effect of exercise training on skeletal muscle histology ......muscle metabolism. similarly,...

Effect of exercise training on skeletal muscle histology and metabolism in peripheral arterial disease

WILLIAM R. HIATT, JUDITH G. REGENSTEINER, EUGENE E. WOLFEL, MICHAEL R. CARRY, AND ERIC P. BRASS Division of Cardiology, Department of Medicine, Section of Vascular Medicine, and Department of Neurology, University of Colorado Health Sciences Center, Denver, Colorado 80262; and Department of Medicine, Harbor- University of California, Los Angeles Medical Center, Torrance, California 90509

Hiatt, William R., Judith G. Regensteiner, Eugene E. Wolfel, Michael R. Carry, and Eric P. Brass. Effect of exercise training on skeletal muscle histology and metabolism in peripheral arterial disease. J. AppZ. PhysioZ. N(2): 780-788, 1996.-Patients with symptomatic peripheral arterial occlusive disease have a claudication-limited peak exercise performance that is improved with exercise training. The effects of training on skeletal muscle metabolism were evaluated in 26 patients with claudication, randomized into a 12-wk program of treadmill training (enhances muscle metabolic activity in normal subjects), strength training (stimu- lates muscle hypertrophy in normal subjects), or a nonexercising control group. Gastrocnemius muscle biopsies were performed at rest and before and after training. After 12 wk, only treadmill training improved peak exercise performance and peak oxygen consumption. Treadmill training did not alter type I or type II fiber area and did not increase citrate synthase activity but was associated with an increase in the percentage of denervated fibers (from 7.6 t 5.4 to 15.6 t 7.5%, P < 0.05). Improvement in exercise performance with treadmill training was associated with a correlative decrease in the plasma (r = -0.67) and muscle (r = -0.59) short-chain acylcarnitine concentrations (intermediates of oxidative metabolism). Patients in the strength and control groups had no changes in muscle histology or carnitine metabolism, but strength-trained subjects had a decrease in citrate synthase activity. Thus treadmill training increased peak exercise performance, but this benefit was associated with skeletal muscle denervation and the absence of a “classic” mitochondrial training response (increase in citrate synthase activity). The present study confirms the relationship between skeletal muscle acylcarnitine content and function in peripheral arterial occlusive disease, demonstrating that the response to treadmill training was associated with parallel improvements in intermediary metabolism.

peripheral vascular diseases; exercise; enzymes; carnitine; muscle denervation

PATIENTS WITH PERIPHERAL arterial occlusive disease (PAOD) have atherosclerotic occlusions in the major arteries supplying their lower extremities. The arterial disease results in a restricted blood flow to skeletal muscle, particularly during exercise. During walking exercise, PAOD patients develop muscle ischemia and the symptom of intermittent claudication that results in a substantial impairment in walking ability (10). Thus the exercise impairment limits the ability of PAOD patients to perform activities essential for daily living (26).

In patients with claudication, the hemodynamic severity of the underlying vascular disease (defined by peripheral blood flow or ankle pressure) is not well correlated with treadmill performance (11). Therefore, in addition to the limited blood flow, other factors must contribute to the functional impairment in patients with PAOD. Previous studies have shown that patients with PAOD developed several histological, neurological, and metabolic changes in the skeletal muscle of their affected legs. These changes included alterations in fiber type distribution and denervation of distal motor nerve axons, which were associated with skeletal muscle weakness (27).

Muscle ischemia may also adversely affect muscle metabolism. Patients with claudication accumulate lactate and intermediates of oxidative metabolism [short-chain (SC) acylcarnitines] in ischemic skeletal muscle at rest (14). Importantly, there was an inverse correlation between muscle SC acylcarnitine content in the diseased leg at rest and subsequent assessment of peak exercise performance (14). Also, in PAOD patients, muscle oxidative enzyme activities were directly correlated with treadmill exercise performance (19,24). Thus muscle enzyme activities and the status of the muscle carnitine pool may be considered indexes of functional disease severity and may serve as markers of the metabolic responses with exercise training.

Exercise rehabilitation, on the basis of walking exercise, is recognized as an important treatment strategy to improve exercise performance and community-based functional status in patients with claudication (13,26). When subjects were tested on a graded-treadmill protocol, exercise training increased claudication-limited peak oxygen consumption (VO,) 20-30%, suggesting an improvement in either oxygen delivery or oxygen utili- zation in skeletal muscle (12, 13). Exercise training in PAOD has been associated with an increase in muscle oxidative enzyme activities, suggesting improved muscle metabolism as a possible mechanism of the training response (4, 16). Exercise training also improves leg oxygen extraction in PAOD (30). An increase in oxygen extraction may be related to a greater number of capillaries per fiber (enhanced by training under isch- emit conditions), which would facilitate the diffusion of oxygen to muscle mitochondria (29,33).

Strength training is a potential alternate rehabilitation strategy that in normal subjects may improve exercise performance by hypertrophy of muscle fibers and increased strength (8). Given the previous findings of muscle weakness in patients with PAOD (27), strate-

780 0161-7567/96 $5.00 Copyright o 1996 the American Physiological Society

EXERCISE TRAINING IN PERIPHERAL ARTERIAL DISEASE 781

gies to improve the strength of the muscle groups in the lower extremities critical for walking were employed in the present study. On the basis of these concepts, this study challenged the hypothesis that, in patients with claudication, an increase in peak exercise performance with a treadmill-walking exercise program would be associated with evidence of an enhancement in skeletal muscle metabolism. Similarly, improvements in exercise performance with strength training would be associated with an increase in muscle strength and fiber hypertrophy. This hypothesis was tested by evaluating the effects of 12 wk of treadmill training or strength training on skeletal muscle function, histology, enzyme activities, and metabolic state as assessed by acylcarnitine metabolism.

METHODS

Patients. All subjects recruited for this study had intermittent claudication, defined as pain in the calf, thigh, or buttocks that limited walking ability and was relieved by rest. Claudication severity by history was stable over the previous 3 mo and was the limiting symptom during community-based activities and during treadmill exercise in the laboratory. PAOD was confirmed by an ankle-to-brachial systolic blood pressure ratio (ABI) of CO.90 at rest that decreased to ~0.80 after exercise. Patients were excluded who had ischemic rest pain, ulceration, or gangrene. Previous vascular surgery or peripheral angioplasty in the past year were also causes for exclusion. Patients were not enrolled who were unable to walk on the treadmill at a speed of at least 2 mph or whose exercise capacity was limited by symptoms of angina, congestive heart failure, chronic obstructive pulmo- nary disease, or arthritis. Diabetics were excluded because hyperglycemia has been suggested to adversely affect the response to a training program (28). Patients taking chronic medications were continued on their drugs, but the dosages were not changed during the study. The study was approved by the University of Colorado School of Medicine Human Subjects Committee, and informed consent was obtained from all enrolled subjects.

The demographics of the study population have been previously reported (13). Twenty-nine patients were enrolled and randomized into the study, of whom 26 consented to muscle biopsies on entry and after 12 wk of treatment. Patients in the three treatment groups were of similar age and weight and had similar cardiovascular risk factor profiles (Table 1). Patients in the strength-training group had fewer years of symptomatic claudication (by self-report) than did patients in the other two groups. However, the resting and postexercise ABI values in the more severely diseased leg were similar among the three groups.

Design. All patients underwent initial testing and were randomized to one of two treatment groups or a control group. Control subjects were instructed to maintain their usual level of activity, whereas treated subjects were enrolled in a 12-wk program of either supervised treadmill-walking exercise or strength training. The focus of this study is the change in skeletal muscle histology and metabolism that occurred after 12 wk. A second phase of the study involved a crossover into different treatment programs, but no additional muscle biopsies were performed. The details of the training regimens and the performance and functional results from the complete study have been previously reported (13,26).

Hemodynamic assessment. At rest and after graded- treadmill exercise, the ABI was measured as previously

Table 1. Patient characteristics on entry

Treadmill Strength Control

Demographics Number 10 8 8 Age, Yr 67?7 6756 6725 Weight, kg 71.5 + 12.3 70.2 7t 8.8 83.5 t 19.0

Risk factor Smoking, pack - yr 59k34 47?22 39+27 Hypertension, % 60 38 25 Hyperlipidemia, % 20 50 13 Diabetes, % 0 0 0

Vascular disease status Years of claudication 4t2 2 k 14: 422 Prior vascular surgery, % 30 0 50 Worse leg

Resting ABI 0.56 5 0.16 0.515 0.27 0.61 IT 0.13 Postexercise ABI 0.19 + 0.10 0.17 + 0.15 0.28 5 0.09

Values are means + SD. packmyr, Packs/day x years of smoking. Twenty-six patients with claudication were randomized to treadmill training, strength training, or a control group. Worse leg refers to limb in each patient with more severe disease as assessed by resting and postexercise ankle-brachial (ABI) indexes. *P < 0.05 for fewer years of claudication in strength group vs. treadmill and control groups.

described (12,13). The leg with “worse”disease was defined as having the lowest resting and postexercise ABI.

?FeadmiZZ testing. Subjects were initially familiarized with the treadmill, with a second test on a subsequent day used for data analysis. All subjects were tested with a graded- treadmill protocol that has been previously described (11,12). Treadmill speed was held constant at 2 mph, and the work rate began at 0% grade and increased 3.5% every 3 min to maximal claudication pain. During exercise, heart rate (by 12-lead electrocardiograph) and arm blood pressure were monitored every minute. The time of onset of claudication pain was recorded when the patient first noticed calf pain during exercise. Peak exercise time was defined as the time on the treadmill at which the patient could no longer continue exercise because of severe claudication pain.

At rest and during exercise, ~oZ and CO2 production (ho2> rates were measured continuously (Ametek/Thermox Instru- ments, Pittsburgh, PA). The instrument was calibrated with known concentrations of 02 and CO2 before each test. Peak vo2 was determined from the average of the final minute of data from the graded-treadmill protocol.

Strength testing. Strength testing of the gastrocnemius muscle was performed supine with the leg extended on a Cybex dynamometer (Lumex, Ronkonkoma, NY). Joints not being tested were stabilized to prevent recruitment of other muscle groups. Each subject flexed and extended the foot with maximal effort on a pedal at a regulated speed of 6O”/s for a total of five repetitions. This sequence was performed a second time, and the single maximal value for each muscle group (expressed in foot-pounds of peak torque) was used for analysis.

?Faining programs. Subjects randomized to the treadmill- walking exercise program were trained in a hospital setting three times per week for 1 h as previously described (12, 13). The training consisted of repetitive walking on a treadmill at a rate and grade that produced moderate claudication pain after 3-5 min of exercise, followed by a rest period to allow claudication pain to subside, then another series of exercise- rest periods. The intensity of the treadmill-training program was advanced on a weekly basis as tolerated by increasing both walking speed and grade.

Supervised strength training was conducted three times per week by a physical therapist and involved resistive

782 EXERCISE TRAINING IN PERIPHERAL ARTERIAL DISEASE

isotonic training of five specific muscle groups in each leg as previously described (13). The training program consisted of applying a resistance to each muscle group with a cuff weight secured to the appropriate part of the leg. Subjects performed six contractions of each individual muscle group, followed by a rest period, and then repeated this circuit for a total of three sets per session per leg. Every 2 wk the exercise prescription was advanced as tolerated.

BZood coLLection and preparation. At rest, three samples of blood were drawn 5 min apart for the analysis of carnitine and acylcarnitines, with the average results reported. Each 5ml sample of blood was immediately placed in a heparinized tube on ice. The blood was centrifuged in chilled tubes at 600 g for 3 min, with plasma aliquots stored at -80°C until subsequent analysis.

Muscle biopsy. At rest, all patients underwent bilateral biopsies of the medial head of the gastrocnemius muscles. The skin was anesthetized with lidocaine, following which a 5-mm biopsy needle (Bergstrom muscle biopsy cannula, DePuy, Warsaw, IN) was used to remove -50-75 mg of tissue. Samples for analysis of enzyme activities and carnitine contents were immediately frozen in liquid nitrogen and stored at -80°C. For histological analyses, the muscle tissue was oriented in gum tragacanth on a wooden block, flash- frozen in liquid nitrogen-cooled isopentane, and stored at -80°C in closed containers.

Muscle histology. Serial lo-urn cryostat sections of muscle tissue were analyzed both for histopathological changes and for fiber type-specific characteristics. Sections of tissue were stained for myosin adenosinetriphosphatase (pH 9.4 and 4.6) and NADH-tetrazolium reductase. These stains were used for qualitative analysis, as well as for evaluation of the presence of angular and target fibers (indicating denervation) or grouped fibers (evidence of reinnervation) (5, 25). The slides were counterstained with eosin for the determination of connective tissue.

The determination of the optimal number of fields needed to be sampled for quantitative analysis has been previously described (27). The cross-sectional area of each fiber type and the amount of connective tissue were estimated by using the point-grid method from cryostat sections stained for myosin adenosinetriphosphatase (pH 9.4) activity. For cross-sectional area calculations, every fifth microscopic field (field size, 0.113 mm2> was photographed and projected to a final magnification of x590 onto a grid. The points (line intersections) falling over type I fibers, type II fibers, and connective tissue were recorded for each microscopic field, and the total number of line intersections from all fields of a single cryostat section were then summed. The data for fiber type and connective tissue area are expressed as the percentage of the total area, which is the sum of the total number of points used for analysis.

Angular and target fiber percentages were determined from every eighth microscopic field (field size, 0.293 mm2> projected to a final magnification of ~364. The number of angular or target fibers was counted in each field and expressed as a percentage of the total number of muscle fibers in the field. The surround factor was determined by counting the number of type I fibers and type II fibers contiguous with a type I fiber. The number of type I fibers was multiplied by one, and the number of type II fibers was multiplied by two. The values were then added, and the sum was divided by the number of fibers. An equal number of type I and type II fibers surrounding a type I fiber would give a surround factor of 1.50. Values deviating from the expected 1.50 indicate fiber type grouping, a measure of reinnervated fibers (18).

Biochemistry. Lactate dehydrogenase (34), citrate synthase (a mitochondrial enzyme) (3 l), and phosphofructokinase (a glycolytic enzyme) (32) activities were quantified by using established spectrophotometric methodologies. Noncollag- enous protein (NCP) was determined by the method described by Lilienthal et al. (22). Protein quantitation was performed by the method of Lowry (23) by using bovine serum albumin as a standard. Enzyme activities were normalized to NCP.

Carnitine was measured by a radioenzymatic assay as previously described (2). Carnitine was directly assayed in the perchloric acid supernatant that contains carnitine and SC acylcarnitines. The supernatant was then subjected to alkaline hydrolysis to convert SC acylcarnitine to carnitine, and carnitine was again measured to determine total acid- soluble (TAS) carnitine. SC acylcarnitine concentration (acyl groups < 10 carbon atoms) was derived from the difference in the concentrations of TAS carnitine and carnitine. The ratio of SC acylcarnitine concentration to TAS carnitine concentration (SUTAS) provides a useful marker of changes in carnitine metabolism (2). Long-chain acylcarnitine concentration (acyl moiety of 10 or more carbons) was measured as carnitine after alkaline hydrolysis of the perchloric acid pellet. Total carnitine was the sum of the carnitine, SC acylcarnitine, and long-chain acylcarnitine concentrations. All assays were performed in duplicate, with the average results reported. In plasma, the carnitine results are expressed in micromoles, and in muscle as nanomoles per gram wet weight.

StatisticaL analysis. A between-subjects analysis of vari- ance was used to compare groups on entry. Student’s t-test was used for paired data analyses before and after the treatment programs. Linear regression was used for calculat- ing correlations between variables. Categorical data were analyzed by using the x2 test. Values are given as means * SD and considered significant when P < 0.05 in a two-tailed test.

RESULTS

Characterization of skeletal muscle in PAOD. This study provided the opportunity to characterize gastrocnemius muscle biopsies from a population of PAOD patients before they entered the training programs (Table 2). The histological status of the two legs was similar in terms of fiber type area and grouping (surround factor). The percentage of angular fibers tended to be greater in the leg with worse disease (P = O.OSS), consistent with previous reports (27). The activities of phosphofructokinase and lactate dehydrogenase were similar, but citrate synthase activity was increased in the leg with worse disease. In the leg with worse disease, the resting ABI was inversely correlated with the citrate synthase activity such that patients with the lowest ABI had the highest citrate synthase activity (Fig. 1). The t o a muscle carnitine content of the two t 1 legs was also similar, but there was an accumulation of SC and long-chain acylcarnitines in the more severely affected legs, and the SUTAS was increased in the legs with worse disease, as previously reported (14).

Effect of training on exercise performance. Twelve weeks of treadmill training increased the time to onset of claudication pain, peak walking time, and peak vo, but had no effect on gastrocnemius muscle strength (Table 3). All t readmill-trained patients had an increase in peak walking time, ranging from 2.5 to 12.6


Table 2. Hemodynamic, muscle histological, and metabolic differences between more and less diseased legs on entry

Worse Less

Hemodynamics (n = 26) Resting ABI 0.56 + 0.19 Exercise AI31 0.212 0.12

Histology (n = 24) Type I fiber area, % 52.8+ 16.9 Type II fiber area, % 40.8 If: 17.8 Connective tissue area, % 6.3k2.7 Angular fibers, % 7.5 26.6 Target fibers, % 6.2 + 6.1 Surround factor 1.42 5 0.17

Enzyme activity (n = 25) Citrate synthase, umol . min-l l g

NCP-’ 79.7 562.0 Phosphofructokinase, umol l rnin-l. g

NCP- l 15.7 t 6.2 Lactate dehydrogenase,

mmol l min -- la g NCP- l 0.5520.29 Carnitine content (n = 26)

Carnitine, nmol/g wet wt 2,600?700 Short-chain acylcarnitine, nmol/g

wet wt 580+400 SC/TAS 0.18 + 0.12 Long-chain acylcarnitine, nmol/g

wet wt 80+60 Total carnitine, nmol./g wet wt 3,330*730

0.84+0.21* 0.59+0.28"'

51.8 -+ 12.5 42.5512.0

5.9k3.2 4.2 24.8 4.8 + 6.2

1.43kO.12

59.3 + 31.0:‘:

14.4k5.1

0.57 kO.28

2,690+1,040

400*350* 0.14+0.10*

60+50* 3,150* 1,190

Values are means ? SD; n, no. of subjects. Gastrocnemius biopsies were obtained at rest in worse and less diseased legs in gastrocnemius muscle. Histological and assay methods are described in text. Two subjects were excluded from histological analysis and 1 from enzyme assay because of tissue loss during preparation. SCYI’AS is ratio of short-chain (SC) acylcarnitine to total acid-soluble (TAS) carnitine. NCP, noncollagenous protein. *P < 0.05 for worse vs. less diseased legs.

min over baseline values. Eight of ten patients also had . an increase in peak VO, (ranging from 0.7 to 7.6 ml. kg -lernin-9, but two had a decrease in peak VO, of 0.7 and 1.6 ml-kg-lemin-l. Patients in the strength- training program had an increase in muscle strength but no improvement in treadmill-exercise duration or

c 300 T 0

.OO 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80

Ankle-Brachial Index

Fig. 1. Relationship between resting ankle-brachial index and citrate synthase activity in leg with worse disease (r = -0.54, P < 0.01; y = -172.8x + 175.8). Each point represents an individual patient. NCP, noncollagenous nrotein.

Table 3. Results of training programs

Entry 12 wk

Claudication onset, min Treadmill 2.6 -+ 1.6 6.0 24.8'" Strength 2.620.9 2.9 + 1.1 Control 3.822.3 3.12 1.3

Peak walking time, min Treadmill 9.325.6 15.1+6.8* Strength 6.523.1 8.72 5.5 Control 7.4 23.3 7.3k2.7

PeakVoz, ml.kg-l*minl Treadmill 14.9 5 2.7 17.4 -+4.6* Strength 14.9 5 3.1 14.6 ? 4.4 Control 14.3 ? 2.3 13.9 + 3.5

Worse leg strength, ft-lb. Treadmill 24.2 + 11.1 27.9 2 9.7 Strength 28.82 7.9 33.12 10.0* Control 30.6k6.9 33.3 k11.4

Values are means + SD. Patients were treated with 12 wk of treadmill training, strength training, or no exercise (control). Peak exercise performance was assessed on a graded-treadmill protocol, with patients stopping exercise because of maximal claudication pain. Gastrocnemius strength testing was performed on a Cybex dynamometer. VOW, 02 uptake. *P < 0.05 for entry vs. 12 wk.

peak VO,. Control patients had no change in any performance measurement.

Changes in skeletal muscle histology with training. In the leg with more disease, 12 wk of treadmill training, strength training, or nonexercising control did not alter type I or type II fiber area (Table 4). However, there was a trend (P = 0.061) for an increase in the amount of connective tissue in patients who

Table 4. Changes in skeletal muscle histology with training

Entry 12 wk

Fiber area Type I area, %

Treadmill 52.027.6 51.4 + 6.8 Strength 47.1 t 19.5 49.3 + 16.1 Control 57.0 t 16.6 59.4k24.4

Type II area, % Treadmill 43.41r8.4 41.7 + 7.4 Strength 47.Ok20.4 44.7516.1 Control 35.42 18.4 34.lk24.6

Connective tissue, % Treadmill 4.721.4 7.022.4 Strength 6.0~12.3 5.8 + 1.9 Control 7.1+ 2.9 6.623.1

Denervation Angular fibers, %

Treadmill 7.6k5.4 15.6 + 7.5* Strength 8.826.6 6.8 k6.2 Control 9.15 8.4 9.1? 6.5

Target fibers, % Treadmill 3.0 k3.3 7.7 56.8:': Strength 7.5 2 6.5 2.223.1 Control 8.3 + 7.6 10.6 + 5.8

Grouped fibers Surround factor

Treadmill 1.46 + 0.09 1.412 0.10 Strength 1.47 -t 0.14 1.47 t 0.21 Control 1.34 + 0.22 1.39 t 0.28

Values are means 5 SD. Skeletal muscle in worse leg was evaluated for fiber type area and evidence of denervated and grouped fibers. *P < 0.05 for entrv vs. 12 wk.


+*

0’ EN EX EN EX EN EX TREADMILL STRENGTH CONTROL

B 40

35 0 /

L 2 5

30--

sz 25--

5

t

+* + + +

0' EN EX EN EX EN EX TREADMILL STRENGTH CONTROL

Fig. 2. Changes in citrate synthase activity (A) and phosphofructokinase activity (B) are shown for worse diseased leg in treadmill, strength, and control groups. Each individual is represented by a dot on entry (EN; at 0 wk) connected by a line to corresponding exit value (EX; at 12 wk). Mean values are represented by horizontal bar. *P < 0.05 for exit vs. entry values.

received treadmill training. Patients in the treadmill- trained group had a doubling of the percentage of angular and target fibers (both P < 0.05) from the exercise program, whereas there was no increase in these denervated fibers in the strength or control groups. The surround factor did not change in any of the groups over 12 wk.

In the leg with less disease, type II fiber area decreased after 12 wk of treadmill training (from 47.2 t 8.9 to 41.4 t 9.1%, P < 0.05), but there were no other changes in fiber area or connective tissue in any of the groups (data not shown). Importantly, there was no increase in angular or target fibers and no change in the surround factor in the leg with less disease in any group with training (data not shown).

Effect of training on skeletal muscle enzyme activities. Muscle enzyme activities, as assessed on entry and exit, demonstrated substantial intersubject heterogene- ity (Fig. 2). Treadmill training had no effect on citrate synthase activity but increased phosphofructokinase activity by 25% from 15.0 2 5.6 to 18.8 t 7.5 pmol min-l l g NCP l (P < 0.05). Lactate dehydrogenase activity was unchanged after treadmill training (0.52 t 0.22 mmolmin-l*g NCP-l on entry to 0.58 ? 0.34 mmol.mir+ l g NCP-l on exit). Strength training was not associated with changes in phosphofructoki-

nase activity, but citrate synthase activity decreased 24% from 84.7 t 41.3 to 64.5 2 33.3 pmolmir+g NCP-l (P < 0.05). Lactate dehydrogenase activity was unchanged after strength training (0.65 5 0.41 mmol minl l g NCP-l on entry to 0.79 * 0.43 mmolmir+g NCP-l on exit). In the control group, there were no changes in citrate synthase or phosphofructokinase activities after 12 wk (Fig. 2) or in lactate dehydrogenase activity (0.47 t 0.19 mmol~min+g NCP-l on entry to 0.63 ? 0.42 mmol l min-l l g NCP-l on exit). In the leg with less disease, there were no changes in the three measured enzyme activities over 12 wk in any group*

Effect of training on carnitine metabolism. The plasma concentrations of carnitine and acylcarnitines were measured on entry in the three treatment groups. After 12 wk, there was no change in any of these mean values in any group. However, consistent with previous reports (12), there was an inverse relationship between change in the plasma SCYI’AS and increase in peak walking time (r = -0.67, P < 0.05; y = -42.7x + 6.1).

There were no changes in the muscle contents of carnitine or acylcarnitines over 12 wk, except for a decrease in SC acylcarnitine content in the strength- trained group (Table 5). This decrease was not associated with any changes in treadmill-exercise performance and was not associated with a decrease in the SC/TAS. In the treadmill-trained group, on the basis of the previously defined relationships between carnitine homeostasis and functional status (12,14), associations between changes in muscle carnitine metabolism and improvements in exercise performance were evaluated. Seven patients improved peak VO, with training and had a decrease in resting muscle SC acylcarnitine content (Fig. 3). The two patients who decreased their

Table 5. Changes in muscle carnitine metabolism with training

Entry 12 wk

Carnitine, nmol/g wet wt Treadmill Strength Control

SC acylcarnitine, nmol/g wet wt Treadmill Strength Control

SUTAS Treadmill Strength Control

Long-chain acylcarnitine, nmol/g wet wt

Treadmill Strength Control

Total carnitine, nmol/g wet wt Treadmill Strength Control

2,320+630 2,130+1,100 2,550+660 2,3402840 2,990+740 2,330+740

5602390 4302610 73Ok400 360+270* 45Ok420 150_+80

0.19 + 0.12 0.15+0.12 0.23kO.14 0.13 20.06 0.12+0.09 0.06t0.03

90270 80+110 60260 50220 70+50 50530

3,140t640 2,720t1,520 3,340?340 2,750?1,030 3,540?1,060 2,600+780

Values are means + SD. Skeletal muscle in worse leg was assayed for carnitine and acylcarnitine contents. *P < 0.05 for entry vs. 12 wk.


Muscle Short-Chain Acylcarnitine (nmollg)

Fig. 3. Relationship betw-een muscle short-chain acylcarnitine content and peak 02 uptake (VO,) in treadmill-trained subject’s worse leg before and after training. Entry value of each individual is represented by a closed circle and is connected to corresponding exit value, represented by an arrow to an open circle. There was an inverse correlation for variables using 20 data points (r = -0.59, P < 0.01).

peak VO, with training had an increase in muscle SC acylcarnitine content. Only one patient had a modest increase in muscle SC acylcarnitine content and an increase in peak VO,. These observations are consistent with the concept that the status of the muscle carnitine pool is predictive of exercise performance and accumulation of acylcarnitines is a marker of metabolic dysfunction (14). When the entry and exit treadmill group were combined, there

data from the was an inverse

relationship between the resting muscle SC acylcarnitine content and exercise performance (Fig. 3, r = -0.59, P < 0.01).

DISCUSSION

Although the clinical benefits of exercise rehabilitation for claudication have been well established, the mechanism of improvement has not been fully eluci- dated. Because exercise training thasn ot been shown to significantly increase leg blood flow ( 12, 30), the present study was designed to test the hypothesis that tread .mill training would improve skeletal muscl .e metabolic function. The resu .lts demonstrated that the increase in claudication-limited peak Voa with treadmill training was associated with an improvement in intermediary metabolism, as reflected in acylcarnitine accumulation.

Effect of PAOD on skeletal muscle. Over time, patients with chronic PAOD acquire several neurological and metabolic abnormalities in the skeletal muscle of their affected legs that may have functional significance. For example, motor nerve denervation has been associated with skeletal muscle weakness (27), low- oxidative enzyme activities were associated with decreased exercise performance (24), and the accumulation of acylcarnitines (intermediates of oxidative metabolism) were associated with reduced exercise performance (14). The present study confirmed that the greatest degree of metabolic and histological alterations w as found in the patie nts’ more severely dis-

eased legs (Table 2). In a previous study, the prevalence of angular fibers in gastrocnemius muscle of healthy older controls was ~2% (27). In the present study, angular fibers were 4.2% in the leg with less disease and tended to be greater (7.5%) in the leg with more disease. Thus skeletal muscle denervation is a feature of PAOD, representing local neurological injury to ischemic muscle.

Citrate synthase activity was elevated in the leg with more disease relative to the leg with less disease (Table 2) and was inversely correlated with disease severity (as defined by the ABI). Although these results are consistent with those from some previous studies (19, 24), other authors have observed no enhancement in oxidative enzyme activities in PAOD (9, 27). The increased citrate synthase activity in the worse leg, the inverse correlation with hemodynamics, and the rela- tivr ‘y large number of subjects evaluated support a PALD-associated metabolic adaptation in the diseased limb. However, the functional impact of the enhanced citrate synthase activity remains unclear. Increased muscle mitochondrial content (reflected by citrate synthase activity) has been suggested as a positive adaptation to increase muscle oxidative capacity and decrease mitochondrial diffusion distances (15).

Carnitine interacts with the cellular and mitochondrial acyl-CoA pool to form acylcarnitines. Changes in the distribution between unesterified carnitine and acylcarnitines provide a marker of changes in cellular metabolism (1, 2). Patients with PAOD accumulate acylcarnitines in their plasma and ischemic skeletal muscle (10,14). This accumulation can be lateralized to the leg with more severe occlusive disease (Table 2; Ref. 14), and patients with the highest levels of acylcarnitines have the most limited exercise performance (10,14).

Histological changes with training. In healthy adults, endurance (aerobic) training does not change muscle fiber size but is associated with a shift from glycolytic to oxidative fibers (15). In the present study, treadmill training did not increase type I fiber area. Although type IIa and IIb fibers were not assessed, the lack of increase in citrate synthase activity is consistent with no change in oxidative fibers with treadmill training. In healthy adults, strength training increases the area of both type I and type II fibers (8). Although strength training increased muscle strength in the PAOD subjects, the resistive exercises prescribed were not associated with fiber hypertrophy. Thus changes in fiber distribution, fiber area, and muscle strength are not a critical component of the training response in patients with claudication. Importantly, capillary density could not be assessed in the present histological analysis. Changes in capillary density might contribute to the functional improvements with training (29).

Patients randomized to the strength and control groups did not experience an increase in the frequency or intensity of their claudication during the 3 mo of training or observation. These patients also had no change in the percentage of angular or target fibers in their skeletal muscle. In contrast, the design of the


treadmill-training program was to encourage patients to exercise to moderate levels of ischemic claudication pain. The increase in muscle ischemia in this group was associated with a marked increase in the percentage of denervated angular and target fibers and a trend toward an increase in the amount of connective tissue in their ischemic skeletal muscle. Thus denervation and loss of muscle fibers are a feature of PAOD and are accelerated by treadmill training (7, 27). However, there was no loss of gastrocnemius muscle strength in the treadmill-trained group, and peak exercise performance was markedly improved by the training program, suggesting little functional significance of the denervation changes during the period of observation. Over the 12 wk, there was no change in the number of grouped fibers (surround factor). These fibers reflect reinnervation of previously denervated fibers, and the process is relatively slow (6,lS). Thus compensation for the denervation injury may occur but would require longer observation periods to detect.

Metabolic changes with training. In healthy individu- als, endurance training is associated with a large increase in skeletal muscle mitochondrial number and density and increases in oxidative enzyme activities (3, 15). The improvement in muscle oxidative capacity is associated with increased exercise performance (15) and thus is an expected and functionally important component of the training response. In healthy adults, the increase in muscle enzyme activities can be greatly augmented by training under conditions of restricted blood flow (20,33).

In contrast to the responses in healthy subjects, treadmill-trained PAOD subjects had no change in gastrocnemius muscle citrate synthase activity. Previ- ous studies in PAOD have shown that training increased the activities of succinic oxidase and cyto- chrome oxidase but did not change citrate synthase activity (4,24). Compared with the responses in normal subjects under conditions of both normal and restricted blood flow, the failure to increment citrate synthase activity with training may be due to the concomitant denervation, which has been shown to decrease the activities of a variety of oxidative enzymes (35). In addition to denervation, patients with PAOD may have a number of other insults to oxidative metabolism because muscle bioenergetics (as assessed by magnetic resonance spectroscopy) are abnormal and display a pattern consistent with a mitochondrial myopathy (21).

In healthy subjects, endurance training is not associated with an increase in the activities of glycolytic enzymes (phosphofructokinase) or lactate dehydrogenase (15). However, under conditions of cuff-induced limitation of blood flow, normal subjects demonstrated an increase in phosphofructokinase activity (20, 33). Subjects with PAOD also increased phosphofructokinase activity with treadmill training. The significance of this glycolytic response is unknown but may reflect a large dependence on glucose instead of fatty acids under conditions of limited oxygen delivery.

In healthy subjects, strength training is not associated with an increase in glycolytic or oxidative enzyme

activity (17). In the PAOD subjects, the decrease in citrate synthase activity with strength training was not seen in the control group and has not been reported in normal subjects. Thus a decrease in oxidative enzyme activity may represent an additional insult to skeletal muscle from strength training.

Acylcarnitine accumulation in tissues reflects the buildup of incomplete oxidative products of metabolism. The association between skeletal muscle accumulation of acylcarnitines and impaired exercise performance in PAOD is consistent with this model of metabolic dysfunction (14). In the present treadmill- trained group, this relationship has been further strengthened. In eight of ten patients, treadmill training was associated with an increase in claudication- limited peak vo2, and in seven of these eight patients, a concomitant decrease in muscle acylcarnitine content was observed. Furthermore, in the two patients in whom peak VO, decreased, acylcarnitine content increased. Thus, under the dynamic conditions of 12 wk of treadmill training, the link between function and muscle metabolic state was maintained. This point is further illustrated in Fig. 3, which demonstrates that the relationship between peak VO, and muscle acylcarnitine content for both entry and exit biopsies was well described by a single regression line. Consistent with previous observations, changes in plasma acylcarnitine metabolism were also inversely related to improvements in exercise performance (12).

Study limitations. Caution should be used in the extrapolation of the study results because a specific duration and intensity of training were employed. Complete “dose-response” curves for exercise training have not been developed, but 6 mo of treadmill training has been shown to be superior to 3 mo of treadmill training (13). Thus, whereas 3 mo of treadmill training did not increase muscle oxidative enzyme activity, exercise programs of greater duration or intensity may have modified these measurements. However, it is also likely that skeletal muscle ischemia impaired these expected metabolic responses to the treadmill-training program. Similarly, longer durations and different muscle-strengthening regimens might result in improved exercise performance from strength training. However, the available data (12, 13) clearly demonstrate the efficacy of treadmill training compared with strength training. Training studies of longer duration will be needed to address these issues. Such studies will be vital, defining the functional importance of the observed denervation injury.

ConeLusions. In patients with claudication, a supervised treadmill-training program is an effective means to improve exercise performance and community-based functional status (12, 13, 26). In contrast, a 12-wk supervised program of muscle strength training was only minimally effective in improving peak exercise performance (13). Thus treadmill training is recom- mended as the preferred mode of exercise rehabilitation in patients with limiting claudication (13). The present studies extend these observations to examine


the changes in muscle histological and metabolic status under the different training conditions.

In patients with claudication, the associations between disease severity and the metabolic markers of 9. citrate synthase activity and acylcarni tine accumulation support the concept that secondary than .ges intrin- sic to skeletal muscle metabolism are an important component of PAOD pathophysiology. The present studies demonstrate that treadmill training was associated with skeletal muscle injury, characterized by denervation of muscle fibers. Over the 12 wk of treatment, these effects were not of sufficient magnitude to prevent the response to the training program but do suggest that increased walking activity over time may be injurious to skeletal muscle fibers. Treadmill training did not result in an improvement in muscle oxidative or glycolytic enzyme activities. However, the increase in peak Vo2 with the walking-exercise program could be correlated with an improvement in skeletal muscle acylcarnitine homeostasis. Also, as noted in Table 3, the increase in walking time with treadmill training ap- peared to be greater than the increase in peak VO,. This apparent dissociation is consistent with the previously reported decrease in VO, during exercise at a defined fixed workload in PAOD patients after treadmill training (13). Taken together, these observations suggest an altered metabolic efficiency of exercise after training. Thus, in patients with PAOD, skeletal muscle dysfunction and the associated muscle metabolic state can be ame liora ted with exercise traini w

The authors thank Michelle Reed, Robert Farrell, and Dr. Tim Moser for providing developmental and technical support in the histological analysis of the muscle tissue.

This study was supported by National Institute on Disability and Rehabilitation Research Grant H-133G90114. W. R. Hiatt is the recipient of a National Institutes of Health Academic Award in Vascular Disease.

Address for reprint requests: W. R. Hiatt, Section of Vascular Medicine, Univ. of Colorado Health Sciences Center, 4200 East Ninth Ave., Box B-180, Denver, CO 80262.

Received 17 October 1995; accepted in final form 22 March 1996.

REFERENCES

8.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

1. 2.

3.

4.

5.

6.

7.

Bieber, L. L. Carnitine. Annu. Rev. Biochem. 57: 261-283,1988. Brass, E. P., and C. L. Hoppel. Relationship between acid- soluble carnitine and coenzyme A pools in vivo. Biochem. J. 190: 495-504,198O. 22. Bylund, A. C., T. Bjuro, G. Cederblad, J. Holm, K. Lund- holm, M. Sjostrom, K. A. Angquist, and T. Schersten. Physical training in man. Skeletal muscle metabolism in relation 23. to muscle morphology and running ability. Eur. J. Appl. Physiol. Occup. Physiol. 36: 151-169, 1977. Dahllof, A., P. Bjorntorp, J. Holm, and T. Schersten. Meta- 24. bolic activity of skeletal muscle in patients with peripheral arterial insufficiency. Effect of physical training. Eur. J. Clin. Invest. 4: g-15,1974. Engel, W. K. Histochemistry of neuromuscular disease- significance of muscle fiber types. In: Proc. VIII Int. Congr. 25. NeuroZ. Vienna: Excerpta Medica, 1965, p. 67-101. England, J. D., M. A. Ferguson, W. R. Hiatt, and J. G. Regensteiner. Progression of neuropathy in peripheral arterial 26. disease. Muscle Nerve 18: 380-387, 1995. England, J. D., J. G. Regensteiner, S. P. Ringel, M. R. Carry, and W. R. Hiatt. Muscle denervation in peripheral arterial 27. disease. NeuroZogy 42: 994-999, 1992.

Frontera, W. R., C. N. Meredith, K. P. O’Reilly, H. G. Knuttgen, and W. J. Evans. Strength conditioning in older men: skeletal muscle hypertrophy and improved function. J. AppZ. PhysioZ. 64: 1038-1044, 1988. Henriksson, J., E. Nygaard, J. Andersson, and B. Eklof. Enzyme activities, fibre types and capillarization in calf muscles of patients with intermittent claudication. Stand. J. CZin. Lab. Invest. 40: 361-369, 1980. Hiatt, W. R., D. Nawaz, and E. P. Brass. Carnitine metabolism during exercise in patients with peripheral vascular disease. J. AppZ. PhysioZ. 62: 2383-2387, 1987. Hiatt, W. R., D. Nawaz, J. G. Regensteiner, and K. F. Hossack. The evaluation of exercise performance in patients with peripheral vascular disease. J. CardiopuZm. Rehab. 12: 525-532,1988. Hiatt, W. R., J. G. Regensteiner, M. E. Hargarten, E. E. Wolfel, and E. P. Brass. Benefit of exercise conditioning for patients with peripheral arterial disease. Circulation 81: 602- 609,199O. Hiatt, W. R., E. E. Wolfel, R. H. Meier, and J. G. Regen- Steiner. Superiority of treadmill walking exercise vs. strength training for patients with peripheral arterial disease. Implica- tions for the mechanism of the training response. CircuZation 90: 1866-1874,1994. Hiatt, W. R., E. E. Wolfel, J. G. Regensteiner, and E. P. Brass. Skeletal muscle carnitine metabolism in patients with unilateral peripheral arterial disease. J. AppZ. Physiol. 73: 346-353,1992. Holloszy, J. O., and E. F. Coyle. Adaptations of skeletal muscle to endurance exercise and their metabolic consequences. J. AppZ. PhysioZ. 56: 831-838, 1984. Holm, J., A. Dahllof, P. Bjorntorp, and T. Schersten. En- zyme studies in muscles of patients with intermittent claudication. Effect of training. Stand. J. CZin. Lab. Invest. 31, SuppZ. 128:201-205,1973. Houston, M. E., E. A. Froese, S. P. Valeriote, H. J. Green, and D. A. Ranney. Muscle performance, morphology and metabolic capacity during strength training and detraining: a one leg model. Eur. J. AppZ. PhysioZ. Occup. Physiol. 51: 25-35, 1983. Iwasaki, Y., and M. Kinoshita. Study on relationship between the muscular pathology and progress in the motor neuron disease. Jpn. J. Med. 26: 335-338, 1987. Jansson, E., J. Johansson, C. Sylven, and L. Kaijser. Calf muscle adaptation in intermittent claudication. Side-differences in muscle metabolic characteristics in patients with unilateral arterial disease. CZin. Physiol. Oxf 8: 17-29, 1988. Kaijser, L., C. J. Sundberg, 0. Eiken, A. Nygren, M. Esbjornsson, C. Sylven, and E. Jansson. Muscle oxidative capacity and work performance after training under local leg ischemia. J. AppZ. Physiol. 69: 785-787, 1990. Kemp, G. J., D. J. Taylor, C. H. Thompson, L. J. Hands, B. Rajagopalan, P. Styles, and G. K. Radda. Quantitative analysis by 31P magnetic resonance spectroscopy of abnormal mitochondrial oxidation in skeletal muscle during recovery from exercise. NMR Biomed. 6: 302-310,1993. Lilienthal, J. L., K. L. Zierler, B. P. Folk, R. Buka, and M. J. Riley. A reference base and system for analysis of muscle constituents. J. BioZ. Chem. 182: 501-508, 1950. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. Protein measurement with the folin phenol reagent. J. BioZ. Chem. 193: 265-275,195l. Lundgren, F., A. G. Dahllof, T. Schersten, and A. C. Bylund- Fellenius. Muscle enzyme adaptation in patients with peripheral arterial insufficiency: spontaneous adaptation, effect of different treatments and consequences on walking performance. CZin. Sci. Lond. 77: 485-493, 1989. Morris, C. J., and J. A. Raybould. Fiber type grouping and end-plate diameter in human skeletal muscle. J. Neural. Sci. 13: 181-187,197l. Regensteiner, J. G., J. F. Steiner, and W. R. Hiatt. Exercise training improves functional status in patients with peripheral arterial disease. J. Vast. Surg. 23: 104-115, 1996. Regensteiner, J. G., E. E. Wolfel, E. P. Brass, M. R. Carry, S. P. Ringel, M. E. Hargarten, E. R. Stamm, and W. R. Hiatt.


Chronic changes in skeletal muscle histology and function in peripheral arterial disease. CircuZation 87: 413-421, 1993.

28. Ronnemaa, T., K. Mattila, A. Lehtonen, and V. Kallio. A controlled randomized study on the effect of long-term physical exercise on the metabolic control in type 2 diabetic patients. Acta Med. Stand. 220: 219-224,1986.

29. Saltin, B., B. Kiens, G. Savard, and P, K. Pedersen. Role of hemoglobin and capillarization for oxygen delivery and extraction in muscular exercise (Abstract). Acta Physiol. Stand. 128, Suppl. 556: 21-32,1986.

30. Sorlie, D., and K. Myhre. Effects of physical training in intermittent claudication. Stand. J. CZin. Lab. Invest. 38: 217- 222,1978.

31. Srere, P.A. Citrate synthase. Methods Enzymol. 13: 3-11,1969.

32. Staal, G. E. J., J. F. Koster, C. J. M. Banziger, and L. van Milligen-Boersma. Human erythrocyte phosphofructokinase: its purification and some properties. Biochim. Biophys. Acta 276: 113-123,1972.

33. Sundberg, C. J. Exercise and training during graded leg ischaemia in healthy man with special reference to effects on skeletal muscle. Actu Physiol. Stand. 150, Suppl. 615: l-50, 1994.

34. Vassault, A. Lactate dehydrogenase. In: Methods of Enzymatic Analysis, edited by H. V Bergmeyer. Weinheim, Germany: Verlag Chemie, 1983, vol. III, p. 118-125.

35. Wicks, K. L., and D. A. Hood. Mitochondrial adaptations in denervated muscle: relationship to muscle performance. Am. J. Physiol. 260 (Cell Physiol. 29): C841-C850,1991.

effect of exercise training on skeletal muscle histology ......muscle metabolism. similarly,...

Documents