Arthrogenic muscle inhibition is an ongoing reflex inhibition of the uninjured musculature surrounding a distended or damaged joint1,2 that affects the ability to activate motor neurons for recruitment during a contraction.3 This reflex inhibition is mediated presynaptically by γ-aminobutyric acid inhibitory interneurons and postsynaptically by Renshaw cells, both of which are influenced by supraspinal centers.2,4 This reflex inhibition decreases neural drive to the muscle, which prevents activation of motor neurons and consequently decreases the capacity for muscle contraction. Although arthrogenic muscle inhibition might be initiated as a protection mechanism to diminish excess forces on a joint after injury,4 this inhibition becomes problematic when muscle-activation deficits persist long after the initial injury5 and might limit the ability to regain optimal muscle function.6

Quadriceps muscle inhibition has been reported with various knee conditions, including anterior cruciate ligament injury,7–9 osteoarthritis,6,10–13 total knee arthroplasty,6,14–16 and meniscectomy.17,18 Quadriceps inhibition not only decreases the ability to perform maximal tasks but also can alter gait19 and shock-attenuation strategies during tasks such as landing.20 Arthrogenic muscle inhibition affecting the quadriceps is a serious clinical impairment that might need to be targeted with specialized therapeutic interventions before muscle strengthening.13,21,22

Conventional rehabilitation programs focusing solely on muscle strengthening might not provide adequate stimuli to effectively treat arthrogenic muscle inhibition5 and might contribute to the perpetuation of muscle dysfunction even when traditional therapeutic exercise is used.23 Researchers have theorized that specialized interventions might be necessary to overcome underlying muscle inhibition modulated by the central nervous system so patients can achieve a greater level of neuromuscular function and, thus, optimal therapeutic outcomes.22,23 Muscle function plays a vital role in shock attenuation at the joint,6 and decreased muscle activation might impair the ability of the neuromuscular system to protect the joint, which might lead to early joint degeneration.20,24 Therefore, an intervention that increases muscle activation during rehabilitation might increase neuromuscular control and improve function.23 Different therapeutic agents, such as transcutaneous electric nerve stimulation22 and focal knee joint cooling,25 have been reported to effectively increase quadriceps muscle activation. Pietrosimone et al23 suggested that improvements in muscle activation using a disinhibitory agent might allow patients to access previously inhibited motor neurons during therapeutic exercise and, hence, develop optimal motor patterns.

Recently, transcranial magnetic stimulation (TMS) used with maximal voluntary isometric contractions (MVICs) has been reported to increase quadriceps activation within an hour in healthy participants26,27 and participants with quadriceps weakness.28 Transcranial magnetic stimulation is a safe, relatively pain-free method of stimulating neural tissue.29,30 The magnetic stimulation is directed onto the region of the motor cortex that generates volitional movement in the periphery and subsequently excites corresponding descending corticospinal tracts that directly project on motor neurons within the targeted muscle.30 Interestingly, TMS has been reported to decrease presynaptic inhibition by affecting interneurons,31 which is a mechanism reported to modulate arthrogenic muscle inhibition.1,32 If this inhibition could be decreased before performing therapeutic exercise using a technique such as TMS, patients might have more motor neurons available to optimize neuromuscular function. Regaining optimal muscular function before developing severe joint degeneration might delay joint breakdown. Yet it remains unknown if TMS would effectively address quadriceps-activation deficits in patients who have inhibition and are at risk for hastened progression of osteoarthritis in the knee.

Patients who have had meniscectomy are good models for studying arthrogenic muscle inhibition because they have relatively standard postoperative joint damage and they tend to acquire quadriceps inhibition after these procedures.33 Consequently, this population also is at heightened risk of developing knee osteoarthritis,34 which might have links to altered muscle function.35 Researchers22,25,28,36,37 have assessed the ability of various disinhibitory agents to immediately activate a chosen muscle group in a time frame that closely mimics that of a clinical therapeutic exercise session. One motive for determining the immediate effects of these modalities is to demonstrate that these agents would be used in conjunction with conventional therapeutic exercise. Therefore, the purpose of our study was to determine the effect of receiving single-pulse TMS over the motor cortex and performing quadriceps MVICs on volitional quadriceps activation in participants after meniscectomy. We hypothesized that quadriceps activation would be greater in patients receiving TMS and performing MVICs than in a control group only performing MVICs.

METHODS

Thirty-one people volunteered for the study, but 11 volunteers were excluded before baseline testing because their quadriceps central activation ratios (CARs) were more than 85%. Twenty participants were assigned randomly to the TMS or the control group (Table 1). Participants were included if they had undergone partial meniscectomy via arthroscopy a minimum of 10 days before and a maximum of 3 years before the study (postoperative time  =  36.62 ± 35.06 weeks) and had quadriceps activation less than 85% when measured by the CAR with the burst-superimposition technique. Exclusion criteria included neurologic disorders, history of depression, history of seizures or migraines, and diagnosed heart conditions that precluded participants from exercising. Both investigators (C.E.G. and B.G.P.) performing the intervention before group assignment were blinded to the randomization, and the investigator (C.E.G.) collecting the muscle-activation data was blinded to group assignment throughout testing. The main outcome measure was volitional quadriceps muscle activation assessed by CAR, whereas the secondary outcome measures included quadriceps MVIC and knee pain during the contraction. All outcome measures were collected at baseline and at 0, 10, 30, and 60 minutes posttest. These time points were chosen (1) to provide a general sense of the muscle activation during a time interval that was similar to the length of a traditional therapeutic exercise session and (2) to resemble those used in previous studies22,25,26,28,37 in which disinhibitory agents were evaluated. Self-reported function was collected using the Lower Extremity Functional Scale38 and International Knee Documentation Committee Subjective Knee Scale.39 All participants provided written informed consent, and our study was approved by the Institutional Review Board (HRS 13399) at the University of Virginia.

Table 1 Group Demographics (Mean ± SD)^a

View Fullscreen

Burst Superimposition

Three submaximal electric stimuli (30, 70, and 100 V) were administered in conjunction with 3 submaximal isometric contractions (25%, 50%, and 75%) before testing to familiarize the participant with the burst-superimposition procedures. In addition to submaximal trials, participants performed 3 to 5 practice MVICs until the investigator was confident that each participant could exert a consistent maximal effort.22,25

The supermaximal exogenous stimulus was applied to the quadriceps when the test administrator observed that a maximal force plateau had been reached (Figure 1). Two trials separated by a 60-second rest were averaged at each time interval (baseline and 0, 10, 30, and 60 minutes posttest) for data analysis.

View Figure

• Download as Powerpoint
• Download Figure

At the beginning of each time interval, an MVIC without an augmented electric stimulation was performed. Each participant was instructed to rate the pain felt in the knee during the MVIC on the 100-mm VAS line.22

TMS Protocol

Before group allocation, participants were instructed to wear swim caps (Water Gear, Inc, Pismo Beach, CA) so we could locate optimal positioning of the motor cortex and to insert earplugs (Aearo Co, Indianapolis, IN) to muffle the sound of the TMS. Two landmark lines were drawn on the swim cap: 1 sagittally separating the hemispheres and 1 intersecting the sagittal line coronally at the external auditory meatus. During all contractions, participants were instructed to cross their arms over their chests and to close their eyes.

Before the TMS treatment, participants were instructed to extend their knees at 25% of their perceived MVICs while a submaximal magnetic stimulus was delivered to the motor cortex contralateral to the involved leg. The first 2 submaximal stimuli were performed at 40% and 60% of 2 T to familiarize the participants with the stimulus. The remaining submaximal stimuli were performed at 80% of 2 T, which induced a motor evoked potential but did not expose the participant to unnecessary maximal stimuli. Submaximal stimuli (average  =  5) were performed until the optimal positioning of the coil over the corresponding section of the motor cortex was located. We used the largest peak-to-peak motor evoked potential of the vastus lateralis (visualized on EMG) to determine the optimal placement for the magnetic coil. When located, the site was marked on the participant's swim cap.

The coil was secured in a holder (Magic Arm; Manfrotto Camera Lighting Supports, Bassano Del Grappa, Italy) and positioned over the previously located section of the motor cortex. Participants performed a contraction at 50% of their perceived MVICs while a stimulus of 90% of 2 T was administered to the motor cortex to ensure that the previously identified stimulation point could be accessed appropriately with the coil secured in the holder. After the submaximal trials, 3 quadriceps MVICs were performed 30 seconds apart, and 1 maximal magnetic stimulus (2 T) was administered when the participant reached a plateau in his or her MVIC (Figure 2). Before administering the magnetic stimulation, the investigator used both hands to guide the head of the participant directly under the coil, then removed his hands.

View Figure

• Download as Powerpoint
• Download Figure

Data Analysis

The CAR was calculated by dividing the force measurements of the MVIC (F_MVIC) by the peak force (sum of the force produced by the burst superimposition [F_SIB] and F_MVIC) as previously performed:

The peak force value and the MVIC value were calculated from the mean of the 2 best separate trials at each time in the series when the supermaximal stimulus was applied. The F_MVIC was calculated from a 0.1-second period immediately before the administration of the exogenous supermaximal stimulus.22 All MVICs were normalized to body mass. Pain scores were calculated by measuring the distance between the mark made by the participant and the end marked Absolutely No Pain. Pain scores were presented in millimeters, with higher scores representing higher levels of pain.

RESULTS

The CAR, MVIC, and VAS scores (means ± SDs) are reported in Table 2. We found no difference in CAR (t₁₈  =  −0.878, P  =  .39), normalized MVIC (t₁₈  =  −0.936, P  =  .36), or VAS (t₁₈  =  −0.803, P  =  .43) between the TMS and control groups at baseline (Table 2). A difference in CAR was seen over time (F_4,72  =  3.025, P  =  .02) but no interaction (F_4,72  =  1.457, P  =  .22, 1 − β  =  .431) or difference between groups (F_1,18  =  0.096, P  =  .76, 1 − β  =  .06; Figure 3). We noted no main effects over time for MVIC (F_4,72  =  0.79, P  =  .53, 1 − β  =  .24) or VAS (F_4,72  =  0.64, P  =  .64, 1 − β  =  .2) and no main effects between groups for MVIC (F_1,18  =  0.32, P  =  .58, 1 − β  =  .08) or VAS (F_1,18  =  0.564, P  =  .46, 1 − β  =  .11). In addition, no interactions were demonstrated for MVIC (F_4,72  =  1.49, P  =  .21, 1 − β  =  .44) or VAS (F_4,72  =  0.184, P  =  .95, 1 − β  =  .09). Moderate effect sizes were found at 10 (Cohen d  =  0.54, 95% CI  =  −0.33, 1.37) and 60 (Cohen d  =  0.50, 95% CI  =  −0.37, 1.33) minutes for CAR in the TMS group compared with baseline values, whereas the control group had small or weak effects at all time points compared with baseline (Table 3). Weak effect sizes were calculated for MVIC at all time levels when compared with baseline in the TMS and control groups (Table 3). When using percentage change scores to calculate effect sizes between groups at posttest, we calculated strong effect sizes for CAR at 10 (Cohen d  =  0.82, 95% CI  =  −0.13, 1.7) and 60 (Cohen d  =  1.06, 95% CI  =  0.08, 1.95) minutes and for MVICs at 10 (Cohen d  =  0.84, 95% CI  =  −0.11, 1.72) and 60 (Cohen d  =  1.22, 95% CI  =  0.21, 2.12) minutes in the TMS group (Table 3). Nine of 11 patients (82%) receiving TMS had an increase of 5 CAR percentage points, but only 5 of 9 patients (56%) in the control group had an increase of at least 5 CAR percentage points (Figures 4 and 5).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Table 2 Central Activation Ratio, Maximal Voluntary Isometric Contraction, and Visual Analog Scale for the Treatment Groups, Mean ± SD

View Fullscreen

Table 3 Effect Sizes (Cohen d) and 95% Confidence Intervals Within and Between Groups

View Fullscreen

DISCUSSION

We are the first to investigate the effects of TMS on muscle activation in patients with quadriceps inhibition after partial meniscectomy. We did not find an increase in CAR or MVIC after TMS when comparing groups (Figure 3). Authors have reported increases in quadriceps activation in healthy participants26,27 and increased quadriceps-contraction forces in patients with total knee arthroplasties after TMS interventions.28 The immediate effect of the TMS intervention in our participants might not have been as robust as that in the healthy26,27 and total knee joint arthroplasty populations28 of the previous studies. In addition, the dosage and nature of the intervention might not have been sufficient to have produced definitive effects in CAR. Only 3 stimuli were administered to the motor cortex in conjunction with MVICs, and this might not have been enough exposure to allow for disinhibition of an inhibited motor-neuron pool.

The objective of our study was to determine the effect of a single-pulsed TMS on CAR. Because other disinhibitory modalities, such as transcutaneous electric nerve stimulation22 and focal joint cooling,22,25 have been reported to elicit strong immediate effects on CAR, we hypothesized that TMS would have strong immediate effects on CAR. However, the sample size for our study might have been too small to reveal differences between groups using an intervention that might not elicit strong immediate effects. Our inability to identify differences between groups might have been due to relatively low power in detecting interactions (1 − β  =  .43) and between-groups effects (1 − β  =  .06). In addition, TMS might require successive interventions to reveal effects on CAR that are similar to those found immediately after focal joint cooling22,25 and transcutaneous electric nerve stimulation.22 Progressive repeated voluntary contractions in addition to a disinhibitory intervention also might be necessary to encourage the motor system to use previously inhibited motor neurons.

Although the level to which quadriceps CAR must increase to alter function has not been defined clearly, 4 of the 11 participants (36%) receiving the TMS had at least 1 posttest data point exceeding 10 CAR percentage points from their baseline scores, whereas an additional 5 of the 11 participants (45%) exceeded 5 CAR percentage points after TMS (Figure 4). Therefore, 9 of the 11 participants (82%) had an increase of 5 CAR percentage points (Figure 4). Conversely, only 5 of the 9 participants (56%) in the control group had at least a 5-CAR percentage-point increase, and only 1 (11%) had an increase of more than 10 CAR percentage points (Figure 5). Some participants in the control group had immediate increases in CAR, which might have been caused by the repetitive MVICs performed during the control intervention and might have been a product of the activation testing, but the TMS group had CAR increases that seemed to be greater and to occur in a larger portion of the population (Figures 4 and 5). Although our study was designed to assess discrete time points within a 60-minute interval, CAR might fluctuate differently among individuals after TMS. Furthermore, this informal qualitative data analysis might suggest that some individuals respond and others do not respond to TMS. Researchers of disinhibition should focus on determining if there is a systematic means of identifying participants who are likely to respond beneficially to a TMS intervention.

Although conventional interferential statistics did not reveal that CAR was different, moderate effect sizes at 10 and 60 minutes for CAR were found after TMS compared with baseline values, and strong effect sizes for TMS were calculated at both 10 and 30 minutes when comparing percentage change scores between groups at all posttests (Table 3). Although moderate to strong effect sizes were found for CAR at 10 and 60 minutes when comparing the TMS posttests with the baseline scores and when comparing TMS percentage change scores with those of the control group, 95% CIs for 3 of these point measures (not the 60-minute between-groups effect size) crossed zero, making it impossible to definitively suggest CAR increased after TMS (Table 3). Small effect sizes were noted for MVIC compared with baseline values in both the TMS and control groups (Table 3). Moderate effect sizes were found between groups at 0 minutes, whereas strong effect sizes were seen at 10 and 60 minutes in the TMS group (Table 3).

Quadriceps inhibition is a serious impairment that has been demonstrated in patients with acute7–9,17,18,42 and chronic joint injuries.12,13,22 The CAR has been reported to be affected by alterations in both motor-unit recruitment and firing frequency,43 which decrease the torque-generating capabilities of the muscle. The inability to optimally activate the muscles might lead to altered neuromuscular control, including altered kinematics during gait and dynamic activities (eg, landing) that might lead to or hasten the progression of joint degeneration. Conventionally, clinicians treat muscle-strength deficits without targeting the underlying arthrogenic muscle inhibition, which has been considered a limiting factor in joint rehabilitation and might require specialized therapy for regaining inhibited muscle.2 Therefore, identifying a disinhibitory agent that can adequately excite the neural system to improve the performance of therapeutic exercise is important in combating arthrogenic muscle inhibition. Moderate to strong effects for CAR immediately after an initial TMS treatment in this small population might be enough evidence to justify future studies in which the effects of TMS on CAR in patients with inhibited musculature are evaluated. In addition, researchers should evaluate potential pathways and neural mechanisms that can be influenced by TMS in patients with neuromuscular deficits after joint injury.

Magnetic stimulation of the motor cortex causes excitation of the neural tissue, which is transmitted through descending pathways to the muscle.44 At this point, the neural mechanisms that contribute to muscle activation in inhibited quadriceps after TMS are unknown. Iles31 reported that presynaptic inhibition from γ-aminobutyric acid interneurons is decreased after magnetic stimulation. A decrease in presynaptic inhibition, which ultimately down-regulates motor-neuron pool excitability, might increase the number of motor neurons available during an MVIC, increasing volitional activation. Authors have suggested that posttetanic potentiation or increased amplitude of postsynaptic potentials might play a role in the increased activation of muscle after TMS.27,28 Posttetanic potentiation is caused by a transient saturation of calcium-buffering systems, leading to excess calcium availability for each action potential.45 Increased calcium can produce increased neurotransmitter release from the presynaptic cell, which has the potential to increase synaptic transmission to corresponding postsynaptic cells. Increased postsynaptic transmission to previously inhibited alpha motor neurons might allow for increased activation of the muscle. Possible consequences of repeated TMS during MVICs might be long-term potentiation, which is thought to be associated closely with the effects described by previous authors.27,28 Long-term potentiation is the enhancement of synaptic transmission after exposure to certain stimuli that can last from hours to weeks. It is based on the Hebb rule, suggesting that “synapses that fire together will wire together.”46 Multiple treatments of TMS in conjunction with MVICs might be needed to cause long-term potentiation.

We noted methodologic differences between our study and previous studies of TMS.26–28 Past research28 on the effects of TMS in a population with quadriceps inhibition involved patients who had total knee arthroplasties and a mean age of 62 years. We studied patients after partial meniscectomy who had a mean age of 38.1 years. The variability in the mean weeks after surgery was large for both groups tested, yet means and SDs for both groups were similar (Table 1). For this initial investigation into the effects of TMS in the meniscectomy population, we were interested in studying patients who had marked clinical quadriceps inhibition (CAR ≤ 0.85) due to meniscectomy, regardless of the time since surgery. In future studies, researchers might consider whether an optimal postoperative time frame exists for administering a TMS intervention.

In addition, we used the burst-superimposition technique, whereas other researchers have used the interpolated-twitch technique.26–28 A similar figure-8–shaped coil was used in our study and in previous research, but our positioning was individualized for each participant. We located the area of the motor cortex that gave us the largest peak-to-peak motor evoked potential of the vastus lateralis. In past research, the coil was placed at a predetermined area for each participant (over the vertex 2 cm contralateral to the quadriceps being investigated between Cz and C3 tangentially to the midline).26–28 The number of treatment stimuli that we gave was the same as in past research.26–28 We performed an additional 5 submaximal stimuli before the treatment session during the location of the motor cortex.

CONCLUSIONS

Quadriceps CAR and MVIC were not increased after TMS compared with the control group. Moderate CAR effect sizes were found at 10 and 60 minutes after TMS compared with baseline scores, whereas strong effect sizes for TMS were noted for both CAR and MVIC at 10 and 60 minutes when comparing percentage change TMS scores with those of the control group.