787
ARTICLE
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
Comparison of glucose tolerance tests to detect the insulin sensitizing effects of a bout of continuous exercise Juan Fernando Ortega, Nassim Hamouti, Valentín Emilio Fernández-Elías, and Ricardo Mora-Rodriguez
Abstract: The aim of the present study was to determine which of the available glucose tolerance tests (oral (OGTT) vs. intravenous (IVGTT)) could more readily detect the insulin sensitizing effects of a bout of continuous exercise. Ten healthy moderately fit young men (V˙O2peak of 45.4 ± 1.8 mL·kg−1·min−1; age 27.5 ± 2.7 yr) underwent 4 OGTT and 4 IVGTT on different days following a standardized dinner and overnight fast. One test was performed immediately after 55 min of cycle-ergometer exercise at 60% V˙O2peak. Insulin sensitivity index was determined during a 50 min IVGTT according to Tura (CISI) and from a 120 min OGTT using the Matsuda composite index (MISI). After exercise, MISI improved 29 ± 10% without reaching statistical significance (p = 0.182) due to its low reproducibility (coefficient of variation 16 ± 3%; intra-class reliability 0.846). However, CISI significantly improved (50 ± 4%; p < 0.001) after exercise showing better reproducibility (coefficient of variation 13 ± 4%; intra-class reliability 0.955). Power calculation revealed that 6 participants were required for detecting the effects of exercise on insulin sensitivity when using IVGTT, whereas 54 were needed when using OGTT. The superior response of CISI compared with MISI suggests the preferential use of IVGTT to assess the effects of exercise on insulin sensitivity when using a glucose tolerance test. Key words: endurance exercise, glucose tolerance, insulin sensitivity, oral glucose tolerance test, intravenous glucose tolerance test. Résumé : Cette étude se propose d'identifier lequel des tests de tolérance au glucose (par voie orale (OGTT) ou par voie intraveineuse (IVGTT)) détecte mieux les effets d'une séance d'exercice continu sur la sensibilisation a` l'insuline. Dix jeunes hommes en bonne santé et de condition physique moyenne (V˙O2de pointe : 45,4 ± 1,8 mL·kg−1·min−1; âge : 27,5 ± 2,7 ans) participent en des jours différents a` 4 OGTT et 4 IVGTT a` la suite d'un repas standard et d'une nuit a` jeun. Les sujets participent a` un test immédiatement après 55 min d'exercice sur un cycloergomètre sollicitant 60 % du V˙O2de pointe. On détermine l'index de sensibilité a` l'insuline au cours d'un IVGTT de 50 min conformément a` Tura (CISI) et au cours d'un OGTT de 120 min en utilisant l'index composite de Matsuda (MISI). Après l'exercice, MISI s'améliore de 29 ± 10 %, mais n'atteint pas le niveau de signification statistique (p = 0,182) a` cause de sa faible reproductibilité (CV : 16 ± 3 %; corrélation intraclasse : 0,846). Toutefois, CISI s'améliore significativement (50 ± 4 %; p < 0,001) après l'exercice a` cause de sa meilleure reproductibilité (CV : 13 ± 4 %; corrélation intraclasse : 0,955). Le calcul de la puissance du test révèle qu'il faut 6 participants pour détecter les effets de l'exercice sur la sensibilité a` l'insuline si on utilise l'IVGTT et 54 participants quand on utilise l'OGTT. La meilleure réponse donnée par CISI comparativement a` MISI suggère l'utilisation préférentielle de l'IVGTT pour évaluer les effets de l'exercice sur la sensibilité a` l'insuline quand on utilise un test de tolérance au glucose. Mots-clés : exercice d'endurance, tolérance au glucose, sensibilité a` l'insuline, test de tolérance au glucose par voie orale, test de tolérance au glucose par voie intraveineuse.
Introduction Exercise is recommended for the treatment and prevention of type 2 diabetes (American-Diabetes-Association 2011) because of its insulin-sensitizing effects (Roden 2012; Mikines et al. 1988; Hawley 2004). Although the euglycemic hyperinsulinemic clamp (EHC) is the gold standard method to assess insulin sensitivity, it requires costly and complex equipment and trained hospital personnel to obtain accurate measures and to avoid complications (Bloomgarden 2006; Muniyappa et al. 2008). Insulin sensitivity indexes derived from oral glucose tolerance tests (OGTT) and intravenous glucose tolerance tests (IVGTT) are affordable surrogates of the EHC frequently reported in the literature. However, to our knowledge, data comparing the reliability and response to exercise of these indexes are scarce.
Although the Matsuda composite index (MISI) based on OGTT shows a fairly good agreement with the EHC (r = 0.73; (Matsuda and DeFronzo 1999)) its reproducibility measured by the coefficient of variation (CV) has been estimated between 14.0% and 20.4%; (Utzschneider et al. 2007). Glucose ingestion, in addition to inducing pancreatic insulin release, stimulates gastrointestinal hormones and affects liver metabolism. Furthermore, the ingested glucose would interact with enteric hormones and neural factors capable of altering glucose absorption and insulin secretion (Kahn 1997). We believe that this may be behind the high within-subject variability when using the OGTT technique. On the other hand, it is worth mentioning that the signals initiated from the gastrointestinal system after an oral load of glucose resemble the signals that an individual faces after a meal, justifying the use
Received 31 October 2013. Accepted 9 January 2014. J.F. Ortega, N. Hamouti, V.E. Fernández-Elías, and R. Mora-Rodriguez.* Exercise Physiology Laboratory at Toledo, University of Castilla-La Mancha, Avda. Carlos III, s/n. 45071 Toledo, Spain. Corresponding author: Ricardo Mora-Rodríguez (e-mail: [email protected]). *All editorial decisions for this paper were made by Michael Riddell and Terry Graham. Appl. Physiol. Nutr. Metab. 39: 787–792 (2014) dx.doi.org/10.1139/apnm-2013-0507
Published at www.nrcresearchpress.com/apnm on 13 January 2014.
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
788
of OGTT for an ecological, but less reproducible, assessment of insulin sensitivity. To avoid the variable interaction between the ingested glucose and gastrointestinal hormones and neural factors (Kahn 1997) the glucose load is often administered through an intravenous catheter using IVGTT. A mathematical model (i.e., minimal model (Bergman et al. 1987)) has been developed to calculate insulin sensitivity from a 180 min IVGTT. When using this model researchers report a high correlation with the EHC (r = 0.89; p < 0.05; (Bergman et al. 1987)) and a CV = 14%; (Ferrari et al. 1991)). To reduce testing time and costs, Tura et al. (2010) have recently developed a simpler mathematical model (CISI) to calculate insulin sensitivity from a 50 min IVGTT. They found that CISI was highly correlated with the EHC (r = 0.91; p < 0.05) and the minimal model (r = 0.92; p < 0.05; (Tura et al. 2010)) when tested in a mixed group of healthy and insulin-resistant participants at rest. However, to our knowledge, CISI response to exercise or its withinsubject reproducibility has not been established. Without knowing the response to exercise of each insulin sensitivity index, researchers often default to use the simpler OGTT technique, perhaps limiting the statistical significance of their findings. The purpose of this study was to compare the insulin sensitivity response to exercise from glucose tolerance tests with different routes of glucose administration (i.e., MISI from OGTT vs. CISI from IVGTT). We compared the response to exercise of each of these indices as well as their reproducibility. We hypothesized that insulin sensitivity measures derived from IVGTT would be more reproducible and responsive to the effects of exercise than when derived from OGTT.
Materials and methods Participants Ten young, healthy and physically active males with a mean ± SEM age of 27.5 ± 2.7 years, and a BMI 24.6 ± 0.9 kg·m−2 were recruited from the University’s student and workers population after medical clearance. After detailed information about the benefits and risks involved with their participation in the study, all gave their written consent, which was approved by the local Hospital’s Ethics Committee and conformed to the latest revision of the Declaration of Helsinki. Preliminary testing A week before participation in the study, participants underwent a maximal graded exercise test under medical supervision, until volitional exhaustion according to the American College of Sports Medicine’s guidelines for exercise testing and prescription (American·College·of·Sports·Medicine 2006), starting at 100 W and increasing 25 W·min−1 in an electrically braked cycle-ergometer (Ergoselect 200, Ergoline, Germany). During the exercise test, O2 consumption was measured by indirect calorimetry (Quark b2, Cosmed, Italy) and peak oxygen consumption and cycling peak power output were assessed. Continuous, standard 12-lead ECG (Quark T12, Cosmed, Italy) was performed to ensure a normal cardiovascular response to exercise. Experimental design All participants underwent 8 experimental trials, 4 OGTT and 4 IVGTT, with a separation of at least 72 h among them. One of the OGTT (i.e., OGEX) and one of the IVGTT (i.e., IVEX) were performed after a standardized bout of cycling exercise. Trial order was fully randomized, starting in the morning (0800–1000 h) after an overnight 10 h fast that followed a standardized dinner, provided by the investigators. Dinner consisted of a frozen lasagna, two apples, and an isotonic beverage (703 kcal; 50% from carbohydrate, 38% from fat, 3% from saturated fatty acids, and 12% from protein; fiber content was 8.4 g). During the 24 h prior to each trial, participants were instructed to ingest at least 7 g of carbohydrate per kilogram of body weight to ensure full glycogen storage. This
Appl. Physiol. Nutr. Metab. Vol. 39, 2014
amount of ingested carbohydrates was suggested to avoid the confounding effect of reduced muscle glycogen levels on insulin sensitivity as previously described (Ivy et al. 1985). In addition, all participants were asked to refrain from vigorous exercise 48 h prior to each trial and to ingest 500 mL of bottled water upon awakening to promote euhydration. Experimental trials Upon the participant’s arrival to the laboratory, nude body weight (Hawk, Mettler Toledo, USA) and urine specific gravity (Uricon-NE, Atago, Japan) were measured to assess hydration status. Following this, the participants rested in a stretcher and a 20G intravenous catheter (BD Insyte, Becton Dickinson, Spain) was inserted in an antecubital vein and a Luer-lock 3-way stopcock was attached (Vitroway, Vitromed Healthcare, India). After 15 min of rest in supine position, 5 mL of blood were withdrawn to obtain a baseline fasting sample. Five minutes later, the participants underwent the glucose tolerance test (either OGTT or IVGTT) or started the aerobic exercise bout according to their assigned trial. The glucose tolerance tests were performed approximately 20 min after exercise. We chose not to delay the glucose tolerance tests beyond 20 min of supine rest because prior data using hyperinsulinemic euglycemic clamp demonstrated no difference in insulin sensitivity when assessed either 30 or 180 min after an exercise bout similar in intensity and duration to the exercise presently used (Howlett et al. 2008) in subjects of similar characteristics (i.e., untrained young men). Both exercise trials (OGEX and IVEX) were followed by 20 min of resting supine after which a new baseline blood sample was withdrawn before staring the glucose tolerance test. Exercise duration and intensity resembled the ones used in a study by Mikines et al. (1988) that showed improved insulin sensitivity in participants of similar characteristics (male, young, and healthy). In addition, the chosen exercise bout followed the recommendations of the last American College of Sports Medicine and American Diabetes Association joint statement of exercise for type 2 diabetes treatment and prevention (Colberg et al. 2010). Specifically, exercise commenced with a warm-up (5 min, 100 W) followed by 45 min at 65% V˙O2peak and ended with a 5 min cool-down (75 W). The average workload for the 55 min of exercise was 60% V˙O2peak. Glucose and insulin analysis Samples were mixed in tubes with 3K-EDTA (Vacuette, Greiner Bio-One GmbH, Austria) and centrifuged at 4000 rpm during 10 min at 4 °C (MPW-350R, Med. Instruments, Poland) to obtain plasma that was stored at –80 °C. Glucose was measured with an enzymatic glucose oxidize assay (Enzymatic Glucose Reagent, Thermoscientific, USA). Insulin concentration was analyzed by chemiluminescence (Architect System Insulin, Abbott Diagnostics Division, Germany). OGTT and IVGTT procedures and insulin sensitivity indexes calculation After collection of the basal blood sample, OGTT started with the ingestion of 75 g of anhydrous glucose (Guinama Laboratorio, Spain) diluted into 250 mL of water. Immediately after all fluid was ingested, a timer was started and 5 mL blood samples were obtained after 30, 60, 90, and 120 min of glucose ingestion. For IVGTT we delivered a glucose load of 0.5 g·kg−1 body mass with a maximal dose of 35 g. We used a 30% glucose solution (Glucosada 30%, Grifols, Spain) manually infused at an even pace during 3 min using two 60 mL syringes (BD Plastipak, Spain). Immediately after delivering the glucose load, the stopcock, catheter, and vein were rapidly flushed with 10 mL of 0.9% saline solution. Next, a 5 mL blood sample was obtained and the catheter flushed with 5 mL of 0.9% saline every 10 minutes (i.e., 10, 20, 30, 40, and 50 min). For the IVGTT procedure we strictly complied with the recommendations of the Islet Cell Antibody Register User’s Group (Bingley et al. Published by NRC Research Press
Ortega et al.
789
Table 1. Baseline hydration and glucose metabolism before glucose load delivery and before exercise.
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
1992)) in all aspects, including the use of a single catheter for glucose infusion and sample collection. Basal glucose and insulin concentrations were used to calculate the homeostasis model assessment of insulin resistance (HOMA2-IR) index (Levy et al. 1998) using a specific software (HOMA calculator, Version 2.2.2 Diabetes Trial Unit, University of Oxford, available at www.dtu.ox.ac.uk/homacalculator). Data from the OGTT technique were used to calculate MISI (Matsuda and DeFronzo 1999) as follows: MISI ⫽ 10000
/兹[(FPI IU · mL
⫺1
Body weight (kg) USG FPG (mM·L−1) FPI (IU·mL−1) HOMA2-IR
CISI ⫽ ␣ [KG /(⌬AUCINS /T)] where ␣ is a scaling factor (0.604), KG is the rate of glucose disappearance (slope of log glucose), ⌬AUCINS is the area under the curve of insulin concentration above basal value, and T is the time interval between 10 and 50 min (i.e., 40 min) when glucose and insulin data are computed. Statistics We used Shapiro–Wilk to test the normal distribution of data. In the 3 resting trials, absolute reliability was calculated using CV and relative reliability using intraclass reliability (Vincent 1999). Student’s t test identified insulin sensitivity differences between rest (1 randomly chosen trial) and post-exercise. Nutritional and hydration status stability among trials was evaluated using repeated measures ANOVA comparing pretrial body weight, urine specific gravity, fasting plasma glucose, insulin, and HOMA2-IR among trials. Correlations between MISI and CISI in all variables were sought using Pearson coefficient. Limits of agreement (LOA) were calculated according to Bland and Altman (1986). Sample size estimates for each index were calculated based on the standard deviations of the differences (t test) between rest and post-exercise values using an ␣ level of 0.05 and a 90% power (Utzschneider et al. 2007). Results are reported as means ± SEM. Significance value was set at p < 0.05. All the tests were performed with SPSS for windows (IBM SPSS Statistics for Windows, Version 19.0. IBM Corp., Armonk, NY).
Results Participant characteristics and pretrial hydration and nutritional status Participants V˙O2peak was 45 ± 2 mL·kg−1·min−1 and they reached a peak pedaling power of 300 ± 18 W before exhaustion. Body weight and urinary specific gravity values were not different before any of the 8 trials evidencing a similar pretest hydration status (Table 1). HOMA2-IR index was not different among any trial (Table 1), suggesting compliance with the prescribed diet and exercise refraining. Reproducibility and response to exercise of insulin sensitivity indices At rest, the within-subject reproducibility of insulin sensitivity measured on 3 different days was superior when using CISI from IVGTT than when using MISI from OGTT, showing lower CV and a higher intraclass reliability (Table 2). The HOMA2-IR from fasting glucose and insulin concentrations obtained at rest before the OGTTs had a CV of 15.9% and an intraclass reliability of 0.827.
p value
78.9±0.2 1.016±0.002 5.0±0.1 6.2±0.3 0.79±0.06
0.420 0.416 0.088 0.111 0.101
Note: Values are means ± SEM for 8 trials in each participant (n = 10). p values from repeated measures ANOVA among all 8 trials. (USG, urine specific gravity; FPG, fasting plasma glucose; FPI, fasting plasma insulin; and HOMA2-IR, homeostasis model assessment of insulin resistance).
· FPG mg · dL⫺1) · (PGMEAN · PIMEAN)]
where FPI is the fasting plasma insulin, FPG is the fasting plasma glucose, PGMEAN and PIMEAN are, respectively, the mean of the glucose and insulin concentration of all the samples collected during the 120 min OGTT, and 10000 is the constant to obtain numbers ranging between 1 and 12. Data from the IVGTT technique were used to calculate CISI (Tura et al. 2010) as follows:
Mean
Table 2. Within-subject reproducibility of the three resting trials when using insulin sensitivity indices derived from oral (OGTT) or intravenous (IVGTT) glucose tolerance test. Trial
OGTT (Matusda index)
IVGTT (Tura index)
Rest 1 Rest 2 Rest 3 Coefficient of variation (%) Intraclass reliability
7.5±0.7 6.8±0.8 7.0±0.7 16.4±2.5 0.846
7.4±1.0 7.5±1.0 7.6±1.0 13.4±3.8 0.955
Note: Values are means ± SEM.
HOMA2-IR before the IVGTTs had a CV of 18.6% and an intraclass reliability of 0.793. After moderate exercise (55 min at 60% V˙O2peak), OGTT-derived MISI rose 29% from rest due to a 17% reduction in the product of the mean glucose and insulin concentration and an 11% reduction in the product of fasting plasma glucose and insulin concentration (Table 3). However, the improvement in MISI did not reach statistical significance (Fig. 1A; p = 0.141), likely due to the inconsistent intersubject response to exercise, because 4 out of the 10 participants did not show improvement in MISI (individual responses are dotted lines in Fig. 1A). Post-exercise CISI increased significantly by 50% (Fig. 1B; p < 0.001) mainly by a 27% reduction in the area under the curve of plasma insulin above basal value (p = 0.034, Table 3). The individual response showed that all 10 participants improved their post-exercise CISI (individual responses are dotted lines in Fig. 1B). After exercise HOMA2-IR did not show significant differences from pre-exercise values despite a 15% improvement (0.78 ± 0.07 to 0.67 ± 0.12 from pre- to post-exercise, p = 0.479). Correlations between techniques During resting conditions MISI and CISI showed a high and significant correlation (r = 0.828; p = 0.003; 95% LOA = 4.1 to –3.9). However, the correlation between MISI and CISI scores after exercise was not significant (r = 0.547; p = 0.102; 95% LOA = 10.2 to –6.2). In addition, the changes in the insulin sensitivity from rest to exercise obtained from both indices (MISI and CISI) were calculated from one randomly chosen resting condition trial (OGREST and IVREST) and the exercise trial (OGEX and IVEX). Correlation of those changes with exercise in each of the techniques (OG and IV) did not show statistical significance (r = 0.337; p = 0.341; 95% LOA = 9.4 to –6.1). Sample size calculation Sample size estimates (Utzschneider et al. 2007) revealed that to find the effect of exercise on insulin sensitivity, 54 participants would be required when using MISI (SD = 4.1), whereas only 6 participants when using CISI (SD = 2.1). Published by NRC Research Press
790
Appl. Physiol. Nutr. Metab. Vol. 39, 2014
Table 3. Components of the Matsuda index from oral (OGTT) and Tura index from intravenous (IVGTT) during rest and after exercise.
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
Matsuda index (OGTT) FPI (IU·L−1)·FPG (mg·dL−1) PGMEAN (mg·dL−1)·PIMEAN (UI·mL−1) Tura index (IVGTT) KG (%·min−1) ⌬AUCINS (IU·mL−1)
Rest
Exercise
p value
537±43 4755±525
455±38 3848±475
0.320 0.136
2.9±0.3 1437±347
3.1±0.2 1042±233
0.138 0.034
Note: Values are means ± SEM. (FPI, fasting plasma insulin; FPG, fasting plasma glucose; PGMEAN, mean plasma glucose; PIMEAN, insulin; KG, rate of glucose disappearance; ⌬AUCINS, and area under the curve of insulin concentration above baseline).
Fig. 1. (A) Matsuda index (MISI) from 2 h oral glucose tolerance tests (OGTT) during rest (OGREST) and after exercise (OGEX). (B) Tura’s insulin sensitivity index (CISI) from 50 min intravenous glucose tolerance tests (IVGTT) during rest (IVREST) and after exercise (IVEX). Individual responses shown as dotted lines. Values are means ± SEM. *Difference from rest (p < 0.05).
Discussion It is well established that exercise stimulates insulin-dependent and insulin-independent glucose uptake in human skeletal muscle (Hayashi et al. 1997). The OGTT and the IVGTT are two dynamic methods to assess insulin sensitivity when the euglycemic hyperinsulinemic clamp technique is not affordable. These surrogate insulin sensitivity indexes from glucose tolerance tests are indistinctly reported in the scientific literature without much regard to questions of reproducibility or responsiveness. We tested the reliability of MISI (derived from OGTT) and CISI (derived from IVGTT) at rest during standardized conditions. In addition, we investigated which index more readily detects the insulin sensitizing effects of a single bout of prolonged (55 min) moderately intense exercise (60% V˙O2peak). Unlike most of the literature in this area, we tested reliability and response in the same participants using a cross-over repeated measures experimental design.
A major finding of this study is that CISI derived from IVGTT showed the highest reproducibility (Table 2). Because of this higher reproducibility and consistency (i.e., every subject improved) we detected the improvements in insulin sensitivity with exercise only with CISI (p < 0.001; Fig. 1B). In contrast, MISI was not responsive enough to significantly detect the effects of exercise on insulin sensitivity (Fig. 1A). To ascertain the effects of exercise using MISI another 44 similar participants would need to be tested. The extra cost and effort of testing that number of participants may not be justified and out of the scope of many research laboratories with restricted budgets. On the other hand, CISI is calculated after a 50 min IVGTT, whereas MISI requires a 120 min OGTT. The reduced time of study is another advantage when using CISI instead of MISI. Thus, our data support the preferential use of the IVGTT technique in comparison with OGTT when evaluating the effects of and exercise bout (or exercise program) on insulin sensitivity. Although most studies show a positive effect of exercise on insulin sensitivity when assessed by OGTT (Balducci et al. 2010; Dumortier et al. 2003; Jenkins and Hagberg 2011; Kishimoto et al. 2002; Venables and Jeukendrup 2008), some studies report no effects. Hansen et al. (2009) found that for obese type 2 diabetes patients, exercise training programs of different intensities improved glycated hemoglobin (HbA1C) and whole body oxidative capacity, whereas the insulin area under the curve measured during OGTT remained unchanged. Mitchell et al. (2011) measured insulin sensitivity using OGTT in obese participants after aerobic exercise and found no improvements. Niakaris et al. (2005) recommended caution in the interpretation of OGTT-derived insulin sensitivity results, because they seem to be population dependent. Our data suggested that the discrepancies about the effects of exercise on OGTT-derived insulin sensitivity calculations originate from the low reproducibility of this technique. In addition, the response was inconsistent, because 4 out of the 10 participants did not improve MISI after exercise. MISI does not only reflect peripheral insulin sensitivity but also hepatic insulin sensitivity (Matsuda and DeFronzo 1999). As exercise is mostly known to affect peripheral insulin sensitivity, it is possible that MISI was not that sensitive to the effects of exercise because a portion of the calculation is devoted to hepatic insulin sensitivity. The fasting glucose and insulin concentration portion of the MISI formulae (see Materials and methods) reflects hepatic insulin sensitivity. When we removed that portion of the formula and only the glucose and insulin concentrations during the OGTT were compared, no significant differences were found between the exercise and rest trials (p = 0.136, Table 3). We interpret this to indicate that the lack of improvement in MISI after exercise is also due to a variable response in peripheral insulin sensitivity. It is well established that one bout of aerobic exercise improves glycemic control and insulin sensitivity (Kennedy et al. 1999; Perseghin et al. 1996). In a recent study, healthy nonobese young men were tested 20 min after a bout of 30 min of continuous pedaling at 50% of their V˙O2max. The authors reported a 45% improvement in insulin sensitivity when using frequently sampled IVGTT (Hayashi et al. 2005). We used a similar energy demanding bout of exercise (i.e., 55 min of exercise at 60% V˙O2peak) and insulin sensitivity improved 50% with CISI (p < 0.001; Fig. 1B). Our results using the 50 min IVGTT technique described by Tura et al. (2010), agree in the magnitude of insulin sensitivity improvement with previous studies (Hayashi et al. 2005). Our study design had several limitations. First we favored the use of IVGTT instead of OGTT. However, it is necessary to note that OGTT better resembles the physiological challenge that an individual faces after a meal, which justifies the use of OGTT especially in large epidemiological studies. Second, we used healthy participants with high baseline insulin sensitivity values very similar to the ones found by Tura et al. (2010) in a similar population (7.5 vs. 5.55 ± 0.25 × 10−4 min−1 [U·mL]−1). However, there is an inverse relationship between baseline insulin sensitivity and inPublished by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
Ortega et al.
sulin sensitivity improvements with exercise (Magkos et al. 2008). So it is possible that if a group of overweight or insulin-resistant participants were to perform this study, the increase in insulin sensitivity with exercise would have been more robust and statistically significant, even when using MISI. Finally, the tolerance tests after exercise were conducted at a time of day that was 1 h after the time of day for the tests in resting condition. Although circadian rhythm could affect insulin sensitivity between morning and afternoon trials (Boden et al. 1996), it is unlikely that circadian rhythm could be a factor when trials are spaced only by one hour (0800 and 0900 h). In addition, this difference in the time at the start of the tests should have not clouded the results of our main aim, which was the comparison of 2 insulin sensitivity indexes with different modes of glucose load delivery (i.e., MISI vs. CISI), considering that the difference in time between the rest and post-exercise tests was the same for MISI and CISI. HOMA2-IR (Levy et al. 1998) stands out as the more cost- and time-efficient insulin sensitivity index, as it requires just fasting glucose and insulin concentrations. However, HOMA2-IR was unable to detect the insulin sensitivity changes induced by the 55 min bout of exercise and it was less reproducible than CISI. Thus, according to our data, HOMA2-IR is not a recommended technique to measure the insulin sensitizing effects of a bout of exercise. In fact, other authors have argued that HOMA-IR is not indicated to assess insulin sensitivity under nonsteady state conditions (Wallace et al. 2004) such as the transitional improvements induced by exercise. In conclusion, an exercise prescription to improve carbohydrate metabolism requires a reliable and responsive test to gauge the effect of exercise on insulin sensitivity. We compared two glucose tolerance tests that could be used for insulin sensitivity assessment out of hospital boundaries. We found improved responsiveness and reproducibility when the glucose load is delivered through an intravenous catheter (CISI, IVGTT derived) in comparison with oral ingestion (MISI, OGTT derived). Using 10 untrained, nondiabetic participants only the insulin sensitivity index based on IVGTT was fit to detect the acute effects of a bout of prolonged aerobic exercise. In most exercise science research environments, where the use of a euglycemic hyperinsulinemic clamp is not possible and sample size is limited, CISI based on 50 min IVGTT is a superior index to gauge the acute effects of exercise.
Acknowledgements We would like to thank the participants for their uninterested collaboration in this project. The help of Alvaro López and Carolina Cruces is greatly appreciated. All authors read and approved the final manuscript. This work was partially supported by a grant from the Spanish Ministry of Economia y Competitividad (DEP201128615). The authors report no conflicts of interest.
References American Diabetes Association. 2011. Standards of medical care in diabetes-2011. Diabetes Care, 34(Suppl. 1): S11–S61. doi:10.2337/dc11-S011. PMID:21193625. American·College·of·Sports·Medicine. 2006. ACSM’s guidelines for exercise testing and prescription. Lippincott Williams & Wilkins, Baltimore, Maryland. Balducci, S., Zanuso, S., Nicolucci, A., De Feo, P., Cavallo, S., Cardelli, P., Fallucca, S., Alessi, E., Fallucca, F., and Pugliese, G. 2010. Effect of an intensive exercise intervention strategy on modifiable cardiovascular risk factors in subjects with type 2 diabetes mellitus: a randomized controlled trial: the Italian Diabetes and Exercise Study (IDES). Arch. Intern. Med. 170(20): 1794– 1803. doi:10.1001/archinternmed.2010.380. PMID:21059972. Bergman, R.N., Prager, R., Volund, A., and Olefsky, J.M. 1987. Equivalence of the insulin sensitivity index in man derived by the minimal model method and the euglycemic glucose clamp. J. Clin. Invest. 79(3): 790–800. doi:10.1172/ JCI112886. PMID:3546379. Bingley, P.J., Colman, P., Eisenbarth, G.S., Jackson, R.A., McCulloch, D.K., Riley, W.J., and Gale, E.A. 1992. Standardization of IVGTT to predict IDDM. Diabetes Care, 15(10): 1313–1316. doi:10.2337/diacare.15.10.1313. PMID:1425095. Bland, J.M., and Altman, D.G. 1986. Statistical methods for assessing agreement
791
between two methods of clinical measurement. Lancet, 1(8476): 307–310. doi:10.1016%2Fj.ijnurstu.2009.10.001. PMID:2868172. Bloomgarden, Z.T. 2006. Measures of insulin sensitivity. Clin. Lab. Med. 26(3): 611–633, vi. doi:10.1016/j.cll.2006.06.007. PMID:16938587. Boden, G., Chen, X., and Urbain, J.L. 1996. Evidence for a circadian rhythm of insulin sensitivity in patients with NIDDM caused by cyclic changes in hepatic glucose production. Diabetes, 45(8): 1044–1050. doi:10.2337/diab.45.8. 1044. PMID:8690150. Colberg, S.R., Albright, A.L., Blissmer, B.J., Braun, B., Chasan-Taber, L., Fernhall, B., Regensteiner, J.G., Rubin, R.R., and Sigal, R.J. 2010. Exercise and type 2 diabetes: American College of Sports Medicine and the American Diabetes Association: joint position statement. Exercise and type 2 diabetes. Med. Sci. Sports Exerc. 42(12): 2282–2303. doi:10.1249/MSS.0b013e3181eeb61c. Dumortier, M., Brandou, F., Perez-Martin, A., Fedou, C., Mercier, J., and Brun, J.F. 2003. Low intensity endurance exercise targeted for lipid oxidation improves body composition and insulin sensitivity in patients with the metabolic syndrome. Diabetes Metab. 29(5): 509–518. doi:10.1016/S1262-3636(07)70065-4. PMID:14631328. Ferrari, P., Alleman, Y., Shaw, S., Riesen, W., and Weidmann, P. 1991. Reproducibility of insulin sensitivity measured by the minimal model method. Diabetologia, 34(7): 527–530. doi:10.1007/BF00403291. PMID:1916059. Hansen, D., Dendale, P., Jonkers, R.A., Beelen, M., Manders, R.J., Corluy, L., Mullens, A., Berger, J., Meeusen, R., and van Loon, L.J. 2009. Continuous lowto moderate-intensity exercise training is as effective as moderate- to highintensity exercise training at lowering blood HbA(1c) in obese type 2 diabetes patients. Diabetologia, 52(9): 1789–1797. doi:10.1007/s00125-009-1354-3. PMID: 19370339. Hawley, J.A. 2004. Exercise as a therapeutic intervention for the prevention and treatment of insulin resistance. Diabetes Metab. Res. Rev. 20(5): 383–393. doi:10.1002/dmrr.505. PMID:15343584. Hayashi, T., Wojtaszewski, J.F., and Goodyear, L.J. 1997. Exercise regulation of glucose transport in skeletal muscle. Am. J. Physiol. 273(6 Pt 1): E1039–E1051. Hayashi, Y., Nagasaka, S., Takahashi, N., Kusaka, I., Ishibashi, S., Numao, S., Lee, D.J., Taki, Y., Ogata, H., Tokuyama, K., and Tanaka, K. 2005. A single bout of exercise at higher intensity enhances glucose effectiveness in sedentary men. J. Clin. Endocrinol. Metab. 90(7): 4035–4040. doi:10.1210/jc.2004-2092. PMID:15840752. Howlett, K.F., Mathews, A., Garnham, A., and Sakamoto, K. 2008. The effect of exercise and insulin on AS160 phosphorylation and 14-3-3 binding capacity in human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 294(2): E401–E407. doi:10.1152%2Fajpendo.00542.2007. PMID:18042670. Ivy, J.L., Frishberg, B.A., Farrell, S.W., Miller, W.J., and Sherman, W.M. 1985. Effects of elevated and exercise-reduced muscle glycogen levels on insulin sensitivity. J. Appl. Physiol. 59(1): 154–159. PMID:4030559. Jenkins, N.T., and Hagberg, J.M. 2011. Aerobic training effects on glucose tolerance in prediabetic and normoglycemic humans. Med. Sci. Sports Exerc. 43(12): 2231–2240. doi:10.1249/MSS.0b013e318223b5f9. PMID:21606871. Kennedy, J.W., Hirshman, M.F., Gervino, E.V., Ocel, J.V., Forse, R.A., Hoenig, S.J., Aronson, D., Goodyear, L.J., and Horton, E.S. 1999. Acute exercise induces GLUT4 translocation in skeletal muscle of normal human subjects and subjects with type 2 diabetes. Diabetes, 48(5): 1192–1197. doi:10.2337/diabetes.48. 5.1192. PMID:10331428. Kishimoto, H., Taniguchi, A., Fukushima, M., Sakai, M., Tokuyama, K., Oguma, T., Nin, K., Nagata, I., Hayashi, R., Kawano, M., Hayashi, K., Tsukamoto, Y., Okumura, T., Nagasaka, S., Mizutani, H., and Nakai, Y. 2002. Effect of short-term low-intensity exercise on insulin sensitivity, insulin secretion, and glucose and lipid metabolism in non-obese Japanese type 2 diabetic patients. Horm. Metab. Res. 34(1): 27–31. doi:10.1055/s-2002-19963. PMID:11832998. Levy, J.C., Matthews, D.R., and Hermans, M.P. 1998. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care, 21(12): 2191–2192. doi:10.2337/diacare.21.12.2191. PMID:9839117. Magkos, F., Tsekouras, Y., Kavouras, S.A., Mittendorfer, B., and Sidossis, L.S. 2008. Improved insulin sensitivity after a single bout of exercise is curvilinearly related to exercise energy expenditure. Clin. Sci. (Lond.), 114(1): 59–64. doi:10.1042/CS20070134. Matsuda, M., and DeFronzo, R.A. 1999. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care, 22(9): 1462–1470. doi:10.2337/diacare.22.9.1462. PMID:10480510. Mikines, K.J., Sonne, B., Farrell, P.A., Tronier, B., and Galbo, H. 1988. Effect of physical exercise on sensitivity and responsiveness to insulin in humans. Am. J. Physiol. 254(3 Pt 1): E248–E259. PMID:3126668. Mitchell, J.B., Phillips, M.D., Yellott, R.C., and Currie, L.M. 2011. Resistance and aerobic exercise: the influence of mode on the relationship between IL-6 and glucose tolerance in young men who are obese. J. Strength Cond. Res. 25(6): 1529–1537. doi:10.1519/JSC.0b013e3182176638. PMID:21487312. Muniyappa, R., Lee, S., Chen, H., and Quon, M.J. 2008. Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am. J. Physiol. Endocrinol. Metab. 294(1): E15–E26. doi:10. 1152%2Fajpendo.00645.2007. PMID:17957034. Niakaris, K., Magkos, F., Geladas, N., and Sidossis, L.S. 2005. Insulin sensitivity derived from oral glucose tolerance testing in athletes: disagreement between Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
792
available indices. J. Sports Sci. 23(10): 1065–1073. doi:10.1080/02640410400023241. PMID:16194983. Perseghin, G., Price, T.B., Petersen, K.F., Roden, M., Cline, G.W., Gerow, K., Rothman, D.L., and Shulman, G.I. 1996. Increased glucose transportphosphorylation and muscle glycogen synthesis after exercise training in insulin-resistant subjects. N. Engl. J. Med. 335(18): 1357–1362. doi:10.1056/ NEJM199610313351804. PMID:8857019. Roden, M. 2012. Exercise in type 2 diabetes: to resist or to endure? Diabetologia, 55(5): 1235–1239. doi:10.1007/s00125-012-2513-5. PMID:22391950. Tura, A., Sbrignadello, S., Succurro, E., Groop, L., Sesti, G., and Pacini, G. 2010. An empirical index of insulin sensitivity from short IVGTT: validation against the minimal model and glucose clamp indices in patients with different
Appl. Physiol. Nutr. Metab. Vol. 39, 2014
clinical characteristics. Diabetologia, 53(1): 144–152. doi:10.1007/s00125-0091547-9. PMID:19876614. Utzschneider, K.M., Prigeon, R.L., Tong, J., Gerchman, F., Carr, D.B., Zraika, S., Udayasankar, J., Montgomery, B., Mari, A., and Kahn, S.E. 2007. Withinsubject variability of measures of beta cell function derived from a 2 h OGTT: implications for research studies. Diabetologia, 50(12): 2516–2525. doi:10.1007/ s00125-007-0819-5. PMID:17928990. Venables, M.C., and Jeukendrup, A.E. 2008. Endurance training and obesity: effect on substrate metabolism and insulin sensitivity. Med. Sci. Sports Exerc. 40(3): 495–502. doi:10.1249/MSS.0b013e31815f256f. PMID:18379212. Vincent, W. 1999. Statistics in kinesiology. Human Kinetics, Champaign. Wallace, T.M., Levy, J.C., and Matthews, D.R. 2004. Use and abuse of HOMA modeling. Diabetes Care, 27(6): 1487–1495. doi:10.2337/diacare.27.6.1487. PMID:15161807.
Published by NRC Research Press
ARTICLE
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
Comparison of glucose tolerance tests to detect the insulin sensitizing effects of a bout of continuous exercise Juan Fernando Ortega, Nassim Hamouti, Valentín Emilio Fernández-Elías, and Ricardo Mora-Rodriguez
Abstract: The aim of the present study was to determine which of the available glucose tolerance tests (oral (OGTT) vs. intravenous (IVGTT)) could more readily detect the insulin sensitizing effects of a bout of continuous exercise. Ten healthy moderately fit young men (V˙O2peak of 45.4 ± 1.8 mL·kg−1·min−1; age 27.5 ± 2.7 yr) underwent 4 OGTT and 4 IVGTT on different days following a standardized dinner and overnight fast. One test was performed immediately after 55 min of cycle-ergometer exercise at 60% V˙O2peak. Insulin sensitivity index was determined during a 50 min IVGTT according to Tura (CISI) and from a 120 min OGTT using the Matsuda composite index (MISI). After exercise, MISI improved 29 ± 10% without reaching statistical significance (p = 0.182) due to its low reproducibility (coefficient of variation 16 ± 3%; intra-class reliability 0.846). However, CISI significantly improved (50 ± 4%; p < 0.001) after exercise showing better reproducibility (coefficient of variation 13 ± 4%; intra-class reliability 0.955). Power calculation revealed that 6 participants were required for detecting the effects of exercise on insulin sensitivity when using IVGTT, whereas 54 were needed when using OGTT. The superior response of CISI compared with MISI suggests the preferential use of IVGTT to assess the effects of exercise on insulin sensitivity when using a glucose tolerance test. Key words: endurance exercise, glucose tolerance, insulin sensitivity, oral glucose tolerance test, intravenous glucose tolerance test. Résumé : Cette étude se propose d'identifier lequel des tests de tolérance au glucose (par voie orale (OGTT) ou par voie intraveineuse (IVGTT)) détecte mieux les effets d'une séance d'exercice continu sur la sensibilisation a` l'insuline. Dix jeunes hommes en bonne santé et de condition physique moyenne (V˙O2de pointe : 45,4 ± 1,8 mL·kg−1·min−1; âge : 27,5 ± 2,7 ans) participent en des jours différents a` 4 OGTT et 4 IVGTT a` la suite d'un repas standard et d'une nuit a` jeun. Les sujets participent a` un test immédiatement après 55 min d'exercice sur un cycloergomètre sollicitant 60 % du V˙O2de pointe. On détermine l'index de sensibilité a` l'insuline au cours d'un IVGTT de 50 min conformément a` Tura (CISI) et au cours d'un OGTT de 120 min en utilisant l'index composite de Matsuda (MISI). Après l'exercice, MISI s'améliore de 29 ± 10 %, mais n'atteint pas le niveau de signification statistique (p = 0,182) a` cause de sa faible reproductibilité (CV : 16 ± 3 %; corrélation intraclasse : 0,846). Toutefois, CISI s'améliore significativement (50 ± 4 %; p < 0,001) après l'exercice a` cause de sa meilleure reproductibilité (CV : 13 ± 4 %; corrélation intraclasse : 0,955). Le calcul de la puissance du test révèle qu'il faut 6 participants pour détecter les effets de l'exercice sur la sensibilité a` l'insuline si on utilise l'IVGTT et 54 participants quand on utilise l'OGTT. La meilleure réponse donnée par CISI comparativement a` MISI suggère l'utilisation préférentielle de l'IVGTT pour évaluer les effets de l'exercice sur la sensibilité a` l'insuline quand on utilise un test de tolérance au glucose. Mots-clés : exercice d'endurance, tolérance au glucose, sensibilité a` l'insuline, test de tolérance au glucose par voie orale, test de tolérance au glucose par voie intraveineuse.
Introduction Exercise is recommended for the treatment and prevention of type 2 diabetes (American-Diabetes-Association 2011) because of its insulin-sensitizing effects (Roden 2012; Mikines et al. 1988; Hawley 2004). Although the euglycemic hyperinsulinemic clamp (EHC) is the gold standard method to assess insulin sensitivity, it requires costly and complex equipment and trained hospital personnel to obtain accurate measures and to avoid complications (Bloomgarden 2006; Muniyappa et al. 2008). Insulin sensitivity indexes derived from oral glucose tolerance tests (OGTT) and intravenous glucose tolerance tests (IVGTT) are affordable surrogates of the EHC frequently reported in the literature. However, to our knowledge, data comparing the reliability and response to exercise of these indexes are scarce.
Although the Matsuda composite index (MISI) based on OGTT shows a fairly good agreement with the EHC (r = 0.73; (Matsuda and DeFronzo 1999)) its reproducibility measured by the coefficient of variation (CV) has been estimated between 14.0% and 20.4%; (Utzschneider et al. 2007). Glucose ingestion, in addition to inducing pancreatic insulin release, stimulates gastrointestinal hormones and affects liver metabolism. Furthermore, the ingested glucose would interact with enteric hormones and neural factors capable of altering glucose absorption and insulin secretion (Kahn 1997). We believe that this may be behind the high within-subject variability when using the OGTT technique. On the other hand, it is worth mentioning that the signals initiated from the gastrointestinal system after an oral load of glucose resemble the signals that an individual faces after a meal, justifying the use
Received 31 October 2013. Accepted 9 January 2014. J.F. Ortega, N. Hamouti, V.E. Fernández-Elías, and R. Mora-Rodriguez.* Exercise Physiology Laboratory at Toledo, University of Castilla-La Mancha, Avda. Carlos III, s/n. 45071 Toledo, Spain. Corresponding author: Ricardo Mora-Rodríguez (e-mail: [email protected]). *All editorial decisions for this paper were made by Michael Riddell and Terry Graham. Appl. Physiol. Nutr. Metab. 39: 787–792 (2014) dx.doi.org/10.1139/apnm-2013-0507
Published at www.nrcresearchpress.com/apnm on 13 January 2014.
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
788
of OGTT for an ecological, but less reproducible, assessment of insulin sensitivity. To avoid the variable interaction between the ingested glucose and gastrointestinal hormones and neural factors (Kahn 1997) the glucose load is often administered through an intravenous catheter using IVGTT. A mathematical model (i.e., minimal model (Bergman et al. 1987)) has been developed to calculate insulin sensitivity from a 180 min IVGTT. When using this model researchers report a high correlation with the EHC (r = 0.89; p < 0.05; (Bergman et al. 1987)) and a CV = 14%; (Ferrari et al. 1991)). To reduce testing time and costs, Tura et al. (2010) have recently developed a simpler mathematical model (CISI) to calculate insulin sensitivity from a 50 min IVGTT. They found that CISI was highly correlated with the EHC (r = 0.91; p < 0.05) and the minimal model (r = 0.92; p < 0.05; (Tura et al. 2010)) when tested in a mixed group of healthy and insulin-resistant participants at rest. However, to our knowledge, CISI response to exercise or its withinsubject reproducibility has not been established. Without knowing the response to exercise of each insulin sensitivity index, researchers often default to use the simpler OGTT technique, perhaps limiting the statistical significance of their findings. The purpose of this study was to compare the insulin sensitivity response to exercise from glucose tolerance tests with different routes of glucose administration (i.e., MISI from OGTT vs. CISI from IVGTT). We compared the response to exercise of each of these indices as well as their reproducibility. We hypothesized that insulin sensitivity measures derived from IVGTT would be more reproducible and responsive to the effects of exercise than when derived from OGTT.
Materials and methods Participants Ten young, healthy and physically active males with a mean ± SEM age of 27.5 ± 2.7 years, and a BMI 24.6 ± 0.9 kg·m−2 were recruited from the University’s student and workers population after medical clearance. After detailed information about the benefits and risks involved with their participation in the study, all gave their written consent, which was approved by the local Hospital’s Ethics Committee and conformed to the latest revision of the Declaration of Helsinki. Preliminary testing A week before participation in the study, participants underwent a maximal graded exercise test under medical supervision, until volitional exhaustion according to the American College of Sports Medicine’s guidelines for exercise testing and prescription (American·College·of·Sports·Medicine 2006), starting at 100 W and increasing 25 W·min−1 in an electrically braked cycle-ergometer (Ergoselect 200, Ergoline, Germany). During the exercise test, O2 consumption was measured by indirect calorimetry (Quark b2, Cosmed, Italy) and peak oxygen consumption and cycling peak power output were assessed. Continuous, standard 12-lead ECG (Quark T12, Cosmed, Italy) was performed to ensure a normal cardiovascular response to exercise. Experimental design All participants underwent 8 experimental trials, 4 OGTT and 4 IVGTT, with a separation of at least 72 h among them. One of the OGTT (i.e., OGEX) and one of the IVGTT (i.e., IVEX) were performed after a standardized bout of cycling exercise. Trial order was fully randomized, starting in the morning (0800–1000 h) after an overnight 10 h fast that followed a standardized dinner, provided by the investigators. Dinner consisted of a frozen lasagna, two apples, and an isotonic beverage (703 kcal; 50% from carbohydrate, 38% from fat, 3% from saturated fatty acids, and 12% from protein; fiber content was 8.4 g). During the 24 h prior to each trial, participants were instructed to ingest at least 7 g of carbohydrate per kilogram of body weight to ensure full glycogen storage. This
Appl. Physiol. Nutr. Metab. Vol. 39, 2014
amount of ingested carbohydrates was suggested to avoid the confounding effect of reduced muscle glycogen levels on insulin sensitivity as previously described (Ivy et al. 1985). In addition, all participants were asked to refrain from vigorous exercise 48 h prior to each trial and to ingest 500 mL of bottled water upon awakening to promote euhydration. Experimental trials Upon the participant’s arrival to the laboratory, nude body weight (Hawk, Mettler Toledo, USA) and urine specific gravity (Uricon-NE, Atago, Japan) were measured to assess hydration status. Following this, the participants rested in a stretcher and a 20G intravenous catheter (BD Insyte, Becton Dickinson, Spain) was inserted in an antecubital vein and a Luer-lock 3-way stopcock was attached (Vitroway, Vitromed Healthcare, India). After 15 min of rest in supine position, 5 mL of blood were withdrawn to obtain a baseline fasting sample. Five minutes later, the participants underwent the glucose tolerance test (either OGTT or IVGTT) or started the aerobic exercise bout according to their assigned trial. The glucose tolerance tests were performed approximately 20 min after exercise. We chose not to delay the glucose tolerance tests beyond 20 min of supine rest because prior data using hyperinsulinemic euglycemic clamp demonstrated no difference in insulin sensitivity when assessed either 30 or 180 min after an exercise bout similar in intensity and duration to the exercise presently used (Howlett et al. 2008) in subjects of similar characteristics (i.e., untrained young men). Both exercise trials (OGEX and IVEX) were followed by 20 min of resting supine after which a new baseline blood sample was withdrawn before staring the glucose tolerance test. Exercise duration and intensity resembled the ones used in a study by Mikines et al. (1988) that showed improved insulin sensitivity in participants of similar characteristics (male, young, and healthy). In addition, the chosen exercise bout followed the recommendations of the last American College of Sports Medicine and American Diabetes Association joint statement of exercise for type 2 diabetes treatment and prevention (Colberg et al. 2010). Specifically, exercise commenced with a warm-up (5 min, 100 W) followed by 45 min at 65% V˙O2peak and ended with a 5 min cool-down (75 W). The average workload for the 55 min of exercise was 60% V˙O2peak. Glucose and insulin analysis Samples were mixed in tubes with 3K-EDTA (Vacuette, Greiner Bio-One GmbH, Austria) and centrifuged at 4000 rpm during 10 min at 4 °C (MPW-350R, Med. Instruments, Poland) to obtain plasma that was stored at –80 °C. Glucose was measured with an enzymatic glucose oxidize assay (Enzymatic Glucose Reagent, Thermoscientific, USA). Insulin concentration was analyzed by chemiluminescence (Architect System Insulin, Abbott Diagnostics Division, Germany). OGTT and IVGTT procedures and insulin sensitivity indexes calculation After collection of the basal blood sample, OGTT started with the ingestion of 75 g of anhydrous glucose (Guinama Laboratorio, Spain) diluted into 250 mL of water. Immediately after all fluid was ingested, a timer was started and 5 mL blood samples were obtained after 30, 60, 90, and 120 min of glucose ingestion. For IVGTT we delivered a glucose load of 0.5 g·kg−1 body mass with a maximal dose of 35 g. We used a 30% glucose solution (Glucosada 30%, Grifols, Spain) manually infused at an even pace during 3 min using two 60 mL syringes (BD Plastipak, Spain). Immediately after delivering the glucose load, the stopcock, catheter, and vein were rapidly flushed with 10 mL of 0.9% saline solution. Next, a 5 mL blood sample was obtained and the catheter flushed with 5 mL of 0.9% saline every 10 minutes (i.e., 10, 20, 30, 40, and 50 min). For the IVGTT procedure we strictly complied with the recommendations of the Islet Cell Antibody Register User’s Group (Bingley et al. Published by NRC Research Press
Ortega et al.
789
Table 1. Baseline hydration and glucose metabolism before glucose load delivery and before exercise.
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
1992)) in all aspects, including the use of a single catheter for glucose infusion and sample collection. Basal glucose and insulin concentrations were used to calculate the homeostasis model assessment of insulin resistance (HOMA2-IR) index (Levy et al. 1998) using a specific software (HOMA calculator, Version 2.2.2 Diabetes Trial Unit, University of Oxford, available at www.dtu.ox.ac.uk/homacalculator). Data from the OGTT technique were used to calculate MISI (Matsuda and DeFronzo 1999) as follows: MISI ⫽ 10000
/兹[(FPI IU · mL
⫺1
Body weight (kg) USG FPG (mM·L−1) FPI (IU·mL−1) HOMA2-IR
CISI ⫽ ␣ [KG /(⌬AUCINS /T)] where ␣ is a scaling factor (0.604), KG is the rate of glucose disappearance (slope of log glucose), ⌬AUCINS is the area under the curve of insulin concentration above basal value, and T is the time interval between 10 and 50 min (i.e., 40 min) when glucose and insulin data are computed. Statistics We used Shapiro–Wilk to test the normal distribution of data. In the 3 resting trials, absolute reliability was calculated using CV and relative reliability using intraclass reliability (Vincent 1999). Student’s t test identified insulin sensitivity differences between rest (1 randomly chosen trial) and post-exercise. Nutritional and hydration status stability among trials was evaluated using repeated measures ANOVA comparing pretrial body weight, urine specific gravity, fasting plasma glucose, insulin, and HOMA2-IR among trials. Correlations between MISI and CISI in all variables were sought using Pearson coefficient. Limits of agreement (LOA) were calculated according to Bland and Altman (1986). Sample size estimates for each index were calculated based on the standard deviations of the differences (t test) between rest and post-exercise values using an ␣ level of 0.05 and a 90% power (Utzschneider et al. 2007). Results are reported as means ± SEM. Significance value was set at p < 0.05. All the tests were performed with SPSS for windows (IBM SPSS Statistics for Windows, Version 19.0. IBM Corp., Armonk, NY).
Results Participant characteristics and pretrial hydration and nutritional status Participants V˙O2peak was 45 ± 2 mL·kg−1·min−1 and they reached a peak pedaling power of 300 ± 18 W before exhaustion. Body weight and urinary specific gravity values were not different before any of the 8 trials evidencing a similar pretest hydration status (Table 1). HOMA2-IR index was not different among any trial (Table 1), suggesting compliance with the prescribed diet and exercise refraining. Reproducibility and response to exercise of insulin sensitivity indices At rest, the within-subject reproducibility of insulin sensitivity measured on 3 different days was superior when using CISI from IVGTT than when using MISI from OGTT, showing lower CV and a higher intraclass reliability (Table 2). The HOMA2-IR from fasting glucose and insulin concentrations obtained at rest before the OGTTs had a CV of 15.9% and an intraclass reliability of 0.827.
p value
78.9±0.2 1.016±0.002 5.0±0.1 6.2±0.3 0.79±0.06
0.420 0.416 0.088 0.111 0.101
Note: Values are means ± SEM for 8 trials in each participant (n = 10). p values from repeated measures ANOVA among all 8 trials. (USG, urine specific gravity; FPG, fasting plasma glucose; FPI, fasting plasma insulin; and HOMA2-IR, homeostasis model assessment of insulin resistance).
· FPG mg · dL⫺1) · (PGMEAN · PIMEAN)]
where FPI is the fasting plasma insulin, FPG is the fasting plasma glucose, PGMEAN and PIMEAN are, respectively, the mean of the glucose and insulin concentration of all the samples collected during the 120 min OGTT, and 10000 is the constant to obtain numbers ranging between 1 and 12. Data from the IVGTT technique were used to calculate CISI (Tura et al. 2010) as follows:
Mean
Table 2. Within-subject reproducibility of the three resting trials when using insulin sensitivity indices derived from oral (OGTT) or intravenous (IVGTT) glucose tolerance test. Trial
OGTT (Matusda index)
IVGTT (Tura index)
Rest 1 Rest 2 Rest 3 Coefficient of variation (%) Intraclass reliability
7.5±0.7 6.8±0.8 7.0±0.7 16.4±2.5 0.846
7.4±1.0 7.5±1.0 7.6±1.0 13.4±3.8 0.955
Note: Values are means ± SEM.
HOMA2-IR before the IVGTTs had a CV of 18.6% and an intraclass reliability of 0.793. After moderate exercise (55 min at 60% V˙O2peak), OGTT-derived MISI rose 29% from rest due to a 17% reduction in the product of the mean glucose and insulin concentration and an 11% reduction in the product of fasting plasma glucose and insulin concentration (Table 3). However, the improvement in MISI did not reach statistical significance (Fig. 1A; p = 0.141), likely due to the inconsistent intersubject response to exercise, because 4 out of the 10 participants did not show improvement in MISI (individual responses are dotted lines in Fig. 1A). Post-exercise CISI increased significantly by 50% (Fig. 1B; p < 0.001) mainly by a 27% reduction in the area under the curve of plasma insulin above basal value (p = 0.034, Table 3). The individual response showed that all 10 participants improved their post-exercise CISI (individual responses are dotted lines in Fig. 1B). After exercise HOMA2-IR did not show significant differences from pre-exercise values despite a 15% improvement (0.78 ± 0.07 to 0.67 ± 0.12 from pre- to post-exercise, p = 0.479). Correlations between techniques During resting conditions MISI and CISI showed a high and significant correlation (r = 0.828; p = 0.003; 95% LOA = 4.1 to –3.9). However, the correlation between MISI and CISI scores after exercise was not significant (r = 0.547; p = 0.102; 95% LOA = 10.2 to –6.2). In addition, the changes in the insulin sensitivity from rest to exercise obtained from both indices (MISI and CISI) were calculated from one randomly chosen resting condition trial (OGREST and IVREST) and the exercise trial (OGEX and IVEX). Correlation of those changes with exercise in each of the techniques (OG and IV) did not show statistical significance (r = 0.337; p = 0.341; 95% LOA = 9.4 to –6.1). Sample size calculation Sample size estimates (Utzschneider et al. 2007) revealed that to find the effect of exercise on insulin sensitivity, 54 participants would be required when using MISI (SD = 4.1), whereas only 6 participants when using CISI (SD = 2.1). Published by NRC Research Press
790
Appl. Physiol. Nutr. Metab. Vol. 39, 2014
Table 3. Components of the Matsuda index from oral (OGTT) and Tura index from intravenous (IVGTT) during rest and after exercise.
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
Matsuda index (OGTT) FPI (IU·L−1)·FPG (mg·dL−1) PGMEAN (mg·dL−1)·PIMEAN (UI·mL−1) Tura index (IVGTT) KG (%·min−1) ⌬AUCINS (IU·mL−1)
Rest
Exercise
p value
537±43 4755±525
455±38 3848±475
0.320 0.136
2.9±0.3 1437±347
3.1±0.2 1042±233
0.138 0.034
Note: Values are means ± SEM. (FPI, fasting plasma insulin; FPG, fasting plasma glucose; PGMEAN, mean plasma glucose; PIMEAN, insulin; KG, rate of glucose disappearance; ⌬AUCINS, and area under the curve of insulin concentration above baseline).
Fig. 1. (A) Matsuda index (MISI) from 2 h oral glucose tolerance tests (OGTT) during rest (OGREST) and after exercise (OGEX). (B) Tura’s insulin sensitivity index (CISI) from 50 min intravenous glucose tolerance tests (IVGTT) during rest (IVREST) and after exercise (IVEX). Individual responses shown as dotted lines. Values are means ± SEM. *Difference from rest (p < 0.05).
Discussion It is well established that exercise stimulates insulin-dependent and insulin-independent glucose uptake in human skeletal muscle (Hayashi et al. 1997). The OGTT and the IVGTT are two dynamic methods to assess insulin sensitivity when the euglycemic hyperinsulinemic clamp technique is not affordable. These surrogate insulin sensitivity indexes from glucose tolerance tests are indistinctly reported in the scientific literature without much regard to questions of reproducibility or responsiveness. We tested the reliability of MISI (derived from OGTT) and CISI (derived from IVGTT) at rest during standardized conditions. In addition, we investigated which index more readily detects the insulin sensitizing effects of a single bout of prolonged (55 min) moderately intense exercise (60% V˙O2peak). Unlike most of the literature in this area, we tested reliability and response in the same participants using a cross-over repeated measures experimental design.
A major finding of this study is that CISI derived from IVGTT showed the highest reproducibility (Table 2). Because of this higher reproducibility and consistency (i.e., every subject improved) we detected the improvements in insulin sensitivity with exercise only with CISI (p < 0.001; Fig. 1B). In contrast, MISI was not responsive enough to significantly detect the effects of exercise on insulin sensitivity (Fig. 1A). To ascertain the effects of exercise using MISI another 44 similar participants would need to be tested. The extra cost and effort of testing that number of participants may not be justified and out of the scope of many research laboratories with restricted budgets. On the other hand, CISI is calculated after a 50 min IVGTT, whereas MISI requires a 120 min OGTT. The reduced time of study is another advantage when using CISI instead of MISI. Thus, our data support the preferential use of the IVGTT technique in comparison with OGTT when evaluating the effects of and exercise bout (or exercise program) on insulin sensitivity. Although most studies show a positive effect of exercise on insulin sensitivity when assessed by OGTT (Balducci et al. 2010; Dumortier et al. 2003; Jenkins and Hagberg 2011; Kishimoto et al. 2002; Venables and Jeukendrup 2008), some studies report no effects. Hansen et al. (2009) found that for obese type 2 diabetes patients, exercise training programs of different intensities improved glycated hemoglobin (HbA1C) and whole body oxidative capacity, whereas the insulin area under the curve measured during OGTT remained unchanged. Mitchell et al. (2011) measured insulin sensitivity using OGTT in obese participants after aerobic exercise and found no improvements. Niakaris et al. (2005) recommended caution in the interpretation of OGTT-derived insulin sensitivity results, because they seem to be population dependent. Our data suggested that the discrepancies about the effects of exercise on OGTT-derived insulin sensitivity calculations originate from the low reproducibility of this technique. In addition, the response was inconsistent, because 4 out of the 10 participants did not improve MISI after exercise. MISI does not only reflect peripheral insulin sensitivity but also hepatic insulin sensitivity (Matsuda and DeFronzo 1999). As exercise is mostly known to affect peripheral insulin sensitivity, it is possible that MISI was not that sensitive to the effects of exercise because a portion of the calculation is devoted to hepatic insulin sensitivity. The fasting glucose and insulin concentration portion of the MISI formulae (see Materials and methods) reflects hepatic insulin sensitivity. When we removed that portion of the formula and only the glucose and insulin concentrations during the OGTT were compared, no significant differences were found between the exercise and rest trials (p = 0.136, Table 3). We interpret this to indicate that the lack of improvement in MISI after exercise is also due to a variable response in peripheral insulin sensitivity. It is well established that one bout of aerobic exercise improves glycemic control and insulin sensitivity (Kennedy et al. 1999; Perseghin et al. 1996). In a recent study, healthy nonobese young men were tested 20 min after a bout of 30 min of continuous pedaling at 50% of their V˙O2max. The authors reported a 45% improvement in insulin sensitivity when using frequently sampled IVGTT (Hayashi et al. 2005). We used a similar energy demanding bout of exercise (i.e., 55 min of exercise at 60% V˙O2peak) and insulin sensitivity improved 50% with CISI (p < 0.001; Fig. 1B). Our results using the 50 min IVGTT technique described by Tura et al. (2010), agree in the magnitude of insulin sensitivity improvement with previous studies (Hayashi et al. 2005). Our study design had several limitations. First we favored the use of IVGTT instead of OGTT. However, it is necessary to note that OGTT better resembles the physiological challenge that an individual faces after a meal, which justifies the use of OGTT especially in large epidemiological studies. Second, we used healthy participants with high baseline insulin sensitivity values very similar to the ones found by Tura et al. (2010) in a similar population (7.5 vs. 5.55 ± 0.25 × 10−4 min−1 [U·mL]−1). However, there is an inverse relationship between baseline insulin sensitivity and inPublished by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
Ortega et al.
sulin sensitivity improvements with exercise (Magkos et al. 2008). So it is possible that if a group of overweight or insulin-resistant participants were to perform this study, the increase in insulin sensitivity with exercise would have been more robust and statistically significant, even when using MISI. Finally, the tolerance tests after exercise were conducted at a time of day that was 1 h after the time of day for the tests in resting condition. Although circadian rhythm could affect insulin sensitivity between morning and afternoon trials (Boden et al. 1996), it is unlikely that circadian rhythm could be a factor when trials are spaced only by one hour (0800 and 0900 h). In addition, this difference in the time at the start of the tests should have not clouded the results of our main aim, which was the comparison of 2 insulin sensitivity indexes with different modes of glucose load delivery (i.e., MISI vs. CISI), considering that the difference in time between the rest and post-exercise tests was the same for MISI and CISI. HOMA2-IR (Levy et al. 1998) stands out as the more cost- and time-efficient insulin sensitivity index, as it requires just fasting glucose and insulin concentrations. However, HOMA2-IR was unable to detect the insulin sensitivity changes induced by the 55 min bout of exercise and it was less reproducible than CISI. Thus, according to our data, HOMA2-IR is not a recommended technique to measure the insulin sensitizing effects of a bout of exercise. In fact, other authors have argued that HOMA-IR is not indicated to assess insulin sensitivity under nonsteady state conditions (Wallace et al. 2004) such as the transitional improvements induced by exercise. In conclusion, an exercise prescription to improve carbohydrate metabolism requires a reliable and responsive test to gauge the effect of exercise on insulin sensitivity. We compared two glucose tolerance tests that could be used for insulin sensitivity assessment out of hospital boundaries. We found improved responsiveness and reproducibility when the glucose load is delivered through an intravenous catheter (CISI, IVGTT derived) in comparison with oral ingestion (MISI, OGTT derived). Using 10 untrained, nondiabetic participants only the insulin sensitivity index based on IVGTT was fit to detect the acute effects of a bout of prolonged aerobic exercise. In most exercise science research environments, where the use of a euglycemic hyperinsulinemic clamp is not possible and sample size is limited, CISI based on 50 min IVGTT is a superior index to gauge the acute effects of exercise.
Acknowledgements We would like to thank the participants for their uninterested collaboration in this project. The help of Alvaro López and Carolina Cruces is greatly appreciated. All authors read and approved the final manuscript. This work was partially supported by a grant from the Spanish Ministry of Economia y Competitividad (DEP201128615). The authors report no conflicts of interest.
References American Diabetes Association. 2011. Standards of medical care in diabetes-2011. Diabetes Care, 34(Suppl. 1): S11–S61. doi:10.2337/dc11-S011. PMID:21193625. American·College·of·Sports·Medicine. 2006. ACSM’s guidelines for exercise testing and prescription. Lippincott Williams & Wilkins, Baltimore, Maryland. Balducci, S., Zanuso, S., Nicolucci, A., De Feo, P., Cavallo, S., Cardelli, P., Fallucca, S., Alessi, E., Fallucca, F., and Pugliese, G. 2010. Effect of an intensive exercise intervention strategy on modifiable cardiovascular risk factors in subjects with type 2 diabetes mellitus: a randomized controlled trial: the Italian Diabetes and Exercise Study (IDES). Arch. Intern. Med. 170(20): 1794– 1803. doi:10.1001/archinternmed.2010.380. PMID:21059972. Bergman, R.N., Prager, R., Volund, A., and Olefsky, J.M. 1987. Equivalence of the insulin sensitivity index in man derived by the minimal model method and the euglycemic glucose clamp. J. Clin. Invest. 79(3): 790–800. doi:10.1172/ JCI112886. PMID:3546379. Bingley, P.J., Colman, P., Eisenbarth, G.S., Jackson, R.A., McCulloch, D.K., Riley, W.J., and Gale, E.A. 1992. Standardization of IVGTT to predict IDDM. Diabetes Care, 15(10): 1313–1316. doi:10.2337/diacare.15.10.1313. PMID:1425095. Bland, J.M., and Altman, D.G. 1986. Statistical methods for assessing agreement
791
between two methods of clinical measurement. Lancet, 1(8476): 307–310. doi:10.1016%2Fj.ijnurstu.2009.10.001. PMID:2868172. Bloomgarden, Z.T. 2006. Measures of insulin sensitivity. Clin. Lab. Med. 26(3): 611–633, vi. doi:10.1016/j.cll.2006.06.007. PMID:16938587. Boden, G., Chen, X., and Urbain, J.L. 1996. Evidence for a circadian rhythm of insulin sensitivity in patients with NIDDM caused by cyclic changes in hepatic glucose production. Diabetes, 45(8): 1044–1050. doi:10.2337/diab.45.8. 1044. PMID:8690150. Colberg, S.R., Albright, A.L., Blissmer, B.J., Braun, B., Chasan-Taber, L., Fernhall, B., Regensteiner, J.G., Rubin, R.R., and Sigal, R.J. 2010. Exercise and type 2 diabetes: American College of Sports Medicine and the American Diabetes Association: joint position statement. Exercise and type 2 diabetes. Med. Sci. Sports Exerc. 42(12): 2282–2303. doi:10.1249/MSS.0b013e3181eeb61c. Dumortier, M., Brandou, F., Perez-Martin, A., Fedou, C., Mercier, J., and Brun, J.F. 2003. Low intensity endurance exercise targeted for lipid oxidation improves body composition and insulin sensitivity in patients with the metabolic syndrome. Diabetes Metab. 29(5): 509–518. doi:10.1016/S1262-3636(07)70065-4. PMID:14631328. Ferrari, P., Alleman, Y., Shaw, S., Riesen, W., and Weidmann, P. 1991. Reproducibility of insulin sensitivity measured by the minimal model method. Diabetologia, 34(7): 527–530. doi:10.1007/BF00403291. PMID:1916059. Hansen, D., Dendale, P., Jonkers, R.A., Beelen, M., Manders, R.J., Corluy, L., Mullens, A., Berger, J., Meeusen, R., and van Loon, L.J. 2009. Continuous lowto moderate-intensity exercise training is as effective as moderate- to highintensity exercise training at lowering blood HbA(1c) in obese type 2 diabetes patients. Diabetologia, 52(9): 1789–1797. doi:10.1007/s00125-009-1354-3. PMID: 19370339. Hawley, J.A. 2004. Exercise as a therapeutic intervention for the prevention and treatment of insulin resistance. Diabetes Metab. Res. Rev. 20(5): 383–393. doi:10.1002/dmrr.505. PMID:15343584. Hayashi, T., Wojtaszewski, J.F., and Goodyear, L.J. 1997. Exercise regulation of glucose transport in skeletal muscle. Am. J. Physiol. 273(6 Pt 1): E1039–E1051. Hayashi, Y., Nagasaka, S., Takahashi, N., Kusaka, I., Ishibashi, S., Numao, S., Lee, D.J., Taki, Y., Ogata, H., Tokuyama, K., and Tanaka, K. 2005. A single bout of exercise at higher intensity enhances glucose effectiveness in sedentary men. J. Clin. Endocrinol. Metab. 90(7): 4035–4040. doi:10.1210/jc.2004-2092. PMID:15840752. Howlett, K.F., Mathews, A., Garnham, A., and Sakamoto, K. 2008. The effect of exercise and insulin on AS160 phosphorylation and 14-3-3 binding capacity in human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 294(2): E401–E407. doi:10.1152%2Fajpendo.00542.2007. PMID:18042670. Ivy, J.L., Frishberg, B.A., Farrell, S.W., Miller, W.J., and Sherman, W.M. 1985. Effects of elevated and exercise-reduced muscle glycogen levels on insulin sensitivity. J. Appl. Physiol. 59(1): 154–159. PMID:4030559. Jenkins, N.T., and Hagberg, J.M. 2011. Aerobic training effects on glucose tolerance in prediabetic and normoglycemic humans. Med. Sci. Sports Exerc. 43(12): 2231–2240. doi:10.1249/MSS.0b013e318223b5f9. PMID:21606871. Kennedy, J.W., Hirshman, M.F., Gervino, E.V., Ocel, J.V., Forse, R.A., Hoenig, S.J., Aronson, D., Goodyear, L.J., and Horton, E.S. 1999. Acute exercise induces GLUT4 translocation in skeletal muscle of normal human subjects and subjects with type 2 diabetes. Diabetes, 48(5): 1192–1197. doi:10.2337/diabetes.48. 5.1192. PMID:10331428. Kishimoto, H., Taniguchi, A., Fukushima, M., Sakai, M., Tokuyama, K., Oguma, T., Nin, K., Nagata, I., Hayashi, R., Kawano, M., Hayashi, K., Tsukamoto, Y., Okumura, T., Nagasaka, S., Mizutani, H., and Nakai, Y. 2002. Effect of short-term low-intensity exercise on insulin sensitivity, insulin secretion, and glucose and lipid metabolism in non-obese Japanese type 2 diabetic patients. Horm. Metab. Res. 34(1): 27–31. doi:10.1055/s-2002-19963. PMID:11832998. Levy, J.C., Matthews, D.R., and Hermans, M.P. 1998. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care, 21(12): 2191–2192. doi:10.2337/diacare.21.12.2191. PMID:9839117. Magkos, F., Tsekouras, Y., Kavouras, S.A., Mittendorfer, B., and Sidossis, L.S. 2008. Improved insulin sensitivity after a single bout of exercise is curvilinearly related to exercise energy expenditure. Clin. Sci. (Lond.), 114(1): 59–64. doi:10.1042/CS20070134. Matsuda, M., and DeFronzo, R.A. 1999. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care, 22(9): 1462–1470. doi:10.2337/diacare.22.9.1462. PMID:10480510. Mikines, K.J., Sonne, B., Farrell, P.A., Tronier, B., and Galbo, H. 1988. Effect of physical exercise on sensitivity and responsiveness to insulin in humans. Am. J. Physiol. 254(3 Pt 1): E248–E259. PMID:3126668. Mitchell, J.B., Phillips, M.D., Yellott, R.C., and Currie, L.M. 2011. Resistance and aerobic exercise: the influence of mode on the relationship between IL-6 and glucose tolerance in young men who are obese. J. Strength Cond. Res. 25(6): 1529–1537. doi:10.1519/JSC.0b013e3182176638. PMID:21487312. Muniyappa, R., Lee, S., Chen, H., and Quon, M.J. 2008. Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am. J. Physiol. Endocrinol. Metab. 294(1): E15–E26. doi:10. 1152%2Fajpendo.00645.2007. PMID:17957034. Niakaris, K., Magkos, F., Geladas, N., and Sidossis, L.S. 2005. Insulin sensitivity derived from oral glucose tolerance testing in athletes: disagreement between Published by NRC Research Press
Appl. Physiol. Nutr. Metab. Downloaded from www.nrcresearchpress.com by DALHOUSIE UNIVER on 06/29/14 For personal use only.
792
available indices. J. Sports Sci. 23(10): 1065–1073. doi:10.1080/02640410400023241. PMID:16194983. Perseghin, G., Price, T.B., Petersen, K.F., Roden, M., Cline, G.W., Gerow, K., Rothman, D.L., and Shulman, G.I. 1996. Increased glucose transportphosphorylation and muscle glycogen synthesis after exercise training in insulin-resistant subjects. N. Engl. J. Med. 335(18): 1357–1362. doi:10.1056/ NEJM199610313351804. PMID:8857019. Roden, M. 2012. Exercise in type 2 diabetes: to resist or to endure? Diabetologia, 55(5): 1235–1239. doi:10.1007/s00125-012-2513-5. PMID:22391950. Tura, A., Sbrignadello, S., Succurro, E., Groop, L., Sesti, G., and Pacini, G. 2010. An empirical index of insulin sensitivity from short IVGTT: validation against the minimal model and glucose clamp indices in patients with different
Appl. Physiol. Nutr. Metab. Vol. 39, 2014
clinical characteristics. Diabetologia, 53(1): 144–152. doi:10.1007/s00125-0091547-9. PMID:19876614. Utzschneider, K.M., Prigeon, R.L., Tong, J., Gerchman, F., Carr, D.B., Zraika, S., Udayasankar, J., Montgomery, B., Mari, A., and Kahn, S.E. 2007. Withinsubject variability of measures of beta cell function derived from a 2 h OGTT: implications for research studies. Diabetologia, 50(12): 2516–2525. doi:10.1007/ s00125-007-0819-5. PMID:17928990. Venables, M.C., and Jeukendrup, A.E. 2008. Endurance training and obesity: effect on substrate metabolism and insulin sensitivity. Med. Sci. Sports Exerc. 40(3): 495–502. doi:10.1249/MSS.0b013e31815f256f. PMID:18379212. Vincent, W. 1999. Statistics in kinesiology. Human Kinetics, Champaign. Wallace, T.M., Levy, J.C., and Matthews, D.R. 2004. Use and abuse of HOMA modeling. Diabetes Care, 27(6): 1487–1495. doi:10.2337/diacare.27.6.1487. PMID:15161807.
Published by NRC Research Press
