The effect of cholinergic enhancement on brain activation in Parkinson*s disease patients with visual hallucinations using fMRI.

Visual hallucinations (VH) are common in Parkinson*s disease (PD) with a
prevalence of approximately 30%. The exact etiology of VH in PD is unknown,
however a combination of impaired visual processing and attention seems to be
involved.
In our previous study, we used movies with images gradually revealed out of
noise (pop-out movies) to investigate if PD patients with VH had a specific
cerebral activation pattern before recognition of the image, compared to PD
patients without VH and healthy controls. We found that PD patients with VH had
decreased activation of the lateral occipito-temporal cortex (middle occipital
and middle temporal gyri) before pop-out, compared to both PD patients without
VH and healthy controls. We hypothesized that this impairment of the
occipito-temporal cortex in PD patients with VH might predispose to VH via a
*release* mechanism of higher order visual cortices.
Other studies have shown that apart from the dopaminergic neurotransmitter
system, the cholinergic system seems to be involved in the pathogenesis of VH
in PD. The cholinergic system plays an important role in conscious awareness. A
decrease in the cortical acetylcholine (Ach) levels impairs the selection of
subcortical information streams, causing unselected and chaotic cortical
activation, which may predispose to hallucinations.
Clinical evidence shows that visual hallucinations can be induced by
anti-cholinergics, while acetylcholinesterase inhibitors (AChE-I) ameliorate
cognitive dysfunction and VH in PD.
Cholinergic projections from the basal forebrain to the cortex include
projections to primary and associative visual cortical regions. Non-demented PD
patients have reduced cholinergic activity in occipital and temporal cortical
regions while PD patients with dementia have a more extensive cholinergic loss,
especially in the middle temporal gyrus, compared to healthy controls.
The previously found reduced activation of occipital and temporal cortex before
visual pop-out in PD patients with VH, compared to PD patients without VH and
healthy controls was independent of cognitive dysfunction. Possibly, this
relative deactivation in PD patients with VH is due to reduced cholinergic
innervation to the occipital and temporal cortex. This might, at least partly,
explain beneficial effects of AChE-I on VH in PD and PD dementia.
Therefore, administration of AChE-I might *normalize* occipital and temporal
cortex activation during a visual pop-out task in PD patients with VH.
Secondly, the extent of change in cerebral activation after administration of
AChE-I might positively correlate with the clinical response of PD patients
with VH on AChE-I.

To investigate cerebral activation patterns after administration of an AChE-I
(rivastigmine), compared to activation after placebo, in PD patients with VH.
Secondly, to investigate whether cerebral activation changes after rivastigmine
correlates with individual clinical response on rivastigmine.

fMRI will be performed during a dynamic presentation of images popping out of
noise after placebo and after administration of rivastigmine. Administration of
placebo or rivastigmine will be single-blind (i.e. blind to the subject). The
first intervention fMRI-session will take place at t=0 and the second
intervention fMRI will take place after several days, up to one week (t=1).
Intervention timing will be balanced, i.e. half of the subjects will receive
placebo at t=0 and rivastigmine at t=1, while the other half will receive
rivastigmine at t=0 and placebo at t=1.
After the last scanning session, PD patients with VH will be treated with a
rivastigmine plaster according to a fixed titration schedule (4.6 mg/24 hours,
after 1 month 9.5 mg/24 hours, highest tolerable dose). During the first
session and after 2 months rivastigmine use, cognition and extent of VH will be
assessed using the SCales for Outcomes in Parkinson's disease - part cognition
(SCOPA-cog) and the University of Miami Parkinson*s disease Hallucinations
Questionnaire (UM-PDHQ), respectively. Also after 2 months, the association
between clinical response on rivastigmine (i.e. improvement on the SCOPA-cog
and the UM-PDHQ) and cerebral activation changes will be investigated.

fMRI will be performed in two scanning sessions during which subjects watch a
dynamic presentation of images popping out of noise either after placebo or
after administration of rivastigmine. Several days before the first session,
subjects will receive a transdermal patch with either rivastigmine (9.5 mg/24
hours) or placebo by post. They will be instructed to apply this patch the
evening before the scan, before going to bed. It has been shown that
rivastigmine concentration rises slowly after application of the patch and
reached a plateau between 8 and 26 hours (median tmax 14.1 hours)12. The
content of the plaster (rivastigmine or placebo) will be single-blind (i.e.
blind to the patient).
The first intervention fMRI-session will take place at t=0 and the second
intervention fMRI will take place after several days, up to one week (t=1).
Intervention timing will be balanced, i.e. half of the subjects will receive
placebo at t=0 and rivastigmine at t=1, while the other half will receive
rivastigmine at t=0 and placebo at t=1.
After the last scanning session, PD patients with VH will be treated with a
rivastigmine patch according to a fixed titration schedule (4.6 mg/24 hours,
after 1 month 9.5 mg/24 hours, highest tolerable dose).

Local irritation patch (minor chance because of relatively short duration of
patch application)
Adverse events of rivastigmine (less chance than with capsule, because of the
slow rise of the blood concentration and absence of a dose peak)