Bilateral motor cortex output with intended unimanual contraction in congenital mirror movements

Methods.

A 15-year-old girl (Patient 1) and a 40-year-old woman (Patient 2), who were both right-handed, presented strong and sustained congenital MM affecting both hands and forearms. Patient 2, whose mother and brother also showed MM, previously underwent detailed clinical and neurophysiologic investigation.3 The family history of Patient 1 was negative. In both patients, developmental milestones and the results of the general and neurologic examinations were normal except for MM. Head and cervical spine MRI was unremarkable. In both patients, focal TMS of either M1 produced bilateral motor evoked potentials (MEP) of normal latency in the resting abductor pollicis brevis (APB) muscles. The control group consisted of 7 healthy subjects (4 females and 3 males, mean age of 34.1 ± 7.8 years). Patients, control subjects, and the parents of Patient 1 gave their informed consent for the study.

Subjects were seated comfortably in an armchair. A surface electromyogram (EMG) (0.1–2-kHz filter) was recorded from both APB. Subjects maintained a brief intended unilateral isometric contraction of either the right or left APB at an intermediate strength level. Focal TMS was delivered using Magstim 200 stimulators (Magstim Co., Whitland, Dyfed, UK) and eight-shaped coils (each loop with an external diameter of 9 cm) centered on the interaural line 7 cm lateral from the midline. The coil handles pointed backward and 45° away from the midline. Motor threshold (MT) was determined using 5% increments of stimulator output until reaching the minimum intensity that produced MEP of >50 μV in at least half of 10 consecutive trials. TMS intensity was set to 20% above MT. Three conditions were tested: unilateral M1 stimulation contralateral and ipsilateral to the intended APB contraction and bilateral simultaneous TMS. In the patients, bilateral TMS produced larger MEP than unilateral TMS (table⇓). To match MEP sizes, bilateral TMS was applied in addition at 10% above MT (see the table⇓). Ten trials were recorded for each condition at intertrial intervals of 10 seconds. CSP duration was determined in the single-trial–rectified and averaged EMG from the stimulus to the point when the mean post-MEP EMG returned to 20% of the mean prestimulus EMG.

Results.

In both patients, intended unilateral contraction of each APB produced bilateral EMG activity (figure). Unilateral stimulation of either M1 during intended unilateral contraction of either APB resulted in a significantly shortened CSP duration in both muscles if compared with the CSP duration in the control subjects (see figure, A and B and the table⇑). In contrast, simultaneous bilateral M1 stimulation during intended unilateral APB contraction resulted in a CSP of normal duration in both APB in both patients (see the table⇑). If stimulated bilaterally at a reduced intensity to match MEP size with the unilateral stimulation conditions, the duration of the CSP was still clearly longer than that of the CSP in the unilateral stimulation conditions (see figure, C and the table⇑).

• Download figure
• Open in new tab
• Download powerpoint

Figure. Cortical silent period (CSP) during an intended unilateral isometric contraction of the right abductor pollicis brevis (APB) in Patient 2. Black lines are the averages of 10 rectified electromyographic (EMG) sweeps recorded from the patient’s right (left column) and left (right column) APB with focal transcranial magnetic stimulation (TMS; 20% above the resting motor threshold [MT]) of either the left (A) or right (B) primary motor cortex (M1) or simultaneous TMS (10% above the resting MT) of both M1 (C). Dark gray areas (mean) and light gray areas (mean + 3 SD) refer to the rectified EMG of the right APB in the control subjects (n = 7; A,B: TMS of left M1; C: simultaneous TMS of both M1; all conditions at 20% above resting MT). Note that unilateral TMS of either M1 (A,B) produced an abnormally short CSP with an early recovery of the EMG (exceeding the mean + 3 SD of the control subjects) in either APB of the patient, whereas simultaneous TMS of both M1 (C) resulted in a normalization of the CSP in either APB.

In the control subjects, the CSP after bilateral TMS was shorter than the CSP after unilateral TMS (see the figure and table⇑).

Discussion.

The novel finding is significant CSP lengthening in bilateral compared with unilateral M1 stimulation in patients with congenital MM during intended unilateral hand contraction. How does this relate to the origin of the EMG activity in the mirror hand?

To answer this question, several TMS effects need to be considered. The CSP was the principal measure here. It reflects activation of inhibitory cortical circuits, resulting in interruption of voluntary EMG in muscles contralateral to the stimulated M1 (see article by Hallett8 for review). Because both M1 were stimulated at the same time, two forms of interhemispheric interaction may have complicated the interpretation of the CSP findings. One is transcallosal inhibition (TI), which refers to short interruption of M1 output by stimulation of the opposite M1, resulting in a silent period ipsilateral to the stimulated M1.9 The other is a shortening of the CSP if conditioned by stimulation of the other M1 10 to 20 milliseconds earlier.10 No such interhemispheric interaction was observed with intervals of 0 to 5 milliseconds, however.10 Therefore, the following discussion is limited to CSP and TI.

If it were correct that the EMG activity in the mirror hand came solely from the active M1 through its access to both sides of the spinal cord,1-3⇓⇓ the short CSP with active M1 stimulation would point to reduced CSP excitability, whereas the short CSP with stimulation of the nonactive opposite M1 would indicate a normal TI (similar values in table⇑ and in Meyer et al.9). In contrast to the current findings, bilateral M1 stimulation would not result in any CSP lengthening unless one assumes abnormal summation of the inhibitory effects of CSP and TI in the active M1. This is unlikely for two reasons. First, in the control subjects, simultaneous bilateral M1 stimulation resulted in CSP shortening, suggesting that the interaction between CSP and TI is normally occlusive. Second, in the patients, the arithmetic sum of CSP and TI in some of the unilateral M1 stimulation conditions was still shorter than the CSP elicited by simultaneous bilateral M1 stimulation, even at the adjusted lower stimulus intensity (see the table⇑). Thus, one would have to assume a more than linear summation to account fully for the observed CSP lengthening.

Therefore, findings strongly suggest a contribution of the mirror cortex to the EMG activity in both hands. The most straightforward scenario assumes a normal CSP and TI. The short CSP with unilateral stimulation of either M1 would then result from the access of the nonstimulated but active M1 to both sides of the spinal cord. The end of the CSP would coincide with the offset of TI of the nonstimulated M1.3,9⇓ CSP lengthening with bilateral M1 stimulation would result from interruption of sustained output from the mirror M1.

In patients with MM with no other neurologic abnormality2 and with X-linked Kallmann’s syndrome,4 PET studies showed abnormal bilateral M1 activation during intended unilateral hand movements. Activation of the mirror M1 was similar to that occurring during passive movements of the mirror hand,4 however, raising the possibility that the activation in the mirror M1 was not motor activity but was instead caused by the sensory feedback from the mirror hand.

Movement-related cortical potential recordings demonstrated that the premovement negativity that is normally distributed in the centroparietal region contralateral to the intended movement was distributed bilaterally in patients with MM.2,6,7⇓⇓ Because sensory feedback does not confound these measures, it appears that both M1 contribute to the preparation of an intended unilateral hand movement. This does not imply that the mirror M1 produced actual motor output, however. The current findings extend these results by suggesting that both M1 of patients with MM produce sustained motor output during the maintenance of an intended unilateral hand contraction.