- Email: [email protected]
31P nuclear magnetic resonance evidence of abnormal skeletal muscle metabolism in patients with congestive heart failure
CONGESTIVE
HEART FAILURE
31PNuclear Magnetic ResonanceEvidence of AbnormalSkeletal Muscle Metabolism in Patients with Congestive Heart Failure BARRY M. MASSIE, MD, MICHAEL CONWAY, BM, MRCPI, RICHARD YONGE, DPhil, SIMON FROSTICK, BM, FRCS, PETER SLEIGHT, DM, FRCP, JOHN LEDINGHAM, DM, FRGP, GEORGE RADDA, DPhil, FRS, and BHEESHMA RAJAGOPALAN, BM, DPhil
In patients with congestive heart failure (CHF), exercise limitation correlates poorly with central hemodynamic abnormalities, suggesting that additional abnormalities in skeletal muscle blood flow or metabolism play an important pathophysiologic role. Therefore, muscle metabolism was examined by 31P nuclear magnetic resonance (NMR) at rest and during repetitive bulb squeeze exercise in 11 patients with New York Heart Association class II to IV CHF and 7 age-matched control subjects. Serial spectra were obtained at rest, at 2 levels of exercise and during recovery. At rest, the only abnormal finding was an elevated inorganic phosphate (Pi) concentration (5.0 f 1.5 vs 3.6 f 0.4 mM, p
0.16 vs 0.53 f 0.10, p
E
xercise intolerance is characteristic of patients with all grades of congestive heart failure (CHF). Despite elevated left ventricular filling pressures during exer-
cise, most patients are limited by muscle fatigue.l-3 Although exercise capacity is related to maximal cardiac index, the correlations of most rest and exercise measurements of left ventricular function with exercise tolerance are weak.3 These findings focus attention on the role of peripheral factors as the limitation to exercise. Blood flow to exercising muscle is frequently reduced in CHF, both as a result of diminished cardiac output and impaired vasodilatory reserve.4-7 However, other findings, such as failure of exercise capacity to rise acutely even when peripheral blood flow is increased,8,g raise the possibility that intrinsic abnormalities of skeletal muscle function and metabolism may also play an important role. Nuclear magnetic resonance (NMR) imaging with 31P permits noninva-
From the Departments of Biochemistry, Cardiology and Medicine, University of Oxford, Oxford, England, and the Department of Medicine and Cardiovascular Research Institute, University of California, San Francisco, California. This work was supported in part by the Medical Research Council [UK], the British Heart Foundation, the Veterans Administration and Grant HL28146 from the National Heart, Lung and Blood Institute, Bethesda, Maryland. Manuscript received January 5,1987; revised manuscript received and accepted March 23,1987. Address for reprints: Barry Massie, MD, Cardiology Division (lllc), Veterans Administration Hospital, 4150 Clement Street, San Francisco, California 94121. 309
310
MUSCLE
METABOLISM
IN HEART
FAILURE
sive evaluation of muscle metabolism at rest and during exercise. lo A previous study evaluated muscle metabolism during low-level steady-state exercise in patients with CHF 11; this study revealed more rapid phosphocreatine (PCr) hydrolysis and greater acidosis relative to work performed. The present investigation was undertaken to confirm these observations, to extend them to more strenuous exercise and to further examine the postexercise recovery period.
Methods Study population: Eleven patients with stable, chronic CHF were studied. All were men, mean age 57 f 7 years (& standard deviation]. The origin of CHF was coronary artery disease in 6 patients and idiopathic dilated cardiomyopathy in 5.,The duration of heart failure ranged from 3 months to 5 years, and at the time of the study, 5, 4 and 2 patients were in New York Heart Association classes II, III and IV, respectively. Mean ejection fraction was 16 f 3% and maximal upright bicycle exercise tolerance ranged from 25 to 125 W. A control group of 7 age-matched (56 f 8 years), untrained normal subjects was also studied. Nuclear magnetic resonance procedure: NMR studies were conducted with ~CI1.8%tesla, %cm-boie superconducting magnet interfaced with a spectrometer operating at frequencies of 32.5 an,d 80.285 MHz for 31P and lH, respectively. The procedures used in our
laboratory have been described.1°J2J3 Patients sat beside the magnet positioned so that the flexor digitorum superficialis muscle of the dominant arm rest on a 2,5cm-diameter surface coil. The magnetic field, was adjusted for homogeneity so that the line width at half maximum of water protons was less than 40 Hz. The 31P spectra were obtained using a 70’ pulse at a 2second interpulse interval. An initial 256-second, time-averaged rest spectrum was obtained; thereafter, 64-second spectra were obtained during rest and recovery, except in the initial 2 minutes of recovery when spectra were acquired at %-second intervals. Study After the rest spectra were obtained, exercise was performed by squeezing a rubber sphygmomanometer bulb at a rate of 22 per minute. The resistance was set SO that, the maximal pressure generated by emptying the bulb was 100 mm Hg during the first 5 minutes of exercise and 300 mm Hg in the final 25 minutes.lO The actual work performed was estimated by integrating the product of the pressure developed times the duration it was maintained and expressed in arbitrary work units. Sp&tiial arialpsik; Quantitative spectral analysis was performed as previously described.1°J2J3 Relative condentrations of intracellular PCr, inorganic phosphate (Pi) and.adenosine triphosphate (ATP) were determined by triangulation of their respective pe,aks with correction for differential saturation. Absolute
ptod:
Exerkise
Rest PCr
[PCtj = 4.5inM pi! = 31.5
PCtj= 30.4mM Ipi] == 6.5mM ,pH=6.?5
PCfl
PCrj = 35.lmM [["iI = 3.0mM
pi
= 23.ImM
= 18.3mM
tpH=6.95,
FIGURE 1. Representative spectra obtained at rest and at the end of exercise in a congestive heart failure (CHF) patient (fop) and a Control subject (boffom). The Pi, PCr, and ATP peaks are identified and the calculated concentrations of PCr and P, are shown. The pH is determined by the chemical shift between the PCr and the Pi peak and the observed values are indicated. Note the higher Pi value at rest and the almost total depletion of PCr during exercise in the CHF patients. The pH declined to 6.25, whereas it remained in the normal range in the control subject.
August
TABLE
I
Measurements
1: 1987
at Rest
THE
AMERICAN
JOURNAL
Volume
60
3lf
120 iOOtorr Controls
n P” [PCr] in mM/kg wet wt [Pi] in m/Id/kg wet wt [PCr]/([PCr]+[Pi]) [PC$[ATP]
OF CARDIOLOGY
CHF
7.02 34.5 3.6 0.90 2.7
f f f f f
0.03 1.5 0.4 0.01 0.1
7.01 34.6 5.0 0.87 2.8
* zk It zk f
0.02 5.4 1.5" 0.03" 0.4
3OOtorr
p”. 05
100
11
7
load
load
P%OOl 0
.-,” c 3
80 0 63
*p
0
Q
0
---+-
metabolite concentrations were calculated by assuming a normal value of 5.5 mM ATP/kg wet weightI and that compounds were uniformly distributed within an intracellular water space of 0.67 ml intracellular water/g wet weight of muscle.15 The rate of PCr utilization was followed as the ratio of [PCr]/([PCr]+[PJ) in order to minimize the effect of fluctuations in total signal due to arm movement. Intracellular pH was calculated from the chemical shift of Pi according to previously published methods.10 When more than 1 Pi peak was present, the pH was determined from the larger peak, or if peaks were the same size, from the mean. Statistical analysis: A Student unpaired t test was used for comparisons between the control subjects and the patients. All values are expressed as mean f standard deviation.
Results
Metabolic measurements at rest: Figure 1shows spectra obtained at rest and during exercise from a CHF patient and a control and illustrates the identification of peaks and measurement of pH from the chemical shift of the Pi peak. The metabolic measurements at rest are shown in Table I. There were no significant differences in pH or [PCr]. Patients did, however, have a significantly higher [Pi], and this produced a lower calculated [PCr]/( [PCr] + [Pi]) ratio. Work performed: Ten of 11patients completed the 7.5-minute exercise protocol. However, patients performed significantly less work than control subjects at both levels of resistance (Fig. 2). This was particularly apparent with the high resistance, at which none of the patients emptied the bulb fully beyond the first minute. In all patients the limitation to exercise was muscle fatigue. Metabolic measurements during exercise: Figures 3 and 4 show the mean changes in the [PCr]/([PCr] + [Pi]) ratio and pH for the 2 groups. There was an initial rapid decline in PCr during the first minute of exercise in both control subjects and patients; during the next 3 spectra at the low workload there was little further change in [PCr] in control subjects, whereas the patients showed substantial further decline. In the final minute of exercise the curves come together, but at this stage control subjects were performing substantially more work than patients. The pH declined only modestly in control subjects, whereas in CHF patients acidification was marked
-3
”
CHF
Controls
Controls
CHF
FIGURE 2. Amount of work performed by patients with congestive heart failure (CHF) and control subjects against the 2 levels of resistance. The work units reflect a pressure-time integral of the pressure developed during bulb-squeezing. This was lower in the patients at both loads.
and rapid. The pH values for the 2 groups were significantly different at all levels of exercise. The differences between the patients and the control group are illustrated by the spectra in Figure 1. In contrast to control subjects, CHF patients showed nearly complete loss of PCr at the end of exercise and a marked decrease in pH. The progression of these changes in another patient is shown in Figure 5. This man displayed a dramatic decline in pH with 2 resolved Pi signals, representing 2 different pH environments. Double Pi peaks were seen in 5 of the 11 CHF patients but were not observed in control subjects.
l.Of- -co1
ns
c.02
c.01
c.02
c.02
ns
Controls
0.6 ipCd pcrl+po 0.4
0.2 CHF OL
, R
I 2.5 Minutes
I 5.0
I 7.5
of Exercise
FIGURE 3. The changes in the PCr ratio during the exercise protocol. Although each spectrum took 64 seconds to acquire, the time delay for storage and for restarting acquisition was such that 1 spectrum was obtained for each 1.2 minutes of exercise. After the first minute, during which PCr fell in both groups, patients with congestive heart failure (CHF) displayed significantly greater decreases in PCr. In the final minute of exercise, when the patients were performing much less work than control subjects, the 2 curves come together. ns = not significant.
312
MUSCLE
1.2
r
ns
MtTAtWLISM
c.05
FAILURE
IN WART
c.01
co2
<.OOl
<.OOl
c.01
PH
6.6 -
6.4 -
6.2 CHF
6.0L
,
I 2.5
R
Minutes
I 1.5
I 5.0 of
Exercise
FIGURE 4. The pH changes during exercise using the same format as Figure 3. Throughout exercise, pH was significantly lower in the patients.
CHF
= congestive
heart
failure;
ns = not significant.
The relation between the change in pH and [PCr] during exercise is shown in Figure 6. In the control group and in normal subjects studied previously,1°J6 marked acidification did not occur until the PCr ratio declined to below 0.50. The hatched area represents the 95% confidence limits derived from the study of 25 subjects. In the studies of 7 of the 11 CHF patients, at least 1 point during exercise fell below this normal region. Metabolic recovery: Some CHF patients exhibited very slow metabolic recovery after exercise. In the patient whose data are shown in Figure 5, PCr recovery was apparently not complete after 10 minutes and pH also remained low at this time. Figure 7 illustrates the findings during recovery. Four patients displayed
A
CONTROL
CHF
PATIENT
delayed PCr recovery, as defined by a time to 50% recovery greater than the mean plus 2 standard deviations of the control group. This only produced differences between the patients and control subjects of borderline statistical significance, as the remaining 6 patients who completed the exercise protocol recovered in a normal fashion, The 4 patients with delayed PCr recovery also displayed between 10 to 20% ATP loss, a finding not seen in the control group or in normal subjects previously studied by the same protocol. The pH recovery was also delayed, but this in part reflected the low end-exercise pH in the patients. The pH recovery rate could not be determined because the Pi peak disappeared in the midportion of recovery in many subjects.
Discussion Findings of the present study: These results reveal abnormal skeletal muscle metabolism during exercise in patients with CHF. Compared with a group of ageand size-matched control subjects, they used PCr at a faster rate despite performing less work. The energy requirement for exercise is met by the hydrolysis of ATP, which is in equilibrium with PCr. The lower [PCr] in the CHF patients indicates reduced ATP and PCr resynthesis, despite their presumed lower rate of ATP utilization due to the substantially lower work rate. Further evidence for impaired oxidative metabolism is provided by loss of ATP during exercise in 4 patients, a finding observed only with much more severe exercise in normal subjects.13 Furthermore, the lower rate of PCr recovery also suggests impaired oxidative phosphorylation.17 Even more prominent was the marked degree of acidification shown in the CHF patients. Indeed, at the lower resistance level, pH declined to a value lower than that seen in any of the normal subjects in 9 of the 11 patients. This finding indicates excessive depen-
IEx2.5
Ret 10
16.291
.30
6.71
.75
FIGURE 5. Serial 64-second spectra obtained at rest, during exercise and during recovery in a control subject and patient with congestive heart failure (CHF). The calculated values for pH and PCr ratio are shown in lower right In addition to the previously described findings of excessive acidification and decline in PCr, note the presence of split Pi peaks and the delayed recovery of PCr in the patient with CHF.
7.05
.91
August
1, 1987
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Controls
FfGURE 6. Relation between pH and the PCr ratio at rest and during exercise at low resistance. Left, data from 25 normal subjects (7 from this study and 18 previously studiedi3). Hafched region represents 95% confidence limits for this group. Righf, results in the 11 patients with congestive heart failure. Seven exhibited at least 1 point outside the normal range.
Volume
313
60
Patients
a . piif@@
6.8
. .
PH
:
6.6 6.4
6.0-1 0
.
0.2
0.4
0.6
0.8 [PCr]/
dence on glycolytic metabolism for ATP synthesis. Another noteworthy finding was the evolution of double Pi peaks during exercise in 5 of 11 heart failure patients. Since both peaks were large, it is likely that both came from working muscle and they therefore indicate 2 populations of fibers with differing intracellular pH. This could result from local variations in blood supply or oxygen diffusion or from differing propensities toward glycolytic metabolism. The rapid decline in intracellular pH may have been the reason for the low levels of work performed, since it has been suggested that muscle fatigue, as defined by the inability to maintain force of contraction, is related to pH.18 Relation to previous studies: Wilson et ali1 found a greater degree of PCr utilization relative to work rate and a greater decline in intracellular pH. However, the degree of acidification that they reported was considerably less than that we observed 16.75 f 0.23 at their highest work rate vs 6.33 f 1.8 at the end of our exercise protocol]. In addition, they did not detect abnormalities during recovery; PCr resynthesis during recovery was abnormal in 4 of our 11 patients. These differing findings can probably be explained by important differences in the NMR methods and exercise protocols. Wilson et al used a 4.5~cm-diameter surface coil, which would generate spectra from a much larger volume than our 2.5~cm coil. It is likely that the larger coil examined both exercising and nonexercising muscle. Wilson’s exercise protocol consisted of dynamic wrist flexion and was performed until steady state was achieved at 3 workloads. The bulbsqueezing routing we used entails primarily finger flexion and includes a component of isometric as well as isotonic work. Thus, the greater changes in our study may have reflected a higher level of exercise in a smaller muscle mass, the flexor digitorum superficialis, and the fact that the measurements were more precisely focused on the working muscle. Mechanism of the metabolic abnormalities: We cannot delineate the underlying mechanism of these metabolic abnormalities from the 31P NMR findings; however, several explanations warrant consideration.
. -.
.
1.0
0
a.
0.2
* e .’
0.6
0.4
0.8
1.0
1 [PCd+[Pg)
Zelis et a15,*g-21showed that forearm blood flow is reduced during exercise in CHF patients and that the postexercise and postischemia hyperemic responses are also impaired. Wilson,6 LeJemte17 and co-workers and others have demonstrated reduced maximal limb blood flow and vasodilatory reserve in exercising muscle in this setting. Reduced O2 delivery may explain the impaired oxidative metabolism and excessive glycolysis detected by 31P NMR. Several lines of evidence suggest that impaired blood flow is not the only explanation for the metabolic changes. The rate of PCr recovery, which is directly proportional to oxidative phosphorylation,17 could be expected to be decreased when blood flow impairment is severe. PCr recovery was abnormal in only 4
OL
CHF n
indicates
OL
Controls CHF pts with >lC%ATP
CHF
Controls
loss
FIGURE 7. Left, phosphocreatine recovery rate, expressed as the time to 50% recovery (T’/z). Four patients show markedly prolonged PCr recovery rates. These same patlents had a more than 10 % loss of ATP. Although the group differences achieve borderline significance, the remaining patients were indistinguishable from control subjects. Right, time for pH to return to >6.90. This was also much longer in patients with congestive heart failure (CHF), although this finding is difficult to interpret because of the significantly lower postexercise pH values in the patients.
314
MUSCLE
METABOLISM
IN HEART
FAILURE
patients, whereas marked acidification and/or excessive PCr utilization were observed in 10. In normal subjects, a consistent relationship between PCr depletion and acidification has been observed in this study and previously.1°J6 This is illustrated in Figure 6, from which it is also obvious that many CHF patients acidify disproportionately to the decline in [PCr]. This suggests there is a greater shift from oxidative to glycolytic metabolism than is to be expected from reduced OZ delivery alone. In recent reports by Wiener et alz2 and our laboratoryz3 in which 31P NMR observations and plethysmographic blood flow determinations were made during the same exercise protocol, no differences in blood flow were found even though metabolic changes similar to those previously discussed were present. Finally, we have noted similar directional differences between CHF patients and control subjects during ischemic exercise.24 Thus, it seems likely that intrinsic changes in skeletal muscle in CHF patients account for at least part of the metabolic abnormalities. A variety of changes have, in fact, been documented in quadriceps biopsies from patients with severe CHF.25 The abnormal relation between PCr utilization and pH suggest changes in the relative oxidative and glycolytic capacity of muscle. Patients with occlusive peripheral vascular disease also have reduced muscle blood flow and exhibit increased glycolytic metabolism.26 They have been shown to have 31P NMR abnormalities similar to those observed in heart failure patients.27 A decrease in oxidative and increase in glycolytic enzyme activity has been observed in these patients,28,2g and in experimental model of peripheral vascular disease by some,30-32 but not by all, workers.33 These changes could be in part explained by selective changes in distribution or size of muscle fibers.28 The metabolic abnormalities seen in the present study may be explained by a greater proportion of type 2B [fast glycolytic) fibers. Such a change has been observed in patients with peripheral vascular disease,28s2g although this point remains controversial.34 Our unusual finding of double Pi peaks, which have not been previously described in CHF, indicates that there is a population of fibers that exhibits excessive glycolysis. Potential limitations: Another possible explanation for. the metabolic differences between the 2 groups is a smaller muscle mass in the patients. While some degree of atrophy cannot be excluded, it is important to note that the patients were in general not severely limited in their activity and that the muscle studied, the flexor digitorum superficialis, is likely to be less subject to deconditioning than weight-bearing muscles. In addition, the 2 groups did not differ in estimated lean body mass or forearm circumference. Finally, the patients performed a much smaller degree of work, so that the changes are unlikely to reflect a disproportion between work rate and muscle mass. Clinical implications: The present findings confirm and extend previous observations of abnormal skeletal muscle metabolism in CHF. The marked variability in the severity of metabolic changes may particularly explain the poor correlation between measurements
of cardiac function and exercise tolerance.1-3 In the present study, there was no apparent correlation between the 31P NMR data and measurements of exercise tolerance. This is not surprising, considering the small number of patients and that the muscle investigated, the flexor digitorum superficialis is a relatively unimportant upper extremity muscle and measurements of exercise capacity involve work of the lower extremity and respiratory muscles. While the mechanism of these changes remains to be fully defined, circumstantial evidence suggest that they reflect, at least in part, intrinsic changes in muscle. This could explain the oft-repeated observation that interventions that improve central hemodynamics and increase peripheral blood flow produce no acute change in exercise capacity or muscle oxygen uptake.8,g,35-38 In contrast, chronic therapy with some of the same medications can produce significant increases in exercise capacity.36-38 The time course of these changes may be determined by changes in muscle composition or biochemistry.
References 1. Franciosa ]A, Ziesche S, Wilen M. Functional capacity of patients with chronic left ventricular failure; relationship of bicycle exercise performance to clinical and hemodynamic characterization. Am J Med 1979;67:460-466. 2. Wilson JR, Ferraro N. Exercise intolerance in patients with chronic left heart failure: relation to oxygen transport and ventilatory abnormalities. Am J Cardiol 1983;51:1358-1363. 3. Szlachcic J, Massie BM, Kramer BL, Topic N, Tubau J. Correlates and prognostic implication of exercise capacity in chronic congestive heart failure. Am J Cardiol 3985;55:1037-1042. 4. Donald KW, Wormald PN, Taylor SH, Bishop JM. Changes in the oxygen content of femoral venous blood and leg blood flow during leg exercise in relation to cardiac output response. Clin Sci 1957;16:567-591, 5. Z&s R, Nellis SH, Longhurst J, Lee G, Mason DT. Abnormalities in the regional circulations accompanying congestive heart failure. Prog Cardiovasc Dis 1975;18:181-199. 6. Wilson JR, Martin JL, Schwartz II, Ferraro N. Exercise intolerance in patients with chronic heart failure: role of impaired nutritive flow to skeletal muscle. Circulation 1984;69:1079-1087. 7. LeJemtel TH, Maskin CS, Lucid0 D, Chadwick B. Failure to augment maximal limb blood flow in response to one-leg versus two-leg exercise in patients. with severe heart failure. Circulation 1986;74:245-251. 8. Wilson JR, Martin JL, Ferraro N, Weber KT. Effect of hydralazine on perfusion and metabolism in the leg during upright bicycle exercise in patients with heart failure. Circulation 1983;68:425-432, 9. Wilson JR, Martin JL, Ferraro N. Impaired skeletal muscle nutritive flow during exercise in patients with congestive heart failure: role of cardiac pump dysfunction as determined by the effect of dobutamine. Am J Cardiol 1984;53:1308-1315. 10. Taylor DJ, Bore PJ. Styles P, Gadian DG, Radda GK. Bioenergetics of intact human muscle. A 31P nuclear magnetic resonance study. Mol BioJ Med 19832:77-94. 11. Wilson JR, Fink L, Maris J, Ferraro N, Power-Vanwart J, Eleff S, Chance B. Evaluation of energy metabolism in skeletal muscle of patients with heart failure with gated phosphorus-31 nuclear magnetic resonance. Circulation 1985;71:57-62. 12. Arnold DL, Matthews PM, Radda GK. Metabolic recovery after exercise and the assessment of mitochondriol function in vivo in human skeletal muscle by means of 31P NMR. Mag Res Med 1984;1:307-315. 13. Taylor DJ, Styles P, Matthews PM, Arnold D, Gadian DG, Bore PJ, Radda GK. Energetics of human muscle: exercise-induced ATP depletion, Mag Res Med 1986;3:44-54. 14. Harris RC, Hultman E, Nordesjo LO. Glycogen, glycolytic intermediates and high-energy phosphates determined in biopsy samples of muscles quadriceps femoris of man at rest. Methods and variance of values. Stand J Clin Lab Invest 1974;33:109-120. 15. Shalin K. Intracellular pH and energy metabolism inn skeletal muscle of man. Acta Physiol Stand i978;suppl 455!1-56. 16. Taylor DJ, Crowe M, Bore PJ, Styles P, Arnold DL, Radda GK. Examination of the energetics of aging skeletal muscle using nuclear magnetic resonance. Gerontology 1984;30:2-7. 17. Sahlin K, Harris RC, Hultman E. Resynthesis of creatine phosphate in human muscle after exercise in relation to intramuscular pH and availability
of oxygen. Stand 1 CJin Lab invest 1979;39:551-558. 18. Mermansen L. Effect of metabolic changes on force generation in skeietal muscle during maximal exercise. Ciba Foundation Symposium 1981;82:75-88. 19. Zelis R, Mason DT, Braunwald E. A comparison of the effects of vasodilatar stimuli on peripheral resistance vessels in normal subjects and in patients with congestive heart failure. J Clin Invest 1968;47:960-970. 20. Zelis R, Longhurst J, Capone RJ. Mason DT. A comparison of regional blood flow and oxygen utilization during dynamic forearm exercise in normal subjects and patients with congestive heart failure. Circulation 1974;50:137143. 21. Longhurst J, Gifford W, Zelis R. Impaired forearm oxygen consumption during static exercise in patients with congestive heart failure. Circulation 1976:54:477-480. 22. Wiener DH, Fink LI, Maris J, Jones RA, Chagne B, Wilson JR. Abnormal skeletal muscle bioenergetics during exercise in patients with heart failure: role of reduced muscle flow. Circulation 1~86;73:1127-1136. 23. Massie BM, Conway M, Rajagopalan B, Yonge R, Frostick S, Sleight P, Ledinham J, Radda G. Abnormal skeletal muscle metabolism in heart failure: Jack of relation to flow (obstr]. /ACC 1987;9:58A. 24. Massie BM, Conway M. Rajagopalan B, Yonge R, Frostick S, Sleight P, Ledingham J, Radda G. Skeletal muscle metabolism in heart failure: abnormalities during ischemic exercise (abstr). Clin Res 1987;35:303A. 25. Lipkin DP, Jones DA, Round JM, Poole-Wilson PA. Maximal force. fibre type, and enzymatic activity in quadriceps of patients with severe heart failure; a mechanism for reduced exercise capacity (abstr]. Br Heart J 1985;54:622. 26. Pernow B, Saltin J. Wahren R, Cronestrand R, Ekestrom S. Leg blood flow and muscle metabolism in occlusive arterial disease of the leg before and after reconstructive surgery. Cfin Sci Mol Med 1975;119:265-275. 27. Hands LJ. Bore PJ, Galloway G, Morris PJ, Radda GK. Muscle metabolism in patients with peripheral vascular disease investigated by P-31 nuclear magnetic resonance spectroscopy. CJin Sci 1986;71:283-290.
28. Henriksson 1, Nygaard E, Andersson J, Eklof B. Enzyme activities, fibre types and capillarization in calf muscles of patients with intermittent claudication. Stand J CJin Lab Invest 1980;40:361-369. 29. Clyne CAC, Mears H, Weller RO, O’Donnell TF. Calf muscle adaptation to peripheral vascular disease. Cardiovasc Research 1985;19:507-512. 30. Ianda J, Urbanova D, Mrhova 0, Linhart J. The effect of muscular work on the activities of certain enzymes in skeletal muscle in chronic muscular ischaemio. Cor Vasa 1972;14:312-329. 31. Bass A, Gutmann E, Hanzlikova V, Teisinger J. Effects of ischaemio on enzyme-activities in the soleus muscle of the rat. Pflugers Arch 1979;379:203208. 32. Hayes DJ. Challiss RAJ, Radda GK. An investigation of arterial insufficiency in rat hindlimb. An enzymatic, mitochondrial and histological study. Biochem J 1986;236:469-473. 33. Elander A, Idstrom JP, Schersten T, Bylund-Fellenius AC. Metabolic adaptation to reduced muscle blood flow. I. Enzyme and metabolite alterations. Am [ Physiof 1985;249:E63-E69. 34. Bylund AC, Hammarsten J, Helm J, Schersten T. Enzyme activities in skeletal muscles from patients with peripheral arterial insufficiency. Eur J CJin Invest 1976;6:425-429. 35. Rubin SA, Chatterjee K, Parmley WW. Metabolic assessment of exercise in chronic heart failure patients treated with short-term vasodilators. Circuiation 1980;61:543-538. 36. Franciosa JA, Goldsmith SR, Cohn JN. Contrasting immediate and longterm effects of isosorbide dinitrate on exercise capacity in congestive heart 1 failure. Am 1 Med 1980;69:559-566, 37. Massie BM, Kramer BL, Haughom F. Acute and long-term effects of vasodilators on rest and exercise hemodynomics and exercise capacity. Circulation 1981;64:1218-1226. 38. Kramer BL, Massie BM, Topic N. Controlled trial of captoprif in chronic heart failure: a rest and exercise hemodynamic study. Circulation 1983;67: 807-816.
HEART FAILURE
31PNuclear Magnetic ResonanceEvidence of AbnormalSkeletal Muscle Metabolism in Patients with Congestive Heart Failure BARRY M. MASSIE, MD, MICHAEL CONWAY, BM, MRCPI, RICHARD YONGE, DPhil, SIMON FROSTICK, BM, FRCS, PETER SLEIGHT, DM, FRCP, JOHN LEDINGHAM, DM, FRGP, GEORGE RADDA, DPhil, FRS, and BHEESHMA RAJAGOPALAN, BM, DPhil
In patients with congestive heart failure (CHF), exercise limitation correlates poorly with central hemodynamic abnormalities, suggesting that additional abnormalities in skeletal muscle blood flow or metabolism play an important pathophysiologic role. Therefore, muscle metabolism was examined by 31P nuclear magnetic resonance (NMR) at rest and during repetitive bulb squeeze exercise in 11 patients with New York Heart Association class II to IV CHF and 7 age-matched control subjects. Serial spectra were obtained at rest, at 2 levels of exercise and during recovery. At rest, the only abnormal finding was an elevated inorganic phosphate (Pi) concentration (5.0 f 1.5 vs 3.6 f 0.4 mM, p
0.16 vs 0.53 f 0.10, p
E
xercise intolerance is characteristic of patients with all grades of congestive heart failure (CHF). Despite elevated left ventricular filling pressures during exer-
cise, most patients are limited by muscle fatigue.l-3 Although exercise capacity is related to maximal cardiac index, the correlations of most rest and exercise measurements of left ventricular function with exercise tolerance are weak.3 These findings focus attention on the role of peripheral factors as the limitation to exercise. Blood flow to exercising muscle is frequently reduced in CHF, both as a result of diminished cardiac output and impaired vasodilatory reserve.4-7 However, other findings, such as failure of exercise capacity to rise acutely even when peripheral blood flow is increased,8,g raise the possibility that intrinsic abnormalities of skeletal muscle function and metabolism may also play an important role. Nuclear magnetic resonance (NMR) imaging with 31P permits noninva-
From the Departments of Biochemistry, Cardiology and Medicine, University of Oxford, Oxford, England, and the Department of Medicine and Cardiovascular Research Institute, University of California, San Francisco, California. This work was supported in part by the Medical Research Council [UK], the British Heart Foundation, the Veterans Administration and Grant HL28146 from the National Heart, Lung and Blood Institute, Bethesda, Maryland. Manuscript received January 5,1987; revised manuscript received and accepted March 23,1987. Address for reprints: Barry Massie, MD, Cardiology Division (lllc), Veterans Administration Hospital, 4150 Clement Street, San Francisco, California 94121. 309
310
MUSCLE
METABOLISM
IN HEART
FAILURE
sive evaluation of muscle metabolism at rest and during exercise. lo A previous study evaluated muscle metabolism during low-level steady-state exercise in patients with CHF 11; this study revealed more rapid phosphocreatine (PCr) hydrolysis and greater acidosis relative to work performed. The present investigation was undertaken to confirm these observations, to extend them to more strenuous exercise and to further examine the postexercise recovery period.
Methods Study population: Eleven patients with stable, chronic CHF were studied. All were men, mean age 57 f 7 years (& standard deviation]. The origin of CHF was coronary artery disease in 6 patients and idiopathic dilated cardiomyopathy in 5.,The duration of heart failure ranged from 3 months to 5 years, and at the time of the study, 5, 4 and 2 patients were in New York Heart Association classes II, III and IV, respectively. Mean ejection fraction was 16 f 3% and maximal upright bicycle exercise tolerance ranged from 25 to 125 W. A control group of 7 age-matched (56 f 8 years), untrained normal subjects was also studied. Nuclear magnetic resonance procedure: NMR studies were conducted with ~CI1.8%tesla, %cm-boie superconducting magnet interfaced with a spectrometer operating at frequencies of 32.5 an,d 80.285 MHz for 31P and lH, respectively. The procedures used in our
laboratory have been described.1°J2J3 Patients sat beside the magnet positioned so that the flexor digitorum superficialis muscle of the dominant arm rest on a 2,5cm-diameter surface coil. The magnetic field, was adjusted for homogeneity so that the line width at half maximum of water protons was less than 40 Hz. The 31P spectra were obtained using a 70’ pulse at a 2second interpulse interval. An initial 256-second, time-averaged rest spectrum was obtained; thereafter, 64-second spectra were obtained during rest and recovery, except in the initial 2 minutes of recovery when spectra were acquired at %-second intervals. Study After the rest spectra were obtained, exercise was performed by squeezing a rubber sphygmomanometer bulb at a rate of 22 per minute. The resistance was set SO that, the maximal pressure generated by emptying the bulb was 100 mm Hg during the first 5 minutes of exercise and 300 mm Hg in the final 25 minutes.lO The actual work performed was estimated by integrating the product of the pressure developed times the duration it was maintained and expressed in arbitrary work units. Sp&tiial arialpsik; Quantitative spectral analysis was performed as previously described.1°J2J3 Relative condentrations of intracellular PCr, inorganic phosphate (Pi) and.adenosine triphosphate (ATP) were determined by triangulation of their respective pe,aks with correction for differential saturation. Absolute
ptod:
Exerkise
Rest PCr
[PCtj = 4.5inM pi! = 31.5
PCtj= 30.4mM Ipi] == 6.5mM ,pH=6.?5
PCfl
PCrj = 35.lmM [["iI = 3.0mM
pi
= 23.ImM
= 18.3mM
tpH=6.95,
FIGURE 1. Representative spectra obtained at rest and at the end of exercise in a congestive heart failure (CHF) patient (fop) and a Control subject (boffom). The Pi, PCr, and ATP peaks are identified and the calculated concentrations of PCr and P, are shown. The pH is determined by the chemical shift between the PCr and the Pi peak and the observed values are indicated. Note the higher Pi value at rest and the almost total depletion of PCr during exercise in the CHF patients. The pH declined to 6.25, whereas it remained in the normal range in the control subject.
August
TABLE
I
Measurements
1: 1987
at Rest
THE
AMERICAN
JOURNAL
Volume
60
3lf
120 iOOtorr Controls
n P” [PCr] in mM/kg wet wt [Pi] in m/Id/kg wet wt [PCr]/([PCr]+[Pi]) [PC$[ATP]
OF CARDIOLOGY
CHF
7.02 34.5 3.6 0.90 2.7
f f f f f
0.03 1.5 0.4 0.01 0.1
7.01 34.6 5.0 0.87 2.8
* zk It zk f
0.02 5.4 1.5" 0.03" 0.4
3OOtorr
p”. 05
100
11
7
load
load
P%OOl 0
.-,” c 3
80 0 63
*p
0
Q
0
---+-
metabolite concentrations were calculated by assuming a normal value of 5.5 mM ATP/kg wet weightI and that compounds were uniformly distributed within an intracellular water space of 0.67 ml intracellular water/g wet weight of muscle.15 The rate of PCr utilization was followed as the ratio of [PCr]/([PCr]+[PJ) in order to minimize the effect of fluctuations in total signal due to arm movement. Intracellular pH was calculated from the chemical shift of Pi according to previously published methods.10 When more than 1 Pi peak was present, the pH was determined from the larger peak, or if peaks were the same size, from the mean. Statistical analysis: A Student unpaired t test was used for comparisons between the control subjects and the patients. All values are expressed as mean f standard deviation.
Results
Metabolic measurements at rest: Figure 1shows spectra obtained at rest and during exercise from a CHF patient and a control and illustrates the identification of peaks and measurement of pH from the chemical shift of the Pi peak. The metabolic measurements at rest are shown in Table I. There were no significant differences in pH or [PCr]. Patients did, however, have a significantly higher [Pi], and this produced a lower calculated [PCr]/( [PCr] + [Pi]) ratio. Work performed: Ten of 11patients completed the 7.5-minute exercise protocol. However, patients performed significantly less work than control subjects at both levels of resistance (Fig. 2). This was particularly apparent with the high resistance, at which none of the patients emptied the bulb fully beyond the first minute. In all patients the limitation to exercise was muscle fatigue. Metabolic measurements during exercise: Figures 3 and 4 show the mean changes in the [PCr]/([PCr] + [Pi]) ratio and pH for the 2 groups. There was an initial rapid decline in PCr during the first minute of exercise in both control subjects and patients; during the next 3 spectra at the low workload there was little further change in [PCr] in control subjects, whereas the patients showed substantial further decline. In the final minute of exercise the curves come together, but at this stage control subjects were performing substantially more work than patients. The pH declined only modestly in control subjects, whereas in CHF patients acidification was marked
-3
”
CHF
Controls
Controls
CHF
FIGURE 2. Amount of work performed by patients with congestive heart failure (CHF) and control subjects against the 2 levels of resistance. The work units reflect a pressure-time integral of the pressure developed during bulb-squeezing. This was lower in the patients at both loads.
and rapid. The pH values for the 2 groups were significantly different at all levels of exercise. The differences between the patients and the control group are illustrated by the spectra in Figure 1. In contrast to control subjects, CHF patients showed nearly complete loss of PCr at the end of exercise and a marked decrease in pH. The progression of these changes in another patient is shown in Figure 5. This man displayed a dramatic decline in pH with 2 resolved Pi signals, representing 2 different pH environments. Double Pi peaks were seen in 5 of the 11 CHF patients but were not observed in control subjects.
l.Of- -co1
ns
c.02
c.01
c.02
c.02
ns
Controls
0.6 ipCd pcrl+po 0.4
0.2 CHF OL
, R
I 2.5 Minutes
I 5.0
I 7.5
of Exercise
FIGURE 3. The changes in the PCr ratio during the exercise protocol. Although each spectrum took 64 seconds to acquire, the time delay for storage and for restarting acquisition was such that 1 spectrum was obtained for each 1.2 minutes of exercise. After the first minute, during which PCr fell in both groups, patients with congestive heart failure (CHF) displayed significantly greater decreases in PCr. In the final minute of exercise, when the patients were performing much less work than control subjects, the 2 curves come together. ns = not significant.
312
MUSCLE
1.2
r
ns
MtTAtWLISM
c.05
FAILURE
IN WART
c.01
co2
<.OOl
<.OOl
c.01
PH
6.6 -
6.4 -
6.2 CHF
6.0L
,
I 2.5
R
Minutes
I 1.5
I 5.0 of
Exercise
FIGURE 4. The pH changes during exercise using the same format as Figure 3. Throughout exercise, pH was significantly lower in the patients.
CHF
= congestive
heart
failure;
ns = not significant.
The relation between the change in pH and [PCr] during exercise is shown in Figure 6. In the control group and in normal subjects studied previously,1°J6 marked acidification did not occur until the PCr ratio declined to below 0.50. The hatched area represents the 95% confidence limits derived from the study of 25 subjects. In the studies of 7 of the 11 CHF patients, at least 1 point during exercise fell below this normal region. Metabolic recovery: Some CHF patients exhibited very slow metabolic recovery after exercise. In the patient whose data are shown in Figure 5, PCr recovery was apparently not complete after 10 minutes and pH also remained low at this time. Figure 7 illustrates the findings during recovery. Four patients displayed
A
CONTROL
CHF
PATIENT
delayed PCr recovery, as defined by a time to 50% recovery greater than the mean plus 2 standard deviations of the control group. This only produced differences between the patients and control subjects of borderline statistical significance, as the remaining 6 patients who completed the exercise protocol recovered in a normal fashion, The 4 patients with delayed PCr recovery also displayed between 10 to 20% ATP loss, a finding not seen in the control group or in normal subjects previously studied by the same protocol. The pH recovery was also delayed, but this in part reflected the low end-exercise pH in the patients. The pH recovery rate could not be determined because the Pi peak disappeared in the midportion of recovery in many subjects.
Discussion Findings of the present study: These results reveal abnormal skeletal muscle metabolism during exercise in patients with CHF. Compared with a group of ageand size-matched control subjects, they used PCr at a faster rate despite performing less work. The energy requirement for exercise is met by the hydrolysis of ATP, which is in equilibrium with PCr. The lower [PCr] in the CHF patients indicates reduced ATP and PCr resynthesis, despite their presumed lower rate of ATP utilization due to the substantially lower work rate. Further evidence for impaired oxidative metabolism is provided by loss of ATP during exercise in 4 patients, a finding observed only with much more severe exercise in normal subjects.13 Furthermore, the lower rate of PCr recovery also suggests impaired oxidative phosphorylation.17 Even more prominent was the marked degree of acidification shown in the CHF patients. Indeed, at the lower resistance level, pH declined to a value lower than that seen in any of the normal subjects in 9 of the 11 patients. This finding indicates excessive depen-
IEx2.5
Ret 10
16.291
.30
6.71
.75
FIGURE 5. Serial 64-second spectra obtained at rest, during exercise and during recovery in a control subject and patient with congestive heart failure (CHF). The calculated values for pH and PCr ratio are shown in lower right In addition to the previously described findings of excessive acidification and decline in PCr, note the presence of split Pi peaks and the delayed recovery of PCr in the patient with CHF.
7.05
.91
August
1, 1987
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Controls
FfGURE 6. Relation between pH and the PCr ratio at rest and during exercise at low resistance. Left, data from 25 normal subjects (7 from this study and 18 previously studiedi3). Hafched region represents 95% confidence limits for this group. Righf, results in the 11 patients with congestive heart failure. Seven exhibited at least 1 point outside the normal range.
Volume
313
60
Patients
a . piif@@
6.8
. .
PH
:
6.6 6.4
6.0-1 0
.
0.2
0.4
0.6
0.8 [PCr]/
dence on glycolytic metabolism for ATP synthesis. Another noteworthy finding was the evolution of double Pi peaks during exercise in 5 of 11 heart failure patients. Since both peaks were large, it is likely that both came from working muscle and they therefore indicate 2 populations of fibers with differing intracellular pH. This could result from local variations in blood supply or oxygen diffusion or from differing propensities toward glycolytic metabolism. The rapid decline in intracellular pH may have been the reason for the low levels of work performed, since it has been suggested that muscle fatigue, as defined by the inability to maintain force of contraction, is related to pH.18 Relation to previous studies: Wilson et ali1 found a greater degree of PCr utilization relative to work rate and a greater decline in intracellular pH. However, the degree of acidification that they reported was considerably less than that we observed 16.75 f 0.23 at their highest work rate vs 6.33 f 1.8 at the end of our exercise protocol]. In addition, they did not detect abnormalities during recovery; PCr resynthesis during recovery was abnormal in 4 of our 11 patients. These differing findings can probably be explained by important differences in the NMR methods and exercise protocols. Wilson et al used a 4.5~cm-diameter surface coil, which would generate spectra from a much larger volume than our 2.5~cm coil. It is likely that the larger coil examined both exercising and nonexercising muscle. Wilson’s exercise protocol consisted of dynamic wrist flexion and was performed until steady state was achieved at 3 workloads. The bulbsqueezing routing we used entails primarily finger flexion and includes a component of isometric as well as isotonic work. Thus, the greater changes in our study may have reflected a higher level of exercise in a smaller muscle mass, the flexor digitorum superficialis, and the fact that the measurements were more precisely focused on the working muscle. Mechanism of the metabolic abnormalities: We cannot delineate the underlying mechanism of these metabolic abnormalities from the 31P NMR findings; however, several explanations warrant consideration.
. -.
.
1.0
0
a.
0.2
* e .’
0.6
0.4
0.8
1.0
1 [PCd+[Pg)
Zelis et a15,*g-21showed that forearm blood flow is reduced during exercise in CHF patients and that the postexercise and postischemia hyperemic responses are also impaired. Wilson,6 LeJemte17 and co-workers and others have demonstrated reduced maximal limb blood flow and vasodilatory reserve in exercising muscle in this setting. Reduced O2 delivery may explain the impaired oxidative metabolism and excessive glycolysis detected by 31P NMR. Several lines of evidence suggest that impaired blood flow is not the only explanation for the metabolic changes. The rate of PCr recovery, which is directly proportional to oxidative phosphorylation,17 could be expected to be decreased when blood flow impairment is severe. PCr recovery was abnormal in only 4
OL
CHF n
indicates
OL
Controls CHF pts with >lC%ATP
CHF
Controls
loss
FIGURE 7. Left, phosphocreatine recovery rate, expressed as the time to 50% recovery (T’/z). Four patients show markedly prolonged PCr recovery rates. These same patlents had a more than 10 % loss of ATP. Although the group differences achieve borderline significance, the remaining patients were indistinguishable from control subjects. Right, time for pH to return to >6.90. This was also much longer in patients with congestive heart failure (CHF), although this finding is difficult to interpret because of the significantly lower postexercise pH values in the patients.
314
MUSCLE
METABOLISM
IN HEART
FAILURE
patients, whereas marked acidification and/or excessive PCr utilization were observed in 10. In normal subjects, a consistent relationship between PCr depletion and acidification has been observed in this study and previously.1°J6 This is illustrated in Figure 6, from which it is also obvious that many CHF patients acidify disproportionately to the decline in [PCr]. This suggests there is a greater shift from oxidative to glycolytic metabolism than is to be expected from reduced OZ delivery alone. In recent reports by Wiener et alz2 and our laboratoryz3 in which 31P NMR observations and plethysmographic blood flow determinations were made during the same exercise protocol, no differences in blood flow were found even though metabolic changes similar to those previously discussed were present. Finally, we have noted similar directional differences between CHF patients and control subjects during ischemic exercise.24 Thus, it seems likely that intrinsic changes in skeletal muscle in CHF patients account for at least part of the metabolic abnormalities. A variety of changes have, in fact, been documented in quadriceps biopsies from patients with severe CHF.25 The abnormal relation between PCr utilization and pH suggest changes in the relative oxidative and glycolytic capacity of muscle. Patients with occlusive peripheral vascular disease also have reduced muscle blood flow and exhibit increased glycolytic metabolism.26 They have been shown to have 31P NMR abnormalities similar to those observed in heart failure patients.27 A decrease in oxidative and increase in glycolytic enzyme activity has been observed in these patients,28,2g and in experimental model of peripheral vascular disease by some,30-32 but not by all, workers.33 These changes could be in part explained by selective changes in distribution or size of muscle fibers.28 The metabolic abnormalities seen in the present study may be explained by a greater proportion of type 2B [fast glycolytic) fibers. Such a change has been observed in patients with peripheral vascular disease,28s2g although this point remains controversial.34 Our unusual finding of double Pi peaks, which have not been previously described in CHF, indicates that there is a population of fibers that exhibits excessive glycolysis. Potential limitations: Another possible explanation for. the metabolic differences between the 2 groups is a smaller muscle mass in the patients. While some degree of atrophy cannot be excluded, it is important to note that the patients were in general not severely limited in their activity and that the muscle studied, the flexor digitorum superficialis, is likely to be less subject to deconditioning than weight-bearing muscles. In addition, the 2 groups did not differ in estimated lean body mass or forearm circumference. Finally, the patients performed a much smaller degree of work, so that the changes are unlikely to reflect a disproportion between work rate and muscle mass. Clinical implications: The present findings confirm and extend previous observations of abnormal skeletal muscle metabolism in CHF. The marked variability in the severity of metabolic changes may particularly explain the poor correlation between measurements
of cardiac function and exercise tolerance.1-3 In the present study, there was no apparent correlation between the 31P NMR data and measurements of exercise tolerance. This is not surprising, considering the small number of patients and that the muscle investigated, the flexor digitorum superficialis is a relatively unimportant upper extremity muscle and measurements of exercise capacity involve work of the lower extremity and respiratory muscles. While the mechanism of these changes remains to be fully defined, circumstantial evidence suggest that they reflect, at least in part, intrinsic changes in muscle. This could explain the oft-repeated observation that interventions that improve central hemodynamics and increase peripheral blood flow produce no acute change in exercise capacity or muscle oxygen uptake.8,g,35-38 In contrast, chronic therapy with some of the same medications can produce significant increases in exercise capacity.36-38 The time course of these changes may be determined by changes in muscle composition or biochemistry.
References 1. Franciosa ]A, Ziesche S, Wilen M. Functional capacity of patients with chronic left ventricular failure; relationship of bicycle exercise performance to clinical and hemodynamic characterization. Am J Med 1979;67:460-466. 2. Wilson JR, Ferraro N. Exercise intolerance in patients with chronic left heart failure: relation to oxygen transport and ventilatory abnormalities. Am J Cardiol 1983;51:1358-1363. 3. Szlachcic J, Massie BM, Kramer BL, Topic N, Tubau J. Correlates and prognostic implication of exercise capacity in chronic congestive heart failure. Am J Cardiol 3985;55:1037-1042. 4. Donald KW, Wormald PN, Taylor SH, Bishop JM. Changes in the oxygen content of femoral venous blood and leg blood flow during leg exercise in relation to cardiac output response. Clin Sci 1957;16:567-591, 5. Z&s R, Nellis SH, Longhurst J, Lee G, Mason DT. Abnormalities in the regional circulations accompanying congestive heart failure. Prog Cardiovasc Dis 1975;18:181-199. 6. Wilson JR, Martin JL, Schwartz II, Ferraro N. Exercise intolerance in patients with chronic heart failure: role of impaired nutritive flow to skeletal muscle. Circulation 1984;69:1079-1087. 7. LeJemtel TH, Maskin CS, Lucid0 D, Chadwick B. Failure to augment maximal limb blood flow in response to one-leg versus two-leg exercise in patients. with severe heart failure. Circulation 1986;74:245-251. 8. Wilson JR, Martin JL, Ferraro N, Weber KT. Effect of hydralazine on perfusion and metabolism in the leg during upright bicycle exercise in patients with heart failure. Circulation 1983;68:425-432, 9. Wilson JR, Martin JL, Ferraro N. Impaired skeletal muscle nutritive flow during exercise in patients with congestive heart failure: role of cardiac pump dysfunction as determined by the effect of dobutamine. Am J Cardiol 1984;53:1308-1315. 10. Taylor DJ, Bore PJ. Styles P, Gadian DG, Radda GK. Bioenergetics of intact human muscle. A 31P nuclear magnetic resonance study. Mol BioJ Med 19832:77-94. 11. Wilson JR, Fink L, Maris J, Ferraro N, Power-Vanwart J, Eleff S, Chance B. Evaluation of energy metabolism in skeletal muscle of patients with heart failure with gated phosphorus-31 nuclear magnetic resonance. Circulation 1985;71:57-62. 12. Arnold DL, Matthews PM, Radda GK. Metabolic recovery after exercise and the assessment of mitochondriol function in vivo in human skeletal muscle by means of 31P NMR. Mag Res Med 1984;1:307-315. 13. Taylor DJ, Styles P, Matthews PM, Arnold D, Gadian DG, Bore PJ, Radda GK. Energetics of human muscle: exercise-induced ATP depletion, Mag Res Med 1986;3:44-54. 14. Harris RC, Hultman E, Nordesjo LO. Glycogen, glycolytic intermediates and high-energy phosphates determined in biopsy samples of muscles quadriceps femoris of man at rest. Methods and variance of values. Stand J Clin Lab Invest 1974;33:109-120. 15. Shalin K. Intracellular pH and energy metabolism inn skeletal muscle of man. Acta Physiol Stand i978;suppl 455!1-56. 16. Taylor DJ, Crowe M, Bore PJ, Styles P, Arnold DL, Radda GK. Examination of the energetics of aging skeletal muscle using nuclear magnetic resonance. Gerontology 1984;30:2-7. 17. Sahlin K, Harris RC, Hultman E. Resynthesis of creatine phosphate in human muscle after exercise in relation to intramuscular pH and availability
of oxygen. Stand 1 CJin Lab invest 1979;39:551-558. 18. Mermansen L. Effect of metabolic changes on force generation in skeietal muscle during maximal exercise. Ciba Foundation Symposium 1981;82:75-88. 19. Zelis R, Mason DT, Braunwald E. A comparison of the effects of vasodilatar stimuli on peripheral resistance vessels in normal subjects and in patients with congestive heart failure. J Clin Invest 1968;47:960-970. 20. Zelis R, Longhurst J, Capone RJ. Mason DT. A comparison of regional blood flow and oxygen utilization during dynamic forearm exercise in normal subjects and patients with congestive heart failure. Circulation 1974;50:137143. 21. Longhurst J, Gifford W, Zelis R. Impaired forearm oxygen consumption during static exercise in patients with congestive heart failure. Circulation 1976:54:477-480. 22. Wiener DH, Fink LI, Maris J, Jones RA, Chagne B, Wilson JR. Abnormal skeletal muscle bioenergetics during exercise in patients with heart failure: role of reduced muscle flow. Circulation 1~86;73:1127-1136. 23. Massie BM, Conway M, Rajagopalan B, Yonge R, Frostick S, Sleight P, Ledinham J, Radda G. Abnormal skeletal muscle metabolism in heart failure: Jack of relation to flow (obstr]. /ACC 1987;9:58A. 24. Massie BM, Conway M. Rajagopalan B, Yonge R, Frostick S, Sleight P, Ledingham J, Radda G. Skeletal muscle metabolism in heart failure: abnormalities during ischemic exercise (abstr). Clin Res 1987;35:303A. 25. Lipkin DP, Jones DA, Round JM, Poole-Wilson PA. Maximal force. fibre type, and enzymatic activity in quadriceps of patients with severe heart failure; a mechanism for reduced exercise capacity (abstr]. Br Heart J 1985;54:622. 26. Pernow B, Saltin J. Wahren R, Cronestrand R, Ekestrom S. Leg blood flow and muscle metabolism in occlusive arterial disease of the leg before and after reconstructive surgery. Cfin Sci Mol Med 1975;119:265-275. 27. Hands LJ. Bore PJ, Galloway G, Morris PJ, Radda GK. Muscle metabolism in patients with peripheral vascular disease investigated by P-31 nuclear magnetic resonance spectroscopy. CJin Sci 1986;71:283-290.
28. Henriksson 1, Nygaard E, Andersson J, Eklof B. Enzyme activities, fibre types and capillarization in calf muscles of patients with intermittent claudication. Stand J CJin Lab Invest 1980;40:361-369. 29. Clyne CAC, Mears H, Weller RO, O’Donnell TF. Calf muscle adaptation to peripheral vascular disease. Cardiovasc Research 1985;19:507-512. 30. Ianda J, Urbanova D, Mrhova 0, Linhart J. The effect of muscular work on the activities of certain enzymes in skeletal muscle in chronic muscular ischaemio. Cor Vasa 1972;14:312-329. 31. Bass A, Gutmann E, Hanzlikova V, Teisinger J. Effects of ischaemio on enzyme-activities in the soleus muscle of the rat. Pflugers Arch 1979;379:203208. 32. Hayes DJ. Challiss RAJ, Radda GK. An investigation of arterial insufficiency in rat hindlimb. An enzymatic, mitochondrial and histological study. Biochem J 1986;236:469-473. 33. Elander A, Idstrom JP, Schersten T, Bylund-Fellenius AC. Metabolic adaptation to reduced muscle blood flow. I. Enzyme and metabolite alterations. Am [ Physiof 1985;249:E63-E69. 34. Bylund AC, Hammarsten J, Helm J, Schersten T. Enzyme activities in skeletal muscles from patients with peripheral arterial insufficiency. Eur J CJin Invest 1976;6:425-429. 35. Rubin SA, Chatterjee K, Parmley WW. Metabolic assessment of exercise in chronic heart failure patients treated with short-term vasodilators. Circuiation 1980;61:543-538. 36. Franciosa JA, Goldsmith SR, Cohn JN. Contrasting immediate and longterm effects of isosorbide dinitrate on exercise capacity in congestive heart 1 failure. Am 1 Med 1980;69:559-566, 37. Massie BM, Kramer BL, Haughom F. Acute and long-term effects of vasodilators on rest and exercise hemodynomics and exercise capacity. Circulation 1981;64:1218-1226. 38. Kramer BL, Massie BM, Topic N. Controlled trial of captoprif in chronic heart failure: a rest and exercise hemodynamic study. Circulation 1983;67: 807-816.