Acute and chronic adaptations of muscle proprioceptors in response to increased use.

Acute adaptations to use have been shown to occur in both the muscle spindle and Golgi tendon organ pathways. This short term activation adaptation increases the excitability of the motoneuron pool, thereby potentially providing a nonvoluntary increase in the excitation of subsequent contractions. It is thought that the intrafusal muscle fibres reset to a higher gain after contraction and that the tendon organ pathway undergoes a brief desensitisation. These phenomena could be important clinically when trying to stretch muscles, i.e. a contraction before a stretch should make the stretch more difficult. Also, this could affect the amount of muscle force generated, thereby altering motor behaviours requiring fine accuracy. Research on the adaptations of proprioceptors during free movement, using locomotion as a model, has found that the hypothesis of consistent alpha-gamma coactivation during motor behaviours is much more complicated and adaptable, depending on the environmental circumstance and the specific motor task. These research findings support the use of selective training, i.e. training to the task, for optimal motor learning. The results of the relatively limited research on chronic adaptations of proprioceptors due to exercise has shown that on a microlevel, the intrafusal muscle fibres may show some metabolic changes but do not show any hypertrophy. However, on a more macro level, with extended training, the latency of the stretch reflex response is found to be decreased and the amplitude is found to be increased in both animals and humans. Through classical conditioning research, proprioceptors may also be necessary for motor learning. Lastly, both primates and humans have been shown to be able to up- or down-grade their stretch reflex responses using operant conditioning techniques and many practice sessions. These findings have important implications both for rehabilitation of persons with abnormal reflex activity and in the training of athletes. This review provides some consolidation of the recent research findings, but much more research needs to be done in order to fully understand the purpose and importance of the proprioceptive sensory system.

Reductions in recruitment force thresholds in human single motor units by successive voluntary contractions.

Recruitment force thresholds of biceps brachii single motor units were studied in 4 male subjects before and after an isometric muscle contraction, passive muscle stretch, or following successive muscle contractions, muscle stretches or during alternations between muscle stretches and muscle contractions. Isometric muscle contractions of 5 s duration decreased subsequent single motor unit force thresholds. These force thresholds could usually be reset at or near precontraction force threshold values by passive muscle stretch induced by elbow extension. Single motor units showing reduced force thresholds following contraction were momentarily derecruited during and/or after muscle stretch. Successive muscle stretches alone did not significantly alter single motor unit force thresholds. In contrast, single motor unit recruitment force thresholds during successive weaker contractions were progressively lowered. Intercontraction muscle stretches maintained the single motor unit force thresholds at or near the initial force threshold level. The mechanism(s) underlying a muscle contraction-induced lowering of single motor unit force thresholds may reside in stretch reflex pathways.

Co-contractions in antagonistic hindlimb muscles during simulated step cycle rates.

To determine whether the step cycle rate during locomotion may be limited by mechanical interactions of antagonistic muscles about the ankle, muscle activation patterns were simulated through electrical stimulation and muscle forces monitored in the rat lateral gastrocnemius-soleus (LGS) and tibialis anterior (TA) muscles. Simulations were made in acute experiments on Nembutal-anesthetized rats. Based on EMG data recorded from free-moving rats at 3 treadmill speeds, SOL mean cycle durations, cycles per second (eps), burst durations, and TA onset times following SOL EMG onset were determined. With muscle nerve stimulation based on these temporal patterns, muscle force was monitored from the detached tendons or directly at the approximate insertion sites. To investigate mechanical interactions attributed to the slow-twitch SOL muscle, forces were measured with the LGS intact, and after the SOL was denervated or the tendon was cut. With tendons detached, TA contraction force temporally overlapped with LGS or LG contraction force over periods ranging between 20 and 50 ms at 3.4 cps, 0-10 ms at 4.1 cps, and 0-10 ms at 4.6 cps. LGS force-time traces were asymmetrical, i.e. return time toward baseline force was more prolonged than the time to reach peak force. Without contributions to force from SOL activation, LG rate of relaxation was consistently increased and rate of rise and fall in force-time patterns were more symmetrical. Co-contractions were not observed when LGS contractions followed TA activation. At all cycle rates, forces remained above baseline (range 0.5-4.0 N) between contractions in LGS.(ABSTRACT TRUNCATED AT 250 WORDS)

Contraction-induced potentiation of human motor unit discharge and surface EMG activity.

In quadrupeds, an electrically induced, moderate to high intensity brief muscle contraction potentiates autogenetic excitation and leads to enhanced recruitment and/or tonic firing frequency of alpha-motor neurons. To determine if similar adaptations occur in humans, single motor units (SMUs) and surface electromyographic activity (EMG) were recorded from the right biceps brachii before and immediately after a 5-s 25% or 50% maximum voluntary contraction (MVC), while subjects held a handle (0-1% MVC) attached to a force transducer or maintained a 2% MVC for 30-60 s. Of 26 SMUs recorded, 15 increased, 4 decreased, and 7 showed no change in firing frequency (mean increase: 5 imp/s, P less than 0.01). Twelve SMUs had lower recruitment force thresholds after contraction. There was no significant treatment effect for the % MVC intensity. The postcontraction surface EMG power spectrum broadened, increased in amplitude, and contained a higher frequency component than the control contraction power spectrum. Changes in recruitment and/or frequency coding were reflected in the raw EMG records. Findings agree with previous reports in animals of contraction-induced potentiation of subsequent submaximal muscle contractions. Such acute adaptations in spinal neuromuscular pathways would function to optimize force output to a submaximal range of neural input frequencies.

Coexistent Hoffmann reflexes in human leg muscles are commonly due to volume conduction.

The existence of "concomitant" (coexistent) electromyographic reflex responses in soleus and tibialis anterior muscles, produced by posterior tibial nerve stimulation, has been cited as evidence for "reciprocal excitation" between these antagonistic muscles normally reflexly linked by reciprocal inhibition. Using the Hoffmann reflex procedure and posterior tibial nerve stimulation, the existence of true reciprocal excitation was tested in six subjects with no neuromuscular impairment. Coexistent EMG responses were observed in all subjects. In no instance, however, could the tibialis anterior EMG volley to posterior tibial nerve stimulation of the soleus muscle be antidromically blocked by common peroneal nerve stimulation applied at 10 to 20 ms offset latencies. A second stimulation pulse applied to the common peroneal nerve at similar offset latencies did antidromically block a tibialis anterior reflex response to common peroneal nerve stimulation. Therefore, volume conduction of reflex activity from the posterior tibial compartment to the anterior tibial compartment was a common observance. We suggest that coexistent EMG reflex responses, presumed to reflect reciprocal excitation, should be tested by the procedure described to reject the possibility of EMG cross-talk as a confounding variable or as the actual variable under investigation.

Post-contraction errors in human force production are reduced by muscle stretch.

1. Based on findings from a previous study of plantar-flexor muscles, the effect of a conditioning 25, 50 or 100% maximum voluntary contraction (m.v.c.) of elbow flexor muscles on the accuracy of reproducing a learned criterion muscle force (2% m.v.c.) was investigated. 2. Each conditioning contraction induced a significant error in reproducing the criterion muscle force under conditions of no visual feed-back of force. As with plantar-flexor muscles, the error was consistently in the direction of a positive bias. The magnitude of the error co-varied with the magnitude of the previous contraction and, in all cases, decayed toward criterion force values over a 35 s period. 3. A brief muscle stretch, induced before subjects attempted the criterion force, reduced the size of the error but did not completely eliminate the bias. 4. The findings provide indirect evidence of post-contraction potentiation of stretch reflex pathways. Residual post-contraction errors in force production after muscle stretch may be attributed to other central or peripheral neural factors such as, for example, potentiating effects of prior activation on submaximal tension production in skeletal muscle.

Soleus muscle electromyographic activity and ankle dorsiflexion range of motion during four stretching procedures.

Volitional muscle contractions are used frequently in some combination with muscle stretching to promote muscle relaxation and to increase range of motion. In this study, muscle lengthening procedures were evaluated in the ankle plantar flexors. Four soleus muscle stretching procedures--static stretch (SS), hold relax (HR) (isometric plantar flexor contraction before stretch), agonist contract (AC) (dorsiflexor contraction assisting stretch), and hold relax-agonist contraction (HR-AC)--were performed in the sagittal plane by 12 physically active adults. The dorsiflexion angle, soleus muscle electromyogram, and soleus muscle motoneuron excitability as determined by the Hoffmann-reflex (H-reflex) amplitude were measured throughout the duration of each stretch. The range of dorsiflexion achieved at the end of the stretch did not differ significantly between stretching procedures, although in 8 of the 12 subjects and in the subject group as a whole, the AC and HR-AC procedures were associated with higher levels of soleus muscle EMG than the levels in the SS and HR procedures (p less than .01). The H-reflex amplitudes during the AC and HR-AC procedures were smaller than the amplitudes during the SS and HR procedures (p less than .001), suggesting the possibility of reciprocal inhibition during the agonist contraction. Increased tonic EMG levels produced by input from other neural pathways affecting alpha motoneurons in the AC and HR-AC procedures may have masked this inhibitory reflex. In healthy adults, a complicated procedure, involving muscle contractions for decreasing active resistance to stretch, may be unnecessary because active resistance to stretch is minimal and muscle relaxation during stretch appears to have little or no direct effect on the ROM achieved.

Tendon vibration-induced inhibition of human and cat triceps surae group I reflexes: evidence of selective Ib afferent fiber activation.

In humans, prolonged vibration of the Achilles tendon produced transient depression or abolition of the soleus H-reflex. Recovery of the electrical reflex threshold to previbration values at a constant lower stimulus intensity usually occurred between 10 to 55 min. Electrical stimulation at higher multiples of the reflex threshold produced reflex EMG amplitudes more immediately comparable to previbration controls. When postvibration H-reflexes were completely abolished, poststimulus averaging of voluntarily maintained tonic EMG activity showed evidence of inhibition at a 46-ms latency in contrast to a 32-ms previbration H-reflex latency. In cat, observation of H-reflexes were rare, but stimulus-evoked changes in EMG activity mimicked the postvibration depression seen in humans. Ventral root postvibration reflexes from triceps surae varied in magnitude but were usually depressed or abolished at 1.0 to 1.2 times the electrical reflex threshold. These responses returned to previbration control amplitudes within 20 to 35 min. Magnitude of depression and time to recovery were dependent on the intensity of the electrical stimulus. In five experiments, depression of postvibration reflex activity and recovery were accompanied by gradual recovery in amplitude of the group I volley to previbration amplitudes. Elevated group Ia axonal electrical thresholds, monitored from seven isolated units, were observed to recover to previbration values in parallel with postvibration reflex recovery to control amplitudes. At electrical stimulus intensities greater than 1.4 times the reflex threshold, postvibration reflex responses were often potentiated, probably reflecting posttetanic potentiation of group Ia pathways activated at their higher axonal thresholds. In two observations, postvibration Ib axonal electrical thresholds did not change. Overall, the findings supported the proposal that postvibration depression of soleus H-reflexes in humans or cats is caused by both disfacilitation and autogenetic inhibition due to withdrawal of Ia afferent activation and increased selectivity of Ib afferent fiber stimulation, respectively.

Kinematic assessment of a functional role for recurrent inhibition and selective recruitment.

It has been proposed that recurrent inhibition, which is presumed to act more powerfully on type S than on FR or FF motoneurons, and a selective recruitment order of synergistic motoneurons may function to prevent slowly contracting muscles from impeding rapid contractions. The possibility that such mechanical interactions do occur was investigated kinematically by analysis of plantar flexion in the rat hind limb under loaded and unloaded muscle conditions. The left limb was denervated to isolate the lateral gastrocnemius-soleus nerve and muscles. Plantar flexion was induced before and after soleus denervation by posterior tibial-nerve stimulation with the ankle in one of three positions: (i) resting (unloaded), (ii) dorsiflexed (held by a restraint bar), or (iii) with ankle position set by a load applied in series to the intact Achilles tendon. With the initial ankle position near maximal dorsiflexion, peak velocities of foot movements associated with passive elastic muscle properties approached one-half or more of the values achieved by active contractile mechanisms. Under unloaded and loaded conditions, no evidence was found in support of mechanical interference by the slower contracting soleus muscle of maximal lateral gastrocnemius shortening responses. Contributions of soleus tension to foot velocity and acceleration began to emerge with loads greater than 3 N. These findings are in agreement with previous reports that the slow-contracting soleus muscle, though fully activated, cannot contribute effectively to plantar flexion movements at speeds above a critical level.

Was Sherrington right about co-contractions?

Differences in electromyographic (EMG) activity between brief isometric maximum voluntary contractions under conditions of agonist contraction and agonist-antagonist co-contractions were assessed for elbow flexors (biceps brachii) and elbow extensors (triceps, long head). Expressing maximum EMG co-contraction activity as a ratio of agonist maximum EMG activity of the same muscle yielded EMG ratios consistently below 1.0 (mean = 0.48 for flexion and 0.76 for extension). Flexor EMG ratios did not co-vary with elbow position but extensor EMG ratios approached 1.0 or sometimes greater at shorter muscle lengths (elbow extension). Findings were in general accord with Sherrington's original proposal of 'double reciprocal innervation'. Neural circuitry designed to limit full muscle activation during co-contractions may serve to protect the joint against excessive tangential or compressive forces.

Stretch sensitivity of Golgi tendon organs in fatigued gastrocnemius muscle.

Alterations in Golgi tendon organs' (Ib afferent receptors) stretch sensitivity were assessed in fatigued gastrocnemius muscle of cats. Muscle fatigue to 60-50% peak tension was induced in 13 cats by electrical stimulation of L7 and S1 ventral roots. Forty-three group Ib afferent fibers were recorded in L7 or S1 dorsal root filaments before and during fatigue. Fiber activity was assessed by applying to the Achilles tendon a 5-mm ramp stretch at 5 or 25-30 mm X s-1. Group Ib afferent responses to both velocities of stretch were either completely abolished or depressed over several seconds compared to pre-fatigue firing frequencies. When responding, longer latencies (measured at 5 mm X s-1 only) to firing onset occurred during fatigue. Increasing the rate of stretch (250 mm X s-1) to produce a 5-mm stretch approximately equivalent to twitch tension time revealed a depression more selective to static stretch sensitivity. Post-excitation depression of the Ib receptor potential appears to be one possible mechanism. These acute adaptations in Ib afferent discharge to muscle stretch were opposite in direction to those recently reported in fatigued muscle for group Ia and group II muscle spindle afferents. Functional implications of these results are discussed.

Dynamic and static stretch responses in muscle spindle receptors in fatigued muscle.

Dynamic and static stretch responses in muscle spindles were investigated in fatigued muscle to determine if acute adaptations do occur in receptor discharge as has been shown after contractions of short duration. Fatigue to 60-50% maximum tetanic tension was induced in the isolated gastrocnemius muscle in 16 cats by sustained, 7 X threshold electrical stimulation (100 Hz) of the cut L7 ventral root and S1 ventral root. Group Ia and II afferent fiber responses to slow ramp stretches (5 mm X s-1) and vibration (100 Hz) applied to the Achilles tendon were monitored before and immediately after muscle tetany to fatigue. Changes in firing characteristics were similar when results from faster (25-30 mm X s-1) ramp stretches were contrasted. During muscle fatigue, decreases in response latency to displacement and increases in resting discharge, mean frequency during stretch, and frequency of firing to vibration were predominant in both afferent fiber types. Static responses were generally lower, indicating a decrease in position sensitivity. Resting muscle force and passive peak muscle stiffness were consistently higher following contraction. The sum effects of these proprioceptive afferent and mechanical muscle responses would be to increase muscle stiffness and thus resist yield in muscle length to perturbations at lower muscle forces. The magnitude of these adaptations in proprioceptive discharge appears dependent on intrafusal muscle fiber activation.

Can Ib axons be selectively activated by electrical stimuli in human subjects?

A tendon-vibration technique, used to raise the electrical threshold of muscle spindle Ia afferent fibers above that of Golgi tendon organ Ib afferent fibers in animals, was tested on human subjects. After prolonged tendon vibration, electrical stimulation of the posterior tibial nerve was ineffective or markedly less effective in eliciting Hoffmann (H-) reflexes in the soleus muscle at previbration threshold intensities. With stimulus intensity held constant at values between 1.0 to 1.4 X threshold, postvibration H-reflex myoelectric amplitudes returned to previbration values usually within 60 min. However, at higher electrical stimulus intensities (1.8 X threshold), postvibration H-reflex amplitudes were produced at or near previbration values irrespective of postvibration recovery time; in contrast, initial postvibration tendon tap reflexes were potentiated. Findings suggest that it is indeed possible to selectively activate Ib afferent fibers in humans by electrical stimuli.

Activation history and constant errors in human force production.

Spinal proprioceptive reflexes and muscle tension are transiently potentiated in the aftermath of isometric muscle contractions. To determine if such alterations may bias ongoing motor commands, subjects were tested for accuracy of matching a criterion force following relaxation or a maximum contraction. Large postcontraction overestimations of the criterion force were consistently found. Errors progressively decayed to postrelaxation control values within 50 s. It is proposed that transient postcontractile potentiation of spinal reflex pathways may summate with previously set motor commands to produce inadvertent errors in perceived effort.

Changes in excitability of tendon tap and Hoffmann reflexes following voluntary contractions.

The effects on the Hoffmann reflex and tendon tap responses in the human soleus and medial gastrocnemius muscles of a brief conditioning volitional contraction have been compared with the objective of distinguishing contributions from the central nervous system and the periphery. The H-wave on successive trials over a 50 sec period following contraction for the most part demonstrated a depression, especially upon the initial trial. The T-wave responses in the first few trials were greater than those in the control period. Using the assumption that responses to tendon taps had been reduced by the amounts reflected in the excitability curve for H-response, it was deducted that a marked enhancement in T-wave excitability had been present at the initial tendon tap and more moderate facilitation thereafter. This is consistent with the post-contraction discharge and increase in stretch sensitivity of spindle receptors seen after contraction of a muscle in experimental animals. It is concluded that such effects on spindles may occur in man, maybe of such force as to detectably influence dynamic movements.

Electromyographic investigation of muscle stretching techniques.

Myotatic reflexes can be enhanced following brief static contractions. Since static contractions are often used as precursors to muscle stretch, rationale underlying these techniques were questioned and re-examined through electromyography (EMG). Twenty-one female gymnasts performed three methods to produce hamstring stretch: static (S), contract-relax (CR), and contract-relax with agonist (hip flexors) contraction (CRAC). Hip joint angles and intra-individual electromyograms were statistically compared across stretch conditions. In 12 subjects, the CRAC method elicited significantly greater hamstring EMG activity (P < 0.05) than the other techniques. A higher level of muscle activation was associated with the S method in only one subject. No significant differences in EMG activity across stretch conditions were found in eight subjects suggesting that the relative effectiveness of the stretch techniques varied across individuals. Involuntary paroxysmal tremor activity was occasionally visible in EMG records of most subjects at the low levels of muscle activation. While apparently contributing to increased muscle stiffness, the CRAC technique produced the largest gains in hip flextion. Rank orderings of minimum pain and maximum perceived stretch effectiveness were significantly related with one another, and with decreasing EMG activity, but not with range of motion.

Postcontractile motoneuronal discharge produced by muscle afferent activation.

Muscle afferent activity is markedly enhanced following isometric contractions in hindlimb musculature of various vertebrates. Such increased and persistent postcontractile sensory discharge (PCSD) consists for the most part of potentiated firing of muscle spindle primary afferents. To determine if a positive feedback function is served, involving recruitment of homonymous and synergistic motor units, 62 motoneurons were isolated in L7 ventral root filaments, categorized according to their reflex response to triceps surae stretch, and monitored before and after isometric contraction for variations in spike frequency. Motoneurons found reflexly coupled to triceps surae stretch receptors significantly increased (P less than .01) their firing frequency during PCSD. Tonic motoneurons, believed to be primarily of alpha origin, accounted for most of the observations. This latter finding is consistent with known effects of spindle afferent input on motoneuronal recruitment. Functionally, this autogenetic control circuit should enhance "muscle stifness" and stabilize the muscle against minor perturbations.

Persisting changes in sensory and motor activity of a muscle following its reflex activation.

A study was made to see if contractions induced by reflex action were followed by an elevation in muscle afferent activity comparable to the post-contraction sensory discharge (PCSD) that occurs after stimulation of the ventral roots. In lightly anesthetized cats, extension reflexes of the triceps surae muscle in the otherwise denervated leg were induced by contralateral stimulation of the popliteal nerve or footpad. The reflex response was followed by an elevated afferent discharge, which disappeared after a brief stretch, as is characteristic also of the PCSD. Crossed-extension responses in fusimotor activity unaccompanied by contraction of the gross muscle were also succeeded by an elevation in sensory discharge and an increased sensitivity to a vibratory stimulus applied to the tendon. Pinna reflexes had a similar effect. In the presence of the elevated discharge, alpha-motoneurons demonstrated increased sensitivity to vibration stimuli. It is suggested that the increase in spindle sensitivity may help stabilize the muscle against minor pertubations in length, at least when background fusimotor activity is relatively quiet.