chronic changes in skeletal muscle histology and function in

413

Chronic Changes in Skeletal Muscle Histologyand Function in Peripheral Arterial Disease

Judith G. Regensteiner, PhD; Eugene E. Wolfel, MD; Eric P. Brass, MD, PhD;Michael R. Carry, PhD; Steven P. Ringel, MD; Melanie E. Hargarten, MS;

Elizabeth R. Stamm, MD; and William R. Hiatt, MD

Background. Peripheral arterial disease (PAD) is associated with an impairment in exercise perfor-mance and muscle function that is not fully explained by the reduced leg blood flow during exercise. Thisstudy characterized the effects of PAD on muscle function, histology, and metabolism.Methods and Results. Twenty-six patients with PAD and six age-matched control subjects were studied.

Ten of the PAD patients had unilateral disease, which permitted paired comparisons between theirdiseased and nonsymptomatic legs. All PAD patients had a lower peak treadmill walking time and peakoxygen consumption than controls. Vascular disease (diseased leg in unilateral patients and the mostseverely diseased leg in bilateral patients) was associated with decreased calf muscle strength comparedwith control values. In patients with unilateral disease, the diseased legs had a greater percentage ofangular fibers (indicating chronic denervation) and a decreased type H fiber cross-sectional area(expressed as percent of total fiber area) compared with the nonsymptomatic, or control, legs. In diseasedlegs, gastrocnemius muscle strength was correlated with the total calf cross-sectional area (r=0.78,p<0.05) and type II fiber cross-sectional area (r=0.63, p<0.05). Activities of citrate synthase, phospho-fructokinase, and lactate dehydrogenase in all 26 PAD patients (most diseased leg) did not differ fromcontrol values. Despite a wide range in citrate synthase activity in PAD patients, activity of this enzymewas not correlated with muscle strength or treadmill exercise performance.

Conclusions. In patients with PAD, gastrocnemius muscle weakness is associated with muscle fiberdenervation and a decreased type II fiber cross-sectional area. In contrast, the PAD patients displayedsubstantial heterogeneity in muscle enzyme activities that was not associated with exercise performance.Denervation and type H fiber atrophy may contribute to the muscle dysfunction in patients with PAD andfurther confirm that the pathophysiology of chronic PAD extends beyond arterial obstruction. (Circulation1993;87:413-421)KEYWoRDs * peripheral arterial disease * muscle, skeletal * exercise * claudication * enzymes

* denervation, muscle * peripheral vascular diseases

P atients with peripheral arterial disease (PAD)develop intermittent claudication with walkingexercise because extremity blood flow is limited

and inadequate to meet the metabolic demand of themuscle.1 This results in an objective impairment inexercise performance, such that the peak oxygen con-sumption during treadmill testing is 50% lower thanvalues in age-matched controls.2 The etiology of theexercise impairment in patients with PAD is multifac-

From the Department of Medicine (J.G.R., E.E.W., M.E.H.,W.R.H.), Section of Vascular Medicine, and the Departments ofNeurology (M.R.C., S.P.R.) and Radiology (E.R.S.), University ofColorado School of Medicine, Denver, Colo.; and the Depart-ments of Medicine and Pharmacology (E.P.B.), Division of Clin-ical Pharmacology, Case Western Reserve University, Cleveland,Ohio.

Supported by a Biomedical Research Support grant (2530538)and a grant from the National Institute on Disability and Reha-bilitation Research (G008635118). E.P.B. is a Burroughs Well-come Scholar in Clinical Pharmacology.Address for correspondence: Judith G. Regensteiner, PhD,

University of Colorado Health Sciences Center, Box B-180, 4200E. 9th Avenue, Denver, CO 80262.

Received February 12, 1992; revision accepted October 13,1992.

torial. Although muscle ischemia is a major contributor,the hemodynamic severity of the disease (assessed bymeasurements of peripheral blood flow or ankle pres-sure) is poorly correlated with exercise performance.3'4Patients with severe ischemic disease develop structuralchanges in skeletal muscle consisting of denervation,fiber atrophy, and a selective loss of type II fibersrelative to type I fibers that may contribute to themuscle dysfunction.5-7 Ambulatory patients with mildPAD also have muscle weakness and reduced muscleendurance,8 but it is not established whether denerva-tion and fiber atrophy also contribute to their impair-ment in muscle function.

In patients with PAD, an increase in skeletal muscleoxidative enzyme activity may be an adaptive responseto the reduced skeletal muscle blood flow.9-1' However,other investigators have observed a decrease in oxida-tive enzyme activity in PAD, particularly in more severeforms of the disease.12"13 Thus, changes in skeletalmuscle enzyme activities in patients with PAD may beheterogeneous, and the functional significance of thesechanges is not well established.To further define the structural and metabolic

changes that occur in skeletal muscle of ambulatory

by guest on September 27, 2016

http://circ.ahajournals.org/D

ownloaded from

http://circ.ahajournals.org/

414 Circulation Vol 87, No 2 February 1993

TABLE 1. Characteristics of Patients With Peripheral ArterialDisease and Control Subjects

PADUnilateral All PAD Controls

DemographicsNo. of patients 10 26 6Age (years) 64±10 65±7 63±5

Peak treadmill performanceWalking time (minutes) 9.8±2.9* 8.1±4.1* 18.6±2.3Vo2 (ml/kg/min) 15.1±1.8* 14.5±2.6* 27.4±3.8

Ankle/brachial indexResting

Diseased 0.72±0.10*t 0.66±0.21* 1.10±0.07Nonsymptomatic 0.97±0.07*

After exerciseDiseased 0.34±0.14*t 0.27±0.14* 1.04±0.09Nonsymptomatic 0.76±0.16*

Patients with unilateral peripheral arterial disease (PAD) andunilateral combined with bilateral diseased patients (all PAD) andcontrol subjects performed graded treadmill exercise protocol todefine peak exercise performance. Measurements of the ankle/brachial systolic blood pressure index were performed at rest andimmediately after exercise. Values are mean±SD.

*p<0.05 for the value in patients compared with controlsubjects.

tp<0.05 for the value in the diseased leg of unilateral patientscompared with their nonsymptomatic leg.

PAD patients as well as the role of these changes inmuscle function and exercise performance, we evalu-ated muscle histology, biochemistry, and exercise per-formance in 26 patients with PAD and in six normalcontrols. Ten of the patients had vascular disease con-fined to one leg, which facilitated comparisons betweendiseased and nonsymptomatic legs, permitting the non-ischemic metabolic and histological states of the indi-vidual's muscle to be defined in the nonsymptomaticlegs. The results demonstrate PAD-associated muscledenervation and atrophy of type II fibers that correlatewith muscle dysfunction in the underperfused muscle.

MethodsSubjects

Twenty-six male patients with PAD were studied, ofwhich 10 had intermittent claudication in only one leg(unilateral patients), and 16 had claudication in bothlegs. Unilateral PAD was confirmed by an abnormalankle-to-brachial systolic blood pressure indexes(ABIs) in the symptomatic leg (see below). In theunilateral patients, four had aortoiliac disease, whereasall 10 had femoropopliteal disease as identified bysegmental limb pressures and pulse volume recordingsusing established criteria.14 Six age- and activity-matched control subjects were also enrolled (Table 1).All participants were sedentary, reporting that they didnot exercise on a regular basis. Patients were excluded ifthey were diabetic or had ischemic rest pain, ulceration,or gangrene. Patients also were excluded if their exer-cise capacity was limited by symptoms of angina, con-gestive heart failure, chronic obstructive pulmonarydisease, or arthritis. Chronic medications were contin-ued at the same dosages. Control subjects had noclinical evidence of chronic or active disease, received

no medications, and had an unremarkable medicalhistory. The study was approved by the University ofColorado School of Medicine Human Subjects Commit-tee, and informed consent was obtained from all en-rolled subjects.

Vascular TestingAll subjects were initially studied at rest in the supine

position with cuffs on each ankle and arm. At eachankle, systolic blood pressures were measured by Dopp-ler ultrasound (model 841, Parks Medical Electronics,Beaverton, Ore.) in duplicate and averaged for thedorsalis pedis and posterior tibial arteries. Systolicblood pressures were also measured in duplicate in eacharm by auscultation. The right and left ABIs at rest werecalculated from the highest arm and ankle pressures.One minute after the completion of the exercise test,systolic blood pressures were measured in the anklesand arm to determine the postexercise ABI. The criteriafor vascular disease were a resting ratio of <0.94 thatdecreased to <0.73 after exercise.15 Data from both legsare presented for the unilateral patients, but only datafrom the most diseased leg of 16 patients with bilateralPAD were used (most diseased leg was determinedfrom the lowest rest and postexercise ABIs).

Treadmill TestingIn patients with PAD, a graded treadmill protocol was

performed to maximally tolerated claudication pain usingpreviously described and validated methods.4 Controlsexercised to exhaustion using the same protocol. Rates ofoxygen consumption (Vo2) and carbon dioxide produc-tion (Vco2) were measured at rest and during treadmillexercise with an Ametek metabolic system (AmetekThermox, Pittsburgh, Pa.). Arm blood pressure (by aus-cultation) and heart rate (by 12-lead ECG) were ob-tained every minute during exercise. Cardiac status wasmonitored throughout the treadmill test by 12-lead ECG.The exercise test began at 2 mph and 0% grade, and thegrade was increased by 3.5% every 3 minutes to peakexertion. In all subjects, peak exercise performance wascharacterized by the longest walking time and the highestVo2 attained during the treadmill test.

Muscle Strength and Endurance TestingStrength testing was performed in all subjects on a

Cybex Dynamometer (Lumex Inc., Ronkonkoma,N.Y.). Testing of the gastrocnemius and anterior tibialmuscles was performed with the subject in the supineposition with the leg fully extended. Joints not beingtested were stabilized to prevent recruitment of othermuscle groups. Each subject flexed and extended thefoot with maximal effort on a pedal at a regulated speedof 60° sec` for a total of five repetitions. This sequencewas performed a second time, and the single maximalvalue for each muscle group (expressed in foot-poundsof peak torque) was used for analysis.16 No patientexperienced claudication during strength testing. Aftera 3-minute rest, muscle endurance was measured in thegastrocnemius muscles. The subjects were in the samepositions as for strength testing, but each muscle con-traction was performed at a rate of 2400 sec`. Muscleendurance was recorded as the number of repetitionsperformed using a particular muscle group until a 50%decrement in peak torque was observed.16 Endurance



ownloaded from


Regensteiner et al Effects of PAD on Muscle Function 415

testing was performed twice, and the highest number ofrepetitions was reported. Claudication was experiencedby five of 10 patients during muscle endurance testing,but in no case did a patient reach a severe level ofclaudication pain (as defined by a standard scale; Ref-erence 17) before the exercise was terminated.

Calf Cross-sectional AreaTo estimate the maximal calf muscle cross-sectional

area, calf circumference (in centimeters) was measuredat the largest part of the muscle and converted tocross-sectional area by the following formula: circum-ference2/41r (assuming the calf to be nearly circular). Inaddition, a subset of six unilateral PAD patients andthree controls underwent computed tomography (CT)scanning (model CT-900S series, Toshiba Corporation,Nasu, Japan) of each calf. A single 1-cm scan wasobtained simultaneously through both calves at the levelof the biopsy sites with the legs parallel. The perimeterof the calf musculature was manually traced (excludingthe tibia, fibula, and subcutaneous tissue), and thecross-sectional area (in centimeters squared) of themuscle was calculated.

Muscle BiopsyBilateral biopsies of the medial head of the gastroc-

nemius muscles were performed at rest in both legs of

TABLE 2. Muscle Strength, Endurance, and CalfCross-sectional Area in Peripheral Arterial Disease Patientsand Control Subjects

PAD

Unilateral All PAD ControlsMuscle performance

Gastrocnemius strength(foot-pounds)

Diseased 31+11*t 30±9* 53+11Nonsymptomatic 38±14*

Gastrocnemius endurance(no. of repetitions)

Diseased 8±4t 8+5* 13±4Nonsymptomatic 13±5

Anterior tibial strength(foot-pounds)

Diseased 9+4* 9+4* 13+3Nonsymptomatic 12±7

Muscle cross-sectional areaCalf measurement (cm2)

Diseased 94.5±+.18.4t 95.2±t16.2 99.7± 11.9Nonsymptomatic 99.5+20.2

Computed tomography(cm,)Diseased 78.5+6.1t 80.5+14.0Nonsymptomatic 82.6±4.3

PAD, peripheral arterial disease. Muscle strength (peak torqueat 60W/sec) and endurance (50% reduction in peak torque at2400/sec) were measured on a Cybex isokinetic dynamometer.Muscle cross-sectional area was estimated from calf circumferenceand, in a subset of six unilateral patients and three controls, fromcomputed tomography (excluding subcutaneous tissue and bone).Values are mean+SD.

*p<0.05 for the value in patients compared with control subjects.tp<0.05 for the value in the diseased leg of unilateral patients

compared with their nonsymptomatic leg.

the patients and controls. After a subcutaneous injec-tion of 3 ml lidocaine, a 5-mm biopsy needle (Bergstrommuscle biopsy cannula, DePuy Inc., Warsaw, Ind.) wasused to remove approximately 40-50 mg of tissue.Samples to be analyzed for enzyme activities wereimmediately frozen in liquid nitrogen and stored at-80°C. For histological analyses, tissue was stored insaline-soaked gauze immediately after the biopsy for<30 minutes. Subsequently, muscle tissue was orientedin gum tragacanth on a wooden block and flash-frozenin liquid nitrogen-cooled isopentane. The muscle sam-ples were stored at -70°C in closed containers. Inpatients with bilateral PAD, only data from the mostdiseased leg were used for analysis.

Histological AnalysisSerial 10-,um cryostat sections of muscle tissue were

analyzed for both histopathological changes and fiber-type specific characteristics.18 Sections of tissue werestained for myosin ATPase (pH 9.4 and 4.6) and nico-tinamide adenine nucleotide dehydrogenase-tetra-zolium reductase (NADH-TR). These stains were usedfor qualitative analysis, which included a general histo-logical description as well as evaluation for the presenceof target or grouped fibers, which indicate denervationand reinnervation, respectively.19-21 In addition, oil-redO was used to stain for extracellular fat, periodicacid-Schiff for glycogen, Verhoeff-van Gieson for fiber

70

n 60

550'

40

=) 30

Z 20

10

o0

,70

0SO0

0

, 30-

i 20m

= 10w

0

/*

0

8

CONTROL

1

DSEASED NON-symPTOiAnC

+ *A]

.

CONTROL DSEASED NON-SYMPTOMAInc

FIGURE 1. Top panel: Gastrocnemius muscle strength incontrol subjects (o) and patients with unilateral peripheralarterial disease (PAD) (c). Each subject is shown as a point,and for the PAD patients the value in each individual'sdiseased leg and nonsymptomatic leg is connected by a line.*p<0.05 for the value in patients compared with controlsubjects, and +p <0.05 for the value in the nonsymptomaticleg of unilateral patients compared with their diseased leg.Bottom panel: Gastrocnemius muscle type II fiber areameasuredfrom cryostat sections stainedfor myosinATPase atpH 9.4.



ownloaded from





ownloaded from



FIGURE 2. Facingpage. PanelA: Micrograph ofgastrocnemius muscle from a control subject using myosin ATPase stain atpH4.6. Type I (dark), IIA oxidative-glycolytic (light), and IIB glycolytic (intermediate stained) fibers are scattered throughout thesection (bar, 100 gm). Panel B: Micrograph ofgastrocnemius muscle from the diseased leg of a patient with unilateral peripheralarterial disease (PAD) stained for myosin ATPase at pH 4.6. Type I (dark) fibers are clustered into a small group (indicatingreinnervation) in the upper centerportion of the figure (bar, 100 gtm). Panel C: Above. Micrograph ofgastrocnemius musclefromthe diseased leg of a unilateral PAD patient stained with NADH-TR Arrows indicate small, dark angular fibers (indicatingdenervation) scattered between healthy type I and II fibers (bar, 50 gm).

morphology and connective tissue, and acid phospha-tase for necrosis.21The diameter, number, and percent cross-sectional

area of each fiber type were calculated using standardmorphometric techniques from cryostat sections stainedfor myosin ATPase (pH 9.4) activity. The optimalnumber of fields that needed to be sampled was deter-mined by an iterative process that minimized the within-subjects standard deviation of each measurement butwithout oversampling.22 To calculate fiber diameter andnumber, every eighth microscopic field (field size, 0.293mm2) was photographed and projected to a final mag-nification of x364. The diameter of each fiber wasdetermined as the maximum measure of the lesser fiberaxis that would correct for any fibers cut at an obliqueangle. The number of type I and II fibers were countedin the same fields and are expressed as a percent of thetotal number. The percent of angular fibers (indicatingdenervation) was also determined from these fields.The cross-sectional area of each fiber type was esti-

mated using the point-grid method.23 For cross-sec-tional area calculations, every fifth microscopic field(field size, 0.113 mm2) was photographed and projectedonto a grid to a final magnification of x590. The points

(line intersections) falling over type I and II fibers wererecorded for each microscopic field. The total numberof line intersections from all fields of a single cryostatsection for each biopsy were then summed for each fibertype. The data are expressed as the percent of the totalfiber cross-sectional area (sum of the number of pointsfor both type I and II fibers) occupied by each fibertype.

BiochemistrySamples for noncollagenous protein determinations

were prepared as described by Lilienthal et al.24 Proteinquantitation was performed by the method of Lowry etal,25 using bovine serum albumin as a standard. Lactatedehydrogenase,26 citrate synthase,27 and phosphofruc-tokinase28 activities were quantified using establishedspectrophotometric methodologies. Enzyme activitieswere normalized to grams of wet weight and grams ofnoncollagen protein. Citrate synthase activity was alsonormalized to the lactate dehydrogenase activity. Thevariability of the assay methods in our laboratory wasinitially assessed in animal studies. For citrate synthaseactivity, four separate homogenates from two ratsyielded a coefficient of variation (standard deviationdivided by the mean) of 18%.



ownloaded from



Statistical AnalysisStudent's t test for paired data or a within-subjects

ANOVA was used for within-subjects comparisons, andan unpaired t test was used for between-subjects com-parisons. Linear regression was used for calculatingcorrelations between variables. Categorical data wereanalyzed using the X 2or Fisher's exact probability tests.Values are given as mean±SD and considered signifi-cant when p<0.05 in a two-tailed test. In control sub-jects, the functional, histological, and biochemical datawere similar between legs; therefore, the results wereaveraged from both legs.

ResultsPatient Characteristics

Patients with PAD were well matched to the controlsubjects for age but had a reduced peak exercise per-formance as defined by treadmill walking time and peakoxygen consumption (Table 1). Patients reported symp-toms of intermittent claudication for 6±8 years with arange of 1-45 years. In all patients with PAD (unilateraland bilateral subjects), the most diseased leg had anabnormal ABI at rest that decreased further afterexercise (Table 1). The average resting and postexerciseABIs in the unilateral patient's nonsymptomatic legswere in the normal range as previously defined andsignificantly greater than in their diseased legs (Table1). However, postexercise ABIs in the nonsymptomaticlegs were significantly lower than in control subjects,suggesting the possibility of mild arterial disease in thenonsymptomatic leg.

Muscle Strength and EnduranceIn all 26 PAD patients, gastrocnemius muscle strength

in their diseased legs was 43% less and anterior tibialstrength was 31% less than the corresponding values inthe control subjects (Table 2). Gastrocnemius muscleendurance in the diseased legs was 38% less than thatobserved in the control subjects. In the subset of patientswith unilateral disease, gastrocnemius strength (Figure 1,top panel) and endurance were less in the diseased thanin the nonsymptomatic legs, but strength in the anteriortibial muscles was not different between legs (Table 2).

Estimates ofMuscle Cross-sectional AreaThe maximal calf cross-sectional area was estimated

by measurements of circumference and by CT scans ofthe calf muscles. In patients with unilateral PAD, theirdiseased legs had 5% reduction (p<0.05 using a pairedanalysis) in cross-sectional area compared with theirnonsymptomatic legs (Table 2). In contrast, the calfcross-sectional area in diseased legs, measured by eithermethod, was not different from control values due to thelarge standard deviation of the measurement in thePAD patients.

Muscle HistologyHistological analysis of the muscle samples was per-

formed only in the unilateral PAD patients and incontrols. Representative micrographs of gastrocnemiusmuscle are shown as follows: from a control subjectusing myosin ATPase stain (Figure 2A), from thediseased leg of a unilateral PAD subject using myosin

unilateral PAD subject stained with NADH-TR (Figure2C). Qualitatively, the muscle samples from patientsand controls showed no evidence of inflammation, fibernecrosis, or unusual accumulations of intercellular fat,connective tissue, or glycogen. However, all diseasedlegs of unilateral patients had histological evidence ofdenervation consisting of angular (Figure 2C), or target,fibers. In addition, fiber-type grouping, suggesting re-innervation, was seen in five of the 10 diseased legs(Figure 2B). Taken together, histological evidence ofdenervation or reinnervation was observed less often inthe nonsymptomatic legs (40%, p<0.05) or in controllegs (36%, p<0.05) than in the diseased legs (100%).

Fiber type distribution, diameter, and area were alsodetermined in biopsies from the unilateral PAD pa-tients and control subjects. Type I fiber diameter andthe average number of type I fibers per field did notdiffer among diseased, nonsymptomatic, and controllegs (Table 3). Type II diameter and the averagenumber of type II fibers per field were also similar forthe diseased and control legs. However, type II fiberarea in the diseased leg (measured independently ofdiameter or number) was 38% of the total fiber area,which was less than the type II area in the nonsymp-tomatic (49%) and control (48%) legs (Figure 1, bottompanel). In the nonsymptomatic leg, type II fiber diam-eter was less, and type II fiber number was greater thanin the diseased leg. However, type II fiber area in thenonsymptomatic legs was not different from type II fiberarea in the control legs.

Muscle BiochemistryThe noncollagenous protein contents (expressed as

grams per gram wet weight) were similar for all diseasedlegs, the nonsymptomatic legs of unilateral patients, andcontrol subjects (Table 4). Activities of citrate synthase,lactate dehydrogenase, and phosphofructokinase in thediseased legs of unilateral patients did not differ from thevalues in their nonsymptomatic legs or from controlvalues (Table 4), whether normalized to grams of noncol-lagenous protein or grams of wet weight (data notshown). However, there was a large variability in enzymeactivities in the diseased legs of the unilateral patients.To determine if this heterogeneous response was char-acteristic of the PAD population, enzyme activities werealso assessed in the most diseased leg of 16 additionalpatients with bilateral PAD. This analysis confirmed thatthere was no difference for citrate synthase, phosphofruc-tokinase, and lactate dehydrogenase activities betweendiseased legs of PAD patients and control subjects.Finally, when citrate synthase activity was normalized tolactate dehydrogenase activity (to control for anychanges in enzyme expression), there remained no dif-ference between the diseased legs of all PAD patientsand control subjects (data not shown).

Predictors of Muscle Strengthand Exercise PerformanceThe ABI is a marker of the hemodynamic severity of

the vascular disease.29 In the diseased legs of theunilateral PAD patients, the ABI was not correlatedwith any changes in muscle histology or enzyme activi-ties. In all 26 PAD patients, there was no correlation ofthe ABI at rest or after exercise with gastrocnemius

ATPase stain (Figure 2B), and the diseased leg of a muscle strength or peak exercise performance on the



ownloaded from



TABLE 3. Histological Characteristics of Gastrocnemius Musclein Patients With Unilateral Peripheral Arterial DiseaseCompared With Control Subjects

Unilateralperipheralarterialdisease Controls

Muscle fiber characteristicsType II fiber diameter (gm)

Diseased 71±11 76+5Nonsymptomatic 63±+11*t

Type II no. of fibers/fieldDiseased 3.5+1.6 3.7+1.0Nonsymptomatic 5.0+±1.8t

Type II fiber area (% of total fiber area)Diseased 38±13*t 48±8Nonsymptomatic 49+ 11

Type I fiber diameter (,um)Diseased 72+ 12 76±9Nonsymptomatic 74+7

Type I no. of fibers/fieldDiseased 5.8±2.9 4.4+±1.3Nonsymptomatic 5.1±+2.0

Angular fibers (% of total fiber number)Diseased 9.7+1.2*t 1.3±1.6Nonsymptomatic 2.5±6.6

Muscle fiber histology determined from cryostat sectionsstained for myosin ATPase at pH 9.4, with the field size describedin "Methods." Values are mean±SD.

*p<0.05 for the value in patients compared with controlsubjects.

tp<0.05 for the value in the nonsymptomatic leg of unilateralpatients compared with their diseased leg.

treadmill. In the patients with unilateral disease, gas-trocnemius muscle strength was correlated with totalcross-sectional area of the calf (Figure 3, top panel) andtype II fiber area (Figure 3, bottom panel), but similarcorrelations were not observed in the nonsymptomaticlegs. Finally, gastrocnemius strength in the diseased legsof all patients was correlated with peak walking time onthe treadmill (r=0.41, p<0.05, y=0.18x+2.75).

Despite the large population variation in citrate syn-thase activity of all diseased legs in the PAD subjects,there was no correlation between citrate synthase activ-ity and peak treadmill walking time (r= -0.23, p=0.26)or with gastrocnemius muscle strength or endurance inthe diseased legs. Phosphofructokinase and lactate de-hydrogenase activities were also not correlated withmuscle strength or treadmill exercise performance. Incontrast, citrate synthase activity in control legs wascorrelated with peak walking time (r=0.86, p<0.05,y=0.09x+ 13.87).

DiscussionPAD is associated with chronic changes in affected

muscle morphology and function. Histological analysisof gastrocnemius muscle from the diseased legs ofpatients with unilateral PAD demonstrated denervationand a reduction in the cross-sectional area of type IIfibers compared with healthy, age-matched controls.

TABLE 4. Enzyme Activities of Gastrocnemius Muscle inPatients With Peripheral Arterial Disease Patients andControl Subjects

PADUnilateral All PAD Controls

Citrate synthase(1£mol/min/g NCP)

Diseased 64.1±27.6 66.1±37.5 54.2±22.1Nonsymptomatic 70.0±24.7

Phosphofructokinase(,umol/min/g NCP)

Diseased 15.6±4.2 15.2±5.8 17.9± 10.1Nonsymptomatic 17.4±6.7

Lactate dehydrogenase(mmol/min/g NCP)


Noncollagen protein(g/g wet wt)


PAD, peripheral arterial disease; NCP, noncollagen protein.Values are mean±SD. There were no significant differences inenzyme activities between patients and controls.

weakness in the diseased legs, and the muscle weaknesscontributed to the impairment in exercise performance.As another control, results from the nonsymptomaticlegs confirmed that the histological and functionalchanges in the diseased legs were not the result ofdeconditioning, aging, or environmental factors.

Previous studies have shown that PAD is associatedwith alterations in skeletal muscle histology. Necroticand regenerating fibers as well as inflammation havebeen seen in the diseased legs of patients with unilateralPAD hospitalized for surgical evaluation30 and in pa-tients with severe ischemic disease.5 In this study,ambulatory patients with less severe symptoms of clau-dication had no findings of fiber necrosis, inflammation,or accumulations of glycogen, fat, or connective tissue intheir diseased legs, and the muscle fibers appearednormal. Thus, milder forms of the disease are notassociated with generalized morphological changes inskeletal muscle. In contrast, signs of denervation (an-gular fibers) and reinnervation (grouped fibers) occurboth in ambulatory patients with claudication7 and, to agreater degree, in patients with ischemic rest pain organgrene.5,31 In this study, histological evidence of de-nervation was found in all diseased legs of PAD pa-tients. Another study, using electrophysiological tech-niques, has confirmed that the muscle denervation inpatients with unilateral claudication was limited to thedistal motor axons, and the degree of denervation wascorrelated with the severity of the vascular disease.32Thus, skeletal muscle denervation in PAD may be dueto ischemic damage of the distal motor nerves and maycontribute to the muscle weakness and dysfunctionobserved in patients with claudication.Type I and type II skeletal muscle fibers are distin-

guished by twitch characteristics (slow versus fast) andmetabolic profiles (oxidative versus glycolytic).21 Type Ifibers have a high mitochondrial content and are well

These histological changes were associated with muscle adapted for prollonged work. Type II fibers, with their



ownloaded from



60.

-0 50.

440

30

d 20

110

0

-50

i40

30

^ 20

10

r-0.78P(0.05

0

0

0~~~~0

i70 90 100 110TOTAL CROSS-SECTIONAL AREA OF CALF (cm2)

120

r-0.63

P(0.05

a

0

0~~~~~~

O . .

0 1 0 20 30 40

TYPE II FIBERS (X crow-sctional area)50 60

FIGURE 3. Toppanel: Relation between cross-sectional area

of the calfand gastrocnemius muscle strength in the diseasedlegs of patients with unilateral peripheral arterial disease.These two parameters were correlated by linear regression(y=0.46x-10.83). Bottom panel: Relation between the cross-

sectional area of type II fibers from histological analysis andgastrocnemius muscle strength in the diseased legs ofpatientswith unilateral peripheral arterial disease. These two param-eters were correlated by linear regression (y=0.56x+9.70).

low aerobic capacity, fatigue more rapidly. Atrophy ofboth types of fibers has been observed in patients withPAD. Fiber loss and a reduced fiber area are correlatedwith disease severity7'12'3' and can become extensive inpatients with severe limb ischemia.5 Furthermore, isch-emia may result in a greater loss of type II fiberscompared with type I fibers,5-7 but other studies havereported no difference in fiber distribution or diameterbetween diseased and nondiseased legs of unilateralpatients.9"13'30 In our study, enrolling patients with uni-lateral disease controlled for the multiple factors thatmay affect fiber number and diameter, such as age,inactivity, and comorbidity. The results confirm thattype I and II fiber diameter and number were similar fordiseased and control legs (Table 3). However, with aseparate measurement, type II fiber area (which reflectsthe net effect of fiber number and diameter on totalfiber content) was reduced in the diseased legs com-pared with the unaffected and control legs. Consistentwith the findings of decreased percent type II fiber area,it has previously been suggested that type II fibers maybe more adversely affected by ischemia than type Ifibers.33'34Taken together, the findings of denervation and a

selective atrophy of type II fibers may account for thedecrease in skeletal muscle cross-sectional area and themuscle weakness in the diseased legs of unilateral PADpatients. The likelihood of these relations is strength-ened by the correlations of type II fiber area and musclecross-sectional area with muscle strength (Figure 3).

Muscle weakness has been reported in PAD patients,but the decrease in strength was only weakly correlatedwith walking tolerance.8 In our study, gastrocnemiusstrength correlated with peak treadmill walking time,but the low correlation coefficient suggests that otherfactors are more important to walking ability thanmuscle strength in patients with PAD.

In the unilateral patients, the nonischemic histologi-cal state of the individual's muscle was determined intheir nonsymptomatic legs. Although type II fiber diam-eter in patients' nonsymptomatic legs was less than inthe diseased or control legs, the average number of typeII fibers in the nonsymptomatic leg was increasedcompared with the diseased leg (Table 3). Therefore,type II fiber area was not different between the non-symptomatic and control legs. The cause of the de-creased type II fiber diameter in the nonsymptomaticlegs is not known but may reflect changes related toinactivity or other environmental factors distinct fromthe effects of ischemia.

In controls, the activity of skeletal muscle oxidativeenzymes reflects the level of habitual physical activity.For example, exercise training results in an increase inseveral oxidative enzyme activities that is directly re-lated to the improvement in exercise performance.35,36In contrast, deconditioning is associated with decreasedskeletal muscle oxidative enzyme activities.37 In patientswith PAD, increased activities of skeletal muscle oxida-tive enzymes have been observed in some studies,10"1and the amount of increase in citrate synthase activitywas correlated with the hemodynamic severity of theocclusive disease.9 The increase in one oxidative en-zyme, cytochrome oxidase, was only weakly correlatedwith treadmill exercise performance," but in otherstudies changes in oxidative enzyme activities were notcorrelated with exercise performance.9'12

Decreases in oxidative enzyme activity in PAD havealso been reported.12'13 In these studies, the more severethe occlusive disease, the lower the oxidative enzymeactivity.12 A decrease in oxidative enzyme activities withincreasing disease severity may simply reflect loss ofambulation and immobility. Also, denervation of skel-etal muscle (as observed in the patients in this study) isassociated with a decrease in oxidative enzyme activitiesand loss of muscle function.6"18 Finally, the higheroxidative enzyme activities in the control group (com-pared with the PAD patients) could also reflect the levelof fitness in the control group.

In the present study, nonsurgical, ambulatory pa-tients with unilateral PAD were enrolled, with thenonsymptomatic leg providing a control for the effectsof activity and other environmental factors on musclemetabolism. A healthy, sedentary control group allowedfor the determination of any absolute differences be-tween normal subjects and the patients with vasculardisease. The results showed that citrate synthase activityin the diseased legs of unilateral patients was notdifferent from values in their nonsymptomatic legs or inthe control group. Furthermore, the citrate synthaseactivities were very heterogeneous in the unilateralPAD population, ranging from 25 to 115 ,umol/min/gnoncollagen protein. This variability could reflect theassay methods; however, when citrate synthase activitywas normalized to grams of wet weight, grams ofnoncollagenous protein, or lactate dehydrogenase (as a



ownloaded from



marker enzyme not expected to change9-11,13), thereremained no differences between diseased and non-symptomatic or control legs. Finally, given the variabil-ity in enzyme activities, additional patients with bilat-eral PAD were enrolled. The results confirm that PADwas not associated with either an increase or decrease inenzyme activities but that the population displays sub-stantial heterogeneity. This heterogeneity is not unex-pected in ambulatory PAD subjects given the multipleinfluences of vascular disease severity, denervation, andlevel of fitness on enzyme activities, as discussed above.Finally, regardless of any absolute changes in oxidativeenzyme activities previously reported, the functionalsignificance has not been well established. The lack ofcorrelation between citrate synthase activity and musclestrength or exercise performance in the PAD subjectssuggests that any adaptations in oxidative enzymes donot compensate for their muscle dysfunction or de-creased exercise performance. In contrast, the expectedcorrelation between citrate synthase activity and exer-cise performance was observed in the control group.

Thus, the chronic response of skeletal muscle in PADis characterized by denervation and a decreased type IIfiber area. These changes are associated with muscleatrophy and a loss of muscle strength that may contrib-ute to the functional impairment in this population.

AcknowledgmentsWe thank Laura Ruff, Kimberly Masterson, Diana Wolf,

Katherine Barriga, and Robert Ferrill for excellent technicalsupport. Kenneth Schneider produced the figures in their finalform.

References1. Holm S, Bylund-Fellenius AC: Continuous monitoring of oxygen

tension in human gastrocnemius muscle during exercise. Clin Phys-iol 1981;1:541-552

2. Hiatt WR, Nawaz D, Brass EP: Carnitine metabolism during exer-cise in patients with peripheral vascular disease. J AppI Physiol1987;62:2383-2387

3. Pernow B, Zetterquist S: Metabolic evaluation of the leg bloodflow in claudicating patients with arterial obstructions at differentlevels. Scand J Clin Lab Invest 1968;21:277-287

4. Hiatt WR, Nawaz D, Regensteiner JG, Hossack KF: The evalua-tion of exercise performance in patients with peripheral vasculardisease. J Cardiopulmonary Rehabil 1988;12:525-532

5. Hedberg B, Angquist KA, Henriksson-Larsen K, Sjostrom M: Fibreloss and distribution in skeletal muscle from patients with severeperipheral arterial insufficiency. EurJ Vasc Surg 1989;3:315-322

6. Makitie J: Peripheral neuromuscular system in peripheral arterialinsufficiency. Scand J Rheumatol 1979;30(suppl):157-162

7. Hedberg B, Angquist KA, Sjostrom M: Peripheral arterial insuf-ficiency and the fine structure of the gastrocnemius muscle. IntAngiol 1988;7:50-59

8. Gerdle B, Hedberg B, Angquist KA, Fugl-Meyer AR: Isokineticstrength and endurance in peripheral arterial insufficiency withintermittent claudication. Scand J Rehab Med 1986;18:9-15

9. Jansson E, Johansson J, Sylven C, Kaijser L: Calf muscle adapta-tion in intermittent claudication: Side-differences in muscle meta-bolic characteristics in patients with unilateral arterial disease. ClinPhysiol 1988;8:17-29

10. Bylund AC, Hammarsten J, Holm J, Schersten T: Enzyme activi-ties in skeletal muscles from patients with peripheral arterial insuf-ficiency. Eur J Clin Invest 1976;6:425-429

11. Lundgren F, Dahllof AG, Schersten T, Bylund-Fellenius AC: Mus-cle enzyme adaptation in patients with peripheral arterial insuffi-

ciency: Spontaneous adaptation, effect of different treatments andconsequences on walking performance. Clin Sci 1989;77:485-493

12. Clyne CAC, Mears H, Weller RO, O'Donnell TF: Calf muscleadaptation to peripheral vascular disease. Cardiovasc Res 1985;19:507-512

13. Henriksson J, Nygaard E, Andersson J, Eklof B: Enzyme activities,fibre types and capillarization in calf muscles of patients withintermittent claudication. Scand J Clin Lab Invest 1980;40:361-369

14. Rutherford RB, Lowenstein DH, Klein MF: Combining segmentalsystolic pressures and plethysmography to diagnose arterial occlu-sive disease of the legs. Am J Surg 1979;138:211-218

15. Hiatt WR, Marshall JA, Baxter J, Sandoval R, Hildebrandt W,Kahn LR, Hamman RF: Diagnostic methods for peripheral arte-rial disease in the San Luis Valley Diabetes Study. J Clin Epidemiol1990;43:597-606

16. Davies GJ: A Compendium of Isokinetics in Clinical Usage, ed 3.Onalaska, Wisconsin, S & S Publishers, 1987, pp 1-36

17. Hiatt WR, Regensteiner JG, Hargarten ME, Wolfel EE, Brass EP:Benefit of exercise conditioning for patients with peripheral arte-rial disease. Circulation 1990;81:602-609

18. Ringel SP, Wilson WB, Barden MT, Kaiser KK: Histochemistry ofhuman extraocular muscle. Arch Ophthalmol 1978;96:1067-1072

19. Engel WK: Histochemistry of neuromuscular disease- Signifi-cance of muscle fiber types. Proceedings of the VIII InternationalCongress of Neurology, Vienna, Excerpta Medica, 1965, pp 67-101

20. Morris CJ, Raybould JA: Fiber type grouping and end-plate diam-eter in human skeletal muscle. J Neurol Sci 1971;13:181-187

21. Dubowitz V: Muscle Biopsy: A Practical Approach, ed 2. East Sus-sex, UK, Bailliere Tindall, 1985, pp 1-73

22. Muller W: Subsarcolemmal mitochondria and capillarization ofsoleus muscle fibers in young rats subjected to an endurance train-ing. Cell Tissue Res 1976;174:367-389

23. Weibel ER: Stereological Methods. London, Academic Press Inc,1979, pp 63-159

24. Lilienthal JL, Zierler KL, Folk BP, Buka R, Riley MJ: A referencebase and system for analysis of muscle constituents. J Biol Chem1950;182:501-508

25. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measure-ment with the folin phenol reagent. J Biol Chem 1951;193:265-275

26. Vassault A: Lactate dehydrogenase, in Bergmeyer HV (ed): Meth-ods of Enzymatic Analysis: Vol III. Weinheim, Germany, VerlagChemie, 1983, pp 118-125

27. Srere PA: Citrate synthase. Methods Enzymol 1969;13:3-1128. Staal GEJ, Koster JF, Banziger CJM, van Milligen-Boersma L:

Human erythrocyte phosphofructokinase: Its purification andsome properties. Biochim Biophys Acta 1972;276:113-123

29. Kiekara 0, Riekkinen H, Soimakallio S, Lansimies E: Correlationof angiographically determined reduction of vascular lumen withlower-limb systolic pressures. Acta Chir Scand 1985;151:437-440

30. Sjostrom M, Angquist KA, Rais 0: Intermittent claudication andmuscle fiber fine structure: Correlation between clinical and mor-phological data. Ultrastruct Pathol 1980;1:309-326

31. Clyne CAC, Weller RO, Bradley WG, Silber DI, O'Donnell TF,Callow AD: Ultrastructural and capillary adaptation of gastrocne-mius muscle to occlusive peripheral vascular disease. Surgery 1982;92:434-440

32. England JD, Regensteiner JG, Ringel SP, Carry MR, Hiatt WR:Muscle denervation in peripheral arterial disease. Neurology 1992;42:994-999

33. Jennische A: Ischaemia-induced injury in glycogen-depleted skel-etal muscle: Selective vulnerability of FG-fibres. Acta PhysiolScand 1985;125:727-734

34. Sjostrom M, Neglen P, Friden J, Eklof B: Human skeletal musclemetabolism and morphology after temporary incompleteischaemia. Eur J Clin Invest 1982;12:69-79

35. Holloszy JO, Coyle EF: Adaptations of skeletal muscle to endur-ance exercise and their metabolic consequences. J Appl Physiol1984;56:831-838

36. Coggan AR, Spina RJ, Rogers MA, King DS, Brown M, NemethPM, Holloszy JO: Histochemical and enzymatic characteristics ofskeletal muscle in master athletes. JAppl Physiol 1990;68:1896-1901

37. Chi MMY, Hintz CS, Coyle EF, Martin WH, Ivy JL, Nemeth PM,Holloszy JO, Lowry OH: Effects of detraining on enzymes ofenergy metabolism in individual human muscle fibers.Am JPhysiol1983;244:C276-C287



ownloaded from


R HiattJ G Regensteiner, E E Wolfel, E P Brass, M R Carry, S P Ringel, M E Hargarten, E R Stamm and W

Chronic changes in skeletal muscle histology and function in peripheral arterial disease.

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1993 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.87.2.413

1993;87:413-421Circulation.

http://circ.ahajournals.org/content/87/2/413World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer available in the

Permissions in the middle column of the Web page under Services. Further information about this process isOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Circulation Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:



ownloaded from

http://circ.ahajournals.org/content/87/2/413

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


chronic changes in skeletal muscle histology and function in

Documents