What keeps the brain alert? - A magnetic resonance study into the neural correlates of vigilance in narcolepsy patients

Narcolepsy with cataplexy is a primary sleep disorder caused by a loss of
hypocretin (orexin)-producing neurons in the lateral hypothalamus and the
perifornical area. Hypocretin-producing neurons project to brain regions
involved in attention, cognition and the regulation of sleep-wake behaviour,
such as the locus coeruleus, the raphe and tuberomammillary nuclei and basal
forebrain regions. Reduction in activity in these regions due to loss of
hypocretin signalling causes the characteristic excessive daytime sleepiness
(EDS, tendency to fall asleep), and other symptoms such as the cataplexy (a
sudden bilateral loss of muscle tone with retention of consciousness provoked
by strong emotions). Several studies have shown that EDS may lead to severely
impaired daytime performance. This can lead to dangerous situations in the
working environment of the patients, or in similarly demanding situations, such
as in traffic.

- Brain anatomy -
By means of neuroimaging techniques, no consistent structural brain
abnormalities have yet been found in narcoleptic patients compared to healthy
control subjects. Studies using magnetic resonance (MR)-based voxel-based
morphometry, which allows statistical comparisons of local tissue composition
(both grey and white matter) across the whole brain, have yielded mixed
results. Whereas several studies found grey matter volume-decreases in the
hypothalamus, thalamus, nucleus accumbens and fronto-temporal cortices, others
could find no differences between patients and controls at all, or only in part
of the aforementioned brain areas.

Diffusion tensor imaging (DTI) provides insight in yet another modality of the
brain anatomy. This relatively quick scan reveals the diffusion direction of
water, which usually is greater parallel to the axon than perpendicular to the
axon, and this therefore believed to reflect the degree of tissue (white
matter) organisation or alignment. In one study full brain DTI has been
performed on patients with idiopathic narcolepsy, with a field strength of 1.5
T, revealing widespread abnormalities in brain regions thought to be involved
in the pathophysiology of idiopathic narcolepsy. Another, complementary study
objectively identified abnormalities in the hypothalamus, consistent with the
hypocretin neuron loss, and the targets of these neurons. A third, recent study
compared shows morphological abnormalities in the left amygdala, the left
inferior frontal gyrus and the left postcentral gyrus in NC patients.

- Brain function -
There is only a limited number of recent studies that have investigated the
resting, awake brain of NC patients at the functional level. Two positron
emission tomography (PET) studies investigated glucose metabolism and provide
us with mixed results. First, a study with 24 NC patients and an equal number
of control subjects indicated hypometabolism in the hypothalamus and thalamus,
in line with the anatomical findings discussed earlier. The other group of
investigators, however, reported a lack of this hypometabolism and instead
suggested an increase of metabolism in the cingulate and visual association
cortices of NC patients compared to controls. In addition, a single photon
emission computed tomography (SPECT) study showed decreased blood flow by in
the hypothalamus and thalamus, in accordance with the first PET study, further
supplemented by decreased blood flow in other brain areas mostly overlapping
structural changes, as reviewed recently.

An additional technique, magnetic resonance spectroscopy (MRS), provides the
ability to analyse the chemical composition of living brain matter
non-invasively. Only a fairly small number of studies using this technique have
been published over the last two decades. After the first study, the
investigators had to conclude that, in narcolepsy patients, there is no
evidence of neuron loss or *gross biochemical abnormality* in the
pontomedullary junction. It was not until six years later that another group
reported loss or damage of neurons in the hypothalamus in NC patients compared
to controls. Whereas in these studies the analyses have only been performed on
one particular region of interest, a third MRS paper appeared, publishing the
results of a more integrative approach by investigating metabolic changes in
multiple regions of interest. Fourteen NC patients and an equivalent number of
matched, healthy controls were subjected to the MRS protocol. The results
indicated that the metabolic rate in the right amygdala was decreased in NC
patients, which supposedly is caused by a loss of integrity of hypothalamic
neurons (indicating that the neurons are damaged in one or another way).

No studies have yet investigated brain activity in NC patients in relation to
alertness and sleep resistance. The aim of the current study is to obtain more
insight into the regulatory mechanisms controlling alertness and sleep
resistance in patients with narcolepsy with cataplexy. To this aim, we will for
the first time, perform combined EEG/fMRI recordings in narcolepsy patients and
matched control subjects.

Following telephone screening for inclusion and exclusion criteria, the
patients and controls will receive information about the purpose and the
methods of this study. If the subjects agree to participate informed consent
will be acquired.

The subjects that meet the criteria and have given informed consent will be
invited to come to the Spinoza Centre for Neuroimaging. The patients are
required to refrain from medication for at least 14 days prior to the study
day. Furthermore, all subjects are not allowed to use caffeine-containing
beverages in the 24 hours preceding the appointment.

The subjects will be tested in the afternoon. Upon arrival, the subjects will
be provided more information on the techniques, if desired. Furthermore, the
subjects will undergo (verbal) IQ assessment using the National Adult Reading
Test (NART) and assessment of daytime sleepiness by means of the Epworth
Sleepiness Scale (ESS). Additionally, the Edinburgh Handedness score will be
assessed, determining whether the particular subject is left- or right handed.
Then, after equipping the subjects with the EEG electrodes, they will be placed
in the MR scanner, and the study protocol will be initiated.

- fMRI and EEG measurements -
Subjects will be scanned using a 3 T Philips MR scanner. This system includes a
computer to generate visual and auditory stimuli and to record the
participants* responses, a beamer to project visual stimuli in the MR room, an
MR-compatible headphone for auditory stimulation, a screen to display visual
stimuli and response boxes for the registration of the manual responses. Blood
oxygenation levels dependent (BOLD) responses will be recorded by the MR
scanner. The electroencephalogram (EEG) will be recorded using an MR-compatible
system. Sleep will be assessed according to the 2007 American Academy of Sleep
Medicine criteria, with sleep stages scored in 30-second epochs according to
the current guidelines. Eyetracking will be used to assess whether subjects
have their eyes open or closed and whether patients* eyes are wandering or
focussed.
Scanning will be initiated with 5 minutes of structural scanning, followed by a
resting state scan (5 minutes) and scanning during the experiment
(approximately 25 minutes), respectively. Furthermore, the white matter
structure will be assessed (DTI, 6 minutes) and the neurotransmitter content of
both amygdalae and the hypothalamus will be determined by MRS (specifically
GABA and glutamate).

- Sustained Attention to Response Task (SART) -
Before and after SART administration a blank screen with a white dot in the
centre will be shown to the subject. The SART is administered comprising blocks
of approximately 30 seconds (27 stimuli in one block with 1150 ms per stimulus,
yielding 31.05 s per block) with two different difficulty levels and blocks of
equal length during which no action is required. Each of the nine target digits
will be shown in an equal number of times in a predetermined quasi-random
order, so that the identical target digits are not clustered. The font size
will be chosen at random from 52, 72 or 144 points. In the easy blocks, each
digit is presented for 250 ms, followed by a blank screen that lasts for 900
ms. A second difficulty level will be constructed and added to the SART to
enable analysis of not just baseline performance and activity-correlates, but
to add the possibility of analysing the changes in activity patterns in
response to an increase of difficulty. In the second difficulty level the
target will be presented for 100 ms. The subjects have the same 900 ms to
respond. Furthermore, the difference between the presentation lengths will be
accounted for by adding 100 ms between the end of the reaction time and the
initiation of the next stimulus. This way, the temporal aspect of the cycle
remains identical in both difficulty levels.
The subjects have to respond to the appearance of each digit by pressing a
button before the next digit appears, except when the digit *3* is shown. They
will be instructed to pay equal attention to speed as to accuracy. The SART
error score consist of the total number of errors, expressed as the cumulative
number of times the button is pressed when a *3* is shown and the number of
times that the button is not pressed when it should have been.

- Sleep resistance -
After the SART the subjects will be asked to fight sleep and to rest quietly in
alternating 30-second blocks while keeping their eyes closed, adding up to 5
minutes. Before every *fighting-sleep*-block, subjects will be woken when
asleep. After the last block, subjects are free to fall asleep and may continue
to sleep for 15 minutes. The presence and staging of sleep will be assessed
using the MR-compatible EEG recording.

- DTI -
In addition to the standard anatomical scan that will be performed before the
resting state scan (as described above) we will analyse the white matter
microstructure by performing diffusion tensor imaging (DTI). In the current
study, data acquisition will be performed at a field strength of 3 T, which
enables us to analyse the white matter organisation in much more detail than
the previous studies managed to do. Furthermore, performing DTI in this study
provides the additional opportunity to use task performance as a regressor in
the DTI analyses. Whole brain DTI in human subjects will cover approximately 6
minutes of scanning time.

- MRS -
Magnetic resonance spectroscopy (MRS) is a technique that allows us to
determine the brain (voxel) content of a particular substance. We would like to
assess the levels of GABA and glutamate in both amygdalae and the hypothalamus,
and to identify potential correlations between vigilance and emotion regulatory
processes and the content of these two important classes of neurotransmitters.
These data will be acquired with a voxel volume of 1 cm^3 (3 voxels in total)
and will take 15-20 minutes.

All subjects will be subjected to a functional MR scan. This scan will take
around half an hour to be completed. There are no risks connected to MR
scanning. There is, however, a chance that the researcher detects anomalies in
the anatomical scan, in which case the attending doctor will be notified.

Preferably non-medicated patients will be included, which abolishes the burden
of treatment cessation. Those patients that will be asked to stop their
treatment for at least 14 days will experience a temporary fall-back in their
symptoms and performance.
All subjects will be asked to fill out a few short questionnaires prior to
scanning. All subjects will be equipped with a net of EEG electrodes and will
undergo MR scanning. These methods bring the burdens of laying still for half
an hour and washing out the gel underneath the EEG electrodes.

The profit for the patients is that more insight will be gained in the
neurobiology of attention and vigilance of in narcolepsy-cataplexy patients
that have the extra burden of poor daytime performance, which might eventually
lead to successive investigations for alternative or additive
medication/treatments improving the quality of life of these patients.