S143

Magnetic Resonance Spectroscopy and Imaging of Muscle — A Physiological Approach R. H. T. Edwards, H. Gibson, N. Roberts, J. E. Clague, P. A. Martin Magnetic Resonance Research Centre and Muscle Research Centre, Department of Medicine, University of Liverpool, Liverpool, United Kingdom

Abstract R. H. T Edwards, H. Gibson, N. Roberts, J. E.

Clague and P. A. Martin, Magnetic Resonance Spectroscopy and Imaging of Muscle — A Physiological Approach. IntJSportsMed,Vol 13,Suppl 1,ppSl43—S146, 1992.

Magnetic resonance spectroscopy (MRS) and imaging (MRI) are now well established techniques for

the study of cellular metabolism and gross structure of muscle. Using non-ferrous materials, we have constructed a

system for the measurement of isometric force of quadriceps in response to percutaneous electrical stimulation and voluntary effort within the bore of a 48 cm diameter 1.5 T General Electric SIGNA whole body MR system. Using this system we have been able to study the relationship be-

tween electromechanical coupling and chemistry of muscle, with 31 MRS for the measurement of high-energy phosphates and pH, during electrically stimulated activity. Image analysis using the Context Vision system enables a distinction to be made, in 1H MRI by Ti/T 2 mapping, of muscle, fat and connective tissue to give force per unit cross

sectional area of muscle. The combination of MR and functional measurements provide a valuable tool for further detailed analysis of human muscle weakness and fatigue.

Key words

Magnetic resonance imaging, magnetic resonance spectroscopy, image analysis, muscle stimulation

Introduction

Magnetic resonance imaging (MRI) and spectroscopy (MRS) have given new and exciting opportunities for the study of gross structure and metabolism of muscle in health and disease despite being of limited value in the diagnosis of myopathy. This paper presents applications of MR to clinical physiology of muscle. The primary function of muscle is to generate force, contraction force being proportional to cross-sectional area of muscle (18). 'Weakness', which may be defined as a lnt.J. SportsMed. 13(1992)S143—S146 GeorgThieme Verlag Stuttgart NewYork

or power output (9), are predominantly the measurable outcomes of muscle disease: the predominant cause of weakness in all myopathies is loss of muscle cross sectional area. These can be studied using objective measures of function employing electrical stimulation techniques which are very reliable and repeatable (10). Because of the magnetic field environment and constraints of space within the magnet bore a satisfactory integrated approach for the study of the interrelations of electromechanical coupling, chemistry and structure of muscle, comparable to that of laboratory investigations (16) has hitherto proved difficult.

We have developed a system for electrical stimulation and measurement of isometric force of quadriceps, a muscle group that is often primarily affected in muscle disease, in a 1.5 T General Electric SIGNA 48 cm bore whole body MR system that permits both MRS and MRI. This paper describes our methods and preliminary results of the integra-

tion of function, MRS and MRI applied to the analysis of fatigue and weakness in healthy and diseased muscle. Inter-Relation of Electro-Mechanical Coupling and Chemistry in the Study of Fatigue

Human muscle metabolism in normal and diseased muscle was one of the first systems to be studied by 31P

MRS (7, 11,23), the gross structure of the tissue and ease of access using relatively simple surface coil techniques being of great advantage. The goal of many early studies was to study the relationships of ATP, PCr, Pi and pH to human muscle function and fatigue. These studies were limited to the small hand or forearm muscles, often being exercised either dynamically by squeezing rubber bulbs (26), performing submaximal isometric contractions (22) or contractions induced by electri-

cal stimulation, via motor nerves, involving parallel inside/outside magnet experiments for the assessment of frequency: force and relaxation characteristics (5). Objective study of larger muscle groups, particularly in disease, has largely been limited by small bore size and the inability to stimulate muscle within the magnet. Hitherto, the relationship between function and chemistry has been investigated in the

laboratory by methods requiring several needle biopsy samples overtheperiodofexercise andrecovery(3, 16, 17).

The large bore of our magnet system has permitted the development of a system for studying muscle func-

tion of the quadriceps. Initial developments took place in a mock-up system with the same dimensions as the internal dimensions of the MR system and from there transferred. With

Downloaded by: National University of Singapore. Copyrighted material.

failure to generate force, and 'fatigue', a failure to sustain force

S144 mt. J. Sports Med. 13 (1992)

R. H. T Edwards, H. Gibson, N. Roberts, J. E. Clague and P. A. Martin

the subject lying supine, the legs were placed over a wedgeshaped polystyrene foam block, such that the knee angle was 105°. Isometric force was obtained from a specially constructed strain gauge made from non-ferrous materials and attached to the ankle via an inextensible strap. Percutaneous stimulation was achieved by the use of two gel defibrillator pads (3M) placed over the proximal and distal portions of the thigh and strapped in place with crepe bandages. Small copper

100

a

% P20 (E.M C)

__________________________________

%PCr (Energy>/' / N

H

foil plates (3 cm sq) were used to pass current from the stimula-

tion leads to the defibrillator pads, thereby minimizing the quantity of metal within the magnet. When set up in the mag-

net, 31p spectra were acquired with pulse and acquire

% P50 (Force)

sequences at a repetition time of 12 seconds, using various home built surface coils for signal sampling. The stimulator, 0 DO

b

Studies were carried out with fatigue produced

under conditions of open or occluded circulation by the methods used previously for the adductor pollicis muscle (6). Occlusion was achieved by inflation of a sphygmomanometer cuff around the upper part of the thigh. Activity consisted of 14 (occluded conditions) or 50 (non-occluded conditions) trains of programmed stimulation myograms (PSM) consisting of impulses delivered at 1, 10, 20, 50 and 100 Hz (1 sec each, 10 Hz for 2 sees). A model of the interrelationship between stimu-

lated force at 50 Hz, electromechanical coupling as represented by the force at 20 Hz and energy status of skeletal muscle as represented by phosphoryl creatine (PCr) for a healthy normal subject is shown in Figure 1. Plotted on the same figure are results from laboratory investigations by Bergstrom & Hultman (3). Such a model allows the visualization of

the relative importance of possible mechanisms underlying fatigue in different forms of exercise: here are shown our data for ischaemic and non-occluded repetitive stimulated contractions and recovery.

Of interest in several studies was the observation of a double Pi peak during activity. The possible contribution of different fibre type populations with differing metabolic characteristics has been alluded to (I), although it is difficult to ensure that contamination of the signal by incompletely stimulated muscle i.e., a partial volume effect, may in part be responsible. Image guided spectroscopy e.g., one dimensional chemical shift imaging (1DCSI) (4) could be performed to ensure that the MR signal is obtained solely from active muscle, but the time costs are considerable with acquisition times increased 64 fold. We are able to check that the muscle studied lies underneath the coil by using standard MRI techniques and

we have developed a new eliptical coil to minimize partial volume effects and to maximize signal to noise.

Inter-Relation of Structure and Function for the Study of Muscle Weakness in Myopathy Force generation in skeletal muscle is proportional to the effective cross-section of all the force generating units in parallel. The most obvious cause of weakness is thus

% P50 (Force)

00

EM C: Eiectro-mechaoical couplin5

Fig. 1 Three-dimensional plots based on the 'catastrophe theory' representation of inter-relations between force, energy and electromechanical-coupling. a) Here are shown the interrelations of stimulated tetanic force at 50 Hz (P50) which depends on excitation; electro-mechanical coupling represented by the force at 20 Hz (P20) which is dependent on excitationcontraction coupling and the energy status of skeletal muscle as represented by phosphory! creatine (PCr), b) Here are shown each interrelation presented on the solid faces of the cube. Solid line non-occluded; dotdashed line ischaemic; bold dashed line in b) data derived from Bergstrom & Hultman (3). All values expressed as a percentage of initial value. This model is presented as a means of visualizing the relative importance of possible mechanisms underlying fatigue in different forms of exercise.

due to a reduction in the number of functional parallel units.

Loss of muscle however means that greater stresses will be placed on what remains, potentially leading to further damage. This may be of particular importance in the wasting diseases such as the inflammatory diseases of muscle and the muscular dystrophies: an understanding of the changes in composition of muscle in relation to function may permit some form of assessment of disease progress.

Here, proton MRI has advantages over other scanning techniques. The analysis of the relation of functional performance to structure is complicated by the inability to discriminate reliably muscle, fat and connective tissues. X-Ray tomography allows the separation of muscle and fat (14), but

Downloaded by: National University of Singapore. Copyrighted material.

sited in the control room, was triggered by a pulse programme on an Apple II computer, itself triggered by the magnet's computer. This system of stimulation allowed tetanic contractions of up to 60% of the maximal voluntary contraction (MVC) to be achieved. Although some contamination of the force trace was seen with use of magnetic field gradients, this was not a problem where only Rf pulses were used.

mt. i Sports Med. 13(1992) S145

Magnetic Resonance Spectroscopy and Imaging of Muscle

Relaxation time and proton density data do, however, take considerably longer to acquire than a single qualitative image; i.e., 40 minutes as compared to 10 minutes, and may not always be possible for reasons of practical economics. Nevertheless, to date useful quantitative analysis has been possible for gastrocnemius and soleus in FSH and Becker dystrophies although detailed studies of quadriceps have not as yet been performed.

12 msecs

1500

Ti msecs Fig. 2 Quantitation of muscle and fat using Ii andT2 mapping techniques. This is a new opportunity of analyse gross muscle structure in disease so that force per unit cross sectional area can be derived based on true

muscle tissue (rather than including fat or fibrous tissue).

owing to the closeness of the Hounsfield number for muscle

and connective tissue quantification of the latter is not possible. This required biopsy sampling for estimation of connective tissue in a detailed study of the composition of muscle in the Duchenne dystrophy (19). However, H MRI yields bet-

ter soft tissue contrast than CT with the advantage of being able to discriminate muscle composition into three compartments and with no hazard of ionizing radiation. We are therefore seeking to apply MRI together with functional measurements to obtain a more reliable measure of intrinsic force generation (force/unit cross section) of muscle in health and disease.

Quantification comes from the measurement

The application of MR to muscle has yet more important information to yield about muscle metabolism and function. Muscle damage can be identified by an increase in Pi as is observed following eccentric exercise and disease such as in polymyositis (2, 21) and visualized by Ti, T2 mapping techniques of MRI, as has been reported in patients following contracture formation resulting from a metabolic energy crisis (13) or where oedema or haematoma have resulted. Recently

developed magnetization transfer contrast imaging techniques can detect changes in signal intensity following exercise at low magnetic fields (27). Three dimensional quantitation of structures is possible using design based stereological methods (15). Specialized computer graphic techniques can be applied serially for detailed and objective visualization of alterations in muscle structures with disease progress. Already

high quality images can be obtained in seconds. As new developments occur time costs will undoubtedly be reduced also. MR imaging with image analysis thus offers a new opportunity for quantifying from muscle structure as a tool for the detailed study of human muscle in health and disease. Acknowledgements We gratefully acknowledge Northwest Cancer Research Fund, Marilyn Houlton Trust for Motor INeurone disease and the Muscular Dystrophy Group of Great Britain and Northern Ireland for financial support.

ofTi (spin lattice) and T2 (spin-spin) relaxation times which allows discrimination between muscle and fat. Fat has relatively

short ii and medium T2 times whereas muscle has relatively long Ti and short T2. Connective tissue is essentially solid with very fast relaxation times. Ti andT2 weighted images therefore may be used to highlight fat and muscle. Ti mapping has been applied to the study of Duchenne dystrophy (20), but combined with T2 mapping quantitative assessment of muscle and fat was shown to be possible (12). To quantify connective tissue we are developing an enhanced procedure in which proton density mapping is performed. Because of its very low mobile proton density, regions of connective tissue appear as a void in the proton density image. Ti, T2 (Figure 2) and proton density maps can therefore be interrogated to provide a three compartment description of muscle composition. Using a Context Vision Image analysis system, histograms are obtained of the various components in the muscle. Statistical image analysis methods can be applied to the study of the three feature images and textural measures may additionally be in-

corporated (8). The spatial distributions of particularly anomalous areas, relative to normals, can be highlighted as colour overlays. Mathematical morphological algorithms can additionally be incorporated for automating the quantitative analysis (24, 25).

References Achten E., van Cauteren M., Willem R., Luypaert R., Malaisse W. J., van Bosch G., Delanghe G., de Meirleir K., Osteaux M.: 31P-

NMR spectroscopy and the metabolic properties of different 2

muscle fibers.JApplPhysiol68: 644—649, 1990. Aldridge R., Cady E. B., Jones D. A., Obletter G.: Muscle pain

after exercise is linked with an inorganic phosphate increase as shown by 31P NMR. Bioscience Reports 6:663—667, i 986.

'

Bergstrom M., Hultman E.: Relaxation and force during fatigue and recovery of the human quadriceps muscle: relations to metabolitechanges.Pflüger'sArch4l8: 153—160,1991. Brown T. R., Kicaid B. M., Ugurbil K.: NMR chemical shift imaging in three dimensions. Proc Nat Acad Sci USA 79: 3523—3526, 1982. Cady E. B., Jones D. A,, Lynn J., Newham D. J.: Changes in force

and intracellular metabolites during fatigue of human skeletal 6

' 8

muscle.JPhysiol(London) 4 i 8: 311— 325, 1989. Cooper R. G., Gibson H., Stokes M. J., Edwards R. H. T.: Human muscle fatigue: frequency dependence of excitation and force generation. JPhysiol(London) 397: 585—599, 1988.

Cresshulll.,DawsonM.J.,EdwardsR.H.T.,GadiaflD.G.,G0r don R. E., Radda G. K., Shaw D., Wilkie D. R.: Human muscle analysed by 31P nuclear magnetic resonance in intact subjects. J Physiol(London)317: i8P, 1981. Duda R. 0., Hart P. E.: Pattern classification and scene analysis. John Wiley, New York, 1973.

Downloaded by: National University of Singapore. Copyrighted material.

Other Developments

R. H. T. Edwards, H. Gibson, N. Roberts, J. E. Clague and P. A. Martin

Edwards R. H. T.: In: Human muscle fatigue: physiological mechanisms, Ciba Foundation Symposium 82. Porter, R. & Whelan, J. (eds). Pitman Medical, London, 1981, pp 324. 10 Edwards R. H. T., Young A., Hosking G. P., Jones D. A.: Human skeletal muscle function: description of tests and normal values. ClinSciMolMed52: 282—290, 1977.

Edwards R. H. T., Dawson M. J., Wilkie D. R., Gordon R. E., Shaw D.: Clinical use of nuclear magnetic resonance in the investigation ofmyopathy. Lancet i: 725—731, 1982. 12 Ehman R. L.: The interpretation of magnetic resonance images. In: Magnetic resonance of the musculoskeletal system. Berquist T.

' ' 15

6

IS

19

H. (eds). Raven Press, 1987, pp 23—64. Fleckenstein J. L., Peshock R. M., Lewis S. F., HaIler R. G.: Magnetic resonance imaging of muscle injury and atrophy in glycolytic rnyopathies.Muscle&Nerve 12:849—855,1989.

Grindrod S., Tofts P., Edwards R. H. T.: Investigation of human skeletal muscle structure and composition by X-ray computerised tomography.EurfCin Invest 13:465 —468, 1983. Gundersen H. J. G., Bendsten T. F., Korbo L., Marcussen N.,

Moller A., Nielsen K., Nyengaard J. R., Pakkenberg B., Sorensen F. B., Vesterby A., West M. J.: Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS96: 379—394,1988. Hultman E., Sjoholm H.: Energy metabolism and contraction force of human skeletal muscle in situ during electrical stimulation. JPhysiol (London) 345:525—532, 1983. Huitman F., Spriet L. L.: Skeletal muscle metabolism, contraction force and glycogen utilization during prolonged electrical stimulation in humans. JPhysiol(London) 374:493—501, 1986. Jones D. A., Edwards R. H. T.: Muscle strength and metabolism. In: recent achievements in restorative neurology 2: Progressive neuromuscular diseases. Basel, Karger, 123—138, 1986. Jones D. A., Round J. M., Edwards R. H. T., Grindrod S. R., Tofts P. 5.: Size and composition of the calf and quadriceps muscles in Duchenne muscular dystrophy. JNeurolSci 60: 307—322, 1983.

20 Matsumura K., Nakano I., Fukuda N., Ikehira H., Tateno Y., Aoki

Y.: Proton spin-lattice relaxation time of Duchenne dystrophy skeletal muscle by magnetic resonance imaging, Muscle & Nerve 11:97—102,1988. McCully K., Argov Z., Boden B. P., Brown L. B., Bank W. J., Chance B.: Detection of muscle injury in humans with 31-P magnetic resonance spectroscopy. Muscle & Nerve 11:212—216, 1988. 22 MillerR.G.,BoskaM.D.,MoussaviR.S.,CarsonP.J.,WeinerM. 21

W.: 31P nuclear magnetic resonance studies of high energy phosphates and pH in human muscle fatigue. J Clin Invest 81:

23

1190—1196,1988. Radda G. K., Bore P. J., Gadian D. G., Ross B. D., Styles P., Taylor

D. J., Morgan-Huges J.: 31P NMR examination of two patients with NADH-C0Q reductase deficiency. Nature 295: 608—609, 24

1982. Serra J.: Image analysis and mathematical morphology. Academic

Press, London, 1982. Voll. Serra I.: Image analysis and mathematical morphology. Academic Press, London, 1988, Vol 2. 26 Taylor D. J., Styles P., Matthews P. M., Arnold D. A., Gadian D. G., Bore P., Radda G. K.: Energetics of human muscle: Exerciseinduced ATP depletion. Magnetic resonance in medicine 3: 44— 54,1986. 27 Zhu X. P., Zhao S., Isherwood I.: Magnetization transfer contrast 25

(MTC) imaging of skeletal muscle at 0.26 Tesla — changes in signal intensity following exercise. BritJRadiol 65: 3 9—43, 1992.

Prof R. H. T Edwards Department of Medicine University of Liverpool P. 0. Box 147 Liverpool L69 3BX United Kingdom

Downloaded by: National University of Singapore. Copyrighted material.

S146 mt. J. Sports Med. 13(1992)

Magnetic resonance spectroscopy and imaging of muscle--a physiological approach.

Magnetic resonance spectroscopy (MRS) and imaging (MRI) are now well established techniques for the study of cellular metabolism and gross structure o...
92KB Sizes 0 Downloads 0 Views