PREVENTIVE CARDIOLOGY
SPRING 2000
The Role of Exercise in Type II Diabetes Mellitus Stephen H. Schneider, MD; Elhadi B. Elouzi, MD
A number of studies have demonstrated a beneficial effect of regular physical activity on levels of HgbA1C in patients with type II diabetes mellitus, largely due to an increase in insulin sensitivity. Benefits are related to short term improvements in insulin sensitivity following individual exercise bouts. Regular exercise can prevent or delay the onset of type II diabetes in high risk populations. The insulin resistant state is associated with a cluster of cardiovascular risk factors all of which improve with regular physical activity. Because of the high incidence of occult coronary disease, patients need a cardiovascular evaluation when initiating an exercise program. High intensity exercise may result in retinal hemorrhage and transient worsening of diabetic proteinuria. The most common complication is hypoglycemia. A combination of aerobic and light resistance exercise is appropriate. Patients should exercise a minimum of three times a week for 30–60 minutes at 50% to 75% of their Vo2max. (Prev Cardiol. 2000;3:77–82) © 2000 by CHF, Inc. he relationship between a sedentary lifestyle T and type II diabetes has been recognized for centuries. Over the last three decades research has clarified the benefits of physical activity in the treatment and prevention of type II diabetes mellitus (Table I). Nevertheless, exercise remains a greatly underutilized therapy with a substantial untapped potential. In this paper we will summarize our current understanding of the role of exercise in the treatment of type II diabetes mellitus. From the Division of Endocrinology, Metabolism and Nutrition, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ Address for correspondence/reprint requests: Stephen H. Schneider, MD, UMDNJ-Robert Wood Johnson Medical School, 51 French Street–MEB 386, New Brunswick, NJ 08903 Manuscript received February 8, 2000; accepted February 28, 2000
EFFECTS OF EXERCISE ON GLUCOSE METABOLISM Type II diabetes mellitus constitutes a heterogeneous disorder associated with abnormalities in both insulin sensitivity and insulin secretion. 1,2 The recognition that a sedentary lifestyle is associated with diabetes is centuries old. In the past 25 years a number of studies have demonstrated the ability of regular exercise to improve glucose control in selected patients. Most studies have utilized moderate levels of activity (50%–75% of Vo2max [maximal oxygen uptake]) for 30–60 minutes 3–5 days a week, which is associated with a decrease of 0.5%–1.5% in HgbA1C.3,4 The effects of regular exercise on insulin sensitivity and glucose disposal are complex. Yet most of the benefits of exercise appear to be mediated by an improvement in insulin sensitivity in skeletal muscle rather than enhanced insulin secretion. 5 These effects appear to be largely the result of the summed influences of individual exercise bouts.3 Following a single bout of moderately intense physical activity, insulin sensitivity and carbohydrate disposal may be enhanced for as long as 24–48 hours.3 As a result, exercise performed ≥ 3 times per week results in a sustained improvement in insulin sensitivity. Evidence that regular physical training alone results in major changes in body composition is limited. When exercise is associated with important changes in body composition, a more long lived metabolic improvement occurs.6 For most individuals, the trained state per se in the absence of major changes in body composition may offer surprisingly little additional benefit to the summed effects of the individual exercise bouts. Studies in which patients have trained for up to one year have demonstrated deterioration of insulin sensitivity back to baseline within one week of discontinuing physical training.7 The mechanism by which an acute bout of exercise improves subsequent insulin sensitivity and carbohydrate tolerance is unclear. During and immediately after exercise an increase in muscle blood flow, and therefore glucose delivery, is one simple mechanism for enhanced glucose uptake.
77
78
PREVENTIVE CARDIOLOGY
TABLE I. POTENTIAL BENEFITS OF REGULAR PHYSICAL ACTIVITY FOR PATIENTS WITH TYPE II DIABETES MELLITUS • Improved insulin sensitivity and glucose disposal • Improved diabetes control and HgbA1C • Decreased concentrations of triglyceride rich lipoproteins • Decreased blood pressure • Improved fibrinolytic activity • Favorable changes in body composition • Improved mood and self image • Prevention of type II diabetes
Nevertheless, insulin sensitivity remains improved long after blood flow has returned to normal. Some studies suggest that changes in muscle glycogen with exercise predict the subsequent improvement. Exercise below the level that results in glycogen depletion does not appear to result in improved insulin sensitivity even when prolonged.8–10 Conversely, glycogen loading, which raises muscle glycogen levels above normal, is associated with insulin insensitivity.11 It is not clear if the glycogen concentration is a direct regulator of insulin sensitivity or simply a marker for other metabolic changes responsible for this effect. Recent research has focused on the role of activated protein kinase (AMP) and changes in intracellular free fatty acid metabolism as possible mediators of exercise effects.12
PREVENTION OF TYPE II DIABETES WITH PHYSICAL ACTIVITY The observation that regular physical activity enhances insulin sensitivity raises the possibility that exercise could be used in the prevention of type II diabetes. In classical type II diabetes, insulin insensitivity appears to be one of the earliest defects.13 Initially, in response to insulin resistance, the pancreas releases additional insulin to restore glucose metabolism to normal. This results in a normoglycemic but hyperinsulinemic state. The development of clinical type II diabetes requires a second defect: progressive failure of beta cell function. The etiology of this defect is unclear, but probably less than half of patients with inherited insulin insensitivity progress to type II diabetes. It has been hypothesized that correction of insulin resistance by regular physical activity could interrupt the cycle that results in eventual beta cell failure, and possibly prevent or delay the onset of clinical disease. A number of epidemiologic and cross sectional studies support this hypothesis. Recently, two large prospective studies have demonstrated a protective effect of moderate physical activity on the progression of impaired glucose tolerance to clinical diabetes over a six year period.14,15 It is important to note that the intensity of exercise in both studies was quite modest. These observations sug-
SPRING 2000
gest that improving physical activity levels in people at high risk for type II diabetes could have major public health implications.
PHYSICAL ACTIVITY AND ARTERIOSCLEROSIS IN TYPE II DIABETES The most important complication of type II diabetes mellitus is the development of premature large vessel disease which accounts for >80% of all premature morbidity and mortality associated with diabetes in the U.S. In addition, recent decreases in the incidence of coronary artery disease in the U.S. have not been shared by the population with diabetes, particularly in women. A large body of epidemiologic and cross sectional data suggests that a sedentary lifestyle is associated with an increased incidence of coronary artery disease.16,17 It is important to note that high levels of physical activity are not required for protection.18 While we have no prospective randomized study showing less coronary disease in patients with diabetes who are more physically active, there are theoretical reasons for considering exercise to be particularly effective for this group of individuals. The mechanisms by which regular physical activity might protect against the development of large vessel disease are under intensive investigation. The relationship between the degree of hyperglycemia and the development of large vessel disease is much weaker than that for the small vessel and neurologic complications of diabetes, suggesting the importance of other risk factors. In patients with mild glucose intolerance or type II diabetes, one of the strongest risk factors for the development of subsequent coronary artery disease is the insulin resistant hyperinsulinemic state, independent of plasma glucose levels.1,19 Controversy persists as to whether hyperinsulinemia per se is an important risk factor for atherosclerosis or whether it is a marker of other underlying pathologies.1,20 Nevertheless, the insulin resistant state is clearly associated with a cluster of cardiovascular risk factors independent of hyperglycemia that could contribute to an increased incidence of coronary artery disease. In addition to various degrees of impaired glucose tolerance, the insulin resistant state is associated with hypertension, a hypercoagulable state with elevated levels of plasminogen activator inhibitor, and a characteristic dyslipidemia with elevated triglyceride rich particles, small dense LDL particles, and decreased levels of HDL cholesterol. It is important to appreciate that these risk factors potentiate each other so that the effect is more than additive when they occur in combination. Additional components of the syndrome include hyperuricemia and, in susceptible women, the polycystic ovarian syndrome. Reversal of the insulin resistant state with regular physical activity is associated with improvement in this entire cluster of cardiovascular risk factors.21 Thus, despite the lack of controlled, randomized, prospective tri-
SPRING 2000
TABLE II. COMPLICATIONS OF EXERCISE IN PATIENTS WITH TYPE II DIABETES MELLITUS General Acute coronary events Foot injury Joint and musculoskeletal injuries Post exercise orthostatic hypotension Microvascular Retinal hemorrhage Proteinuria Metabolic Paradoxical hyperglycemia Acute and delayed exercise related hypoglycemia
als, it is reasonable to encourage physical activity for patients with diabetes in the hope of preventing premature coronary disease.
COMPLICATIONS OF EXERCISE PROGRAMS FOR PATIENTS WITH DIABETES For the great majority of patients with type II diabetes, it is possible to design safe and effective exercise programs.22 Unfortunately, by the time these individuals present to their physicians, they are often older sedentary individuals who may have multiple complications of their disease. Table II lists some of the complications which can occur in patients with diabetes in exercise programs. One of the most common is injury to the diabetic foot. While vascular disease contributes to this risk, it is primarily related to diabetic neuropathy. Decreased pain and pressure sensation leads to the use of poorly fitting footwear and lack of recognition of foot injuries. Denervation and atrophy of interosseous muscles result in change in the shape of the foot and a maldistribution of weight bearing. Another important and often overlooked factor is the loss of proprioception. This results not only in an increased likelihood of injury to the foot, but may also contribute to a possible increase in degenerative joint disease.23 A careful evaluation of blood flow and neurologic status of the lower extremities should be part of any preexercise assessment. Careful attention needs to be paid to the selection of proper footwear. Patients with severe disease may need to limit activities, such as jogging, in favor of lower impact forms of exercise. Finally, it should be noted that autonomic dysfunction may place some patients at risk for developing postexercise orthostatic hypotension, particularly during the early phase of an exercise program when autonomic dysfunction is not apparent at rest. Microvascular complications of diabetes could be aggravated in some circumstances. Evidence suggests that moderately intense exercise does
PREVENTIVE CARDIOLOGY
not induce retinopathy and may even have a protective effect.24 However, in the presence of established severe background or proliferative disease, certain types of exercise may predispose the patient to retinal hemorrhage. 25,26 Direct trauma to the eyes can be avoided with protective eyewear. Rapid head movements that can result in dangerous forces should be avoided as another potential cause of hemorrhage. Finally, high intensity exercises, particularly those involving Valsalva maneuvers, should be avoided. In addition to eye disease, there is concern about the potential effects of exercise on the progression of diabetic nephropathy. Many patients with no proteinuria at rest will develop protein in their urine for as long as 12–24 hours following a single intensive bout of exercise. 27 The degree of proteinuria is roughly proportional to the rise in systolic blood pressure that occurs during the activity. In type I diabetes, there is some evidence that this may predict those individuals more likely to develop overt diabetic nephropathy.28 Using current criteria for evaluating proteinuria, it is important to instruct patients to remain sedentary for 24 hours prior to the collection. Whether or not the hemodynamic changes resulting in proteinuria might contribute to the progression of diabetic nephropathy is unknown.29 Metabolic complications of exercise are most likely to occur in patients with type I diabetes mellitus. Exercise related hypoglycemia is rare in patients with type II diabetes, but may occur in those taking insulin or some of the more potent sulfonylureas. Exercise of the muscles directly underlying an insulin injection can result in accelerated absorption of insulin which in turn can contribute to hypoglycemia.30 This occurs primarily with short acting insulins and only when the exercise follows the insulin injection within a short period of time. It is easily avoided by selecting injection sites away from the area of exercising muscle. In normal circumstances, exercise results in an acute suppression of pancreatic insulin release and a decrease of serum insulin levels. In the presence of an exogenous insulin depot the normal suppression in insulin levels does not occur, resulting in a relative hyperinsulinemia. In addition, some patients with long standing diabetes may develop deficiencies in glucagon and epinephrine release that could also limit their ability to maintain a normal glucose level during activity. Nonselective ß-blockers rarely contribute to hypoglycemia in patients with type II diabetes who generally have a normal ability to secrete glucagon, but have been implicated in predisposing of hypoglycemia in patients with long standing type I diabetes who are glucagon deficient.10 Most episodes of significant exercise related hypoglycemia occur during or immediately after exercise. In addition to this acute glucose lowering response, patients should also be aware
79
80
PREVENTIVE CARDIOLOGY
TABLE III. EXERCISE RECOMMENDATIONS Exercise testing for sedentary patients age >35 years or 10 year duration of type II diabetes mellitus Warm up Stretching (avoid Valsalva) Aerobic exercise for 20–60 minutes 50%–70% VO2max at least four times a week High volume resistance exercises Cool down
that following a brisk bout of exercise delayed bouts of hypoglycemia can occur anywhere from 6–10 hours following activity and is probably related to changes in insulin sensitivity. Exercise performed at night can result in hypoglycemic episodes during early morning hours when the patient is still asleep.31 In patients with type I diabetes with severe insulin deficiency, exercise may actually result in a paradoxical rise in both the blood glucose and ketone levels. In this situation, lack of the restraining effect of insulin on hepatic glucose production and lipolysis leads to overproduction of glucose and ketoacids that can exceed their rate of utilization. In general, a fasting glucose of >300 mg per deciliter is a marker of sufficient insulin deficiency to be a risk.32 This situation is not well documented in patients with type II diabetes. Poorly controlled patients, particularly those exhibiting dehydration, may also demonstrate deterioration of glucose control. Therefore, exercise should be avoided in patients who have poor glycemic control.
RECOMMENDATIONS FOR EXERCISE IN PEOPLE WITH TYPE II DIABETES MELLITUS The American Diabetes Association has developed specific recommendations for exercise programs intended for patients with type II diabetes mellitus (Table III). 33,34 Unfortunately, physical activity levels in the U.S. are generally low. Also the majority of patients who present with type II diabetes mellitus have been sedentary for years. These patients require careful pre-exercise evaluations, including a cardiovascular examination, a complete retinal exam after dilatation, and a careful assessment of the vascular and neurologic status of the feet. For patients >35 years who have been sedentary, or patients of any age who have had diabetes for >10 years, it is recommended that they undergo some kind of formal exercise testing prior to starting more than a low intensity exercise program. Recommendations for exercise stress testing in asymptomatic diabetic patients remain controversial. In particular, the need for formal cardiac stress testing with or without imaging is currently under review.
SPRING 2000
Type of Activity A wide variety of exercise regimens are available and flexibility in selecting a regimen that fits a patient’s lifestyle and makes allowances for social, financial, and medical limitations is important. Most of the studies demonstrating the benefits of formal exercise programs have used aerobic exercise of moderate intensity (50%–75% VO2max) for 30–60 minutes performed 3–5 times per week. Recent smaller studies suggest that modest resistance exercise is safe and may also improve glycemic control and cardiovascular risk factors.35–38 In addition, the increase in lean body mass resulting from such a program may increase the basal metabolic rate and could be useful in weight maintenance. Recent recommendations suggest adding a component of so called high volume resistance exercise to aerobic exercise recommendations for patients with diabetes who have no contraindications. 39 Recent programs have stressed less formal types of exercise, especially early on and behavioral approaches using concepts such as “stages of change” show promise in some large studies.40 The Exercise Session Warm up. Patients should engage in a brief warm up prior to stretching. Stretching is important in this population because they may be predisposed to musculoskeletal injuries, and should be done smoothly without breath holding or bouncing motions. Intensity. Exercise of ≥ 50% maximal aerobic capacity is recommended when practical. While some epidemiologic studies suggest that less intensive exercise may protect against coronary disease, it is clear that the acute effect of exercise on insulin sensitivity requires exercise of at least moderate intensity. Modification of other risk factors may require different levels of exercise intensity. For example, some studies have shown an effect of low intensity exercise on reducing triglycerides, while exercise of much greater intensity and duration appears to be necessary to consistently increase levels of HDL cholesterol.41 The exercise prescription is generally monitored by self measurement of heart rate. Resistance exercise is prescribed as a series of sets of 8–12 repetitions with brief rest periods in between. The degree of resistance should be adjusted so that the patient can complete the exercise without exhaustion.39 It is important to remember that most middle age individuals with type II diabetes mellitus have a very poor aerobic exercise capacity even when compared to sedentary age matched controls. In general, the maximal aerobic exercise capacity of these patients is about 15% below that of their peers. 42 Interestingly, studies have shown a similar decrease in aerobic exercise ca-
PREVENTIVE CARDIOLOGY
SPRING 2000
pacity in the offspring of patients with type II diabetes mellitus, as well as abnormalities in muscle composition and capillary density.43,44 This raises the possibility that a modest impairment of aerobic capacity may precede the onset of type II diabetes and may contribute to the sedentary lifestyle of these individuals. Nevertheless, their aerobic capacity does increase with physical training. In patients with significant microvascular disease, an exercise intensity is established which, during exercise testing, does not raise the systolic blood pressure above 180–200 mm Hg. Duration. The duration of activity is generally 20–60 minutes. Short bouts of exercise may not result in significant increases in insulin sensitivity, although exercise of >60 minute duration seems to be associated with a higher incidence of musculoskeletal injuries. There is some data to suggest that two brief exercise sessions may be as effective as a single more prolonged episode, and this approach is often preferred by some patients. 40 Brief rest periods of
