JEPDavey

JEPonline
Journal of
Exercise Physiologyonline
Official Journal of the American
Society of Exercise Physiologists (ASEP)
ISSN 1097-9751
An International Electronic Journal
Volume 3 Number 2 April 2000

Exercise and Health
Effects of Exercise on Insulin Resistance in South Asians and Europeans
G.J.G. DAVEY¹, J.D. ROBERTS², S. PATEL³, T. PIERPOINT³, I.F. GODSLAND⁴, B. DAVIES⁵ and P.M. MCKEIGUE³
¹Department of Public Health Science, St George’s Hospital Medical School, Cranmer Terrace, London, ²British Olympic Medical Centre, Northwick Park Hospital, ³Epidemiology Unit, Department of Epidemiology & Population Sciences, London School of Hygiene & Tropical Medicine, ⁴Wynn Department of Metabolic Medicine, St Mary’s & Imperial College of Science & Technology, ⁵B. Davies, British Olympic Medical Centre, Northwick Park Hospital

G.J.G. DAVEY, J.D. ROBERTS, S. PATEL, T. PIERPOINT, I.F. GODSLAND, B. DAVIES and P.M. McKEIGUE. Effects Of Exercise On Insulin Resistance In South Asians And Europeans. JEPonline, Vol 3, No 2, 2000. Insulin resistance underlies coronary heart disease and type II diabetes in South Asians and Europeans. We investigated whether exercise training could ameliorate insulin resistance, and whether any benefit depended on the timing of measurement in relation to exercise. Ninety-two sedentary South Asian and European men and women aged 35-49 years were recruited. After baseline measurements, subjects were randomized to one of three groups: no change in daily activity (NE), exercise with follow up insulin sensitivity measured within 24 hours of last exercise session (E1), and exercise with follow up insulin sensitivity measured five days after last session (E5). Insulin sensitivity was determined by minimal model analysis of glucose and insulin concentrations. Maximal oxygen uptake was measured using a graded exercise treadmill test based on a modified Bruce protocol. Data from E1 and E5 showed a significant increase in cardiorespiratory fitness compared to NE (+4.15 vs. -0.003 mL/kg/min, p<0.001). Insulin sensitivity was significantly improved only in E1 compared to NE or E5 (+0.67 vs. +0.30 min/pmol/L, p = 0.05), representing a 40% mean increase on initial values. In conclusion, exercise improved insulin sensitivity by 40% among those in whom it was measured within 24 hours of the final exercise session. An effect of this magnitude has considerable implications for the prevention of non-insulin-dependent diabetes at the population level.
Key Words: Exercise, Insulin resistance, Maximal oxygen uptake, Ethnic, Intravenous glucose tolerance test

INTRODUCTION
The cluster of metabolic disturbances termed the insulin resistance syndrome was first described in association with non-insulin-dependent diabetes mellitus (NIDDM) and coronary heart disease in Europeans and Native Americans (1). Hyperinsulinemia, hypertriglyceridemia, hypertension, low HDL-cholesterol, and a tendency to accumulate intra-abdominal fat were later found to be prominent in people originating from India, Pakistan and Bangladesh (“South Asians”) (2,3). These abnormalities are apparent from a young age (4), and together with type II diabetes underlie the high risk of coronary heart disease in this population (5,6).
Interventions to diminish insulin resistance and hence the risk of NIDDM and coronary heart disease (CHD) are urgently needed. Several interventions have been proposed, including weight loss (7), exercise (8,9), thiazolidinediones (10) and omega-3 fatty acids (11). We tested the effect of an exercise programme among South Asians and Europeans with features of insulin resistance, but with normal glucose tolerance according to World Health Organization standard classification.
METHODS
The study was a randomized controlled trial. The duration of intervention was 12 weeks, and five successive but overlapping cohorts were formed to eliminate seasonal variation, thus the intervention part of the study ran from May 1995 to April 1996.
Ethical Approval
The study design and informed consent was approved by committees at the London School of Hygiene and Tropical Medicine, the Wynn Department of Metabolic Medicine, and Ealing, Hammersmith and Hounslow Health Agency.
Sample Size
A sample size of 15 subjects in each group was calculated to have 90% power to detect at 5% significance a change in insulin sensitivity of 36% (assuming 3-month within-subject coefficient of variation for insulin sensitivity of 30% - unpublished data). We recruited 90 participants to allow for potential dropouts.
Recruitment
Participants were recruited in a three-stage process. Lists of people registered with ten family practices in West London were used for an initial mailshot. Those responding to the mailshot were invited to attend an examination at a nearby health center. Self-reported ethnicity was established at this visit. Those with hyperlipidemia, abnormal glucose tolerance, ischemic changes on resting ECG, body mass index >40 kg/m² and intolerance of venepuncture were excluded. Sedentary people in the highest third of an index for insulin resistance (based on fasting insulin, glucose and triglycerides, waist hip ratio and family history of type II diabetes) were invited to participate. Those who agreed to participate performed a familiarization session on an exercise treadmill, which included the completion of the first three stages of the Bruce protocol (12).
Baseline Measures
The same procedures were performed on participants before randomization, in order to minimize post-randomization dropout. The procedures consisted of an intravenous glucose tolerance test (IVGTT) and an exercise test for determination of maximal oxygen uptake (VO₂max).
The IVGTT was performed using a high dose of glucose (0.5 g/kg), no tolbutamide, and a reduced sampling schedule. Participants prepared for this by consuming a carbohydrate-rich (at least 200 g/day) diet in the 3 days leading up to the test, and fasting from 9 pm the night before. Fasting samples were taken for glucose, insulin and lipids, and a 50% D-glucose was then given via an antecubital vein over 3 minutes, and blood samples were collected at 3, 5, 7, 10, 15, 20, 30, 45, 60, 75, 90, 120, 150, and 180 minutes. Height and weight were measured, and a questionnaire on tobacco and alcohol consumption, diet, family history and daily activity was administered.
Insulin sensitivity (S_i) was measured using Bergman’s minimal model of glucose disappearance (13), using programs written in Fortran 77 run on a PDP-11/83 microcomputer. The relatively high glucose dose (0.5 g/kg rather than 0.3 g/kg) employed provides for a sufficient endogenous insulin response in non-diabetic volunteers without recourse to additional augmentation of pancreatic insulin secretion. This is apparent in the high rate of model identification (96%) and good correlation with measures of insulin sensitivity derived from the euglycaemic clamp (r = 0.92) that we have found at the higher glucose dose (14,15).
The test of VO₂max was completed using a symptom limited graded exercise treadmill test (GXT) with a modified Bruce protocol (12). Speed and elevation were increased incrementally every 3 minutes. Breath-by-breath analyses of oxygen, carbon dioxide and ventilation were recorded using an on-line computerized gas analysis system (Mjinhardt Oxycon Champion System, Jaeger, UK). Heart rate, blood pressure, rating of perceived exertion and blood lactate were measured every 3 minutes, the latter being measured using a YSI Lactate Analyzer (Yellow Springs, OH). VO₂max was established when at least two of the following conditions were met: (a) a plateau or decrease in oxygen uptake associated with an increase in workload, or (b) a respiratory exchange ratio (RER) >1.15.
Randomization
Ninety-two participants were coded, the codes entered into a computer randomization routine and randomized to three groups stratified by sex and ethnicity. These groups were, no change in physical activity/wait list (NE), exercise intervention with follow up tests 24 hours after last session (E1) and exercise intervention with follow up tests 5 days after last session (E5).
Exercise Intervention
Participants randomized to exercise were given individually tailored exercise programs, requiring them to perform three half-hour sessions of interval walk/jogging (or later jog/running) to 65-75% of VO₂max assessed by heart rate extrapolated from the exercise test, plus one supervised aerobic circuit session per week. Compliance was monitored by weekly diary records and waistband actometers.
Follow-up Measures
IVGTT and GXT were repeated at the end of the intervention, with the timing of the follow-up IVGTT being related to a participant’s final exercise session as determined by randomization group for exercisers (E1 or E5). Anthropometry was repeated at the IVGTT appointment, and questionnaire-led checks were made on changes in general health, diet and activity during the intervention period. Five people dropped out during the intervention, so follow-up measures were performed on 87 participants.
Data Analysis
Data was analysed using Stata4.0 for Windows. Natural log or square root transformations were performed on any quantitative variables whose distribution was skewed. Chi-squared tests were performed for differences in proportions of binary variables. T-tests were used to compare means of quantitative variables whose standard deviations were equal, and Wilcoxon’s Rank Sum test for those whose standard deviation were significantly different. One-way analysis of variance was employed to compare means across more than two groups where the standard deviations were equal, and the F-test was used to evaluate the significance of these comparisons. Data are presented as mean± SD.
RESULTS
Eight-seven participants completed the intervention. Data for baseline variables, grouped by gender, are shown for each of the groups in Table 1.
Table 1. Mean±SD baseline variables by sex and ethic group.

Variable Men Women

European South Asian European South Asian

Height
(cm) 175±4.8 171.6±6.9 164.9±4.9 159.5±4.9

Weight
(kg) 79.1±6.8 74.5±8.0 65.4±7.8 65.6±7.2

Systolic BP (mmHg) 115.3±12.4 114.0±12.4 108.5±10.4 107.5±12.9

Diastolic BP (mmHg) 73.0±8.3 76.8±10.4 68.1±11.6 68.4±8.2

BMI
(kg/m²) 25.7±2.0 25.2±2.1 24.1±2.5 25.8±2.7

WHR 0.92±0.04 0.94±0.04 0.80±0.05 0.84±0.05

S_i
(min/pmol/L) 2.182±0.29 1.626±0.23 2.901±0.38 1.325±0.12

VO₂max (ml/kg/min 39.4±4.2 35.4±4.1 30.7±4.7 24.3±4.0

S_i = insulin sensitivity
Baseline characteristics of participants did not differ significantly between randomization groups, and are shown in Table 2.
Table 2. Mean±SD baseline variables by exercise group.

Variable NE E1 E5

Male: female 19:11 19:11 17:10

Age (years) 41.6±3.8 42.4±3.9 41.9±4.4

European:
South Asian 15:15 16:14 16:11

Proportion current smokers 0.13 0.1 0.22

Proportion family history DM 0.1 0.3 0.3

BMI
(kg/m²) 24.4±2.7 25.4±2.3 24.8±1.9

WHR 0.89±0.01 0.89±0.01 0.89±0.01

S_i
(min/pmol/L) 2.09±0.28 1.66±0.32 2.36±16.1

VO₂max (ml/kg/min) 33.6±7.1 34.4±6.6 34.1±6.1

S_i = insulin sensitivity
However, mean waist-hip ratio was lower among Europeans than South Asians within each sex (0.92 vs. 0.94, p = 0.036 for men; 0.80 vs. 0.84, p = 0.026 for women). Maximal oxygen uptake (VO₂max) at baseline was significantly higher among men than women. VO₂max was higher among Europeans than South Asians (35.7± 6.2 vs. 32.0± 6.5 ml/kg/min, p = 0.009), with significance persisting after adjustment for sex and age.
Table 3 shows changes in variables during the intervention, by randomization group. Fitness increased significantly in E1and E5 compared to NE (+4.15 vs.-0.003 ml/kg/min, p<0.0001), representing a mean change of 12.1% on baseline levels (Figure 1). This persisted after controlling for age, sex, ethnicity and baseline BMI. There were no significant between-group differences in fitness change comparing E1 and E5.
Table 3. Mean (95% CI) changes in variables during trials by exercise group.

Variable NE E1 E5

Weight
(kg) 0.34 (-0.23 to 0.91 0.14 (-0.47 to 0.75) -0.03 (-0.63 to 0.57)

BMI
(kg/m²) 0.09 (-0.10 to 0.28) 0.07 (-0.13 to 0.23) -0.03 (-0.24 to 0.19)

WHR
(x10²) -0.6 (-1.2 to -0.1) -1.3 (-2.0 to -0.7) -1.5 (-2.2 to -0.9)

VO₂max (ml/kg/min) 0.00 (-0.82 to 0.81)^a 3.95 (2.94 to 4.97)^a 4.36 (3.13 to 5.61)^a

S_i
(min/pmol/L) 0.31 (-0.35 to 0.98)^b 0.67 (0.06 to 1.27)^b 0.30 (-0.27 to 0.87)^b

Key: ^aE1 and E5>NE, p<0.001; ^bE1>NE and E5, p = 0.05. S_i = insulin sensitivity
Insulin sensitivity was found to increase significantly in E1 compared to NE and E5 (0.67 vs. 0.31 and 0.30 min/pmol/L, p = 0.05, Figure 2), the significance increasing (to p=0.031) after adjustment for age, sex, ethnicity and baseline BMI. The increase in E1 represented a mean increase of 40%.

Figure 1. Changes in VO₂max by group.

Figure 2. Changes in insulin sensitivity by exercise group.
DISCUSSION
Before addressing the substantial problems represented by population-based interventions, it is important first to establish whether a proposed intervention is worthwhile in principle and likely to be effective. This study has demonstrated that a sizeable and significant increase in insulin sensitivity is achievable with exercise in a group of motivated South Asians and Europeans.
The 12% increase in VO₂max with twelve weeks of supervised exercise is consistent with other interventions of similar duration (9,16,17). Mean insulin sensitivity increased by 40% among those randomised to exercise whose IVGTTs were performed within 24 hours of the final session. This is comparable to other exercise interventions (9,16,18) in which follow-up insulin sensitivity was measured at this time interval. The loss of this improvement among participants whose IVGTT was performed five days after exercise has been demonstrated in other studies (18), and seems to mirror the effect of stopping training among trained athletes (8,19,20). The 40% increase in insulin sensitivity seen within 24 hours of an exercise session in this trial is approximately the same as that which accompanies weight loss of 6-10% (7,21). The Oslo Diet and Exercise Study (22) demonstrated more modest decreases in insulin resistance (measured by homeostatic modelling) of 9% with weight loss of 4.4% in the diet group, and a decrease of 20% with weight loss of 13% in the diet plus exercise group.
The precise frequency of exercise necessary to maintain increased insulin sensitivity cannot be determined from this study in which the increase is seen at 24 hours but not at five days. Other studies have found the effect to have worn off five days after exercise (18,19), but differ as to the effect at three days; King describing an increase maintained at three days (18), but Schneider, in Koivisto (9) not showing the same retention. The implications of our trial in the context of these smaller studies are that exercise training must be maintained for benefits in insulin sensitivity to be maintained.
CONCLUSIONS
We have demonstrated that a rigorous exercise programme undertaken by South Asians and Europeans can ameliorate insulin resistance. The improvement related to exercise is demonstrable 24 hours but not five days after an exercise session. Such an effect extrapolated to the population level would have important implications for reducing the risk of development of non-insulin-dependent diabetes.

ACKNOWLEDGEMENTS: This study was commissioned by the NHS Research and Development Directorate, following a pilot study funded by the British Heart Fund. We acknowledge the support of SmithKline Beecham, who provided Lucozade for the oral glucose tolerance tests; and Reebok UK for a number of pairs of training shoes. We thank Jan Mazar for assistance with recruitment and interpretation, and Sheelagh Kerr for creating and managing the main study database. We acknowledge the help of Jaspal Kooner, Melvyn Hillsdon, Keith Oppenshaw and Tim Anstiss at Ealing Hospital. Thanks also go to Dr Guha for the use of premises in Old Southall, as well as to the other general practitioners who allowed us to approach their patients. We are particularly thankful to the people who participated in the study, finding time to attend for tests, interviews and training. It is a tribute to their enthusiasm that the weekly exercise classes continue even beyond the official end of the study.
REFERENCES
1. Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37:1595-1607.
2. McKeigue PM, Marmot MG, Syndercombe Court YD, Cottier DE, Rahman S, Riemersma RA. Diabetes, hyperinsulinaemia and coronary risk factors in Bangladeshis in east London. Br Heart J 1988;60:390-396.
3. Sharp PS, Mohan V, Levy JC, Mather HM, Kohner EM. Insulin resistance in oatients of Asian Indian and European origin with non-insulin dependent diabetes. Horm Metab Res 1987;19:84-85.
4. Gelding SV, Nuththyananthan R, Chan SP, Skinner E, Robinson S, Gray IP, et al. Insulin sensitivity in non-diabetic relatives of patients with non-insulin-dependent diabetes from two ethnic groups. Clin Endocrinol 1994;40:55-62.
5. McKeigue PM, Shah B, Marmot MG. Relation between central obesity and insulin resistance with high diabetes prevalence and cardiovascular risk in South Asians. Lancet 1991;337:382-386.
6. Dhawan J, Bray CL, Warburton R, Ghambhir DS, Morris J. Insulin resistance, high prevalence of diabetes and cardiovascular risk in immigrant Asians. Br Heart J 1994;72:413-421.
7. Olefsky JM, Reaven GM, Farquhar JW. Effects of weight reduction on obesity: studies of carbohydrate and lipid metabolism. J Clin Invest 1974;53:64-76.
8. Health GW, Gavin JR III, Hinderliter JM, Hagberg JM, Bloomfield SA, Holloszy JO. Effects of exercise and lack of exercise on glucose tolerance and insulin sensitivity. J Appl Physiol 1983;55:512-517.
9. Koivisto VA, Yki-Jarvinen H, DeFronzo RA. Physical training and insulin sensitivity. Diabetes Metab Rev 1986;1:445-481.
10. Keen H. Insulin resistance and the prevention of diabetes mellitus. N Engl J Med 1994;331:1226-1227.
11. Popp-Snijders C, Schouten JA, Heine RJ, Vander Meer H, VanderVeen EA. Dietary supplementation of omega-3 polyunsaturated fatty acids improves insulin sensitivity in non-insulin-dependent diabetes mellitus. Diabetes Res 1987;4:141-147.
12. Bruce RA. Exercise testing of patients with coronary heart disease. Principles and normal standards for evaluation. Ann Clin Res 1971;3:323-332.
13. Bergman RN, Ziya Ider Y, Bowden CR, Cobelli C. Quantitative estimation of insulin sensitivity. Am J Physiol 1979;236:E667-677.
14. Walton C, Godsland IF, Proudler A, Felton C, Wynn V. Evaluation of four mathematical models of glucose and insulin dynamics with analysis of effects of age and obesity. Am J Physiol 1992;262:E755-762.
15. Swan JW, Walton C, Godsland IF. Assessment of insulin sensitivity in man: a comparison of minimal-model- and euglycaemic clamp-derived measures in health and heart failure. Clin Science Colch 1994;865:317-322.
16. Kirwan JP, Kohrt WM, Wojta DM, Bourey RE, Holloszy JO. Endurance exercise training reduces glucose-stimulated insulin levels in 60- to 70-year-old men and women. J Gerontol 1993;48:M84-90.
17. Rauramaa R, Salonen JT, Kukkonen-Harjula K, Seppanen K, Seppala E, Vapaatalo H, et al. Effects of mild physical exercise on serum lipoproteins and metabolites of arachidonic acid: a controlled randomised trial in middle aged men. BMJ 1984;288:603-606.
18. King DS, Baldus PJ, Sharp RL, Kesl LD, Feltmeyer TL, Riddle MS. Time course for exercise-induced alterations in insulin action and glucose tolerance in middle-aged people. J Appl Physiol 1995;78:17-22.
19. Mikines KJ, Sonne B, Tronier B, Galbo H. Effects of acute exercise and detraining on insulin action in trained men. J Appl Physiol 1989;66:704-711.
20. Burstein R, Polychronakos C, Toews CJ, MacDougall JD, Guyda HJ, Posner BI. Acute reversal of the enhanced insulin action in trained athletes: association with insulin receptor changes. Diabetes 1985;34:756-760.
21. Slabber M, Barnard H, Kuyl J, Dannhauser A, Schall R. Effects of a low-insulin-response, energy-restricted diet on weight loss and plasma insulin concentrations in hyperinsulinemic obese females. Am J Clin Nutr 1994;60:48-53.
22. Torjesen PA, Birkeland KI, Andersen SA, Hjermann I, Holme I Urdal P. Lifestyle changes may reverse development of the insulin resistance syndrome. Diabetes Care 1997;20:26-31.
Address for correspondence:G.J.G. Davey, Department of Public Health Science, St George’s Hospital Medical School, Cranmer Terrace, London SW17 0RE, Phone number: (44)-181-725-2796, Fax number: (44)-181-725-3584, email: gdavey@sghms.ac.uk

Copyright ©1997-2000 American Society of Exercise Physiologists. All rights reserved.

ASEP Table of Contents
PEPonline
JEPonline -April 2000