Excessive Daytime Sleepiness in Parkinson Disease: A SPECT Study

PARKINSON’S DISEASE
Excessive Daytime Sleepiness in Parkinson Disease: A SPECT Study
Hideaki Matsui, MD1; Kazuto Nishinaka, MD1; Masaya Oda, MD, PhD1; Narihiro Hara, MD2; Kenichi Komatsu, MD1; Tamotsu Kubori, MD, PhD1 Fukashi Udaka, MD, PhD1
Department of Neurology and 2Department of Radiology, Sumitomo Hospital, Osaka, Japan
1
Study Objectives: The underlying pathologic mechanism of excessive
daytime sleepiness (EDS) in Parkinson disease and the relative contributions of brain function to this process are poorly understood. We compared
brain perfusion images between patients with Parkinson disease and EDS
and those without EDS using n-isopropyl-p-123I iodoamphetamine single
photon emission computed tomography.
Design: Clinical study.
Setting: Sumitomo Hospital.
Patients: Thirteen patients with Parkinson disease with EDS (EDS group)
and 27 patients with Parkinson disease without EDS (no-EDS group) were
studied. Whether or not each case had EDS was determined according to
the response to the Epworth Sleepiness Scale: patients with an Epworth
Sleepiness Scale score ≥ 10 were included in the EDS group, and patients with an Epworth Sleepiness Scale score ≤ 9 were included in the
no-EDS group.
Measurements and Results: There were significant hypoperfusions in
the left parietal and temporal association cortex in the EDS group. In the
multivariable logistic regression model, attention and decreased regional
cerebral blood flow of the left parietal association cortex and right caudate
and increased regional cerebral blood flow of the right thalamus were the
independent and significant factors.
Conclusions: The cortical hypofunction relative to hyperfunction of the
brain stem may relate to EDS in Parkinson disease. This is the first imaging study about EDS in Parkinson disease, and further studies are required.
Keywords: Parkinson disease, SPECT, Excessive daytime sleepiness,
Sleep disorder
Citation: Matsui H; Nishinaka K; Oda M et al. Excessive daytime sleepiness in parkinson disease: a SPECT study. SLEEP 2006;29(7):917-920.
INTRODUCTION
METHODS
SLEEP DISTURBANCE HAS BEEN RECOGNIZED AS AN
INTRINSIC PART OF PARKINSON DISEASE (PD). REPORTED PREVALENCE RATES RANGE FROM 75% to 98% of patients with PD.1,2 Sleep disturbance in PD can be classified into
disorders of sleep initiation and maintenance, excessive daytime
sleepiness (EDS), and so on. Especially EDS has received a great
deal of attention due to the sudden and irresistible sleep attacks,
with resulting automobile accidents in 9 patients with PD.3 However, the underlying pathologic mechanism of the development of
EDS in PD and the relative contributions of brain function to this
process are poorly understood.
Low levels of orexin or the influence of dopaminergic drugs
have been discussed as the culprit of EDS in PD, but these do not
seem to be the primary factors.4,5 On the other hand, EDS in PD
has hardly been discussed in the context of brain perfusion imaging.
This study aimed to compare brain perfusion images between
patients PD and EDS and those without EDS using n-isopropylp-123I iodoamphetamine (123I-IMP) single photon emission computed tomography (SPECT) and to investigate the mechanism of
EDS in this disease from the point of brain function.
Subjects
Forty consecutive patients with Hoehn Yahr stage 3or 4 PD
admitted to the Department of Neurology, Sumitomo Hospital,
were studied. Patients who were admitted for other diseases such
as pneumonia or bone fractures were excluded. Thirteen patients
with PD with EDS (EDS group) and 27 patients with PD without
EDS (no-EDS group) were studied. Whether or not each case had
EDS was determined according to the response to questions in
the Epworth Sleepiness Scale (ESS): patients with an ESS score
of 10 or greater were included in EDS group, and patients with
an ESS score of 9 or less were included in the no-EDS group. All
of the patients fulfilled the UK Parkinson Disease Society Brain
Bank criteria for idiopathic PD,6 and dopaminergic treatment was
effective for parkinsonian symptoms in all patients. No patients
had any other central nervous diseases. Patients suspected of
having other forms of parkinsonism, such as diffuse Lewy body
disease7 or multiple system atrophy8, were not enrolled. We differentiated patients with diffuse Lewy body disease from those
with PD with dementia by the 1-year rule.7 Brain magnetic resonance imaging, including T2 and T2* sequence, was conducted
in all patients for differential diagnosis. We examined the Mattis Dementia Rating Scale, Mini-Mental State Examination, revised version of Hasegawa’s Dementia Scale, Clinical Dementia
Rating, Hamilton Depression Scale, ESS, and 123I-IMP SPECT
within 1 day and during the on-medication and on-motor state if a
patient showed motor fluctuations. The ESS is a simple, standardized, 8-item questionnaire that is employed in clinical practice to
measure the level of daytime sleepiness.9
Written informed consent was received, and all patients agreed
to participate in this study. Details of patient profiles are shown in
Table 1.
Disclosure Statement
This was not an industry supported study. Drs. Matsui, Nishinaka, Oda,
Hara, Komatsu, Kubori, and Udaka have indicated no financial conflicts of
interest.
Submitted for publication September 14, 2005
Accepted for publication April 9, 2006
Address correspondence to: Hideaki Matsui, Department of Neurology, Sumitomo Hospital, 5-3-20 Nakanoshima, Kita-ku, Osaka, 530-0005, Japan; Tel:
81 6 6443 1261; Fax: 81 6 6444 3975; E-mail: [email protected]
SLEEP, Vol. 29, No. 7, 2006
917
EDS in PD: SPECT Study—Matsui et al
Table 1—Patients Profiles
PD with EDS
No.
Total
13
Men
2
Age, y
71.5 ± 6.9 (62-84)
Disease duration, y 12.0 ± 7.7 (1-30)
Hoehn-Yahr stage
3.2 ± 0.4 (3-4)
UPDRS motor
40.4 ± 11.1 (28-64)
score
CDR score
4.3 ± 5.2 (0-16)
DRS score
121.7 ± 19.0
(86-143)
DRS1, attention
27.4 ± 1.9 (23-29)
DRS2, initiation/ 29.0 ± 6.4 (16-37)
perseveration
DRS3, construction 5.8 ± 0.6 (4-6)
DRS4,
33.3 ± 6.7 (18-39)
conceptualization
DRS5, memory
26.3 ± 6.9 (12-34)
HDS-R score
21.6 ± 6.9 (9-30)
MMSE score
22.8 ± 5.2 (13-29)
HAMD score
13.2 ± 6.3 (4-24)
ESS score
15.6 ± 5.8 (10-24)
L-dopa, mg
322.7 ± 181.4
(100-600)
L-dopa equivalent
459.2 ± 257.4
dosage, mg
(100-1100)
Agonist dosage, mg 136.5 ± 179.9
(0-500)
Table 2—Volume of Interest Analysis
PD without EDS p Value
27
4
71.0 ± 9.1 (53-86)
8.7 ± 5.4 (2-19)
3.3 ± 0.4 (3-4)
31.1 ± 17.3 (8-63)
Parietal
association
cortex
Temporal
association
cortex
Frontal
association
cortex
Occipital
association
cortex
Posterior
cingulate
cortex
Anterior
cingulate
cortex
Medial
frontal
cortex
Medial
parietal
cortex
Caudate
nucleus
Pons
NS
.171
NS
.048
1.6 ± 3.4 (0-16)
134.0 ± 11.7
(91-144)
28.7 ± 1.2 (23-30)
31.9 ± 5.9 (16-37)
.096
.047
6.0 ± 0.0 (6-6)
36.9 ± 2.9 (26-39)
.190
.082
.043
.179
30.2 ± 3.3 (20-34) .072
25.2 ± 5.3 (10-30) .114
26.6 ± 3.3 (18-30) .031
13.0 ± 8.3 (4-33)
NS
2.6 ± 2.9 (0-8) < .0001
257.4 ± 158.5
NS
(0-550)
367.1 ± 230.4
NS
(0-925)
109.7 ± 121.5
NS
(0-375)
PD with EDS PD without EDS p Value
0.943 ± 0.071
0.990 ± 0.055
.047
0.956 ± 0.083
0.996 ± 0.067
.146
left
right
0.939 ± 0.038
0.971 ± 0.021
0.968 ± 0.045
0.976 ± 0.056
.046
NS
left
right
0.977 ± 0.011
0.993 ± 0.049
0.992 ± 0.054
1.005 ± 0.052
NS
NS
left
right
1.006 ± 0.071
1.015 ± 0.062
1.034 ± 0.059
1.028 ± 0.059
NS
NS
left
right
0.949 ± 0.053
0.954 ± 0.045
0.956 ± 0.053
0.970 ± 0.060
NS
NS
left
right
0.855 ± 0.063
0.846 ± 0.070
0.875 ± 0.053
0.873 ± 0.054
NS
NS
left
right
1.021 ± 0.053
1.038 ± 0.058
1.035 ± 0.048
1.032 ± 0.047
NS
NS
left
right
1.062 ± 0.109
1.074 ± 0.109
1.107 ± 0.073
1.104 ± 0.079
.196
NS
left
right
1.055 ± 0.084
1.012 ± 0.071
0.965 ± 0.051
1.050 ± 0.081
1.059 ± 0.092
0.922 ± 0.087
NS
.087
.055
Data are presented as mean ± SD. PD refers to Parkinson disease;
EDS, excessive daytime sleepiness. For comparisons between the 2
groups, unpaired and 2-tailed t-tests were used. Significance was set
at p < .05. NS, not significant (p > .2)
Data are presented as mean ± SD unless otherwise specified. PD refers
to Parkinson disease; EDS, excessive daytime sleepiness; UPDRS,
Unified Parkinson Disease Rating Scale; CDR, Clinical Dementia
Rating; DRS, Mattis Dementia Rating Scale; HDS-R, revised version
of Hasegawa’s Dementia Scale, MMSE, Mini-Mental State Examination; HAMD, Hamilton Depression Scale; ESS, Epworth Sleepiness
Scale. Levodopa (L-dopa) equivalent daily doses were calculated as
follows: 100 mg of standard levodopa equals 10 mg of bromocriptine
or 1 mg of pergolide, cabergoline, or pramipexole. In comparing the 2
groups, the unpaired and 2-tailed t-test was used. Significance was set
at p < .05. NS, not significant (p > .2)
Data Analysis
We performed volume of interest (VOI) analysis and compared
regional cerebral blood flow (rCBF) between the 2 groups of patients. We used computer software iNEUROSTAT (Nihon MediPhysics Co., Ltd., Nishinomiya, Hyogo, Japan) for VOI analysis.
It is a common practice in SPECT analysis to normalize a data set
to a reference region. In the present work, we used the formula:
Single Photon Emission Computed Tomography
Normalized perfusion rate = Perfusion rate / Global perfusion
rate.
Measurements were carried out during the on-medication
state in a quiet and dimly lit room with the subjects at rest and
awake in a supine position with their eyes closed. We directed
every patient to stay awake and monitored whether the patient
stayed awake while we acquired SPECT images. Whether or
not the patient could keep awake was confirmed by questioning soon after the measurement had ended. As a consequence,
all patients stayed awake. SPECT imaging was performed using
a Starcam3000XR/T (General Electric Company, Fairfield, CT).
Resolution was 10.5 mm full width half maximum, and the computer slice width was 6 mm. SPECT data were obtained in a 128
× 128 format for 64 angles with 30 seconds per angle. The study
was initiated 15 minutes after the intravenous injection of 167
MBq of 123I-IMP, and the total period of data acquisition was 32
minutes. A filtered back-projection method was used for image
reconstruction after preprocessing projection data with a Butterworth filter. A series of slices was reconstructed to be parallel to
the orbitomeatal line.
SLEEP, Vol. 29, No. 7, 2006
left
right
Secondly, we performed a 3-dimensional stereotactic surface
projection (3D-SSP) analysis10,11 using the computer software
iSSP35_2tZ in iSSP3 (Nihon Medi-Physics Co., Ltd.) set for comparison between the 2 patient groups. The extracted cortical activity of patients with EDS was compared with that of the no-EDS
group using a 2-sample Student t-test on a pixel-by-pixel basis.
Calculated t values were converted to Z values using a probability integral transformation. The pixels showing brain perfusion of
the EDS group that were significantly decreased compared with
no-EDS group were expressed on standardized brain MRI images.
Thirdly, we used multiple logistic regression analysis to make
comparisons between the 2 groups. We selected the Unified Parkinson Disease Rating Scale (UPDRS) motor score and DRS1,
2, 3, 4, 5 scores as variables from clinical parameters. We also
selected rCBF of the regions that showed differences (p < .2) in
918
EDS in PD: SPECT Study—Matsui et al
ment, depression, dementia, higher treatment doses of medication, and functional status, that have a noticeable impact on the
occurrence of EDS in patients with PD.15 In this study, we also
showed higher UPDRS motor scores and worse cognitive function, especially in attentional deficits, in patients with PD and
EDS. The close correlation between EDS and a more advanced or
demented person with PD shows that patients with a more widespread cerebral disturbance have an increased risk for developing somnolence. Several studies indicate that dopamine agonists
play a role in the genesis of EDS in PD, most indicating that the
sedating effect of dopamine agonists is related to the stimulation
of the inhibitory D2-like autoreceptors at the level of the ventral
tegmental area.16,17 The sedating effect is not unique to dopamine
agonists, and almost all parkinsonian drugs may cause EDS.18
However, because patients with new-onset PD who have not taken dopaminergic drugs also show EDS,4 EDS is not a secondary
phenomenon but, instead, is an intrinsic character of PD. In this
study, although there was a tendency for dopaminergic drug dosage to be higher in patients with EDS, we failed to demonstrate
a significant relationship between dopaminergic treatment and
EDS. This may be because of the small number of patients in this
study or because dopaminergic drugs are not a central factor in
EDS.
EDS in PD has rarely been discussed in the context of brain
perfusion image. In this study, we demonstrated cortical hypoperfusion, especially in the left parietal association cortex, and
hyperperfusion in the right thalamus in parkinsonian patients with
EDS. Because a lot of lesions were compared, there may be type
1 error. After correction for multiple comparisons, any lesions described in Table 2 did not remain significant. There may be also
type 2 error because of the relatively small number of patients
in this study. These are the study limitations; however, we may
suggest that the cortical hypofunction relative to hyperfunction of
the brainstem may relate to EDS in PD. Impaired neural network
between the cortex and brainstem may induce EDS and attention
deficits in patients with EDS.
Another study limitation is that we did not perform polysomnography in this study. Therefore, both imaging and polysomnography evidence are required in future studies.
In Kleine-Levin syndrome, thalamic hyperperfusion was reported in the hypersomnia period.19 Other authors have reported
left fronto-temporal hypoperfusion in both the hypersomnia period and the asymptomatic interval.20 In that case, EDS in PD and
Kleine-Levin syndrome may have some common mechanisms.
Figure 1—Three-dimensional stereotactic surface projection analysis.
The regions with Z values over 2.0 were shown. Data were obtained
using 3-dimensional brain structure-from-motion computer software.
the VOI analysis from rCBF parameters, except for global cortical
rCBF. JMP version 5.1 (SAS Institute, Inc., Cary, NC) was used
for statistical analysis, and the significance level was set at p <
.05.
RESULTS
Results of the VOI analysis are shown in Table 2. There were
significant hypoperfusions in the left parietal and temporal association cortex in the EDS group. Global cortical rCBF was also
decreased significantly in the EDS group compared with the noEDS group. This hypoperfusion was also demonstrated by 3-dimensional stereotactic surface projection analysis (Figure 1).
Multivariable logistic regression analysis (stepwise forward selection, p <.25 was enrolled) selected UPDRS motor score, DRS1,
DRS4, and rCBF of the bilateral parietal association cortex; left
temporal association cortex; right caudate; and right thalamus (p
= .0009, R2 = 0.526). In this logistical regression model, DRS1
(attention) (p = .043) and decreased rCBF of the left parietal association cortex (p = .049) and right caudate (p = .030) and increased rCBF of the right thalamus (p = .0085) were independent
and significant factors. The rCBF of the other lesions such as the
pons and left temporal association cortex did not remain significant.
CONCLUSION
DISCUSSION
We demonstrated that left parietal hypoperfusion and right
thalamus hyperperfusion were significant and independent factors in EDS in PD. This is the first imaging study about EDS in
PD, and further studies are needed.
At present, there is no universally accepted definition of EDS.
Some researchers define the disorder as a score of 7 or greater on
the ESS, whereas others use an ESS score of 10 or more.12 Significant EDS is found in 15% to 20% of those who have PD compared with a rate of 1% in healthy elderly individuals.12 Gjerstad,
et al reported that the rate of EDS in patients with PD increases
by 6% per year.13 In 1999, the scientific community was alerted
to the sleep-attack phenomenon in PD.3 High ESS scores were the
main risk factors for the occurrence of sleep attacks.14 However,
during this study, there was only 1 patient who had such a sleep
attack.
There are several risk factors, such as age, cognitive impairSLEEP, Vol. 29, No. 7, 2006
REFERENCES
1.
2.
3.
919
Olanow CW, Watts RL, Koller WC. An algorithm (decision tree)
for the management of Parkinson’s disease: treatment guidelines.
Neurology 2001;56(Suppl 5):S1-88.
Lees AJ, Blackburn NA, Campbell VL. The nighttime problems of
Parkinson’s disease. Clin Neuropharmacol 1988;11:512-9.
Frucht S, Rogers JD, Greene PE, Gordon MF, Fahn S. Falling asleep
at the wheel: motor vehicle mishaps in persons taking pramipexole
and ropinirole. Neurology 1999;52:1908-10.
EDS in PD: SPECT Study—Matsui et al
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Fabbrini G, Barbanti P, Aurilia C, Vanacore N, Pauletti C, Meco G.
Excessive daytime sleepiness in de novo and treated Parkinson’s
disease. Mov Disord 2002;17:1026-30.
Overeem S, van Hilten JJ, Ripley B, Mignot E, Nishino S, Lammers
GJ. Normal hypocretin-1 levels in Parkinson’s disease patients with
excessive daytime sleepiness. Neurology 2002;58:498-9.
Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical
diagnosis of idiopathic Parkinson’s disease: a clinico-pathological
study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181-4.
McKeith IG, Galasko D, Kosaka K, et al. Consensus guidelines for
the clinical and pathologic diagnosis of dementia with Lewy bodies
(DLB): report of the consortium on DLB international workshop.
Neurology 1996;47:1113-24.
Gilman S, Low PA, Quinn N, et al. Consensus statement on the diagnosis of multiple system atrophy. J Neurol Sci 1999;163:94-8.
Johns MW. A new method for measuring daytime sleepiness: the
Epworth Sleepiness Scale. Sleep 1991;14:540-5.
Minoshima S, Frey KA, Koeppe RA, Foster NL, Kuhl DE. A diagnostic approach in Alzheimer’s disease using three-dimensional stereotactic surface projections of fluorine-18-FDG PET. J Nucl Med
1995;36:1238-48.
Minoshima S, Giordani B, Berent S, Frey KA, Foster NL, Kuhl DE.
Metabolic reduction in the posterior cingulate cortex in very early
Alzheimer’s disease. Ann Neurol 1997;42:85-94.
Larsen JP. Sleep disorders in Parkinson’s disease. In: Gordin A,
Kaakkola S, Teräväinen H, eds. Parkinson’s Disease: Advances in Neurology, vol. 91. Philadelphia: Lippincott Williams &
Wilkins;2003;329-34.
Gjerstad MD, Larsen JP. Development of daytime somnolence over
time in Parkinson’s disease. Neurology 2002;58:1544-6.
Paus S, Brecht HM, Koster J, Seeger G, Klockgether T, Wullner U.
Sleep attacks, daytime sleepiness, and dopamine agonists in Parkinson’s disease. Mov Disord 2003;18:659-67.
Simuni T. Somnolence and other sleep disorders in Parkinson’s
disease: the challenge for the practicing neurologist. Neurol Clin
2004;22:S107-26.
Rye DB, Jankovic J. Emerging views of dopamine in modulating sleep/wake state from an unlikely source: PD. Neurology
2002;58:341-6.
Ferreira JJ, Galitzky M, Thalamas C, et al. Effect of ropinirole on
sleep onset: a randomized placebo controlled study in healthy volunteers. Neurology 2002;58:460-2.
The European Agency for the Evaluation of Medicinal Products.
Evaluation of Medicines for Human Use. CPMP Position Statement: dopaminergic substances and sudden onset of sleep. Available at: http://www.emea.eu.int/pdfs/human/press/pos/057802.pdf.
Accessed: 1, Sep, 2005.
Ogawa Y, Kanbayashi T, Shimizu T. Kleine-Levin syndrome. Nihon
Rinsho (Ryouikibetushoukougun) 2003;39:61-3.
Landtblom AM, Dige N, Schwerdt K, Safstrom P, Graner. A case of
Kleine-Levin syndrome examined with SPECT and neuropsychological testing. Acta Neurol Scand 2002;105:318-21.
SLEEP, Vol. 29, No. 7, 2006
920
EDS in PD: SPECT Study—Matsui et al