Eur J Appl Physiol (2010) 108:641–648 DOI 10.1007/s00421-009-1261-9
ORIGINAL ARTICLE
Quantifying the mechanical work of resting quadriceps muscle tone William Paul Mckay • Philip D. Chilibeck Brian L. F. Daku • Brendan Lett
•
Accepted: 17 October 2009 / Published online: 3 November 2009 Ó Springer-Verlag 2009
Abstract Our laboratory has shown that resting muscle, commonly thought to be mechanically inert, is actually mechanically active. We report a study of the mechanics of resting quadriceps muscle in adult surgical patients that determines how much metabolic activity can be attributed to quadriceps resting-muscle mechanical work. This was calculated by studying the motion of relaxed supine subjects’ instrumented legs dropped onto a pillow before and after anesthesia with muscle paralysis. By subtracting the acceleration of the dropping leg of the conscious subject before the quadriceps is paralysed from that found after paralysis, resting muscle tensile force and power of the quadriceps muscles can be calculated. Mechanomyography was also recorded, using an accelerometer. Paralysis produced an increase in acceleration in all cases (pre-paralysis 6.99 ± 1.51 m s-2; post-paralysis 7.65 ± 1.51 m s-2; P = 0.00007) and a decrease in mechanomyographic mean absolute amplitude (pre-paralysis 10.6 ± 3.7 mm s-2; postparalysis 4.5 ± 2.6 mm s-2; P = 0.00003). Calculated Communicated by Roberto Bottinelli. W. P. Mckay (&) Department of Anesthesia, University of Saskatchewan, RUH 103 Hospital Drive, Saskatoon, SK S7N 0W8, Canada e-mail:
[email protected] P. D. Chilibeck College of Kinesiology, University of Saskatchewan, Saskatoon, Canada B. L. F. Daku Department of Electrical and Computer Engineering, University of Saskatchewan, Saskatoon, Canada B. Lett Faculty of Medicine, University of Saskatchewan, Saskatoon, Canada
force exerted by resting quadriceps was 22.6 ± 16.8 N; power 0.34 ± 0.17 W, corresponding to a daily caloric expenditure of 7.0 ± 3.6 kcal day-1. This corresponds to approximately 205 kcal day-1 for all skeletal muscle. Knowledge of the phenomenon of resting muscle mechanical activity may be of clinically importance in the study and treatment of obesity and of disorders of muscle tone. Keywords Mechanomyography Resting muscle Skeletal muscle Muscle power Muscle tone
Introduction The accuracy of measurement of human metabolic activity has been improved by the doubly-labelled water method, which has shown that metabolic activity is greater than formerly believed (Birmingham and Jones 2002). The cause of this additional metabolic activity, in excess of that measured by older calorimetric methods, remains unexplained. Our laboratory has shown that resting muscle, commonly thought to be mechanically inert, is actually mechanically active; more so after vigorous exercise (McKay et al. 1998, 2004). We report an experiment to study the mechanics of resting muscle in surgical patients, which measures the metabolic activity that can be attributed to resting-muscle mechanical activity; specifically, the resting muscle mechanical tensile force representing work of the quadriceps muscles. This can be measured by subtracting the tensile force after the quadriceps is paralysed from that found in the conscious subject before the muscles are paralysed. Knowledge of the phenomenon of resting muscle mechanical activity may be of clinically importance in the study and treatment of obesity and of disorders of muscle tone.
123
642
It has been known for a long time that contracting muscle makes low frequency ‘‘muscle sound’’ vibrations, normal to the long axis of the muscle (Oster 1984), now known as mechanomyographic (MMG) activity. Resting muscle has been shown in our lab to produce MMG activity, indicating mechanical activity, which decreases with neuromuscular blockade. (McKay et al. 1998) Prior to this work, it had long been conventional wisdom that resting muscle is mechanically inactive because the electromyogram (EMG) is silent. However, early work with a sensitive EMG and careful electrical shielding showed resting muscle EMG activity (deVries 1965; Jacobson 1943; 1951). Correlation with VO2 also has been shown with a sensitive EMG system (deVries et al. 1976). We have shown (McKay et al. 2004) that resting MMG activity is increased after vigorous exercise, and that this activity correlates well with oxygen utilisation (VO2), and that low levels of work also induce measurable increases in resting MMG activity (McKay et al. 2006). How much mechanical work does resting muscle do? This study will answer four questions about this phenomenon. First, what is the tensile (shortening) force that accompanies resting MMG activity? Second, what is the energy expenditure (power) represented by maintaining this force? Third, how is the force/work related to MMG activity? Finally, is the work that is done mechanically by resting muscle enough to be a significant consumer of calories? Hypothesis If resting muscle is performing tensile active work, a relaxed outstretched leg will fall slower before the muscle is paralysed compared to after paralysis. The force and power of resting muscle can be derived from this difference. Rationale Tensile shortening of resting muscle fibres is presumably related to muscle tone, and would be expected to disappear with neuromuscular blockade. Resting muscle MMG activity likewise decreases with neuromuscular blockade (McKay et al. 1998). Muscle tone has been quantified by instrumented models of the Wartenberg test, a test of muscle tone that depends upon allowing the relaxed leg to drop freely from the horizontal position and studying the movement that results (Fee and Miller 2004; Wartenberg 1951). Because it is the most direct approach to measurement of muscle force, we used a modification of the instrumented Wartenberg test applied to the leg-quadriceps unit that studies only the time it takes for the foot to land on a pillow 16 cm below the horizontal.
123
Eur J Appl Physiol (2010) 108:641–648
An experiment just completed in our laboratory (McKay et al. 2009) shows that spinal anesthesia also, as expected, decreases resting VO2 and MMG activity. Oxygen consumption (VO2) decreased 25% with spinal anesthesia, and the mean absolute amplitude of the MMG decreased by 37%. These experiments show that the mechanical activity in resting muscle demonstrated by MMG is due to the innervated contraction of some fibres in the muscle. The power of that contraction can be estimated by measuring its effect on the work exerted on a sudden eccentric load moved over a small incremental distance before and after the muscles are paralysed. A suitable eccentric load for proximal resting muscle is the free fall of a distal part of the relaxed resting limb, since it can reliably stretch resting relaxed muscle. The leg acts as a lever in falling. Knowing the dimensions of the lever, the force provided by these actively contracting fibres can then be calculated by weighing the limb at its centre-of-gravity (COG) and measuring the rate of fall before and after paralysis. Ignoring the very small air resistance, when the limb is released and dropped, any deviation in acceleration from that expected with a frictionless pendulum with the same dimensions are due to retarding forces acting through the knee joint and patellar tendon. The change in those forces before and after denervation are due to innervated muscle activity. We are interested in the accelerations during the first few degrees of drop because they measure what the quadriceps muscles were doing at rest before they were significantly passively stretched. The leg and foot is an irregular solid, not a simple pendulum (which would require that the mass is concentrated at the foot end, and suspended by a weightless rod). Because the mass of the leg is distributed unevenly along the limb, it is necessary to find the COG. The distance of the COG from the knee pivot point, the weight of the limb measured at COG, and the dimensions of the angled lever that is formed by the anatomic relations of the tibia and femur are needed in order to calculate the force (fquad) exerted by the quadriceps. The COG location can be estimated by interpolating from a series of measurements of lower limb circumference at measured distances from the tibial plateau (Fig. 1). It is necessary to know only the distance of the COG from the knee; not where in the transverse section through the leg that the COG is found. To find the COG, the distance from the tibial plateau of each circumference measurement is weighted by the area, which approximately varies as the mass. A suitable approximation can be made by taking circumferences at the tibial plateau, around the foot from mid-heel to around the toes, and a suitable number of equal intervals between (Fig. 1). We estimated that 10 circumferential measurements would give a good approximation. The distance, d, of the COG from the tibial plateau is calculated using
Eur J Appl Physiol (2010) 108:641–648
Fig. 1 Circumferential leg measurements to derive centre of gravity (COG)
P li :ai d¼ P ai
Fig. 2 The tibial lever pivots about point c, with short arm bc, and long arm ab. Tibia with solid outline before dropping; with dashed outline as it drops
Calculating force and power
where li is the length from the tibial plateau to the ith circumference measurement, ai = c2i /4p is the area of the ith measurement and ci is the ith measured circumference. For an earthbound rod (representing the leg) hinged at one end with a frictionless hinge (representing the knee), the COG of the rod falls to earth with an acceleration due to gravity g. The hinged end does not fall, and the parts closer to the hinge fall with acceleration slower than g, while those farther from the hinge than the COG fall with acceleration faster than g, all in proportion to length from the hinge. The COG of the limb will fall a distance Dy in time t according to the equation: 1 Dy ¼ vy0 t þ gt2 2 where vy0 is the initial velocity, in this case, zero, t is time, and g is the acceleration due to gravity at the earth’s surface. Thus, in dropping the foot 16 cm, we can expect the COG to fall about 8 cm, and this would take 0.12 s in a friction-free system. In fact, viscous forces of the stretching muscle, and the frictional forces of the knee joint will slow the acceleration. To calculate the quadriceps tendon movement corresponding to a drop of the COG, it is necessary to include the dimensions of the lever that is formed by the leg. The lever dimensions (Fig. 2) can be obtained by measuring the distance of the COG from the tibial plateau (long arm of the lever; ‘‘a’’ to ‘‘b’’ = [ab] in Fig. 2) and the distance from the femoral condyle pivot on the tibial plateau to the patellar tendon (short arm of the lever; ‘‘b’’ to ‘‘c’’ = [bc] in Fig. 2). Points ‘‘b’’ and ‘‘c’’ are easily visualised with ultrasound, a non-invasive technique with no known adverse effects. Then displacement, by dropping the leg, of the COG (Fig. 3) from d0 to d1 results in a displacement of the patellar tendon and stretching of the quadriceps dq0 to dq1 of approximately (valid for small angles): dq0 dq1 ¼ ðd0 d1 Þ
643
½bc ½ab
Force Force = mass 9 acceleration. The force fquad exerted on the patellar tendon by the activity of quadriceps can be calculated by multiplying the change in acceleration of the COG resulting from neuromuscular blockade by the weight of limb measured at COG corrected for the dimensions of the lever as described above. Thus, fquad ¼ ðaccpre accpost Þðweight at COGÞðab=bcÞ Power Power = work/time = energy expenditure. We assume an insignificant change in heat during the first few degrees of passive motion of the knee. Estimating power then depends upon determining the incremental change in work done to move the tibia relative to the femur at the beginning of knee flexion. When dropped, the leg behaves as a pendulum, and the COG falls through an arc. The attached accelerometer measures acceleration along the path of the arc. The mass of the leg equals the weight measured at COG. Then kinetic energy imparted to the leg by gravity equals mass 9 displacement 9 measured acceleration, and work done by the quadriceps equad equals the difference in e before and after paralysis. Preliminary experimentation in our lab showed that the dropped-leg acceleration, as expected, is less than acceleration due to gravity. A further deceleration due to the dynamic stretch reflex was not observed. Muscle paralysis Muscle paralysis is a routine part of general anesthesia for many surgical procedures. For general anesthesia, neuromuscular blocking drugs are used as a normal part of
123
644
Eur J Appl Physiol (2010) 108:641–648
Fig. 3 Instrumented Wartenburg setup. Above: a leg supported, below: b leg dropped (dashed lines)
clinical anesthesia care for operations requiring muscle relaxation. They function at the neuromuscular junction, and their effect can be monitored by using a nerve stimulator.
Method Subjects With approval of the University of Saskatchewan Research Ethics Board, and signed informed consent, subjects were recruited who met the following inclusion criteria: (1) adults less than 70 years of age, (2) having elective surgery, (3) in the supine position, (4) under general anesthesia with muscle relaxation, (5) with no neuromuscular or joint disorder in the right lower limb, (6) where the right lower limb could be made available for the study for 5–10 min prior to commencing surgery, and (7) in good general health. Equipment The subject was made comfortable on the operating table, the leg section of the operating table was flexed down, and the feet supported on small padded rests attached to the supporting frame. The right-foot rest was supported by a trap-door mechanism to release it and allow the right foot to drop onto a soft sponge pillow on a shelf below,
123
positioned to allow a free-fall of the foot of 16 cm (Fig. 3). The supine patient’s right leg, supported as shown in Fig. 3a), was weighed at the COG by a strap suspended from a load cell (Interface Model SML-5 load cell driven by an Interface SGA; Interface Advanced Force Measurement, Scottsdale, AZ, USA) attached to a metal supporting frame. The instruments for the modified Wartenberg test were applied as follows. MMG over the anterior mid-thigh was recorded with a Bruel & Kjaer (B&K) #4381 accelerometer (54gm, 2 9 2 cm; Bru¨el & Kjær, Nærum, Denmark), powered by a B&K #2635 charge amplifier, and calibrated with a B&K Type 4294 Calibration Exciter. The charge amplifiers were set to provide bandpass pre-filtering of the signal from 0.2 to 100 Hz. Electrocardiogram (ECG) leads for lead II ECG were attached to enable digital pulse removal from the MMG signal. The ECG was recorded from the hospital monitor (M-series—various models, Hewlett-Packard, Palo Alto, CA, USA). Leg acceleration motion was measured with a second Bruel & Kjaer (B&K) #4381 accelerometer. Surface EMG was obtained with a Delsys model Bagnoli 2-Channel EMG (Delsys Inc., Boston, MA, USA). The skin was prepared for the EMG by scrubbing with an alcohol swab, then with gauze, in order to maximise electrical conductivity. As reference electrode (ground), a commercially available silver–silver chloride electrode (Meditrace 200, The Ludlow Company LP, Chicopee, MA,
Eur J Appl Physiol (2010) 108:641–648
USA) was attached over the patella. The EMG includes single differential amplifier with a bandwidth of 20 ± 5 Hz to 450 ± 50 Hz, a 12 dB/octave cut-off slope, with a Common Mode Rejection Ratio (CMRR) of 92 dB, and a maximum output voltage range ±5 volts. The overall amplification or gain is 10009. The system noise is \1.2 lV (rms) for the specified bandwidth. The sensing electrodes are two silver bars (10 mm 9 1 mm diameter) spaced 10 mm apart. The EMG was recorded as raw EMG (volts) and digitised for analysis. Digitised signal recordings employed a National Instruments BNC 2090 DAQ board (National Instruments, Austen, TX, USA) attached to a laptop computer. Analogue outputs were sampled at 1 kHz. Biomechanics of the Wartenberg test The classic Wartenberg test depends upon allowing a supported limb to drop freely under the force of gravity, assessing its rate of fall, and also assessing its behaviour as a pendulum that is allowed to swing freely by the subject. Wartenberg’s original description is of a bedside test without instrumentation that relies upon the judgement of the clinician. We have modified the Wartenberg test to overcome several issues as follows: 1.
2.
3.
4.
It requires voluntary relaxation by the subject prior to the test. This is easily achievable for most people, and is easily monitored by MMG and EMG, which both register a large transient increase in amplitude that is an order of magnitude or more in size with any voluntary motion. It requires that there be no ‘‘tensing up’’ as the subject senses that the limb is about to be released; the subject must remain relaxed, suppressing the voluntary tensing that is induced by the sudden fall of the limb, during the fall. This is made easy by the knowledge that the foot is dropping a short distance into a soft pillow, and by a few practice drops. The dynamic stretch reflex, mediated by muscle spindle fibres and type Ia proprioceptive nerves, induces contraction with sufficient rapid stretching of skeletal muscle (Guyton and Hall 2001), and can be detected by EMG. We placed surface EMG electrodes on mid-quadriceps and on mid-hamstring muscles. The dynamic stretch reflex can then be distinguished because the quadriceps EMG becomes active at its onset, when its amplitude would greatly exceed that of the hamstring EMG signal, which can be expected to be actively suppressed by the dynamic stretch reflex. Visco-elastic tissue friction (of deforming skin, fat, and connective tissue) slows the dropping leg by an
645
5.
6.
unknown amount that is unique to each individual. However, it is unaffected by paralysis. The passive properties of stretched denervated muscle, which have been extensively studied—reviewed in Proske and Morgan (1999), come into play. The nonlinear nature of this impedance to stretching is a confounder that is removed by studying only the first small increment of drop. Full extension of the hamstring muscles prior to the drop may result in a stretched-spring effect from the elastic component of the hamstring muscle which could impart an acceleration greater than gravity to the falling leg. This was eliminated by beginning with knee flexed to 5 degrees.
Procedures Following recruitment, and prior to entering the operating room: 1. 2. 3.
4.
The patient’s leg was marked and measured as in Fig. 1. An ultrasound of the knee was used to find the dimensions of the lever’s short arm as in Fig. 2. After entering the operating room, subjects lay supine upon the operating table, and were encouraged to relax. The hinged leg-support section of the operating table was lowered and both legs supported by the padded foot-supports. EMG, MMG, and ECG monitors were attached. Any visible motion, no matter how small, results in large, obvious transient signals from the MMG. When these are absent, we consider the subject relaxed. The experimental limb was instrumented with the accelerometer over the COG, the MMG and EMG over the mid-thigh, and the second EMG over mid-hamstrings, all with tape or doublesided tape. The leg was released to fall on a pillow supported on a shelf 16 cm lower than the initial support position, and the MMG, EMG, and acceleration recorded (repeated three times). The subject was anesthetised for surgery and paralysed with rocuronium. Paralysis was ensured with a nerve stimulator applied to the left ulnar nerve at the wrist (Digistim model III, Neuro Technology, Houston, TX, USA). The paralysed limb was replaced on the foot support, and the experimental leg-drop repeated three times. Then the equipment was removed, and the surgery performed as usual.
The experiment added about 5–10 min to the operating room time.
123
646
Statistics and analysis Number of subjects Because there is no literature describing the variance of resting muscle tensile force, formal sample size calculation was not possible. Thus, sample size is based upon our previous studies, for which ten subjects provided significant results. In the 1998 study, ‘‘Resting muscle sounds in anesthetized patients’’, the effect size on MMG amplitude of paralysis was to decrease it by 54% (P \ 0.05) with ten subjects. None of the subjects of this earlier study was elderly, so we avoided recruiting elderly subjects. Thus, we planned to study ten consenting healthy adult volunteers 18–69 years.
Eur J Appl Physiol (2010) 108:641–648 Table 1 Leg drop measurements Measurement
Mean ± SD [95% Confidence intervals]
Leg weight at COG
5.16 ± 1.67 kg [4.13–6.2]
Tibial plateau to COG
18.1 ± 2.14 cm [16.8–19.4]
Femoral condyle pivot to patellar tendon Leg length
2.95 ± 0.6 cm [2.57–3.32]
Acceleration of COG pre-paralysis
6.99 ± 1.51 m s-2
Acceleration of COG post-paralysis 7.65 ± 1.51 m s-2 Acceleration difference
0.66 ± 0.33 m s-2, P = 0.00007 [0.424–0.903 m s-2]
Acceleration percent change
110 ± 5.5%
Force of quadriceps at rest
22.6 ± 16.8 N [12.1–33]
Power
0.34 ± 0.17 W [0.23–0.44]
Analysis Weight and acceleration measurements were made from the digitised graphed signals. Tensile resting muscle force and power were calculated as described above. MMG signals were processed with pulse removal as described previously in detail (McKay et al. 2004) and the mean absolute value of the zeroed signal found. EMG signals were examined and root-mean-square voltage and time after leg drop determined for areas of increased activity.
38.4 ± 3.17 cm [36.4–40.4]
= 29.2 ± 14.9 KJ day-1 MMG pre-paralysis
10.6 ± 3.7 mm s-2
MMG post-paralysis MMG difference
4.5 ± 2.6 mm s-2 -6.1 ± 2.7 mm s-2; P = 0.00003 [-4.178 to -8.007]
Discussion
Statistics
Force
Pre- and post-paralysis measurements were compared by paired t test.
The resting muscle tensile force of the quadriceps muscle group in this sample of middle-aged people presenting for elective surgery was 22.6 ± 16.8 N. Hsu et. al. report that in normal subjects in their twenties, standing on one leg with knee flexed to 45 degrees and all motor units recruited, the quadriceps generates a force equivalent to twice the body weight (Hsu et al. 1993). For our subjects, this would be 1,508 ± 310 N. Thus, the force maintained by resting quadriceps muscle represents activation of about 1.5% of motor units.
Results Demographics Ten subjects were recruited, age (mean ± standard deviation) 47 ± 11 years; 8 female, 2 male; height 166.8 ± 9.4 cm; weight 77.4 ± 15.8 kg. Surgical procedures: 2 laparotomies, 4 laparoscopies, and 4 hernia repairs. The experiment was completed on all ten subjects. Force and power See Table 1 for measurements. All had an increase in COG acceleration with paralysis. MMG See Table 1 for measurements. Mean absolute MMG decreased after paralysis to 57% of the awake level.
123
Power Mechanical energy expenditure varies as force; generating this force in the resting quadriceps would burn 29.2 ± 14.9 kJ day-1, or 7 Cal day-1. Thus, the energy expenditure we found with activation of about 1.5% of motor units in relaxed resting quadriceps muscle accords with early evidence that the phenomenon of resting muscle mechanical activity, which is found in homeotherms, but not poikilotherms, may account for maintaining body temperature at ambient room temperature (Roracher calculated that body temperature could be maintained if 2% of skeletal
Eur J Appl Physiol (2010) 108:641–648
muscle was active.) (deVries et al. 1976; Rohracher 1959, 1962; Sugano 1960). This experiment suggests that resting-muscle mechanical activity constitutes a small but measureable part of resting metabolic activity. Our results agree within the limits of resolution with Roracher’s calculations. MMG Resting muscle MMG mean absolute amplitude decreased by 57% P = 0.001 following paralysis. This is greater than the 37% decrease found after spinal anesthesia (McKay et al. 2009). This is to be expected from the more complete paralysis with neuromuscular blocker than is to be found after a spinal anesthetic that has had some time to wear off. EMG The EMG was attached over mid-quadriceps rectus femoris and over the muscle belly of biceps femoris. The biceps femoris EMG did not show measurable activity. The EMG over rectus femoris did not show activity during the footdrop that would have signified a response to the dynamic stretch reflex.
647
play. Third, the recordings showed no sudden deceleration of the leg before it struck the pillow. Resting muscle active force involves only 1.5% of maximum force (see below), a very small quantity. Did our measurements have sufficient resolution to measure this? Worst case analysis involves the additive or multiplicative effect of the worst errors of each measurement upon an outcome measure derived from them. Force and power in this experiment are such outcomes that involve all the measurements (leg dimensions, mass, and acceleration) in a multiplicative fashion. The force and power results cannot be negative (a muscle can only pull, it cannot push). Thus, if the lower 95% CIs for either force or power were negative (that is, if the CIs bracket zero), it would be good evidence that our measurement or instrument errors were greater than the required level of resolution, and would cast doubt on the results. This is not the case for either the primary measurements or the derived outcomes, as the CIs in Table 1 show. As well, and importantly, all of the subjects showed changes in the same direction; that is to say, all accelerated faster after muscle paralysis. Had any behaved otherwise, it would have required some other plausible explanation, or would have cast doubt on all the results. Metabolism and obesity
Possible weaknesses of the experiment The stretch reflex is a possible confounder that could cause deceleration of the dropped leg in the awake subject. There are three reasons for discounting the stretch reflex as a confounder. First, the surface EMG over rectus femoris did not show activity during the foot-drop that would have signified a response to the dynamic stretch reflex. Although we recorded EMG only over mid-anterior thigh, where the strongest signal would come only from vastus intermedius and rectus femoris, we expect that the muscles most likely to produce a signal suggesting quadriceps contraction in response to the sudden stretching would be those most stretched—vastus intermedius and rectus femoris. It seems unlikely that these muscles would maintain EMG silence while their less-stretched neighbours, vastus medialis and lateralis, were activated. Second, we compared the time course of the leg drop to data from a paper by Kamen (Kamen and Koceja 1989), who recorded leg movement in response to a patellar tap (with a reflex hammer) in young (mean age 25.4 years) and old (mean age 72.5 years) adults. The time for the leg to fall to the pillow [95% CI: 35–43 ms] was well below the lower CI of the time to onset of leg movement in Kamen’s paper, even in young adults (young adults [54–62 ms]; old adults [91–99 ms]). Thus, the leg in our experiment would have dropped before the stretch reflex could come into
We found that the right quadriceps at rest performs mechanical work that consumes 7 kcal day-1. How much energy utilisation might this represent for total skeletal muscle? We were not able to measure quadriceps or total muscle volume, but useful estimates can be found in the literature. A recent study published a graph summarising quadriceps volumes measured in a variety of adults by various methods (Tate 2006). We took measurements from Fig. 1 of that article and found a mean volume of 0.86 ± 0.15 L. Using the method of Clarys et al. (1999), we calculated total skeletal muscle of 25.2 ± 7.2 L. The energy expenditure of the quadriceps thus suggests that resting skeletal muscle mechanical activity may account for consumption of 205 kcal day-1 under the circumstances (ambient temperature, hospital clothing) pertaining to our subjects. All of these calculated estimations and assumptions must be tested with further direct experimental research. However, the estimates suggest that such research should be pursued. The phenomenon of resting muscle mechanical activity has only recently been clearly demonstrated. It is an area of muscle physiology whose ultimate importance is unknown. Our lab has clearly demonstrated its existence; has shown that it increases almost threefold after vigorous exercise; that it shows a proportionate increase after mild exercise; that it exhibits exponential decay after exercise with a time constant of about 7 min; and that the increased activity
123
648
persists for more than 5 h after very vigorous aerobic exercise. Thus, it seems plausible that it may play a significant role in resting metabolism. This experiment is one of a set of studies to characterise and quantify the mechanical work being done by apparently resting muscle that may contribute to the understanding of obesity and that may suggest studies to improve exercise prescription. Implications for metabolism Tensile resting muscle mechanical work has been shown here to account for a small but significant caloric expenditure. Further questions immediately follow. First, as to the pathophysiology of obesity: is the resting muscle work different in naturally thin than in naturally stout people? Does exercise affect it differently? Second, what is the effect of various exercise regimens on increasing resting muscle activity? For example, if, as we have shown, resting muscle activity is increased for a time after exercise, do frequent short workouts several times a day burn more calories than the equivalent total exercise done once a day? Most of the large literature on the optimal time-course of physical training applies to training athletes rather than to treating obesity. This study may have implications for exercise prescription in obesity. This paper reports, as a first step, measuring the work of a single resting muscle group. Implications for muscle tone The passive resistance of resting muscle to stretch has long been an important physical sign of neurological disorders. Muscle tone is notoriously difficult to quantify at the bedside (Ward 2000). This experiment may lead to easier objective clinical measurement of muscle tone. It seems plausible that what we call muscle tone is largely due to the phenomenon of resting muscle mechanical activity measurable by MMG. Resting muscle MMG, or a modification thereof may be useful in monitoring treatment of patients with disabling muscle tone disorders, for example patients with stroke, multiple sclerosis, or Parkinson’s disease.
References Birmingham CL, Jones PJ (2002) Clinical nutrition: 5. How much should Canadians eat? CMAJ 166:767–770
123
Eur J Appl Physiol (2010) 108:641–648 Clarys JP et al (1999) Human body composition: a review of adult dissection data. Am J Human Biol 11:167–174 deVries HA (1965) Muscle tonus in postural muscles. Am J Physical Med 44:275–291 deVries HA, Burke RK, Hopper RT, Sloan JH (1976) Relationship of resting EMG level to total body metabolism with reference to the origin of ‘‘tissue noise’’. Am J Phys Med 55:139–147 Fee JW Jr, Miller F (2004) The leg drop pendulum test performed under general anesthesia in spastic cerebral palsy. Dev Med Child Neurol 46:273–281 Guyton AC, Hall JE (2001) Textbook of medical physiology, 10th edn. WB Saunders Company, Philadelphia, p 626 Hsu AT, Perry J, Gronley JK, Hislop HJ (1993) Quadriceps force and myoelectric activity during flexed knee stance. Clin Orthop Relat Res 288:254–262 Jacobson E (1943) Innervation and tonus of striated muscle in man. J Nerv Ment Dis 97:197–203 Jacobson E (1951) Muscular tension and the estimation of effort. Am J Psychol 64:112–124 Kamen G, Koceja DM (1989) Contralateral influences on patellar tendon reflexes in young and old adults. Neurobiol Aging 10:311–315 McKay WP, Gregson PH, McKay BW, Blanchet T (1998) Resting muscle sounds in anesthetized patients. Can J Physiol Pharmacol 76:401–406 McKay WP, Chilibeck PD, Chad KE, Daku BL (2004) Resting mechanomyography after aerobic exercise. Can J Appl Physiol 29:743–757 McKay WP, Jacobson P, Chilibeck PD, Daku BL (2006) Effects of graded levels of exercise on ipsilateral and contralateral postexercise resting rectus femoris mechanomyography. Eur J Appl Physiol 98:566–574 McKay WP, Lett B, Chilibeck PD, Daku BL (2009) Effects of spinal anesthesia on resting metabolic rate and quadriceps mechanomyography. Eur J Appl Physiol 106:583–588 Oster G (1984) Muscle sounds. Sci Am 250:108–114 Proske U, Morgan DL (1999) Do cross-bridges contribute to the tension during stretch of passive muscle? J Muscle Res Cell Motil 20:433–442 Rohracher H (1959) Staendige Muskelaktivitaet (Mikrovibration), Tonus und Konstanz der Korpertemperatur. Z f Biol 111:38–53 Rohracher H (1962) Permanente rhythmische Mikrobewegungen des Warmbluter-Organismus ‘‘Mikrovibration’’ (review; German only). Die Naturwissenschaften 49:145–150 Sugano H (1960) Studies on minor tremor. Japanese J Physiol 10:246–257 Tate CMWG (2006) Lower extremity muscle morphology in young athletes: an MRI-based analysis. Med Sci Sports Exerc 38:122– 128 Ward AB (2000) Assessment of muscle tone. Age Ageing 29:385– 386 Wartenberg R (1951) Pendulousness of the legs as a diagnostic test. Neurology 1:18–24
