(Circulation. 2006;113:2642-2650.)
© 2006 American Heart Association, Inc.
Contemporary Reviews in Cardiovascular Medicine |
From the Center for Exercise Science, College of Health and Human Performance and the College of Medicine, University of Florida, Gainesville (R.W.B.); and the Division of Cardiology, Johns Hopkins School of Medicine, Baltimore, Md (K.J.S.).
Correspondence to Randy Braith, PhD, PO Box 118206, University of Florida, Gainesville, FL 32611. E-mail rbraith{at}hhp.ufl.edu
Key Words: arteriosclerosis cardiovascular diseases exercise risk factors
| Introduction |
|---|
|
|
|---|
This review will critically evaluate whether RT modifies CV risk factors and improves characteristics of CV structure and function. The topics will be limited to the effects of RT on major and independent risk factors for CV disease including diabetes mellitus, hypertension, dyslipidemia, and advancing age.4 The quantitative relation between these risk factors and CV events has been elucidated by the Framingham Heart Study4 and other studies. The topics will also include 2 predisposing risk factorsobesity and physical inactivitythat are designated as major risk factors by the American Heart Association.1,2,4 To the extent possible, this review will examine the separate and independent effects of RT in studies that did not include a concomitant aerobic exercise component. However, in those instances where the data from RT studies are equivocal, studies that combined RT and aerobic exercise will be acknowledged to help the clinician formulate recommendations for their patients. Additionally, the review will focus mainly on primary prevention, for example, risk reduction in persons without established CV disease. Many low- to moderate-risk patients with established CV disease should be encouraged to incorporate RT into their physical conditioning program, especially those who rely on their upper extremities for work or recreational pursuits. However, the safety and effectiveness of RT in other populations of CV patients (eg, women, older patients with low aerobic fitness, patients with severe left ventricular dysfunction) have not been well studied. Accordingly, these patient subsets may require more careful evaluation and initial monitoring, and RT guidelines and recommendations must be modified accordingly. Moreover, there is only a limited body of literature assessing the independent benefits of RT on CV risk factors in patients with established CV disease. Studies conducted in cooperation with comprehensive cardiac rehabilitation programs typically include the confounding influences of aerobic activity, initiation of vasoactive and lipid-lowering drugs, and nutritional education with subsequent dietary modifications.
| Rationale for Resistance Training |
|---|
|
|
|---|
|
Long-term adaptation to RT lowers cortisol response to acute stress,20 increases total energy expenditure and physical activity in healthy18,21 and frail older adults,22 and relieves anxiety, depression, and insomnia in clinical depression.23 RT has beneficial effects on bone density,18,24 osteoarthritic symptoms,22,25 hypertension,26,27 lipid profiles,28 and exercise tolerance in coronary artery disease.29 Conversely, the loss of skeletal muscle mass and contractile function that accompanies aging, for example, sarcopenia, is linked to peripheral insulin resistance, dyslipidemia, and increased adiposity.30 Sarcopenia is also accelerated in patients with chronic heart failure, a condition characterized by peripheral skeletal muscle abnormalities and muscle wasting. Several studies have shown beneficial effects of RT on muscle mass and strength in patients with chronic heart failure.31,32 However, the relative mitigating effects of RT on primary and secondary CV risk factors remains controversial.
| Resistance Training and Diabetes |
|---|
|
|
|---|
Muscle contraction increases glucose uptake in skeletal muscle,34 thereby forming the basis for recommending RT for individuals with abnormal glucose metabolism. Aerobic exercise uses large muscle groups for extended periods of time, but whole-body RT programs may provide equally high or higher recruitment of muscle mass over a comparable period of time. The American College of Sports Medicine has recommended the use of progressive RT as part of a well-rounded exercise program for individuals with type 2 diabetes.35 Similarly, in the absence of contraindications, the American Diabetes Association also recommends RT for those with type 2 diabetes.11 These recommendations are supported by evidence that RT is an integral component in the therapeutic management of glycemic control in both young and older type 2 diabetics,28,3638 particularly if the RT is performed in a supervised versus home-based program.39
Glucose Tolerance and Insulin Sensitivity
A frequent postulate is that RT and subsequent increases in skeletal muscle mass may improve glucose and insulin responses to a glucose load.6,40 However, there are little data available showing that RT prevents type 2 diabetes. RT does not usually alter glucose tolerance or glycemic control regardless of age,57,16,41,42 unless baseline glucose tolerance is abnormal.36,38,4044 Nevertheless, RT reduces acute insulin responses during an oral glucose tolerance test in healthy young, middle-aged, and older men in most studies.5,6,45 RT also reduces acute insulin responses during glucose tolerance testing in diabetic men36,38 and women36,38,44 and improves insulin sensitivity during hyperglycemic and hyperinsulinemic-euglycemic clamps in diabetic and/or insulin-resistant middle-aged6,46,47 and older men42 and diabetic middle-aged48 and older women.42
Glycemic Control
RT decreases glycosylated hemoglobin (HbA1c) levels in diabetic men28,36,37,43 and women,28,36,37 regardless of age. This effect is observed even in the absence of a lasting effect of RT on fasting glucose or insulin levels.37 Improved glycemic control and decreased HbA1c levels are important for reducing the microvascular and macrovascular complications of diabetes. For example, the UK Prospective Diabetes Study49 reported that each percentage point reduction in HbA1c was associated with a 35% reduction in microvascular complications, whereas the European Prospective Investigation of Cancer and Nutrition (EPIC)Norfolk prospective population study showed that an increase of 1 percentage point in HbA1c was associated with a 28% increase in mortality risk, independent of other CV risk factors.50 RT-induced improvements in glycemic control, however, appear to be intensity-dependent, with beneficial effects occurring when subjects train at 70% to 90% of the 1-repetition maximum strength (1-RM). One-repetition maximum is the maximal weight lifted in 1 attempt during strength testing. In studies where the RT intervention was less than 2 months and/or the exercise intensity was less than 50% of the 1-RM, improvements in HbA1c were modest or undetected.47,51,52
| Resistance Training and Hypertension |
|---|
|
|
|---|
Systemic Blood Pressure
The rationale for RT as an adjunct to aerobic exercise for controlling BP stems from multiple studies. Two meta-analyses of RT and hypertension are noteworthy.26,27 Inclusion criteria, consistent across both reviews, were (1) inclusion of a randomized nonexercise control group; (2) RT as the only intervention; (3) training for a minimum of 4 weeks; and (4) participants who were sedentary normotensive and/or hypertensive adults with no other concomitant disease. Kelley and Kelley27 examined the effects of RT on resting BP in studies published between January 1966 and December 1998. A total of 11 studies met the inclusion criteria and represented initial and final BP assessments in 182 RT subjects and 138 control subjects. Decreases (P
0.05) of approximately 3 mm Hg were found for both systolic and diastolic BP across all BP categories as the result of RT. These changes represented a 2% decrease for resting systolic BP and 4% for resting diastolic BP. No differences were found for changes in resting BP between studies that used conventional RT compared with a circuit RT protocol. A conventional RT protocol generally consists of lifting heavier weights with longer rest periods, whereas a circuit RT protocol consists of lifting lighter weights with shorter rest periods between exercises. By moving quickly between exercises and by using lighter weight with higher repetitions, circuit training introduces an aerobic component to the workout.54
In the more recent meta-analysis, Cornelissen and Fagard26 pooled data from studies published between 1996 and 2003 that included 9 randomized controlled trials involving 341 participants. The overall effect of RT was a decrease of 3.2 mm Hg (P=0.10) in systolic BP and a decrease of 3.5 mm Hg (P
0.05) in diastolic BP. Results from these meta-analyses are consistent with conclusions generated by narrative reviews.17,55,56 Although these reductions seem modest, a systolic BP reduction of 3 mm Hg in average populations has been estimated to reduce cardiac morbidity by 5% to 9%, stroke by 8% to 14%, and all-cause mortality by 4%.57 The lack of data on the effects of RT on ambulatory BP warrants further investigation because this may be more indicative of future CV disease morbidity and mortality.58
Control of BP is even more important in individuals who already have hypertension. Although there is general agreement that endurance training lowers resting BP in patients with mild to severe hypertension,27,57 there is a paucity of data on the effects of RT alone on BP in individuals with hypertension. Only 20% of the outcomes in the 2 meta-analysis reviews were based on a mean initial resting systolic BP >140 mm Hg, whereas only 13% had a mean initial resting diastolic BP >90 mm Hg.26,27 One study that used RT in combination with aerobic exercise in middle-aged hypertensive men for 10 weeks demonstrated reductions of 13 mm Hg for both systolic and diastolic BP.59 Conversely, a similar program for 6 months in older adults with hypertension showed mean decreases in systolic and diastolic BP of 5.3 and 3.7 mm Hg, respectively.60 The change in systolic BP, although significantly lower than study entry, was not different from values in the control group. Although any reduction in BP is desirable, the available studies do not answer the question regarding the independent benefit of RT in persons initially classified as being hypertensive or prehypertensive.
Arterial Stiffness
With aging, hypertension, insulin resistance, and diabetes, there is increased arterial stiffness from degeneration of the arterial media, increased collagen and calcium content, and arterial dilation and hypertrophy. These factors lead to increased systolic BP and an increased risk of cardiac events.61 Several studies have shown that aerobic exercise is associated with reduced arterial stiffness in healthy subjects of all ages,62 competitive aerobic athletes,63 patients with coronary artery disease,64 and hemodialysis patients.65 Moreover, when aerobic exercise is combined with RT, there is no evidence of increased arterial stiffness.66 However, less is known about the independent effects or RT on arterial stiffness. Two cross-sectional studies have suggested that young and middle-aged men who participate in regular RT have greater arterial stiffness than age-matched sedentary control subjects.67,68 However, only 3 interventional studies have examined the effect of RT on arterial function.6971 Miyachi et al70 reported that RT 3 days per week for 4 months decreased carotid arterial compliance by 19% (P<0.05) in young, healthy men who were novice weight trainers. Interestingly, carotid arterial compliance returned to baseline values within 2 months after RT was discontinued. Cortez-Cooper et al69 reported that high intensity RT for 4 days per week for 3 months in young, healthy women who were novice weight trainers (n=23; age 29±1 years, mean±SD) increased carotid augmentation index (a measure of arterial wave reflection and arterial stiffness) from 8±13% to 1±18% (P<0.05), and carotid-femoral pulse-wave velocity increased (P
0.05) from 791±88 to 833±96 cm/s. Paradoxically, neither study reported increases in systolic or diastolic BP secondary to RT.
Contradictory results were recently reported by Rakobowchuk and coworkers.71 By using similar vascular measurement techniques, they found that central arterial compliance was unaltered after 3 months of RT in young men (n=28; age 23±3.9 years, mean± SD). The discrepancy between studies may be explained by differences in the RT protocols. The 2 studies in which central arterial stiffness increased used RT protocols consisting of high-intensity super sets and an extremely high volume (up to 6 sets per exercise),69,70 both of which are not commonly recommended for the majority of the population and are usually performed by competitive athletes.9,10 Conversely, the study in which arterial stiffness was unchanged used a progressive training protocol that increased intensity but not the volume of exercise over 12 weeks.
Mechanisms for Change in Arterial Compliance
Studies reporting adverse effects of RT on the arterial system have only speculated about mechanisms responsible for the changes.6770 The elastic properties of the arterial wall are determined by both structural components (eg, relative composition of elastin and collagen) and functional components (eg, vasoconstrictor tone exerted by the vascular smooth muscle cells). Because 3 to 4 months of RT are unlikely to cause marked structural changes in the arterial wall, changes in the functional components of the arterial wall need to be considered. One potential mechanism is endothelial dysfunction manifested as a reduction in the bioavailability of nitric oxide. Recent evidence, however, indicates that 4 months of RT in healthy young men does not impair endothelium-dependent vasodilation in the brachial artery.72 Another mechanism for functional change in the arterial wall is increased sympathetic tone. There is evidence that RT increases resting humoral norepinephrine levels, a surrogate marker of sympathetic nervous system activity.73 However, increased sympathetic nervous system vasoconstrictor tone is likely to be greater in peripheral muscular arteries than in central elastic arteries. Surprisingly, both studies reporting increases in stiffness of central conduit arteries after RT did not show changes in peripheral muscular arteries.69,70 Results from narrative15,17,55,56 and meta-analytical reviews26,27 do not support the contention the RT increases vascular resistance. Moreover, those findings are compatible with the absence of hypertension observed among isometric and power athletes.74,75
| Resistance Training and Obesity |
|---|
|
|
|---|
Obesity Prevention
Epidemiological evidence supports the use of increased exercise in preventing age-associated weight and fat gains.1 Exercise recommendations to treat or prevent obesity have focused mainly on aerobic activities.1 However, RT is a behaviorally feasible and efficacious alternative to endurance exercise for weight control. For example, resting energy expenditure (REE) decreases with aging, and this decrease is closely correlated to losses in skeletal muscle mass.76 RT increases muscle mass by a minimum of 1 to 2 kg in studies of sufficient duration.77 Theoretically, a gain of 1 kg in muscle mass should result in an REE increase of approximately 21 kcal/kg of new muscle.78 In practice, RT intervention studies report REE increases in the range of 28 to 218 kcal/kg of muscle.7982 RT, when sustained over years or decades, translates into clinically important differences in daily energy expenditure and age-associated fat gains. However, even without a change in REE, maintenance of muscle mass through midlife years may prevent age-associated fat gains by promoting an active lifestyle.15,83,84
Visceral Adipose Tissue
RT can reduce total body fat mass in men84,85 and women,83,8587 independent of dietary caloric restriction. However, regional distribution of fat may be more important to health than the total amount of body fat. Excessive central obesity and especially visceral adipose tissue have been linked with the development of hyperlipidemia, hypertension, insulin resistance and glucose intolerance, diabetes, and heart disease.15,17,88,89 Fat distributed in the arms and legs, however, appears to impose little or no risk.15,88,89 Although there may be a genetic predisposition for visceral adipose tissue, increasing age, high fat diets, and a sedentary lifestyle are also important determinants.
Several studies have demonstrated decreases in visceral adipose tissue after RT programs.84,85,87,90,91 Treuth and coworkers84,87 assessed body composition in older men by using dual-energy x-ray absorptiometry84 and in older women by using computed tomography87 and observed significant decreases in visceral fat after 16 weeks of RT. Ross et al90,91 used magnetic resonance imaging to measure regional fat losses after exercise combined with diet interventions. In their first study,90 both diet plus aerobic exercise and diet plus RT elicited similar losses of visceral fat that were greater than losses of whole-body subcutaneous fat. In a follow-up study,91 they isolated the effects of endurance exercise training and RT by comparing the responses to diet alone and diet combined with each training modality in middle-aged obese men. All 3 groups lost significant amounts of total body fat, and all 3 groups experienced a significantly greater visceral fat loss compared with whole-body subcutaneous fat loss. The changes amounted to a 40% reduction in visceral fat in the RT and diet group, 39% in the endurance training and diet group, and a 32% reduction in the diet-only group.
One study has raised the possibility of gender specificity in visceral fat reduction in response to RT. Hunter et al85 studied older women and men (age, 61 to 77 years) after 25 weeks of supervised RT. Both genders significantly increased muscle mass but men increased muscle more than women (2.8 versus 1.0 kg, respectively). Similar decreases in total body fat mass were found for the men (1.8 kg) and women (1.7 kg). However, women lost a significant amount of visceral adipose tissue (131 to 116 cm2), whereas the men did not (143 to 152 cm2). Similarly, women also lost a significant amount of subcutaneous adipose tissue (254 to 239 cm2), but men did not (165 to 165 cm2). Conversely, in a 6-month study of RT combined with aerobic exercise, men lost more visceral adipose tissue than women, but losses of total and subcutaneous adipose tissue were similar.60 Although more research is needed to clarify these possible gender-specific responses, the overall available body of literature supports the use of RT, with or without aerobic exercise, and with or without diet modification, as an effective intervention that contributes to the reduction of abdominal obesity.
Obesity Reduction
Studies of the efficacy of RT in the context of total body weight loss have had mixed results. Studies that use more severe caloric intake restriction have not shown gains in muscle mass,92,93 whereas RT studies with less severe caloric restriction have shown muscle mass gains with only modest losses in body weight.87,90,91,94 RT studies that attempt to maintain caloric balance during the intervention typically do not observe major changes in body weight in either gender, despite significant reductions in fat mass and percent body fat.15,84,85,87 In essence, body weight does not change much because loss of fat mass is generally offset by the gain in muscle mass. Conversely, endurance traininginduced decreases in fat mass are more likely to be associated with reductions in body weight because there is no offsetting gain in muscle mass.
| Resistance Training and Dyslipidemia |
|---|
|
|
|---|
Most interventional studies have failed to adequately control for normal variations in lipoproteins and lacked proper dietary controls and/or lacked statistical power. When these factors are controlled, most studies show no improvement in lipid profiles after RT in either middle-aged or older adults.40,87,98,99 However, the lack of significant lipoprotein-lipid changes with RT may be due to the fact that total cholesterol values for most study groups has been
200 mg/dL at study entry. Individuals with normal lipoprotein-lipid profiles may require greater exercise stimulus and energy expenditure coupled with significant reductions in body weight to further improve lipid profiles. Alternatively, Shoup and Durstine100 postulated that for changes in blood lipids to occur, lipoprotein lipase levels need to increase or remain elevated postexercise, or associated hepatic lipase must be suppressed to reduce conversion of high-density lipoprotein cholesterol (HDL-C) subfraction HDL2-C to HDL3-C. It is possible that the RT stimulus in most studies does not generate these cellular changes.
| Screening and Precautions |
|---|
|
|
|---|
Selected individuals should consult their healthcare practitioners before beginning a vigorous exercise program. For those at moderate risk or higher, such as men age 45 years and over, women age 55 years and over, those with major risk factors for arteriosclerosis, and those with diabetes at any age, the American College of Sports Medicine recommends a medical history and physical examination including an exercise stress test before initiating a vigorous exercise program.95 Because of the marked rise in BP with RT, those with uncontrolled hypertension (systolic BP
160 mm Hg and/or diastolic BP
100 mm Hg) should be controlled to lower levels before starting an exercise program.102 High-intensity RT should also be avoided by individuals who have active proliferative retinopathy or moderate or worse nonproliferative diabetic retinopathy. To minimize excessive BP responses, individuals should be told to avoid extended breath-holding during their workouts.35,80,103
| Resistance Training Exercise Prescription |
|---|
|
|
|---|
|
|
| Summary |
|---|
|
|
|---|
(1) RT does not appear to alter glucose tolerance or glycemic control regardless of age, unless baseline glucose tolerance is abnormal. Nonetheless, most studies show that RT improves insulin action either through reductions in acute insulin responses during an oral glucose tolerance test or increased glucose uptake during glycemic clamp procedures. Moreover, RT significantly decreases HbA1c in diabetic men and women regardless of age, and this effect is observed even in the absence of a lasting effect of RT on fasting glucose.
(2) In healthy, normotensive persons, RT elicits reductions of approximately 3 mm Hg for both systolic and diastolic BP. Future studies are needed in individuals initially classified as hypertensive or prehypertensive to determine the extent to which RT lowers BP when it is elevated at baseline. Until these studies are performed, an RT program combined with aerobic exercise should be recommended for lowering BP in hypertensive adults.
(3) There is some evidence that RT can increase central arterial stiffness during high-intensity and high-volume training regimens, but an explanation for this effect has not been determined. No studies have found increased BP or peripheral vascular resistance secondary to RT.
(4) There is good evidence that RT reduces total body fat mass in men and women, independent of dietary caloric restriction. There is also good evidence that RT reduces visceral adipose tissue in older men and women.
(5) There is little evidence that RT improves lipoprotein-lipid profiles. However, total cholesterol values for most study groups have been
200 mg/dL at study entry. Individuals with normal lipoprotein-lipid profiles may require greater exercise stimulus and energy expenditure coupled with significant reductions in body weight in order to further improve lipid profiles.
(6) Although RT by itself may have limited beneficial effects on CV disease risk factors, this mode of exercise is beneficial in the prevention and management of musculoskeletal injuries and disorders, osteoporosis, and sarcopenia. RT also reduces susceptibility to falls and prevents or delays impaired physical function in frail and elderly persons.
(7) Although performing RT by itself rather than in combination with aerobic exercise appears to contribute to some aspects of CV disease reduction, the available data do not permit accurate estimation of the magnitude of the risk reduction. Thus, for the individual without existing cardiac disease whose goal is to improve their CV health and prevent disease, there is little evidence herein to challenge existing exercise guidelines that call for moderate-intensity RT to be performed in combination with aerobic exercise.
| Acknowledgments |
|---|
None.
| References |
|---|
|
|
|---|