437

Journal of Parkinson’s Disease 4 (2014) 437–452 DOI 10.3233/JPD-130316 IOS Press

Research Report

Cortical Mechanisms of Mirror Activation during Maximal and Submaximal Finger Contractions in Parkinson’s Disease Simon A. Sharplesa , Quincy J. Almeidaa,b and Jayne M. Kalmara,∗ a Department

of Kinesiology, Wilfrid Laurier University, Waterloo, ON, Canada Life Financial Movement Disorders Research & Rehabilitation Centre, Wilfrid Laurier University, Waterloo, ON, Canada

b Sun

Abstract. Background: Mirror movements are often reported in the early stages of Parkinson’s disease (PD) and have been attributed to bilateral activation of the primary motor cortex; however, the precise cortical mechanisms are still unclear. Subclinical mirror activation (MA) that accompanies mirror movement has also been reported in healthy aging adults. Objective: To characterize mirror activation and determine the cortical mechanisms of MA in individuals with PD who demonstrate mirror movements. Hypothesis: 5 Hz rTMS to the supplementary motor area (SMA) will reduce MA by increasing interhemispheric inhibition (IHI) of the ipsilateral motor cortex. Methods: MA was assessed using surface electromyography during maximal and submaximal unimanual contractions of the first dorsal interosseous in 7 individuals with PD with mirror movements (PD-MM: 70.9 ± 13.9 years; UPDRS III: 28.0 ± 8.2), 7 individuals with PD without mirror movements (PD-NM: 71 ± 10.1 years; UPDRS III: 27.8 ± 6.7) and 7 healthy controls (74.4 ± 6.0 years). IHI of the ipsilateral motor cortex was assessed using paired-pulse transcranial magnetic stimulation. Results: MA was enhanced in both PD groups during submaximal contractions, with the latest onset of activation in PD-NM. Ipsilateral motor cortex excitability was the highest in PDMM; however, IHI did not differ between PD and controls. 5 Hz rTMS to the SMA reduced IHI in PD-NM; however, did not affect MA. Conclusions: IHI may not be the sole contributor to the expression of overt mirror movements in PD. Expression of overt mirror movement may be due to the combination of enhanced ipsilateral motor cortex excitability and an earlier onset of electromyographic activation in the mirror hand (mirror activation) in PDMM. Keywords: Parkinson’s disease, transcranial magnetic stimulation, primary motor cortex, cortical inhibition

INTRODUCTION Mirror movements are involuntary movements that occur on the opposite side of the body during intended unilateral movement, whereas mirror activation is involuntary activation of the resting limb that can only be detected using EMG [1]. Both have been reported in early, asymmetric Parkinson’s disease (PD) [2–4] and ∗ Correspondence to: Dr. Jayne M. Kalmar, Department of Kinesiology, Wilfrid Laurier University, 75 University Ave West, Waterloo, Ontario, N2L 3C5, Canada. Tel.: +1 519 884 0710/Ext. 2033; Fax: +1 519 747 4594; E-mail: [email protected].

have been associated with reduced interhemispheric inhibition [5], possibly contributing to bilateral activation of the primary motor cortex. Mirror activation, however, is a natural phenomenon commonly reported in healthy individuals during isometric unimanual contractions and it tends to increase with contraction strength [6] and age [7, 8]. Higher levels of mirror activation have been reported during rhythmic submaximal contractions in individuals with PD compared to healthy controls; however, it is unclear why overt mirror movement is not expressed in all individuals with PD [4]. Therefore, the objective of this investigation

ISSN 1877-7171/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved

438

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

was to further examine the cortical mechanisms of mirror activation in individuals with PD that demonstrate overt mirror movements. Recent investigations on the cortical mechanisms of mirror movements in PD have suggested that motor overflow between hemispheres of the motor cortex may be due to altered interhemispheric inhibition of the ipsilateral motor cortex [5]. However, healthy unimanual control involves the interaction of a distributed network of cortical structures including not only both hemispheres of the primary motor cortex but also the supplementary motor area and premotor cortex [9]. The supplementary motor area in particular has been suggested to play an important role in the coordination of movement in the upper limbs through modulation of corticospinal neurons within the motor cortex [10, 11] and involves a great deal of interhemispheric cross talk at the level of this supplementary motor area and primary motor cortex during the movement planning and execution phases [12]. The notion of the supplementary motor area promoting unimanual movement through the recruitment of callosally-projecting neurons has been proposed based on investigations in the nonhuman primate. Callosally-projecting neurons located in the supplementary motor area and primary motor cortex have been shown to participate in the generation of unimanual and bimanual tasks and unimanual [13, 14] lesions to the SMA results in impairments in the ability of the animals to perform tasks with each hand independent of one another. Interestingly, this deficit is restored following callosal transection [15]. Bimanual coordination impairments reported in PD [16–18] may therefore be a product of reduced supplementary motor area activity [19, 20] due to its tight connectivity to the basal ganglia [11, 21]. It is therefore possible that coordination of movement in the upper limbs and suppression of mirror movement may be accomplished through activation of cortical circuitry acting within and between hemispheres of the primary motor cortex by the supplementary motor area. Interhemispheric inhibition of the ipsilateral motor cortex can be assessed using paired-pulse transcranial magnetic stimulation (TMS), where a suprathreshold conditioning stimulus is applied to the contralateral motor cortex 10–50 ms prior to a second suprathreshold magnetic stimulus in the ipsilateral motor cortex [22]. The resultant decline in motor evoked potential (MEP) amplitude elicited by the contralaterallyevoked conditioned stimulus, compared to a single unconditioned stimulus, provides a measure of the strength of the interhemispheric inhibitory pathway. Previous investigations of interhemispheric inhibition

and mirror movements in PD were assessed with the muscle at rest [5]; however, the potential contribution of premotor regulation of interhemispheric inhibition and mirror movements cannot be examined when these structures are inactive. In the present investigation, interhemispheric inhibition was assessed in a “precontraction” protocol with the muscle at rest, but 500 ms prior to the onset of rhythmic voluntary contraction when premotor structures are active [23]. As suggested by Li et al. [5], it would be expected that mirror movements in PD are expressed in part as a result of reduced interhemispheric inhibition of the ipsilateral motor cortex. In addition, the potential role of the supplementary motor area in the suppression of mirror movements was explored in the present investigation by offsetting reduced supplementary motor area activity in PD using high-frequency (5 Hz) repetitive TMS (rTMS) [24]. Repetitive TMS is a technique that has been extensively employed in clinical and laboratory settings to manipulate the excitability of small regions of the cortex [25]. When applied at high frequencies (>1 Hz) this method has been shown to not only increase the excitability of the stimulated region, but also influence other cortical structures that connect to the stimulated site [24]. Although not fully understood, the post-stimulation increases in excitability elicited by high-frequency rTMS may be mediated by mechanisms of long term potentiation [26] or posttetanic potentiation [27]. It was therefore hypothesized that offsetting reduced supplementary motor area activity would decrease recorded mirror activation during unimanual contractions by increasing interhemispheric inhibition between hemispheres of the primary motor cortex. METHODS Participants Thirty-two participants with PD who presented with mirror movements in the hand, mouth or lower limb associated with unimanual tasks were identified through the Sun Life Financial Movement Disorders Center during standard symptom assessments. Participants with PD were excluded if they presented with any neurological disorder other than Parkinson’s disease, demonstrated axial tremor or levodopa-induced dyskinesia. One participant did present with mild dyskinesia following enrolment but completed the study. Of the thirty-two participants identified, seven presented with overt mirror movements (PD-MM) in the hands only (70.9 ± 13.9 years; total UPDRS III: 28.0 ± 8.2) and

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

439

Table 1 Participant characteristics. 21 participants participated in this study. Motor symptom lateralization was quantified by calculating a Lateralized UPDRS score and Assymmetry Index. The Lateralized UPDRS score was expressed as the difference between left and right hand scores on items 20–25. The Asymmetry index was calculated as the ratio between the difference and sum of left and right hand scores (L–R/L+R) on items 20–25. For both the Lateralized UPDRS and Asymmetry index, a greater value indicates stronger asymmetry of motor symptoms in the upper limbs and a positive value identifies left side affected and negative values indicate right side affected. Data are presented as mean ± SD Group PD-MM PD-NM Control

Age

WHQ score

Total UPDRS III

Lateralized UPDRS

Asymmetry Index

70 ± 13.9 71 ± 10.1 74.4 ± 6.0

39 ± 15 36 ± 37 48.6 ± 6.8

28.1 ± 8.17 27.8 ± 6.7

1.5 ± 2.9 0.21 ± 2.5

0.13 ± 0.2 0.031 ± 0.2

were enrolled in this study. Seven age- and disease severity-matched participants with Parkinson’s disease that did not demonstrate overt mirror movements in the hands, mouth or lower limbs (PD-NM) were also recruited (71 ± 10.1 years; total UPDRS III: 27.8 ± 6.7), along with a group of 7 healthy agematched controls (74.4 ± 6.0 years). We actively recruited participants with minimal tremor. Only one participant in the PD-MM and one in the PD-NM actually had observable tremor. The PD-MM participant had a tremor score of 3 in the right and 2 in the left and in the PD-NM group had a score of 3 in the right 1 in the left. All other participants had tremor scores of 0 or 1. The average tremor score between groups was (PD-MM: Right: 0.92; Left 0.71; PD-NM Right: 1.04 Left 0.76). Eight participants with PD were left side affected (4 PD-MM, 4 PD-NM), 3 were right side affected (2 PD-MM, 1 PD-NM) and 3 participants demonstrated symmetric symptom severity (1 PD-MM, 2 PD-NM) in upper limbs assessed using the lateralized UPDRS and asymmetry index generated from the upper limb scores on items 20–25 of section III of the UPDRS. Upper limb rigidity was assessed using item 22 of section III of the UPDRS. Individuals in the PDNM group demonstrated greater upper limb rigidity in the right upper limb compared to those in the PDMM group (PD-MM: 1.3 ± 0.76, PD-NM: 2.1 ± 0.6; t(12) = 2.14, p = 0.053). Left upper limb rigidity was the same between PD-MM and PD-NM groups (PDMM: 1.86 ± 0.7, PD-NM: 1.64 ± 0.85). Twenty of the 21 participants were right hand dominant assessed with the Waterloo Handedness Questionnaire (WHQ) [28]. Prior to enrollment, all participants were screened for contraindications to TMS [29]. Participants attended one orientation session and two experimental testing sessions lasting approximately 1.5 to 2 hours in duration each. All experimental sessions were conducted at the same time of day, separated by at least 2 days to reduce any conditioning effects from the previous experiment. Lower incidence of mirror movement has been associated with increased symptom severity

in PD [2, 3] and a higher incidence reported in individuals who respond better to dopaminergic treatment [30]. The present investigation was conducted while participants with PD were on their dopaminergic medication. As such, the timing of medication intake relative to the start of each experiment was controlled for each experimental participant between days. Motor severity of PD symptoms was assessed using section III of the Unified Parkinson’s Disease Rating Scale (UPDRS) within one week and at the same time of day as experimental sessions to ensure an accurate representation of motor severity at the time experiments were conducted. Detailed participant demographic information is presented in Table 1 and the type, dose and elapsed time from medication intake on UPDRS assessment, experimental and control days are presented in Table 2. Protocol Participants attended two sessions in this repeated measures study; an experimental session and a control session that were randomized and counter-balanced. During each session, baseline measures of maximal voluntary contraction (MVC) strength of each hand and intracortical and interhemispheric inhibition were made followed by either 750 pulses of high frequency rTMS to the supplementary motor area on the experimental day or no rTMS on the control day where participants sat quietly for a time-matched duration of six and a half minutes. Post-rTMS measures of MVC in each hand and intracortical and interhemispheric inhibition were the same as baseline measures and completed within ten minutes of the completion of the rTMS protocol. Experimental setup Participants were seated in a modified automobile seat with the left and right forearm secured in thermomoldable splints mounted on armrests. Elbows were positioned at approximately 90◦ , with forearms pronated, and the thumb and index finger maintained at a constant angle of sixty degrees.

440

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

Table 2 Medication information. Individuals with PD participated in this study while ON all medication. Participants 1–7 were in the PD-MM group and 8–14 in the PD-NM group. All experiments and UPDRS assessments were conducted at the same time of day relative to medication intake for each participant and are presented as mean ± SD Participant

1 2 3 4 5 6

7

Average 8

9 10 11

12 13 14 Average

Medication type

Dosage

Sinemet Sinemet Sinemet Stalevo Rasagiline Sinemet Mirapex Stalevo Cipralex Sinemet Domperidone Lorazepam

25/100 25/100 25/100 200/100/25 1 mg 100/25 0.25 mg 200/100/2 10 mg 25/100 10 mg 1 mg

Venlafaxine Sinemet Eltroxine Gabapentin Sinemet Sinemet Sinemet Requip Rasagiline Sinemet Rasagiline Requip Sinemet

75/50 100/25 0.05 mg 300 mg 100/25 100/25 100/25 2 mg 1 mg 100/25 1 mg 8 mg 100/25

# per day

Time elapsed since last intake to start of session (minutes) Experimental

Control

UPDRS

4 4 4 4 2×4

120 210 20 60 130

120 30 120 60 130

240 345 10 165 515

3 4 4 1.2 × 3 1 1

60

60

150

145

180

160

95 ± 44 220

111.42 ± 69 100

226.42 ± 162.5 210

60 165 115

65 180 140

190 300 100

60 150

60 225

180 205

90 122.8 ± 59.1

15 112.1 ± 73.8

105 184.3 ± 69.2

Electromyographic and force data acquisition Surface electromyography (EMG) was recorded from the first dorsal interosseous muscle (FDI) of the left and right hand using silver-silver chloride electrodes that were epoxy-embedded with a preamplifier (40×) (EQ Inc., Chalfont, PA). A third surface EMG electrode was also positioned on the abductor digiti minimi (ADM) of the left hand to detect motor overflow in non-homologous muscles (Fig. 1A). Recording surfaces of the EMG electrodes were 0.8 centimetres in diameter with an interelectrode distance of 2 cm and positioned over each muscle belly. Preamplifiers fed into a variable-gain second stage amplifier (20×) for a total gain of 800×. Force data were recorded with a custom force dynamometer positioned over the lateral aspect of the first interphalangeal joint of the right and left index finger. The force signal was amplified with a custom-built amplifier (10×). Experimental protocols were sequenced and data collected using frame-based data acquisition software (Signal, Version 4, Cambridge Electronics Design, UK). EMG

1 4 1 1 4 4 6 1 1 4 1 3 8

and force signals were digitized at 5 kHz (Micro3 1401, Cambridge Electronics Design, UK). Following data acquisition, surface EMG was band-bass filtered from 10 to 1000 Hz and force low-pass filtered at 50 Hz. Signals were then stored on a laboratory computer for offline analysis. Baseline measurements began with maximal abduction of the left and right index fingers to record maximal voluntary contraction strength. Participants were given three attempts with three minutes between contractions to prevent fatigue. The peak force attained was defined as the participant’s maximum. Transcranial magnetic stimulation TMS of the right motor cortex was performed by delivering single monophasic magnetic pulses through a 9 cm figure of eight coil connected to two Magstim 200 magnetic stimulators via a BiStim module (Magstim Company, UK). The handle of the coil was positioned posteriorly at an angle of 45 degrees to the midsagittal line with the induced current in a

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

441

Fig. 1. Surface EMG was recorded in bilateral FDI and left ADM during low level isometric voluntary contractions of the right FDI (A). Grey lightning bolts denote conditioning and test pulses from the TMS coils. Long and short interhemispheric inhibition was measured from the active to mirror hemisphere (B) by delivering a suprathreshold conditioning pulse to the left hemisphere 10 and 40 ms (CCS10/40) prior to a test pulse to the right hemisphere (TS). Short-interval-intracortical inhibition was assessed in the resting hemisphere (B) by delivering a subthreshold conditioning stimulus to the right hemisphere 2 and 3 ms (CS2/3) prior to the test pulse (TS). In the TMS protocol (C), measures of cortical excitability and inhibition were made 500 ms prior to rhythmic low-level isometric contractions (5% MVC) of the right FDI. Contractions were maintained for 2 seconds with 3 seconds rest between contractions to prevent neuromuscular fatigue. The frame-based software allowed static cursors to be placed at 500 ms and 2.5 s to be used as cues to contract and relax the right FDI. This allowed for contractions to be rhythmic and predictable in nature.

posterior to anterior direction. Optimal coil position was determined by placing the coil tangentially to the scalp by moving the coil in 1 cm increments around the presumed FDI hotspot in the primary motor cortex (M1). The location that elicited finger abduction and the largest MEP was considered optimal and was digitized and saved using a coil tracking system (TMS Manager, Northern Digital Instruments, Waterloo, CAN). This experiment required the use of 2 coils, one to stimulate each hemisphere. Coil position for the left motor cortex was determined using the same method as the right motor cortex. Single biphasic magnetic stimuli were delivered through a 9 cm figure of eight coil over the left hemisphere using a Magstim Rapid2 Stimulator (Magstim Company, UK) (Fig. 1A).

monophasic magnetic stimulus and was defined as the minimum stimulator output required to elicit an MEP with a peak-to peak amplitude ≥ 50␮V in 5 out of 10 trials. Active motor threshold was determined in the left motor cortex with single biphasic magnetic stimuli and was defined as the minimum stimulator output required to elicit an MEP with a peak-to-peak amplitude ≥ 200␮V in 5 out of 10 trials during a low-level (5% MVC) contraction. Motor threshold was also measured at rest, during a 500 ms epoch prior to the onset of an anticipated isometric contraction of the right FDI. This threshold is referred to as “pre-contraction motor threshold” [23]. This threshold was determined in the right motor cortex with single monophasic magnetic stimuli using the same criteria as at rest.

Stimulation parameters

Assessment of interhemispheric inhibition and intracortical inhibition

Stimulation intensities were determined at the beginning of each experiment. Resting motor threshold was measured in the right motor cortex with a

Paired-pulse TMS was used to assess interhemispheric and intracortical inhibitory circuits of the

442

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

Table 3 rTMS parameters. Participants attended two sessions on separate days; an Experimental session where high frequency (5 Hz) rTMS was applied to the supplementary motor area and a Control session where the stimulation period was replaced with rest for a time-matched duration of the stimulation protocol. Stimulator intensity was set to 110% of active motor threshold (AMT) which was determined at the start of each session in the left motor cortex during a submaximal (5% maximal) contraction of the right FDI. Active motor threshold and rTMS stimulator intensity was consistent across days and groups. Data are presented as mean ± SD as a percent of maximal stimulator output (%MSO) Group PD-MM PD-NM Control

AMT (%MSO)

110% AMT (%MSO)

Experimental

Control

Experimental

Control

47.9 ± 4.5 42.7 ± 8.4 47.3 ± 8.8

47.6 ± 4.1 42.6 ± 3.7 48.1 ± 7.5

52.6 ± 4.9 47.0 ± 9.3 52.0 ± 9.7

52.3 ± 4.5 46.8 ± 4.3 53.0 ± 8.2

primary motor cortex. Short-interval intracortical inhibition was assessed in the right (mirror) motor cortex by delivering a subthreshold conditioning stimulus at both 2 ms and 3 ms prior to a suprathreshold test stimulus from the same coil [31] (Fig. 1B). The conditioning stimulus was set to 80% of the pre-contraction motor threshold. Interhemispheric inhibition was assessed by delivering a suprathreshold biphasic conditioning stimulus from the Rapid2 to the left (active) motor cortex at one of two intervals (10 ms or 40 ms) prior to the test pulse delivered to the right motor cortex (mirror) using a monophasic magnetic stimulus from the Bistim.2 This protocol (Fig. 1B) allowed us to assess both short latency (IHI10) and long latency (IHI40) interhemispheric inhibition. The interhemispheric conditioning pulse was set to a stimulator output to evoke a MEP with a peak-to-peak amplitude of 1 mV in the right FDI 500 ms prior to a low level contraction [22, 32]. Intracortical and interhemispheric inhibition were both assessed at two test pulse intensities as the rTMS protocol has previously been reported to increase MEP amplitude [24], which also affects measures of inhibition [33]. The test pulse was therefore set to a stimulator output to elicit an unconditioned MEP of 1 mV and 3 mV in the left FDI 500 ms prior to a low level contraction in the right FDI. Cortical inhibition was measured with the muscle at rest, but 500 ms prior to the onset of a unimanual isometric contraction of the right FDI (Fig. 1 C). Participants were provided with visual feedback of the force recording of the right hand on a computer screen positioned at eye-level three feet in front of them. Frame-based software was used with frames that were 5 s in duration. This allowed for static vertical cursors to be placed at 500 ms and 2.5 s which were used to provide cues for participants to contract and relax the right FDI [23]. Magnetic stimuli were delivered at the start of each frame, approximately 500 ms prior to the onset of each contraction (voluntary EMG onset: PD-MM: 392 ± 90 ms, PD-NM: 356 ± 90 ms Control:

Average rTMS stimulator intensity 52.5 ± 4.7 46.9 ± 6.1 52.5 ± 8.9

369 ± 80 ms). The rhythmic nature of the contraction made the task predictable and more internally generated in attempt to increase supplementary motor area activation [34, 35]. The target force of 5% MVC was provided with a horizontal cursor on the force feedback screen. In total, participants were required to perform contractions for 2 seconds with 3 seconds rest in between contractions to reduce the effects of fatigue (Fig. 1 C). Repetitive transcranial magnetic stimulation Repetitive TMS was employed to alter activity of the supplementary motor area using a 9 cm aircooled figure of eight coil connected to a Rapid2 Magnetic stimulator (Magstim Company, UK). The supplementary motor area was located and stimulated by positioning the focal point of the coil 3 cm anterior to vertex with the handle pointing to the left, perpendicular to the midsagittal line to preferentially activate the left supplementary motor area [24, 36, 37]. A total of 750 pulses (5 trains of 150 pulses with an inter train interval of 60 s) were delivered to the supplementary motor area at 110% of active motor threshold [24]. rTMS stimulator intensities applied on each day are presented in Table 3. Data analysis MEP analysis was performed using frame-based software (Signal, Cambridge Electronics Design) by generating average waveforms from ten trials for each condition measuring cortical excitability and inhibition. Peak-to-peak MEP amplitudes are reported in millivolts (mV) and conditioned MEP amplitudes for measures of cortical inhibition are reported as a ratio between the conditioned MEP and the unconditioned test MEP amplitude. Trials were excluded when the MEP was superimposed on voluntary or involuntary EMG activity to ensure measures were a reflection of pre-contraction excitability [23]. Stimulator output

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

intensities for motor thresholds and test MEPs are reported as percent of the maximal stimulator output (%MSO). Root mean square (RMS) was used as an amplitude measurement of electromyographic activity during voluntary and involuntary contractions. The amplitude of the EMG during the submaximal voluntary contraction in the right FDI and involuntary contraction in the left FDI is expressed as a percentage of the maximal surface EMG of the respective FDI. All data are reported as mean ± standard deviation. Excitability indices were generated for each participant using the slope of the regression line generated by plotting MEP amplitude against stimulator output required to elicit responses at resting motor threshold and unconditioned test MEPs of 1 mV and 3 mV from data collected at baseline. The correlation coefficient of the regression line fitted to this data for each participant was 0.93 ± 0.08. This analysis provides a region of the TMS input-output curve (for example see [38]) that permits corticospinal excitability to be assessed across a greater portion of the corticospinal motor neuron pool. The effects of rTMS to the supplementary motor area on cortical excitability and inhibition were determined by changes in unconditioned and conditioned MEP amplitude. Mirror activation was measured bi-directionally during maximal voluntary contractions of the right and left FDI and quantified as a ratio of the maximal voluntary EMG amplitude of the respective FDI (i.e., left mirror activation measured during right MVC/left voluntary activation measured during left MVC). EMG amplitude was also measured in the left abductor digiti minimi muscle to measure motor overflow between non-homologous muscles during maximal contractions of the right and left FDI and is expressed relative to the voluntarily contracting FDI. Submaximal mirror activation was measured during rhythmic submaximal contractions (5% MVC) of the right FDI and identified when the EMG amplitude of the left FDI, measured during voluntary contraction of the right FDI, exceeded the EMG amplitude of the left FDI, measured 1 second after the cue to relax for the right FDI. Submaximal mirror activation incidence is presented as percent of all trials before and after rTMS. Average waveforms were generated from the identified frames and sub maximal mirror activation amplitude measured by calculating the ratio between RMS of the left (mirror) and right FDI (voluntary) normalized to the respective RMS max Submaximal mirror activation = (RMSmirror /RMSmax )/(RMSvoluntary /RMSmax )

443

A value of 1.0 indicates the left (mirror) FDI was activated to the same extent of the right (voluntary) FDI. Voluntary and mirror EMG onset was defined as the time when the EMG amplitude exceeded twice the baseline signal and sustained for greater than 500 ms. Overall, PD motor symptom severity was assessed using Section III of the Unified Parkinson’s disease rating scale (UPDRS). A lateralized UPDRS score and Asymmetry index were calculated for all participants as mirror movements in PD have been described as a characteristic of asymmetric PD [2]. The lateralized UPDRS score was expressed as the difference between left and right hand scores on items 20–25. The Asymmetry index was calculated as the ratio between the difference and sum of left and right hand scores (L-R/L+R) on items 20–25. For both the Lateralized UPDRS and Asymmetry index, a greater value indicates stronger asymmetry of motor symptoms in the upper limbs and a positive value identifies left side affected and negative values indicate right side affected. Statistical analysis Data were pooled at baseline between days to determine differences in overall mirror activation and cortical excitability between groups (PD-MM, PD-NM and Control). Mixed analyses of variance (ANOVA) were conducted to determine the effect of hand dominance or affected side on mirror activation in homologous and non-homologous muscles during maximal contractions of the right or left hand. The effect of contraction strength intensity on mirror activation was determined by conducting a mixed ANOVA. One-way ANOVAs were performed to determine differences in submaximal mirror activation amplitude, incidence (percent of trials) and mirror onset during rhythmic unimanual contractions of the right FDI and also differences in cortical excitability (stimulator output at resting and pre-contraction motor thresholds, unconditioned test MEPs of 1 mV and 3 mV and excitability index slope), as well as interhemispheric and intracortical inhibition between groups. Pearson correlations were conducted to determine the association between maximal or submaximal mirror activation and handedness scores (WHQ score), interhemispheric inhibition, intracortical inhibition and cortical excitability, as well as clinical measures of disease severity and asymmetry (total UPDRS III, Lateralized UPDRS, Asymmetry Index) within PD groups. The average of the baseline data for the experimental and control days were used for these

444

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

correlations. Pearson correlations were performed with all three groups pooled. Separate analyses were also conducted within each group to determine correlates of enhanced mirror activation in individuals that demonstrate overt mirror movements. The effect of stimulating the supplementary motor area on submaximal mirror activation and measures of cortical excitability and inhibition was determined by conducting a two factor mixed ANOVA with day (5 Hz and Control) and time (pre-rTMS and post-rTMS) as within subject factors. The effect of rTMS to the supplementary motor area on maximal mirror activation was determined by performing a three factor mixed ANOVA with day, time and hand dominance (dominant and non-dominant) as within subject variables. RESULTS Is mirror activation only observed in people with PD that demonstrate overt mirror movements? Mirror activation of the left FDI occurred during rhythmic voluntary submaximal contractions of the right FDI in all three groups. Although voluntary activation of the right FDI was consistent between groups (PD-MM: 3.4 ± 2.0%RMSmax, PD-NM: 3.9 ± 3.6%RMSmax, Control: 4.4 ± 2.3%RMSmax) the incidence of mirror activation was significantly higher in individuals with PD that demonstrate overt mirror movements (Fig. 2A; F(2,39) = 3.86, p = 0.03) and only occurred between homologous muscles. The amplitude of mirror activation of the left FDI was highest in the PD-MM group (Fig. 2B; F(2,39) = 3.3, p = 0.048). The onset of mirror activation was the earliest in the PD-MM group and the latest in individuals with PD that do not demonstrate overt mirror movements (Fig. 2 C; F(2,39) = 4.25, p = 0.02). Mirror activation was also assessed during brief maximal contractions of the right and left hand. There was no difference in the amount of mirror activation during maximal contractions between groups in either the dominant or non-dominant hand (Fig. 3A). This was true whether mirror activation was assessed in the contralateral homologous muscle (FDI, Fig. 3A) or a contralateral non-homologous muscle (ADM). The degree of mirror activation during maximal contractions did not differ between the most and least affected hand as indicated using the lateralized UPDRS and asymmetry index in the PD-NM and PD-MM groups. Maximal mirror activation across all three groups was greater in individuals with a higher resting motor threshold (Fig. 3B; r = 0.49, p = 0.02). In

Fig. 2. Submaximal mirror activation incidence(A) and amplitude (B) assessed in the left FDI during rhythmic low-level (5% MVC) unimanual contractions of the right FDI was the greatest in individuals with PD who demonstrate overt mirror movements (PD-MM). Overt mirror movement may not be present in others with PD (PDNM) due to later onset of mirror activation (C). Data are presented as mean ± SE. Significant differences are shown with asterisks (p < 0.05).

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

445

Fig. 3. Mean ± SE mirror activation assessed during maximal unimanual contractions of the dominant and non-dominant FDI was consistent across all three groups and unaffected by hand dominance (A). Maximal mirror activation in the non-dominant hand was associated with lower corticospinal excitability (higher resting motor threshold) in the corresponding right motor cortex (B).

Fig. 4. Healthy older adults demonstrate an increase in maximal mirror activation with age.

healthy controls, maximal mirror activation was positively correlated with age (Fig. 4; r = 0.83, p = 0.02). Individuals with PD do not demonstrate any relationships between mirroring during maximal contractions and age, handedness, overall motor severity or symptom lateralization. A mixed ANOVA revealed a significant group x contraction intensity (maximal vs. submaximal) interaction (F(2,39) = 3.5, p = 0.04), where healthy controls demonstrated greater mirror activation during maximal contractions (p = 0.09) and individuals with PD-NM demonstrated greater mirror activation during submaximal contractions (p = 0.08); however, these did not reach significance. Mirror activation was the same during maximal and submaximal contractions in the PD-MM group (Fig. 5).

Fig. 5. Healthy controls demonstrated higher mirror activation during maximal contractions compared to submaximal contractions where PD-NM demonstrate higher mirror activation during submaximal contractions compared to maximal contractions; however, these did not reach statistical significance. Individuals with overt mirror movements did not demonstrate any difference in the amount of mirror activation between tasks.

Do alterations in cortical excitability and inhibition contribute to mirror movements? Resting motor threshold assessed in the right motor cortex and active motor threshold assessed in the left motor cortex (during a 5% contraction) were consistent between days and the same between groups. We therefore assessed excitability at three different stimulator intensities (resting motor threshold and test MEP amplitudes of 1 mV and 3 mV) to index excitability across a greater portion of the corticospinal motor neuron pool. The slope of the excitability index indicated

446

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

Fig. 6. Regression lines generated for each participant between data of stimulator output and motor response amplitude at resting motor threshold and unconditioned MEPs of 1 mV and 3 mV pooled at baseline across two days (A) demonstrate that corticospinal excitability in the ipsilateral (mirror) motor cortex is enhanced in individuals with PD who demonstrate overt mirror movements (PD-MM) compared to those who do not (PD-NM) and healthy controls. The excitability index is indicated by the slope of the regression lines and is plotted in (B). Data are shown as mean ± SE. Significant differences are shown with asterisks (p < 0.05).

that ipsilateral motor cortex excitability was significantly higher in individuals with PD that demonstrate overt mirror movements compared to those with PD that do not demonstrate overt mirroring or healthy controls (Fig. 6B; F(2,39) = 6.56, p = 0.03). Paired-pulse TMS was used to assess interhemispheric inhibition of the ipsilateral (mirror) motor cortex by the contralateral (active) motor cortex and intracortical inhibition within the ipsilateral motor cortex prior to unimanual contractions of the right FDI. As expected, the peak-to-peak amplitude of the test MEP elicited in the left FDI was reduced via the interhemispheric inhibition pathway when conditioned by a suprathreshold MEP of 1 mV in the right FDI (PD-MM: 1.15 ± 1.17 mV, PD-NM: 1.32 ± 0.83 mV, Control: 0.79 ± 0.5 mV) at both 10 ms and 40 ms interstimulus intervals. Test MEP amplitude was also reduced in the ipsilateral motor cortex when preceded by a subthreshold conditioning stimulus applied through the same coil to the right motor cortex at both 2 ms and 3 ms interstimulus intervals. There was no effect of test MEP amplitude on interhemispheric or intracortical inhibition within any group (Fig. 7). Interhemispheric inhibition assessed with a 10 ms interstimulus interval appeared to be reduced in PDMM and PD-NM groups compared to healthy controls when assessed with a 1 mV test MEP and also at a 3 mV test MEP in the PD-MM group; however, these differences were not statistically significant (Fig. 7). We therefore pooled the two PD groups and performed a t-test comparing PD and healthy control groups for IHI10 at 1 mV. Individuals with PD (PD-NM and

Fig. 7. Interhemispheric inhibition was assessed from the contralateral (active) to ipsilateral (mirror) primary motor cortex at interstimulus intervals of 10 ms (IHI10) and 40 ms (IHI40) and intracortical inhibition within the ipsilateral motor cortex at interstimulus intervals of 2 ms (SICI2) and 3 ms (SICI3) using test MEPs of 1 mV (A) and 3 mV (B) in three groups (PD-MM, PD-NM and Control). Inhibition is presented as the conditioned MEP/Unconditioned MEP where a value of 1.0 represents no inhibition. Bars represent data pooled at baseline on the two days and are presented as mean ± SE.

S.A. Sharples et al. / Cortical Mechanisms of Mirror Movements in PD

447

Table 4 Mirror activation during maximal contractions. Mirror activation in the FDI of the dominant and non-dominant hand during maximal unimanual contractions were consistent across groups and unaffected by high frequency rTMS applied over the supplementary motor area (SMA). Mirror activation is expressed as the ratio between RMS of mirror activation (during maximal contraction of the contralateral FDI) and maximal voluntary activation. Data are presented as mean ± SD on experimental (rTMS to SMA) and control days. F and p values are presented from a three factor mixed ANOVA with day, time and hand as within subject factors Dominant hand PD-MM PD-NM Control Non-dominant hand PD-MM PD-NM Control

rTMS to SMA

Control

Pre

Post

Pre

Post

F(2,18)

p value

0.18 ± 0.17 0.16 ± 0.12 0.2 ± 0.25

0.17 ± 0.14 0.08 ± 0.06 0.14 ± 0.15

0.16 ± 0.14 0.19 ± 0.18 0.3 ± 0.4

0.1 ± 0.13 0.2 ± 0.27 0.2 ± 0.27

0.92

0.42

0.29 ± 0.23 0.14 ± 0.16 0.29 ± 0.39

0.25 ± 0.16 0.17 ± 0.2 0.46 ± 0.55

0.24 ± 0.17 0.16 ± 0.14 0.26 ± 0.3

0.25 ± 0.23 0.17 ± 0.22 0.7 ± 0.3

Table 5 Mirror activation during submaximal contractions. Submaximal mirror activation measured in the left FDI during rhythmic submaximal contractions (5% MVC) of the right FDI was not affected by high frequency rTMS applied over the supplementary motor area (SMA). Mirror activation is expressed as a ratio between RMS of the left (mirror) and right FDI (voluntary) normalized to the RMS during respective maximal voluntary contractions. Data are presented as mean ± SD on experimental (rTMS to SMA) and control days. F and p values are presented from a two factor mixed ANOVA with day and time as within subject factors Submax MA PD-MM PD-NM Control

rTMS to SMA

Control

Pre

Post

Pre

Post

F(2,18)

p value

0.44 ± 0.33 0.45 ± 0.38 0.15 ± 0.08

0.5 ± 0.38 0.1 ± 0.07 0.18 ± 0.12

0.27 ± 0.2 0.15 ± 0.06 0.11 ± 0.09

0.38 ± 0.29 0.12 ± 0.03 0.11 ± 0.05

2.4

0.12

PD-MM) did not differ significantly from controls with respect to interhemispheric inhibition (t(40) = 1.88, p = 0.067). Is mirror activation reduced by 5 Hz rTMS to the supplementary motor area through alterations in interhemispheric inhibition? 5 Hz rTMS of the supplementary motor area did not alter the amount of mirror activation during maximal (Table 4) or submaximal (Table 5) unimanual contractions in any group. The results from a mixed ANOVA was a three-way interaction between group (PD-MM, PD-NM, control), day (stimulation and control) and time (pre and post) for both IHI and SICI (3 ms only) when assessed with a high intensity test pulse (3 mV). Specifically, 5 Hz rTMS to the supplementary motor area reduced interhemispheric inhibition at 10 ms (Fig. 8A; F(2,18) = 5.4, p = 0.014) and 40 ms (Fig. 8B; F(2,18) = 4.4, p = 0.03). It also reduced intracortical inhibition assessed with a 3 ms (Fig. 10 C; F(2,18) = 6.4, p

Cortical mechanisms of mirror activation during maximal and submaximal finger contractions in Parkinson's disease.

Mirror movements are often reported in the early stages of Parkinson's disease (PD) and have been attributed to bilateral activation of the primary mo...
802KB Sizes 0 Downloads 4 Views