In adults with type 1 diabetes mellitus (T1DM), basal–bolus therapy using rapid and long-acting insulin analogs offers improved glycemic control versus regular human insulin (14). The health benefits of intensive diabetes control (17) may be offset by a significant increase risk of severe hypoglycemia at rest (10) and possibly even more during exercise. Physical activity must be integrated into diabetes management for its numerous benefits (20), but exercise-induced hypoglycemia may survene if diabetic treatment is not optimally coordinated (insulin reduction/carbohydrate (CHO) supplement) when T1DM individuals become active. Thus, it is critical to test strategies to prevent hypoglycemia, in particular, during and after moderate-intensity exercise, a strong recommendation in these patients at high risk of cardiovascular disease (20).
Many factors such as intensity, duration, and timing of the exercise; composition and time of the last meal before exercise; preexercise blood glucose (BG) level; insulin type and dose; and time and site of injection influence glucose kinetics during exercise (15, 9, 18, 7). The American Diabetes Association (ADA) (1) suggests that additional CHO may be required for unplanned exercise and proposes that approximately 10–15 g of CHO per hour of moderate physical activity would be sufficient for a 70-kg person. However, this recommendation is not precise with respect to the insulin regimen involved. Few reports have been published on newer insulin combinations based on widely used insulin analog, such as Lispro with either intermediate insulin Humulin (19) or long-acting insulin U (12) with respect to exercise. We have already studied the effect of exercise under a regimen of N-Lispro and found that for 60 min of late postbreakfast exercise followed by 60 min of recovery, an estimated 40 g of a liquid glucose supplement, ingested 15 min before exercise, maintained safe BG levels (4). In adult type 1 diabetes patients, glulisine and Lispro have comparable efficacy in terms of glycemic control and rates of symptomatic hypoglycemia when administered as a basal–bolus regimen with insulin glargine. Glulisine is a recently available rapid-acting insulin analog, and its effect in combination with glargine, a long-acting insulin analog, on exercise BG and specific precautions with glucose supplement in prevention of hypoglycemia during exercise are not well documented. Insulin glulisine may have properties that merit to be specifically studied. Indeed, glulisine has been useful in the treatment of allergy to rapid-acting insulin and its rapid acting analogs (11). To our knowledge, information evaluating the combination of insulin analogs glargine/glulisine in an exercise context is limited. Moreover, the combination of glargine/glulisine should be studied in specific exercise situations, that is, moderate versus intermittent high intensity.
A regimen of intermittent high-intensity exercise has been proposed to attenuate the decline in BG levels and has been recommended to minimize the risk of exercise-induced hypoglycemia in complication-free individuals with T1DM (6). Guelfi et al. (6) used the 4-s sprints performed every 2-min model to simulate the activity patterns of team sports. Other groups (16,5) have tested the effects of other sprint modalities (1 min with 1-min rest and 20-s efforts every 2 min, respectively) with interesting results. Therefore, we combined sprints duration of 10 s performed every 2 min for a longer period, i.e., 60 min. This exercise duration was also chosen to compare it with 60-min moderate-intensity session, a duration we already studied in previous trials with other insulin regimen. This exercise duration also mimics the usual duration of many sports (spinning session, ice hockey, etc.), but little is known about the effect of this regimen in glargine/glulisine insulin users.
Considering that no study has been performed with glargine/glulisine insulin in active patients with T1DM, the effect of a postprandial exercise supplement (30G) and intensity of exercise (moderate and high intensity) strategies on BG must be tested so that these participants may perform 60-min exercise with a reduced risk of hypoglycemia. Therefore, the objective of this study was to evaluate the effect of a preexercise liquid glucose snack or a variable intensity exercise protocol on BG when a 60 min-exercise session is performed 120 min after lunch in glargine/glulisine insulin users.
Eleven moderately active participants with T1DM (five men and six women) participated in this study. The women, who were all taking oral contraceptives, were studied in the follicular menstrual phase (3). Participants were free of diabetic complications and had no contraindication for exercise. All participants were on the basal–bolus insulin regimen using an insulin analog glargine at bedtime and glulisine before every meal (insulin was provided by Sanofi-aventis Canada Inc., Laval, QC, Canada). This study was approved by the ethics committee of Centre Hospitalier Universitaire de Québec, and all patients gave their informed written consent.
Premeal insulin (mean prelunch dose was 10.4 ± 3.9 U) was determined by a series of test meal to obtain a 2-h postlunch glucose value less than 2 mmol·L−1 above premeal levels. Participants were instructed not to engage in any preceding unusual or intense exercise on the days of exercise testing. Exercise tolerance was evaluated for each subject by using an incremental protocol of 15 W·min−1 (women) and 30 W·min−1 (men) after a warm-up period of 2 min performed on an electromagnetically braked cycle ergometer (Corival, Lode, The Netherlands) at a pedaling rate of 50 to 70 rpm. Participants were given strong verbal encouragement to exercise to the highest tolerated symptoms of fatigue or dyspnea. Exercise was terminated at this point and/or when participants were unable to maintain speed or ≥40 revolutions per minute on the ergocycle. Expired air was continuously recorded on a breath-by-breath basis for the determination of V˙O2, carbon dioxide production (V˙CO2), V˙E, and the RER (V˙CO2/V˙O2). The HR was obtained from electrocardiographic monitoring. Blood pressure was measured every 2 min by using an automated sphygmomanometer with a headphone circuit option (Model 412; Quinton Instrument, Bothell, WA). V˙O2peak was defined as the mean V˙O2 recorded in the last 15 s of the incremental exercise protocol concurrent with an RER of 1.15 or greater.
Each participant performed 60 min of aerobic exercise on a cycle ergometer (Organic V3; Bodyguard Fitness, QC, Canada) at 50% of their previously determined V˙O2peak on two occasions and on one occasion with interspersed 10-s sprints at maximal effort every 2 min of the exercise period. Exercise sessions were monitored for HR and workload (W) and BG levels. Each participant performed in random order all three exercise experiments. Each experiment was separated by approximately 1 wk, during which time participants maintained their usual lifestyle and dietary regimen. Participants refrained from physical activity 24 h before every trial, and tests were conducted only in the absence of hypoglycemic episodes in the previous 24 h.
In an outpatient setting, each experiment started at 11:30 a.m. Participants were advised to maintain their routine in terms of meal, snack, and insulin dose. The subject’s BG had to range between 4 and 12 mmol·L−1 to proceed to the experiment. After baseline sampling and 5 min before lunch, the participants injected their premeal insulin into the abdomen. Insulin doses remained identical for each participant in all experiments. A standard lunch provided 8 kcal·kg−1 (50% CHO, 30% lipids and 20% proteins) consisting of vegetarian lasagna, natural almonds, and unsweetened apple sauce and remained identical for each subject in all experiments. The study was performed using a repeated-measures design with three randomly ordered exercise conditions. Participants completed 60 min of moderate-intensity ergocycle exercise at 50% V˙O2peak for all conditions. The three exercise conditions varied as follows: participants ingested 0 g of CHO preexercise (sucaryl in 10% water solution) (0G + MOD), 30 g of CHO preexercise (30G + MOD), or 0 g of CHO preexercise but performed brief high-intensity intervals interspersed every 2 min (0G + MOD/INT) during the exercise trial. The exercise protocol was scheduled 120 min after lunch. The CHO beverage consisted of 8 mg·kg−1·min−1 of exercise of dextrose in 10% water solution given 15 min before the beginning of exercise period. All exercise sessions were followed by a 30-min recovery period.
During the experimental protocols, a catheter (Cathlon Clear, Johnson & Johnson, New Brunswick, NJ) was placed in an antecubital vein for sampling and kept patent by saline drip. Blood samples were collected from time −15 min at 5- to 30-min intervals for 2 h and then every 5 min during and after the exercise period to measure BG concentrations by using a Freestyle glucometer (Abbott, California). Blood samples were centrifuged for plasma and stored at −20°C for later analyses of glucose using a hexokinase method (13). Plasma C-peptide levels were assessed to document insulin secretion defect in this population. They were measured by a modification of the method of Heding (8) using polyclonal antibody (Fitzgerald, Acton, MA) and polyethylene glycol precipitation (13). Another catheter (Cathlon Clear, Johnson & Johnson) was placed in a contralateral antecubital vein for dextrose infusion, if needed. Therefore, a glucose clamp procedure adapted from DeFronzo et al. (2) was followed with a variable infusion of a 20% dextrose solution. If at any time during the testing conditions glucose levels fell below 4 mmol·L−1 or symptomatic hypoglycemia occurred, the dextrose infusion was initiated at a rate based on the drop in BG to maintain BG between 4 and 5 mmol·L−1. Each participant had to note his/her glycemia postexercise for a period of 24 h. Intensity of exercise was estimated by the Borg scale of perceived exertion with scores going from 6 to 20. Area under the curve was calculated for glycemia ≥8 mmol·L−1 from time 120 min (beginning of exercise) until the end of the experiment.
All values are expressed as means ± SE. Wilcoxon test was used to test group differences (P < 0.05). A multiple comparison procedure was performed using a Bonferroni correction to determine specific group differences with overall P values <0.016 considered as significant using JMP 7 of SAS.
Five men and six women participated in this study because both sexes are equally vulnerable to hypoglycemia and could benefit similarly from strategies tested in this study. Mean age was 26.5 ± 6.6 yr, body mass index was 25.8 ± 2.7 kg·m−2, weight was 74.1 ± 6.7 kg, C-peptide levels was 109 ± 121 pmol·L−1, and duration of diabetes was 12.2 ± 5.1 yr. Participants had a mean HbA1C of 7.3% ± 0.4% (range, 6.8–7.8) (Varia II; Bio-Rad, California). They were considered as moderately active as shown by their V˙O2peak values in Table 1. BG was similar between conditions at arrival as demonstrated in Table 2. The preexercise beverage of 30 g of glucose (30G + MOD) resulted in the higher preexercise BG of 11.7 ± 2.7 mmol·L−1 compared with 7.8 ± 4.0 and 9.2 ± 3.5 mmol·L−1, for 0G + MOD and 0G + MOD/INT conditions, respectively, P < 0.05. Glycemia excursions in time are presented in Figure 1. BG was significantly different between conditions with higher values in the 30G + MOD compared with 0G + MOD and 0G + MOD/INT between time 125 and 210 min (all P < 0.02).
The 60-min exercise period induced a BG change of −2.5 ± 3.3, −3.0 ± 2.2, and −3.7 ± 1.6 mmol·L−1 for the 0G + MOD, 0G + MOD/INT, and 30G + MOD, respectively, P = 0.49. The first half of exercise induced a similar BG change between conditions of −1.7 ± 2.0, −1.4 ± 1.2, and −1.1 ± 1.3 mmol·L−1 for the 0G + MOD, 0G + MOD/INT, and 30G + MOD, respectively, P = 0.70. The second half of exercise induced a mean BG change of −0.8 ± 1.8, −1.5 ± 1.6, and −2.6 ± 1.4 mmol·L−1 for 0G + MOD, 0G + MOD/INT, and 30G + MOD, respectively, P = 0.04. The exercise-induced BG drop was also calculated before dextrose infusion, and there was no significant difference with −4.9 ± 2.7, −3.5 ± 1.7, and −3.8 ± 1.5 mmol·L−1 for 0G + MOD, 0G + MOD/INT, and 30G + MOD, all P = 0.38.
The proportion of participants requiring a dextrose infusion was significantly different between sessions with 7/11, 4/11, and 1/11 participants for 0G + MOD, 0G + MOD/INT, and 30G + MOD, respectively, P < 0.02, but with no significant difference between 0G + MOD/INT and 30G + MOD. The duration of dextrose infusion was significantly longer for 0G + MOD condition and shorter for 0G + MOD/INT and 30G + MOD, P < 0.01 in all conditions, but not different between 0G + MOD/INT versus 30G + MOD. As expected, the quantity of dextrose infused was significantly higher for 0G + MOD condition and lower for 0G + MOD/INT and 30G + MOD, P < 0.01 in all conditions, but not different between 0G + MOD/INT versus 30G + MOD.
Mean scores on the Borg scale of perceived exertion was not significantly different between conditions during the exercise periods except at time 165 min (10.9 ± 2.0, 13.8 ± 2.4, and 11.9 ± 2.5) for the 0G + MOD, 0G + MOD/INT, and 30G + MOD conditions, respectively, all P = 0.02.
Area under the curve was calculated for glycemia ≥8 mmol·L−1 from time 120 min (beginning of exercise) until 210 min (end of study period), and there was a significant difference between conditions with mean values of 46.2 ± 94.0, 102.1 ± 149.4, and 192.8 ± 116.7 mmol·L·min−1 for the 0G + MOD, 0G + MOD/INT, and 30G + MOD conditions, respectively, all P = 0.03.
The effects of exercise in participants with T1DM on a basal–bolus regimen of glargine and glulisine are not well documented. Therefore, we examined the effect of three exercise protocols on BG values when a 60-min exercise session is performed, 120 min after lunch, in glargine/glulisine users. We found that intermittent high-intensity exercise and preexercise CHO liquid snack were both strategies that limited BG decrease during and following exercise and therefore offered strategic choices in the prevention of exercise-induced hypoglycemia.
Among the three exercise sessions, it was no surprise to note that the 0G + MOD condition was the modality with the most hypoglycemia and with higher BG decrease, higher proportion of participants needing a dextrose infusion with infusion starting earlier in the exercise period, and consequently higher quantity of dextrose infused. Our results indicated that moderate-intensity exercise, executed for 60 min in the postprandial state, induced a significant BG decrease of about 5 mmol·L−1 and required a large quantity of dextrose in the majority of participants to prevent hypoglycemia. Therefore, patients with T1DM using glargine/glulisine who practiced moderate-intensity physical activity for a long period (60 min) without any nutritional supplement would rapidly face hypoglycemia during and after exercise.
Traditionally, the ADA proposes that approximately 10–15 g of CHO per hour of moderate physical activity would be sufficient for a 70-kg person (1). However, this recommendation has not been linked to any insulin regimen. We have already shown that 40 g of a liquid glucose supplement helped maintain safe BG levels during 60 min of moderate-intensity exercise and recovery in adult participants with T1DM using N-Lispro (4). In the present study, we tested the effect of a 30-g CHO snack taken before moderate-intensity exercise on hypoglycemia prevention in adult participants using glargine/glulisine. Indeed, compared with the placebo snack, ingestion of 30 g of CHO lessened the exercise-induced BG drop, limited markedly the proportion of participants requiring a dextrose infusion, and limited the quantity of dextrose infused as the duration of this infusion. These results suggest that this option prevents hypoglycemia with a mean BG value at the end of exercise of about 8 mmol·L−1. However, the downside of CHO supplementation is preexercise BG hyperglycemia with values approaching 12 mmol·L−1 and a greater exposure to prolonged hyperglycemia with higher area under the curve for glycemia ≥8 mmol·L−1 during and after the exercise. Moreover, this CHO quantity provides extra calories to individuals who are also on average in the overweight category for body mass index. Our results suggest that 30 g of CHO before exercise may have been an excessive quantity to compensate for exercise-induced BG decrease. We figure that a smaller CHO snack, as recommended by ADA, may have resulted in a smaller increase in BG.
Intermittent high-intensity exercise has been proposed to reduce the decline in glycemia during exercise in participants with T1DM because of a greater increment in endogenous glucose production during exercise and attenuated glucose use during exercise and early recovery (6). To our knowledge, such an exercise strategy has not been tested in a basal/bolus insulin regimen of glargine/glulisine. The 0G + MOD/INT condition resulted in a BG decreases intermediate to 0G + MOD and 30G + MOD (Fig. 1). This condition ensured a lower mean BG value at the end of the experiment compared with 30G + MOD, thus keeping the participant in a more optimal BG range. The 0G + INT condition produced similar delay in dextrose infusion, quantity of dextrose infused as well as proportion of participants requiring a dextrose infusion compared with the 30G + MOD. These results suggest that the 0G + INT could be an alternative to the 30G + MOD in terms of preventing hypoglycemia without the downside of preexercise excessive rise in BG. Although beneficial to prevent hypoglycemia, the 0G + INT condition designed with moderate intensity interspersed with several sprints was harder to perform compared with moderate-intensity exercise conditions with or without CHO snack as reflected by the Borg scale scores. Therefore, this condition appears to prevent both exercise-induced hypoglycemia and hyperglycemia before, during, and immediately after exercise but requires a certain level of fitness.
Our results suggest that 60-min exercise of moderate-intensity executed with a 30-g preexercise beverage or interspersed with bouts of intermittent high-intensity may be safe strategies to prevent hypoglycemia in participants using a basal/bolus regimen of glargine/glulisine. Depending on personal preference for preexercise snack or high-intensity sprints, participants with T1DM may practice safe moderate-intensity exercise and choose the option that best fits their lifestyle at a particular time. Further studies are required to better estimate the quantity of glucose supplement or of intense exercise required to prevent exercise-induced hypoglycemia in patients with T1DM.
This study was supported by Sanofi-aventis Canada Inc.
The authors have no conflict of interest to declare.
Results of the present study do not constitute endrosement by the American College of Sports Medicine.
1. Bantle JP, Wylie-Rosett J, Albright AL, et al.. Nutrition recommendations and interventions for diabetes: a position statement of the American Diabetes Association. Diabetes Care
. 2008; 31: (Suppl 1): S61–78.
2. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol
. 1979; 237 (3): E214–23.
3. Diamond MP, Simonson DC, DeFronzo RA. Menstrual cyclicity has a profound effect on glucose homeostasis. Fertil Steril
. 1989; 52 (2): 204–8.
4. Dube MC, Weisnagel SJ, Prud’homme D, Lavoie C. Exercise and newer insulins: how much glucose supplement to avoid hypoglycemia? Med Sci Sports Exerc
. 2005; 37 (8): 1276–82.
5. Ford T, Berg K, Latin R, Thomas L. The effect of exercise intensity on blood glucose in persons with type 1 diabetes. Int Sports J
. 1999; 3: 91–100.
6. Guelfi KJ, Ratnam N, Smythe GA, Jones TW, Fournier PA. Effect of intermittent high-intensity compared with continuous moderate exercise on glucose production and utilization in individuals with type 1 diabetes. Am J Physiol Endocrinol Metab
. 2007; 292 (3): E865–70.
7. Hayes C. Pattern management: a tool for improving blood glucose control with exercise. J Am Diet Assoc
. 1997; 97 (10 Suppl 2): S167–71.
8. Heding LG. Radioimmunological determination of human C-peptide in serum. Diabetologia
. 1975; 11 (6): 541–8.
9. Koivisto VA, Pelkonen R, Nikkila EA, Heding LG. Human and porcine insulins are equally effective in the regulation of glucose kinetics of diabetic patients during exercise. Acta Endocrinol (Copenh)
. 1984; 107 (4): 500–5.
10. McCrimmon RJ, Frier BM. Hypoglycaemia, the most feared complication of insulin therapy. Diabetes Metab
. 1994; 20 (6): 503–12.
11. Mollar-Puchades MA, Villanueva IL. Insulin glulisine in the treatment of allergy to rapid acting insulin and its rapid acting analogs. Diabetes Res Clin Pract
. 2009; 83 (1): e21–2.
12. Rabasa-Lhoret R, Bourque J, Ducros F, Chiasson JL. Guidelines for premeal insulin dose reduction for postprandial exercise of different intensities and durations in type 1 diabetic subjects treated intensively with a basal–bolus insulin regimen (ultralente-lispro). Diabetes Care
. 2001; 24 (4): 625–30.
13. Richterich R, Dauwalder H. Determination of plasma glucose by hexokinase-glucose-6-phosphate dehydrogenase method. Schweiz Med Wochenschr
. 1971; 101 (17): 615–8.
14. Rossetti P, Porcellati F, Fanelli CG, Perriello G, Torlone E, Bolli GB. Superiority of insulin analogues versus human insulin in the treatment of diabetes mellitus. Arch Physiol Biochem
. 2008 Feb; 114 (1): 3–10.
15. Rowland TW, Swadba LA, Biggs DE, Burke EJ, Reiter EO. Glycemic control with physical training in insulin-dependent diabetes mellitus. Am J Dis Child
. 1985; 139 (3): 307–10.
16. Sills IN, Cerny FJ. Responses to continuous and intermittent exercise in healthy and insulin-dependent diabetic children. Med Sci Sports Exerc
. 1983; 15 (6): 450–4.
17. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New Engl J Med
. 1993; 329 (14): 977–86.
18. Trovati M, Anfossi G, Vitali S, Mularoni E, Massucco P, De Facis R, et al.. Prevention of exercise-induced hypoglycaemia in type 1 (insulin-dependent) diabetic patients on conventional intensified insulin therapy: timing of exercise and role of counter-regulatory hormones. Ann Med Interne (Paris)
. 1988; 139 (2): 149–51.
19. Tuominen JA, Karonen SL, Melamies L, Bolli G, Koivisto VA. Exercise-induced hypoglycaemia in IDDM patients treated with a short-acting insulin analogue. Diabetologia
. 1995; 38 (1): 106–11.
20. Zinman B, Ruderman N, Campaigne BN, Devlin JT, Schneider SH. Physical activity/exercise and diabetes. Diabetes Care
. 2004; 27 (Suppl 1): S58–62.