Equal proximal and distal lengthening of rat extensor digitorum longus (EDL) were studied. Tibialis anterior, extensor hallucis longus, and EDL were active maximally. The connective tissues around these muscle bellies were left intact. Proximal EDL forces differed from distal forces, indicating myofascial force transmission to structures other than the tendons. Higher EDL distal force was exerted (ratio ≈118%) after distal than after equal proximal lengthening. For proximal force, the reverse occurred (ratio ≈157%). Passive EDL force exerted at the lengthened end was 7–10 times the force exerted at the nonlengthened end. While kept at constant length, synergists (tibialis anterior + extensor hallucis longus: active muscle force difference ≈ −10%) significantly decreased in force by distal EDL lengthening, but not by proximal EDL lengthening. We conclude that force exerted at the tendon at the lengthened end of a muscle is higher because of the extra load imposed by myofascial force transmission on parts of the muscle belly. This is mediated by changes of the relative position of most parts of the lengthened muscle with respect to neighboring muscles and to compartment connective tissues. As a consequence, muscle relative position is a major codeterminant of muscle force for muscle with connectivity of its belly close to in vivo conditions.
- anterior crural compartment
- connective tissue
- length-force characteristics
- muscle proximo-distal force difference
length is generally recognized as one of the major determinants of force exerted by muscle. As a consequence, since Blix (4-7) reported his experiments more than a century ago, isometric length-force characteristics have been the subject of extensive research. Work was performed at all levels of organization, i.e., muscle (e.g., Refs. 8, 14, 16,35, 44, 45, 49,55, 57, 58, 72), small muscle fascicles (e.g., Ref. 75), and single-muscle fibers (e.g., Refs. 17, 33, 34,36, 48, 56).
In most of such work, even if performed in situ, the muscular elements of interest were experimented on after substantial dissection to free it as much as possible from surrounding connective tissues and other structures. Force exerted at one end of the fiber, fascicle, or muscle was measured, based on the mostly implicit assumption that force exerted at either end would be equal.
For in situ rat muscle, with most of its connective tissue at its belly intact, we recently showed that this is predominantly not the case (25, 28, 39). The size of any difference in force (ΔF) exerted at proximal and distal tendons of EDL muscle was shown to be quite dependent on the actual length of the muscle tendon complex (25, 28, 39) (i.e., −22% < ΔF < +22.5% and 0% < ΔF < +20.0%, respectively). In addition, blunt dissection of intermuscular connective tissue, as well as full compartmental fasciotomy, did alter muscle length-force characteristics (27,59). These effects are explained in terms of myofascial force transmission from muscle i.e., transmission by other paths than the myotendinous ones. In most of the studies referred to in this paragraph, the effects were studied in either proximal lengthening of rat extensor digitorum longus (EDL) muscle or distal lengthening of its synergists, i.e., tibialis anterior (TA) and extensor hallucis longus (EHL) muscles exclusively.
EDL is a unipennate muscle with one proximal aponeurosis from which all of its muscle fibers originate. The insertions of these muscle fibers are divided over four distal aponeuroses. Therefore, EDL is distally a multitendon muscle with connections to digits II–IV (e.g., Refs.1, 9). EDL extends the toes and is also a dorsal flexor of the ankle. In rat, the unique proximal EDL tendon crosses the knee joint (knee extensor). Such arrangement allows experimental access to the EDL proximal tendon without dissecting the anterior crural compartment. Having access to both proximal and distal tendons allows convenient simultaneous proximal and distal force measurement. It also allows changes in muscle-tendon complex length to be easily imposed on EDL at either end of the muscle. Short intermuscular connective tissue connects the muscle belly of EDL to that of TA and EHL. TA completely surrounds EDL as well as EHL. TA has a very complex architecture, with muscle fibers originating both from bony origins and from a proximal aponeurosis and inserting on a distal tendon. TA is predominantly an ankle dorsal flexor. All muscle fibers of EHL originate from the anterior intermuscular septum and insert on a distal aponeurosis that is connected to a tendon, which is in very close proximity to the TA tendon, before passing to the big toe. As these muscles do not share aponeuroses, but do have connective tissue links to each other and to the compartment boundaries, they are very useful for studying myofascial force transmission.
In the present work, we studied the effects of symmetric lengthening of rat EDL to identical muscle-tendon complex lengths. Identical target muscle-tendon complex lengths were reached by lengthening EDL at its proximal or at its distal tendon before isometric contraction. The effects on isometric force exerted at each of the proximal and distal EDL tendons and on the expected proximodistal EDL ΔF are studied. In addition, the effects of such symmetric lengthening of EDL to identical muscle-tendon complex lengths on neighboring synergists (TA+EHL) are considered as well.
The principal aims of the present work are to compare the effects of equal proximal and distal lengthening of EDL muscle and test the hypotheses that such effects of equal lengthening on 1) proximal and distal force exerted at the EDL tendons and on2) the expected EDL proximodistal ΔF are equal in magnitude and thus exclusively a function of muscle-tendon complex length. 3) A similar hypothesis is tested for the effects of changes of EDL muscle-tendon complex length on neighboring synergists.
MATERIALS AND METHODS
Surgical and experimental procedures were in strict agreement with the guidelines and regulations concerning animal welfare and experimentation set forth by Dutch law and were approved by a Committee on Ethics of Animal Experimentation at the Vrije Universiteit, Amsterdam. Immediately after all experiments, double-sided pneumothorax was performed, and animals were killed with an overdose of pentobarbital.
Surgical Procedure and Preparation for Experiment
Male Wistar rats (n = 6, mean ± SD body mass: 301 ± 16.25 g) were anesthetized by intraperitoneal injection of a urethane solution (initial dose, 150 mg/100 g body mass). Supplementary doses of the anesthetic agent (0.62 mg) were injected intraperitoneally (maximally three times), if necessary, to maintain deep anesthesia. The animals were placed on a heated water pad (37°C) during surgery and experimentation.
The left femoral compartments were opened to 1) cut the femur transversely to allow later fixation within the experimental setup; 2) reach the insertion of the proximal tendon of EDL muscle; and 3) dissect the sciatic nerve and cut it as proximally as possible. The sural, tibial, and articular branches of the sciatic nerve were cut so that, by stimulating the sciatic nerve during the experiment, the full motor segment of the common peroneal nerve would be excited exclusively.
Anterior crural compartment.
In the rat, connective tissue associated with the biceps muscle covers the anterior compartment to reach a very long insertion along the tibia. Removing the skin, parts of the crural fascia, and the biceps femoris muscle exposed the anterior crural compartment of the left leg. The very distal part of the anterior tibial compartment had to be opened (local fasciotomy) to reach the distal tendons of EDL and of the EHL and TA muscles. With the knee joint at ∼100° and the angle between the footplate and the tibia at 90° (referred to as the reference position), two sets of distal tendons were tied together:1) the four adjacent distal tendons of EDL (using polyester thread); and 2) the distal EHL tendon, which was tied to the adjacent distal tendon of TA (using polyester thread). The complex of TA and EHL created this way will further be referred to as TA+EHL.
Matching markers were placed on the distal tendons of EDL and of TA and EHL, as well as on a fixed location on the lower leg.
Similar to the anterior crural compartment, the peroneal compartment was opened only distally to reach the distal tendons of the peroneal muscle group. This group consists of the four peroneal muscles (i.e., mm. peroneus longus, brevis, quarti, and quinti), which fill most of the compartment. Their distal tendons were dissected free from surrounding tissues, leaving the compartmental borders and connective tissues around the muscle bellies fully intact. The four adjacent distal peroneal tendons were tied together by using polyester and Kevlar threads.
Further treatment of the tendons of target muscles.
The retinaculae at the ankle (i.e., transverse crural ligament and the cruciate ligament) were removed while under observation through a dissection microscope (Wild, magnification ×6–40). Subsequently, tenotomy was performed as distally as possible on the distal EDL and TA+EHL as well as peroneal tendon complexes. The severed tendons were removed from their retinaculae near the ankle joint (transverse crural ligament and cruciate ligament). The proximal tendon of EDL was cut loose from the femur, with a small piece of the lateral femur condyle still attached.
With the use of polyester threads, each of the distal tendon complexes as well as the proximal EDL tendon were sutured to Kevlar threads with a small loop at their end (proximal or distal, respectively).
Subsequently, the left foot of the rat was attached firmly to a plastic footplate by using a Kevlar thread.
If necessary, the tendons were irrigated with an isotonic saline solution to prevent drying.
All Kevlar threads used (Goodfellow) are characterized by a diameter of 0.2 mm and a tensile modulus of 58 GPa. All polyester threads used (Gütermann) had a diameter of 0.3 mm.
Mounting the Animal in the Experimental Apparatus
The cut femur was secured by means of a self-tapping screw (material: electrolytic zinc-plated steel, diameter: 2 mm). This screw was screwed into the femur through the knee toward the tibia and connected with a stiff aluminum rod (diameter: 8 mm) to the experimental setup (for a schematic view, see Fig.1 A).
The plate with the left foot attached was manipulated such that the ankle was in extreme plantar flexion. Such a position makes room for passage of distal tendon complexes and their attached Kevlar threads (Fig. 1). Subsequently, the plate was firmly attached to the experimental setup with the ankle joint in extreme plantar and ∼40° of supination.
The TA+EHL complex was brought to a muscle-tendon complex length corresponding to a summed distal active force of ∼3 N. Subsequently, the peroneal muscles were set a muscle-tendon complex length corresponding to a summed distal active force of ∼5 N. Such setting of muscle-tendon complex lengths as defined by force exerted has been shown to yield highly reproducible settings.
The EDL distal tendon group was set at a position corresponding approximately to the reference position.
All four Kevlar threads attached to tendons were connected to force transducers (Hottinger Baldwin, maximal output error <0.1%, compliance 0.0048 mm/N). For proximal EDL as well as TA+EHL and the peroneal complex, the Kevlar wires were guided over low-friction pulleys, which were shown not to affect force measurements before the experiments.
The three-dimensional coordinates of the force transducers were manipulated to obtain orthogonal orientations with respect to the Kevlar threads.
The severed end of the sciatic nerve was placed on a bipolar stimulation electrode.
Experimental Procedure and Data Collection
To make sure that any differences in force transducers and their calibration before the experiment introduced no artifact, the two force transducers to be used for measurement of EDL forces were connected to each other with a compliant spring. The output was recorded with the same measurement system [i.e., amplifiers, analog-to-digital (A/D) converters] used in the animal experiment. It is concluded that any major difference in force (>1.36%) at these transducers cannot be ascribed to the measurement system used. In addition to that, the locations of these proximal and distal force transducers for EDL were exchanged in one-half of the experiments.
During the experiments, ambient temperature was kept constant at 22 ± 0.5°C, and air humidity was kept at 80 ± 2% by a computer-controlled air-conditioning system (Holland Heating), creating a downflow of air onto the experimental table. The surface of the anterior crural compartment was rinsed regularly with saline to prevent fluid loss.
TA+EHL and EDL muscles (innervated via the deep peroneal nerve) and also the whole peroneal muscle group (innervated by the superficial peroneal nerve) were excited simultaneously. This was done by stimulating the distal end of the severed, sciatic nerve supramaximally, by using a pair of silver electrodes connected to a constant-current source (square pulse width 100 μs, pulse train 400 ms, 100 Hz). In the preparatory phase of each experiment, current was increased in small steps until no further increase in force was attained. In this condition, currents of ∼3 mA were necessary. The constant-current mode of the stimulator delivers the set amplitude of current, even if changes in nerve impedance should occur during the experiment, thereby helping to maintain maximal excitation of the nerve during the course of the experiment. To prevent drying of the nerve, the exposed part was covered with tissue paper and saturated with isotonic saline, which itself was covered by a thin layer of latex.
After lengthening of EDL muscle to any desired target length, two twitches were evoked (200 ms apart). Passive force was determined ∼600 ms after the second twitch. Almost immediately after that, the muscles were excited tetanically. During the tetanic plateau (i.e., 275 ms after evoking tetanic stimulation), total isometric muscle force was determined. All force signals were acquired by using an A/D converter (sample frequency 1,000 Hz, resolution of force 0.01 N) and recorded on a microcomputer. A special-purpose microcomputer controlled timing of events related to stimulus generation as well as A/D conversion.
After each isometric contraction, the muscles were allowed to recover for 2 min. For EDL, recovery was allowed to occur near active slack length. The positions of distal tendons of the TA+EHL complex, as well as of the peroneal complex, were kept constant throughout the experiment. As the origins of these muscles were not treated in any way, the muscle-tendon complex length of these muscle groups was constant during experiments.
For the experimental protocol applied, three phases can be distinguished, which were performed in the order shown below.
Initial contractions to eliminate effects of previous activity at high length.
Specific care had been taken that experimental muscles did not attain the previous high length of this part of the experiment. Isometric tetanic EDL force was measured at a test length (i.e., a length at which ∼0.5 N of active distal force was exerted) with simultaneous measurements of force of the TA+EHL complex, while at constant complex length. Subsequently, EDL was brought to high muscle-tendon complex length (i.e., ∼1 mm over optimum length) and activated isometrically. After the recovery period near active slack length, EDL was brought again to the test length and activated. Because of the previous activity at high length, force at the test length was decreased substantially (see also Ref. 28) without changing optimum force.
The procedure described was repeated (∼4–5 times) until the previous activity at high length no longer changed the force exerted at the test length.
Distal lengthening for determination of EDL length-force characteristics.
EDL muscle-tendon complex length changes were imposed first by moving the distal force transducer (1-mm increments, as determined on a vernier mechanism read to the nearest one-tenth of a millimeter) in between contractions. Proximal tendons were kept at reference position. Length-force data were obtained, starting from active slack length [i.e., the lowest length at which active muscle force (Fma) approaches zero] and ending at ∼2 mm over optimum length.
Proximal lengthening for determination of EDL length-force characteristics.
Subsequently, EDL muscle-tendon complex length changes were imposed by moving the distal force transducer (1-mm increments, as determined on a vernier mechanism read to the nearest one-tenth of a millimeter) in between contractions. Distal tendons were kept at reference position. Length-force data were obtained starting from active slack length and ending at ∼2 mm over optimum length.
Treatment of Data
The individual length-force data sets for passive muscle force (Fmp) and muscle length were fitted with an exponential curve (Eq. 1 ), using a least squares criterion Equation 1where y is Fmp, x is passive muscle-tendon complex length [i.e., deviation from minimal length (Δl)], and a 1 anda 2 are coefficients determined in the fitting process. Fma was estimated by subtracting passive force calculated according to Eq. 1 for the appropriate active muscle-tendon complex length from the total force exerted by the muscle at that length.
Data for active EDL force (Fma) in relation to active muscle-tendon complex length (Δl) were fitted by using a polynomial Equation 2where y is Fma; x is length of the active muscle-tendon complex; n is the order of the polynomial; and b 0, b 1,b 2 … bn are coefficients determined in the least squares fitting process. The fitting started with a first-order polynomial, and the power was increased up to and including the sixth order. Polynomials that best described the experimental data were selected (see below). These polynomials were used for three purposes: 1) determining EDL optimal force and 2) optimum length, and 3) averaging data and calculating standard errors. For each individual muscle, optimal muscle force (Fmao) is defined as the maximum of the fitted polynomial for Fma, and optimum muscle-tendon complex length is defined as the corresponding active length.
Individual data for muscle-tendon complex length are expressed as deviations from minimal EDL length.
In the fitting procedure, one-way ANOVA (50) was used to select the lowest order of the polynomials that still added a significant improvement of the description of changes of muscle-tendon complex length and muscle force data for EDL.
Two-way ANOVA for repeated measurements (factors: EDL muscle-tendon complex length and location of EDL force measurement) was performed to test for effects on muscle length-force characteristics measured simultaneously. The same procedure (factors: EDL muscle-tendon complex length and location of EDL lengthening) was applied also on data on distally exerted EDL force, as well as on proximally exerted force, to test for differential effects of proximal and distal lengthening.
If significant effects were found, post hoc tests were performed by using the Bonferroni procedure for multiple paired comparisons, to further locate significant differences.
To test for any differences in proximal and distal optimum length as well as forces exerted at reference length (l ref) and position or to test further for length effects, one-way ANOVA for repeated measurements was performed.
Any differences at P ≤ 0.05 were considered significant.
Differences in EDL Length-Force Characteristics Measured Simultaneously at Proximal and Distal Tendons
To facilitate comparison of proximal and distal EDL forces, proximal force is represented as a reaction force (i.e., −Fma or −Fmp).
Figure 2 shows isometric length-force curves as measured simultaneously at both distal and proximal EDL tendons and obtained after distal lengthening (Fig. 2 A) and after proximal lengthening (Fig. 2 B) of EDL muscle, respectively. This classic way of plotting emphasizes a comparison of force for equal muscle-tendon complex lengths. Corresponding values of muscle-tendon complex length normalized for optimum length are shown in Table 1. For both active and passive EDL forces, ANOVA indicates significant effects for factor length and factor proximal or distal location of force measurement, as well as interaction between these factors.
At the tendon at which EDL was lengthened, isometric active force rises much faster with length changes than active force measured at the other (fixed) tendon. This is true for both distal and proximal EDL active forces. Note, however, that optimum lengths for length-force characteristics measured at the tendon at the lengthened end of EDL or at its nonlengthened end are similar (Fig. 2, A andB, respectively).
Proximodistal differences of force.
Active forces exerted at proximal and distal tendons of EDL are not equal, except at one particular length (crossover of curves). Note that crossover length is similar for proximal and distal lengthening (Fig.2 A). For the present experimental conditions, no significant differences could be shown for proximal and distal active force atl ref. This means that proximal and distal forces are equal at a length very close to EDL l ref(mean ± SE = 41.27 ± 0.49 mm).
Over crossover length, for distal lengthening, distal EDL active force is higher than proximal force, but for proximal lengthening the reverse is true. This means that, for high lengths, active force exerted at the location of lengthening dominates the active force exerted at the fixed end of EDL over a wide length range.
In contrast, for the smaller length range from active slack lengths up to the crossover length, isometric active force measured at the location of lengthening remained lower initially (−Fmameasured at the proximal tendon < Fma measured at the distal tendon on proximal lengthening and vice versa).
It is concluded that proximal and distal lengthening affect the relation of force exerted at the proximal tendon and at the distal tendon in opposite ways.
Any proximodistal difference in force indicates myofascial force transmission (extra- and or intermuscular) from or onto EDL. Such myofascial force transmission can even cause one EDL tendon to be slack (force = 0), whereas, at the other tendon, a substantial force is still exerted (Fig. 2, A and B; compare values at the respective active slack lengths). As the tendon at the location of lengthening approaches slack, force exerted at the fixed tendon may be substantial (Fma = 0.27 N, Fig. 2 A; Fma = 0.19 N, Fig. 2 B). This represents ∼16.5 and 9% of the optimal value of proximal force after distal lengthening and of distal force after proximal lengthening, respectively. In these specific conditions, it is clear that all force exerted at the tendon at which the muscle is not lengthened is transmitted from the muscle, without myotendinous force transmission to the other tendon.
Values of the proximodistal active ΔF are quite dependent on the location of lengthening of the muscle. For distal lengthening, this difference ranges from approximately −0.27 N < ΔFma < +0.90 N. For proximal lengthening, the range of ΔF is more limited (i.e., −0.19 N < ΔFma< +0.41 N). Even though these absolute values do not seem high, compared with the active force actually exerted by the muscle, the proximodistal active ΔF values represent very substantial fractions of the actual highest force exerted at a particular length.
It is concluded that the opposite effects of proximal and distal lengthening on active force are not symmetric. More active force is transmitted via myofascial pathways from or onto EDL after distal lengthening than after proximal lengthening.
Also for passive force at higher lengths, EDL force exerted at the tendon at the location of lengthening is considerably higher than that at the fixed tendon. At high lengths, passive force may be as much as 9–10 times higher (Fig. 2, C and D) than exerted at the fixed tendon. This indicates that, throughout the length range over crossover length, most of the passive force exerted at the tendon of the lengthening location is transmitted to or from extramuscular tissues (extramuscular myofascial force transmission) or other muscles (intermuscular force transmission).
Note that, as for active force, the proximodistal ΔF is higher after distal than after proximal lengthening. Therefore, more passive myofascial force transmission from or onto the muscle did occur as EDL was lengthened distally.
It is concluded that symmetric lengthening of EDL muscle-tendon complex at proximal or distal tendons to equal lengths causes not only opposite but also quite asymmetric effects on active as well as passive length-force characteristics and on the proximodistal differences of force.
Therefore, for muscle length change, not only effects of its magnitude should be considered, but also effects of its direction.
Further directional analysis: comparison of force exerted at a specific EDL tendon in different conditions.
The muscle-tendon complex lengths of EDL muscle can also be expressed as deviation from its l ref, i.e., its length with both proximal and distal tendons at their reference positions (corresponding to EDL length encountered with ankle joint at 90° and the knee joints at 100°). For proximal and distal lengthening of the muscle to equal muscle-tendon complex lengths, using that format, Fig.3 compares EDL with either distal (A) or proximal isometric active forces (B). It should be noted that, in such a representation, the effect of relative position of the muscle (or parts thereof) is emphasized. Successive points of any one curve are affected by 1) muscle-tendon complex length and 2) the location of lengthening. The latter effect involves changes of the relative position of (parts of) the muscle with respect to surrounding muscle and other tissues. In such a case, the x-axis also represents the deviation of a major part of the muscle from reference position and the direction of such deviation. For comparisons made for equal absolute values of Δl (i.e., symmetric with respect tol ref), EDL lengths are identical (see also Table1) but were obtained by lengthening of the muscle-tendon complex by moving tendons at opposite ends and in opposite directions.
For active force (Fig. 3) as well as passive force (Fig.4), ANOVA showed significant effects of length and location of lengthening and a significant interaction between these factors. Such a result was found for both distal and proximal EDL force.
The length range between optimum and active slack lengths is smaller for distal lengthening of EDL than for proximal lengthening. This is true for distal as well as proximal EDL length-active force characteristics (Fig. 3). This higher length range is related predominantly to a higher optimum length after the proximal lengthening than after distal lengthening (Fig. 3, see also Fig. 2 A: optimum length after proximal lengthening ≈ 49 mm, and Fig.2 B: optimum length after distal lengthening ≈ 47 mm). Smaller but opposite differences of active slack lengths cannot fully abolish this effect on length range.
EDL force measurements in the direction of lengthening yield higher forces (e.g., distal lengthening increases distal force more than proximal lengthening). For distal force such an asymmetry, the response is already apparent (Fig. 3 A). This muscle position-related asymmetry is substantial, particularly at high lengths [Fdist (Δl dist)/Fdist(Δl prox) ≈ 118%, where Fdist is Fma measured at distal tendons, Δl dist is distal lengthening, and Δl prox is proximal lengthening]. At high length, the enormous asymmetry of the proximal EDL active length-force curves after distal and proximal lengthening of EDL is striking (Fig.3 B). After distal lengthening, much lower isometric forces are exerted at the proximal EDL tendon than after equal distal lengthening [Fprox(Δl prox)/Fprox(Δl dist) ≈ 157%, where Fprox is Fma measured at proximal tendons].
Note that the crossover point for the curves of active force exerted at a specific tendon (i.e., equal force after proximal or distal lengthening) of EDL distal as well proximal length-force curves is quite near to EDL l ref. This similarity is due to the fact that, exclusively for this data point, one extra factor is similar (in addition to the identical EDL lengths). This additional factor is the relative position of EDL with respect to surrounding muscles and connective tissue. It should be noted that these conditions are satisfied only for the data atl ref.
It is concluded that, if EDL has a similar length as well as relative position, forces exerted at a specific tendon are similar. However, if only EDL length is similar, but not EDL relative position, then that is clearly not the case.
Figure 4 indicates that an extremely high degree of asymmetry, effects of distal and proximal lengthening, is also true for passive EDL force. For distal lengthening, distal force increases more than for proximal lengthening, and proximal force increases more for proximal lengthening than for distal lengthening. The asymmetry increases readily with increasing lengths. Passive forces exerted at a particular tendon increase much more if lengthening was performed at that site rather than at the other end of the muscle (>7–10 times as high for distal and proximal force, respectively).
These results indicate that, at lengths other thanl ref, relative position of the muscle with respect to its surroundings is a factor of considerable importance, codetermining (with length) isometric length-force characteristics.
Differential Effects of Proximal or Distal EDL Lengthening on Neighboring Muscles
Synergists: anterior crural group.
Because TA+EHL muscle-tendon complex length was constant, TA+EHL active force is plotted as a function of EDL length (i.e., deviation from itsl ref: Fig. 5). ANOVA of these data indicates significant main effects (factors: EDL muscle-tendon complex length and location of lengthening), as well as significant interaction between these factors, on TA+EHL active force.
Distal lengthening of EDL affected TA+EHL active force significantly. Post hoc test located significant length effects on TA+EHL force at higher EDL lengths after distal lengthening: active force exerted at the length range Δl dist > 2 mm differed from that for the length range Δl prox < −2 mm (ΔFma≈ −0.3 N, i.e., by approximately −10% of initial force). In contrast, proximal EDL lengthening to equal muscle-tendon complex lengths did not affect TA+EHL force significantly. Such results once more emphasize the asymmetric effects of proximal and distal lengthening of EDL, this time on force exerted by its synergists, that were kept at constant muscle-tendon complex length.
It is concluded that, particularly for distal lengthening, we have evidence of intermuscular myofascial force transmission between the active synergists. This means that active force is transmitted from TA+EHL onto EDL to be exerted at its distal tendon.
ANOVA did not show either significant EDL length effects or interaction on passive TA+EHL force, but did show significant effects of location of EDL lengthening (Fig. 5 B).
It is concluded that symmetric lengthening of the adjacent synergist (EDL), at either the proximal or distal tendon to attain equal muscle-tendon complex lengths, has very asymmetric effects on active as well as passive force exerted at the distal tendons of TA and EHL.
It is concluded that symmetric proximal or distal lengthening of the EDL muscle-tendon complex causes differential effects on active force exerted at distal tendons by synergists TA+EHL kept at constant muscle-tendon complex length. In addition, passive force of TA+EHL is affected differentially as well. The asymmetry of these responses to symmetric lengthening at EDL proximal or distal tendons is striking and is related to differences in relative position of EDL with respect to neighboring muscles and extramuscular connective tissue.
The almost permanently present difference in proximal and distal EDL forces after proximal lengthening of EDL reported in the present work confirms the phenomenon for somewhat different experimental conditions compared with previous work (26, 28, 39). The reversal of sign of the difference, as well as the differences in magnitude after and symmetric lengthening to equal EDL length, is a novel and surprising finding. These results could not be explained if muscle fibers transmitted their force exclusively in series from sarcomere to sarcomere and then to the tendon.
Myofascial Force Transmission
The general idea of this concept is that, in addition to myotendinous transmission, force is transmitted also between sarcomeres and the collagen fiber-reinforced extracellular matrix of muscle (i.e., the intramuscular connective tissues). Active force generated within the contractile proteins of an individual sarcomere will lead to shortening, unless opposed by an external force. If the opposing force counteracting the sarcomere shortening does not originate from sarcomeres in series, but from the extracellular matrix, we call it active myofascial force transmission. In passive muscle, an external force may stretch sarcomeres. The external force exerted onto a stretched muscle may be transmitted via the myotendinous junction to the sarcomeres or onto the muscle's extracellular matrix. Based on an old observation by Banus and Zetlin (2), before the 1970s the generally accepted view was that passive force was borne by the structures of the extracellular matrix. However, after that time it became clear (e.g., Ref. 41) also that the sarcomeres are stretched, and most of the force exerted on the muscle is borne by stretched passive components [e.g., titin filaments (18)] of the sarcomere. In such a stretched condition, the tendency of the sarcomere to shorten elastically to its initial length is also prevented by the external force. If the external force is transmitted via the extracellular matrix to the sarcomere (rather than via the myotendinous junction and sarcomeres in series), we speak about passive myofascial force transmission.
Once the force is exerted on an endomysium, there are again two potential paths. The first path is tensile transmission in longitudinal direction. As the endomysium is continuous with a similar network surrounding groups of tendon collagen fibers, force can be transmitted onto the tendon.
The second path is shear transmission in cross-fiber directions. Force can also be transmitted further on the continuous endomysial stroma, i.e., the endomysium of neighboring muscle fibers of a fascicle. As the endomysial network is also continuous with the perimysium, force can be transmitted again in the two principal directions, either onto the tendon or neighboring fascicles (28, 43) or to the epimysium, as the perimysial network is also continuous with the epimysium.
If the muscle is dissected fully from other structures, intramuscular myofascial force transmission to the tendon is the only path, in addition to direct myotendinous force transmission. In effect, the full stroma of intramuscular connective tissues of a muscle acts as an integrator for active and/or passive force exerted by muscle fibers that have sufficiently stiff connections to their endomysium. This property allows groups of muscle fibers that are no longer connected to their tendon at one end still to contribute to muscle force (29,31, 32). These contributions are made at lower mean sarcomere lengths than those of fibers with connections to the tendon fully intact, because fiber shortening occurs until sufficient shear strain has accumulated in the basal lamina and endomysium to stiffen these structures to be able to transmit the additional force (43). This introduces or enhances a distribution of fiber mean sarcomere length of the muscle. However, for fully isolated muscle, force exerted at the proximal or distal tendon is equal as the muscle is built of parallel myofascial units exerting identical force at proximal and distal ends.
Extra- and intermuscular transmission.
Any proximodistal ΔF for a muscle, as reported also in the present work, indicates that force is transmitted from or onto the muscle-tendon complex anywhere between the locations of force measurement.
In fact, also in such conditions, force originating from sources outside of the muscle is integrated with force exerted by the target muscle. However, the locations of the points of application of the external forces being intermediate between the ends of the muscle fibers causes this integration to create or enhance distribution of lengths of sarcomeres arranged in series within the muscle fibers (73). Also, as a consequence, the distribution of fiber mean sarcomere lengths is expected to be affected.
If active or passive force is transmitted between the intramuscular connective tissues of two muscles, we refer to it as intermuscular myofascial force transmission. In such a case, the paths of transmission are the very short collagen fiber connections between the intermuscular stroma of the two muscles. Sometimes in literature, evidence may be found in general terms that some awareness existed of intermuscular connections of muscles or their conceivable effects (e.g., that muscle may not be fully independent units) (e.g., Refs.51, 69, 70, 71). A most expressive example may be a short note by Pond (51) in response to a review article by Stein (62). The ideas presented in that note are quite compatible with the concepts of intermuscular myofascial force transmission presented here. Unfortunately, at the time, no follow up was performed by Pond or others to provide any experimental evidence.
If force is transmitted, in either direction, between the intramuscular connective tissue of a muscle and extramuscular connective tissues, we refer to it as extramuscular myofascial force transmission. Examples of extramuscular connective tissue are the fascia constituting the compartmental boundaries or the neurovascular tract (i.e., the string of connective tissue containing bundles of nerves and vessels with its connections to compartment boundaries).
In most physiological experiments, dissection is performed on inter- and extramuscular connective tissues to gain access to the muscle. So far we have been able to find only one reference that stated explicitly that dissection was performed to remove (for the particular experiment) undesired effects of extramuscular and intermuscular connections on the properties of the experimental muscle (11).
Length and Position Effects of EDL
As a novel aspect of our present work, we systematically studied aspects of myofascial force transmission by imposing symmetrical length changes on the EDL muscle tendon complex to equal lengths at either its distal or proximal tendon. For a muscle with the connectivity of its belly to the extracellular matrix of connective tissues and to nearby muscles maintained close to their in vivo condition, as in the present work, three major conclusions can be drawn. 1) Lengthening of the muscle-tendon complex is always accompanied by changes in configuration of the intra- and extramuscular connective tissues, because of changes of relative positions of most of the muscle.2) The combined effects of changing length and muscular relative position are not symmetric for proximal and distal lengthening. 3) These effects do also change force exerted at distal tendons of synergist and even antagonist muscles, even if the lengths of their muscle-tendon complexes are kept constant.
In combination with muscle-tendon complex length effects, the factor relative muscular position is shown to be a major codeterminant of muscle force: for comparisons at equal length obtained by either proximal or distal lengthening, force actually exerted by EDL may be quite different (e.g., Fig. 3). A most simple representation of this idea is shown in Fig. 6. Note that the length change of the muscle-tendon complex is accompanied by changes in relative position of the muscle with respect to its surroundings. Also note that the conditions as represented in Fig. 6would have yielded opposite, but symmetrical, effects on the proximodistal ΔF.
The changes in relative position should not be expected to be similar for different segments of the muscle. For distal lengthening, the most distal segments of the muscle fibers are repositioned most with respect to their surroundings, and the most proximal, less. Therefore, the effects should be expected to have a high local variation within muscle fibers and/or their associated endomysial network. It is concluded that imposing a length change on the muscle-tendon complex with intact extra- and intermuscular connections inseparably involves some changes in relative position. It has also proven feasible to assess experimentally effects of relative position without involving length change of the target muscle (for a preliminary report, see Ref. 40). An advantage of the present approach is that effects of different EDL positions were assessed at many EDL lengths by symmetric proximal and distal lengthening of EDL.
Why are effects of EDL relative position so asymmetric?
The asymmetry of myofascial force transmission effects may have two major sources.
The first source is extramuscular structures. Because EDL muscle has no insertions of muscle fibers on the walls of the anterior tibial compartment, we argued previously (28) that the neurovascular tract is a major pathway. This may seem counterintuitive, as it is likely that the collagen fibers have been laid down to protect these sensitive structures. The nerves and blood vessels themselves will bear some of the force. However, it should be realized that, in a collagen fiber-reinforced composite (as muscle is), a relatively weak and compliant matrix (including the neurons and blood vessels) will be unloaded by a high concentration of collagen fibers (i.e., the fiber reinforcement). This happens because, at any intersection of paths of force transmission, such collection of fibers “attracts” a high force because of their high stiffness, i.e., most force is transmitted along the stiffest path (24). In this way, the work performed on the structure is minimized.
The proximal-to-distal decreasing size of the neurovascular tract due to the branching pattern of nerves and blood vessels may contribute to the asymmetry. However, this cannot be the only reason as it cannot explain the reversal of the sign of the effect of equal proximal and distal lengthening on proximal as well as distal EDL force.
The second source is intermuscular sources. If the muscles are shifted with respect to each other in such a way that intermuscular connections with EDL are loaded preferentially in one direction, asymmetries due to proximal and distal lengthening will arise. Because we know little about the neutral positions of muscle within a compartment with respect to each other, such contributions are presently fairly hard to judge. However, for the experimental conditions of the present study, it seems likely that, at high lengths, this factor will play a role because of high EDL lengths and the lower TA+EHL length.
Effects on Neighboring Muscles Kept at Constant Lengths
The mechanical interaction between immediately adjacent synergists is most likely intermuscular myofascial force transmission. However, the possibility cannot be excluded that extramuscular pathways also play a role in this phenomenon. There are a number of conceivable paths for extramuscular mechanical interaction in this case.
First, there are connections between both muscles via their neurovascular tracts and the intermuscular septum. EDL as well as the TA+EHL muscles are innervated via nerve branches (from the peroneus profundus nerve). This nerve enters into the anterior crural compartment from the peroneal compartment, through a fenestration within the anterior intermuscular septum (e.g., Fig. 5, Aand B, of Ref. 27). Both muscles receive circulation via branches of blood vessels (e.g., tibialis artery) that enter each compartment from the proximal side. These delicate structures are embedded in a structure of rather stiff connective tissues, with the purpose of protecting them from excessive strain. The connective tissues around nerves and blood vessels of each muscle have connections to the intermuscular septum (Fig.7). We refer to the complex of all of these elements as the neurovascular tract. For anatomic photographs of this neurovascular tract, see the following references (e.g., see Fig. 4 of Ref. 28, Fig. 5C of Ref.27, and figures of Ref. 30).
Such connections, in principle, do allow mechanical interaction between the muscles by means of extramuscular force transmission. Therefore, this pathway is one of the prime candidates for explaining interaction between adjacent synergists by extramuscular myofascial force transmission.
Second, it should be realized that direct connections of the intramuscular connective tissues of muscle bellies to the septum, presumably with myotendinous junctions, exist in muscle for fibers of EHL (Fig. 7). Because rat EDL muscle fibers have no origin or insertion of muscle fibers onto that structure, it is clear that interactions between these muscles must be mediated by intermuscular connective tissues or the extramuscular neurovascular tract.
It is concluded that substantial mechanical interaction by myofascial force transmission may be encountered between synergists. Those muscles cannot be considered mechanically as fully independent actuators.
Changes of Muscle Relative Position Are a Common Occurrence in Vivo
Any length change of a muscle will lead to changes in position of the muscle belly with respect to fixed extramuscular elements of the compartment (e.g., bones as well as compartment boundaries), which will strain the connective tissue elements that connect muscle to these structures.
In our present experiments, the length and position of EDL were changed exclusively. In vivo, simultaneous length changes in synergists are determined individually by changes of joint angle and activity of the muscle. It should be realized that any difference in moment arm between synergists would lead to changes in relative position of the muscles. Differences in moment arm for synergists are quite usual and sometimes substantial (e.g., Refs. 12, 13,15, 19, 38, 43,68). For example, for the mouse, the moment arm of TA at the ankle is reported to be 17% higher than for EDL (37). Therefore, differences in moment arm may contribute to considerable changes in in vivo muscular relative positions.
In addition to the above, a distinction between monoarticular and bi- or polyarticular muscles should be made. Because length changes of the latter group are determined in two or more joints, these muscles are likely to change their relative position with respect to monoarticular muscles and extramuscular connective tissues.
The size of the mechanical effect of in vivo changes of muscle relative position is codetermined by the actual stiffness of extra- and intermuscular connections. In our present experimental conditions, a group of synergists as well as a group of antagonist muscles were activated maximally (i.e., synchronized maximal activity of all muscles studied). This is not a very likely occurrence in vivo. Nevertheless, coactivation of synergists and antagonists is a common feature of in vivo movement (e.g., Refs. 3, 61). However, as extramuscular myofascial force transmission is expected to be a prominent feature of muscle, higher strains and thus stiffness may have been induced, particularly in extramuscular tissues and possibly also in intermuscular connections, than would be encountered with lower degrees of activation. On the other hand, our results also show sizable effects for EDL relative position in the passive condition also, in which stiffness of different pathways was not very high. Therefore, we expect that in vivo myofascial force transmission and effects of muscle relative position will have effects of sufficient magnitude to be taken into account, even if levels of recruitment are considerably lower than in our present study.
It is concluded that substantial changes of in vivo muscle position with respect to neighboring muscles and extramuscular structures are likely. Nevertheless, quantification of such effects awaits application of in vivo imaging techniques to this problem.
Some Functional Implications of Effects of Changes of Muscle Relative Position
A major effect of myofascial force transmission is that individual muscle should not be considered anymore as fully independent units of force generation and movement and that muscular relative position is a major codeterminant of force exerted at individual tendons of synergists and antagonists. This altered view is likely to have substantial effects on ideas within a large number of fields.
Joint range of movement.
Some of the expected functional effects are mediated by changes in joint angle-moment curves. The joint range of active force exertion of a muscle tendon complex is expected to increase because of increased distributions of sarcomere lengths within the muscle (23). Particularly at lower and higher muscle-tendon complex lengths, even limited enhancement of the active force may, for individuals, have quite important effects, as limitations to movement are removed by enhancing the length range of active force generation. In an isolated muscle, such increases in distribution of sarcomere lengths will always lower active force exerted at optimum length (23). However, in vivo intermuscular force transmission may counteract such effects.
Even within a midrange of joint movement, it is likely that extramuscular forces and possibly also intermuscular forces will cause high strains locally in muscles and connective tissue of a compartment, particularly at higher deviation of muscle relative position from equilibrium positions. Such local deformations may play a major role in the etiology of afflictions of the human locomotion apparatus, such as repetitive strain injury, tennis elbow, musician's arm, etc., and the adaptation of muscle and connective tissues to such conditions.
Individual muscles have been distinguished anatomically according to identifiable morphology after dissection, as well as different mechanical effects after being isolated. Our results regarding myofascial force transmission indicate that the mechanical basis underlying motor control is different. Neighboring synergists (EDL, EHL, and TA) form more of a functional unit than foreseen previously.
Of course, even with the view of myofascial force transmission, the basic unit of control remains the motor unit (e.g., Ref.47). Explaining how the central nervous system controls posture and movement in vivo requires an understanding of how excitation of motor neurons is organized. The rules that have been shown for regulating the control of recruitment of motor units of one muscle (e.g., Refs. 20, 21, 22) have been called the size principle: motor units of a muscle are predominantly recruited in order of their size, which correlates with many other motor unit properties.
Our present results could yield an expectation that the concept of motor unit pool (traditionally limited to the motor units of one muscle) should be extended to include motor units of synergists. In an exclusively muscle-based scheme, the size principle would organize only those motor units within individual muscles, leaving the nervous system with the substantial task of coordinating the relative activities of motor units from different muscles. Cope and Sokoloff (10) reviewed advantages of organization at a higher level than the muscle and speculated about such extension of the concept of the pool. They argued that this would simplify the neural process involved in organizing active motor units: in a muscle ensemble-based scheme, orderly recruitment of all motor units according to the size principle would automatically coordinate motor units both within and across motor nuclei. In subsequent experimental work (60), cat medial and lateral gastrocnemius were distinguished as two separate muscles on the basis of their motor nuclei. In the decerebrate cat, they dissected these heads of the gastrocnemius muscle from surrounding tissues but not from each other. They demonstrated that the size principle applies most of the time to the pool of motor units from these structures, if stretch were applied to the gastrocnemius tendon. However, such recruitment order of motor units may not be applicable to all active synergists within a limb, as they could not consistently show such order of control of motor units for synergist cat medial gastrocnemius and posterior biceps femoris muscles (both knee flexors), if activated by the cutaneous reflex induced by stimulation of the cutaneous crural nerve.
For intramuscular coordination, it has already been pointed out (46) that integration of multiple levels of organization may yield unique mechanical properties of motor units. Our present work indicates that such an idea should be extended to a process of integration across muscles. By varying exclusively relative positions of synergists (40), very different mechanical effects can be obtained. It is expected that varying recruitment would also cause such variable effects because of the mechanics of extra- and intermuscular force transmission.
Perfect coordination of muscular activity is only possible if relative stiffness of different intra-, extra-, and intermuscular pathways, which determine the partition of force transmitted via the major pathways, are optimal for exertion of moments and forces at the joints. However, joint stability is a conditio sine qua non for movement. Myofascial force transmission may help to exert forces at the joint. It has been pointed out that joint capsule and capsular ligaments cannot be described as separate entities but should be considered in unity with muscles that are arranged mechanically in series with them (70, 42).
It seems likely that a major role of motor control is the tuning of the stiffness of many connective tissue elements of a limb. This may seem to be an incredibly complex task for the central nervous system because the number and degree of interactions seem almost limitless. Two factors should aid to simplify this problem somewhat.
First, intramuscular and muscle-related receptors of different kinds are reported to be located at specialized positions within muscles to monitor muscle-connective tissue units in a way that does not coincide with the topography of muscles as morphological entities (Ref.70, chapters 5 and 6). Rather a lot of receptors and nerve endings are located within the walls of the compartment (general fascia, intermuscular septum, and interosseal membrane). The same author (in his chapter 8) also concluded that extramuscular receptors of proprioception aid the perception of movement.
Second, some specific interactions are likely to occur more frequently and may have been learned by humans in the course of many years of development or training.
The actual stiffness of the general fascia and other borders of compartments seems to be very important for the quantity of myofascial force transmission (25, 28, 39). This is also implicitly apparent from the fact that muscle fibers of special muscles (e.g., tensor fascia latea muscles) insert fully on these connective tissue structures. Such a situation may be more common than usually realized, because muscles (such as gluteus maximus muscle, e.g., Ref.25) may insert the greater majority (and in individual cases even 100%) of their muscle fibers onto these fascia. Of the muscles of our present study, the muscle fibers of rat EHL muscle originate from the anterior intermuscular septum. For humans, there are also indications that the superficial lamina of the thoracolumbar fascia will be tensed by activity of the latissimus dorsi, gluteus maximus, and erector spinae muscle, and even EDL muscle (69). For the cubital region, already in 1984 van Mameren and Drukker (46) viewed compartment walls as stress-conveying structures arranged, not in parallel, but in series with muscles. We propose that stiffening of the connective tissues constituting the walls of different compartments creates a firm origin or insertion for muscle fibers contained within the compartment. A stiff system of muscular compartments may act as a secondary skeleton. Therefore, activity of muscle fibers of this type of muscle is expected much more frequently than only for activities in which they may play a role as prime movers of a joint. Within such a stiff system, changes of length and relative position of muscle may allow very fine control of forces exerted via different pathways.
In summary, for passive and fully active synergists, our present work shows that relative position, with respect to each other and with respect to extramuscular connective tissues of the compartment in which they are located, is an important codeterminant of active and passive force exerted at a specific tendon of muscles. Because this is true also for EDL muscle, which has no connections to its neighboring synergists other than its myofascial connections, these effects are mediated by extra- and intermuscular myofascial force transmission.
Address for reprint requests and other correspondence: P. A. Huijing, Faculteit Bewegingswetenschappen, Vrije Universiteit, Van de Boechorststraat 9, 1081 BT Amsterdam, The Netherlands (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published November 27, 2002;10.1152/japplphysiol.00173.2002
- Copyright © 2003 the American Physiological Society