Diet factors are now assumed to be the leading cause of chronic disease worldwide.1 One in three Americans born in 2000 or later and every second person in high-risk ethnic populations will develop type 2 diabetes mellitus (T2DM).2 T2DM is a significant cause of morbidity related to cardiovascular disease (CVD), blindness, kidney and nerve disease, arthritis, obstructive sleep apnoea syndrome (OSAS), psychological ill health, amputation and premature mortality.3
Chronic overconsumption of energy leads to weight gain and excess intra-abdominal fat, which predispose to insulin resistance,4 probably the strongest single predictor for the development of 2DM.5,6 Appropriate dietary measures and physical activity as part of a healthy lifestyle show substantial and clinically relevant effects in both the prevention and treatment of insulin resistance and T2DM. Lifestyle measures that aim to increase insulin sensitivity are of particular interest in this context. As weight loss and reduction of abdominal fat mass are almost invariably associated with improved insulin sensitivity,4 any sustainable energy-reduced and safe diet, ideally in combination with increased exercise levels, may be used both in patients with T2DM and in subjects at high risk. However, apart from weight loss, isoenergetic changes in the quality of ingested foods and in the macronutrient composition appear to exert additional important effects on insulin ensitivity,7–10 with adverse, neutral or protective effects of specific foods.1,11–13 However, this remains a controversial topic.
Chronic reduction in energy expenditure attributed to decreased physical activity is assumed to be another major contributor to the global epidemic of overweight and T2DM.14 Exercise has been shown to increase insulin sensitivity both acutely and in the long term, in addition to reducing weight and body fat mass, lowering blood glucose levels and improving cardiovascular function.15 However, many patients with T2DM are not physically active.16
We review current concepts and controversies regarding the modulation of glucose metabolism and diabetes risk using lifestyle measures.
We searched PubMed for original papers and review articles up to April 2012 using a combination of query terms that included ‘type 2 diabetes’, ‘nutrition’, ‘exercise’, ‘metabolic syndrome’, ‘insulin resistance’, ‘dyslipidaemia’, ‘adipokines’, ‘gut hormones’, ‘pro-inflammatory factors’, ‘obesity’ and many others that were assumed to be relevant. Relevant articles were also selected among reference lists in published papers. Using these search criteria, more than 1000 relevant original papers and review articles were identified. We then hand selected studies and review articles covering the relevant areas, excluding articles that provided similar information to the selected ones. When data from larger or more robust trials were available, studies with small sample sizes were excluded, as well as studies that showed high losses to follow-up and/or differential losses between the comparison groups.
The role of nutrition in the prevention and treatment of type 2 diabetes mellitus
Recommendations in current guidelines
Nutritional recommendations in current guidelines17 for the treatment of patients with T2DM and subjects at high risk of developing diabetes generally recommend weight loss of at least 7% in overweight/obese patients; a cholesterol intake < 200 mg/day; restriction of the intake of saturated fats to < 7% of energy intake; a high-fibre intake of at least 14 g/1000 kcal; and, in newer guidelines, relaxed restrictions on protein intake, e.g. protein intake of 15–20% of energy as long as kidney function is normal.17 Furthermore, intake of trans fat should be minimized.17 The use of low glycaemic index (GI) and glycaemic-load carbohydrates may provide a modest additional benefit for glycaemic control over that observed when total carbohydrate is considered alone.17 Finally, routine supplementation with antioxidants, such as vitamins E and C and carotene, is not advised because of lack of evidence of efficacy and increasing concerns related to long-term safety.17,18
Effects of weight loss
Obesity, and particularly accumulation of intra-abdominal fat mass, is the most common cause of insulin resistance and T2DM. Simply being overweight [body mass index (BMI) > 25 kg/m2 raises the risk of developing T2DM by a factor of 3,19 whereas even relatively modest weight reduction in obese patients with poorly controlled T2DM can markedly reduce plasma glucose concentrations. As weight loss progresses and is maintained, an improvement in glycaemic control is reflected by reductions in markers of glycaemic control such as glycated haemoglobin (HbA1c).20,21 Thus, a guiding principle in the treatment of patients with T2DM has been the recommendation to lose weight.22
However, even in the general overweight population, sustained weight loss appears too difficult to achieve. Weight-loss trials in diabetic patients are relatively consistent, showing initial success which typically plateaus after 4–6 months, followed by weight regain.23 In an 18-month study that compared the use of different dietary guidelines for T2DM, no significant changes in body weight were observed.24 In a 2-year study in 811 overweight adults, energy-reduced diets resulted in clinically meaningful weight loss regardless of the macronutrient composition (fat, protein or carbohydrates), with comparable effects of all diets on feelings of satiety and hunger, satisfaction with the diet, and attendance rates at group sessions; however, attendance at instructional sessions was strongly associated with weight loss (0.2 kg per session attended), indicating that adherence to any chosen diet may be a crucial factor and could be even more important than the macronutrient composition of the diet per se.25 This phenomenon has been described previously: in a study investigating the effectiveness of four popular diets (Atkins, Ornish, Weight Watchers and Zone) on weight loss, each diet modestly reduced body weight and several cardiac risk factors at 1 year, with low overall dietary adherence rates to all diets. However, increased adherence was associated with significantly greater weight loss and cardiac risk factor reductions in each diet group.26
Even under well-controlled conditions, sustained weight loss is in patients with T2DM is generally moderate (4% or 4.6 kg) and typically results only in a small decrease in HbA1c of ∼0.5%.27 At least in the short term, low-carbohydrate, high-protein diets may result in better weight loss than traditional low-fat diets.28 However, regardless of the dietary strategy, it appears to be difficult to achieve sustained therapeutic weight loss, particularly in patients with T2DM as compared with the general (overweight/obese) population. Generally, in obese individuals energy expenditure begins to drop as soon as body weight starts to decline,23,29,30 and powerful hypothalamic hormonal responses are induced in an effort to prevent further weight loss.23 However, in patients with diabetes, additional factors have been proposed that may further prevent sustained relevant weight loss. Energy expenditure is increased in the hyperglycaemic state as a result of increased protein turnover and could drop towards normal after improvement of glycaemic control.23,31,32
Furthermore, improved glycaemic control results in reduced glucosuria and thus decreased energy excretion in the urine. This increased retention of energy may lead to weight regain if energy intake does not drop further.23 Furthermore, many obese patients with T2DM are typically sedentary and may have barriers to exercising, including neuropathy, foot ulcers, heart disease23 and anxiety about experiencing hypoglycaemia. In addition, medication with certain antidiabetic drugs, including insulin, sulfonylureas, and thiazolidinediones, causes weight gain.23 Thus, achieving and maintaining both weight loss and relevant exercise levels is often not realistic in overweight/obese patients with T2DM. Furthermore, typically recommended exercise levels that are probably sufficient to improve glycaemic control and cardiovascular risk (e.g. 150 min/week of brisk walking) are usually inefficient to achieve relevant weight loss.33 The optimal volume of exercise to achieve sustained major weight loss appears to be much larger, some 60 min/day or more when relying on exercise alone as a weight-loss strategy.3,33
Little is known about the composition of body weight during weight regain, particularly in patients with T2DM. In the European multicentre Diet, Obesity and Genes (DioGenes) trial in overweight non-diabetic subjects, a modest increase in protein content and a modest reduction in the GI led to a more successful maintenance of weight loss after an initial energy-reduced diet.34 However, after 6 months, maintenance of weight loss was only marginally better with a high protein intake than with low protein intake (−0.71 kg or −1.1. kg, depending on combination with a low- or high-GI diet) and failed to reach significance in the full model, despite the large number (548) of completers in study.34 Furthermore, when investigating the body composition in postmenopausal women after intentional weight loss, followed by weight regain, for every 1 kg fat lost during weight loss intervention, 0.26 kg lean tissue was lost; but for every 1 kg fat regained over the following year, only 0.12 kg lean tissue was regained. These results indicate that fat mass is regained to a greater degree than is lean mass in those who experience weight regain after initial weight loss.35
Although weight loss and reduction of abdominal fat mass in patients with T2DM are in principle powerful tools for reducing insulin resistance, sustained relevant weight loss in these patients appears to be difficult to achieve. Therefore, it seems reasonable to explore specific metabolic effects of different (isoenergetic) foods and macronutrients on insulin sensitivity both in patients with T2DM and in individuals who are at high risk of developing T2DM.7–10
Effects of dietary components on insulin resistance and diabetes risk
An increasing number of studies indicate that isoenergetic changes in the macronutrient composition and the quality of ingested foods may exert additional important effects on insulin sensitivity, independent of weight loss. The effects of different nutrients on diabetes risk and glycaemic control are summarized in Table 1.
|Nutrient||Effect on glucose metabolism||Level of evidence||American Diabetic Association recommendation|
|Total carbohydrate||Unclear||Moderate||Minimum recommended daily allowance of 130 g/day to provide adequate glucose to central nervous system44|
|Simple sugars||Adverse||Moderate||Avoid excessive intake17|
|Low glycaemic index carbohydrate||Beneficial||Moderate||Encouraged. May produce modest additional benefts45|
|Monounsaturated fatty acids||Beneficial||Moderate||Encouraged17|
|Polyunsaturated fatty acids||Beneficial||Moderate||Two or more servings of fish per week46|
|Saturated fatty acids||Adverse||Strong||Should be < % of total energy intake17|
|Trans-unsaturated fatty acids||Adverse||Moderate||Should be minimized44|
|Protein||Can increase insulin response||Moderate||0.8 g high-quality protein/kg body weight. Less than 20% of energy44|
|Fibre||Beneficial||Strong||At least 14 g/1000 kcal44|
|Alcohol, moderate||Possibly beneficial||Moderate||Men – two drinks/day or fewer. Women – one drink/day or fewer17|
|Alcohol, excessive||Adverse||Moderate||Not advised17|
|Vitamins C, E, carotene||Unclear||Moderate||Not advised47|
|Fructose (sweetener)||Unclear||Moderate||May adversely affect lipid profle48|
|Sorbitol, xylitol, tagalose, lactitol||Unclear||Moderate||No adverse efects48|
Excessive intake of total fat drives insulin resistance, irrespective of the composition of fatty acids (FAs) in the diet,36 and possibly related to increases in intramyocellular lipid content and interference of binding of insulin to its receptors.9 However, many overweight/obese patients have difficulties adhering to these diets, particularly in the longer term, resulting in only limited success. Apart from the total amount of dietary fat consumed, the type of FA appears to exert relevant additional effects on the modulation of insulin resistance, especially under conditions of a more moderate fat intake (< 30%).36
Effects of specific dietary fatty acids on insulin resistance and diabetes risk
Dietary fat is a heterogeneous blend of different FAs, with monounsaturated fatty acids (MUFAs), polyunsaturated fatty acids (PUFAs), saturated fatty acids (SFAs) and trans-unsaturated fatty acids (TFAs) as the main components.9 A high intake of TFA may lead to insulin resistance and shows adverse effects on CVD.8,37,38 Many bakery products and high-energy pre-packed foods contain sufficient amounts of TFAs and SFAs to increase insulin resistance and risk of diabetes.39 Epidemiological studies indicate a direct relation of dietary SFA with the incidence of insulin resistance or T2DM,40,41 whereas replacing SFAs by MUFAs reduces insulin resistance36 and has a beneficial effect on blood pressure, low-density lipoprotein (LDL)-cholesterol and triacylglycerols.42 However, recent research indicates that specific SFAs differ widely in function, structure and metabolic effects, with some SFAs having important and specific biological roles.43
The underlying mechanisms involved in MUFA-induced improvement of insulin sensitivity are not completely understood but may involve effects on cell membrane FA composition,8 with functional effects on membrane fluidity, ion permeability, insulin receptor binding/affinity8 and up-regulation of glucose transporters.9 Further involved mechanisms might be related to alterations in incretin responses and beta-cell function.9 Despite these findings, an increased intake of MUFA was not associated with reduced risk of T2DM in prospective cohort studies.49
There also appear to be relevant interspecies differences when comparing metabolic effects of specific FAs. In humans, n-6 PUFAs may improve insulin resistance and diabetes risk,8 whereas the reported beneficial effects of n-3 PUFAs consumption from marine origin, as shown in rodent models, do not appear to apply in humans.4,8 Importantly, no long-term randomized trials have been published to date that have investigated the effect of dietary fat composition on diabetes risk.
Effects of dietary fibre and carbohydrates
The modest success of low-fat diets has prompted research on alternative strategies, including low-carbohydrate diets, which are often high in dietary protein, and carbohydrate-rich diets, which are aimed at modulating the post prandial glucose and insulin responses and thus the GI of foods. Carbohydrates with a high GI lead to rapid onset and pronounced increases in post prandial glucose and insulin concentrations that may compromise fat oxidation, fuel partitioning and metabolic flexibility.11,50 High-GI diets have been linked to insulin resistance in epidemiological observations, whereas low-GI diets improved insulin sensitivity in patients with T2DM.11 However, separating the effects of single nutrients in complex foods on metabolic outcomes is not straightforward: many low-GI diets are also rich in cereal fibres which are insoluble in water and have only negligible effects on the GI, but are strongly and independently associated with reduced risk of T2DM in prospective cohort studies.10,12,13,51 It cannot be excluded that at least some of the effects attributed to a low GI of carbohydrate-rich foods may be related to the cereal fibre content of the diet.10,52 Indeed, meta-analyses of large prospective cohort studies consistently show a 20–30% reduction in the risk of developing T2DM in subjects consuming diets high in cereal fibre [relative risk for extreme quintiles (RR) 0.67; 95% confidence interval (CI) 0.62 to 0.72],13 whereas results regarding the protective effects of low-GI [and glycaemic load (GL)] foods are less consistent.10 Notably, main sources of cereal fibre in the large prospective cohort studies in the USA are cellulose and hemicelluloses from wheat bran10 that are insoluble in water, non-viscous and only moderately fermentable,10,53 whereas the main sources of soluble, viscous and fermentable fibre are typically fruit and vegetables. As many assume that the strong associations between fibre intake and reduced diabetes risk12,13 are mainly related to the viscous and/or gel-forming properties of soluble fibre from fruit and vegetables, as such beneficially influencing GI and blood lipids, and to metabolic effects of short-chain fatty acids (SCFAs) derived from colonic fermentation of non-digested fibre by the gut microbiota,10,54–57 it is surprising that the intake of neither fruit (RR 0.96; 95% CI 0.88 to 1.04) nor vegetables (RR 1.04; 95% CI 0.94 to 1.15) shows any significant associations with reduced risk of developing T2DM.13 This is an unexpected but consistent finding, strongly indicating that other mechanisms are likely to be involved.
Indeed, insoluble cereal fibre intake, under isoenergetic conditions, increases whole-body insulin sensitivity in both short-term and more prolonged studies, as measured using euglycaemic–hyperinsulinaemic clamps.10,58,59 These effects appear to be dose dependent10 but independent of colonic fermentation, changes in dominant groups of the gut microbiota or circulating glucagon-like peptide-1 (GLP-1).10,53,59 We have recently proposed a novel concept that could help explain the improved insulin sensitivity resulting from cereal fibre intake, showing that cereal fibre may simply hinder the digestion and/or absorption of dietary protein in the upper gut, thereby preventing amino acid-induced activation of the mammalian target of rapamycin (mTOR)/translation initiation factor serine-kinase-6-1 (S6K1) signalling pathway that is known to drive insulin resistance.59–61 However, further, as yet unknown, mechanisms may also be involved and need to be investigated in future studies. Examples include cereal fibre-induced modulation of gut hormones, adipokines, bile acid binding and metabolite profiles.
Effects of dietary protein
Many energy-reduced diets are difficult to follow because they require elimination of certain foods, leading to poor adherence and limited success. The only moderate effects of low-fat diets on weight reduction62 have led to a renaissance of various alternative concepts with the aim of promoting weight loss. The most common concepts include carbohydrate restriction and increasing the intake of dietary protein. Low-carbohydrate diets are attractive because they promise rapid weight loss without having to count calories, despite allowing the consumption of many palatable foods.62 High-protein diets have beneficial effects on blood lipids, body composition and weight loss, at least in the short term.28 Better weight loss with high-protein diets may be explained by the satiating effects of dietary protein, reduced choice of foods and aversion to high-dietary-fat contents in the absence of carbohydrates. Reducing the protein content of the diet from 15% to 10% results in a higher total energy intake, predominantly from savoury-flavoured foods available between meals,63 reinforcing the finding that a higher dietary protein intake may help to reduce energy intake.
However, sustained weight loss with any diet is difficult to achieve, and the long-term safety of high-protein diets remains to be investigated34,64 In the DiOGenes trial, weight regain at 1 year was only marginally lower with a high-protein intake, but both the high-protein and the high-GI diets were shown to increase low-grade inflammation,65 which could result in worsening of insulin resistance. Indeed, recent data indicate that low-carbohydrate plant-based high-protein diets, even with an amino acid profile that is assumed to have beneficial metabolic effects, may induce insulin resistance, at least under isoenergetic conditions.59 Furthermore, healthy humans who are exposed to amino acid infusions rapidly develop insulin resistance,60 with inhibition of glucose uptake being driven through phosphorylation of downstream factors of the insulin signalling cascade by S6K1.60,61 In agreement with this, long-term high protein intake has been shown to result in whole-body insulin resistance,59,66 associated with up-regulation of factors involved in the mTOR/S6K1 signalling pathway,59 increased stimulation of glucagon and insulin within the endocrine pancreas, high glycogen turnover66 and stimulation of gluconeogenesis.59,66 In the short-term, these negative effects on insulin sensitivity may be compensated by high-protein diet-induced weight loss, and, at least in physically active people, relevant increases in lean mass that are also mediated via the mTOR/S6K1 pathway.61 However, most subjects on weight loss diets are overweight/obese and typically sedentary; therefore, relevant increases in lean mass under these conditions are unlikely. In further agreement that high-protein diets cause deterioration of glucose metabolism, it was recently shown in a large prospective cohort with 10 years' follow-up that consuming 5% of energy from both animal and total protein at the expense of carbohydrates or fat increases diabetes risk by 30%.64 This reinforces the theory that high-protein diets can have adverse effects on glucose metabolism.
Brand-Miller and colleagues67 have proposed the carnivore connection hypothesis, suggesting that during human evolution a scarcity of dietary carbohydrates together with high intake from animal proteins led to insulin resistance. This may have provided a survival and reproductive advantage by redirecting glucose from maternal use to fetal metabolism, thus increasing birth weight and survival of the offspring,67 but could be deleterious in a high-carbohydrate environment. In this context it is interesting that populations which have only recently switched from traditional high-protein hunter–gatherer-type diets to modern high-carbohydrate diets show an excessively high prevalence of insulin resistance and T2DM compared with European populations that switched to a higher carbohydrate intake some 12 000 years ago.67,68
The role of exercise in the prevention and treatment of type 2 diabetes mellitus
Regular exercise is another cornerstone of diabetes management, along with dietary and pharmacological interventions.3,17 Both aerobic exercise and resistance training improve insulin action, at least acutely, and may have a beneficial effect on quality of life, glucose control, blood lipids, systolic blood pressure, cardiovascular risk and mortality.3 Current guidelines recommend that patients with T2DM should perform at least 150 min per week of moderate-intensity aerobic exercise and should perform resistance exercise three times per week.3,17 The US Department of Health recommends that adults > 65 years or those with disabilities follow the above recommendations if possible, or be as physically active as they are able.69
The National Institutes of Health (NIH)-AARP Diet and Health Study, in 1995 to 1996, enrolled 114 996 men and 92 483 women, aged 50–71 years without evidence of heart disease, cancer or diabetes. Incident self-reported, physician-diagnosed diabetes was identified with a follow-up survey in 2004–2006. Men and women whose diet score, physical activity level, smoking status and alcohol use were all in the low-risk group had odds ratios for diabetes of 0.61 (CI 0.56 to 0.66) and 0.43 (CI 0.34 to 0.55), respectively. When absence of overweight or obesity was added, the corresponding odds ratios were 0.28 (CI 0.23 to 0.34) and 0.16 (CI 0.10 to 0.24) for men and women, respectively. This indicates that lifestyle factors, when considered in combination, are associated with a substantial reduction in the risk of diabetes.70 Independent of exercise levels, sedentary behaviour, especially excessive television watching, is associated with a significantly elevated risk of obesity and T2DM, whereas even light to moderate activity may lower this risk.71,72
In a Danish study, 4031 patients with impaired fasting glucose or impaired glucose tolerance underwent oral glucose tolerance tests at baseline and after five years to reassess their status.73 Leisure-time physical activity at baseline was assessed by questionnaire. Physical activity was associated with a lower risk of progression to diabetes in the total study population and in individuals with impaired glucose tolerance, a condition primarily characterized by muscle insulin resistance. However, it did not predict progression to diabetes in individuals with impaired fasting glucose.73 There is some evidence that recreational physical activity performed during the year before and during the first 20 weeks of index pregnancy was also associated with significant reductions in risk of gestational diabetes.74 However, although physical activity is a key element in the prevention and management of T2DM,75–82 many people with T2DM are not active.16 Furthermore, reported differences in the effects of aerobic versus resistance training, and of unstructured physical activity versus structured exercise programmes, on the risk of developing T2DM and metabolic control in patients with T2DM are incompletely understood and controversial.
Exercise intensity, aerobic exercise and resistance training
Physical activity can result in acute improvement in insulin sensitivity lasting 2–72 hours.3 Physical activity must be regular to have continued beneficial effects, and probably needs to include varying types of exercise.3 A combination of aerobic and resistance training may be more effective for the management of glucose control than either type of exercise alone.83,84 Any increase in muscle mass that may result from resistance training could contribute to glucose uptake, whereas aerobic exercise improves insulin action independent of changes in muscle mass or aerobic capacity.83 The most successful programmes for long-term weight control have involved combinations of diet, exercise and behaviour modification.3 Physical exercise, and especially combined aerobic/resistance exercise, in T2DM patients with the metabolic syndrome may also result in improvement of markers of insulin resistance and inflammation, independent of weight loss, although results are controversial.85–87
Although the beneficial effects of regular exercise on physical and mental well-being are beyond doubt, less is known about their effects on weight loss, insulin resistance and diabetes risk. Aerobic exercise has been shown to improve whole-body insulin sensitivity as measured by euglycaemic–hyperinsulinaemic clamp, but there may be no significant improvement in hepatic insulin sensitivity.88 Several studies have investigated the effect of aerobic exercise on surrogate markers of cardiovascular risk in patients with T2DM. A recent meta-analysis showed significant improvements in systolic blood pressure (−6.08 mmHg; 95% CI −10.79 mmHg to −1.36 mmHg) and triglycerides (−0.3 mmol/l; 95% CI −0.48 mmol/l to −0.11 mmol/l). Combined aerobic/resistance exercise also improved these parameters, but to a lesser extent.89 Aerobic training also results in a modest increase in high-density lipoprotein-cholesterol (HDL-C) concentrations.90 However, this could have a significant effect on coronary heart disease, since a 1% increase in HDL-C levels is associated with a 3.5% decrease in mortality.91,92 Aerobic exercise of at least 3 months' duration has also been shown to decrease arterial stiffness in large arteries in a group of older adults with T2DM, hypertension and hypercholesterolaemia.93 In a study involving middle-aged obese women, high-intensity (more than lactate threshold) aerobic exercise effectively reduced total abdominal fat, subcutaneous abdominal fat and abdominal visceral fat.94 However, this benefit of aerobic exercise was not replicated in other studies.95,96 In non-dieting, overweight subjects, a higher amount of activity appears to be necessary for weight maintenance.
Resistance training, even when not associated with weight loss, improves insulin sensitivity and fasting glucose and has favourable impact on body composition.97,98 There might be subtle differences in the effect of resistance exercise in different ethnic groups. In a study involving Asian Indians, insulin resistance improved significantly following 12 weeks of training without any change in BMI, levels of total body fat, abdominal fat or lean body mass or cross-sectional skeletal muscle area of the extremities.99 A recent randomized trial examined the effect of aerobic and resistance exercise in a group of Caucasian and Afro-Caribbean patients with T2DM. Afro-Caribbean subjects responded more favourably to resistance training than did Caucasians, showing reductions in BMI (−2.57% ± 0.90% vs 2.57% ± 1.09%; P < 0.01) and improved markers of insulin resistance (−19.15% ± 9.00% vs 13.12% ± 11.86%; P < 0.05).100
Table 2 summarizes results of studies that reported effects of aerobic and/or resistance training on HbA1c.
|Source||Type of intervention||Dietary intervention||Intervention (number)||Control (number)||Frequency and duration||Dropout (%)||HbA1c change (%)|
|Church et al. (2010)105||Aerobic training||No||72||41||3/week, 39 weeks||Intervention 4; control 10||−0.23 (−0.30 to −0.16)|
|Sigal et al. (2007)84||Aerobic training||No||60||63||3/week, 26 weeks||Intervention 20; control 5||−0.50 (−1.22 to +0.22)|
|Sridhar et al. (2010)108||Aerobic training||No||55||50||5/week, 52 weeks||Not reported||−2.76 (−3.13 to −2.39)|
|Church et al. (2010)105||Resistance training||No||73||41||3/week, 39 weeks||Intervention 5; control 10||−0.15 (−0.22 to −0.08)|
|Sigal et al. (2007)84||Resistance training||No||64||63||3/week, 26 weeks||Intervention 11; control 5||−0.37 (−1.08 to +0.34)|
|Balducci et al. (2004)109||Aerobic and resistance training||No||51||53||3/week, 52 weeks||Intervention 18; control 9||−1.24 (−1.88 to +0.60)|
|Church et al. (2010)105||Aerobic and resistance training||No||76||41||3/week, 39 weeks||Intervention 5; control 10||−0.34 (−0.41 to −0.27)|
|Sigal et al. (2007)84||Aerobic and resistance training||No||64||63||3/week, 26 weeks||Intervention 13; control 5||−0.97 (−1.69 to −0.25)|
|Christian et al. (2008)110||Advice||Yes||141||132||52 weeks||Intervention 9; control 13||−0.60 (−1.00 to −0.20)|
|Di Loreto et al. (2003)111||Advice||Yes||182||158||104 weeks||Intervention 2; control 2||−0.50 (−0.62 to −0.38)|
|Hordern et al. (2009)112||Advice||Yes||88||88||52 weeks||Intervention 21; control 21||−0.70 (−1.08 to −0.32)|
|Jakicic et al. (2009)113||Advice||Yes||2486||2575||52 weeks||Intervention 4; control 4||−0.50 (−0.56 to −0.44)|
|Mayer-Davis et al. (2004)114||Advice||Yes||49||56||52 weeks||Total 19||−0.44 (−0.87 to −0.01)|
|van Rooijen et al. (2004)115||Advice||Yes||75||74||12 weeks||Intervention 6; control 4||+0.62 (−0.14 to +1.38)|
Effects of structured exercise interventions
Structured interventions that consist of aerobic exercise, resistance training, or both, combined with modest weight loss, may lower risk of T2DM by up to 58% in high-risk populations.3 Exercise training of more than 150 min per week is associated with greater HbA1c reduction than that training for 150 min or less per week. Previous meta-analyses103,104 found that structured exercise including aerobic and resistance training reduces HbA1c levels by approximately 0.6%. However, only one previous review separately analysed associations of aerobic exercise, resistance training and the combination of aerobic exercise and resistance training on change in HbA1c levels;103 differences among the effects of aerobic, resistance and combined training on HbA1c were marginal, with all forms of exercise training producing small improvements in HbA1c.104 Since the publication of this meta-analysis, two large randomized trials84,105 have reported contradictory findings regarding the types of structured exercise associated with declines in HbA1c levels. Sigal et al.84 found that aerobic or resistance exercise training alone improved glycaemic control, but the effects were more pronounced when combined. In contrast, Church et al.105 observed that only the combination, but not aerobic and resistance training alone, reduced HbA1c levels. Although high-intensity exercise has been previously shown to have an association with HbA1c reduction,106 others have failed to demonstrate that more intensive exercise is associated with greater declines in HbA1c.107
A recent systematic review and meta-analysis of 47 randomized controlled trials of structured exercise programmes with or without dietary intervention in subjects with T2DM showed overall HbA1c reduction of −0.67% (−0.84% to −0.49%) for subjects in an intervention group involving exercise and diet when compared with controls. In addition, structured aerobic exercise (−0.73%), structured resistance training (−0.57%) and both combined (−0.51%) were each associated with declines in HbA1c. Structured exercise of total duration more than 150 min per week was associated with HbA1c reductions of 0.89%, whereas structured exercise durations of 150 min or less per week were associated with HbA1c reductions of 0.36%. Physical activity alone without dietary input did not have any impact on glycaemic control.107 An earlier Cochrane review analysed 14 randomized controlled trials involving 377 participants comparing exercise with no exercise in subjects with T2DM. Trials ranged in duration from 8 weeks to 12 months. Compared with the control group, exercise intervention significantly improved glycaemic control, as indicated by a decrease in HbA1c of 0.6% (−0.9% to −0.3%; P < 0.05). There was no significant difference between groups in whole-body mass, probably because of an increase in fat-free mass (muscle) with exercise.104
Supervised brisk walking is possibly the easiest and most cost-effective way to improve physical activity. However, randomized controlled trials involving brisk walking three times a week in subjects with T2DM have found no significant improvement in HbA1c or other metabolic parameters, possibly because of the observed high dropout rates of up to 50%, which were attributed to injuries and lack of motivation. However, among participants who attended at least 50% of walking sessions, discontinuation or reduction in anti-diabetic drugs was achieved in one-third of subjects.116,117
Use of a pedometer with a recommendation to walk at least 10 000 steps a day may give a more objective assessment of active lifestyle. Randomized controlled trials have found a significant improvement in physical activity, but no change in metabolic parameters.118–120 Again, the dropout rates are reported to be considerable, especially in the subgroups of subjects who have poor physical fitness, suggesting that the persons most in need of physical exercise are the least compliant in exercise programmes.118–120 After discontinuation of exercise programme, the effects tend to taper off over the following 6–12 months, presumably because of lack of adherence. Additional support in the form of community centre-based exercise or physical therapist-directed counselling seems to maintain physical activity following cessation of supervised exercise.121–123
Unstructured physical activity
In contrast to structured exercise training, physical activity is defined as any bodily movement produced by skeletal muscle contractions resulting in increased energy expenditure.124 Although structured exercise training may be available to a subset of patients with T2DM, physical activity advice is more feasible and should be offered to most of these patients. However, meta-analyses have not been performed to determine whether physical activity advice is associated with similar declines in HbA1c as those associated with structured exercise. In one study, a recommendation to increase physical activity was beneficial (0.43% HbA1c reduction), but only if combined with dietary instructions.107 In 1030 patients involved in the Finnish diabetic nephropathy study, leisure-time physical activity was assessed by a 12-month validated questionnaire expressed in metabolic equivalent units. There was a weak correlation between physical activity (r = 0.12; P = 0.007) and HbA1c level in women, but not in men. Age, obesity, smoking, insulin dose, social class, diabetic nephropathy or CVD did not explain the results.125 An earlier study evaluated 50 patients with type 1 diabetes mellitus (T1DM), 50 patients with T2DM and 70 control subjects using a questionnaire to determine the type, duration and intensity of exercise. There was no correlation between the degree of activity and HbA1c levels or hypoglycaemic events. A negative correlation was found between physical activity and daily insulin usage (r = −0.27; P < 0.05).126 A cross-sectional survey of physical activity in 221 patients with T1DM did not find any correlation with glycaemic control, but insulin requirements were lower (r = −0.20; P = 0.002) in more active subjects.127
In summary, a structured exercise programme appears to be superior to advice alone in improving physical activity levels. However, there are obvious cost and logistic implications in providing such a programme for all patients with diabetes.
Exercise and improvement in quality of life
Rejeski et al.128 found that, among overweight and obese adults with T2DM, an intensive lifestyle intervention led to a relative reduction of 48% in the severity of mobility related disability, as compared with diabetes support and education. This effect was mediated by both weight loss and better fitness. Even without changes in body weight and glycaemic control, exercise programmes have consistently shown to result in improvements in self-reported physical activity and health-related quality of life. Depressive symptoms and mental health status with aerobic and resistance exercise may also improve.129–131 A relatively short duration (3 months) of vigorous aerobic exercise improves orthostatic tolerance and arterial baroreflex function in elderly patients at high cardiovascular risk.132,133 A balance exercise programme in older adults with diabetic neuropathy improves balance and trunk proprioception possibly preventing falls.134 Even though many overweight/obese patients with T2DM and coexistent morbidities may not be able to perform exercise at a level that is sufficient to result in relevant weight loss and major improvements in glycaemic control, additional benefits of even moderate increases in physical activity in these patients appear to be important.
Excessive energy intake in relation to physical activity leads to adiposity and insulin resistance and has been proposed as the strongest single predictor for T2DM,6 whereas weight loss is almost always associated with improved insulin sensitivity.4 Thus, any lifestyle measures that lead to weight loss and can be sustained in the long term have the potential to improve glycaemic control. Energy-reduced diets should be balanced to avoid potential detrimental effects on health, and long-term success appears to be mainly determined by the adherence to the diet, regardless of the chosen dietary strategy. However, particularly in patients with T2DM, long-term sustained weight loss appears to be difficult to achieve. In this situation isoenergetic changes in the macronutrient composition and the quality of ingested foods may exert additional important effects on insulin sensitivity, independent of weight loss. Dietary measures that may improve insulin sensitivity include using a Mediterranean-like dietary pattern, but avoiding excess intake of dietary fat, replacing SFAs and TFAs with MUFAs and n-6 PUFAs and increasing cereal fibre intake and exercise levels, particularly when choosing a low-carbohydrate, high-protein diet. It is also noteworthy that intentional weight loss through any diet and/or exercise causes not only loss of excess fat mass, but also some loss of lean body mass. An ideal weight management programme should therefore maximize fat loss and minimize the loss of muscle mass.
Most people with T2DM can perform exercise safely, but precautions must be taken in patients with severe obesity, CVD or diabetes-related neuropathic damage. Structured exercise programmes appear to be superior to simple advice to increase physical activity levels.
Weight loss, the macronutrient composition of the diet, aerobic exercise and resistance training all appear to improve insulin resistance, by distinct mechanisms. Therefore, a combination of these interventions tailored to the requirements of each subject should be one of the cornerstones of management of overweight/obese patients with T2DM.3,17,135