Physical Activity/Exercise and Type 2 Diabetes

Physical Activity/Exercise and Type 2DiabetesRONALD J. SIGAL, MD, MPH

1,2,3

GLEN P. KENNY, PHD2,3

DAVID H. WASSERMAN, PHD4

CARMEN CASTANEDA-SCEPPA, MD, PHD5

For decades, exercise has been consid-ered a cornerstone of diabetes man-agement, along with diet and

medication. However, high-quality evi-dence on the importance of exercise andfitness in diabetes was lacking until recentyears. The last American Diabetes Associ-ation (ADA) technical review of exerciseand type 2 diabetes (formerly known asnon–insulin dependent diabetes) waspublished in 1990. The present workemphasizes the advances that have oc-curred since the last technical review waspublished.

Major developments since the 1990technical review include:

● Advances in basic science, increasingour understanding of the effects of ex-ercise on glucoregulation.

● Large clinical trials demonstrating thatlifestyle interventions (diet and exer-cise) reduce incidence of type 2 diabe-tes in people with impaired glucosetolerance (IGT).

● Meta-analyses of structured exercise in-terventions in type 2 diabetes showing:1) effectiveness of exercise in reducingHbA1c, independent of body weight;and 2) association between exercisetraining intensity and change in HbA1c.

● Large cohort studies showing that lowaerobic fitness and low physical activitylevel predict increased risk of overall

and cardiovascular disease (CVD) mor-tality in people with diabetes.

● Clinical trials showing effectiveness ofresistance training (such as weight lift-ing) for improving glycemic control intype 2 diabetes.

● New data on safety of resistance train-ing in populations at high risk for CVD.

Based on this new evidence, we haverefined the recommendations on the de-sired types, amounts, and intensities ofaerobic physical activity for people withdiabetes. Resistance training will now berecommended in a broader group of pa-tients and at a broader range of intensitythan done previously. There are other ar-eas in which new evidence is lacking, butwe feel that previous recommendationsmay have been more conservative thannecessary. These areas include indica-tions for exercise stress test before begin-ning an exercise program and precautionsregarding exercise in the presence of somespecific complications or suboptimalmetabolic control. The levels of evidenceused are defined by the ADA (see Ref. 1).

A new Handbook of Exercise was pub-lished in 2002 by the ADA, including 40articles by leading experts on specifictopics related to exercise and diabetes.Space limitations do not allow the presentwork to be comprehensive and, where ap-propriate, we refer the reader to chapters

in the Handbook of Exercise and other re-view articles for additional details. Thepresent review focuses on type 2 diabetes.Issues primarily germane to type 1 diabe-tes will be covered in a subsequent techni-cal review.

DefinitionsThe following definitions are based onthose outlined in “Physical Activity andHealth,” the 1996 report of the SurgeonGeneral (2).Physical activity. Bodily movementproduced by the contraction of skeletalmuscle that requires energy expenditurein excess of resting energy expenditure.Exercise. A subset of physical activity:planned, structured, and repetitive bodilymovement performed to improve ormaintain one or more components ofphysical fitness. In the present review, theterms “physical activity” and “exercise”will be used interchangeably.Physical fitness. This includes cardiore-spiratory fitness, muscular fitness, andflexibility.Cardiorespiratory fitness (also knownas cardiorespiratory endurance or aer-obic fitness). The ability of the circula-tory and respiratory systems to supplyoxygen during sustained physical activity.The gold standard for measurement ofcardiorespiratory fitness is a test of maxi-mal oxygen uptake (VO2max), typicallyperformed using indirect calorimetry on atreadmill or bicycle ergometer. Cardiore-spiratory fitness can be estimated accu-rately using graded maximal exercisetesting on standard treadmill or bicycleergometer protocols without indirect cal-orimetry (3).Aerobic exercise. This consists ofrhythmic, repeated, and continuousmovements of the same large musclegroups for at least 10 min at a time. Ex-amples include walking, bicycling, jog-ging, continuous swimming, wateraerobics, and many sports. When per-formed at sufficient intensity and fre-quency, this type of exercise increasescardiorespiratory fitness.Intensity of aerobic exercise. This willbe described as “moderate” when it is at40–60% of VO2max (�50–70% of maxi-

● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

From the 1Department of Medicine, University of Ottawa, Ottawa, Canada; the 2School of Human Kinetics,University of Ottawa, Ottawa, Canada; the 3Clinical Epidemiology Program, Ottawa Health Research Insti-tute, Ottawa, Canada; the 4Department of Molecular Physiology and Biophysics, Vanderbilt University,Nashville, Tennessee; and the 5Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts Uni-versity, Boston, Massachusetts.

Address correspondence and reprint requests to Ronald J. Sigal, Clinical Epidemiology Program, OttawaHealth Research Institute, 1053 Carling Ave., Ottawa, Ontario, Canada K1Y 4E9. E-mail: [email protected].

Received and accepted for publication 16 June 2004.Abbreviations: ACSM, American College of Sports Medicine; ADA, American Diabetes Association;

CAD, coronary artery disease; CVD, cardiovascular disease; DPP, Diabetes Prevention Program; ECG, elec-trocardiogram; EGP, endogenous glucose production; IGT, impaired glucose tolerance; IRS, insulin receptorsubstrate; MAP, mitogen-activated protein; MET, metabolic equivalent; NEFA, nonesterified fatty acid; PI,phosphatidylinositol I; RM, repetition maximum.

A table elsewhere in this issue shows conventional and Systeme International (SI) units and conversionfactors for many substances.

© 2004 by the American Diabetes Association.

R e v i e w s / C o m m e n t a r i e s / A D A S t a t e m e n t sT E C H N I C A L R E V I E W

2518 DIABETES CARE, VOLUME 27, NUMBER 10, OCTOBER 2004

mum heart rate) and “vigorous” when it isat �60% of VO2max (�70% of maximumheart rate).Muscular fitness. This refers to strength(the amount of force a muscle can exert)and muscular endurance (the ability ofthe muscle to continue to perform with-out fatigue).Resistance exercise. Activities that usemuscular strength to move a weight orwork against a resistive load. Examplesinclude weight lifting and exercises usingweight machines. When performed withregularity and moderate to high intensity,resistance exercise increases muscularfitness.Intensity of resistance exercise. Thiswill be described as “high” if the resistanceis �75% of the maximum that can belifted a single time (�75% of 1-RM [rep-etition maximum]) and “moderate” if re-sistance is 50–74% of 1-RM.Flexibility. This term refers to the rangeof motion available at joints.Flexibility exercise. This is exercise(typically stretching) aimed at increasingor maintaining range of motion at joints.MET (metabolic equivalent). A MET isa unit of intensity equal to energy expen-diture at rest. Physical activity at 3 METsuses three times as much energy as sta-tionary sitting. MET-hours are units of ex-ercise volume in which intensity in METsis multiplied by duration of the activity inhours.

PHYSIOLOGY OF FUELMETABOLISM DURINGEXERCISEFor more detailed reviews of the physiol-ogy and basic science of exercise, see Refs.4–6.

Exercise results in a shift in fuel usageby the working muscle from primarilynonesterified fatty acids (NEFAs) to ablend of NEFAs, glucose, and muscle gly-cogen. Muscle glycogen is the chief sourceof energy during the early stages of stren-uous exercise, while with increasing exer-cise duration the contribution ofcirculating glucose and particularly NE-FAs become more important as muscleglycogen gradually depletes. The origin ofcirculating glucose also shifts from he-patic glycogenolysis to gluconeogenesis.With increasing exercise intensity the bal-ance of substrate usage shifts to greatercarbohydrate oxidation. Although themetabolic response to exercise is influ-enced by numerous factors (e.g., nutri-tion, age, type of exercise, and physicalcondition), the most important factors af-fecting fuel utilization are generally workintensity and duration.

Regulation of fuel mobilizationduring aerobic exerciseFuel mobilization is controlled duringaerobic exercise largely by the neuroen-docrine system. If exercise is sustained, adecrease in insulin secretion and in-

creases in glucagon, catecholamines, cor-tisol secretion, and other hormones areobserved. Postulated signals causingthese hormonal and neural changes maybe related to deficits in fuel availability,neural feed-forward mechanisms, or in-creased afferent nerve activity from theworking limb, splanchnic bed, or carotidsinus (7,8). In addition to neuroendo-crine factors, other parameters such asblood-flow shifts and subtle changes inglycemia or metabolic state play a rolein the control of fuel metabolism duringexercise.Endogenous glucose production. En-dogenous glucose production (EGP) isclosely coupled to the increase in muscleglucose uptake during moderate exercise.Studies conducted in animal models andhuman subjects have defined the impor-tance of insulin and glucagon in the stim-ulation of EGP during light- andmoderate-intensity aerobic exercise (7).The exercise-induced increment in gluca-gon stimulates glycogenolysis and glu-coneogenes is (7) . Glucagon alsostimulates hepatic amino acid metabo-lism (9) and fat oxidation (7), providingprecursors for gluconeogenesis and en-ergy to fuel it. The decrease in insulin dur-ing exercise is necessary for the fullglycogenolytic response (7). When thedecline in portal vein insulin is elimi-nated, the increase in EGP is reduced by�50%. The importance of the exercise-induced increase in EGP to whole-bodyglucose homeostasis is illustrated in Fig.1. If the liver did not release more glucosein response to exercise, hypoglycemiawould result.

In contrast, in very intense aerobic ex-ercise (�80% of V O2max), the cat-echolamines likely play a more importantrole. In this situation, norepinephrine andepinephrine levels rise as much as 15-foldfrom baseline, and glucose production inyoung fit subjects rises about 7-fold dur-ing exercise (10–12). In intense exerciseunder islet cell clamp, where the gluca-gon-to-insulin ratio does not rise, glucoseproduction increases as much as it doeswithout islet cell clamp (12). Infusion ofnorepinephrine (13), epinephrine (14),or both (15) during moderate exercise in-duces glucose production similar to thatcharacteristic of intense exercise. Studiesof intense exercise during �-adrenergicblockade with propranolol in normal (11)and type 1 diabetic (16) subjects foundthat glucose production during exercise

Figure 1—Potential impact of impaired EGP on glucose homeostasis during exercise. If EGP is notcoupled to the increase in glucose utilization, hypoglycemia may result.

Sigal and Associates

DIABETES CARE, VOLUME 27, NUMBER 10, OCTOBER 2004 2519

was greater than that in subjects exercis-ing without �-blockade. Conversely, an-other study (17) found that subjectsperforming intense exercise during �-ad-renergic blockade with phentolamine hadslightly less glucose production than sub-jects exercising at similar intensity with-out adrenergic blockade. No study hasused combined �- and �-adrenergicblockade during very intense exercise.These studies are difficult to interpret dueto the lack of specificity of these pharma-cological blockers. A method was devel-oped in the dog that uses intraportalpropranolol and phentolamine infusionto selectively block hepatic adrenergic re-ceptors (18). Results obtained in thismodel were consistent with those usingsystemic adrenergic blockade, showingthat EGP was not reliant on hepatic ad-renergic receptor stimulation duringheavy exercise (18). In normal subjects,plasma insulin doubles soon after the endof a very intense exercise session, restor-ing glycemia to baseline within an hour(10). In contrast, in type 1 diabetes, inwhich endogenous insulin cannot in-crease, hyperglycemia after very intenseexercise lasts at least several hours(19,20).

Type 2 diabetic patients with a mildto moderate elevation in glucose levelsmay experience a fall in glucose duringexercise due to impaired endogenous glu-cose output. This population, whenmaintained on diet therapy alone or diet

and sulfonylurea therapy with postab-sorptive plasma glucose in excess of 200mg/dl and normal basal insulin, shows afall in glycemia of �50 mg/dl during a45-min exercise bout (21). High-intensityintermittent exercise performed in post-prandial type 2 diabetic subjects (22) hasthe same plasma glucose– and insulin-lowering effect as moderate-intensity ex-ercise of equivalent caloric requirement(22).Fat metabolism. Moderate exercise isassociated with an �10-fold increase infat oxidation. This is due to increased en-ergy expenditure coupled with greaterfatty acid availability. The increase in fattyacid availability is due both to an increasein lipolysis and decreased re-esterificationof NEFA to triglycerides (23). AcuteNEFA release from adipose tissue is regu-lated primarily by the actions of insulinand the catecholamines. When the exer-cise-induced fall in insulin is prevented,the increase in NEFA levels is prevented(7). NEFA levels are diminished duringexercise by �-blockade, presumably dueto a suppression of lipolytic activity. Notonly are catecholamines increased by ex-ercise, but adipocytes taken followingexercise have increased lipolytic respon-siveness to �-adrenergic actions of thecatecholamines (24). Besides fat mobili-zation from adipocytes, there is evidence(25,26) that intramuscular triglyceridesrepresent an important fuel for workingmuscle. Metabolism of fats during exer-

cise is quantitatively different in obesetype 2 diabetic subjects in relation tohealthy subjects. In this population, utili-zation of plasma free fatty acids is re-duced, while intramuscular triglycerideutilization is increased (27). Interestingly,lean type 2 diabetic individuals do nothave this adaptation to exercise (28).Muscle glycogenolysis. Glycogenbreakdown is regulated by glycogenphosphorylase. It is interesting that al-though muscle glycogenolysis increaseswith increasing work rate, phosphorylasetransformation to its active phosphory-lated form is not (29). This suggests thatallosteric regulators may be important ac-tivators of glycogen phosphorylase dur-ing exercise (30). �-Adrenergic receptorstimulation by catecholamines plays amajor role in the mobilization of muscleglycogen during exercise (7).

Exercise-induced muscle glucoseuptakeMuscle glucose uptake requires three se-rial steps (Fig. 2). These are the delivery ofglucose from the blood to the muscle,transport of glucose across the musclemembrane, and phosphorylation of glu-cose within the muscle. The effect of ex-ercise on these specific steps thatcomprise muscle glucose uptake is dis-cussed below and illustrated in Fig. 1.Glucose delivery from blood to muscle.Muscle interstitial glucose would fall pre-cipitously and the glucose transport gra-

Figure 2—Schematic representation of the control of muscle glucose uptake during exercise. Control of glucose is distributed between three serialsteps. These are extracellular glucose delivery from blood to muscle membrane, muscle membrane glucose transport, and intracellular glucosephosphorylation. Below each gear are known mechanisms of regulation. Arrows associated with those mechanisms reflect exercise adaptations. Notethat acute exercise increases hexokinase II protein; however, this adaptation is slow in onset. Modified from Wasserman and Halseth (47).

Exercise in type 2 diabetes


dient would be insufficient to sustainglucose uptake if it were not for themarked increase in blood flow to workingmuscle. The exercise-induced increase inglucose delivery is so effective at main-taining interstitial glucose that an increasein muscle fractional glucose extraction isnot required for the increase in muscleglucose uptake (31). The importance ofglucose supply is supported by the closecorrelation of muscle blood flow to glu-cose uptake by the working limb (32). Inaddition, an increase in perfusion of theisolated rat hindlimb is necessary for fullcontraction-induced glucose uptake (33).Membrane glucose transport. Exerciseincreases glucose transport by stimulatingGLUT4 translocation to the muscle cellsurface (34). A possible mechanism in-volves sensing of an increase in muscleAMP, which stimulates AMP kinase, caus-ing a number of metabolic changes, in-cluding increased glucose transport (35).Muscle AMP kinase measurements are, infact, consistent with stimulation by exer-cise (36,37). Such a role for this enzyme issupported by the demonstration thatpharmacological activation of AMP kinasestimulates GLUT4 translocation (38) andglucose uptake (36,39,40) and is linkedto other changes in enzyme activities(35,41) and gene transcription (42,43)associated with exercise. AMP kinase ac-tivation is not the sole mechanism of con-traction-stimulated muscle glucoseuptake (40). In addition to AMP kinaseactivation, data suggest that nitric oxide(NO) may mediate contraction-inducedglucose uptake. Electrical stimulation in-creases muscle NO synthesis, and phar-macologic inhibition of NO synthesisdecreases uptake of a glucose analog (44).Another study (45) showed that NO syn-thase inhibition at the working leg causeda reduction in leg glucose uptake withoutaffecting blood flow. NO synthase hasalso been implicated as a mediator of AMPkinase–induced glucose uptake (46).Muscle glucose phosphorylation. Thefirst step in glucose metabolism is phos-phorylation by a hexokinase. There is ev-idence that glucose phosphorylation isthe primary limitation to glucose uptakeduring exercise (47). Hexokinase II over-expression improves the ability to con-sume glucose in mice, and reducing theglucose 6-phosphate pool by a glycogen-depleting fast enhances uptake of glucoseeven further (48). In contrast to the exten-sive work on glucose transport, very little

is known regarding the effects of exerciseon hexokinases. The percentage of hex-okinase activity associated with mito-chondria, where its specific activity isgreater, does not appear to be increasedwith exercise in human muscle (49,50).Exercise has been shown to stimulatemuscle hexokinase II gene transcription(51), leading to an increase in its protein(50). The increase in hexokinase activitylags behind the increase in hexokinase IImRNA (51) and may be more significantto the persistent increase in insulin actionfollowing exercise or adaptations that oc-cur with training.

Insulin-independent and insulin-sensitive muscle glucose uptakeduring exerciseThe preceding section described the stepsrequired for muscle glucose uptake. Theflux through these steps is controlled byinsulin-independent signals generatedwithin working muscle, but can be great-ly modified by actions of circulating in-sulin. Exercise increases both insulin-independent muscle glucose uptake andinsulin sensitivity. The focus below is onthese two important characteristics ofexercise.Insulin-independent glucose uptake.The cell signaling pathway of contraction-stimulated glucose transport is distinctfrom that for insulin-stimulated glucosetransport (52,53). Although the increasein membrane transporters in response toboth insulin and exercise result from anincrease in GLUT4 translocation, thesestimuli recruit GLUT4 from different in-tracellular pools (54). Evidence that cellsignaling for glucose uptake is differentfor exercise and insulin is supported bythe demonstration that muscle contrac-tion does not increase phosphorylation ofinsulin receptor substrate (IRS)-1 and -2or phosphatidylinositol I (PI) 3-kinase(55), all of which are involved in insulinsignaling. In addition, wortmannin, an in-hibitor of PI 3-kinase, eliminates insulin-stimulated glucose uptake but not glucosetransport in isolated contracting muscle(34). The importance of insulin-independent mechanisms in control ofexercise-stimulated muscle glucose up-take is further exemplified by studies intype 2 diabetic individuals. Although type2 diabetic individuals are usually insulinresistant, they are not resistant to thestimulatory effects of exercise on glucoseutilization. Type 2 diabetic subjects retain

the capacity to translocate GLUT4 to thesarcolemma in response to exercise (56).The functional recruitment of GLUT4transporters coupled to elevated circulat-ing glucose levels can actually lead to agreater rate of glucose utilization by mus-cle of people with type 2 diabetes.Insulin-dependent glucose uptake (in-sulin sensitivity). Exercise and insulinstimulate glucose utilization synergisti-cally (7). The primary route of insulin-mediated glucose metabolism at rest andin the postexercise state is nonoxidativemetabolism (7). Exercise, however, shiftsthe route of insulin-stimulated glucosedisposal so that all glucose consumed bymuscle is oxidized (7). The effects of thisincrease in insulin action are probablymost important in the postprandial stateand in the intensively treated diabeticstate, when insulin levels are higher thanthose that normally accompany exercise.Several mechanisms have been proposedto explain how exercise enhances insulinaction (6,57,58). Hemodynamic adjust-ments increase capillary surface area inworking muscle, increasing the availabil-ity of insulin. Exercise may also increaseinsulin-stimulated glucose utilization by amechanism secondary to insulin’s sup-pressive effect on non-NEFAs. Insulin ac-tion is also directly enhanced at workingmuscle by activation of postinsulin recep-tor signaling (32).

Carbohydrate ingestion and exerciseGlucose feeding has been shown (7) toimprove exercise endurance. The under-lying mechanism for this improvement isprobably related to increased glucoseavailability to working muscle. Theamount, form, and timing of an oral car-bohydrate load, along with the durationand intensity of exercise, will determinehow effective glucose ingestion is at sus-taining glucose availability to the workingmuscle.

Carbohydrate ingestion slows themobilization of endogenous fuels duringprolonged exercise. It also slows the rateof fall of circulating glucose that wouldotherwise occur or leads to an overt in-crease in circulating glucose (7). At leasttwo important endocrine changes accom-pany the increase in glucose availability.The exercise-induced fall in insulin andrise in glucagon are attenuated or elimi-nated altogether. The absence of the fall ininsulin attenuates the increases in lipoly-sis and EGP, whereas a reduction in glu-



cagon will reduce the latter (7). Althoughinsulin acts to suppress glycogen break-down, multiple signals are present inworking muscle, and glycogen is gener-ally not spared by carbohydrate ingestion(59).

The metabolic availability of ingestedcarbohydrate depends on the composi-tion and quantity of the substrate load. Inaddition, exercise parameters (i.e., workintensity, duration, and modality) alsodetermine the availability of ingested glu-cose. As a consequence, it is difficult toascribe an exact metabolic efficiency ofingested glucose. In any case, a reasonableestimate might be that �40% of a 50-gglucose load, ingested at the start of mod-erate exercise, is metabolized during thefirst hour (60). There is probably little dif-ficulty in delivering adequate amounts ofingested glucose to the muscle duringlight exercise. Because glucose oxidationby muscle increases at higher work rates,it may no longer be possible to absorbadequate amounts of ingested glucose. Asa consequence, limitations to the oxida-tion of ingested glucose may shift at dif-ferent work intensities: low muscledemand limits glucose oxidation at lightwork rates, whereas absorption from thegastrointestinal tract limits it at high workrates (61). In humans, the maximum ab-sorption rate of carbohydrate in the gutduring exercise is about 1 g/min (62).

Postexercise glucose metabolismExercise leads to diverse adaptations thathave significant impact on glucoregula-tion, even after the cessation of exercise.These adaptations largely share the com-mon purpose of replenishing fuel stores,particularly muscle and liver glycogen.Muscle. Stimulation of muscle glucoseuptake persists well after exercise. Theadded glucose taken up after exercise ischanneled into glycogen (32). Glycogenrepletion is characterized by a markedand persistent increase in insulin action(32). This increase in insulin action oc-curs without increasing the tyrosinephosphorylation of the insulin receptor,IRS-1, IRS-2, and shc (63,64). Moreover,PI 3-kinase activity is not increased (65).The presence of the muscle insulin recep-tor may not even be necessary for the in-creased effects of insulin on muscleglucose uptake and glycogen synthase af-ter exercise (66). This implies that the ef-fects of prior exercise are mediated by

nonmuscle cells or by downstream signal-ing (66). It has also been proposed that apostreceptor modification may be linkedto the glycogen depleting effect of exer-cise. However, the improved effect of in-sulin on glucose uptake can persist afterexercise, even when pre-exercise glyco-gen levels have been restored (32). Thecellular basis for the persistent increase ininsulin sensitivity may, at least in part,relate to increases in skeletal muscleGLUT4 (67), glycogenin (68), and hex-okinase II (51) during exercise recovery.It is noteworthy that AMP kinase activa-tion leads to a subsequent increase in in-sulin sensitivity, much like exercise (69).

Muscle contraction activates a num-ber of signaling pathways, in addition tothose involved with glucose transport, ina manner that is influenced by the inten-sity and duration of work (70,71). Theseinclude the activation of the mitogen-activated protein (MAP) kinase (64,72),Akt (71), and p70s6k (70) pathways. Con-traction inhibits the glycogen synthase ki-nase-3 pathway, at least in rodent muscle(73), promoting glycogen synthesis. It islikely that the activation of some of thesesignaling cascades are important to thepersistent adaptations to exercise and notthe acute metabolic response. Activationof MAP kinase, Akt, p70s6k, and AMP ki-nase pathways and deactivation of the gly-cogen synthase kinase-3 pathway all arecapable of stimulating gene transcriptionor protein synthesis (53).Liver. It was recently shown (74) thatprior exercise increases the capacity of theliver to consume glucose. These data areconsistent with studies (75) using 13Cmagnetic resonance spectroscopy thatshowed that ingestion of glucose immedi-ately after prolonged exercise increasedliver glycogen resynthesis. The liver, likemuscle, is more insulin sensitive after ex-ercise (76). Also like muscle, a greaterfraction of glucose taken up by the liverafter exercise is nonoxidatively metabo-lized (77).

Metabolic adaptations to regularphysical activityAdaptations to chronic exercise dependon the exercise parameters (intensity, du-ration, frequency, and mode) and thecharacteristics of the individual (presenceof disease, fitness, and genetic determi-nants). Endurance and resistance exerciselead to biochemical adaptations specific

to the training regimen. Adaptations toendurance exercise enable the muscle touse O2 and blood-borne fuels, whereasthose for resistance exercise lead to im-proved force generation (e.g., hypertro-phy, contractile properties). Of specificinterest to people with diabetes are thoseadaptations that directly affect the metab-olism of glucose.

The adaptation of the pancreatic�-cell to exercise training has been themost widely assessed of the endocrine or-gans. Basal and glucose-stimulated insu-lin levels are both reduced in response toregular exercise due to reduced secretion(7). Training results in decreases in themRNA for proinsulin and glucokinase inthe pancreas (78). This suggests that thereare at least two potential cellular mecha-nisms for decreased insulin secretion.First, the reduction in proinsulin mRNAsuggests that the synthesis of insulin isreduced. Second, because glucokinase isnecessary for glucose sensing in the pan-creas, the reduction in glucokinasemRNA may explain the decreased sensi-tivity of the �-cell to glucose.

Exercise training, whether endurance(79) or resistance (80), leads to increasedmuscle GLUT4. This increase in GLUT4probably contributes to the increased ca-pacity for insulin-stimulated glucosetransport in trained subjects. This, ofcourse, has important therapeutic impli-cations for people with insulin resistance.Exercise training has been shown to stim-ulate insulin-stimulated PI 3-kinase inmuscle (81–83). There is evidence thatthis increase is due to IRS-1–associated PI3-kinase activity (81,83). Because PI 3-ki-nase is an important step in the recruit-ment of GLUT4 by insulin to the musclecell surface, it is reasonable to postulatethat this is one site at which regular phys-ical activity may affect insulin signaling.Regular physical activity leads to an in-crease in basal and insulin-stimulatedMAP kinase pathway activity (84). Al-though an increase in the activity of thissignaling pathway is not thought to benecessary for the exercise-induced in-crease in insulin-stimulated glucose up-take, it may be associated with otheradaptive changes in muscle. The mecha-nisms through which aerobic and resis-tance exercise increase glucose disposalare similar (85). Although resistance ex-ercise has greater propensity than aerobicexercise to increase muscle mass andthereby glucose storage space (5), this is



only one of many factors explaining itseffects on glucose disposal (85).

Trained subjects have an increasedability to mobilize and store NEFAs (86).The increased ability to mobilize NEFAsoccurs at least in part due to increasedadipocyte catecholamine sensitivity and ismediated by an increased formationand/or improved effectiveness of cAMP(87). Training increases the capacity ofmuscle to extract NEFAs from the bloodand oxidize them (7). The mechanism forthis adaptation may pertain to the en-hanced capacity of trained muscle to oxi-dize fat (88,89) or to increased number orfunction of muscle fatty acid transport orbinding proteins (89,90). It has been hy-pothesized that an excess accumulation ofintramuscular lipid is associated with in-sulin resistance. This concept is inconsis-tent with results from exercise-trainedsubjects. Compared with sedentary controlsubjects, athletes with high aerobic fitnesshave increased intramuscular lipids, but aremore insulin sensitive, not less (91).

EVALUATION OF THEDIABETIC PATIENT BEFORERECOMMENDING ANEXERCISE PROGRAMFor a more detailed review on this sub-ject, see Ref. 92.

Before beginning a program of phys-ical activity more vigorous than briskwalking, people with diabetes should beassessed for conditions that might contra-indicate certain types of exercise or predis-pose to injury (e.g., severe autonomicneuropathy, severe peripheral neuropathy,or preproliferative or proliferative retinopa-thy), which require treatment before begin-ning vigorous exercise, or that may beassociated with increased likelihood ofCVD. The patient’s age and previous phys-ical activity level should be considered.

One potential area of controversy isthe circumstances under which a gradedexercise electrocardiogram (ECG) stresstest should be considered medically indi-cated. We unfortunately did not find anyrandomized trials or large cohort studiesevaluating the utility of exercise stresstesting specifically in people with diabe-tes; the lack of such studies is an impor-tant gap in the literature. Previous ADAguidelines (93) have suggested that beforebeginning a vigorous or moderate exer-cise program, an exercise ECG stress testshould be done in all diabetic individuals

aged �35 years and in all individualsaged �25 years in the presence of evenone additional CVD risk factor (diabetesduration �10 years for type 2 diabetes or�15 years for type 1 diabetes, hyperten-sion, dyslipidemia, smoking, proliferativeretinopathy, nephropathy including mi-croalbuminuria, peripheral vascular dis-ease, or autonomic neuropathy). If thisprevious recommendation were followedstrictly, the great majority of people withdiabetes, including a large number ofyounger individuals with very low abso-lute risk of CVD, would require formalexercise stress testing before beginningeven a moderate-intensity exercise pro-gram. The costs of such widespread stresstesting might be prohibitive (92). Theprevalence of both symptomatic andasymptomatic coronary artery disease(CAD) is higher in both type 1 and type 2diabetic individuals compared with non-diabetic individuals of the same age-group. However, many younger diabeticpatients have relatively low absolute riskfor a coronary event. For example, a 38-year-old Caucasian nonsmoking manwith diabetes for 5 years, HbA1c 7.5%,systolic blood pressure 130 mmHg, totalcholesterol 5.2 mmol/l, and HDL choles-terol 1.1 mmol/l would have a 10-yearCAD risk of only 7.3% or �0.7% per year,calculated using the U.K. Prospective Di-abetes Study (UKPDS) Risk Engine (94)(www.dtu.ox.ac.uk/riskengine/download.htm). The lower the absolute CADrisk, the higher the likelihood of a false-positive test. A recent systematic reviewfor the U.S. Preventive Services TaskForce came to the conclusion that stresstests should usually not be recommendedto detect ischemia in asymptomatic indi-viduals at low CAD risk (�10% risk of acardiac event over 10 years) because therisks of subsequent invasive testing trig-gered by false-positive tests outweighedthe expected benefits from the detectionof previously unsuspected ischemia(95,96).

There is, however, some value to per-forming a maximal aerobic exercise test ina broader range of individuals. In additionto screening for exercise-induced isch-emia, a maximal exercise test can provideuseful information regarding maximalheart rate and blood pressure responses todifferent exercise levels, initial perfor-mance status, and prognosis, and there-fore is potentially of some benefit to anyindividual, diabetic or otherwise. With-

out a maximal exercise test, one cannotknow a given individual’s maximum heartrate or the heart rate associated with agiven percentage of the maximum. Use ofthe Borg scale (Rating of Perceived Exer-tion [97]), with target perceived intensi-ties of “moderate,” “somewhat hard,” or“hard,” is sometimes recommended as apossible alternative to heart rate–basedtargets based on maximal exercise testing.A large long-term cohort study found thatexercising habitually at perceived inten-sity of “moderate,” “somewhat strong,”“strong,” or more intense than “strong”were associated with adjusted relativerisks for coronary heart disease of 0.86,0.69, and 0.72, respectively, comparedwith exercising at perceived intensity of“weak” or less intense (98). However,there is a great deal of variability amongindividuals in terms of the perceived ex-ertion associated with performing thesame exercise at the same objectively de-fined exercise intensities (99). Likewise,the same individuals often have differentratings of perceived exertion when per-forming different exercises at the same in-tensities (e.g., running or bicycling at thesame percentage of heart rate reserve) andeven at equivalent stages of differenttreadmill protocols (Bruce versus Balke)(100).

Therefore, available clinical evidencedoes not support any specific definitiverecommendations regarding which indi-viduals should undergo stress testing. Po-tential benefits must be weighed againstrisks and costs. Our recommendationsshould be considered in this context.

A stress test is most useful in terms ofpositive predictive value for coronaryischemia when the probability of CAD isat least moderate. When the probability ofCAD is low (e.g., �10% over 10 years),the number of false-positive tests is likelyto be substantially greater than the num-ber of true-positive tests.

Therefore, we propose the followingrevised criteria for deciding when a stresstest is indicated for detection of ischemia.These criteria would encompass virtuallyall people with diabetes with a 10-yearCAD risk of at least 10% (1% per year).

Recommendations: indications forgraded exercise test with ECGmonitoringIn the absence of contraindications(101,102), maximal exercise testingcould be considered in all diabetic indi-



viduals in order to assess maximal heartrate, set exercise intensity targets, and as-sess functional capacity and prognosis. Agraded exercise test with ECG monitoringshould be seriously considered before un-dertaking aerobic physical activity withan intensity exceeding the demands of ev-eryday living (more intense than briskwalking) in previously sedentary diabeticindividuals whose 10-year risk of a coro-nary event is �10%. This risk could beestimated directly using the UKPDS RiskEngine (www.dtu.ox.ac.uk/riskengine/download.htm) (94) and would corre-spond approximately to meeting any ofthe following criteria:

● Age �40 years, with or without CVDrisk factors other than diabetes

● Age �30 years and● Type 1 or type 2 diabetes of �10

years’ duration● Hypertension● Cigarette smoking● Dyslipidemia● Proliferative or preproliferative reti-

nopathy● Nephropathy, including microalbu-

minuria● Any of the following, regardless of age

● Known or suspected CAD, cerebro-vascular disease, and/or peripheralvascular disease

● Autonomic neuropathy● Advanced nephropathy with renal

failure

The above should not be construed asa recommendation against stress testingfor individuals without the above risk fac-tors or for those who are planning less-intense exercise.Level of evidence: E.

PHYSICAL ACTIVITY ANDPREVENTION OF TYPE 2DIABETESRecent clinical trials, and a number oflarge cohort studies, provide strong evi-dence for the value of physical activity inreducing the incidence of type 2 diabetes.The Da Qing IGT and Diabetes Study(103) was the first randomized trial eval-uating lifestyle interventions for the pre-vention of type 2 diabetes. In this study,577 people with IGT from 33 clinics wererandomized, by clinic, to diet only, exer-cise only, diet plus exercise, or control.After 6 years of follow-up, cumulative in-

cidence of type 2 diabetes was 68% incontrol, 44% in diet only, 41% in exerciseonly, and 46% in diet plus exercisegroups. This study provides evidence thatboth diet and exercise can be effective di-abetes prevention modalities, althoughtheir effects were not additive. The DaQing study’s subjects were far leaner(mean BMI �23 kg/m2) than most peoplewith IGT in the western world.

More compelling evidence for the ef-fectiveness of lifestyle interventionscomes from two randomized controlledtrials: the Finnish Diabetes PreventionStudy (104,105) and the U.S. DiabetesPrevention Program (DPP) (106,107). Inthe Finnish Diabetes Prevention Study(104,105), 522 overweight subjects, aged40–65 years, with IGT were randomlyassigned to a lifestyle intervention or con-trol group. The goals were to reduceweight by at least 5%; perform moderate-intensity exercise at least 30 min/day;limit total and saturated fat intake to �30and �10%, respectively, of energy con-sumed; and increase fiber intake to �15g/1,000 kcal. Intervention group subjectshad 1-h meetings with a dietitian seventimes in the first year and every 3 monthssubsequently. Subjects in the interven-tion group were also offered an individu-alized exercise plan, thrice-weeklysupervised facility-based aerobic and re-sistance exercise for 6–12 months free ofcharge, but the proportion of subjectsparticipating in this facility-based trainingwas not stated in the publications. Thecumulative incidence of type 2 diabeteswas 11% in the intervention group and23% in the control group. In the DPP(106,107), 3,234 individuals with IGTaged �25 years were randomly allocatedto intensive lifestyle intervention, met-formin, or placebo. (A fourth arm, trogli-tazone, was terminated early whentroglitazone was withdrawn from themarket.) Subjects in the intensive lifestylegroup followed an individually super-vised program aimed at achieving andmaintaining a �7% weight loss through alow-calorie, low-fat diet and to engage inphysical activity of moderate intensity,such as brisk walking, for �150 min/week. A 16-lesson curriculum was taughtby case managers one-on-one during thefirst 24 weeks, and individual and groupsessions occurred at least monthly for theremainder of the program. After a mean of2.8 years of follow-up, the incidence ofdiabetes was 11.0, 7.8, and 4.8 cases/100

person-years in the placebo, metformin,and lifestyle groups, respectively. Relativerisk reductions in the Finnish DiabetesPrevention Study and the DPP were iden-tical at 58% with lifestyle interventionscompared with usual care or placebo. TheDPP was notable because it included alarge, racially and socioeconomically di-verse sample.

In the Malmo study (108,109), a non-randomized trial, 161 people with IGTwho participated in a diet-and-exerciseintervention were compared after 6 yearswith 56 individuals with IGT who wereoffered the same intervention and de-clined. The cumulative 6-year incidenceof type 2 diabetes was 11% in the inter-vention group and 21% in the controlgroup (108). After 12 years of follow-upin the Malmo study, overall mortalityamong IGT subjects was 6.5 per 1,000person-years in the lifestyle interventiongroup, less than one-half of the 14.0 per1,000 person-years in the IGT/no lifestyleintervention (109). Large cohort studies(110–118) have consistently found thathigher levels of physical activity and/orcardiorespiratory fitness were associatedwith reduced risk of developing type 2diabetes. This was true in most studies,regardless of the presence or absence ofadditional risk factors for diabetes such ashypertension, parental history of diabe-tes, and obesity. Comparable magnitudesof risk reduction were seen with walkingcompared with more vigorous activitywhen total energy expenditures are simi-lar (114).

Therefore, there is firm and consis-tent evidence that programs of increasedphysical activity and modest weight lossreduce the incidence of type 2 diabetes inindividuals with IGT. The two strongeststudies, the Finnish Diabetes PreventionStudy (104) and the U.S. DPP (106), donot permit one to determine the relativeimportance of physical activity versusdiet.

Recommendations: lifestylemeasures for prevention of type 2diabetesIn people with IGT, a program of weightcontrol is recommended, including atleast 150 min/week of moderate to vigor-ous physical activity and a healthful dietwith modest energy restriction.Level of evidence: A (104–107).



AEROBIC FITNESS ANDAEROBIC EXERCISE IN TYPE2 DIABETES

Effects of structured exerciseinterventions on glycemic controland body weight in type 2 diabetesFor details of the individual aerobic exer-cise clinical trials, see Ref. 119.

Most clinical trials on the effects ofphysical activity interventions in type 2diabetes have had small sample sizes andtherefore inadequate statistical power todetermine the effects of exercise on glyce-mic control and body weight. Boule et al.(119) undertook a systematic review andmeta-analysis on the effects of structuredexercise interventions in clinical trials ofduration �8 weeks on HbA1c and bodymass in people with type 2 diabetes.Twelve aerobic training studies and tworesistance training studies were included(totaling 504 subjects), and the resultswere pooled using standard meta-analyticstatistical methods. The exercise and con-trol groups did not differ at baseline inHbA1c or body weight. PostinterventionHbA1c was significantly lower in exercisethan control groups (7.65 vs. 8.31%,weighted mean difference �0.66%; P �0.001). In contrast, postinterventionbody weight did not differ between exer-cise and control groups. Meta-regressionconfirmed that the beneficial effect of ex-ercise on HbA1c was independent of anyeffect on body weight. Therefore, struc-tured exercise programs had a statisticallyand clinically significant beneficial effecton glycemic control, and this effect wasnot mediated primarily by weight loss.

Although the significant effect of ex-ercise on HbA1c in these studies is encour-aging, the lack of overall effect of exerciseon body weight in these studies is disap-pointing but not surprising. The exercisevolumes and program durations (mean53 min/session, mean 3.4 sessions/week,mean duration 15 weeks) may have beeninsufficient to achieve the energy deficitnecessary for major weight loss. Most ofthese studies did not examine body com-position, and loss of fat might have beenpartially offset by increased lean bodymass (119a).

Boule et al. (120) later undertook ameta-analysis of the interrelationshipsamong exercise intensity, exercise vol-ume, change in cardiorespiratory fitness,and change in HbA1c. This analysis was

restricted to aerobic exercise studies inwhich VO2max was either directly mea-sured or estimated from a maximal exer-cise test using a validated equation.Exercise intensities during trainingranged from �50% of VO2max to �75% ofVO2max, exercise volume 8.75–24.75MET-hours/week. Meta-analysis revealeda clinically significant 11.8% increase inVO2max in exercising groups, comparedwith a 1% decrease in control groups. Ex-ercise intensity predicted postinterven-tion weighted mean difference in HbA1c

(r � �0.91, P � 0.002) to a larger extentthan exercise volume (r � �0.46, P �0.26).

Consistent with the above, the great-est effect of exercise on HbA1c (mean ab-solute postintervention HbA1c differenceof 1.5% between exercise and controlgroups) was seen in the single study withthe highest exercise intensity (121). Inthis study, subjects exercised at 75% ofVO2max, with intervals at even higher in-tensity, for 55 min three times a week,including 5 min of warm-up and 5 min ofcool down. VO2max increased 41% in theexercising subjects versus 1% in controlsubjects. Abdominal visceral fat assessedby magnetic resonance imaging was re-ported to decline by 48% and abdominalsubcutaneous fat by 18% in the exercisinggroup in this study, which are muchlarger fat losses than seen in most exercisestudies and surprising in light of the rela-tively moderate total energy expenditureon exercise.

This meta-analysis provides supportfor higher-intensity aerobic exercise inpeople with type 2 diabetes as a means ofimproving HbA1c. The analysis, however,is limited by the fact that only one study(121) featured an unequivocally high-intensity exercise program at 75% ofVO2max. This intensity might be difficultto sustain or even hazardous for manypreviously sedentary people with type 2diabetes. Nevertheless, there was a strongdose-response relationship between exer-cise intensity across studies and both car-diorespiratory fitness and HbA1c change.These results would provide support forencouraging type 2 diabetic individualswho are already exercising at moderateintensity to consider increasing the inten-sity of their exercise to obtain additionalbenefits in both aerobic fitness and glyce-mic control.

Physical activity, aerobic fitness,and risk of cardiovascular andoverall mortalityThe effects of exercise on glycemic con-trol, while statistically and clinically sig-nificant, might be perceived as modest inrelation to the time and effort involved.After all, a similar degree of glucose low-ering could be achieved in many cases byadding a single oral hypoglycemic medi-cation. The effects of aerobic exercise onlipids and blood pressure are also rela-tively modest. However, large cohortstudies have found that higher levels ofhabitual aerobic fitness and/or physicalactivity are associated with significantlylower subsequent cardiovascular andoverall mortality, to a much greater extentthan could be explained by glucose low-ering alone.

Wei et al. (122) reported on 1,263type 2 diabetic men, a subsample of�20,000 men in the Aerobics CenterLongitudinal Study who underwent a de-tailed examination, including a maximaltreadmill exercise test with ECG monitor-ing, physical exam, blood tests, and ex-tensive health and lifestyle questionnairesbetween 1970 and 1993 and followed formortality through 31 December 1994 us-ing the National Death Index. Cardiore-spiratory fitness was classified as lowwhen treadmill time was at the bottom20% of the overall cohort (including non-diabetic subjects) for the subject’s age-group (30–39 years, 40–49 years, etc.),moderate if performance was in the 21stto 60th percentile for age-group, and highif in the highest 40% for age-group.Among the diabetic subjects, 42% wereclassified as “low fit” and 58% as “moder-ate” or “high-fit.” The 50% of diabeticsubjects who reported any participationin walking, jogging, or other aerobic ex-ercise programs in the previous 3 monthswere classified as “active,” and the other50% were classified as “inactive.” After amean of 11.7 years of follow-up, therewere 180 deaths. Mortality in the moder-ate-fit men was �60% lower than in low-fit men. Even after adjustment for age,examination year, baseline CVD, hyper-cholesterolemia, hypertriglyceridemia,BMI, hypertension, parental CVD, smok-ing, and baseline fasting glucose levels,low-fit diabetic men had a 2.2-fold greatermortality risk compared with men withmoderate or high fitness (95% CI 1.6–3.2). Similarly, after adjusting for thesame list of confounders, mortality in di-



Table 1—Clinical trials of resistance exercise with people with diabetes

Source

Sessionsper

week Duration No. of exercisesNo. of sets, rest interval, and

intensity of exercise HbA1c (mean � SD)

Eriksson et al.,1997 (152)

2 13 weeks 11 (including both upper andlower body, performed atstations arranged in acircuit)

1 set, maximum 30-s rest intervalbetween each exercise, 15–20repetitions performed at �50%of maximal exertion.

No control groupIntervention: 8.8 � 1.4 to 8.2

� 1.4%*

Honkola et al.,1998 (154)

2 5 months 8–10 (including both upperand lower body)

2 sets, total of 16–20 setsperformed, rest intervalbetween sets was �60 s, 12–15repetitions performed at amoderate intensity.

Control (type 2 diabetes): 7.7� 0.3 to 8.1 to 0.3%†

Intervention group (type 2diabetes): 7.5 � 0.3 to 7.5� 0.3%

Ishii et al., 1998(153)

5 4–6 weeks 4 (lower body) and 5 (upperbody)

2 sets, rest interval between setsof �1 min, 10 and 20repetitions for upper and lowerbody exercises, respectively,performed at 40–50% of 1-RM.

Control group (type 2diabetes): 8.8 � 2.1 to 7.6� 1.9%

Intervention group (type 2diabetes): 9.6 � 2.8 to 7.6� 1.3% (All subjects werehospitalized.)

Dunstan et al.,1998 (155)

3 8 weeks 3 (lower body) and 7 (upperbody)

3 sets; rest interval between sets30 s, with active recovery oncycle ergometer; 10–15repetitions performed at 50–55% of 1-RM.

Control group (type 2diabetes): 8.1 � 0.6 to 8.3� 0.7%

Intervention group: 8.2 � 0.5to 8.0 � 0.8%

Maiorana et al.,2002 (156)

3 8 weeks 7 (lower and upper bodyexercises)

Number of sets was increasedfrom 1 to 3; rest intervalbetween sets of 45 s, with activerecovery on a cycle ergometeror treadmill; resistance trainingintensity commenced at 55% of1-RM and increased to 65% of1-RM by week 4. Aerobicexercise was performed at 70%of peak heart rate, increasing to85% by week 6.

Control group (type 2 diabetes):8.0 � 0.5 to 8.4 � 0.6%

Intervention group (type 2diabetes): 8.5 � 0.4 to 7.9� 0.3%

Dunstan et al.,2002 (159)

3 26 weeks 8–10 (including both upperand lower body)

Intervention group: Initial 3sessions involved 2 sets, 8–10repetitions performed at 50–60%of 1-RM and increased to 3 setsin the subsequent 3 sessions. 3sets, 8–10 repetitions performedat 75–85% of 1-RM wasperformed for the duration of theintervention.

Control group (type 2diabetes, placebo exercise):7.5 � 1.1 to 7.1 to 0.8%

Intervention group (type 2diabetes): 8.1 � 0.9 to 6.9� 0.9% (All subjectsfollowed a moderate energy-restriction diet.)

Control group: placebo exercisetraining program for 3 days/week, incorporating 18different stretching/flexibilityexercises.

Castaneda et al.,2002 (160)

3 16 weeks 3 (lower body) and 2 (upperbody)

3 sets of 8–10 repetitions,progression to 75% of 1-RM,2–3 min of rest between sets.

Control group (type 2 diabetes):increased 0.4 � 1.2%

Intervention group (type 2diabetes): decreased 1.0 �1.1%‡ (All subjects wereHispanic.)

Continued on following page



abetic men reporting no physical activityparticipation in the previous 3 monthswas 1.8-fold higher than in those report-ing any participation in such activity(95% CI 1.3–2.5). Low cardiorespiratoryfitness was at least as strong a risk factorfor mortality as smoking, hypertension,hypercholesterolemia, and any of theother listed risk factors. The same re-search team has now published an up-dated study (123) with more subjects andlonger follow-up. This later analysisfound that moderate or high cardiorespi-ratory fitness was associated with lowermortality than low fitness independentlyof body composition and that essentiallyall of the association between higher BMIand higher mortality was explained byconfounding with cardiorespiratory fitness.

Because moderate fitness was associ-ated with vastly lower mortality than lowfitness, it is of interest to know the activitylevels associated with moderate fitness.Over 17,000 mainly nondiabetic partici-pants in the Aerobic Center LongitudinalStudy completed detailed physical activ-ity logs and a maximal exercise test.

Among moderately fit subjects (21st to60th percentile for age) whose only exer-cise was walking, the mean time spent perweek on exercise was 130 min for menand 148 min/week for women. Thesetimes are consistent with recommenda-tions from the U.S. Surgeon General(124) and other respected bodies (125–127) to accumulate about 150 min/weekof moderate-intensity exercise. Moder-ately fit subjects whose only exercise wasjogging or running reported a mean of 90min/week for men and 92 min/week forwomen. These times are consistent withan alternative and equally valid recom-mendation for vigorous activity 30 minthree times a week.

Hu et al. (128) reported on 5,125 fe-male nurses with type 2 diabetes whocompleted detailed health questionnairesevery 2 years, of whom 323 developednew CVD events over 14 years of follow-up. Age-adjusted relative risks accordingto average hours per week of moderate orvigorous activity were 1.0 for �1 h (ref-erence group), 0.93 for 1–1.9 h, 0.82 for2–3.9 h, 0.54 for 4–6.9 h, and 0.52 for

�7 h. There was little change with adjust-ment for BMI, smoking, and other CVDrisk factors (relative risks 1.0, 1.02, 0.87,0.61, and 0.55 for �1, 1–1.9, 2–3.9,4 – 6.9, and �7 h/week, respectively).Faster usual walking pace was associatedwith reduced CVD risk, independently oftotal physical activity score. Therefore, therewas some cardioprotection associated with2–3.9 h/week of physical activity, but agreater degree of cardioprotection associ-ated with at least 4 h/week of such activity.

Myers et al. (129) reported on 6,213consecutive men referred for treadmill ex-ercise testing for clinical reasons, includ-ing �500 with diabetes (J. Myers, e-mailcommunication, August 2003). Theyfound that in people with and in thosewithout baseline CVD, absolute exercisecapacity was a better predictor of mortal-ity risk than percentage of age-predictedexercise capacity. Every 1-MET incre-ment in treadmill performance was asso-ciated with 12% greater probability ofsurvival.

To our knowledge, no meta-analysisof the effects of exercise training on lipids

Table 1—Continued

Source

Sessionsper

week Duration No. of exercisesNo. of sets, rest interval, and

intensity of exercise HbA1c (mean � SD)

Cuff et al., 2003(158)

3 16 weeks Control group: no exercise;usual care.

Control group: usual care. Control group (type 2diabetes): decreased 0.03 �0.43

Combined aerobic andresistance group: warm up,aerobic phase (continuousaerobic activity), resistancephase (three lower bodyand two upper bodyexercises), and cool down.

Combined aerobic and resistancegroup: total duration ofsessions, 75 min. Strengthcomponent comprised 2 sets,12 repetitions, light initialintensity that increasedthereafter, techniquepermitting.

Combined group (type 2diabetes): decreased 0.1 �1.4%

Aerobic only group: warmup, aerobic phase(continuous aerobicactivity), warm down.

Aerobic component comprised60–75% heart rate reserve(progression not defined).Aerobic only group: totalduration of session was 75 min.Aerobic component comprised60–75% heart rate reserve(progression not defined). Low-impact, low-intensity aerobicactivity was used during thetime period allocated for theresistance phase in thecombined treatment group.

Aerobic only group (type 2diabetes): decreased 0.1 �1.6%

*P � 0.05 vs. baseline; †P � 0.001 vs. baseline; ‡P � 0.0001 vs. control group.



or blood pressure in people with diabeteshas been published. In the general, pre-dominantly nondiabetic population, theeffects of exercise training on blood pres-sure and lipids are relatively modest. Ameta-analysis of the effects of aerobic ex-ercise training on blood pressure (130)(54 trials, total 2,419 participants) founda weighted mean blood pressure changethrough exercise interventions of �3.84mmHg systolic and �2.58 mmHg dia-stolic. A review of the effects of super-vised, structured aerobic exercise trainingon lipids (51 trials of duration �12weeks, �4,700 subjects) (131) found amean increase in HDL cholesterol of 4.6%(P � 0.05), and reductions in plasma trig-lycerides and LDL and total cholesterol of3.7% (P � 0.05), 5.0% (P � 0.05), and1.0% (NS), respectively. Greater increasesin HDL cholesterol and decreases inplasma triglycerides have been seen withexercise programs that are more rigorousin terms of both volume and intensitythan those that have been evaluated indiabetic subjects (132). Aerobic exercisetraining increases lipoprotein lipase activ-ity and reduces the number of apoli-poprotein B–containing particles (133–136).

Potential mechanisms through whichexercise could improve cardiovascularhealth were reviewed recently by Stewart(137). These include decreased systemicinflammation, improved early diastolicfilling (reduced diastolic dysfunction),improved endothelial vasodilator func-tion, and decreased abdominal visceral fataccumulation.

Frequency of exerciseThe U.S. Surgeon General’s report recom-mended that most people accumulate�30 min of moderate intensity activity onmost, ideally all, days of the week. How-ever, most clinical trials evaluating exer-cise interventions in people with type 2diabetes have used a three times per weekfrequency (119), and many people find iteasier to schedule fewer longer sessionsrather than five or more weekly shortersessions. The effect on insulin sensitivityof a single bout of aerobic exercise lasts24–72 h, depending on the duration andintensity of the activity (138). Because theduration of increased insulin sensitivity isgenerally not �72 h, we recommend thatthe time between successive sessions ofphysical activity be no more than 72 h(i.e., there should not be more than 2 con-

secutive days without aerobic physical ac-tivity). There is some evidence that theeffect of resistance exercise training on in-sulin sensitivity is somewhat longer(139), perhaps because some of its effectsare mediated by increases in muscle mass.

Exercise for weight loss and weightmaintenanceThe most successful programs for long-term weight control have involved com-binations of diet, exercise, and behaviormodification (140). University-basedobesity research programs using combi-nations of diet, exercise, and behaviormodification have typically producedweight losses of 9–13.6 kg after 20 weeks,and �60% of this weight loss is main-tained over 1 year of follow-up (140). Ex-ercise alone, without concomitant dietarycaloric restriction and behavior modifica-tion, tends to produce only modestweight loss of �2 kg. Weight loss is typ-ically this small primarily because obesepeople often have difficulty performingsufficient exercise to create a large energydeficit, and it is relatively easy to counter-balance increased energy expenditurethrough exercise by eating more or be-coming less active outside of exercise ses-sions (140). However, in a randomizedtrial, high-volume aerobic exercise (700kcal/day, about 1 h/day of moderate-intensity aerobic exercise) produced atleast as much fat loss as the equivalentdegree of caloric restriction (141). Fur-thermore, exercise-induced weight lossresulted in greater improvements in insu-lin sensitivity than diet-induced weightloss (141).

The optimal volume of exercise toachieve sustained major weight loss isprobably much larger than that needed toachieve improved glycemic control andcardiovascular health. In the NationalWeight Control Registry (142), a study ofindividuals who lost at least 13.6 kg(mean 30 kg) and maintained the weightloss for at least 1 year (mean 5 years), theaverage self-reported energy expenditureon exercise was 2,545 kcal/week amongwomen and 3,293 kcal/week among men.These amounts would correspond to �7h/week of moderate-intensity exercise.Similarly large amounts of exercise havebeen associated with long-term mainte-nance of weight loss in other studies(143–146).

Recommendations: aerobic exerciseThe amount and intensity recommendedfor aerobic exercise vary according togoals.

● To improve glycemic control, assistwith weight maintenance, and reducerisk of CVD, we recommend at least150 min/week of moderate-intensityaerobic physical activity (40–60% ofVO2max or 50–70% of maximum heartrate) and/or at least 90 min/week of vig-orous aerobic exercise (�60% ofVO2max or �70% of maximum heartrate). The physical activity should bedistributed over at least 3 days/weekand with no more than 2 consecutivedays without physical activity.

● Performing �4 h/week of moderate tovigorous aerobic and/or resistance ex-ercise is associated with greater CVDrisk reduction compared with lowervolumes of activity (128).

● For long-term maintenance of majorweight loss (�13.6 kg [30 lb]), largervolumes of exercise (7 h/week of mod-erate or vigorous aerobic physical activ-ity per week) may be helpful.

Levels of evidence: A, for improved gly-cemic control (119,120); B, for CVD pre-vention (122,128); and B, for long-termmaintenance of major weight loss (142–146).

RESISTANCE EXERCISEFor more detailed reviews on this topic,see Refs. 148 and 149.

The proven value of aerobic exercisenotwithstanding, it does have some limi-tations. Some find aerobic exercise mo-notonous. Most forms of aerobic exercisewould not be advisable with advanced pe-ripheral neuropathy and are challengingin people with severe obesity. Resistanceexercise training, by increasing musclemass and endurance, often causes morerapid changes in functional status andbody composition than aerobic trainingand might therefore be more immediatelyrewarding. Because each session involvesmany different resistance exercises, somefind it less monotonous than aerobic ex-ercise. Resistance exercise improves insu-lin sensitivity to about the same extent asaerobic exercise (150).

Because of the increased evidence forhealth benefits from resistance trainingduring the past 10–15 years, the Ameri-can College of Sports Medicine (ACSM)



Table

2—T

hepositions

ofother

major

professionalassociationson

therecom

mended

typesand

amounts

ofphysicalactivity

Reference

Aerobic

trainingR

esistancetraining

FrequencyIntensity

Duration/m

odeFrequency

No.ofexercises

Sets/repetitions

Health

yad

ults

1995C

DC

/AC

SMPublic

Health

Statement

(126)

Daily

Moderate

Accum

ulate30

min/day

continuousphysical

activity

—A

ddressedbut

notspecified

—

1996U

.S.SurgeonG

eneral’sR

eport(2)

Near

dailyM

oderate/intenseA

ccumulate

�30

min

2days/w

eek8–10

exercisesinvolving

major

muscle

groups1–2

sets,8–12reps

1998H

ealthC

anada(186)

Daily

Light/moderate

Accum

ulate60

min/day

with

lightphysical

activity,candecrease

to30

min

with

moderate

intensity

2–4days/

week

Involveallm

ajorm

usclegroups

2–4sets,10–15

reps

2000A

CSM

PositionStand

(127)3–5

days/week

65–90%H

Rm

axor

50–85%V

o2

max

orH

Rm

axreserve;

startat

lower

intensityif

initiallyunfit

orsedentary

20–60m

incontinuous

aerobicactivities;20–30

min

minim

um

2days/w

eek8–10

exercisesinvolving

major

muscle

groups1

set,8–12reps

Eld

erlyp

eople

(aged>

65years)

1998A

CSM

PositionStand

(151)3–5

days/week

50–85%V

o2

max

or40–80%

HR

max

reserve30–60

min

continuousaerobic

activity2

days/week

8–10exercises

involvingm

ajorm

usclegroups

1set,10–15

reps

Card

iacp

atients

1995A

HA

Exercise

Standards(187)

Minim

umof3

days/week

50–60%V

o2

max

orH

Rm

axreserve

Minim

umof30

min

continuousphysical

activity

2–3days/

week

8–10exercises

involvingm

ajorm

usclegroups

1set,10–15

reps

1999A

AC

VPR

(188)3–5

days/week

50–60%V

o2

max

orH

Rm

axreserve

30–45m

inofcontinuous

orinterm

ittentaerobic

activity

2–3days/

week

Incorporateallm

ajorm

usclegroups

1set,10–15

reps

Ind

ividu

alsw

ithtyp

e2

diabetes

2003C

anadianD

iabetesA

ssociation(189)*

At

least3

nonconsecutivedays/w

eekM

oderateor

intenseA

ccumulate

atleast

150m

in/week

(�h/w

eek,ifw

illing)

3days/w

eek�

8exercises

involvingm

ajorm

usclegroups

Initially1

set,10–15reps

Progressto

3sets,8–

12reps

2000A

CSM

PositionStand

(127)A

tleast

3nonconsecutive

days/week

40–70%V

o2

max

10–15m

incontinuous

physicalactivityin

earlystages,graduallyincreased

to30

min.

Sessionscan

bedivided

intothree

10-min

sessions

At

least2

days/week

Minim

umof8–10

exercisesinvolving

major

muscle

groups

Minim

umof1

set,10–15

reps

PreviousA

DA

guidelines(190)

3–5days/w

eek55–79%

ofHR

max

or40–74%

HR

max

reserve20–60

min

continuousor

intermittent

aerobicactivity

(min

of10-min

bouts)

High

resistanceexercise

usingw

eightsm

aybe

acceptablefor

youngindividuals

orthose

with

longstandingdiabetes.M

oderatew

eighttraining

programs

thatutilize

lightw

eightsand

highrepetitions

canbe

usedfor

maintaining

orenhancing

upperbody

strengthin

nearlyall

patientsw

ithdiabetes

*The

2003C

anadianD

iabetesA

ssociationC

linicalPracticeG

uidelines,publishedin

Decem

ber2003,w

eredrafted

bytw

oofthe

authorsofthe

presenttechnicalreviewand

reviewed

bythe

Canadian

Diabetes

Association

Expert

Com

mittee.T

heC

anadianD

iabetesA

ssociationguidelines

arebased

onsom

eofthe

same

authorsexam

iningthe

same

updated-evidencereview

edin

thepresent

work,and

thereforethe

recomm

endationsare

similar

tothe

newproposed

AD

Arecom

mendations.A

HA

,Am

ericanH

eartA

ssociation;AA

CV

PR,A

merican

Association

ofC

ardiovascularand

Pulmonary

Rehabilitation;C

DC

,U.S.

Centers

forD

iseaseC

ontrolandPrevention;H

R,heart

rate.



now recommends resistance training beincluded in fitness programs for healthyyoung and middle-aged adults (125),older adults (151), and adults with type 2diabetes (127). With increased age, thereis a tendency to progressive declines inmuscle mass, leading to “sarcopenia,” de-creased functional capacity, decreasedresting metabolic rate, increased adipos-ity, and increased insulin resistance, andresistance training can have a major pos-itive impact on each of these (151).

Studies of resistance exercise in type2 diabetesNotwithstanding the above, the previousADA Position Statement on physical ac-tivity and exercise took a conservativeview on resistance exercise, stating thathigh-intensity resistance training withweights might be appropriate for youngerdiabetic patients but not for older patientsor those with longstanding disease; in-stead lower-intensity resistance exercisewith light weights was endorsed (93).This was due to a lack of published stud-ies on high-intensity resistance exercise inolder diabetic individuals and concernsabout the safety of this type of exercise inthe absence of published data (Table 1).

Before 1997 there were no publishedstudies of resistance exercise in type 2 di-abetic subjects. The first such publishedexperiment was by Eriksson et al. (152),who studied eight moderately obese type2 diabetic patients aged 55 � 9 years(�SD) before and after a 3-month pro-gram of moderate-intensity weight train-ing. Muscle endurance increased by 32%.HbA1c decreased from 8.8 to 8.2% (P �0.05), and there was a strong negativecorrelation between HbA1c and musclecross-sectional area (r � �0.73). Therewas no control group. Ishii et al. (153)studied nine nonobese middle-aged type2 diabetic subjects before and after 4–6weeks of high-volume, moderate-intensity weight training. They were com-pared with control subjects unable toexercise because of orthopedic disorders.Insulin sensitivity rose 48% in exercisersbut remained unchanged in control sub-jects. HbA1c declined from 9.6 to 7.6% inthe weight training group, but also inex-plicably declined from 8.8 to 7.6% in thesedentary subjects. In a nonrandomizedtrial (154), 18 subjects with type 2 diabe-tes (12 men and 6 women; mean age 62years) underwent 5 months of moderate-intensity resistance training and were

compared with 5 men and 15 women(mean age 67 years) with type 2 diabeteswho did not exercise during this time.HbA1c in the exercise group was 7.5% atbaseline and 7.4% at 20 weeks, whereasHbA1c in control subjects increased from7.7 to 8.1% (P � 0.05 between groups).Interpretation of this study is complicatedby a lack of randomization and imbal-ances at baseline in age and sex betweenthe exercisers and control subjects. Thefirst randomized controlled trial evaluat-ing resistance training on glycemic con-trol in type 2 diabetic patients was doneby Dunstan et al. (155) in which 27 type 2diabetic patients were randomized tononexercise control or 8 weeks of circuittraining in which subjects alternated be-tween 30 s at a time of moderate-intensityweight lifting and 30 s at a time of lightstationary cycling following each 30 s ofweight lifting (155). In the exercising sub-jects, both the insulin and glucose areasunder the oral glucose tolerance test curvedecreased nonsignificantly, and there wasno significant effect on HbA1c. In a similarstudy, Maiorana et al. (156) randomized16 subjects in a crossover design to non-exercise control, followed by 8 weeks ofthree times per week circuit training orvice versa. During each circuit trainingsession, subjects alternated 45 s of aerobicexercise at a moderate-intensity station-ary cycling station with 45 s of moderate-intensity weight lifting. Mean HbA1c was8.5% following sedentary periods and7.9% following exercise periods. Thisstudy and the 1998 Dunstan et al. study(155) shared two limitations. First, dura-tion of the intervention was insufficient tosignificantly affect body compositionthrough resistance training because 3–6months of training are required for clini-cally significant muscle hypertrophy(157). Second, the mixed resistance andaerobic training design precluded distin-guishing the independent effects of eachmodality.

In recent trial, Cuff et al. (158) ran-domized 28 well-controlled, obese, post-menopausal type 2 diabetic women tocombined aerobic and resistance training,aerobic training alone, or a nonexercisingcontrol group. Subjects in the exercisinggroups participated in three 75-min gymsessions per week for 16 weeks. The aer-obic exercise was at 60–75% of heart ratereserve, whereas the resistance trainingprogram included two sets of 12 repeti-tions of five exercises. The aerobic-only

group spent additional time on very-low-intensity warm-up and cool-down activ-ity that was not expected to affect glucosemetabolism. HbA1c was excellent in allgroups before training (6.3–6.9%) anddid not change with exercise training.However, insulin sensitivity assessed withglucose clamp was increased significantlymore in the combined aerobic and resis-tance exercise group than in the aerobicexercise only or control groups. Body fatdeclined significantly and similarly inboth exercise groups, but muscle mass in-creased significantly only in the com-bined aerobic and resistance exercisegroup.

Two clinical trials published in late2002 (159,160) provided much strongerevidence for the value of resistance train-ing in type 2 diabetes. Dunstan et al. (159)randomized 36 Australian sedentary,overweight, type 2 diabetic subjects aged60–80 years to 6 months of moderateweight loss plus high-intensity resistancetraining (RT/WL group; progressing tothree sets of 8–10 repetitions of 8–10 ex-ercises three times per week at 75–80% ofmaximum) or moderate weight loss plusflexibility exercise (control/WL group).Absolute HbA1c declined 1.2% in theRT/WL group compared with just 0.4% inthe control/WL group (P � 0.05 betweengroups). Mean weight loss and fat losswere similar in both groups, but meanlean body mass increased by 0.5 kg inRT/WL subjects while decreasing 0.4 kgin control/WL subjects (P � 0.05 be-tween groups). Castaneda et al. (160) ran-domized 62 older sedentary Hispanicadults (40 women and 22 men; mean age66 years) to 16 weeks of individually su-pervised high-intensity resistance exer-cise (RT group, progressing to three setsof eight repetitions of five exercises threetimes per week at 70–80% of maximum)or sedentary control. Mean HbA1c de-clined from 8.7 to 7.6% in RT but did notchange in control subjects (P � 0.01 be-tween groups), even though 72% of RTsubjects (versus 3% of control subjects)had hypoglycemic medications reducedand 42% of control subjects (versus 7% ofRT subjects) had hypoglycemic medica-tions increased. Mean systolic blood pres-sure declined 9.7 mmHg in RT subjectsand rose 7.7 mmHg in control subjects(P � 0.05 between groups). Free fattyacid concentrations declined significantlyby 27% in the RT group compared withcontrol subjects, in whom circulating free



fatty acids increased by 10% (161). Therewas a significant positive correlation be-tween the changes in glycosylated hemo-globin and plasma free fatty acidconcentrations in the groups combined.The interventions in these two studies in-volved higher exercise intensity (70 –85% of one repetition maximum versus40 – 60% of one repetition maximum)and more sets of each exercise (three setsvs. one to two sets) than the other studiesdescribed above. Both studies enrolledonly older subjects, with mean age of �66years in both.

Resistance exercise improves bonedensity, muscle mass, strength, balance,and overall capacity for physical activityand therefore is potentially important forprevention of osteoporotic fractures inthe elderly (162,163).

The ACSM recommends a resistancetraining regimen for type 2 diabetic indi-viduals whenever possible. It recom-mends “a minimum of 8 –10 exercisesinvolving the major muscle groups. . .with a minimum of one set of 10–15 rep-etitions to near fatigue. Increased inten-sity of exercise, additional sets, orcombinations of volume and intensitymay produce greater benefits and may beappropriate for certain individuals.”These recommendations were publishedin 2000, before the 2002 Dunstan et al.(159) and Castaneda et al. (160) resultswere known. Given the superiority ofCastaneda et al. and Dunstan et al.’s re-sults in programs requiring three sets ofeach exercise compared with the other tri-als evaluating programs requiring just oneto two sets of each exercise, we advocate aresistance program similar to theirs: pro-gressing to three sets of 8–10 repetitionsof the heaviest weight that can be lifted8–10 times to near fatigue. Although oneset of each exercise may be sufficient toincrease muscle strength (164), it appearsthat three sets of each exercise producethe greater metabolic benefit in type 2diabetes.

A conservative approach is to beginwith one set of 10–15 repetitions two tothree times per week at moderate inten-sity for several weeks, then two sets of10–15 repetitions two to three times perweek for several weeks, and then progressto three sets of 8 –10 repetitions at aweight that cannot be lifted more than8–10 times (8–10 RM). In the studies byDunstan et al. (159) and Castaneda et al.(160), intensity of resistance exercise was

increased more rapidly than this. Eachworkout should be preceded by 5 min ofwarm up and followed by 5 min of cooldown, each consisting of light aerobic ac-tivity with or without flexibility exercises.Initial supervision and periodic reassess-ment by a qualified exercise specialist isrecommended to optimize benefits whileminimizing risk of injury; such supervi-sion was included in all of the above pub-lished studies.

Safety of resistance trainingSome medical practitioners have con-cerns about the safety of higher-intensityresistance exercise in middle-aged andolder people who are at risk of CVD. Themain concern is often that the acute risesin blood pressure associated with higher-intensity resistance exercise might beharmful, possibly provoking stroke, myo-cardial ischemia, or retinal hemorrhage.We have found no evidence that resis-tance training actually increases theserisks. No serious adverse events havebeen reported in any research study ofresistance training in patients with type 2diabetes, although the total number ofsubjects enrolled in these studies (152–155,159,160) was small. A review of 12resistance exercise studies in a total of 246male cardiac rehabilitation patients (166)found no angina, ST depression, abnor-mal hemodynamics, ventricular dys-rhythmias, or other cardiovascularcomplications. A study of 12 men withknown coronary ischemia and ECGchanges inducible by moderate aerobicexercise found that even maximal-intensity resistance exercise did not in-duce ECG changes (167). Therefore,moderate- to high-intensity resistancetraining was found to be safe even in menat significant risk of cardiac events. Lom-bardi et al. (168) analyzed data from theNational Electronic Injury SurveillanceSystem (NEISS, a nationwide representa-tive sample of hospital emergency depart-ments), the National Death Index, theU.S. Consumer Products Safety Commis-sion, and two other databases in an effortto determine the morbidity and mortalityassociated with resistance training in theU.S. Between 1999 and 2001 there were20 deaths in the U.S. associated withweight lifting. All of them were related toweights or barbells falling and causing in-jury. Fourteen of the deaths were due toasphyxia, primarily from dropping a bar-bell on the chest and neck. Most of these

occurred in subjects’ homes. There wereno cases of death from myocardial infarc-tion or stroke associated with resistanceexercise.

The reason why resistance exerciseappears less likely to induce ischemiathan aerobic exercise has not been clearlydemonstrated. A number of reasons seemplausible. First, in resistance exercise atleast as much time is spent resting be-tween sets as is spent lifting, lifting gener-ally does not last �60 s at a time. Incontrast, with aerobic exercise, there isgenerally no rest during the exercise ses-sion. Second, during resistance exercise,systolic and diastolic blood pressure risein parallel, possibly helping to maintaincoronary perfusion, whereas in aerobicexercise systolic pressure rises signifi-cantly more than diastolic pressure (169).Third, the rise in cardiac output withhigh-intensity resistance exercise is signif-icantly less than that associated with high-intensity aerobic exercise (170).

Although it is well known that bloodpressure rises while lifting a heavy weight,it is often not appreciated that blood pres-sure can also rise considerably in healthyolder people performing aerobic exercise.Benn et al. (171) studied the responses toaerobic and resistance exercise in 17healthy men aged 64 � 6 years. Subjectsperformed each of the following exerciseswith continuous monitoring of heart rateand intra-arterial blood pressure: one-arm military press, one-arm curl at 70%of 1-RM (moderate intensity), single- anddouble-leg press at 80% of 1-RM (highintensity), horizontal walking for 20 minat 2.5 mph carrying 20 lbs in minutes4–6 then 30 lbs in minutes 8–10, 4-mintreadmill walk at 3 mph up an 8% incline,and 192 steps on a Stairmaster in �3 min.The highest peak systolic blood pressuresoccurred in stair climbing (271 mmHg)and military press (261 mmHg), whereasthe lowest peaks were in the single-armcurl (224 mmHg) and carrying a 20-lbweight (220 mmHg). Peak diastolic pres-sures were 128–151 mmHg in the resis-tance exercises, 121 mmHg in stairclimbing, and 101–118 mmHg in the var-ious treadmill exercises. Rate-pressureproduct, which is significantly correlatedwith myocardial oxygen demand, washighest at 41,000 for stair climbing; it was22,000–30,000 for resistance exercisesand 23,000–28,000 for treadmill exer-cises. Therefore, the myocardial demandsof high-intensity resistance exercise are



not out of line with those occasionallyneeded for activities of daily living, suchas climbing stairs, walking up a hill, orcarrying 20–30 lbs of groceries.

There is little or no evidence to guidepractitioners in terms of whether stresstesting before undertaking resistancetraining is necessary. One might askwhether such testing should use resis-tance exercise, rather than the usual aer-obic exercise, during a stress test in suchcircumstances. Very few test centerswould currently be equipped for suchtesting, and such tests have not been stan-dardized. In contrast, aerobic exercisestress testing is widely available, stan-dardized, and of proven prognostic value.

Recommendations: resistanceexerciseIn the absence of contraindications, peo-ple with type 2 diabetes should be en-couraged to perform resistance exercisethree times a week, including all majormuscle groups, progressing to three setsof 8–10 repetitions at a weight that can-not be lifted �8–10 times (8–10 RM).Level of evidence. A. In order to ensureresistance exercises are performed cor-rectly, maximize health benefits, andminimize the risk of injury, we recom-mend initial supervision and periodic re-assessments by a qualified exercisespecialist, as was done in the clinical trials(159,160).

FLEXIBILITY EXERCISEFlexibility exercise (stretching) has fre-quently been recommended as a means ofincreasing range of motion and hopefullyreducing risk of injury. However, two sys-tematic reviews (172,173) have foundthat flexibility exercise does not reducerisk of exercise-induced injury. It shouldbe noted that most studies included inthese systematic reviews evaluatedyounger subjects undertaking very vigor-ous activity programs, such as those inmilitary basic training; these results maynot be generalizable to older subjects.Flexibility exercise has been successfullyused in clinical trials as a “placebo” exer-cise (159,174), since there is no evidencethat flexibility exercise affects metaboliccontrol or quality of life. We found twosmall studies providing indirect supportfor flexibility exercise in reducing risk offoot ulceration. In a case-control study(175), 25 diabetic patients with a historyof neuropathic foot ulceration had higher

pressure on the plantar aspect of the footand lower ankle joint flexibility than 50control subjects without neuropathy orfoot ulceration. In a small randomizedtrial, 19 diabetic subjects were random-ized to unsupervised active and passiverange of motion exercises of the joints infeet or an inactive control group. After 1month, the nine who performed range ofmotion exercises had a 4.2% decrease inpeak plantar pressures compared with a4.4% increase in peak plantar pressures inthe control group. We found no studiesthat directly evaluated whether flexibilitytraining reduced the risk of ulceration orinjury in people with diabetes. Therefore,we feel that there is insufficient evidenceto recommend for or against flexibilityexercise as a routine part of the exerciseprescription.

EXERCISE IN THE PRESENCEOF NONOPTIMALGLYCEMIC CONTROL

HyperglycemiaWhen people with type 1 diabetes are de-prived of insulin for 12–48 h and ketotic,exercise can worsen the hyperglycemiaand ketosis (176). Previous ADA exerciseposition statements have suggested thatphysical activity be avoided if fasting glu-cose levels are �250 mg/dl and ketosis ispresent and performed with caution ifglucose levels are �300 mg/dl even if noketosis is present (93). We agree that vig-orous activity should probably be avoidedin the presence of ketosis. However, therecommendation to avoid physical activ-ity if plasma glucose is �300 mg/dl, evenin the absence of ketosis, is probably morecautious than necessary for a person withtype 2 diabetes, especially in a postprandialstate. In the absence of very severe insulindeficiency, light- or moderate-intensity ex-ercise would tend to decrease plasma glu-cose. Therefore, provided the patient feelswell and urine and/or blood ketones arenegative, it is not necessary to postpone ex-ercise based simply on hyperglycemia.

HypoglycemiaIn individuals taking insulin and/or insu-lin secretagogues, physical activity cancause hypoglycemia if medication dose orcarbohydrate consumption is not altered.This is particularly so at times when ex-ogenous insulin levels are at their peaksand if physical activity is prolonged. Hy-poglycemia would be rare in diabetic in-

dividuals who are not treated with insulinor insulin secretagogues. Previous ADAguidelines suggest that added carbohy-drate should be ingested if pre-exerciseglucose levels are �100 mg/dl (93). Weagree with this recommendation for indi-viduals on insulin and/or an insulin secre-tagogue. However, the revised guidelinesclarify that supplementary carbohydrateis generally not necessary for individualstreated only with diet, metformin, �-glu-cosidase inhibitors, and/or thiazo-lidinediones without insulin or asecretagogue.

Those who take insulin or secreta-gogues should check capillary blood glu-cose before, after, and several hours aftercompleting a session of physical activity,at least until they know their usual glyce-mic responses to such activity. For thosewho show a tendency toward hypoglyce-mia during or after exercise, several strat-egies can be used. Doses of insulin orsecretagogues can be reduced before ses-sions of physical activity, extra carbohy-drate can be consumed before or duringphysical activity, or both. For a detaileddiscussion of medication adjustments toreduce risk of hypoglycemia, see Ref. 177.

Concomitant medicationsDiabetic patients frequently take diuret-ics, �-blockers, ACE inhibitors, aspirin,and lipid-lowering agents. In most type 2diabetic individuals, medications will notinterfere with the physical activities theychoose to perform, but patients andhealth care personnel should be aware ofpotential problems to minimize their im-pact. Diuretics, especially in higher doses,can interfere with fluid and electrolytebalance. �-Blockers can blunt the adren-ergic symptoms of hypoglycemia, possi-bly increasing risk of hypoglycemiaunawareness. They can reduce maximalexercise capacity to �87% of what itwould be without �-blockade (11)through their negative inotropic andchronotropic effects. However, most peo-ple with type 2 diabetes do not choose toexercise at very high intensity, so this re-duction of maximum capacity is generallynot problematic. In people with CAD,�-blockade actually increases exercise ca-pacity by reducing coronary ischemia.ACE inhibitors may modestly increase in-sulin sensitivity, and both ACE inhibitorsand aspirin (especially in high doses [4–6g/day]) may increase risk of hypoglycemiain some individuals through unclear



mechanisms. In rare cases, statins, espe-cially in combination with fibrates, canproduce myositis. For additional discus-sion of the impact of concomitant medi-cations on physical activity, see Ref. 92.

EXERCISE IN THE PRESENCEOF SPECIFIC LONG-TERMCOMPLICATIONS OFDIABETESThere is a paucity of research on risks andbenefits of exercise in the presence of di-abetes complications. Therefore, recom-mendations in this section are basedlargely on “expert opinion.”

RetinopathyExercise and physical activity are notknown to have any adverse effects on vi-sion or the progression of nonprolifera-tive diabetic retinopathy or macularedema (178). This applies to resistancetraining as well as aerobic training. How-ever, in the presence of proliferative orsevere nonproliferative diabetic retinopa-thy, vigorous aerobic or resistance exer-cise may be contraindicated because ofthe risk of triggering vitreous hemorrhageor retinal detachment (178). We found noresearch studies that would provide guid-ance as to an appropriate time interval be-tween successful laser photocoagulationand initiation or resumption of resistanceexercise. Opthalmologists with whomone of us (C.C.-S.) consulted suggestedwaiting 3–6 months after laser photoco-agulation before initiating or resumingthis type of exercise.

Peripheral neuropathyWe are unaware of research studies as-sessing the risk of exercise-induced injuryin people with peripheral sensory neu-ropathy. Common sense, however, wouldindicate that decreased pain sensation inthe extremities would result in increasedrisk of skin breakdown and infection andof Charcot joint destruction. Therefore, inthe presence of severe peripheral neurop-athy, it may be best to encourage non–weight-bearing activit ies such asswimming, bicycling, or arm exercises(179,180).

Autonomic neuropathyAutonomic neuropathy can increase therisk of exercise-induced injury by de-creasing cardiac responsiveness to exer-cise, postural hypotension, impairedthermoregulation due to impaired skin

blood flow and sweating, impaired nightvision due to impaired papillary reaction,impaired thirst increasing risk of dehy-dration, and gastroparesis with unpre-dictable food delivery (179). Autonomicneuropathy is also strongly associatedwith CVD in people with diabetes (181).People with diabetic autonomic neuropa-thy should definitely undergo cardiac in-vestigation before beginning physicalactivity more intense than that to whichthey are accustomed. Some experts advo-cate thallium scintigraphy as the pre-ferred screening technique for CVD inthis high-risk population (179).

Microalbuminuria and nephropathyPhysical activity can acutely increase uri-nary protein excretion. The magnitude ofthis increase is in proportion to the acuteincrease in blood pressure. This findinghas led some experts to recommend (182)that people with diabetic kidney diseaseperform only light or moderate exercise,such that blood pressure during exercisewould not rise to �200 mmHg. However,there is no evidence from clinical trials orcohort studies demonstrating that vigor-ous exercise increases the rate of progres-sion of diabetic kidney disease. Severalrandomized trials in animals with diabe-tes and proteinuria showed that aerobicexercise training decreased urine proteinexcretion (183,184), possibly in part dueto improved glycemic control, bloodpressure, and insulin sensitivity. Resis-tance training also may be of benefit. Inone clinical trial (185), 26 older peoplewith renal disease treated with a low-protein diet, including 10 with diabeticnephropathy, were randomized to 12weeks of high-intensity resistance train-ing or an inactive control group. Thoseallocated to resistance training had signif-icant improvements in muscle mass, nu-tritional status, functional capacity, andeven glomerular filtration rate comparedwith control subjects. Because of these en-couraging findings, we believe there maybe no need for any specific exercise re-strictions for people with diabetic kidneydisease. However, because microalbu-minuria and proteinuria are associatedwith increased risk for CVD, it is impor-tant to perform an exercise ECG stress testbefore beginning exercise significantlymore intense than the demands of every-day living.

COMPARISON BETWEENTHE PROPOSED NEW ADAGUIDELINES AND THE2000 ACSM POSITIONSTAND ON TYPE 2DIABETES AND EXERCISEThere is substantial agreement betweenour recommendations and those of the2000 ACSM Position Stand. Our positionon the recommended frequency, dura-tion, and intensity of aerobic exercise issimilar to the ACSM recommendations.The ACSM document was written beforethe publication of most resistance exer-cise trials in type 2 diabetes, but neverthe-less endorsed resistance training. Ournew recommendations for three sets of8–10 repetitions of a range of resistanceexercises are based on trials published in2002 (159,160) in which results were su-perior with this type of regimen com-pared with other trials evaluating less-intense regimens. The ACSM PositionStand is similar to ours in firmly recom-mending aerobic and resistance exercise,but not explicitly recommending for oragainst flexibility exercise. See Table 2 fordescriptions of the positions of other ma-jor professional associations on the rec-ommended types and amounts ofphysical activity.

Our position on which individualsshould undergo stress testing is based ona re-evaluation of the evidence rather thanon new evidence. The ACSM PositionStand, similarly to the previous ADA po-sition, advocates stress testing for all dia-betic individuals aged �35 years beforeparticipating in most physical activity.Our position advocates stress testing forthose age �30 years with additional riskfactors, and age �40 years regardless ofadditional risk factors, who wish to un-dertake vigorous exercise. We see stresstesting as less essential for detection ofischemia in younger individuals becauseof their relatively low absolute risks ofCVD. However, we continue to recognizethe potential value of maximal exercisetesting for the purposes of setting appro-priate exercise intensities and assessingphysical capacity and prognosis.

References1. American Diabetes Association: Intro-

duction. Diabetes Care 27 (Suppl. 1):S1–S2, 2004

2. U.S. Department of Health and HumanServices: Physical Activity and Health: A



Report of the Surgeon General. Atlanta,GA, U.S. Department of Health and Hu-man Services, Centers for Disease Con-trol and Prevention, National Center forChronic Disease Prevention and HealthPromotion, 1996

3. Pollock ML, Foster C, Schmidt D, Hell-man C, Linnerud AC, Ward A: Compar-ative analysis of physiologic responses tothree different maximal graded exercisetest protocols in healthy women. AmHeart J 103:363–373, 1982

4. Wasserman DH, Davis SN, Zinman B:Fuel metabolism during exercise inhealth and diabetes. In Handbook of Ex-ercise in Diabetes. Ruderman N, DevlinJT, Schneider SH, Eds. Alexandria, VA,American Diabetes Association, 2002, p.63–99

5. Ivy JL: Exercise physiology and adapta-tions to training. In Handbook of Exercisein Diabetes. 2nd ed. Ruderman N, DevlinJT, Schneider SH, Kriska A, Eds. Alexan-dria, VA, American Diabetes Associa-tion, 2002, p. 23–62

6. Goodyear LJ, Horton ES: Signal trans-duction and glucose transport in muscle.In Handbook of Exercise in Diabetes. 2nded. Ruderman N, Devlin JT, SchneiderSH, Kriska A, Eds. Alexandria, VA,American Diabetes Association, 2002, p.101–124

7. Wasserman DH, Cherrington AD: Regu-lation of extramuscular fuel sources dur-ing exercise. In Handbook of Physiology.Rowell LB, Shepherd JT, Eds. Columbia,MD, Bermedica Production, 1996

8. Koyama Y, Coker RH, Denny JC, LacyDB, Jabbour K, Williams PE, WassermanDH: Role of carotid bodies in the neu-roendocrine response to exercise. Am JPhysiol 281:E742–E748, 2001

9. Krishna MG, Coker RH, Lacy DB, ZinkerBA, Halseth AE, Wasserman DH: Gluca-gon response to exercise is critical foraccelerated hepatic glutamine metabo-lism and nitrogen disposal. Am J Physiol279:E638–E645, 2000

10. Marliss EB, Sigal RJ, Manzon A, HalterJB, Vranic M: Glucoregulation in intenseexercise, and its implications for personswith diabetes mellitus. In Proceedings ofthe IDF Satellite Symposium on Exercise,Fuel Homeostasis and Diabetes Mellitus,Nara, Japan, 12–13 November 1994. Lon-don, Smith-Gordon, 1995

11. Sigal RJ, Purdon C, Bilinski D, Vranic M,Halter JB, Marliss EB: Glucoregulationduring and after intense exercise: effectsof beta-blockade. J Clin Endocrinol Metab78:359–366, 1994

12. Sigal RJ, Fisher S, Halter JB, Vranic M,Marliss EB: The roles of catecholaminesin glucoregulation in intense exercise asdefined by the islet cell clamp technique.Diabetes 45:148–156, 1996

13. Kreisman SH, Mew NA, Halter JB,Vranic M, Marliss EB: Norepinephrineinfusion during moderate-intensity ex-ercise increases glucose production anduptake. J Clin Endocrinol Metab 86:2118–2124, 2001

14. Kreisman SH, Mew NA, Arsenault M,Nessim SJ, Halter JB, Vranic M, MarlissEB: Epinephrine infusion during moder-ate intensity exercise increases glucoseproduction and uptake. Am J Physiol278:E949–E957, 2000

15. Kreisman SH, Halter JB, Vranic M, MarlissEB: Combined infusion of epinephrineand norepinephrine during moderate ex-ercise reproduces the glucoregulatory re-sponse of intense exercise. Diabetes52:1347–1354, 2003

16. Sigal RJ, Fisher SJ, Halter JB, Vranic M,Marliss EB: Glucoregulation during andafter intense exercise: effects of beta-ad-renergic blockade in subjects with type 1diabetes mellitus. J Clin Endocrinol Metab84:3961–3971, 1999

17. Sigal RJ, Fisher SJ, Manzon A, Morais JA,Halter JB, Vranic M, Marliss EB: Glu-coregulation during and after intense exer-cise: effects of alpha-adrenergic blockade.Metabolism 49:386–394, 2000

18. Coker RH, Krishna MG, Lacy DB, AllenEJ, Wasserman DH: Sympathetic driveto liver and nonhepatic splanchnic tissueduring heavy exercise. J Appl Physiol 82:1244–1249, 1997

19. Sigal RJ, Purdon C, Fisher SJ, Halter JB,Vranic M, Marliss EB: Hyperinsulinemiaprevents prolonged hyperglycemia afterintense exercise in insulin-dependentdiabetic subjects. J Clin Endocrinol Metab79:1049–1057, 1994

20. Purdon C, Brusson M, Nyreen SL, MilesPDG, Halter J, Vranic M, Marliss EB: Theroles of insulin and catecholamines inthe glucoregulatory response during in-tense exercise and early recovery in in-sulin-dependent diabetic and controlsubjects. J Clin Endocrinol Metab 76:566–573, 1993

21. Minuk HL, Vranic M, Hanna AK, Albis-ser AM, Zinman B: Glucoregulatory andmetabolic response to exercise in obesenoninsulin-dependent diabetes. Am JPhysiol 240:E458–E464, 1981

22. Larsen JJ, Dela F, Kjaer M, Galbo H: Theeffect of moderate exercise on postpran-dial glucose homeostasis in NIDDM pa-tients. Diabetologia 40:447–453, 1997

23. Wolfe RR, Klein S, Carraro F, Weber JM:Role of triglyceride-fatty acid cycle incontrolling fat metabolism in humansduring and after exercise. Am J Physiol258:E382–E389, 1990

24. Harant I, Marion-Latard F, Crampes F,deGlisezinski I, Berlan M, Stich V,Riviere D: Effect of a long-duration phys-ical exercise on fat cell lypolytic respon-

siveness to adrenergic agents and insulinin obese men. Int J Obes 26:1373–1378,2002

25. Sacchetti M, Saltin B, Osada T, Hall Gv:Intramuscular fatty acid metabolism incontracting and non-contracting humanskeletal muscle. J Physiol 540:387–395,2002

26. Horowitz JF, Klein S: Lipid metabolismduring endurance exercise. Am J ClinNutr 72 (Suppl.):558S–563S, 2000

27. Blaak EE, van Aggel-Leijssen DP, Wagen-makers AJ, Saris WH, van Baak MA: Im-paired oxidation of plasma-derived fattyacids in type 2 diabetic subjects duringmoderate-intensity exercise. Diabetes49:2102–2107, 2000

28. Borghouts LB, Wagenmakers AJ, GoyensPL, Keizer HA: Substrate utilization innon-obese type II diabetic patients atrest and during exercise. Clin Sci (Lond)103:559–566, 2002

29. Howlett RA, Parolin ML, Dyck DJ, Hult-man E, Jones NL, Heigenhauser GF,Spriet LL: Regulation of skeletal muscleglycogen phosphorylase and PDH atvarying exercise power outputs. Am JPhysiol 275:R418–R425, 1998

30. Rush JW, Spriet LL: Skeletal muscle gly-cogen phosphorylase a kinetics: effectsof adenine nucleotides and caffeine.J Appl Physiol 91:2071–2278, 2001

31. Hargreaves M, Kiens B, Richter EA: Ef-fect of increased plasma free fatty acidconcentrations on muscle metabolism inexercising men. J Appl Physiol 70:194–201, 1991

32. Richter EA: Glucose utilization. In Hand-book of Physiology. Rowell LB, ShepherdJT, Eds. New York, Oxford UniversityPress, 1996, p. 912–951

33. Hespel P, Vergauwen L, VandenbergheK, Richter EA: Important role of insulinand flow in stimulating glucose uptakein contracting skeletal muscle. Diabetes44:210–215, 1995

34. Lund S, Holman GD, Schmitz O, Ped-ersen O: Contraction stimulates translo-cation of glucose transporter GLUT4 inskeletal muscle through a mechanismdistinct from that of insulin. Proc NatlAcad Sci U S A 92:5817–5821, 1995

35. Kemp BE, Mitchelhill KI, Stapleton D,Michell BJ, Chen ZP, Witters LA: Deal-ing with energy demand: the AMP-acti-vated protein kinase. Trends Biochem Sci24:22–25, 1999

36. Chen ZP, McConnell GK, Michell BJ,Snow RJ, Canny BJ, Kemp BE: AMPKsignaling in contracting human skeletalmuscle: acetyl-CoA carboxylase and NOsynthase phosphorylation. Am J Physiol279:E1202–E1206, 2000

37. Stephens TJ, Chen ZP, Canny BJ, MichellBJ, Kemp BE, McConnell GK: Progres-sive increase in human skeletal muscle



AMPK�2 activity and ACC phosphory-lation during exercise. Am J Physiol 282:E688–E694, 2002

38. Kurth-Kraczek EJ, Hirshman MF, Good-year LJ, Winder WW: 5AMP-activatedprotein kinase activation causes GLUT4translocation in skeletal muscle. Diabetes48:1667–1671, 1999

39. Ai H, Ihlemann J, Hellsten Y, LauritzenHPMM, Hardie DG, Galbo H, Ploug T:Effect of fiber type and nutritional stateon AICAR- and contraction-stimulatedglucose transport in rat muscle. Am JPhysiol 282:E1291–E1300, 2001

40. Mu J, Brozinick JT, Valladares O, BucanM, Birnbaum MJ: A role of AMP-acti-vated protein kinase in contraction- andhypoxia-regulated glucose transport inskeletal muscle. Mol Cell 7:1085–1094,2001

41. Saha AK, Schwarsin AJ, Roduit R, Masse F,Kaushik V, Tornheim K, Prentki M, Rud-erman NB: Activation of malonyl-CoA de-carboxylase in rat skeletal muscle bycontraction and the AMP-activated pro-tein kinase activator 5-aminoimidazaole-4-carboxamide-1-B-D-furanoside. J BiolChem 275:24279–24283, 2000

42. Stoppani J, Hildebrandt AL, SakamotoK, Cameron-Smith D, Goodyear LJ,Neufer PD: AMP-activated protein ki-nase activates transcription of the UCP3and HKII genes in rat skeletal muscle.Am J Physiol 283:E1239–E1248, 2002

43. Zheng D, MacLean PS, Pohnert SC,Knight JB, Olson AL, Winder WW,Dohm GL: Regulation of muscle GLUT-4 transcription by AMP-activated pro-tein kinase. J Appl Physiol 91:1073–1083, 2001

44. Balon TW, Nadler JL: Nitric oxide re-lease is present from incubated skeletalmuscle preparations. J Appl Physiol 77:2519–2521, 1994

45. Bradley SJ, Kingwell BA, McConnell GK:Nitric oxide synthase inhibition reducesleg glucose uptake but not blood flowduring dynamic exercise. Diabetes 48:1815–1821, 1999

46. Fryer LG, Hajduch E, Rencurel R, Salt IP,Hundal HS, Hardie DG, Carling D: Acti-vation of glucose transport by AMP-acti-vated protein kinase via stimulation ofnitric oxide synthase. Diabetes 49:1978–1985, 2000

47. Wasserman DH, Halseth AE: An over-view of muscle glucose uptake duringexercise: sites of regulation. In Advancesin Experimental Medicine and Biology.Richter EA, Kiens B, Galbo H, Saltin B,Eds. New York, Plenum Press, 1998, p.1–16

48. Halseth A, Bracy D, Wasserman D: Over-expression of hexokinase II increases in-sulin- and exercise-stimulated muscleglucose uptake in vivo. Am J Physiol 276:

E70–E77, 199949. Cusi K, Pratipanawatr T, Koval J, Printz

R, Ardehali H, Granner DK, DefronzoRA, Mandarino LJ: Exercise increaseshexokinase II mRNA, but not activity inobesity and type 2 diabetes. Metabolism50:602–606, 2001

50. Koval JA, DeFronzo RA, O’Doherty RM,Printz RL, Ardehali H, Granner DK,Mandarino LJ: Regulation of hexokinaseII activity and expression in human mus-cle by moderate exercise. Am J Physiol274:E304–E308, 1998

51. O’Doherty RM, Bracy DP, Osawa H,Wasserman DH, Granner DK: Rat skele-tal muscle hexokinase II mRNA and ac-tivity are increased by a single bout ofacute exercise. Am J Physiol 266:E171–E178, 1994

52. Wojtaszewski JF, Nielsen JN, RichterEA: Effect of acute exercise on insulinsignaling and action in humans. J ApplPhysiol 93:384–392, 2002

53. Sakamoto K, Goodyear LJ: Intracellularsignaling in contracting skeletal muscle.J Appl Physiol 93:369–383, 2002

54. Coderre L, Kandror KV, Vallega G, PilchPF: Identification and characterizationof an exercise-sensitive pool of glucosetransporters in skeletal muscle. J BiolChem 270:27584–27588, 1995

55. Goodyear LJ, Giorgino F, Balon TW,Condorelli G, Smith RJ: Effects of con-tractile activity on tyrosine phosphopro-teins and PI 3-kinase activity in ratskeletal muscle. Am J Physiol 268:E987–E995, 1995

56. Kennedy JW, Hirshman MF, GervinoEV, Ocel JV, Forse RA, Hoenig SJ, Aron-son D, Goodyear LJ, Horton ES: Acuteexercise induces GLUT4 translocation inskeletal muscle of normal human sub-jects and subjects with type 2 diabetes.Diabetes 48:1192–1197, 1999

57. Ivy JL, Zderic TW, Fogt DL: Preventionand treatment of non-insulin-dependentdiabetes mellitus. Exerc Sport Sci Rev 27:1–35, 1999

58. Mandarino LJ, Bonadonna RC, McGuin-ness OP, Halseth AE, Wasserman DH:Regulation of muscle glucose uptake invivo. In The American Physiological Soci-ety Handbook of Physiology-EndocrinePancreas. Jefferson LS, Cherrington AD,Eds. Rockville, MD, Waverly Press,2001, p. 803–848

59. Coyle EF, Coggin AR, Hemmert MK, IvyJL: Muscle glycogen utilization duringprolonged strenuous exercise when fedcarbohydrate. J Appl Physiol 61:165–172, 1986

60. Adopo E, Peronnet F, Massicotte D, Bris-son GR, Hillaire-Marcel C: Respective ofoxidation of exogenous glucose andfructose given in the same drink duringexercise. J Appl Physiol 76:1014–1019,

199461. Pirnay F, Crielaard JM, Pallikarakis N,

Lacroix M, Mosora F, Luyckx A, Lefeb-vre PJ: Fate of exogenous glucose duringexercise of different intensities in hu-mans. J Appl Physiol 43:258–261, 1982

62. Duchman SM, Ryan AJ, Schedl HP,Summers RW, Bleiler TL, Gisolfi CV:Upper limit for intestinal absorption of adilute glucose solution in men at rest.Med Sci Sports Exerc 29:482–488, 1997

63. Howlett KF, Sakamoto K, Hirshman MF,Aschenbach WG, Dow M, White MF,Goodyear LJ: Insulin signaling after ex-ercise in insulin receptor substrate-2–deficient mice. Diabetes 51:479–483,2002

64. Sherwood DJ, Dufresne SD, Markus JF,Cheatham B, Moller DE, Aronson D,Goodyear LJ: Differential regulation ofMAP kinase, p709(s6k) and akt by con-traction and insulin in rat skeletal mus-cles. Am J Physiol 276:E870–E878, 1999

65. Wojtaszewski JF, Hansen BF, Urso B,Richter EA: Wortmannin inhibits bothinsulin-stimulated and contraction-stimulated glucose uptake and transportin rat skeletal muscle. J Appl Physiol 81:1501–1509, 1996

66. Wojtaszewski JF, Higaki Y, HirshmanMF, Michael MD, Dufresne SD, KahnCR, Goodyear LJ: Exercise modulatespostreceptor insulin signaling and glu-cose transport in muscle-insulin recep-tor knockout mice. J Clin Invest104:1257–1264, 1999

67. Ren JM, Semenkovich CF, Gulve EA,Gao JG, Holloszy JO: Exercise inducesrapid increases in GLUT4 expression,glucose transport capacity, and insulin-stimulated glycogen storage in muscle.J Biol Chem 269:14396–14401, 1994

68. Kraniou Y, Cameron-Smith D, Misso M,Collier G, Hargreaves M: Effects of exer-cise on GLUT-4 and glycogenin gene ex-pression in human skeletal muscle.J Appl Physiol 88:794–796, 2000

69. Iglesias MA, Ye JM, Frangioudakis G,Saha AK, Tomas E, Ruderman NB,Cooney GJ, Kraegen EW: AICAR admin-istration causes an apparent enhance-ment of muscle and liver insulin actionin insulin-resistant high-fat-fed rats. Di-abetes 51:2886–2894, 2002

70. Nader G, Nader KA: Intracellular signal-ing specificity in skeletal muscle in re-sponse to different modes of exercise.J Appl Physiol 90:1936–1942, 2001

71. Sakamoto K, Hirshman MF, AschenbachWG, Goodyear LJ: Contraction regula-tion of Akt in rat skeletal muscle. J BiolChem 277:11910–11917, 2002

72. Osman AA, Pendergrass M, Koval J, Ma-ezono K, Cusi K, Pratipanawatr T, Man-darino LJ: Regulation of MAP kinasepathway activity in vivo in human skel-



etal muscle. Am J Physiol 278:E992–E999, 2000

73. Markuns JF, Wojtaszewski JFP, Good-year LJ: Insulin and exercise decreaseglycogen synthase kinase-3 activity bydifferent mechanisms in rat skeletalmuscle. J Biol Chem 274:24896–24900,1999

74. Galassetti P, Coker RH, Lacy DB, Cher-rington AD, Wasserman DH: Prior exer-cise increases net hepatic glucose uptakeduring a glucose load. Am J Physiol 276:E1022–E1029, 1999

75. Casey A, Mann R, Banister K, Fox J, Mor-ris PG, MacDonald IA, Greenhaff PL: Ef-fect of carbohydrate ingestion on glycogenresynthesis in human liver and skeletalmuscle, measured by 13C MRS. Am JPhysiol 278:E65–E75, 2000

76. Pencek RR, James FD, Lacy DB, JabbourK, Williams PE, Fueger PT, WassermanDH: Interaction of insulin and prior ex-ercise in control of hepatic metabolismof a glucose load. Diabetes 52:1897–1903, 2003

77. Hamilton KS, Gibbons FK, Bracy DP,Lacy DB, Cherrington AD, WassermanDH: Effect of prior exercise on the parti-tioning of an intestinal glucose load be-tween splanchnic bed and skeletalmuscle. J Clin Invest 98:125–135, 1996

78. Koranyi LI, Bourey RE, Slentz CA, Hol-loszy JO: Coordinate reduction of ratpancreatic islet glucokinase and proin-sulin mRNA by exercise training. Diabe-tes 40:401–404, 1991

79. Lee JS, Bruce CR, Tunstall RJ, Cameron-Smith D, Hugel H, Hawley JA: Interac-tion of exercise and diet on GLUT-4protein and gene expression in type Iand type II rat skeletal muscle. ActaPhysiol Scand 175:37–44, 2002

80. Terada S, Yokozeki T, Kawanaka K,Ogawa K, Higuchi M, Ezaki O, Tabata I:Effects of high-intensity swimmingtraining on GLUT-4 and glucose trans-port activity in rat skeletal muscle. J ApplPhysiol 90:2019–2024, 2001

81. Kirwin JP, Aguila LFD, Hernandez JM,Williamson DL, O’Gorman DJ, Lewis R,Krishnan RK: Regular exercise enhancesactivation of IRS-1-associated PI3-ki-nase in human skeletal muscle. J ApplPhysiol 88:797–803, 2000

82. Houmard JA, Shaw CD, Hickey MS,Tanner CJ: Effect of short-term exercisetraining on insulin-stimulated PI3-ki-nase activity in human skeletal muscle.Am J Physiol 277:E1055–E1060, 1999

83. Chibalin AV, Yu M, Ryder JW, Song XM,Galuska D, Krook A, Wallberg-Hen-ricksson H, Zierath JR: Exercise-inducedchanges in expression and activity ofproteins involved in insulin signal trans-duction in skeletal muscle: differentialeffects on insulin receptor substrates 1

and 2. Proc Natl Acad Sci U S A 97:38–43, 2000

84. Osman A, Hancock J, Hunt DG, Ivy JL,Mandarino LJ: Exercise training in-creases ERK3 activity in skeletal muscleof obese Zucker rats. J Appl Physiol 90:454–460, 2001

85. Holten MK, Zacho M, Gaster M, Juel C,Wojtaszewski JF, Dela F: Strength train-ing increases insulin-mediated glucoseuptake, GLUT4 content, and insulin sig-naling in skeletal muscle in patients withtype 2 diabetes. Diabetes 53:294–305,2004

86. Romijn JA, Klein S, Coyle EF, SidossisLS, Wolfe RR: Strenuous endurancetraining increases lipolysis and triglycer-ide-fatty acid cycling at rest. J ApplPhysiol 75:108–113, 1993

87. Izawa T, Komabayashi T, Mochizuki T,Suda K, Tsuboi M: Enhanced couplingof adenylate cyclase to lipolysis in per-meabilized adipocytes from trained rats.J Appl Physiol 71:23–29, 1991

88. Schrauwen P, Aggel-Leijssen DPv, HulG, Wagenmakers AJ, Vidal H, Saris WH,Baak MAv: The effect of a 3-month low-intensity endurance training program onfat oxidation and acetyl-CoA carboxy-lase-2 expression. Diabetes 51:2220–2226, 2002

89. Tunstall RJ, Mehan KA, Wadley GD,Collier GR, Bonen A, Hargreaves M,Cameron-Smith D: Exercise training in-creases lipid metabolism gene expres-sion in human skeletal muscle. Am JPhysiol 283:E66–E72, 2002

90. Kiens B, Kristiansen S, Richter EA, Tur-cotte LP: Membrane associated FABP inhuman skeletal muscle is increased byendurance training. Biochem Biophys ResCom 231:463–465, 1997

91. Goodpaster BH, He J, Watkins S, KelleyDE: Skeletal muscle lipid content and in-sulin resistance: evidence for a paradoxin endurance-trained athletes. J Clin En-docrinol Metab 86:5755–5761, 2001

92. Ganda OP: Patients on various drugtherapies. In Handbook of Exercise in Di-abetes. 2nd ed. Ruderman N, Devlin JT,Schneider SH, Kriska A, Eds. Alexan-dria, VA, American Diabetes Associa-tion, 2002, p. 587–599

93. Zinman B, Ruderman N, Campaigne BN,Devlin JT, Schneider SH: Physical activ-ity/exercise and diabetes mellitus (Posi-tion Statement). Diabetes Care 26 (Suppl1):S73–S77, 2003

94. Stevens RJ, Kothari V, Adler AI, StrattonIM: The UKPDS Risk Engine: a model forthe risk of coronary heart disease in typeII diabetes (UKPDS 56). Clin Sci (Lond)101:671–679, 2001

95. Fowler-Brown A, Pignone M, PletcherM, Tice JA, Sutton SF, Lohr KN: Exercisetolerance testing to screen for coronary

heart disease: a systematic review for thetechnical support for the U.S. PreventiveServices Task Force. Ann Intern Med 140:W9–W24, 2004

96. US Preventive Services Task Force:Screening for coronary heart disease:recommendation statement. Ann InternMed 140:569–572, 2004

97. Borg GA: Psychophysical bases of per-ceived exertion. Med Sci Sports Exerc 14:377–381, 1982

98. Lee IM, Sesso HD, Oguma Y, Paffen-barger RS Jr: Relative intensity of physi-cal activity and risk of coronary heartdisease. Circulation 107:1110–1116,2003

99. Whaley MH, Brubaker PH, KaminskyLA, Miller CR: Validity of rating of per-ceived exertion during graded exercisetesting in apparently healthy adults andcardiac patients. J Cardiopulm Rehabil17:261–267, 1997

100. Whaley MH, Woodall T, Kaminsky LA,Emmett JD: Reliability of perceived ex-ertion during graded exercise testing inapparently healthy adults. J CardiopulmRehabil 17:37–42, 1997

101. Gibbons RJ, Balady GJ, Bricker JT, Chait-man BR, Fletcher GF, Froelicher VF,Mark DB, McCallister BD, Mooss AN,O’Reilly MG, Winters WL Jr, AntmanEM, Alpert JS, Faxon DP, Fuster V, Gre-goratos G, Hiratzka LF, Jacobs AK, Rus-sell RO, Smith SC Jr: ACC/AHA 2002guideline update for exercise testing:summary article: a report of the Ameri-can College of Cardiology/AmericanHeart Association Task Force on PracticeGuidelines (Committee to Update the1997 Exercise Testing Guidelines). Cir-culation 106:1883–1892, 2002

102. Gibbons RJ, Balady GJ, Beasley JW,Bricker JT, Duvernoy WF, FroelicherVF, Mark DB, Marwick TH, McCallisterBD, Thompson PD, Winters WL Jr,Yanowitz FG, Ritchie JL, Cheitlin MD,Eagle KA, Gardner TJ, Garson A Jr,Lewis RP, O’Rourke RA, Ryan TJ: ACC/AHA guidelines for exercise testing: ex-ecutive summary: a report of theAmerican College of Cardiology/Ameri-can Heart Association Task Force onPractice Guidelines (Committee on Ex-ercise Testing). Circulation 96:345–354,1997

103. Pan XR, Li GW, Hu YH, Wang JX, YangWY, An ZX, Hu ZX, Lin J, Xiao JZ, CaoHB, Liu PA, Jiang XG, Jiang YY, Wang JP,Zheng H, Zhang H, Bennett PH, HowardBV: Effects of diet and exercise in pre-venting NIDDM in people with impairedglucose tolerance: the Da Qing IGT andDiabetes Study. Diabetes Care 20:537–544, 1997

104. Tuomilehto J, Lindstrom J, Eriksson JG,Valle TT, Hamalainen H, Ilanne-Parikka



P, Keinanen-Kiukaanniemi S, Laakso M,Louheranta A, Rastas M, Salminen V,Uusitupa M: Prevention of type 2 diabe-tes mellitus by changes in lifestyleamong subjects with impaired glucosetolerance. N Engl J Med 344:1343–1350,2001

105. Lindstrom J, Louheranta A, Mannelin M,Rastas M, Salminen V, Eriksson J, Uusi-tupa M, Tuomilehto J: The Finnish Dia-betes Prevention Study (DPS): lifestyleintervention and 3-year results on dietand physical activity. Diabetes Care 26:3230–3236, 2003

106. Diabetes Prevention Program ResearchGroup: Reduction in the incidence oftype 2 diabetes with lifestyle interven-tion or metformin. N Engl J Med 346:393–403, 2002

107. Diabetes Prevention Program ResearchGroup: The Diabetes Prevention Pro-gram (DPP): description of lifestyle in-tervention. Diabetes Care 25:2165–2171, 2002

108. Eriksson KF, Lindgarde F: Prevention oftype 2 (non-insulin-dependent) diabetesmellitus by diet and physical exercise:the 6-year Malmo feasibility study. Dia-betologia 34:891–898, 1991

109. Eriksson KF, Lindgarde F: No excess 12-year mortality in men with impaired glu-cose tolerance who participated in theMalmo Preventive Trial with diet and ex-ercise. Diabetologia 41:1010–1016, 1998

110. Helmrich SP, Ragland DR, Leung RW,Paffenbarger RS Jr: Physical activity andreduced occurrence of non-insulin-de-pendent diabetes mellitus. N Engl J Med325:147–152, 1991

111. Manson JE, Nathan DM, Krolewski AS,Stampfer MJ, Willett WC, HennekensCH: A prospective study of exercise andincidence of diabetes among US malephysicians. JAMA 268:63–67, 1992

112. Manson JE, Rimm EB, Stampfer MJ,Colditz GA, Willett WC, Krolewski AS,Rosner B, Hennekens CH, Speizer FE:Physical activity and incidence of non-insulin-dependent diabetes mellitus inwomen. Lancet 338:774–778, 1991

113. Wei M, Gibbons LW, Mitchell TL, Kam-pert JB, Lee CD, Blair SN: The associa-tion between cardiorespiratory fitnessand impaired fasting glucose and type 2diabetes mellitus in men. Ann Intern Med130:89–96, 1999

114. Hu FB, Sigal RJ, Rich-Edwards JW,Colditz GA, Solomon CG, Willett WC,Speizer FE, Manson JE: Walking com-pared with vigorous physical activityand risk of type 2 diabetes in women: aprospective study. JAMA 282:1433–1439, 1999

115. Hu FB, Leitzmann MF, Stampfer MJ,Colditz GA, Willett WC, Rimm EB:Physical activity and television watching

in relation to risk for type 2 diabetes mel-litus in men. Arch Intern Med 161:1542–1548, 2001

116. Hu FB, Manson JE, Stampfer MJ, ColditzG, Liu S, Solomon CG, Willett WC: Diet,lifestyle, and the risk of type 2 diabetesmellitus in women. N Engl J Med 345:790–797, 2001

117. Burchfiel CM, Sharp DS, Curb JD, Rod-riguez BL, Hwang LJ, Marcus EB, YanoK: Physical activity and incidence of di-abetes: the Honolulu Heart Program.Am J Epidemiol 141:360–368, 1995

118. Lynch J, Helmrich SP, Kakha TA, KaplanGA, Cohen RD, Salonen R, Salonen JT:Moderately intense physical activitiesand high levels of cardiorespiratory fit-ness reduce risk of non-insulin-depen-dent diabetes mellitus in middle-agedmen. Arch Intern Med 156:1307–1314,1996

119. Boule NG, Haddad E, Kenny GP, WellsGA, Sigal RJ: Effects of exercise on gly-cemic control and body mass in type 2diabetes mellitus: a meta-analysis ofcontrolled clinical trials. JAMA 286:1218–1227, 2001

119a.Holloszy JO, Constable SH, Young DA:Activation of glucose transport in mus-cle by exercise. Diabetes Metab Rev1:409–423, 1986

120. Boule NG, Kenny GP, Haddad E, WellsGA, Sigal RJ: Meta-analysis of the effectof structured exercise training on cardio-respiratory fitness in type 2 diabetesmellitus. Diabetologia 46:1071–1081,2003

121. Mourier A, Gautier JF, De Kerviler E,Bigard AX, Villette JM, Garnier JP, Du-vallet A, Guezennec CY, Cathelineau G:Mobilization of visceral adipose tissuerelated to the improvement in insulinsensitivity in response to physical train-ing in NIDDM: effects of branched-chainamino acid supplements. Diabetes Care20:385–391, 1997

122. Wei M, Gibbons LW, Kampert JB,Nichaman MZ, Blair SN: Low cardiore-spiratory fitness and physical inactivityas predictors of mortality in men withtype 2 diabetes. Ann Intern Med 132:605–611, 2000

123. Church TS, Cheng YJ, Earnest CP, Bar-low CE, Gibbons LW, Priest EL, BlairSN: Exercise capacity and body compo-sition as predictors of mortality amongmen with diabetes. Diabetes Care 27:83–88, 2004

124. U.S. Department of Health and HumanServices: Physical Activity and Health: AReport of the Surgeon General. Atlanta,GA, U.S. Department of Health and Hu-man Services, Centers for Disease Con-trol and Prevention, National Center forChronic Disease Prevention and HealthPromotion, 1996

125. American College of Sports Medicine:American College of Sports MedicinePosition Stand: the recommended quan-tity and quality of exercise for develop-ing and maintaining cardiorespiratoryand muscular fitness, and flexibility inhealthy adults. Med Sci Sports Exerc 30:975–991, 1998

126. Pate RR, Pratt M, Blair SN, Haskell WL,Macera CA, Bouchard C, Buchner D, Et-tinger W, Heath GW, King AC, et al.:Physical activity and public health: a rec-ommendation from the Centers for Dis-ease Control and Prevention and theAmerican College of Sports Medicine.JAMA 273:402–407, 1995

127. Albright A, Franz M, Hornsby G, KriskaA, Marrero D, Ullrich I, Verity LS: Amer-ican College of Sports Medicine PositionStand: exercise and type 2 diabetes. MedSci Sports Exerc 32:1345–1360, 2000

128. Hu FB, Stampfer MJ, Solomon C, Liu S,Colditz GA, Speizer FE, Willett WC, Man-son JE: Physical activity and risk for car-diovascular events in diabetic women. AnnIntern Med 134:96–105, 2001

129. Myers J, Prakash M, Froelicher V, Do D,Partington S, Atwood JE: Exercise capac-ity and mortality among men referred forexercise testing. N Engl J Med 346:793–801, 2002

130. Whelton SP, Chin A, Xin X, He J: Effectof aerobic exercise on blood pressure: ameta-analysis of randomized, controlledtrials. Ann Intern Med 136:493–503,2002

131. Leon AS, Sanchez OA: Response ofblood lipids to exercise training alone orcombined with dietary intervention.Med Sci Sports Exerc 33:S502–S515,2001 [discussion in Med Sci Sports Exerc33:S528–S509, 2001]

132. Kraus WE, Houmard JA, Duscha BD,Knetzger KJ, Wharton MB, McCartneyJS, Bales CW, Henes S, Samsa GP, OtvosJD, Kulkarni KR, Slentz CA: Effects ofthe amount and intensity of exercise onplasma lipoproteins. N Engl J Med 347:1483–1492, 2002

133. Despres JP: Visceral obesity, insulin re-sistance, and dyslipidemia: contributionof endurance exercise training to thetreatment of the plurimetabolic syn-drome. Exerc Sport Sci Rev 25:271–300,1997

134. Despres JP, Lamarche B: Low-intensityendurance exercise training, plasma li-poproteins and the risk of coronaryheart disease. J Intern Med 236:7–22,1994

135. Despres JP, Lamarche B, Bouchard C,Tremblay A, Prud’homme D: Exerciseand the prevention of dyslipidemia andcoronary heart disease. Int J Obes RelatMetab Disord 19 (Suppl. 4):S45–S51,1995



136. Despres JP, Couillard C, Bergeron J,Lamarche B: Regional body fat distribu-tion, the insulin resistance-dyslipidemicsyndrome, and the risk of type 2 diabe-tes and coronary heart disease. In Hand-book of Exercise in Diabetes. 2nd ed.Ruderman N, Devlin JT, Schneider SH,Eds. Alexandria, VA, American DiabetesAssociation, 2002, p. 197–234

137. Stewart KJ: Exercise training and the car-diovascular consequences of type 2 dia-betes and hypertension: plausiblemechanisms for improving cardiovascu-lar health. JAMA 288:1622–1631, 2002

138. Wallberg-Henriksson H, Rincon J, Zie-rath JR: Exercise in the management ofnon-insulin-dependent diabetes melli-tus. Sports Med 25:25–35, 1998

139. Zachwieja JJ, Toffolo G, Cobelli C, BierDM, Yarasheski KE: Resistance exerciseand growth hormone administration inolder men: effects on insulin sensitivityand secretion during a stable-label intra-venous glucose tolerance test. Metabo-lism 45:254–260, 1996

140. Wing RR: Exercise and weight control.In Handbook of Exercise in Diabetes. 2nded. Ruderman N, Devlin JT, SchneiderSH, Kriska A, Eds. Alexandria, VA,American Diabetes Association, 2002, p.355–364

141. Ross R, Dagnone D, Jones PJ, Smith H,Paddags A, Hudson R, Janssen I: Reduc-tion in obesity and related comorbidconditions after diet-induced weightloss or exercise-induced weight loss inmen: a randomized, controlled trial. AnnIntern Med 133:92–103, 2000

142. Klem ML, Wing RR, McGuire MT,Seagle HM, Hill JO: A descriptive studyof individuals successful at long-termmaintenance of substantial weight loss.Am J Clin Nutr 66:239–246, 1997

143. Pavlou KN, Krey S, Steffee WP: Exerciseas an adjunct to weight loss and mainte-nance in moderately obese subjects. Am JClin Nutr 49:1115–1123, 1989

144. Schoeller DA, Shay K, Kushner RF: Howmuch physical activity is needed to min-imize weight gain in previously obesewomen? Am J Clin Nutr 66:551–556,1997

145. Weinsier RL, Hunter GR, Desmond RA,Byrne NM, Zuckerman PA, Darnell BE:Free-living activity energy expenditurein women successful and unsuccessful atmaintaining a normal body weight. Am JClin Nutr 75:499–504, 2002

146. Saris WH, Blair SN, van Baak MA, EatonSB, Davies PS, Di Pietro L, Fogelholm M,Rissanen A, Schoeller D, Swinburn B,Tremblay A, Westerterp KR, Wyatt H:How much physical activity is enough toprevent unhealthy weight gain? Out-come of the IASO 1st Stock Conferenceand consensus statement. Obes Rev

4:101–114, 2003147. Reference deleted.148. Willey KA, Fiatarone-Singh MA: Battling

insulin resistance in elderly obese peoplewith type 2 diabetes: bring on the heavyweights. Diabetes Care 26:1580–1588,2003

149. Hornsby WG, Chetlin RD: Resistancetraining. In Handbook of Exercise in Dia-betes. 2nd ed. Ruderman N, Devlin JT,Schneider SH, Kriska A, Eds. Alexan-dria, VA, American Diabetes Associa-tion, 2002, p. 311–319

150. Ivy JL: Role of exercise training in theprevention and treatment of insulin re-sistance and non-insulin-dependent di-abetes mellitus. Sports Med 24:321–336,1997

151. American College of Sports Medicine:American College of Sports MedicinePosition Stand: exercise and physical ac-tivity for older adults. Med Sci Sports Ex-erc 30:992–1008, 1998

152. Eriksson J, Taimela S, Eriksson K, Parvi-ainen S, Peltonen J, Kujala U: Resistancetraining in the treatment of non-insulin-dependent diabetes mellitus. Int J SportsMed 18:242–246, 1997

153. Ishii T, Yamakita T, Sato T, Tanaka S,Fujii S: Resistance training improves in-sulin sensitivity in NIDDM subjectswithout altering maximal oxygen up-take. Diabetes Care 21:1353–1355, 1998

154. Honkola A, Forsen T, Eriksson J: Resis-tance training improves the metabolicprofile in individuals with type 2 diabe-tes. Acta Diabetol 34:245–248, 1997

155. Dunstan DW, Puddey IB, Beilin LJ,Burke V, Morton AR, Stanton KG: Ef-fects of a short-term circuit weight train-ing program on glycaemic control inNIDDM. Diabetes Res Clin Pract 40:53–61, 1998

156. Maiorana A, O’Driscoll G, Goodman C,Taylor R, Green D: Combined aerobicand resistance exercise improves glyce-mic control and fitness in type 2 diabe-tes. Diabetes Res Clin Pract 56:115–123,2002

157. Baechle TR, Earle RW: Essentials ofStrength Training and Conditioning.Champaigne, IL, Human Kinetics, 2000

158. Cuff DJ, Meneilly GS, Martin A, Ignas-zewski A, Tildesley HD, Frohlich JJ: Ef-fective exercise modality to reduceinsulin resistance in women with type 2diabetes. Diabetes Care 26:2977–2982,2003

159. Dunstan DW, Daly RM, Owen N, JolleyD, de Courten M, Shaw J, Zimett P:High-intensity resistance training im-proves glycemic control in older patientswith type 2 diabetes. Diabetes Care 25:1729–1736, 2002

160. Castaneda C, Layne JE, Munoz-OriansL, Gordon PL, Walsmith J, Foldvari M,

Roubenoff R, Tucker KL, Nelson ME: Arandomized controlled trial of resistanceexercise training to improve glycemiccontrol in older adults with type 2 dia-betes. Diabetes Care 25:2335–2341,2002

161. Castaneda C, Layne JE, Gordon PL,Huyng T, Foldari M, Walsmith J, Mu-noz-Orians L, Roubenoff R, Nelson ME:Skeletal muscle hypertrophy in poorlycontrolled Hispanic older adults withtype 2 diabetes in response to progres-sive resistance exercise training, FASEB J24:A24, 2002

162. Evans WJ: Exercise and aging. In Hand-book of Exercise in Diabetes. 2nd ed. Ru-derman N, Devlin JT, Schneider SH,Kriska A, Eds. Alexandria, VA, AmericanDiabetes Association, 2002, p. 567–585

163. Nelson ME, Fiatarone MA, MorgantiCM, Trice I, Greenberg RA, Evans WJ:Effects of high-intensity strength train-ing on multiple risk factors for osteopo-rotic fractures: a randomized controlledtrial. JAMA 272:1909–1914, 1994

164. Carpinelli RN, Otto RM: Strength train-ing: single versus multiple sets. SportsMed 26:73–84, 1998

165. Reference deleted166. Wenger NK, Froelicher ES, Smith LK,

Ades PA, Berra K, Blumenthal JA, CertoCM, Dattilo AM, Davis D, DeBusk RF, etal.: Cardiac rehabilitation as secondaryprevention: Agency for Health Care Pol-icy and Research and National Heart,Lung, and Blood Institute. Clin PractGuidel Quick Ref Guide Clin 17:1–23,1995

167. Featherstone JF, Holly RG, AmsterdamEA: Physiologic responses to weight lift-ing in coronary artery disease. Am J Car-diol 71:287–292, 1993

168. Lombardi VP, Troxel RA: US deaths andinjuries associated with weight training(Abstract). Med Sci Sports Exerc 35:S203,2003

169. McCartney N: Role of resistance trainingin heart disease. Med Sci Sports Exerc 30:S396–S402, 1998

170. Gordon NF, Kohl HW 3rd, Pollock ML,Vaandrager H, Gibbons LW, Blair SN:Cardiovascular safety of maximalstrength testing in healthy adults. Am JCardiol 76:851–853, 1995

171. Benn SJ, McCartney N, McKelvie RS:Circulatory responses to weight lifting,walking, and stair climbing in oldermales. J Am Geriatr Soc 44:121–125,1996

172. Shrier I: Stretching before exercise doesnot reduce the risk of local muscle in-jury: a critical review of the clinical andbasic science literature. Clin J Sport Med9:221–227, 1999

173. Yeung EW, Yeung SS: Interventions forpreventing lower limb soft-tissue inju-



ries in runners. Cochrane Database SystRev 3:CD001256, 2001

174. Beniamini Y, Rubenstein JJ, ZaichkowskyLD, Crim MC: Effects of high-intensitystrength training on quality-of-life param-eters in cardiac rehabilitation patients.Am J Cardiol 80:841–846, 1997

175. Bennett PJ, Stocks AE, Whittam DJ:Analysis of risk factors for neuropathicfoot ulceration in diabetes mellitus. J AmPodiatr Med Assoc 86:112–116, 1996

176. Berger M, Berchtold P, Cuppers HJ,Drost H, Kley HK, Muller WA,Wiegelmann W, Zimmerman-TelschowH, Gries FA, Kruskemper HL, Zimmer-mann H: Metabolic and hormonal ef-fects of muscular exercise in juveniletype diabetics. Diabetologia 13:355–365,1977

177. Berger M: Adjustment of insulin and oralagent therapy. In Handbook of Exercise inDiabetes. 2nd ed. Ruderman N, DevlinJT, Schneider SH, Kriska A, Eds. Alexan-dria, VA, American Diabetes Associa-tion, 2002, p. 365–376

178. Aiello LP, Wong J, Cavallerano JD,Bursell S-E, Aiello LM: Retinopathy. InHandbook of Exercise in Diabetes. 2nd ed.Ruderman N, Devlin JT, Schneider SH,Kriska A, Eds. Alexandria, VA, AmericanDiabetes Association, 2002, p. 401–413

179. Vinik AI, Erbas T: Neuropathy. In Hand-book of Exercise in Diabetes. 2nd ed. Ru-derman N, Devlin JT, Schneider SH,Kriska A, Eds. Alexandria, VA, AmericanDiabetes Association, 2002, p. 463–496

180. Levin ME: The diabetic foot. In Hand-book of Exercise in Diabetes. 2nd ed. Ru-derman N, Devlin JT, Schneider SH,Kriska A, Eds. Alexandria, VA, AmericanDiabetes Association, 2002, p. 385–399

181. Valensi P, Sachs RN, Harfouche B,Lormeau B, Paries J, Cosson E, Paycha F,Leutenegger M, Attali JR: Predictivevalue of cardiac autonomic neuropathyin diabetic patients with or without si-lent myocardial ischemia. Diabetes Care24:339–343, 2001

182. Mogensen CE: Nephropathy: early. InHandbook of Exercise in Diabetes. 2nd ed.Ruderman N, Devlin JT, Schneider SH,Kriska A, Eds. Alexandria, VA, AmericanDiabetes Association, 2002, p. 433–449

183. Ward KM, Mahan JD, Sherman WM:Aerobic training and diabetic nephropa-thy in the obese Zucker rat. Ann Clin LabSci 24:266–277, 1994

184. Albright AL, Mahan JD, Ward KM, Sher-man WM, Roehrig KL, Kirby TE: Dia-betic nephropathy in an aerobicallytrained rat model of diabetes. Med SciSports Exerc 27:1270–1277, 1995

185. Castaneda C, Gordon PL, Uhlin KL,Levey AS, Kehayias JJ, Dwyer JT, Field-ing RA, Roubenoff R, Singh MF: Resis-tance training to counteract thecatabolism of a low-protein diet in pa-tients with chronic renal insufficiency: arandomized, controlled trial. Ann InternMed 135:965–976, 2001

186. Health Canada: Canada’s Physical Activ-ity Guide to Healthy Active Living. Ottawa,Government of Canada, 1998

187. Fletcher GF, Balady G, Froelicher VF,Hartley LH, Haskell WL, Pollock ML:Exercise standards: a statement forhealthcare professionals from the Amer-ican Heart Association Writing Group.Circulation 91:580–615, 1995

188. American Association of Cardiovascularand Pulmonary Rehabilitation: Guide-lines for Cardiac Rehabilitation Programs.Champaign, IL, Human Kinetics Pub-lishers, 1999

189. Canadian Diabetes Association ClinicalPractice Guidelines Expert Committee:Physical activity and diabetes. Can J Di-abetes 27:S24–S26, 2003

190. American Diabetes Association: Diabe-tes mellitus and exercise (Position State-ment). Diabetes Care 25 (Suppl. 1):S64,2002



Documents

Physical Activity/Exercise and Type 2 Diabetes