role in the defense against insulin-induced hypoglycemia in patients with poorly controlled type 2 diabetes.
Footnotes
This work was supported by the University of New Mexico General Clinical Research Center (NIH NCRR GCRC Grant 5 M01-RR-00997) and by NIH NIDDK Grant 1-K23-DK-02680-01.
Received December 11, 2000.
Accepted July 24, 2001.
References
cell defect. Science 182:171–173This article has been cited by other articles:
 |
|
 |
|
  S. N. Davis, S. Mann, V. J. Briscoe, A. C. Ertl, and D. B. Tate Effects of Intensive Therapy and Antecedent Hypoglycemia on Counterregulatory Responses to Hypoglycemia in Type 2 Diabetes Diabetes, March 1, 2009; 58(3): 701 - 709. [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 |
