Coherence Analyses in Hyperkinetic Movement Disorders

Hyperkinetic neurological movement disorder (HMD) encompass, among others,
dystonia, myoclonus and tremor. The exact pathophysiology of most of these
abnormal involuntary movements is unknown. Occasionally, HMD is due to lesions
in the nervous systems, but usually it is due to functional changes in
dysfunctional neural networks including the so called *cortico-basal ganglia
thalamo-cortical (CBGTC) loop* (Lehéricy et al., 2013).
To increase our knowledge of the pathophysiology of HMD we will study
functional changes in the motor systems in HMD by using analysis techniques,
i.e. frequency domain (coherence) and time domain (cumulant density) analyses
of recordings at different levels of the motor pathway.
Coherence is the phase synchrony of two or more signals and can be quantified
using the cross-correlation function (Windhorst & Johansson, 1999). To better
determine whether signal A causes B in coherent signals or vice versa,
additional cumulant density analyses can be performed (Windhorst & Johansson,
1999).
Further understanding of the pathophysiology of HMD could lead to a better
application of the current DBS techniques by stimulating where neural activity
is most disturbed. Next to this, further understanding of the pathophysiology
of HMD could also lead to future, adaptive DBS algorithms (Little et al.,
2013), that only stimulate when neural activity is disturbed.
DBS of different basal ganglia targets has proven effective for several
movement disorders. The internal part of the globus pallidus (GPi) has become
an established treatment for primary generalised dystonia. The same holds for
stimulation of the ventral intermediate nucleus (Vim) of the thalamus in ET. PD
tremor is treated with either GPi or subthalamic nucleus (STN).
In dystonia, converging evidence shows that excessive oscillatory activity in
the brain drives dystonic muscles (Foncke et al., 2007a; Sharott et al., 2008;
Tijssen et al., 2000; Tsang et al., 2012). This so called *dystonic drive* was
detected in idiopathic dystonia in which 4-7 Hz coherent, phase locked EMG
activity was present in cervical musculature (Tijssen et al., 2000). This
pathological coherence pattern was also seen between LFP*s of the GPi and
dystonic muscles (Foncke et al., 2007a; Sharott et al., 2008). These findings
support the disrupted oscillatory activity in the previous mentioned CBGTC loop
(Lehéricy et al., 2013). Little data is available on the effect of DBS
post-operatively on the intra-muscular dystonic drive and whether DBS
electrodes with the highest coherence in these lower 4-7 Hz frequencies are
most suitable for (adaptive) stimulation.
In ET, cortico-muscular coherence is also present. The frequency of this
coherence is at the tremor frequency of around 6-8 Hz (Hellwig et al., 2001).
Similar subcortico-muscular coherence patterns are also present between the
VIM and the involved muscles (Pedrosa et al., 2012). However, it is not known
whether strong cortico-muscular coherence patterns predict DBS outcome and
whether LFP*s with the highest subcortico-muscular coherence are best suitable
for (adaptive) stimulation.
PD tremor is typically a resting tremor among 3-7 Hz. Recently, a direct
relationship among STN oscillations at tremor frequency and tremor
manifestation was discovered (Hirschmann et al. Brain 2013). This coherence at
tremor frequency was also present between cortex and STN. However, it is not
known whether DBS works most effective in patients with the highest LFP-EMG or
EEG-EMG coherence profiles.
By studying coherence at 3 points of the motor pathway: cortex (EEG), basal
ganglia nuclei (LFP) and muscle (EMG) before, during (LFP) and after DBS
surgery in HMD we not only increase our understanding of the pathophysiology of
HMD but also of the relation between neurophysiological and clinical changes in
HMD.

1. Investigate whether per-operative subcortico-muscular (LFP-EMG) coherence
can be used as a predictor for favourable surgical outcome in DYS and PD tremor
patients undergoing DBS surgery
2. Investigate whether pre-surgical EEG-EMG-coherence and cumulant density
profiles alone predict the efficacy of DBS in DYS and PD tremor.
3. Investigate whether the change in pre- and post-surgical cortico-muscular
(EEG-EMG) and intra-muscular (EMG-EMG) coherence can predict the efficacy of
DBS in DYS and PD tremor.
4. Investigate whether intra-operative subcortico-muscular coherence assist in
selecting stimulation contacts in DYS, ET and PD patients undergoing DBS.
5. Investigate whether there is a relation between cortico-muscular,
subcortico-muscular and intra-muscular coherence and whether a combination of
these three coherence methods can result in a strong predictor of DBS efficacy
in DYS and PD.

A prospective cohort study with blinded endpoint evaluation will be conducted
at the department of Neurology and Neurosurgery of the University Medical
Center Groningen.

Besides the risks involved in DBS surgery for clinical care as usual, the risks
involved for the participating patients are estimated to be negligible. During
the recording interval that comprises circa 10 minutes and coincides with
clinical testing, sterile recording electrodes are temporary connected to the
implanted DBS leads. This procedure does not involve invasive techniques. For
this reason there is no increased risk for intra-cerebral haemorrhage or other
sequalae like seizures or brain shift. The theoretical risk of infection is
assessed to be negligible since the electrodes which connect the implanted DBS
leads to the EEG amplifier are sterile. Moreover, these electrodes are also
used for teststimulation and the actual extra (temporary) connection made for
measurements, will take place in the non-sterile environment. Additionally, the
EMG electrodes are not located in the sterile operation area and therefore do
not contribute to an infection risk. For above mentioned reasons, the DBS
working-group concluded the patient*s risk for these intra-operative
measurements to be marginal. The extra pre- and post-surgery investigations in
PD and DYS will require approximately 2 hours of the patient*s time in which
the subjects will undergo a brief neurological examination and EEG-EMG. In ET
DBS battery replacement patients the added risk of measurements is expected to
be marginal, since the same reasons mentioned above are applicable here. There
are no potential benefits for the participating patients.