| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Endocrinology and Diabetes (S.K.M., N.R., L.M.Y., E.A.D., T.W.J.) and Clinical Biochemistry, Pathwest (L.M.Y.), Princess Margaret Hospital for Children, Perth, Western Australia 6840, Australia; Schools of Paediatrics and Child Health (S.K.M.) and Human Movement and Exercise Science (L.D.F., R.J.D., P.A.F.), University of Western Australia, Perth, Western Australia 6009, Australia; and Centre for Child Health Research (N.R., E.A.D., T.W.J.), Telethon Institute for Child Health Research, University of Western Australia, Perth, Western Australia 6008, Australia
Address all correspondence and requests for reprints to: Timothy W. Jones, Department of Endocrinology and Diabetes, Princess Margaret Hospital for Children, P.O. Box D184, Perth, Western Australia 6840, Australia. E-mail: tim.jones{at}health.wa.gov.au.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: This study aimed to investigate how the amount of glucose required to prevent an exercise-mediated fall in glucose level changes over time in adolescents with type 1 diabetes.
Setting: The study took place at a tertiary pediatric referral center.
Design, Participants, and Intervention: Nine adolescents with type 1 diabetes mellitus (five males, four females, aged 16 ± 1.8 yr, diabetes duration 8.2 ± 4.1 yr, hemoglobin A1c 7.8 ± 0.8%, mean ± SD) were subjected on two different occasions to a rest or 45 min of exercise at 95% of their lactate threshold. Insulin was administered iv at a rate based on their usual insulin dose, with similar plasma insulin levels for both studies (82.1 ± 19.0, exercise vs. 82.7 ± 16.4 pmol/liter, rest). Glucose was infused to maintain euglycemia for 18 h.
Main Outcome Measures: Glucose infusion rates required to maintain euglcycemia and levels of counterregulatory hormones were compared between rest and exercise study nights.
Results: Glucose infusion rates to maintain stable glucose levels were elevated during and shortly after exercise, compared with the rest study, and again from 7–11 h after exercise. Counterregulatory hormone levels were similar between exercise and rest studies except for peaks in the immediate postexercise period (epinephrine, norepinephrine, GH, and cortisol peaks: 375.6 ± 146.9 pmol/liter, 5.59 ± 0.73 nmol/liter, 71.9 ± 14.8 mIU/liter, and 558 ± 69 nmol/liter, respectively).
Conclusions: The biphasic increase in glucose requirements to maintain euglycemia after exercise suggests a unique pattern of early and delayed risk for nocturnal hypoglycemia after afternoon exercise.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A number of factors contribute to the early onset of exercise-mediated hypoglycemia in type 1 diabetes mellitus (T1DM). Insulin-treated patients with T1DM remain relatively hyperinsulinemic during exercise, in part due to the absence of a physiological decrease in insulin secretion and an increase in the absorption of insulin if previously injected in the exercising limbs (8, 9). As a result there is a rise in the rate of muscle glucose transport (10) that increases the risk of hypoglycemia. In addition, release of counterregulatory hormones such as cortisol, GH, glucagon, and norepinephrine has been shown in various circumstances to be suboptimal in adults and children with T1DM (8, 9, 11, 12, 13, 14). As a consequence of the resulting elevated portal insulin to glucagon ratio, the hepatic glucose output may be insufficient to maintain euglycemia during exercise, thus resulting in hypoglycemia. A recent report from the DirecNet Study Group found 22% of adolescents became hypoglycemic during a standardized exercise regimen (7).
Late-onset postexercise hypoglycemia, which may occur up to 6–15 h after exercise (6), is also multifactorial in origin. A contributing factor may be impaired counterregulation in response to hypoglycemia in T1DM (6). In addition, the increase in glucose uptake by skeletal muscles for the replenishment of muscle glucose stores and a rise in insulin sensitivity after exercise also contribute to the delayed risk of hypoglycemia (6, 15). Finally, because exercise in children often takes place in the afternoon after school and counterregulatory responses to hypoglycemia are impaired during sleep (16), it is likely that the risk of late-onset postexercise hypoglycemia is further increased overnight in adolescents with T1DM. To date, literature on late-onset postexercise hypoglycemia in children has been confined to studies reporting incidence of hypoglycemia and more recently studies observing the counterregulatory response to postexercise hypoglycemia (6, 7).
One approach to prevent blood glucose levels from falling during and after exercise is to ingest adequate amounts of carbohydrates, but the increase in glucose intake that is required to maintain euglycemia after exercise still remains to be established. For this reason, the primary aim of this study was to determine the increased glucose requirements to maintain euglycemia after a defined period of moderate intensity exercise and determine the temporal pattern of change over which these additional substrates are required. Because increased glucose consumption and glucose requirements are important determinants of hypoglycemia risk in the insulin-treated person with diabetes we also aimed to indirectly evaluate the risk of early and late-onset postexercise hypoglycemia in these subjects.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Nine adolescents (five males, four females) with mean age 16.0 ± 1.8 yr (mean ± SD) and duration of diabetes 8.2 ± 4.1 yr were recruited. Insulin treatment was by multiple daily injections or continuous sc insulin infusion. The mean hemoglobin (Hb) A1c at the time of the first study was 7.8 ± 0.8% and body mass index was 24.5 ± 4.5 kg/m2. The peak rate of oxygen consumption (
O2 peak) of the participants and their mean lactate thresholds expressed relative to power output or O2 peak were 37.99 ± 2.92 ml/kg·min, 115 ± 12 W, and 54.9 ± 2.8% O2 peak, respectively. No subjects had evidence of microvascular complications. The institution’s ethics committee approved the study; informed consent was obtained from the parents and assent was obtained from the participants.
Exercise clamp studies
After a familiarization session during which the anthropometric characteristics, lactate threshold, and O2 peak of all participants were determined, they were required to attend the laboratory on two further occasions. These visits consisted of an exercise (exercise study) and a control (rest study) testing session and took place approximately 4 wk apart using a counterbalanced, paired design with subjects acting as their own controls. Subjects attended for the main study days only if they had not had any known hypoglycemic episodes for 48 h and had not taken part in any exercise for 24 h. In addition, females were studied only in the midfollicular phase of the menstrual cycle. To ensure that the diet was matched between studies, all subjects kept a food diary for 24 h before the first study and then consumed similar foods before the second study. Finally, on the morning of the testing, all subjects consumed their normal breakfast at home and had their usual doses of either insulin Aspart or insulin Lispro. No intermediate or long-acting insulin was administered on the day of the study.
After arrival of the participant in the laboratory at 1100 h, a cannula was inserted in a retrograde fashion in a vein in the dorsum of one hand for sampling blood and a second cannula was inserted into a vein in the contralateral antecubital fossa for infusion of insulin and glucose. Insulin (Lispro) was infused at a constant rate and at a dose based on 50% of the subject’s usual total daily insulin dose to mimic basal insulin. Blood glucose levels were clamped for the duration of the studies at 5–6 mmol/liter by titrating infusion of a 20% (wt/vol) dextrose solution. To facilitate this, blood samples were collected every 15 min for glucose assays and glucose infusion rates were adjusted accordingly. At 1200 h, a standard lunch of one to two sandwiches designed to match the participant’s usual lunch size was given with a bolus of iv insulin based on the participant’s usual lunchtime boluses for the amount of carbohydrates consumed. The subjects fasted for the remainder of the study.
At 1600 h, subjects either exercised for 45 min on a cycle ergometer (Evolution, Geelong, Australia) at an intensity of 95% of their lactate threshold (exercise study), which corresponded to 54.9 ± 2.8% O2 peak, or sat on the bike without pedaling (rest study). Throughout recovery and for the remainder of the study, all participants rested in a seated or supine position in bed. Also, before, during and hourly after exercise, the rate of O2 consumption, CO2 production, and respiratory exchange ratio were measured, using a mask during exercise and a canopy after exercise, both linked to an indirect calorimetry system (Sensormedic, Viasys Australia). Finally, blood glucose samples were obtained every 15 min and were analyzed using a YSI analyzer (YSI, Yellow Springs, OH). Samples for the assessment of hormones and metabolites were obtained at 15-min intervals during exercise and hourly after exercise from arterialized venous blood. The study ceased at 0600 h, when subjects were given breakfast with their usual morning insulin.
Lactate threshold and O2 peak determination
The lactate threshold and O2 peak were determined concurrently by exercising each subject on the same cycle ergometer as that used for testing. Initial workload was set at 50 W and subsequently increased by 25 W every 3 min until exhaustion. Arterialized capillary blood samples were collected every 3 min for lactate levels, and rates of oxygen consumption and CO2 production were also determined throughout testing as described above.
Biochemical analyses
Plasma C-peptide and serum progesterone were measured by competitive chemiluminescent immunoassay using the Immulite and Immulite 2000, respectively (Diagnostic Products Corp., Los Angeles, CA). Whole-blood HbA1c was measured using latex immunoagglutination inhibition (DCA 2000 analyzer; Bayer, Indianapolis, IN). Serum testosterone and serum estradiol were analyzed by Coat-a-Count solid-phase RIA and double antibody RIA, respectively, both by Diagnostic Products Corp. The remainder of the hormones and metabolites were analyzed as described previously (17).
Statistical analyses
Data were analyzed using SPSS V12.0 for Windows (SPSS Inc., Chicago, IL) with P < 0.05 considered statistically significant. Hormone and substrate levels at specific time points were compared for each subject between the exercise and rest studies using the Wilcoxon signed ranks test. The mean hormone and substrate levels for all subjects were calculated. Mixed-effect models were used to assess glucose infusion rate by analyzing the repeated measure data of glucose infusion rates taken at various time points (18, 19). The model was used to adjust for possible confounding effects of demographic and biochemical variables. The model included age, gender, HbA1c, rate of insulin infusion, and work performed during exercise on the exercise day because it was not known how these variables would affect the difference in glucose infusion rates between the rest and exercise studies for individual subjects. To control for the confounding effect of repeated measures of outcome variables, several correlation structures including compound symmetry, autoregressive (18), and unstructured were assessed using Akaike’s information criterion to select the most appropriate correlation structure (20). All mixed-model fitting was done using SAS V9 PROC MIXED (SAS Institute, Cary, NC). Unless otherwise stated, all results are expressed as means ± SEM.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
, A and B).
|
During the rest study, the glucose infusion rate to maintain euglycemia remained relatively stable (Fig. 2A). However, in response to exercise, glucose infusion rates increased during and for the first 90 min after exercise, when average glucose requirements were elevated (2.57 ± 0.28 mg/kg·min), compared with the rest study requirements (1.58 ± 0.08 mg/kg·min, P < 0.05). For the next 5 h, mean glucose infusion rates were similar for both studies (1.76 ± 0.05 mg/kg·min, exercise, and 1.84 ± 0.02 mg/kg·min, rest) and not significantly different from preexercise glucose infusion rates. However, between 7 and 11 h after the completion of exercise (from 2330 until 0400 h), a second period of higher average glucose infusion rates on the exercise (1.97 ± 0.04 mg/kg·min), compared with rest (1.45 ± 0.03 mg/kg·min), study was noted (Fig. 2A). Eight of the nine subjects required a greater glucose infusion during this later time period.
|
). The difference in glucose infusion rate between exercise and rest was calculated for each subject at each time point. As shown in Fig. 2B, there was a higher difference in glucose infusion rate during exercise and in the first hour of recovery, compared with rest (mean difference = 1.58 ± 0.42 mg/kg·min) and also between both studies over the time period extending between 2400 and 0400 h (mean difference = 1.02 ± 0.03 mg/kg·min). The additional amount of glucose required to maintain euglycemia throughout recovery (252 ± 204 mg/kg) was lower than the additional carbohydrate oxidized during exercise (707 ± 114 mg/kg), with 55% of this extra glucose being administered between 7 and 11 h of recovery. Fuel metabolism
Before exercise, oxygen consumption rates (0.27 ± 0.02 vs. 0.26 ± 0.02 l/min), carbon dioxide production rates (0.25 ± 0.01 vs. 0.23 ± 0.02 l/min), and respiratory exchange ratio (0.90 ± 0.02 vs. 0.91 ± 0.01) were similar for exercise and rest days, respectively. During exercise, the absolute rate of carbohydrate oxidation increased, compared with the rest study (17.20 ± 2.51 vs. 2.32 ± 0.35 mg/kg·min, Fig. 2C) and returned to basal levels within 1 h of recovery. The rate of fat oxidation increased in the exercise study, compared with the rest study, both during exercise (4.30 ± 0.64 vs. 0.85 ± 0.08 mg/kg·min, Fig. 2D) and 3 h after exercise (1.03 ± 0.13 vs. 0.62 ± 0.09 mg/kg·min, Fig. 2D).
Hormones and metabolites
Before, during, and late after exercise, there was no difference in plasma nonesterified fatty acid (NEFA) levels between exercise and rest studies. However, plasma NEFA levels on the exercise night (0.122 ± 0.025 g/liter) were elevated 2 h after the completion of exercise, compared with the rest night (0.053 ± 0.018 g/liter, P < 0.05. Fig. 1C). In female subjects, serum estradiol levels were 82.5 ± 30.0 and 70.0 ± 23.6 pmol/liter (exercise vs. rest) and serum progesterone levels were 1.3 ± 0.29 mmol/liter for both exercise and rest studies, consistent with levels in the midfollicular phase of the menstrual cycle. Serum cortisol (558 ± 69 vs. 286 ± 28 nmol/liter, P < 0.01) and GH levels (27.7 ± 5.7 vs. 11.1 ± 3.0 ng/ml, P < 0.05) peaked at the end of exercise, compared with the rest study (Fig. 1, D and E), respectively, and serum cortisol levels remained elevated for the next 2 h (244 ± 69 vs. 108 ± 20 nmol/liter, P < 0.05). After this, there was no significant difference in cortisol or GH levels between the exercise and control studies (Fig. 1, D and E). Norepinephrine concentrations also peaked at the end of exercise (5.59 ± 0.73 nmol/liter), compared with rest (1.36 ± 0.14 nmol/liter, P < 0.01; Fig. 1F). In contrast, there was no significant difference in epinephrine levels between exercise (375.6 ± 146.9 pmol/liter, end exercise) and rest studies (184.4 ± 31.6 pmol/liter, NS) at any time point (Fig. 1G). There was no significant increase in glucagon levels in response to exercise (58.0 ± 4.8 vs. 52.2 ± 3.4 ng/liter, exercise vs. rest, Fig. 1H).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The increase in the amount of exogenous glucose required to maintain euglycemia during exercise is likely the result of a mismatch between endogenous rates of glucose production and use. This mismatch is probably due, at least in part, to the lack of suppression of insulin levels (12, 21). In nondiabetic individuals, insulin levels fall with exercise and the levels of counterregulatory hormones are increased, resulting in increased hepatic glucose output (22, 23, 24). Here, however, insulin was infused at a constant rate throughout and remained at stable levels, thus probably opposing any increase in hepatic glucose production. In addition, these elevated insulin levels are likely to have contributed further to the high rates of glucose infusion during exercise due to the additive effect of insulin and muscle contraction on peripheral glucose disposal. In support of this, a recent study from the DirecNet group (25) has demonstrated a reduction in hypoglycemia rates during exercise when basal rates on the insulin pump are suspended rather than continued during exercise. Finally, it is also possible that an insufficient increase in epinephrine and glucagon in response to exercise might have contributed to an impaired increase in hepatic glucose output as suggested by the results of others showing impaired or absent counterregulatory response to exercise in T1DM (12, 13, 14, 21).
The finding that exercise performed late in the afternoon does not increase the amount of carbohydrate required to maintain euglycemia during the evening was surprising, given that both the exercise-mediated fall in muscle glycogen stores and increase in insulin sensitivity (22, 26) were expected to result in a rise in peripheral rates of glucose disposal (6, 24, 27). It is possible that the elevated plasma NEFA levels during that time might have countered the exercise-mediated rise in glucose use via either increased NEFA oxidation, as suggested by the increased rate of fat oxidation during early recovery, or via NEFA-dependent but oxidation-independent mechanisms (28). Because cortisol increases lipolysis and therefore levels of NEFA (29) and GH increases lipid oxidation and decreases both muscle glucose uptake and muscle glycogen synthase activity (30), this raises the possibility that the elevated levels of these hormones at the onset of recovery might have had a long-lasting effect in opposing, at least in part, the expected exercise-mediated rise in postexercise rate of glucose use.
In contrast to the early postexercise period, there was a difference in glucose infusion rates between the exercise and rest studies from 2400 to 0400 h. This suggests an imbalance between glucose production and use during that time, resulting in a higher need for exogenous glucose. Given that there was no difference between studies with respect to absolute glucose oxidation rates during night time, it is possible that the relative increase in glucose requirements to maintain euglycemia late at night in the exercise study serves to support the repletion of muscle glycogen stores. It is also important to note that the delayed increase in glucose requirements on the exercise night, compared with the rest night, was coincident with the onset of sleep and therefore that sleep may be a contributing factor to the increase in risk for hypoglycemia. Indeed, postexercise hypoglycemia at night has been shown to have an early onset in younger patients, and this may be related in timing to the onset of sleep (6).
On clinical grounds, the measurement of the amount of glucose required to maintain euglycemia in response to exercise might provide an indirect and novel approach for assessing some aspects of the risk of hypoglycemia associated with exercise while maintaining euglycemia. In this respect, a recently published study has confirmed that episodes of hypoglycemia were indeed more frequent after afternoon exercise in juveniles with diabetes (7). It is important to note, however, that a further contributing factor to the risk of late-onset postexercise hypoglycemia is the impaired counterregulation to hypoglycemia after exercise (6) because antecedent exercise diminishes the responses of glucagon, norepinephrine, and epinephrine to subsequent hypoglycemia (31, 32). Given that all subjects remained euglycemic for the duration of the experimental protocol adopted here, this current study was not designed to evaluate the counterregulatory responses to hypoglycemia. We have previously shown, however, that the counterregulatory responses of children to hypoglycemia during sleep are blunted (16).
Conclusions
This study shows that adolescents with T1DM taking part in moderate intensity exercise in the afternoon have increased glucose requirements at the time of and shortly after the completion of exercise and also from approximately 2400 to 0400 h. This together with the reported diminished counterregulatory responses to hypoglycemia after exercise may lead to a greater risk of hypoglycemia overnight. It remains to be established, however, the extent to which the ingestion of one or several meals after exercise as opposed to the continuous infusion of glucose is likely to affect the temporal pattern of change in both the amount of glucose required to stabilize glucose levels after exercise and the associated risk of hypoglycemia.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
This paper was published in abstract form at the 65th Scientific Sessions of the American Diabetes Association, June 2005; the Annual Scientific Meeting of the Australian Diabetes Society, September 2005; and at the 41st Annual Meeting of the European Association for the Study of Diabetes, September 2005.
Disclosure Statement: The authors have nothing to disclose.
First Published Online November 21, 2006
Abbreviations: Hb, Hemoglobin; NEFA, nonesterified fatty acid; T1DM, type 1 diabetes mellitus; O2 peak, peak rate of oxygen consumption.
Received October 17, 2006.
Accepted November 13, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  The Highs and Lows of Exercise and Diabetes Journal Watch Pediatrics and Adolescent Medicine, April 18, 2007; 2007(418): 3 - 3. [Full Text]
|
|
|
|
 |
|
 W. V. Tamborlane Triple Jeopardy: Nocturnal Hypoglycemia after Exercise in the Young with Diabetes J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 815 - 816. [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 |
