Functional and morphometric evaluation of end-to-side neurorrhaphy for muscle reinnervation

This study was undertaken to quantify the effect of motor collateral sprouting in an end-to-side repair model allowing end organ contact. Besides documentation of the functional outcome of muscle reinnervation by end-to-side neurorrhaphy, this experimental work was performed to determine possible downgrading effects to the donor nerve at end organ level. In 24 female New Zealand White rabbits, the motor nerve branch to the rectus femoris muscle of the right hindlimb was dissected, cut, and sutured end-to-side to the motor branch to the vastus medialis muscle after creating an epineural window. The 24 rabbits were divided into two groups of 12 each, with the second group receiving additional crush injury of the vastus branch. After a period of 8 months, maximum tetanic tension in the reinnervated rectus femoris and the vastus medialis muscles was determined. The contralateral healthy side served as control. The reinnervated rectus femoris muscle showed an average maximum tetanic force of 24.9 N (control 26.2 N, p = 0.7827), and the donor- vastus medialis muscle 11.0 N (control 7.3 N, p = 0.0223). There were no statistically significant differences between the two experimental groups (p = 0.9914). The average number of regenerated myelinated nerve fibers in the rectus femoris motor branch was 1,185 +/- 342 (control, 806 +/- 166), and the mean diameter was 4.6 +/- 0.6 microm (control, 9.4 +/- 1.0 microm). In the motor branch to the vastus medialis muscle, the mean fiber number proximal to the coaptation site was 1227 (+/-441), and decreased distal to the coaptation site to 795 (+/-270). The average difference of axon counts in the donor nerve proximal to distal regarding the repair site was 483.7 +/- 264.2. In the contralateral motor branch to the vastus medialis muscle, 540 (+/- 175) myelinated nerve fibers were counted. In nearly all cross-section specimens of the motor branch to the vastus medialis muscle, altered nerve fibers could be identified in one fascicle distal and proximal to the repair site. The results show a relevant functional reinnervation by end-to-side neurorrhaphy without functional impairment of the donor muscle. It seems to be evident that most axons in the attached segment were derived from collateral sprouts. Nonetheless, the present study confirms that end-to-side neurorrhaphy is a reliable method of reconstruction for damaged nerves, which should be applied clinically in a more extended manner.     

Experimental free-muscle transplantation with microneurovascular anastomoses

In 13 rabbits the rectus femoris muscle was freely transplanted from the left to the right side using microneurovascular anastomoses. About 7 months after surgery the muscle transplants were assessed functionally by force measurements. On the average, the transplanted muscles regained 55 percent of the maximal tetanic tension of unoperated, normal rectus femoris muscles, expressed as force per gram of muscle weight even 68 percent. After functional assessment, the muscles were weighed and then used for histologic, histochemical, planimetric, and biochemical evaluation. H&E-stained cross sections showed a high content of healthy muscle fibers; only some small atrophic and single fat cells were scattered over the cross sections. Good reinnervation over the sutured muscle nerve was confirmed by the type-grouping of muscle fibers in the NADH and myofibrillar ATPase staining. There was an excellent correlation between the functional results and the histologic picture as well as the content of choline acetyltransferase (CAT). A certain parallelism was found between the function of the transplants and the content of hexokinase, but none for the other estimated muscle enzymes, such as malate dehydrogenase (MDH), creatine kinase (CK), and lactic dehydrogenase (LDH). All enzyme levels were lower than in normal muscles. The results of this experimental series underline the utility of muscle transplantation with microneurovascular anastomoses to restore lost muscle function, even in the extremities, when strong forces are needed.     

Task-dependent inhomogeneous muscle activities within the bi-articular human rectus femoris muscle

The motor nerve of the bi-articular rectus femoris muscle is generally split from the femoral nerve trunk into two sub-branches just before it reaches the distal and proximal regions of the muscle. In this study, we examined whether the regional difference in muscle activities exists within the human rectus femoris muscle during maximal voluntary isometric contractions of knee extension and hip flexion. Surface electromyographic signals were recorded from the distal, middle, and proximal regions. In addition, twitch responses were evoked by stimulating the femoral nerve with supramaximal intensity. The root mean square value of electromyographic amplitude during each voluntary task was normalized to the maximal compound muscle action potential amplitude (M-wave) for each region. The electromyographic amplitudes were significantly smaller during hip flexion than during knee extension task for all regions. There was no significant difference in the normalized electromyographic amplitude during knee extension among regions within the rectus femoris muscle, whereas those were significantly smaller in the distal than in the middle and proximal regions during hip flexion task. These results indicate that the bi-articular rectus femoris muscle is differentially controlled along the longitudinal direction and that in particular the distal region of the muscle cannot be fully activated during hip flexion.     

Neuromuscular rehabilitation by treadmill running or electrical stimulation after peripheral nerve injury and repair.

Numerous studies have been devoted to the regeneration of the motor pathway toward a denervated muscle after nerve injury. However, the regeneration of sensory muscle endings after repair by self-anastomosis are little studied. In previous electrophysiological studies, our laboratory showed that the functional characteristics of tibialis anterior muscle afferents are differentially affected after injury and repair of the peroneal nerve with and without chronic electrostimulation. The present study focuses on the axonal regeneration of mechano- (fibers I and II) and metabosensitive (fibers III and IV) muscle afferents by evaluating the recovery of their response to different test agents after nerve injury and repair by self-anastomosis during 10 wk of treadmill running (LSR). Data were compared with control animals (C), animals with nerve lesion and suture (LS), and animals with lesion, suture, and chronic muscle rehabilitation by electrostimulation (LSE) with a biphasic current modulated in pulse duration and frequency, eliciting a pattern mimicking the activity delivered by the nerve to the muscle. Compared with the C group, results indicated that 1) muscle weight was smaller in LS and LSR groups, 2) the fatigue index was greater in the LS group and smaller in the LSE group, 3) metabosensibility remained altered in the LS and LSE groups, and 4) mechanosensitivity presented a large increase of the activation pattern in the LS and LSE groups. Our data indicated that chronic muscle electrostimulation partially favors the recovery of muscle properties (i.e., muscle weight and twitch response were close to the C group) and that rehabilitation by treadmill running also efficiently induced a better functional muscle afferent recovery (i.e., the discharge pattern was similar to the C group). The effectiveness of the chronic electromyostimulation and the treadmill exercise on afferent recovery is discussed with regard to parameters listed above.     

Functional properties of palmaris longus muscles of rhesus monkeys transplanted as index finger flexors

This experiment with skeletal muscle autografts in monkeys was designed to retest previous findings that transplanted skeletal muscle can regenerate to a functional degree in primates without predenervation and to test a new hypothesis that increased functional demands on regenerated muscle grafts in monkeys may result in improved functional capacity of the grafts. Rhesus monkey index flexors were replaced with free palmaris longus muscle autografts with microneural anastomoses between the graft motor nerve and the severed profundus motor nerve. One monkey was taught selective index flexion before grafting and continued with this program after grafting to test the effect of training on the graft. Mature grafts were evaluated for in vivo contractile properties and by histology and histochemistry and were compared with a group of normal Rhesus palmaris longus muscles. The results reconfirm the capacity of nonpredenervated monkey skeletal muscle grafts to regenerate and to achieve some contractile ability and suggest that training of free muscle grafts may enhance recovery of their functional and structural properties.     

Neurovascularized free short head of the biceps femoris muscle transfer for one-stage reanimation of facial paralysis

The single-stage technique for cross-face reanimation of the paralyzed face without nerve graft is an improvement over the two-stage procedure because it results in early reinnervation of the transferred muscle and shortens the period of rehabilitation. On the basis of an anatomic investigation, the short head of the biceps femoris muscle with attached lateral intermuscular septum of the thigh was identified as a new candidate for microneurovascular free muscle transfer. The authors performed one-stage transfer of the short head of the biceps femoris muscle with a long motor nerve for reanimation of established facial paralysis in seven patients. The dominant nutrient vessels of the short head were the profunda perforators (second or third) in six patients and the direct branches from the popliteal vessels in one patient. The recipient vessels were the facial vessels in all cases. The length of the motor nerve of the short head ranged from 10 to 16 cm, and it was sutured directly to several zygomatic and buccal branches of the contralateral facial nerve in six patients. One patient required an interpositional nerve graft of 3 cm to reach the suitable facial nerve branches on the intact side. The period required for initial voluntary movement of the transferred muscles ranged from 4 to 10 months after the procedures. The period of postoperative follow-up ranged from 5 to 42 months. Transfer of the vascularized innervated short head of the biceps femoris muscle is thought to be an alternative for one-stage reconstruction of the paralyzed face because of the reliable vascular anatomy of the muscle and because it allows two teams to operate together without the need to reposition the patient. The nerve to the short head of the biceps femoris enters the side opposite the vascular pedicle of the muscle belly, and this unique relationship between the vascular pedicle and the motor nerve is anatomically suitable for one-stage reconstruction of the paralyzed face. As much as to 16 cm of the nerve can be harvested, and the nerve is long enough to reach the contralateral intact facial nerve in almost all cases. The lateral intermuscular septum, which is attached to the short head, provides "anchor/suture-bearing" tissue, allowing reliable fixations to the zygoma and the upper and lower lips to be achieved. In addition, the scar and deformity of the donor site are acceptable, and loss of this muscle does not result in donor-site dysfunction.     

Experimental study of results of nerve suture under tension vs. nerve grafting.

Evoked electromyograms and myelinated fiber caliber histograms were used to compare the results of end-to-end nerve suture and nerve grafting, performed with an operating microscope. The best results were achieved after direct suture without tension. Direct suture under a tension of 25 grams produced results sometimes comparable the that of the zero tension group, but mostly quite inferior to those and even to the grafted group. The regeneration of axons in the nerve graft group was delayed more than in the sutured groups, but by the tenth month the number of regenerating fibers had increased with stable and comparatively good results. After a nerve suture under a tension of 25 grams, there is impairment of the blood flow at the suture site, and the results are not as good. In such situations I believe nerve grafting should be performed. This amount of tension also seems to be the limit which will allow accurate coaptation of a nerve by microsurgical techniques.     

Enhancement of nerve regeneration and orientation across a gap with a nerve graft within a vein conduit graft: a functional, stereological, and electrophysiological study

In this study, the right sciatic nerves of 40 rats were used to determine whether a nerve graft within a vein graft might accelerate and facilitate axonal regeneration, compared with a nerve graft alone. The animals were separated into four groups, as follows: group 1, sham control; group 2 (control), segmental nerve resection and no repair; group 3, segmental nerve resection and nerve grafting; group 4, segmental nerve resection and reconstruction with a nerve graft within a vein conduit graft. For all groups, sciatic functional indices were calculated before the operation and on postoperative days 7 and 90. On postoperative day 90, the sciatic nerves were reexposed and nerve conduction velocities were recorded. The sciatic nerves were harvested from all groups for counting of the myelinated axons with a stereological method. No statistically significant differences with respect to return of gait function, axon count, or nerve conduction were noted between groups 3 and 4 (p > 0.05). However, functional recovery in group 4 on postoperative day 90 was significant, compared with group 2 (p < 0.05); the recovery difference between groups 2 and 3 was not significant (p > 0.05). This study was not able to demonstrate any functional benefits with the use of a nerve graft within a vein graft, compared with standard nerve grafting.     

Effect of the tendon reflex on motoneurons of the antagonist muscle in man

The knee jerk was elicited during regular firing of relatively low-threshold motor units of the biceps femoris muscle (during weak voluntary contraction). Besides the reflex response of the rectus femoris muscle, synchronous discharges of motor units of the biceps femoris muscle and activation of new motor units also were observed. Poststimulus histograms and statistical analysis of interspike intervals of motor units of the biceps femoris muscle revealed well-marked excitatory influences synchronous with the reflex response of the rectus femoris. This result can be explained by the presence of excitatory inputs of Ia afferents on motoneurons of the antagonist muscle. In the knee jerk, excitation of motoneurons of the antagonist was followed by later inhibitory influences which evidently correspond to the "silent period" of motoneurons of the agonist muscle during the elicitation of its tendon reflex.Institute for Problems of Information Transmission, Academy of Sciences of the USSR, Moscow. Translated from Neirofiziologiya, Vol. 8, No. 6, pp. 624–632, November–December, 1976.     

Task-dependent spatial distribution of neural activation pattern in human rectus femoris muscle

Compartmentalization of skeletal muscle by multiple motor nerve branches, named as neuromuscular compartment (NMC), has been demonstrated in animals as well as humans. While different functional roles among individual NMCs were reported in the animal studies, no studies have clarified the region-specific functional role within a muscle related with NMCs arrangement in human skeletal muscle. It was reported that the rectus femoris (RF) muscle is innervated by two nerve branches attached at proximal and distal parts of the muscle. The purpose of the present study is to clarify the possible region-specific functional role in the human RF muscle. Multi-channel surface electromyography (SEMG) were recorded from the RF muscle by using 128 electrodes during two different submaximal isometric contractions that the muscle contributes, i.e. isometric knee extension and hip flexion, at 20%, 40%, 60% and 80% of maximal voluntary contraction (MVC). Results indicated that the central locus activation for the amplitude map of SEMG during hip flexion located at more proximal region compared with that during knee extension. Significant higher normalized root mean square (RMS) values were observed at the proximal region during the hip flexion in comparison to those at middle and distal regions at 60% and 80% of MVC (p<0.05). In while, significant higher normalized RMS values were demonstrated at the distal region comparing with that at the proximal region at 80% of MVC (p<0.05). The results of the present study suggest possible region-specific functional role in the human RF muscle.     

Homologous Muscle Contraction during Unilateral Movement Does Not Show a Dominant Effect on Leg Representation of the Ipsilateral Primary Motor Cortex

Co-activation of homo- and heterotopic representations in the primary motor cortex (M1) ipsilateral to a unilateral motor task has been observed in neuroimaging studies. Further analysis showed that the ipsilateral M1 is involved in motor execution along with the contralateral M1 in humans. Additionally, transcranial magnetic stimulation (TMS) studies have revealed that the size of the co-activation in the ipsilateral M1 has a muscle-dominant effect in the upper limbs, with a prominent decline of inhibition within the ipsilateral M1 occurring when a homologous muscle contracts. However, the homologous muscle-dominant effect in the ipsilateral M1 is less clear in the lower limbs. The present study investigates the response of corticospinal output and intracortical inhibition in the leg representation of the ipsilateral M1 during a unilateral motor task, with homo- or heterogeneous muscles. We assessed functional changes within the ipsilateral M1 and in corticospinal outputs associated with different contracting muscles in 15 right-handed healthy subjects. Motor tasks were performed with the right-side limb, including movements of the upper and lower limbs. TMS paradigms were measured, consisting of short-interval intracortical inhibition (SICI) and recruitment curves (RCs) of motor evoked potentials (MEPs) in the right M1, and responses were recorded from the left rectus femoris (RF) and left tibialis anterior (TA) muscles. TMS results showed that significant declines in SICI and prominent increases in MEPs of the left TA and left RF during unilateral movements. Cortical activations were associated with the muscles contracting during the movements. The present data demonstrate that activation of the ipsilateral M1 on leg representation could be increased during unilateral movement. However, no homologous muscle-dominant effect was evident in the leg muscles. The results may reflect that functional coupling of bilateral leg muscles is a reciprocal movement.     

Donor-site morbidity of the pedicled rectus femoris muscle flap

The rectus femoris muscle flap is well known for its reliable anatomy, the ease with which it can be harvested, and its great versatility. As a pedicled or free flap, it is used to cover soft-tissue defects and to recreate motor function. Although the muscle is very reliable, it is not well respected because of its assumed donor-site morbidity, such as weakened knee extension force and decreased range of motion of the knee. To date, these clinical assumptions have only rarely been quantified in terms of objective scores, concerning force deficit and functional or aesthetic outcome. From 1995 to 2002, the authors treated 24 patients with pedicled rectus femoris muscle flaps. Fourteen patients were followed up. Follow-up time ranged from 3 to 56 months postoperatively. The results were evaluated by a standard questionnaire in which pain in relation to walking distance, subjective feeling of weakness, sensibility disorders, everyday function, and aesthetic aspects were assessed. Range of motion in the hip and the knee was measured. For objective verification of a decrease of maximal voluntary contraction force of the remaining quadriceps muscle and for detecting differences in true muscular capacity and voluntary activation, 10 patients with unilateral rectus femoris flaps were tested using the twitch interpolation technique. The authors' patients assessed the remaining function and the aesthetic result of the thigh as at least satisfactory. Two patients complained about pain and weakness in the thigh. Eight patients reported hypesthesia in the lateral suprapatellar region. The maximal voluntary contraction and true muscular capacity values were reduced by 21.8 percent and 18 percent, respectively, when compared with the healthy leg. The range of motion in the knee and hip was not influenced by muscle harvesting. The twitch interpolation technique revealed a mild voluntary activation deficit, probably caused by inhibitory regulation in the spinal cord. In conclusion, donor-site morbidity of the rectus femoris muscle flap is evident but well compensated. There is no decrease in active range of motion in the knee and hip. Patient satisfaction with the functional and aesthetic outcome was high.     

Lateral intermuscular septum of the thigh and short head of the biceps femoris muscle: an anatomic investigation with new clinical applications.

An anatomic study was performed to reappraise the vasculature of the lateral intermuscular septum of the thigh and muscles associated with it using 12 preserved cadaver legs. Several possible new clinical applications of the lateral intermuscular septum and the short head of the biceps femoris were identified as follows: (1) short head of biceps femoris muscle or musculoseptal flap based on the second and/or third profunda perforating vessels, or based on the superior lateral genicular vessels, with or without the iliotibial tract and the deep fascia, and with or without the motor nerve of the short head; (2) transverse extension of the fascial portion of the tensor fasciae latae muscle or musculocutaneous flap to include the lateral intermuscular septum; (3) combination use of items 1 and 2, above; and (4) free septofascial graft using the lateral intermuscular septum and iliotibial tract.It is anticipated that the distally based short head of the biceps femoris muscle flap will be an additional option for repairing defects around the knee, and that a free short head of the biceps femoris muscle flap based on the profunda femoris perforating vessels will be useful in functional reconstruction such as reanimation of the paralyzed face. The lateral intermuscular septum can be incorporated into the short head of biceps femoris muscle flap or into the tensor fasciae latae flap, and it also can be used as a free fascial graft. Functional deficit resulting from harvesting the short head of the biceps femoris and the lateral intermuscular septum is minimal, and donor wound at the lateral lower thigh seems to be acceptable.     

The role of reflex mechanisms in postinhibitory rebound studied by analysis of human single motor unit activity

The mechanism of onset of rebound after inhibition induced by electrical stimulation of a nerve of maximal and submaximal strength for M-response was studied in single motor units of normal human soleus, rectus femoris, and hand muscles. Poststimulus histograms and changes in the duration of interspike intervals were compared with mechanical recordings of muscle contractions. In all muscles tested, during strong isotonic contraction, the increase in motor unit activity after a silent period was partly due to synchronization of their emergence from inhibition. However, it also contained a component of true facilitation of motoneurons, which was evidently a reflex response to lengthening of the muscle in the relaxation phase after evoked contraction. The latent period of this facilitation in the soleus and rectus femoris muscles coincided in value with the latent period of the monosynaptic spinal reflex, whereas in the hand muscles, in which a monosynaptic response to electrical nerve stimulation could not be evoked, the latent period of facilitation as a result of spindle activation during muscle relaxation was significantly longer than the latent period of the monosynaptic reflex. These findings support the hypothesis of presynaptic suppression of monosynaptic connections of Ia afferents with the motoneurons of some human muscles by descending tonic influences and of the use of information coming from spindles by supraspinal levels of the CNS.     

Effects of aging and life-long physical training on collagen in slow and fast skeletal muscle in rats

Summary Intramuscular collagen in a slow (m. soleus) and a fast (m. rectus femoris) skeletal muscle was studied by biochemical, morphometric, and immunohistochemical methods. Wistar white rats of 1, 4, 10, and 24 months were used as experimental animals. Our aim was to evaluate the effects of life-long physical training (treadmill running, 5 days a week for 1, 3, 9, and 23 months depending on the age attained). The biochemical concentration of collagen was higher in m. soleus than in m. rectus femoris and it increased in youth and in old age in m. soleus. The trained rats had higher concentrations of collagen than the untrained rats at 10 and 24 months. The morphometrically measured area-fractions of both the endomysium and perimysium were higher in m. soleus than in m. rectus femoris. The age-related increase in intramuscular connective tissue was of endomysial origin. The immunohistochemical staining of type-I, -III, and -IV collagens indicated the more collagenous nature of m. soleus as compared with m. rectus femoris for all major collagen types; this was most marked for type-IV collagen of basement membrane. The results indicate that both age and endurance-type physical training further distinguish the slow and fast muscles with respect to their connective tissue.     

Nerve growth factor receptor (NGFR) immunoreactivity in skeletal muscles of the rat.

The peroxidase-antiperoxidase (PAP) method, and a specific monoclonal antibody (192-IgG) were used to determine the localization of nerve growth factor receptor (NGFr) in the skeletal muscles of the adult rats. The rectus femoris and the gastrocnemius (medialis and lateralis) muscles were analyzed. Occurrence of NGFr immunoreactivity was observed in: 1) a subpopulation of myelinated nerve fibers within muscle nerve trunks; 2) the vascular adventitia and nerve-like profiles around the blood vessels; 3) the outer capsule and the surface of the intrafusal muscle fibers of muscle spindles. Conversely, images, suggesting the presence of NGFr on muscle fibers or in motor end-plates, were not found. Our results suggest the presence of NGF-binding sites in sensory and sympathetic nerve fibers, and/or their target tissues localized on the skeletal muscles of the rat, whereas the motor nerve fibers lack of NGFr. The dependence of sympathetic neurons, proprioceptive primary sensory neurons, and motoneurons innervating the mammalian muscles upon NGF or other neurotrophic factors is discussed.     

Quantitative EMG analysis to investigate synergistic coactivation of ankle and knee muscles during isokinetic ankle movement. Part 1: time amplitude analysis.

Synergy generally refers to the coordinated action of several motor elements to produce a specific motor task, either intentionally or automatically. One example is motor irradiation, a sudden spread of synergistic muscular coactivation resulting from a forceful single joint movement. To investigate this type of synergy pattern, a quantitative EMG approach was employed to characterize explicit neuromuscular synergy in the ankle-knee complex during maximal ankle isokinetic contraction. In the present study, isokinetic ankle contractions, both dorsiflexion and plantarflexion, at four different speeds (30, 60, 120, and 240 degrees/s) were studied in a normal adult population (N=11) to assess synergistic coactivation of the prime movers (tibialis anterior and gastrocnemius) and irradiated muscles (ipsilateral and contralateral rectus femoris and biceps femoris) of the ankle-knee complex. Electromyographic signals were collected with surface EMG electrodes and processed with traditional time-amplitude analysis to examine specific neural control strategies. The data generally supported several empirical assumptions common to neurological facilitation techniques. (1) Motor irradiation to the knee muscles due to ankle muscle isokinetic contraction was strongly directionally dependent. (2) Motor irradiation to the ipsilateral knee muscles due to ankle isokinetic contraction was speed dependent. (3) The prime movers demonstrated a similar control strategy, irrespective of different contraction speeds.     

Parallel-fibered muscles transplanted with neurovascular repair into bipennate muscle sites in rabbits.

Experiments were performed on 20 New Zealand White male rabbits. Our hypotheses were that (1) latissimus dorsi (LTD) muscles transplanted into the site of a bipennate rectus femoris (RFM) muscle with neurovascular repair would retain their parallel-fibered structure and (2) the parallel-fibered structure of latissimus dorsi grafts would reduce their total fiber cross-sectional area and adversely affect force development relative to that of bipennate rectus femoris grafts and muscles. Compared with their respective donor muscles, 120 to 150 days after grafting, latissimus dorsi and rectus femoris grafts showed no change in the number of fibers and a decrease in the mean single-fiber cross-sectional area to approximately 70 percent. The latissimus dorsi grafts, which remained parallel-fibered, developed maximum forces 34 and 23 percent of the values for fully activated rectus femoris grafts and muscles, respectively. The deficit in the maximum force of the latissimus dorsi grafts resulted primarily from the smaller total-fiber cross-sectional area as a result of the parallel-fibered structure.     

Plasticity of skeletal muscle: regenerating fibers adapt more rapidly than surviving fibers

The properties of mammalian skeletal muscle demonstrate a high degree of structural and functional plasticity as evidenced by their adaptability to an atypical site after cross-transplantation and to atypical innervation after cross-innervation. We tested the hypothesis that, regardless of fiber type, skeletal muscles composed of regenerating fibers adapt more readily than muscles composed of surviving fibers when placed in an atypical site with atypical innervation. Fast muscles of rats were autografted into the site of slow muscles or vice versa with the donor muscle innervated by the motor nerve to the recipient site. Surviving fibers in donor muscles were obtained by grafting with vasculature intact (vascularized muscle graft), and regenerating fibers were obtained by grafting with vasculature severed (free muscle graft). Our hypothesis was supported because 60 days after grafting, transposed muscles with surviving fibers demonstrated only a slight change from the contractile properties and fiber typing of donor muscles, whereas transplanted muscles with regenerating fibers demonstrated almost complete change to those of the muscle formerly in the atypical site.