INTRODUCTION — A single session of exercise increases non-insulin-mediated and insulin-mediated glucose uptake. Insulin sensitivity can be enhanced for up to 48 hours after exercise. While the acute effects of exercise on insulin sensitivity are temporary, regular sessions of aerobic and resistance exercise promote adaptations in the vasculature, skeletal muscle mass and oxidative capacity, adipose tissue, and the liver that further enhance insulin sensitivity [1].
These same adaptations likely occur with exercise training in type 2 diabetes [2], which tend to improve overall glycemic management and also result in other favorable outcomes, including reductions in body weight and improvements in cardiovascular risk profile [3]. The general benefits of regular exercise with regards to weight loss and improving insulin sensitivity and glucose disposal apply predominantly to adults with type 2 diabetes and to the increasingly overweight and obese type 1 population. In non-obese individuals with type 1 diabetes, exercise training may have more modest effects on insulin sensitivity, glycemia, and body weight, but can still improve cardiorespiratory fitness and the cardiovascular risk profile [4].
This topic will review when exercise has differential salutary effects and side effects depending on diabetes type and also provide exercise guidance specifically for people with diabetes. The glycemic benefits of intensive lifestyle interventions in patients with diabetes, exercise for the prevention of type 2 diabetes, and sample cases illustrating problems that can occur when exercise is performed in patients treated with insulin are discussed separately. The effects of exercise in people without diabetes is also reviewed elsewhere.
●(See "Prevention of type 2 diabetes mellitus", section on 'Exercise'.)
●(See "Effects of exercise on lipoproteins and hemostatic factors".)
●(See "The benefits and risks of aerobic exercise".)
EXERCISE AND MUSCLE METABOLISM — Exercise has both short- and long-term effects on energy metabolism. Skeletal muscle is the largest organ for glucose uptake, storage, and oxidation. Several types of exercise (including traditional aerobic, resistance, and high-intensity interval, low-volume training) have been shown to improve skeletal muscle oxidative capacity and glucose uptake in people without diabetes [5-10].
For most individuals living with diabetes, exercise training also improves muscle mechanics, muscle insulin sensitivity, and increases glucose disposal (mitochondrial oxidation and storage as muscle glycogen) [2,3]. However, the physiologic response to exercise (both short- and long-term) may be different in people with diabetes.
Short-term effects — During the first few seconds of muscle contractions, striated muscles use free adenosine triphosphate (ATP), which is immediately resynthesized from phosphocreatine, and then over the next few minutes, anaerobic glycolysis to sustain muscle contractions. After approximately five minutes of exercise, ATP resynthesis occurs primarily by catabolizing other fuel sources (eg, carbohydrates and lipids), both from within and outside of the muscle. The longer the exercise duration, the greater the body's reliance on blood glucose and blood lipids as fuel (figure 1) [11].
●Skeletal muscle glucose uptake – A single session of exercise increases glucose uptake into skeletal muscle severalfold, largely via mechanisms that are independent of insulin signaling [1]. Exercise-mediated glucose uptake occurs largely via muscle contraction-mediated increases in glucose transporter 4 (GLUT4) translocation to the plasma membrane, complemented by increased rates of glucose delivery to muscle via increased blood flow and increased rates of muscle glucose metabolism (ie, glycolysis and glucose oxidation) (figure 1) [12-14]. The activation of the contraction-sensitive GLUT4 proteins likely occurs via activation of several kinases that "sense" and transduce signals related to changes in the intracellular environment during muscular contractions, including calcium ion levels and adenosine monophosphate levels [12,15,16]. The non-insulin-mediated uptake and oxidation of glucose is largely responsible for the acute blood glucose-lowering effect of exercise. (See 'Exercise benefit in diabetes' below.)
During exercise recovery, lasting one to four hours depending on the nature of the activity, there is a gradual decay of muscle contraction-mediated glucose uptake. At the same time, however, insulin sensitivity in the previously active muscle increases and remains elevated for up to 48 hours after exercise, depending on the intensity and duration of the activity [17,18]. The increase in insulin sensitivity and glucose uptake after exercise are thought to be important for muscle glycogen replenishment; however, the window of time of enhanced insulin sensitivity is unrelated to muscle glycogen synthase activity, muscle glycogen content, or degree of glycogen utilization during the preceding session of exercise [19]. This enhancement of peripheral glucose uptake and increased insulin sensitivity is particularly beneficial for individuals with type 2 diabetes, but can increase hypoglycemia risk for individuals with type 1 diabetes or insulin-requiring type 2 diabetes [20]. (See 'Exercise benefit in diabetes' below and 'Glycemic management during exercise' below.)
While the acute effects of exercise on insulin sensitivity are temporary, regular training promotes adaptations in the vasculature, skeletal muscle, adipose tissue, and the liver that further enhances insulin sensitivity [1]. (See 'Long-term effects' below.)
•Type 2 diabetes – In the resting state, there is a reduction in insulin sensitivity in skeletal muscle in people with type 2 diabetes (see "Pathogenesis of type 2 diabetes mellitus"). The molecular mechanisms behind insulin resistance in skeletal muscle involve a complex network of inter-tissue cross-talk and several indirect and direct effects within the myocyte that impair insulin signaling, including the following [21]:
-Increased intramyocellular lipid uptake and lipid droplet formations in muscle with the accumulation of lipid intermediates (diacylglycerol and ceramides) that inhibit the insulin receptor signaling cascade and GLUT4 translocation capacity (in the resting state)
-Reduced glucose uptake, glucose oxidation rates, and rates of glycogen synthesis
-Reduced skeletal muscle mitochondrial function with chronically elevated reactive oxygen species (ROS) levels that further reduce muscle insulin sensitivity and impair glucose oxidative capacity
-A low maximal oxygen uptake (VO2max) during exercise
Exercise overcomes the defects in glucose transport caused by insulin resistance. During exercise, skeletal muscle contraction-mediated (non-insulin-dependent) GLUT4 translocation occurs normally in people with type 2 diabetes [22]. During recovery, insulin sensitivity in the previously active muscle increases and remains elevated for up to 48 hours after exercise, depending on the intensity and duration of the activity [17,18]. These acute effects of exercise can impact glycemia in diabetes, often resulting in a lowering of glucose concentrations during the activity and for several hours after the activity, unless very intensive short bursts of exercise are performed where glucose levels may transiently rise [23,24]. (See 'Exercise benefit in diabetes' below.)
•Type 1 diabetes – In the resting state, muscle glucose kinetics are largely intact in people with type 1 diabetes. During exercise, the glucose response is variable depending on time of day, type of exercise, timing and amount of insulin administered, and food intake [4]. In the fasting state and when insulin levels are at a basal concentration, glucose uptake into exercising muscle is relatively predictable [4,25,26]. When the circulating insulin levels are higher (eg, two- to threefold above basal as can occur after meals), there is an increase in insulin-mediated glucose uptake during exercise [27], which likely contributes to elevated risk for hypoglycemia. (See 'Glycemic management during exercise' below.)
●Insulin-to-glucagon ratio – Based on an exercise-mediated increase in sympathetic drive and as the blood glucose concentration begins to fall, the secretion of insulin decreases, while that of glucagon rises directly in the portal vein. The net effect of the decrease in the insulin-to-glucagon ratio in the portal circulation is increased hepatic glucose production due both to glycogenolysis and to gluconeogenesis (in which glucose is formed from lactate, pyruvate, alanine and other amino acids, and glycerol) (figure 2). Skeletal muscle differs from liver in that it lacks the enzyme glucose-6-phosphatase, which converts glucose-6-phosphate (derived from glycogen) to glucose; as a result, muscle glycogen can only be used as an energy source for muscle via the metabolism of glucose-6-phosphate to pyruvate (figure 2). In other words, glucose cannot be transferred out of muscle to prevent exercise-induced hypoglycemia.
In type 1 and insulin-treated type 2 diabetes, insulin levels do not fall physiologically and may even increase owing to increased absorption of exogenous insulin from exercising muscle [25,28]. The physiologic response to exercise, therefore, depends on the serum insulin concentration at the time of exercise which, in turn, is influenced by the site and timing of recent basal and or bolus insulin administrations (multiple daily injections [MDI] or continuous subcutaneous insulin infusion [CSII]).
Even people with well-controlled type 1 diabetes and adequate serum insulin concentrations at the time of exercise are vulnerable to falling blood glucose concentration that is much larger than that in individuals without diabetes [29]. Several factors contribute to this response:
•Exogenous insulin cannot be shut off, resulting in prolonged inhibition of hepatic glucose output and potentially increases in glucose disposal rates [27].
•Exogenous insulin concentrations may even rise with exercise, due to increased temperature and blood flow associated with exercise, which may speed insulin absorption from subcutaneous depots. This effect is most prominent if the injection was recent, was given into an arm or leg that is being exercised [30], or was inadvertently given intramuscularly [31].
•Exercise one to two hours after a meal with the usual prandial insulin bolus is particularly problematic for those with type 1 diabetes since this aggravates the relative hyperinsulinemia, resulting in increased hypoglycemia risk [32].
●Other counterregulatory hormones – As the duration of moderate-intensity exercise increases beyond an hour or more, other counterregulatory hormones (ie, epinephrine, norepinephrine, growth hormone, and cortisol) rise and help sustain hepatic glucose production, blood glucose levels, and shift substrate utilization more towards lipid oxidation. Epinephrine and norepinephrine stimulate hepatic glucose production to some extent, but their major effect is to stimulate lipolysis and to limit skeletal muscle glucose uptake and utilization. These processes help to protect against exercise-associated hypoglycemia.
Triglycerides are broken down into both free fatty acids (which are utilized as fuel by exercising muscles) and glycerol (which can be converted to glucose in the liver). These changes in hormone release and muscle metabolism become more pronounced with prolonged exercise; insulin secretion continues to fall, while there is a further increase in the release of the counterregulatory hormones. The net effect is a gradual reduction in whole-body carbohydrate utilization (muscle and blood glucose) and a greater reliance on lipid mobilization (ie, lipolysis) and oxidation (figure 1) [5,33].
Mild- to moderate-intensity exercise can cause a paradoxical elevation in blood glucose concentrations in patients with type 1 diabetes, but typically only if insulin is withheld for a prolonged period of time and severe hypoinsulinemia occurs [29]. This phenomenon is less likely to occur in the modern age of basal insulins with longer-acting profiles but might occur if there is a prolonged (ie, hours) insulin pump disconnect [4]. Elevations in glucose counterregulatory hormones, along with diminished insulin levels in diabetes, enhance lipolysis but may also result in excessive ketone production during or after exercise [29].
In addition to prolonged mild- to moderate-intensity exercise, short bursts of very high-intensity (ie, explosive) exercise [23] and high-intensity interval exercise [24] can also promote a rise in glucose levels and hyperglycemia because of elevations in glucose counterregulatory hormones and high sympathetic drive (figure 3). (See 'Type and frequency of exercise' below.)
Long-term effects — Moderate aerobic exercise on a regular, long-term basis (ie, endurance training) has several beneficial effects on skeletal muscle mass and function that lead to more efficient use of energy (glucose and lipid), a greater fuel oxidative capacity, and a reduction in whole-body insulin needs. Regular exercise also promotes adaptations in adipose tissue and the liver that further enhance insulin sensitivity [1]. Much of the long-term effects of regular exercise also occur for people with type 1 [34] and type 2 diabetes [2,35]. These changes include [5,36]:
●Increases in the number and quantity of mitochondria and mitochondrial enzymes and the relative expression of "slow-twitch" muscle fibers.
●The development of new muscle capillaries with enhanced vasoactive capabilities.
●Increases in the translocation of both the insulin-responsive and exercise-responsive GLUT4 proteins from intracellular stores to the cell surface.
GLUT4 appears to be a rate-limiting step in muscle glucose uptake, which probably explains much of the overall increase in insulin sensitivity. Improvements in blood flow, increases in muscle mass, enhanced skeletal muscle mitochondrial function, and increased intracellular glucose flux (including glucose oxidation and glycogen replenishment rate) also contribute to the beneficial effect [13].
With repeated exercise sessions over the longer term (ie, weeks to months) there are also favorable changes in body composition (eg, muscle and adipose tissue mass changes), intramyocellular lipid deposition, hepatic and blood lipid levels, and whole-body basal (resting) and meal-associated insulin signaling [37]. There are also training-induced improvements in skeletal muscle mitochondrial expression and function (thereby improving muscle energetics), muscle capillary recruitment, cardiac output, and maximal aerobic capacity (VO2max) [9,38-41].
Emerging evidence in type 2 diabetes also demonstrates that exercising muscle releases health-promoting factors (myokines) that can improve insulin sensitivity in other tissues including adipose tissue, liver, pancreas, and even regions of the central nervous system [37]. While weight loss via dietary restriction also reduces insulin resistance, some evidence suggests that exercise training-induced weight loss may have a greater effect than weight loss without exercise [42,43]. Typically, the greatest improvements in insulin sensitivity in type 2 diabetes is observed when the exercise programming includes both endurance (aerobic) and resistance (anaerobic) exercise [44-46]. (See 'Type and frequency of exercise' below.)
EXERCISE BENEFIT IN DIABETES
Type 2 diabetes — Regular sessions of aerobic and resistance exercise maintain skeletal muscle mass and oxidative capacity, increase whole-body insulin sensitivity, and increase overall muscle glucose uptake, thereby improving glycemia [15,45].
●Acute glycemia – A single session of exercise increases glucose uptake into skeletal muscle severalfold, largely via mechanisms that are independent of insulin signaling [1]. After exercise, insulin sensitivity in the previously active muscle is elevated for up to 48 hours in untrained adults, depending on the intensity and duration of the activity [17]. These acute effects of exercise can impact glycemia in diabetes, often resulting in a lowering of glucose concentrations during the activity and for several hours after the activity, and also decrease hepatic lipogenesis over the next several hours [47]. The overall improvement in glucose metabolism lasts anywhere from 2 to 72 hours after a single bout of exercise [15,38,48,49].
●Chronic glycemia – In several meta-analyses, improvement in exercise regimens over time reduced glycated hemoglobin (A1C) values by approximately 0.5 to 0.7 percentage points compared with control participants [16,35,50-52] and generally independent of weight loss. In a subsequent meta-analysis, the improvements in A1C were proportional to the volume of exercise performed in supervised exercise training studies, with maximal benefit observed at approximately 100 min/week [53]. (See 'Type and frequency of exercise' below.)
Traditional aerobic, resistance, and possibly high-intensity, low-volume interval-type exercise regimens can improve glycemia. When the duration of total accumulated physical activity is fixed (ie, 150 minutes/week), a program that combines aerobic exercise and resistance training is thought to be optimal [44,54]. Higher levels of exercise intensity are associated with greater improvements in A1C [52] and more time in the glycemic target range with less meal-related glucose excursions, as assessed by continuous glucose monitoring (CGM) metrics [55]. Long-term compliance is essential to achieve the glycemic benefits of exercise [56].
Although one large meta-analysis comparing different exercise modalities revealed only "trivial" differences in their beneficial effects on glycemia [16], a few trials suggest that combined aerobic and resistance training may improve A1C more than either type of exercise alone [44,54]. As examples:
•In a trial of 251 adults with type 2 diabetes randomly assigned to resistance, aerobic, combined exercise, or control groups, all three exercise groups were associated with improvements in A1C compared with the control group (absolute change in A1C -0.38 to -0.97 percentage points) [44]. The combined exercise program was associated with the greatest reduction in A1C (approximately 1 percentage point compared with control participants). However, the combined program was also associated with a longer duration of exercise than the other groups, which may account for the greater improvement in glycemic control.
•In another trial of resistance training, aerobic exercise, combined aerobic and resistance training, or no exercise in 262 sedentary adults with type 2 diabetes, the duration of exercise (140 to 150 minutes/week) was similar in all exercise groups [54]. After nine months, mean A1C decreased modestly in all exercise groups (-0.04 to -0.23 percentage points compared with a 0.11 percentage point increase in the control group). The absolute difference in A1C was significantly improved only in those assigned to the combined program compared with control participants (between-group difference 0.34 percent). The combined group also lost significantly more weight than the control and resistance training groups (-1.5, +0.4, and -0.3 kg, respectively), a finding that may account for the greater reduction in A1C.
High-intensity (low-volume) exercise may also improve glycemia in patients with type 2 diabetes [40,57,58]. The effect of the type of exercise on blood glucose is reviewed in more below. (See 'Type and frequency of exercise' below.)
●Cardiovascular disease (CVD) – In several randomized trials, exercise has been shown to improve cardiovascular risk factors (dyslipidemia, blood pressure, and body composition) in patients with type 2 diabetes [51]. There are no clinical trial data that demonstrate a reduction in major cardiovascular endpoints or mortality. The Look AHEAD controlled clinical trial, which included an exercise component in its intensive lifestyle intervention, failed to demonstrate a cardiovascular benefit over more than 10 years of study [59]. (See "Initial management of hyperglycemia in adults with type 2 diabetes mellitus", section on 'Intensive lifestyle modification'.)
In several follow-up studies and secondary analyses of the Look AHEAD cohort, lifestyle intervention with exercise improved several health-related metrics in people with type 2 diabetes [60,61], and an increase in physical activity levels by more than 100 metabolic equivalents (MET) min/week was associated with decreased risk of cardiovascular outcomes [62]. Similarly, in prospective cohort studies, regular exercise was associated with improvement in cardiovascular outcomes and a reduction in cardiovascular and overall mortality in people with type 2 diabetes [63-66]. In meta-analyses of prospective cohort studies examining the impact of exercise and cardiorespiratory fitness on all-cause mortality in type 2 diabetes, the highest level of physical activity was associated with lower overall mortality risk than the lowest (relative risk [RR] 0.60, 95% CI 0.52-0.70) [66,67]. Any positive findings from these and other cohort studies or post-hoc, secondary analyses must be tempered by the negative, long-term Look AHEAD clinical trial results where the benefit of intensive lifestyle intervention on CVD mortality was not observed [59].
Type 1 diabetes
●Chronic glycemia – In contrast to the response in type 2 diabetes, there is less evidence that regular exercise is associated with improved glycemia in patients with type 1 diabetes, presumably due to the lesser importance of insulin resistance in many of these patients [34,68-71]. In a meta-analysis of studies assessing the overall effects of exercise on chronic glycemia, aerobic exercise (12 studies) was associated with a small improvement in A1C (0.3 percentage points), whereas resistance training (two studies), combined aerobic and resistance training (four studies), and high-intensity exercise (one study) did not significantly improve chronic glycemia [68]. In a subsequent trial in overweight or obese individuals with type 1 diabetes, 12 weeks of high-intensity interval training did not significantly reduce A1C [69]. Percent time in range (a CGM metric), however, does appear to be better on active days as compared with sedentary days [70].
●CVD – There may be substantial beneficial effects of exercise on general well-being, hypertension, and other cardiovascular risk factors, independent of glycemic control [34]. As an example, in a prospective cohort study of 2639 individuals with type 1 diabetes, high- compared with low- or moderate-intensity exercise was associated with a reduction in overall and cardiovascular mortality (eg, cardiovascular mortality rates 6.7, 1.9, and 0.2 percent in the low-, moderate-, and high-intensity exercise groups) [72].
EXERCISE GUIDANCE — Exercise is recommended for adults with type 1 and type 2 diabetes to improve glycemia (mostly for type 2 diabetes) and insulin sensitivity, manage weight, and to reduce cardiovascular risk factors [3,73,74]. (See 'Exercise benefit in diabetes' above and "The benefits and risks of aerobic exercise", section on 'Benefits of exercise'.)
Guidance specifically for people with diabetes is discussed below. Guidance related to exercise in general is reviewed in detail elsewhere. (See "Exercise prescription and guidance for adults", section on 'Prescribing an exercise program' and "Exercise prescription and guidance for adults", section on 'Helping patients advance their exercise program'.)
Evaluation prior to recommending an exercise regimen — We typically perform a history (with assessment of cardiovascular risk factors and for conditions that may preclude some types of exercise) and physical examination in sedentary adults (age >50 years) with diabetes prior to beginning an exercise program. All cardiovascular disease (CVD) risk factors (dyslipidemia, hypertension, smoking) should be evaluated and treated (whether patients are exercising or not). (See "Overview of hypertension in adults" and "Low-density lipoprotein cholesterol-lowering therapy in the primary prevention of cardiovascular disease", section on 'CVD Risk Assessment'.)
For most individuals planning to participate in a low- to moderate-intensity physical activity (eg, brisk walking), no additional pre-exercise medical evaluation is needed unless symptoms of CVD or microvascular complications are present [3]. This approach is largely consistent with guidelines from the American Diabetes Association and is based on clinical experience and limited observational evidence [3,75].
●Asymptomatic, low risk – We do not typically perform exercise stress testing in asymptomatic individuals at low risk of coronary artery disease [3,74]. We counsel initiation of a gentle exercise program with gradual progression to a more vigorous program as tolerated and evaluate if typical or atypical signs or symptoms of atherosclerotic CVD (ASCVD) develop with exercise or are evident on examination. Routine exercise stress testing does not improve outcomes beyond medical management of cardiac risk factors [3,74-78]. As an example, in a cohort study evaluating 1705 asymptomatic adults with type 2 diabetes who attended an exercise intake program, sedentary individuals with one or more cardiac risk factors (40 percent of the cohort) were referred for pre-exercise stress testing [75]. There was no difference in the composite cardiovascular outcome (revascularization, cardiovascular-related hospital admission, or cardiovascular-related death) within one year (2.8 versus 1.9 percent) or after one year (3.1 versus 4.6 percent) in those who did or did not undergo stress testing. (See "Screening for coronary heart disease in patients with diabetes mellitus", section on 'Does screening for subclinical CHD improve outcomes?'.)
●Symptomatic, higher risk – We individualize the decision to perform stress testing prior to beginning an exercise program in patients at higher cardiovascular risk (eg, evidence of peripheral or carotid atherosclerotic vascular disease, renal disease, abnormal resting electrocardiogram [ECG], multiple diabetes complications), or with concerning symptoms (eg, a recent decrease in exercise tolerance, unexplained dyspnea), or who plan to initiate a more rigorous exercise program.
It is well established that sudden exercise in sedentary subjects can precipitate myocardial infarction [79-81]. The increased risk for asymptomatic coronary artery disease in those with diabetes and other risk factors suggests that the decision to perform cardiac evaluation should be individualized, with consideration given to those at very high risk. In the Look AHEAD trial, only older age was associated with an increased prevalence of all abnormalities during maximal exercise stress testing [82]. (See "Overview of general medical care in nonpregnant adults with diabetes mellitus", section on 'Screening for coronary heart disease'.)
The American College of Sports Medicine (ACSM) 2022 Consensus Guidelines recommend prior medical clearance from a health care professional for anyone with diabetes who is currently sedentary and desires to begin a new program of vigorous exercise (aerobic, resistance, or high-intensity interval exercise) [83,84].
The ACSM also recommends exercise stress testing more broadly in individual with diabetes who meet selected criteria (eg, age >40 years, age >30 years with CVD risk factors) [83]. This recommendation may be too conservative and may pose as a barrier to exercise for some patients with diabetes.
Type and frequency of exercise — Regular physical activity is recommended throughout life. There is not one exercise prescription for all individuals, and some individuals may even be considered as "nonresponders" to exercise training, at least with respect to aerobic fitness gains, insulin sensitivity improvements, and/or weight loss [85]. Nonetheless, most individuals with diabetes likely derive some benefit of regular physical activity [86]. Exercise programs should be individualized according to baseline fitness level and underlying comorbidities. Most sedentary individuals should begin a gentle exercise program with gradual progression as tolerated. (See "Exercise for adults: Terminology, patient assessment, and medical clearance", section on 'Terminology and common types of exercise'.)
●We encourage most adults with diabetes to perform at least 150 minutes of moderate-intensity aerobic exercise per week (30 to 60 minutes [at 40 to 60 percent maximal oxygen uptake (VO2max)] on most days of the week).
A reasonable initial regimen is 10 minutes of stretching and warm-up, followed by 20 minutes of gentle aerobic exercise such as walking, cycling, or rowing. The exercise should be performed regularly (at least three to five times per week) and preferably at the same time in relation to meals and insulin injections in patients treated with insulin. The duration and intensity of exercise should be increased gradually, as tolerated by the patient, to achieve a moderate intensity (eg, brisk walking).
●A shorter duration of more vigorous aerobic exercise (75 minutes per week of jogging 9.6 km/hour) is an alternative for physically fit individuals. This type of exercise regimen should only be used in adults who have been regularly exercising. Patients who are initiating an exercise program should begin with a gentle exercise program as described above with gradual progression as tolerated.
Another example of a low-volume, high-intensity training (HIT) program is cycling at 90 percent of individual maximal heart rate for 60 seconds, followed by 60 seconds of rest, repeated 10 times (total training session 20 minutes with 10 minutes of cycling and 10 minutes resting) [6]. The long-term health effects of low-volume, high-intensity training are unknown [3], however, this type of training has short-term improvements on glycemia in type 2 diabetes [40,57,58] and may be more effective than moderate-intensity continuous training in reducing abdominal fat in postmenopausal women with type 2 diabetes [87]. (See 'Type 2 diabetes' above.)
●In the absence of contraindications (ie, moderate to severe proliferative retinopathy), we also encourage adults to participate in resistance training exercises (free weights or weight machines) two to three times per week. The exercises should include the large muscles of the core, upper, and lower body (approximately 10 repetitions per set).
●For older adults, flexibility and balance training (eg, yoga, tai chi) may be helpful for fall prevention. (See "Falls: Prevention in community-dwelling older persons", section on 'Exercise'.)
●Younger patients with type 1 diabetes generally can tolerate even more vigorous exercise, including participation in competitive triathlons and marathons [88,89]. The exercise should optimally be performed at the same time of day in relation to meals and insulin injections (or boluses for insulin pumps). When that is done, the change in blood glucose concentrations is usually somewhat predictable and reproducible, particularly if the exercise is done with no prandial insulin "onboard" [90]. The effect of exercise at other times on blood glucose is highly variable [91], but exercise should not be discouraged as long as the individual is motivated and educated on glucose management during exercise.
Traditional insulin pumps allow one to reduce the basal rate prior to the activity, typically best done 90 minutes before the onset of aerobic exercise [92,93]; newer insulin delivery devices often have an exercise mode that increase the glucose target and reduce basal rates shortly before or during exercise, but these systems still have some challenges with exercise control [94]. (See 'Glycemic management during exercise' below.)
In people without diabetes, it has generally been shown that many of the health-related outcomes associated with regular exercise (blood pressure, weight control, lipid profiles, etc) are similar when comparing strategies of frequent, short-duration exercise sessions versus fewer, longer-duration exercise sessions [95,96]. In type 1 diabetes specifically, one study found that individuals achieve similar continuous glucose monitoring (CGM)-based outcomes (eg, percent time in range, time below range) whether they choose to perform the recommended 150 minutes of weekly physical activity during five short sessions (lasting approximately 30 minutes each) or two long sessions (lasting approximately 85 minutes each) [97]. Since the glycemic benefits of a 30-minute exercise session tend to last up to approximately 24 hours in type 1 diabetes [70], spacing out exercise training days with shorter exercise sessions (30 to 45 minutes) might be slightly better for overall glucose control, as compared with training less frequently for longer durations. In people with type 2 diabetes or obesity, small bursts of activity throughout the day to help break up periods of prolonged sitting are associated with enhanced insulin sensitivity and better glycemia [98].
Evidence is emerging, albeit limited in scope, that the time of day of exercise might impact the glycemic effects of exercise in diabetes [99-102]. As an example, in one small study of males with type 2 diabetes, performing high-intensity exercise in the afternoon rather than the morning was associated with greater improvement in blood glucose [99]. In a small study in adults with type 1 diabetes, performing moderate-intensity exercise in the morning conferred a lower rate of hypoglycemia than in the afternoon [100].
Contraindications and precautions — The ability to exercise may be limited by the presence of macro- or microvascular complications, as well as by prevailing glycemia [3].
●Exercise should be avoided or stopped immediately in the setting of symptoms of acute ASCVD (eg, chest pain, dyspnea, stroke symptoms).
●Vigorous exercise (eg, aerobic, resistance) should be avoided in patients with [3]:
•Severe retinopathy – Severe nonproliferative and unstable (recent or active retinal bleeding) proliferative retinopathy, recent panretinal photocoagulation, or other recent surgical eye treatment. There are occasional reports of patients with proliferative retinopathy developing retinal bleeding during vigorous or high-impact exercise (diving, boxing, or involving Valsalva maneuvers), which have led many clinicians to advise caution with regard to high-intensity activities that involve breath holding (eg, weightlifting, isometrics, or overhead lifting). Activities that lower the head (eg, yoga, gymnastics) or that jar the head are also not recommended. Intense isometric exercise (such as weightlifting) can cause a marked increase in blood pressure that might precipitate intraocular bleeding [83].
•Hyperglycemia – Substantial hyperglycemia (≥270 mg/dL [15 mmol/L]) with (ie, for people with type 1 diabetes) or without ketones. It is not necessary to defer exercise based on milder hyperglycemia, as long as the patient feels well and there is no ketonemia or ketonuria.
If the pre-exercise glucose is moderately elevated, individuals with type 1 diabetes [73] or type 2 diabetes [83] are advised to only begin light activity if they are asymptomatic and properly hydrated. For patients with type 1 diabetes or insulin-requiring type 2 diabetes, pre-exercise hyperglycemia due to missed doses of insulin (or interruption of insulin infusion from a pump) should be corrected before any activity begins.
•Hypoglycemia – Severe hypoglycemia within the previous 24 hours. Severe hypoglycemia is defined as blood glucose <54 mg/dL (3 mmol/L) or a hypoglycemic event with cognitive impairment requiring external assistance. Antecedent severe hypoglycemia impairs the hormonal counterregulatory response during exercise, thus increasing the risk for recurrent hypoglycemia. (See "Physiologic response to hypoglycemia in healthy individuals and patients with diabetes mellitus", section on 'Exercise'.)
●Exercise precautions in specific settings include the following:
•Hypertension – Individuals with poorly controlled hypertension should avoid heavy weightlifting or breath holding during exercise. These individuals should perform aerobic exercises that use large muscle groups, such as walking and cycling at a low to moderate intensity (ie, rating of perceived exertion [RPE] 10 to 12 on the 6 to 20 Borg Scale, or 3 to 4 on the Modified 0 to 10 Borg Scale).
•Peripheral neuropathy – Patients with distal polyneuropathy should avoid traumatic repetitive impact (long-distance running or prolonged downhill skiing), which may precipitate stress fractures in the small bones of the foot and ankle and the development of pressure ulcers on the toes and feet. Non-weightbearing exercises (eg, cycling, chair exercises, swimming) may be more appropriate. Well-fitting protective footwear and comfortable shoes are needed. Water activities should be avoided for those with unhealed plantar surface ulcers. Patients should continue to examine their feet regularly. (See "Evaluation of the diabetic foot", section on 'Preventive foot care'.)
•Autonomic neuropathy – Individuals with autonomic neuropathy may have an increased likelihood of developing exercise-associated hypoglycemia, abnormal blood pressure responses, and impaired thermoregulation, as well as elevated resting and blunted maximal heart rate responses to exercise. Exercise intensity should be monitored using RPE rather than heart rate [83].
•Diabetic kidney disease – People with chronic kidney disease may engage in aerobic and resistance exercise, as long as the exercise is started at a low intensity and volume and slowly increased as tolerated [103,104]. Although exercise can transiently increase urinary protein excretion, there is no evidence that exercise increases progression of chronic kidney disease [3]. People with kidney disease and severe retinopathy should refrain from high-intensity activities that can increase blood pressure or that involve breath holding (eg, weightlifting, isometrics, or overhead lifting) to avoid precipitating intraocular bleeding [83]. (See "Uremic myopathy and deconditioning in patients with chronic kidney disease (including those on dialysis)".)
•Heat conditions – Individuals with diabetes should be cautious when exercising in hot environments as they may be more prone to heat stress because of an impaired ability to thermoregulate [105]. This may promote excessive dehydration and hyperglycemia. Nonetheless, some heat acclimation has been shown to be possible in adults with type 2 diabetes engaging in aerobic [106] or resistance training [107].
Glycemic management during exercise — Fluid intake should be maintained at a relatively high level before, during, and after exercise.
●Type 1 diabetes or insulin-requiring type 2 diabetes – Insulin and carbohydrate regimens are adjusted to prevent hypoglycemia. Inadequate replacement of carbohydrate before, during, and after exercise is the most common cause of exercise-associated hypoglycemia in patients taking insulin [108]. (See "Hypoglycemia in adults with diabetes mellitus", section on 'Exercise-induced hypoglycemia'.)
•Patients who take insulin (particularly those with type 1 diabetes), should be advised to measure the blood glucose before, during, and after exercise so that the changes in blood glucose can be documented and then predicted for subsequent exercise sessions [90]. CGM can and should be maintained during exercise and is a useful adjunct in reducing risk for hypoglycemia including during exercise. (See "Glucose monitoring in the ambulatory management of nonpregnant adults with diabetes mellitus", section on 'Benefits of CGM' and "Cases illustrating the effects of exercise in intensive insulin therapy for type 1 diabetes mellitus", section on 'Case 2'.)
Insulin dose adjustments before, during, and after exercise are often empiric and aided by the results of glucose monitoring (with fingersticks every 30 to 45 minutes or using CGM) [109].
•If the pre-exercise glucose is <100 mg/dL (5.6 mmol/L), patients should ingest extra food, in the form of 15 to 30 grams of quickly absorbed carbohydrate (such as glucose tablets, hard candies, or juice), which should be taken 15 to 30 minutes before exercise and approximately every 30 minutes during exercise, based on repeat glucose testing during the exercise.
•If CGM is used, carbohydrate feeding should be initiated during exercise if the glucose drops below approximately 125 mg/dL, particularly in individuals who are at greater risk for exercise-induced hypoglycemia [109]. The use of directional trend arrows in CGM systems can also be used to better guide carbohydrate feeding protocols [109].
•Morning high-intensity exercise may also cause hyperglycemia in type 1 diabetes (eg, when serum insulin concentrations are inadequate), and a bolus insulin correction (50 to 100 percent of the usual correction dose) may be required to restore glycemia in recovery [24].
•In the setting of adequate or high serum insulin concentrations, late (or nocturnal) hypoglycemia may occur four to eight hours after the termination of exercise due to replenishment of depleted glycogen stores. To reduce the risk, the daily basal insulin dose may be reduced by approximately 20 percent [73,110]. In addition, slowly absorbed carbohydrates (dried fruit, fruit jerky, granola bars, or trail mix) can be ingested immediately after exercise. Some individuals increase the bedtime snack (long-acting carbohydrate with protein) and omit bolus insulin [73].
For insulin pump users who exercise in the afternoon, the basal rate can be reduced by 20 percent for six hours (starting at bedtime) to reduce nocturnal hypoglycemia [111]. As an alternative, the patient can use a reduced temporary basal rate ("exercise mode") and higher glucose target starting up to 90 to 120 minutes before exercise [112]. If the temporary basal rate function of the insulin pump is used, this reduced rate should be continued until after exercise, the exact timing depending on CGM trend arrows [109]. (See "Cases illustrating the effects of exercise in intensive insulin therapy for type 1 diabetes mellitus", section on 'Case 3'.)
•Some individuals may decrease the insulin dose that affects time of the day when exercise will be performed (by 25 to 75 percent). This is especially important if exercise is of long duration (more than 60 minutes) [108]. As an example, if an individual is exercising within three hours after a meal, the prandial insulin for that meal can be reduced by 25 to 50 percent to avoid hypoglycemia during subsequent exercise. Some individuals may prefer to exercise before eating, with or without a basal insulin reduction, and then perform the usual or a slightly reduced (approximately 25 percent) premeal bolus insulin, based on the prevailing post-exercise glucose level and their residual increase in insulin sensitivity.
•To prevent increased insulin absorption, some individuals inject the insulin in a site other than the muscles to be exercised [30]. As an example, the arm is a suitable site for cycling. On the other hand, the abdomen is preferred with tennis or racquetball where the exercise involves both the arms and legs.
●Type 2 diabetes treated with insulin secretagogues – For patients with type 2 diabetes, who take sulfonylureas or meglitinides, the blood glucose can be measured before, during, and after exercise so that the changes in blood glucose can be documented and then predicted for subsequent exercise sessions. If hypoglycemia occurs, the exercise should be stopped, rapid-acting carbohydrates should be consumed, and the individual should be instructed to decrease the dose of the insulin secretagogue on exercise days [3]. The correction of exercise-induced hypoglycemia with a rapid-acting carbohydrate (eg, glucose tablets, hard candies, juice) is particularly important in patients treated concomitantly with a glucagon-like peptide 1 (GLP-1) receptor agonist, which may delay gastric emptying.
●Type 2 diabetes, not taking insulin or insulin secretagogues – Hypoglycemia is uncommon in patients with type 2 diabetes not treated with insulin or insulin secretagogues (sulfonylureas, glinides), and therefore, monitoring of glucose during exercise, ingestion of extra carbohydrates, and medication adjustments are not typically necessary. On the other hand, CGM may be a useful tool (for educational and biofeedback reasons) to help demonstrate the expected improvements in glycemia associated with being more physical activity.
Addressing barriers to exercise — Several perceived barriers may contribute to the inability to sustain an exercise regimen in patients with diabetes. As expected, patients with low perceived barrier to exercise do more physical activity and have better overall glycemic control [113]. Unfortunately, perceptions of barriers held by physicians increase the perception of barriers in their patients and can reduce activity habits [113]. In general, most patients with type 2 diabetes are trying to change a lifetime of sedentary behavior, and continued compliance is a major problem [114]. In addition to decreased exercise capacity, potential movement discomforts (eg, joint pain, foot pain), occult coronary or peripheral vascular disease, and/or diabetic neuropathy can limit exercise tolerance.
Despite these difficulties, maintenance of an exercise program in patients with type 2 diabetes remains a worthwhile goal because compliance is associated with long-term improvement in cardiovascular risk factors. (See 'Type 2 diabetes' above.)
To increase compliance, the clinician should attempt to help break down perceived barriers to exercise and help patients choose a type of exercise they will enjoy and maintain. In some studies, simple behavioral counseling by clinicians during routine clinic visits to promote sustained commitment to an exercise program has shown encouraging results [115,116]. The use of step counters and technology-based support (eg, internet-based programs) have also been shown to increase physical activity [117,118]. (See "Exercise prescription and guidance for adults", section on 'Improving compliance with a basic aerobic exercise program'.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diabetes mellitus in adults".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Type 1 diabetes (The Basics)" and "Patient education: Type 2 diabetes (The Basics)")
●Beyond the Basics topics (see "Patient education: Type 1 diabetes: Overview (Beyond the Basics)" and "Patient education: Type 2 diabetes: Overview (Beyond the Basics)" and "Patient education: Preventing complications from diabetes (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Exercise and muscle metabolism – A single session of exercise increases non-insulin-mediated and insulin-mediated glucose uptake. Insulin sensitivity can be enhanced for up to 48 hours after exercise. For most individuals living with diabetes, regular exercise training increases glucose disposal (mitochondrial oxidation and storage as muscle glycogen) and improves muscle mass and morphology, whole-body insulin sensitivity, and cardiovascular health. (See 'Exercise and muscle metabolism' above.)
•Type 2 diabetes – In the resting state, individuals with type 2 diabetes and insulin resistance have several defects in glucose metabolism (eg, decreased number and function of skeletal muscle insulin receptors and glucose transporters), reducing the ability of muscle cells to take up glucose. Exercise overcomes the defects in glucose transport caused by insulin resistance by increasing both non-insulin-mediated and insulin-mediated glucose transport during and after the exercise session. In the long-term, additional training-induced adaptations in skeletal muscle, adipose tissue, and the liver further enhance insulin sensitivity.
•Type 1 diabetes – Patients with type 1 diabetes typically have normal muscle glucose kinetics; however, the inability to shut off previously administered exogenous insulin may result in inhibition of hepatic glucose output (with continued muscle glucose uptake) and a subsequent fall in blood glucose concentration during exercise that is much larger than that in individuals without diabetes.
●Exercise benefit in diabetes – Exercise improves glycemic indices (particularly in type 2 diabetes), assists with weight maintenance, and improves other cardiovascular risk factors (eg, dyslipidemia, hypertension). (See 'Exercise benefit in diabetes' above.)
●Pre-exercise evaluation – We typically perform a history (with assessment of cardiovascular risk factors and for conditions that may preclude some types of exercise) and physical examination in sedentary adults (age >50 years) with diabetes prior to beginning an exercise program. All cardiovascular disease (CVD) risk factors (dyslipidemia, hypertension, smoking) should be evaluated and treated (whether patients are exercising or not). (See 'Evaluation prior to recommending an exercise regimen' above.)
We do not typically perform exercise stress testing in asymptomatic patients as long as they are beginning a gentle exercise program with gradual progression as tolerated. We evaluate if typical or atypical signs or symptoms of atherosclerotic CVD (ASCVD) develop with exercise or are evident on examination. However, the increased risk for asymptomatic coronary artery disease in those with diabetes and other risk factors suggests that an exercise tolerance test should be individualized. We perform stress testing prior to beginning a new exercise program in patients at higher cardiovascular risk (eg, evidence of peripheral or carotid atherosclerotic vascular disease, renal disease, abnormal resting electrocardiogram [ECG], multiple diabetes complications), or with concerning symptoms (eg, a recent decrease in exercise tolerance, unexplained dyspnea), or who plan to initiate a more rigorous exercise program. (See 'Evaluation prior to recommending an exercise regimen' above.)
●Type and frequency of exercise – Regular physical activity is recommended throughout life. For adults with diabetes, we encourage at least 150 minutes of moderate-intensity (eg, brisk walking) aerobic exercise per week. In the absence of contraindications (eg, moderate to severe proliferative retinopathy), we also encourage resistance training (exercise with free weights or weight machines) at least twice per week. For adults with diabetes who are generally fit and have higher aerobic capacity, a shorter duration of more vigorous aerobic exercise (eg, 75 minutes per week of jogging 9.6 km/hour), or brief periods of high-intensity interval training, may be preferable. (See 'Type and frequency of exercise' above.)
●Contraindications to exercise – Exercise should be avoided in the presence of symptoms of acute ASCVD. In addition, vigorous exercise should be avoided in patients with any of the following:
•Severe nonproliferative and unstable (recent or active retinal bleeding) proliferative retinopathy
•Severe hypoglycemia (glucose <54 mg/dL [3 mmol/L] or a hypoglycemic event with cognitive impairment requiring external assistance) within the previous 24 hours
•Substantial hyperglycemia (≥270 mg/dL [15 mmol/L])
It is not necessary to defer exercise based on moderate hyperglycemia (eg, <270 mg/dL [15 mmol/L]), as long as the patient feels well and there is no ketonemia or ketonuria. (See 'Contraindications and precautions' above.)
●Glycemic management during exercise
•Patients taking insulin or insulin secretagogues – For patients who take insulin (particularly those with type 1 diabetes), adjustments of their insulin regimen before, during, and after exercise are often empiric and aided by the results of glucose monitoring (with fingersticks every 30 to 45 minutes, or using a continuous glucose monitor [CGM]). For patients who take insulin or insulin secretagogues (sulfonylureas, glinides), blood glucose should be measured before, during, and after exercise so that the changes in blood glucose can be documented and then predicted for subsequent exercise sessions. (See 'Glycemic management during exercise' above.)
If the pre-exercise blood glucose is <100 mg/dL (5.6 mmol/L), insulin- or insulin secretagogue-treated patients should ingest extra food, in the form of 15 to 30 grams of quickly absorbed carbohydrate (such as glucose tablets, hard candies, or juice), which should be taken 15 to 30 minutes before exercise and approximately every 30 minutes during exercise, based on repeat blood glucose testing during the exercise. Such patients are also at risk of late hypoglycemia (ie, four to eight hours after the termination of exercise) due to replenishment of depleted glycogen stores. (See 'Glycemic management during exercise' above.)
•Type 2 diabetes not taking insulin or insulin secretagogues – Hypoglycemia is uncommon in patients with type 2 diabetes not treated with insulin or insulin secretagogues (sulfonylureas, glinides), and therefore, monitoring of glucose during exercise, ingestion of extra carbohydrates, and medication adjustments are not typically necessary. (See 'Glycemic management during exercise' above.)
