exercise in the management of non—insulin-dependent diabetes mellitus

Exercise in the Management ofNon–Insulin-Dependent Diabetes MellitusHarriet Wallberg-Henriksson, Jorge Rincon and Juleen R. Zierath

Department of Clinical Physiology, Karolinska Hospital, Karolinska Institute, Stockholm, Sweden

ContentsSummary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251. Clinical Characteristics of Patients with Non–Insulin-Dependent

Diabetes Mellitus (NIDDM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272. Pathophysiology of NIDDM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.1 β-Cell Dysfunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272.2 Hepatic Insulin Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282.3 Peripheral Insulin Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282.4 Glucose Metabolism in Skeletal Muscle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

3. Effect of Acute Exercise in Patients with NIDDM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293.1 Blood Glucose Levels and Insulin Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293.2 Hormone Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303.3 Glucose Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

4. Long Term Effect of Exercise in Patients with NIDDM . . . . . . . . . . . . . . . . . . . . . . . . . . . 314.1 Metabolic Control and Insulin Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314.2 Bodyweight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314.3 Cardiovascular Risk Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

5. Exercise in the Prevention of NIDDM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315.1 Relationship Between Physical Fitness and Insulin Sensitivity . . . . . . . . . . . . . . . . . . . 315.2 First Degree Relatives of Patients with NIDDM . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325.3 Strategies to Prevent NIDDM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

6. Exercise as a Therapeutic Tool in Clinical Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336.1 Medical Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336.2 Type of Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Summary The incidence of non–insulin-dependent diabetes mellitus (NIDDM) has in-creased worldwide during the last decades, despite the development of effectivedrug therapy and improved clinical diagnoses. NIDDM is one of the major causesof disability and death due to the complications accompanying this disease. Forthe well-being of the patient, and from a public healthcare perspective, the devel-opment of effective intervention strategies is essential in order to reduce theincidence of NIDDM and its resulting complications. For the patient, and forsociety at large, early intervention programmes are beneficial, especially from a

REVIEW ARTICLE Sports Med. 1998 Jan; 25 (1): 25-350112-1642/98/0001-0025/$05.50/0

© Adis International Limited. All rights reserved.

cost-benefit perspective. Physical activity exerts pronounced effects on substrateutilisation and insulin sensitivity, which in turn potentially lowers blood glucoseand lipid levels. Exercise training also improves many other physiological andmetabolic abnormalities that are associated with NIDDM such as lowering bodyfat, reducing blood pressure and normalising dyslipoproteinaemia. Clearly, reg-ular physical activity plays an important role in the prevention and treatment ofNIDDM.

Since physical activity has been shown in prospective studies to protect againstthe development of NIDDM, physical training programmes suitable for individ-uals at risk for NIDDM should be incorporated into the medical care system to agreater extent. One general determinant in a strategy to develop a preventiveprogramme for NIDDM is to establish a testing programme which includesV.O2max determinations for individuals who are at risk of developing NIDDM.Before initiating regular physical training for people with NIDDM, a complete

physical examination aimed at identifying any long term complications of diabe-tes is recommended. All individuals above the age of 35 years should perform anexercise stress test before engaging in an exercise programme which includesmoderate to vigorously intense exercise. The stress test will identify individualswith previously undiagnosed ischaemic heart disease and abnormal blood pres-sure responses. It is important to diagnose proliferative retinopathy, micro-albuminuria, peripheral and/or autonomic neuropathy in patients with NIDDMbefore they participate in an exercise programme. If any diabetic complicationsare present, the exercise protocol should be modified accordingly.

The exercise programme should consist of moderate intensity aerobic exer-cise. Resistance training and high intensity exercises should only be performedby individuals without proliferative retinopathy or hypertension. Once enrolledin the exercise programme, the patient must be educated with regard to properfootwear and daily foot inspections. Fluid intake is of great importance whenexercising for prolonged periods or in warm and humid environments. With theproper motivation and medical supervision, people with NIDDM can enjoy reg-ular physical exercise as a means of enhancing metabolic control and improvinginsulin sensitivity.

Despite the development of effective drug ther-apy and improved clinical diagnoses, the inci-dence of non–insulin-dependent diabetes mellitus(NIDDM) has increased world-wide during the lastdecades. Because of complications accompanyingthis disease, NIDDM is a major cause of disabilityand death.[1] Today, the therapeutic tools com-monly used to treat NIDDM include diet, anti-diabetic drug therapy and insulin treatment. Thesetherapies have been successful in keeping metabo-lic control within a range that makes it possible forthe patient to enjoy everyday life. However, the

management of patients with NIDDM has failedwith respect to the prevention of the disease. Fur-thermore, strategies to fully normalise metaboliccontrol and prevent the development of severecomplications in the cardiovascular, nervous andrenal systems are, at best, satisfactory. Thus, in-creasing costs of hospital care programmes andlong term treatment of complications represent theburden placed on society by diabetes mellitus.

Physical activity exerts pronounced effects onsubstrate utilisation and insulin sensitivity, whichin turn potentially lowers blood glucose and lipid

26 Wallberg-Henriksson et al.

Adis International Limited. All rights reserved. Sports Med. 1998 Jan; 25 (1)

levels.[2] Thus, exercise appears to play a promis-ing role in the prevention and therapy of NIDDM.Exercise training may also improve many otherphysiological and metabolic abnormalities that areassociated with NIDDM such as lowering body fat,reducing blood pressure and normalising dys-lipoproteinaemia.

In many patients, NIDDM is likely to evolvefrom a combination of genetic and environmentalfactors, which together, lead to a progressive alter-ation in glucose homeostasis from impaired glu-cose tolerance to impaired β-cell function and/ordecreased peripheral insulin resistance.[3] How-ever, in some cases, environmental factors such asinactivity, obesity and stress are more importantfor the manifestation of the disease. When thesefactors are abolished, insulin sensitivity can im-prove, and the disease may regress.[4] This reviewsummarises the present knowledge in the area ofexercise and NIDDM. Emphasis has been placedon the effects of exercise on peripheral insulin sen-sitivity and the importance of exercise in the pre-vention and clinical management of NIDDM.

1. Clinical Characteristics of Patientswith Non–Insulin-Dependent Diabetes Mellitus (NIDDM)

Although the primary factor in the pathogenesisof NIDDM is unknown, a combination of geneticand environmental factors contribute to the mani-festation of this progressive metabolic disorder,which is not usually clinically apparent until laterin life. Patients with NIDDM are characterised byfasting hyperglycaemia, and by elevated, normalor low levels of insulin. These alterations arecaused mainly by defects at 3 major sites: alteredinsulin secretion from the β-cell, elevated hepaticglucose production, due to decreased insulin sen-sitivity in the liver, and diminished peripheral glu-cose utilisation due to insulin resistance in skeletalmuscle.[5]

Due to the progressive nature of the disease,elucidation of the ‘primary defect’ of this metabo-lic disorder has been difficult to establish in a pa-tient newly diagnosed with NIDDM.[3] Many of

the affected individuals with fasting plasma glu-cose levels in the range of 6 to 12 mmol/L are oftenasymptomatic and undiagnosed.[3] Once the fast-ing glucose levels reach 12 mmol/L or higher, theseindividuals present various clinical symptoms andare diagnosed with NIDDM. Yet, at this point thedisease has progressed from impaired glucose tol-erance to overt NIDDM, and the disease has devel-oped as a consequence of an insulin secretory dys-function[6] and/or insulin resistance.[7] Defects ininsulin secretion have been noted in patients withNIDDM who are lean or of normal bodyweight,[8]

whereas peripheral insulin resistance has been re-ported in patients with NIDDM regardless of obe-sity.[7] However, not all individuals with NIDDMdemonstrate peripheral insulin resistance;[9,10]

thus, decreased insulin action in peripheral tissuesmight not be a primary defect in patients withNIDDM.

Presently, 3 major therapeutic approaches areused in the clinical management of individualswith NIDDM: diet, oral sulphonylureas and insulinadministration. Bodyweight reduction improvesthe metabolic derangement associated with NIDDMand obesity by improving both hepatic and periph-eral insulin sensitivity.[2] Sulphonylureas can ef-fectively lower fasting plasma glucose and reducebasal hepatic glucose production.[11] Furthermore,when diet and sulphonylurea treatment is not suf-ficient, intermittent or chronic insulin treatmentcan be used to overcome insulin resistance and im-prove β-cell responsiveness to glucose.[12]

2. Pathophysiology of NIDDM

2.1 β-Cell Dysfunction

The first phase insulin response to an intrave-nous glucose tolerance test is generally abolishedin patients with NIDDM who have fasting hyper-glycaemia.[13] In contrast, the response to oral glu-cose is more varied; insulin levels are elevated inmildly glucose-intolerant individuals, but fall pro-gressively as the hyperglycaemia deteriorates. Inpatients with NIDDM who have pronounced fast-ing hyperglycaemia, insulin secretion is markedly

Exercise and NIDDM 27


reduced compared with nondiabetic individu-als.[14] Although longitudinal studies are not avail-able, hypersecretion of insulin occurs early in thecourse of NIDDM, whereas later in the course ofthe disease, the β-cell function decreases and insu-lin secretion is insufficient.

2.2 Hepatic Insulin Resistance

Basal hepatic glucose production is markedlyincreased in individuals with NIDDM and fastinghyperglycaemia, irrespective of the presence ofobesity.[15] Furthermore, hepatic glucose produc-tion is not altered in individuals with impaired glu-cose tolerance.[15] Since the degree of the elevationin hepatic glucose production demonstrates a tight,positive correlation with the increase in fastingblood glucose levels, the increased hepatic glucoseproduction appears to be responsible for fastinghyperglycaemia. Under basal conditions, non–insulin-sensitive tissues, such as the brain, accountfor approximately 80% of the total glucose dis-posal,[16] thus insulin resistance in peripheral tis-sues would only have minor effects on fastingblood glucose levels. Consequently, in the basalstate, hepatic overproduction of glucose appears tobe a major determinant for hyperglycaemia. Thehepatic defect appears to be a secondary compo-nent of insulin resistance, since a complete nor-malisation can be obtained by antidiabetic therapy,such as insulin[17] or bodyweight loss.[18]

2.3 Peripheral Insulin Resistance

In vivo studies reveal that, in the postprandialstate, insulin-mediated glucose uptake and utilisa-tion is markedly impaired in patients with dia-betes[19,20] and in first degree relatives who are glu-cose-tolerant.[21] Skeletal muscle is the principlesite of glucose uptake under insulin-stimulatedconditions and accounts for approximately 75% ofglucose disposal following glucose infusion.[19,22]

Under euglycaemic-hyperinsulinaemic conditions,skeletal muscle has been identified as the most im-portant site for insulin resistance in patients withNIDDM.[19] From in vivo tracer studies, the defectin whole body glucose uptake observed in patients

with NIDDM has been localised to the nonoxida-tive pathway for glucose metabolism.[20,23] Thus,in vivo methods have provided indirect evidence tosuggest that the reduced rate of nonoxidative glu-cose metabolism observed in patients with NIDDMmay result from a defect at the level of glucosetransport, glucose phosphorylation or glycogensynthase.

2.4 Glucose Metabolism in Skeletal Muscle

Both in vitro[24-26] and in vivo[27,28] studies re-veal that the decrease in insulin-mediated periph-eral glucose uptake can be localised to a defect inthe mechanism involved in insulin action at thelevel of glucose transport in skeletal muscle. Adose-response relationship for insulin-stimulated3-O-methylglucose transport has been assessed by

1.0

0.8

0.6

0.4

0.2

030 minutes 2 hours

3-O

-Met

hylg

luco

se tr

ansp

ort (

µmol

/ml/h

)

*

Healthy individuals

Individuals with NIDDM

Fig. 1. Normalisation of the decreased capacity for insulin-stim-ulated 3-O-methylglucose transport in vitro. Skeletal musclestrips from healthy individuals (n = 8) and those with non–insu-lin-dependent diabetes mellitus (NIDDM) [n = 7] were exposedfor 2 hours to media containing insulin 100 µU/ml and glucose4 mmol/L. Thereafter, an equimolar concentration of 3-O-methylglucose was substituted for glucose and glucose trans-port was assessed as described.[30] Values are reported asmean ± SEM (reproduced from Zierath et al.,[26] with permis-sion). Symbol: *p < 0.05 significantly different from control mus-cle.



Andréasson et al.[25] in skeletal muscle biopsies ob-tained from individuals with NIDDM and from acontrol group. The dose-response curve for insu-lin-stimulated 3-O-methylglucose transport wasmarkedly decreased in muscle specimens fromthose with NIDDM compared with the controlgroup, suggesting the presence of a post receptordefect in the activation of glucose transport. Re-cently glucose transport has been assessed in skel-etal muscle from individuals with NIDDM andfrom controls using the euglycaemic clamp tech-nique, in conjunction with tracer techniques.[28]

Strikingly, regardless of technique (i.e. in vitro ver-sus in vivo), insulin increases 3-O-methylglucoseuptake 2-fold in control skeletal muscle, whereas,insulin has a minimal effect in increasing glucosetransport in NIDDM skeletal muscle. Furthermore,using both techniques, the rate of insulin-stimu-lated 3-O-methylglucose transport at the level ofskeletal muscle has been reported to be reduced by50% in individuals with NIDDM.[25,26,28] Sinceplasma glucose levels are negatively correlatedwith insulin-mediated whole body glucose uptakeand cellular glucose transport,[29] we have hypo-thesised that hyperglycaemia per se may down-regulate glucose transport. We have shown that theimpaired insulin-stimulated glucose transport inmuscle from individuals with NIDDM can benormalised in vitro following exposure to glucose4 mmol/L [26] (fig. 1). Thus, a component of muscleinsulin resistance is due to an acquired defect re-sulting form disturbances in the metabolic milieu.

Since total skeletal muscle glucose transporterprotein (GLUT4) content is not altered byNIDDM, [31,32] the defect in glucose transport islikely to result from a defect in the traffic or trans-location of GLUT4 from an intracellular site to theskeletal muscle plasma membrane or alternativelyfrom a defect in the insulin signal-transductionpathway. We have shown that insulin-mediatedGLUT4 recruitment is altered in muscle tissuetaken from patients with NIDDM[29] (fig. 2). Otherpotential targets for this defect included reducedphosphorylation and activity of IRS-1 and phos-phatidylinositol 3-kinase, 2 key proteins involved

in the insulin signal-transduction pathway leadingto GLUT4 translocation.[33] Overcoming these de-fects in GLUT4 translocation or in the insulin sig-nal transduction pathway is likely to improve glu-cose homeostasis in the diabetic patient.

3. Effect of Acute Exercise in Patients with NIDDM

3.1 Blood Glucose Levels and Insulin Sensitivity

In healthy, nondiabetic individuals, exercise hasvery little impact on blood glucose levels. How-ever, in individuals with NIDDM, exercise of mod-erate to heavy intensity is usually associated witha decrease in blood glucose levels. Thus, a singlebout of exercise often results in decreased plasmaglucose levels[34,35] which persist into the post-

2.5

2.0

1.5

1.0

0.5

0Control NIDDM

GLU

T4/

Na+

-K+A

TP

ase

α 1-s

ubun

it

*

Basal

Insulin-stimulated

Fig. 2. Effect of insulin on plasma membrane glucose trans-porter protein (GLUT4) levels. Appearance of GLUT4 per α1-subunit of Na+-K+-ATPase plasma membranes prepared fromskeletal muscle obtained at basal insulin levels [68 ± 18 and 48± 7 pmol/L for non–insulin-dependent diabetes mellitus (NIDDM)and control group, respectively, and following in vivo infusion ofinsulin 666 ± 72 and 588 pmol/L for NIDDM and control group,respectively]. Optical density values were corrected per α1-sub-unit of Na+-K+-ATPase in each sample and normalised to thevalue observed for a control group under basal conditions (re-produced from Zierath et al.,[29] with permission). Values arepresented as mean ± SEM. Paired observations were obtainedfor basal and insulin-stimulated conditions. Symbol: *p < 0.05.



exercise period.[34,36] The blood glucose loweringeffect of exercise in individuals with NIDDM canbe explained by the insulin-independent activationof glucose transport by exercise,[37] as well as byincreased insulin sensitivity.[38] However, post-exercise hyperglycaemia and hyperinsulinemiahave been found in individuals with NIDDM fol-lowing maximal dynamic exercise.[39] This alteredglycaemic and insulinaemic response may resultfrom an exaggerated counter-regulatory hormonalresponse.

3.2 Hormone Levels

Moderate to heavy exercise reduces insulin se-cretion, and consequently, plasma insulin levels arereduced.[40] Conversely, glucagon and adrenalinelevels increase during exercise.[41] Glucose homeo-stasis is generally maintained during exercise be-cause the increase in glucose utilisation is matchedby an increase in hepatic glucose production. Theexercise-induced increase in glucagon and the fallin insulin levels can control almost the entire in-crease in glucose production. Despite the importantrole of pancreatic hormones in the control of he-patic glucose production during exercise, the in-crease in glucose uptake in peripheral tissues iscontrolled primarily by insulin-independent mech-anisms.[37,42]

3.3 Glucose Transport

One of the few states in which GLUT4 levelsare modulated is through exercise. GLUT4 levelsare increased in young athletes compared with sed-entary individuals.[43] Furthermore, GLUT4 levelsare increased in exercise-trained middle aged indi-viduals with either normal glucose tolerance, im-paired glucose tolerance, or in individuals withNIDDM, compared with sedentary individu-als.[44,45] Thus, these individuals have improved in-sulin sensitivity, which may be partially explainedby increased GLUT4 expression. Exercise per se isalso a profound stimulator of glucose transport.[30]

In the insulin-resistant streptozotocin-induced dia-betic rat, muscle contractions can increase glucoseuptake in the absence of insulin.[37,42,46] These

early studies were the first to suggest that exercisemay increase glucose transport in muscle via a sep-arate pathway from that of insulin, a classical reg-ulator of glucose transport in skeletal muscle. Inhumans, insulin sensitivity is related to the degreeof physical activity.[47] Furthermore, exercise en-hances insulin sensitivity in obese individuals andthose with NIDDM.[48,49] We have recently shownthat exercise can increase glucose transport in mus-cle tissue from healthy individuals (fig. 3).[50] Theseparate or combined effects of bicycle exercise toexhaustion and insulin action on glucose transportwere investigated in vastus lateralis skeletal mus-cle using an open muscle biopsy technique.[50] Ourresults show that in humans, exercise is a potentregulator of glucose transport activity. Further-more, the combined effects of insulin and exerciseare additive on glucose transport activity, a finding

6

5

4

3

2

1

0Basal Insulin Exercise Exercise

+ insulinCalculatedcombined

effect

3-O

-Met

hylg

luco

se tr

ansp

ort (

µmol

/ml/h

)

Fig. 3. The separate or combined effects of bicycle exercise toexhaustion and insulin 1000 µU/ml on the rate of 3-O-methylglucose activity in human skeletal muscle. A musclespecimen was obtained by means of an open biopsy[50] following90 minutes of bicycle exercise to exhaustion at 65% V

.O2max.

Muscle strips were prepared and incubated as described[50] and3-O-methylglucose was assessed.[30] Glucose transport is ex-pressed per ml of intracellular water. The theoretical basal rateof glucose transport is indicated by the shaded areas (n = 30).The rate of insulin-stimulated glucose transport was assessedin sedentary individuals (n = 30). The right-hand bar indicatesthe calculated combined effects of exercise and insulin on 3-O-methylglucose transport (calculated from the sum of the actualeffects of exercise and insulin above the basal rate). All otherpoints represent the mean ± SEM for 3-5 observations (repro-duced from Zierath et al.,[50] with permission).



which is consistent with studies performed usingisolated rat epitrochlearis muscle.[30,42] Thus, inpatients with NIDDM, regular exercise trainingmay also enhance cellular glucose uptake. Theseimprovements in insulin sensitivity may overcomedefects in the insulin signal transduction pathwaynoted in muscle tissue taken from individuals withNIDDM.[33]

4. Long Term Effect of Exercise inPatients with NIDDM

4.1 Metabolic Control and Insulin Sensitivity

When individuals with NIDDM participate inregular exercise programmes, metabolic control isusually improved, particularly in younger individ-uals.[2,51] An improvement in metabolic control hasbeen observed in those with a mean age of between40 and 54 years, whereas in groups with a meanage of between 57 and 61 years, no response tophysical training programmes was observed.[2]

The apparent difference in metabolic response toregular exercise between younger and olderNIDDM individuals is not clear, but possible ex-planations may be differences in training intensity,degree of complications and pretraining level ofmetabolic control and bodyweight.

4.2 Bodyweight

In individuals with NIDDM and obesity a diet-induced decrease in bodyweight is associated withthe beneficial effects of improved metabolic con-trol and reduced risk of ischaemic heart disease.[35]

Unfortunately, patient compliance with weightloss programmes is often low and diet therapyalone is generally not sufficient to maintain weightloss on a long term basis.[52] Programmes whichcombine diet and exercise are generally more suc-cessful in achieving bodyweight reduction in theparticipants.[53]

4.3 Cardiovascular Risk Factors

Physical training is associated with a decreasein serum triglyceride levels, particularly very low-density lipoproteins, and an increase in high den-

sity lipoprotein-2 cholesterol.[54] Furthermore,physical activity has been reported to lower bloodpressure in hypertensive individuals.[55,56] The lat-ter adaptation occurs independent of weight lossand changes in body composition. Furthermore,physical training has been shown to increase car-diovascular fitness and physical working capacityin individuals with NIDDM.[57]

5. Exercise in the Prevention of NIDDM

5.1 Relationship Between Physical Fitnessand Insulin Sensitivity

Physical work capacity is a major determinantof whole body insulin and skeletal muscle insulinsensitivity. Several groups have utilised the hyper-insulinaemic euglycaemic clamp technique todemonstrate a strong correlation between wholebody insulin-mediated glucose uptake and V

.O2max

in healthy individuals.[58] Several factors may con-tribute to this relationship. Physical training leadsto changes in body composition, with a reductionin body fat and an increase in muscle tissue. Thesechanges in body composition lead to an increasedinsulin-stimulated whole body glucose dispo-sal.[59] Furthermore, exercise enhances the effectof insulin to stimulate blood flow,[60] the contentof the insulin-sensitive glucose transport molecule,GLUT4[61] and the activity of glycogen syn-thase.[61] Collectively, these factors may also con-tribute to the increased insulin sensitivity in trainedindividuals. Due to the strong correlation betweenphysical fitness and whole body insulin sensitivity,it is tempting to speculate that inactivity per seconstitutes an important risk factor for the devel-opment of NIDDM. Conversely, physical trainingmay be a potential tool to prevent the developmentof NIDDM, especially in groups which are at ahigher risk (fig. 4).

A prospective study performed in Pima Indiansrevealed that insulin resistance is a major risk fac-tor for the development of NIDDM in this popula-tion.[5] Interestingly, in this study, neither in-creased glucose output from the liver duringinsulin infusion nor reduced insulin secretion from



the β-cell in response to glucose was shown to pre-dict for NIDDM.[5] However, an association be-tween a decreased level of physical activity and anincreased occurrence of NIDDM has been shownin 2 large prospective studies.[62,63] In another pro-spective population-based study of middle-agedWhite men, both physical fitness and the level ofphysical activity were shown to correlate with thedevelopment of diabetes.[64]

5.2 First Degree Relatives of Patients with NIDDM

In glucose-tolerant first degree relatives of indi-viduals with NIDDM, reduced peripheral insulinsensitivity occurs in the presence[65-67] as well as inthe absence[21] of hyperinsulinaemia. One interpre-tation arising from these studies suggests that pe-ripheral insulin resistance precedes the develop-ment of NIDDM and constitutes a primary factorin the pathogenesis of NIDDM. Nevertheless, dif-ferences in lean body mass or in the degree of phys-ical fitness (V

.O2max) can contribute to insulin sen-

sitivity.[47] Consequently, if the first degreerelatives of individuals with NIDDM in the latterstudies were matched to control individuals by de-gree of physical fitness (V

.O2max), the differences

in insulin sensitivity would be less apparent.In an attempt to examine the possible role of

reduced physical fitness in developing NIDDM, 21first degree relatives of individuals with NIDDMand 22 controls without any family history of dia-

betes were investigated with respect to maximalworking capacity and whole body insulin-mediatedglucose uptake.[68] V

.O2max was lower in the rela-

tives of patients with NIDDM than in the controlgroup. Furthermore, a highly significant cor-relation was found between the rate of insulin-mediated whole body glucose uptake and V

.O2max

in both relatives and control group.[68] Thus, insu-lin resistance in healthy first degree relatives ofindividuals with NIDDM is associated with a di-minished physical work capacity.

5.3 Strategies to Prevent NIDDM

Clearly, the finding of decreased working ca-pacity in first degree relatives of individuals withNIDDM [68] underscores the importance of match-ing participants for V

.O2max in order to investigate

whether peripheral insulin resistance precedes thedevelopment of NIDDM. However, this findingalso gives rise to the possibility of utilising physi-cal training as a means of preventing NIDDM. Inpractical terms, this could be achieved in severaldifferent ways. However, such a strategy would de-pend on the type of medical care system availableto the individual.

One general determinant in a strategy to developa preventive programme for NIDDM is testing forV.O2max. Typically this type of testing is not per-

formed in the general population or routinely onindividuals at risk for developing NIDDM. From apractical point of view, administration of this test

Geneticpredisposition

+

Risk factors: inactivity obesity stress poor diet

Risk factors: decreased insulin sensitivity

NIDDM

NIDDM +complications

Regular exercise: improves insulin sensitivity activates glucose transport

Fig. 4. The role of exercise in the prevention and therapy of NIDDM.



to the general population is not feasible. However,it should be possible to establish a testing pro-gramme which includes V

.O2max determinations for

individuals who are at risk of developing NIDDM.

6. Exercise as a Therapeutic Tool inClinical Practice

6.1 Medical Considerations

Before initiating an physical training programme,individuals with NIDDM are recommended toundergo a complete physical examination which isdirected towards identifying any long term compli-cations of diabetes. According to the AmericanCollege of Sports Medicine Guidelines,[69] all in-dividuals above the age of 35 years should performan exercise stress test before engaging in an exer-cise programme which includes moderate to vigor-ously intense exercise. The stress test will identifyindividuals with previously undiagnosed isch-aemic heart disease and abnormal blood pressureresponses. It is important to diagnose proliferativeretinopathy, microalbuminuria, peripheral and/orautonomic neuropathy in individuals with NIDDMbefore they participate in an exercise programme.If any diabetic complications are present, the exer-cise protocol should be modified accordingly.

6.2 Type of Exercise

The exercise programme should consist of mod-erate intensity aerobic exercise. Resistance train-ing and high-intensity exercises should only beperformed by individuals who do not have prolif-erative retinopathy or hypertension. The individualshould choose a suitable activity which he or shecan perform on a regular basis. A flexible pro-gramme based on lifestyle-oriented exercise ismore conducive to long term exercise adherencethan a structured programme.[70] Once enrolled inthe exercise programme, the individual must beeducated with regard to proper footwear and dailyfoot inspections. Fluid intake is of great impor-tance when exercising for prolonged periods or inwarm and humid environments.

7. Conclusions

Clearly, regular physical activity plays an im-portant role in the prevention and treatment ofNIDDM. In various populations, physical inactiv-ity and abdominal obesity have been linked to thedevelopment of NIDDM.[71] In prospective stud-ies, physical activity, independent of other risk fac-tors such as obesity, hypertension and family his-tory of NIDDM, has a protective effect against thedevelopment of NIDDM.[63,72,73] Furthermore, aninverse relation between energy expenditure in lei-sure-time physical activity and the development ofNIDDM over 15 years has been shown.[62] Thesestudies[62,63,71-73] provide convincing evidence thatregular physical activity serves as protectionagainst NIDDM. Thus, physical training pro-grammes suitable for individuals at risk forNIDDM should be incorporated into the medicalcare system to a greater extent. In people withNIDDM, regular physical activity results in sev-eral beneficial changes, such as increased insulinsensitivity, improved glycaemic control, improvedlipid profile, lower blood pressure and increasedcardiovascular fitness. Furthermore, exercise in-creases the energy expenditure, which, when com-bined with diet, can lead to bodyweight loss anddecreased body fat content.

Acknowledgements

This work was supported by grants from the SwedishMedical Research Council (9517, 12211 and 11823), and theSwedish Diabetes Association. Juleen Zierath is a recipientof a Junior Individual Grant from the Swedish Foundationfor Strategic Research. We are especially grateful to MrsYvonne Stridsberg for her skilful secretarial assistance.

References1. Deckert T, Poulsen JE, Larsen M. Prognosis of diabetics with

diabetes onset before the age of 31: I. Survival, causes ofdeath, and complications. Diabetologia 1978; 14: 363-70

2. Zierath JR, Wallberg-Henriksson H. Exercise training in obesediabetic patients: special considerations. Sports Med 1992;14: 171-89

3. Turner RC, Hattersley AT, Shaw JTE, et al. Type II diabetes:clinical aspects of molecular biological studies. Diabetes1995; 44: 1-10

4. Friedman JE, Dohm GL, Leggett-Frazier N, et al. Restorationof insulin responsiveness in skeletal muscle of morbidlyobese patients after weight loss. J Clin Invest 1992; 89: 701-5



5. Lillioja S, Mott DM, Spraul M, et al. Insulin resistance andinsulin secretory dysfunction as precursors of non-insulin-de-pendent diabetes mellitus. N Engl J Med 1993; 329: 1988-92

6. Porte Jr D. β-cells in type II diabetes mellitus. Diabetes 1991;40: 166-80

7. DeFronzo RA. The triumvirate: β-cell, muscle or liver - a col-lusion responsible for NIDDM. Diabetes 1988; 37: 667-87

8. Kelley DE, Mokan M, Mandarino LJ. Metabolic pathways ofglucose in skeletal muscle of lean NIDDM patients. DiabetesCare 1993; 16: 1158-66

9. Efendic S, Kahn A, Östenson CG. Insulin release in Type 2diabetes mellitus. Diabete et Metabolisme 1994; 20: 81-6

10. Galuska D, Nolte LA, Zierath JR, et al. Effect of metformin oninsulin-stimulated glucose transport in isolated muscle ob-tained from patients with NIDDM. Diabetologia 1994; 37:826-32

11. Lebovitz HE. Oral hypoglycemic agents in diabetes mellitus.In: Rifkin H, Porte D, editors. Diabetes mellitus: theory andpractice, 4th ed. Elsevier 1990; 554-74

12. Edelman SV, Henry RR. Intensive insulin therapy for patientswith type II diabetes. In: Le Roith D, Taylor SI, Olefsky JM,editors. Diabetes mellitus. Philadelphia: Lippincott-Raven,1996; 647-60

13. Efendic S, Östenson CG. Hormonal responses and future treat-ment of non-insulin-dependent diabetes mellitus (NIDDM). JIntern Med 1993; 234: 127-38

14. DeFronzo RA, Ferrannini E. Influence of plasma glucose andinsulin concentration on plasma glucose clearance in man.Diabetes 1982; 31: 683-8

15. Kolterman OG, Gray RS, Griffin J, et al. Receptor andpostreceptor defects contribute to the insulin resistance innon-insulin dependent diabetes mellitus. J Clin Invest 1981;68: 957

16. Nolan JJ, Olefsky JM. Multiple etiologies of insulin resistancein NIDDM. In: Kawamori R, Vranic M, Horton ES, et al.London: Smith-Gordon, 1995: 193-200

17. Garvey WT, Olefsky JM, Griffin J, et al. The effects of insulintreatment on insulin secretion and action in Type II diabetesmellitus. Diabetes 1985; 34: 222-34

18. Henry RR, Wallace P, Olefsky JM. The effects of weight losson the mechanisms of hyperglycemia in obese non-insulin-dependent diabetes mellitus. Diabetes 1986; 35: 990-8

19. DeFronzo RA, Gunnarsson R, Björkman O, et al. Effects ofinsulin on peripheral and splanchnic glucose metabolism innon-insulin-dependent (type II) diabetes mellitus. J Clin In-vest 1985; 76: 149-55

20. Shulman GI, Rothman DL, Jue T, et al. Quantitation of muscleglycogen synthase in normal subjects and subjects with non-insulin-dependent diabetes mellitus by 13C nuclear magneticresonance spectroscopy. N Engl J Med 1990; 322: 223-8

21. Henriksen JE, Alford F, Handberg A, et al. Increased glucoseeffectiveness in normoglycemic but insulin-resistant relativesof patients with non-insulin-dependent diabetes mellitus. JClin Invest 1994; 94: 1196-204

22. Nuutila, P, Koivisto VA, Knuuti J, et al. Glucose-free fatty acidcycle operates in human heart and skeletal muscle in vivo. JClin Invest 1992; 89: 1767-74

23. Groop L, Bonadonna RC, DelPrato S, et al. Glucose and freefatty acid metabolism in non-insulin-dependent diabetes mel-litus: evidence for multiple sites of insulin resistance. J ClinInvest 1989; 84: 205-13

24. Dohm GL, Tapscott EB, Pories WJ, et al. An in vitro humanskeletal muscle preparation suitable for metabolic studies.Decreased insulin stimulation of glucose transport in muscle

from morbidly obese and diabetic subjects. J Clin Invest 1988;82: 486-94

25. Andréasson K, Galuska D, Thörne A, et al. Decreased insulin-stimulated 3-O-methylglucose transport in in vitro incubatedmuscle strips from type II diabetic subjects. Acta PhysiolScand 1991; 142: 255-60

26. Zierath JR, Galuska D, Nolte LA, et al. Effects of glycaemia onglucose transport in isolated skeletal muscle from patientswith NIDDM: in vitro reversal of muscular insulin resistiance.Diabetologia 1994; 37: 270-7

27. Fink RI, Wallace P, Brechtel G, et al. Evidence that glucosetransport is rate-limiting for in vivo glucose uptake. Metabo-lism 1992; 41: 897-902

28. Bonadonna RC, Del Prato S, Saccomani MP, et al. Transmem-brane glucose transport in skeletal muscle of patients with non-insulin-dependent diabetes. J Clin Invest 1993; 92: 486-94

29. Zierath JR, He L, Gumá A, et al. Insulin action on glucosetransport and plasma membrane content in skeletal musclefrom patients with NIDDM. Diabetologia 1996; 39: 1180-9

30. Wallberg-Henriksson H. Glucose transport into skeletal muscle:influence of contractile activity, insulin, catecholamines anddiabetes mellitus. Acta Physiol Scand Suppl 1987; 564: 1-80

31. Pedersen O, Bak JF, Andersen PH, et al. Evidence against al-tered expression of GLUT1 or GLUT4 in skeletal muscle ofpatients with obesity or NIDDM. Diabetes 1990: 39: 865-70

32. Handberg A, Vaag A, Damsbo P, et al. Expression of insulin-regulatable glucose transporters in skeletal muscle from Type2 (non-insulin-dependent) diabetic patients. Diabetologia1990; 33: 625-7

33. Björnholm M, Kawano Y, Lehtihet M, et al. Insulin receptorsubstrate-1 phosphorylation and phosphatidylinositol 3-ki-nase activity in skeletal muscle from NIDDM subjects afterin vivo insulin stimulation. Diabetes 1997; 46: 524-7

34. Hübinger A, Franzen A, Gries FA. Hormonal and metabolicresponse to physical exercise in hyperinsulinaemic and non-hyperinsulinaemic type 2 diabetics. Diabetes Res Clin Pract1987; 4: 57-61

35. Schneider SH, Vitug A, Ruderman NB. Atherosclerosis andphysical activity. Diabetes Metab Rev 1986; 1: 514-53

36. Minuk HL, Vranic M, Marliss EB, et al. Glucoregulatory andmetabolic response to exercise in obese non-insulin-depend-ent diabetes. Am J Physiol 1981; 240: E458-64

37. Wallberg-Henriksson H, Holloszy JO. Contractile activity in-creases glucose uptake by muscle in severely diabetic rats. JAppl Physiol 1984; 57: 1045-9

38. Burstein R, Epstein Y, Shapiro Y, et al. Effect of an acute boutof exercise on glucose disposal in human obesity. J Appl Phys-iol 1990; 69: 299-304

39. Kjaer M, Hollenbeck CB, Frey-Hewitt B, et al. Glucoregulationand hormonal responses to maximal exercise in non-insulin-dependent diabetes. J Applied Physiol 1990; 685: 2067-74

40. Galbo H. Hormonal and metabolic adaptation to exercise. NewYork: Thieme-Stratton Inc., 1983

41. Wasserman DH, Spalding JA, Lacy DB, et al. Glucagon is aprimary controller of hepatic glycogenolysis and gluconeo-genesis during muscular work. Am J Physiol 1989; 257:E108-17

42. Wallberg-Henriksson H, Holloszy JO. Activation of glucosetransport in diabetic muscle: responses to contraction and in-sulin. Am J Physiol 1985; 249: C233-7

43. Andersen PH, Lund S, Schmitz O, et al. Increased insulin-stim-ulated glucose uptake in athletes: the importance of GLUT4mRNA, GLUT4 protein and fibre type composition of skele-tal muscle. Acta Physiol Scand 1993; 149: 393-404



44. Houmard JA, Egan PC, Neufer PD, et al. Elevated skeletal mus-cle glucose transporter levels in exercise-trained middle-agedmen. Am J Physiol 1991; 261: E437-43

45. Hughes V, Fiatarone M, Fielding R, et al. Exercise increasesmuscle GLUT4 levels and insulin action in subjects with im-paired glucose tolerance. Am J Physiol 1993; E855-62

46. Ploug T, Galbo H, Richter EA. Increased muscle glucose uptakeduring contractions: no need for insulin. Am J Physiol 1984;247: E726-31

47. Seals DR, Hagberg JM, Hurley BF, et al. Effects of endurancetraining on glucose tolerance and plasma lipid levels in oldermen and women. JAMA 1984; 252: 645-9

48. Björntorp P, Holm G, Jacobsson B. Physical training in humanhyperplastic obesity: effect on the normal status. Metabolism1997; 26: 319-28

49. Trovati M, Carta Q, Cavalot F, et al. Influence of physical train-ing on blood glucose control, glucose tolerance, insulin se-cretion, and insulin action in non-insulin-dependent diabeticpatients. Diabetes Care 1984; 7: 416-20

50. Zierath JR. In vitro studies of human skeletal muscle: hormonaland metabolic regulation of glucose transport. Acta PhysiolScand Suppl 1995; 626: 1-96

51. Wallberg-Henriksson H. Exercise and diabetes mellitus. ExercSports Sci Rev 1992; 20: 339-68

52. Schneider SH, Ruderman NB. Exercise and NIDDM. DiabetesCare 1990; 13: 785-9

53. Bogardus C, Ravussin E, Robbins DC, et al. Effects of physicaltraining and diet therapy on carbohydrate metabolism in pa-tients with glucose intolerance and non-insulin-dependent di-abetes mellitus. Diabetes 1984; 33: 311-8

54. Huttunen JK, Lansimies E, Voutilainen E, et al. Effect of mod-erate physical exercise on serum lipoproteins. Circulation1979; 60: 1220-9

55. Boyer J, Kasch F. Exercise therapy in hypertensive men. JAMA1970; 211: 1668-71

56. Leon AS, Conrad JC, Casal DC, et al. Exercise for diabetics:effects of conditioning at constant body weight. J Card Rehab1984; 4: 278-86

57. Holloszy JO, Schultz J, Kusnierkiewics J, et al. Effects of ex-ercise on glucose tolerance and insulin resistance. Acta MedScand 1987; 711 Suppl.: 55-65

58. Koivisto VA, Yki-Järvinen H, DeFronzo RA. Physical trainingand insulin sensitivity. Diabetes Metab Rev 1986; 1: 445-81

59. Yki-Järvinen H. Koivisto VA. Effects of body composition oninsulin sensitivity. Diabetes 1983; 32: 965-9

60. Laakso M, Edelman SV, Brechtel G, et al. Decreased effect ofinsulin to stimulate skeletal muscle blood flow in obese man:a novel mechanism for insulin resistance. J Clin Invest 1990;85: 1844-52

61. Ebeling P, Bourey R, Koranyi L, et al. Mechanism of enhancedinsulin sensitivity in athletes: increased blood flow, muscle

glucose transport protein (GLUT-4) concentration, and gly-cogen synthase activity. J Clin Invest 1993; 92: 1623-31

62. Helmrich SP, Ragland DR, Leung RW, et al. Physical activityand reduced occurrence on non-insulin-dependent diabetesmellitus. N Engl J Med 1991; 325: 147-52

63. Manson JE, Rimm EB, Stampfer MJ et al. Physical activity andincidence of non-insulin-dependent diabetes mellitus inwomen. Lancet 1991; 338: 774-8

64. Eriksson KF, Lindgärde F. Poor physical fitness, and impairedearly insulin response but late hyperinsulinaemia, as predic-tors of NIDDM in middle-aged Swedish men. Diabetologia1996; 39: 573-9

65. Vaag A, Henriksen JE, Beck-Nielsen H. Decreased insulin ac-tivation of glycogen synthase in skeletal muscles in youngnonobese Caucasian first-degree relatives of patients withnon-insulin-dependent diabetes mellitus. J Clin Invest 1992;89: 782-8

66. Eriksson J, Koranyi L, Bourey R, et al. Insulin resistance in type2 (non-insulin-dependent) diabetic patients and their relativesis not associated with a defect in the expression of the insu-lin-responsive glucose transporter (GLUT4) gene in humanskeletal muscle. Diabetologia 1992; 35: 143-7

67. Martin BC, Warram JH, Krolewski AS, et al. Role of glucoseand insulin resistance in development of type diabetes melli-tus: results of a 25-year follow-up study. Lancet 1992; 340:925-9

68. Nyholm B, Mengel A, Nielsen S, et al. Insulin resistance inrelatives of NIDDM patients: the role of physical fitness andmuscle metabolism. Diabetologia 1996; 39: 813-22

69. American College of Sports Medicine. Guidelines for exercisetesting and prescription, 4th ed. Philadelphia: Lea & Febiger,1991: 5-9

70. Epstein LH, Wing RR, Koeske R, et al. A comparison of life-style exercise, aerobic exercise, and calisthenics on weightloss in obese children. Behav Ther 1985; 16: 345-56

71. Dowse GK, Zimmer PZ, Gareeboo H, et al. Abdominal obesityand physical activity are risk factors for NIDDM and im-paired glucose tolerance in Indian, Creole, and ChineseMauritians. Diabetes Care 1991; 14: 271-82

72. Frisch RE, Wyshak G, Albright TE, et al. Lower prevalence ofdiabetes in female former college athletes compared withnonathletes. Diabetes 1986; 35: 1101-5

73. Manson JE, Nathan DM, Krolewski AS, et al. A prospectivestudy of exercise and incidence of diabetes among US malephysicians. JAMA 1992; 268: 63-7

Correspondence and reprints: Associate Prof. HarrietWallberg-Henriksson, Department of Clinical Physiology,Karolinska Hospital, S-171 76 Stockholm, Sweden.



exercise in the management of non—insulin-dependent diabetes mellitus

Documents