Eur J Appl Physiol (1997) 76: 472±479
Ó Springer-Verlag 1997
ORIGINAL ARTICLE
R.D. Herbert á J. Crosbie
Rest length and compliance of non-immobilised and immobilised rabbit soleus muscle and tendon
Accepted: 7 May 1997
Abstract The ®rst aim of this study was to measure the contributions of muscle and tendon to the total compliance of resting muscle-tendon units. A second aim was to determine whether the decrease in muscle-tendon unit rest length produced by prolonged immobilisation in a shortened position is mediated primarily by adaptations of the muscle or tendon. One ankle joint from each of ®ve rabbits was immobilised in a plantar¯exed position for 14 days. The passive length-tension properties of soleus muscle fascicles and tendons from both hindlimbs were measured using a video-based tensile-testing system. In non-immobilised muscles, muscle fascicle strains exceeded tendon strains by up to four times. However, because the rest length of tendon was much greater than that of muscle fascicles, changes in tendon length accounted for nearly half of the total change in muscletendon unit length. The rest length of immobilised muscle-tendon units was less than that of non-immobilised muscle-tendon units from contralateral limbs. Most of this dierence was attributable to a change in the rest length of the tendon; there was little change in the rest length of muscle fascicles. It is concluded that the tendon is responsible for a large part of the compliance of rabbit soleus muscle-tendon units at physiological resting tensions, and that adaptation of tendon rest length is the primary mechanism by which the rabbit soleus shortens in response to immobilisation at short lengths. Key words Skeletal muscle á Immobilisation á Tendon á Rest length á Compliance
Introduction The elastic properties of contracting muscles and their tendons have been studied extensively (e.g. Rack and Westbury 1984), but much less is known about resting R.D. Herbert (&) á J. Crosbie School of Physiotherapy, University of Sydney, P.O. Box 170, Lidcombe 2141 New South Wales, Australia
(non-contracting) muscle and tendon. The elastic properties of resting muscle-tendon units are functionally important because they (along with the periarticular connective tissues) determine the range of movement available at the joints. This paper investigates two questions about the elastic properties of resting muscletendon units. The ®rst question concerns the origin of the compliance of normal muscle-tendon units. It is not clear whether the compliance of resting muscle-tendon units resides primarily in muscle fascicles or tendons. The issue could be resolved by measuring the compliance of resting muscle fascicles and tendons at physiological resting tensions, but such measurements have not yet been made. The few studies in which the compliance of both resting muscle and whole tendon has been measured have not dierentiated the intramuscular part of the tendon from that of muscle fascicles (Stolov and Weilepp 1966; Hawkins and Bey 1997), or they have been primarily concerned with the compliance of muscle and tendon at lengths and tensions many times greater than those attained by resting muscle in vivo (Lieber et al. 1991; Trestik and Lieber 1993). The ®rst aim of the present study was to determine whether the compliance of resting muscle-tendon units resides primarily in muscle fascicles or tendons. To address this question we measured changes in the length of muscle fascicles, and in all of the tendon in series with the muscle fascicles of rabbit soleus muscle-tendon units as they were stretched through a physiological range of resting tensions. The second issue addressed in the present study concerns the eects of immobilisation on the elastic properties of resting muscle and tendon. It has been shown that prolonged immobilisation at short lengths reduces both the rest length and compliance of whole muscle-tendon units (Herbert and Balnave 1993). This result probably underlies the clinical phenomenon of muscle contracture, commonly seen following prolonged immobilisation of joints (Herbert 1993). A reduction in the rest length and compliance of muscle-
473
tendon units could be produced by changes in either muscle fascicles or tendon, but it is not yet clear which is of most importance. Immobilisation at short lengths is known to produce changes in muscle ®bres such as a decrease in the number of sarcomeres in series (Tabary et al. 1972; Williams and Goldspink 1978; Witzmann et al. 1982) and changes in the amount and architecture of the intramuscular connective tissue (Williams and Goldspink 1984; JoÂzsa et al. 1990). However, it is not certain that these morphological changes produce a decrease in muscle fascicle rest length (see Discussion). Huijing and his colleagues have shown that immobilisation at short lengths produces large decreases in the ``active slack length'' (a proxy for rest length) of the proximal tendon plate in rat medial gastrocnemius muscle (Heslinga and Huijing 1993), suggesting that adaptations of the intramuscular part of the tendon may contribute to decreases in the rest length of immobilised muscle. The second aim of the present study was to determine whether the decreases in the rest length of rabbit soleus muscle-tendon units produced by prolonged immobilisation occur primarily in muscle fascicles or tendon.
Methods Five female New Zealand white rabbits [mean (SD) weight 2.2 (0.2) kg; age 117±119 days] were sedated with acepromazine, and one hindlimb of each rabbit, chosen at random, was immobilised with a cast that extended from the end of the toes to the knee. The cast held the ankle joint in a position of full plantar¯exion so that the soleus muscle was held at its most shortened length. Once the casts had been applied the rabbits were allowed unrestricted cage activity and were provided with food and water for consumption ad libitum. All procedures were approved by the University Ethics Committee. After 14 days of immobilisation the animals were deeply anaesthetised with a pentobarbitone/urethane mixture. The gastrocnemius was severed from the soleus just proximal to the part of the distal tendon that the two muscles share. The soleus muscle-tendon unit was excised with small pieces of the head of the ®bula and the calcaneum still attached, and then coated with paran oil to prevent drying. With the aid of a magnifying lens, small spherical markers (radius 1.75 mm) were attached with a cyanoacrylate adhesive to the surface of the muscle at the proximal and distal ends of the proximal and distal tendon plates (Fig. 1). As little adhesive as possible was used, and the adhesive was applied to the markers, not the muscle, so as not to alter the compliance of either the muscle or the tendon. Markers located on the clamps of the test apparatus were used to locate the insertions of muscle-tendon units. Measurement of muscle-tendon unit length and tension Once the markers were attached, the muscle-tendon unit was mounted vertically in a purpose-built tensile-testing system. The bony origin and insertion of the muscle were held in slotted clamps to prevent slippage, and a small clasp held the distal bony attachment of the muscle ®rmly in its clamp to prevent artefacts from the compression of soft tissues occurring during the test. Muscle-tendon unit length (the distance from the top of the superior clamp to the bottom of the inferior clamp) was measured using a linear potentiometer mounted in parallel with the muscle, and tension was
Fig. 1 Rabbit soleus muscle-tendon unit, drawn to scale from nonimmobilised muscles. The top of the muscle-tendon unit is its proximal end. The hollow circles represent the spherical markers that were placed on the proximal and distal ends of proximal and distal tendon plates. Note that the ``lengths'' of the muscle-tendon components represent the lengths projected onto the long axis of the muscle measured using a force transducer (Grass FT03C, Quincy, Mass., USA, range 0±2N) that was connected in series with the proximal clamp. Immediately prior to testing, the muscle-tendon unit was suspended from its proximal attachment for 5 min. It was then lengthened at a constant velocity of 0.2 mm á s)1 (approximately 0.01 muscle fascicle lengths á s)1) from a length at which the distal tendon was visibly slack, until the tension in the muscle exceeded 1400 mN. The force and length signals were ampli®ed, ®ltered (10 Hz low-pass), and sampled at 640-ms intervals with 12-bit resolution using a personal computer. Soleus muscles from both hindlimbs of each animal were tested in random order. Testing of both muscles was always completed within 1 h of disruption of the muscle's blood supply, and usually within 40 min. This is much less than the time to onset of rigour, which has been shown to be several hours for the psoas muscles of rabbits killed under anaesthesia (Bate-Smith and Bendall 1949). Marker kinematics Three-dimensional kinematic data were obtained by using simultaneously three video cameras (National Panasonic M7) positioned anteriorly, posteriorly and laterally to the muscle at the height of the mid-point of the muscle. Each camera was orientated so that the long (vertical) axis of the muscle unit was located along the long (horizontal) axis of the camera ®eld, giving optimal image resolution. The image of the fully extended muscle occupied about the middle twothirds of the ®eld of view. Throughout testing, all markers were seen by at least two cameras, and some were seen by three. After testing, the spatial coordinates of the muscle marker centroids were realised using a manual digitising tablet and a videoplayback unit arranged in a con®guration similar to that described by Abraham (1987). Images from the video unit and from an overhead camera, which captured the active area of a digitising tablet, were superimposed using a video mixer and displayed on a monitor at approximately three times life size. Every 16th frame was digitised, giving a sampling interval of 640 ms. The sampling interval was checked with a time-code (precision 0.01 s) recorded on the video ®lm. Marker coordinates were merged and scaled, and transformations were applied to correct for parallax and for the oset of marker centroids from the surface of the muscle. The resulting three-dimensional marker coordinates were used to de®ne the lengths and relative angulations of the proximal and distal muscle fascicles, tendon plates and tendons, as well as the length of the
474 whole muscle-tendon unit (Fig. 1). The true length of each muscletendon component was multiplied by the cosine of the angle made with the muscle-tendon unit's long axis, giving the component's length projected onto the long axis of the muscle. The projected lengths of the proximal tendon, distal tendon plate and distal tendon were summed to give the projected length of the tendon in series with the proximal muscle fascicles. Similarly, the projected lengths of the proximal tendon, proximal tendon plate and distal tendon were summed to give the projected length of the tendon in series with the distal muscle fascicles. The system was tested prior to data processing. An elastic model simulating a muscle with markers located at known points on its surface was digitised on ten separate occasions, and the distances between markers and the angles subtended by the markers were calculated and compared with known values. Discrepancies between known and measured distances were less than 0.3 mm, and discrepancies between known and measured angles were less than 2°. The repeatability of the digitising system was tested by repeatedly digitising randomly selected video frames. Coecients of variation were less than 2.3% for all length measurements, and less than 6% for all angle measurements. Synchronisation of data from the force and displacement transducers with kinematic data derived from the video was achieved by the inclusion of a set of light-emitting diodes which were mounted on the test apparatus and visible to all three video cameras. The diodes were activated at the same time as the onset of data collection from the force and linear displacement transducers. When the muscle-tendon unit length data derived from the linear potentiometer and from the video®lm analysis were compared they were found to dier by less than 0.2 mm. Data reduction and analysis Muscle-tendon unit length-tension data were digitally smoothed and corrected for small baseline errors, as described previously (Herbert and Balnave 1993). The muscle-tendon unit was said to be under zero tension when suspended under its own weight. Highly linear and reproducible deformations of the testing system, located primarily in the force transducer, increased the apparent compliance of the proximal tendon and the whole muscle-tendon unit, so corrections for system compliance were made to measurements of proximal tendon and muscle-tendon unit length. Kinematic data for immobilised and contralateral muscles were averaged at 100 equally spaced tensions from 20 to 1100 mN using a linear interpolation procedure. Previously published measurements on New Zealand rabbits of the same size and age indicate that the range of tensions experienced in vivo by the normal resting rabbit soleus muscle is up to about 500 mN (Herbert and Balnave 1993), so only this range of tensions was considered in the analysis of non-immobilised muscles. Maximal isometric tension (Po), estimated from muscle mass and mean fascicle rest length, and assuming a stress of 30 N á cm)2, was 18.9 N; the range of tensions investigated was therefore up to approximately 2.6% Po.
Table 1 Mean rest lengths of muscle-tendon components from nonimmobilised and immobilised limbs. Dierences have been calculated by subtracting the rest lengths of immobilised muscles from the rest lengths of their contralateral counterpart; positive dier-
Muscle-tendon unit Proximal muscle fascicles Distal muscle fascicles Tendon in series with distal muscle fascicles Tendon in series with proximal muscle fascicles
The rest lengths of muscle-tendon units and muscle-tendon components were de®ned as their lengths at a tension of 20 mN. A coecient of muscle-tendon stiness, K, was determined from the least-squares regression of the equation T 20 eK
lÿL (20 £ T £ 1400 mN), where T is tension, l is muscle-tendon unit length and L is muscle-tendon rest length. The rest lengths of immobilised and contralateral muscles were compared using paired-samples t-tests. Except where stated otherwise, results are given as means (SD) of the sample.
Results The angles formed between muscle fascicles or tendon plates and the long axis of the muscle were always small (less than 12° for muscle fascicles, and less than 6° for tendon plates at the muscle-tendon unit rest length) and did not dier between contralateral and immobilised muscles. Angles as small as these have little eect (
